Maritime Archaeology

A Reader of Substantive and

Theoretical Contributions
The Plenum Series in Underwater Archaeology
Series Editor:
J. Barto Arnold III
Institute of Nautical Archaeology
Texas A&M University
College Station, Texas

Maritime Archaeology: A Reader of Substantive and Theoretical

Contributions
Edited by Lawrence E. Babits and Hans Van Tilburg

The Persistence of Sail in the Age of Steam: Underwater Archaeological

Evidence from the Dry T ortugas
Donna J. Souza

A Continuation Order Plan is available for this series. A continuation order wi11 bring delivery of each
new volume immediately upon publication. Volumes are bi11ed only upon actual shipment. For further
information please contact the publisher.
Maritime Archaeology
A Reader of Substantive and
Theoretical Contributions

Edited by

Lawrence E. Babits
East Carolina University
Greenville, North Carolina

and

Hans Van Tilburg

University of Hawaii at Manoa
Honolulu, Hawaii

Springer Science+Business Media, LLC

 Library of Congress Cataloglng-ln-Publ1 c a t i o n Data

Maritime archaeology : a r e a d e r of s u b s t a n t i v e and t h e o r e t i c a l

c o n t r i b u t i o n s / e d i t e d by Lawrence E. B a b l t s and Hans Van T i l b u r g .
p. cm. — (The Plenum s e r i e s in underwater archaeology)
I n c l u d e s b i b l i o g r a p h i c a l r e f e r e n c e s and index.

1. Underwater archaeology. I . B a b i t s , Lawrence Edward. I I . Van

T i l b u r g , Hans. I I I . S e r i e s .
CC77.U5M366 1998
930. 1'028'04~dc21 97-49977
CIP

ISBN 978-0-306-45331-1 ISBN 978-1-4899-0084-5 (eBook)

DOI 10.1007/978-1-4899-0084-5

© Springer Science+Business Media New York 1998

Originally published by Plenum Press, New York in 1998
Softcover reprint of the hardcover 1st edition 1998
http://www.plenum.com

1098765432 1

All rights reserved

No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any
means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written
permission from the Publisher
 This work is respectively dedicated to four maritime archaeologists who first saw
the importance of underwater cultural resources and helped create a new discipline

George F. Bass
Keith Muckelroy
Reynold Ruppe
Peter Throckmorton

and to the next generation of maritimers

who will continue in their tradition of excellence and professionalism
 Series Preface

This volume initiates a new series of books on maritime or underwater archaeology,

and as the editor of the series I welcome its appearance with great excitement. It is
appropriate that the first book of the series is a collection of articles intended for gradu-
ate or undergraduate courses in underwater archaeology, since the growth in academic
opportunities for students is an important sign of the vitality of this subdiscipline. The
layman will enjoy the book as well.
Academic and public interest in shipwrecks and other submerged archaeological
sites is indicated by a number of factors. Every year there are 80 to 90 research papers
presented at the Society for Historical Archaeology's Conference on Historical and
Underwater Archaeology, and the Proceedings are published. Public interest is shown
by extensive press coverage of shipwreck investigations.
One of the most important advances in recent years has been the passage of the
Abandoned Shipwreck Act of 1987, for the first time providing national-level law con-
cerning underwater archeological sites. The legislation has withstood a number of legal
challenges by commercial treasure salvors, a very hopeful sign for the long-term pres-
ervation of this nonrenewable type of cultural resource.
The underwater archaeological discoveries of 1995 were particularly noteworthy.
The Texas Historical Commission discovered the Belle, one of La Salle's ships, and the
CSS Hunley was found by a joint project of South Carolina and a private nonprofit
organization called NUMA.
In many areas there is active participation of avocationals with underwater archae-
ologists is researching and investigating shipwreck sites. The recreational dive commu-
nity has the potential to provide assistance in studying underwater archaeological sites. It
is the responsibility of the archaeological community to embrace this opportunity and
provide the guidance and education needed by the divers. Volunteers are particularly

vii
viii Series Preface

adapted to the predisturbance mapping and site-recording phases of fieldwork. Publica-

tions in the Plenum series will be particularly useful for avocationals.
As a field of study, maritime archaeology continues to grow and expand. Like
other archaeologists, underwater archaeologists have an obligation to publish their
results; the series is a great new channel to fulfilling that obligation. This book and the
initiation of The Plenum Series in Underwater Archaeology will make 1998 a milestone.

J. Barto Arnold III

 Preface

Refinement of the self-contained underwater breathing apparatus (SCUBA) during

World War II gave more and more people an opportunity to spend time underwater. On
repeated occasions, divers encountered the remains of past human activity, ranging
from eroded prehistoric sites through sunken vessels to entire towns. As word of these
accidental discoveries spread, site visitation increased and damage occurred.
Eventually, news of the sunken sites reached the archaeological community and
divers already trained in archaeological techniques began to work underwater. In many
cases, people trained as archaeologists underwent dive training in order to work on
sites related to their academic interests. As underwater technical knowledge increased,
more archaeologists and historians, both amateur and academic, began to dive. For
many, the mask, fins, tanks, and slates were simply an outgrowth of trowels, shovels,
and other equipment used on land.
Development of specific techniques for underwater research occupied most exca-
vators and few thought to delve into the theory behind the archaeology. This practical
approach was not a problem for those trained as archaeologists. For those trained as
historians, or diving simply to recover something of the past, a basic understanding of
what is meant by archaeology, its assumed meanings, and its relation to the scientific
method was missing. Theoretical problems grew as students and amateurs alike
engaged in the recovery of submerged cultural artifacts.
While implications of a specific anthropological orientation permeating the field
can be seen throughout this text, development of a scientific methodological frame-
work allowing replicative testing is heavily emphasized. Although a scientific para-
digm often shifts slightly, its use allows others to check and verify conclusions which
should push investigators toward accurate data recovery and interpretive conclusions.
In some ways, this development paralled Historical Archaeology. Many terres-
trial archaeologists said, often in writing, that working underwater was not archaeol-
ogy, possibly because of the publicity, the financial riches of some sites, and the rogue

ix
x Preface

image associated with divers and presented in the media. The legacy of the cowboy
image persists today but diver training, both in techniques of working underwater and
the academic aspects of archaeology have improved greatly.
Diving archaeologists need a foundation in theoretical issues and the "why" of
archaeological endeavor if they are to carry the field of archaeology into the next cen-
tury. This book is written for those who wish to convey a knowledge of the past within
a scientific framework, especially students with serious interests in locating and inter-
preting past human behavior.
The first part (I) includes an introduction to the field. This is followed by a series
of articles discussing what the field of archaeology actually involves and how it relates
to the past. The sometimes acrimonious division between salvors and archaeologists is
included here because they represent two different extremes of the same effort.
Part II presents information and sources on several areas of the world. An emphasis
on writings outside the mainstream suggests areas which need additional effort, particu-
larly native African and Asian maritime material culture. The well-known Caribbean,
Mediterranean, and northern European areas are covered with introductory bibliographic
materials which are readily available.
The two parts that follow deal with the theory behind the scientific method (IV)
and how one finds sites (V). The part on scientific theory includes two classic articles
by Chamberlin and Platt which are available only in microfilm. Finding sites requires
knowledge of theory, the site formation process, and the sophisticated equipment (VI)
used to search below the surface for sites. The readings on survey apply theoretical and
practical matters, especially in Part VI, which includes material on the sophisticated
technology used to locate sites. Since this technology changes so rapidly, early, basic
articles are presented. These are supported by a more current bibliography.
The importance of any site varies from person to person depending on their back-
ground and interests. Part VII provides readings on how one determines significance,
an image-laden word related to funding, protection, and publicity about the site. With-
out some idea ofa site's significance, and its relationship to other sites, generating sup-
port for a research effort is problematic.
How information is recovered and conserved is presented in Parts VIII and IX.
Methods of examining a site before it is disturbed, how the site is recorded during
excavation, and working in zero visibility conditions are covered. A section (IX) on
how artifacts are conserved and why follows. The last two parts deal with the legality
of working on underwater sites (X) and how a site is interpreted, and the information
given to the public (XI). In some ways, these two areas represent both ends of the
underwater spectrum. Salvors and archaeologists have confronted each other in the
courts on numerous occasions and the often spectacular message of treasure hunting
salvors seems to reach more of the public than does archaeological work. The last part,
on interpreting underwater sites and creating exhibitions is designed so the archaeologi-
cal community'S next generation realizes the ultimate responsibility lies in informing
the public, rather than their peers, of what they have learned. Two appendices provide
information on additional source materials, the archaeological journals which publish
underwater findings, and a chronologically oriented bibliographic listing of sites, ships,
and research. The entire text often shifts from theoretical to practical methodology and
experiences in the underwater realm. This is deliberate because it encourages those new
Preface xi

to the field to recognize interrelationships between theoretical positions and the thought
processes involved in solving practical problems.
This collection is not an attempt to be complete, nor up-to-date, but to provide a
selection of certain hard-to-find and/or out-of-print articles that address aspects of
maritime archaeological theory and perspective.
 Acknowledgments

This book is the result of student requirements for a text on maritime archaeology. The
students who went through three years of History 6805 without a text and made com-
ments about their readings played a major role in creating this book.
The authors and publishers who granted us permission to reprint their material
made the book a reality. We owe special thanks to Eliot Werner of Plenum for guiding
us gently into the publishing business and yet still holding us to a functional timeline.
Most importantly, we owe our wives and children a special note of thanks because
they put up with long telephone calls and our absences. We must specifically acknow-
ledge Maria, Nancy, Sabina (the "Bean"), and John because they were so patiently
understanding while we created this book.
At the Program in Maritime History, East Carolina University, Frank Cantelas
aided tremendously in retrieving and printing hard copy and solving computer prob-
lems. Karen Underwood tracked down addresses, kept interruptions to a minimum, and
sent out letters.
One and all, you have our thanks.

xiii
 ~

'.~1,; .' Contents

~l
~/:.
.

Introduction

PART I: THE FIELD 5

LA: Underwater Archaeology. . . . . . . . . . .... . .. . . .................. . . 7

I. The Fos Underwater Excavations . . . . . . . . . . .. . . . .. . . . . . .. . . . . .. . . . . . . 9

Rene Baucaire

2. The Technical Importance of Shipwreck Archaeology. . . . . . . . . . . . . . . . .17. . . .

William A. Baker

3. Introducing Maritime Archaeology . . ... . ... . .... . .. . ...... . ... . ... . .. , 23

Keith Muckelroy

4. Integrating Archaeological and Historical Records in

Dutch East India Company Research .. .. . ... . . . .. . .... . . . .. ... .. . . . . 39
J. Bas Kist

5. A Discussion of Maritime Archaeology . . . . . . . . . . . . . .. .... . . . . . 47

.. . .. . .. .
Paul Fontenoy

Additional Bibliography for Part I.A 53

I.B: The Secular Debate. . . . . . . . . . ....... . . . . . . . . . . . . . . . . . ..

. . . .55
...

6. Underwater Archaeology, European versus American . . . . . . ........ ... . 57

...
Joel L. Shiner

xv
xvi Contents

7. The South Carolina Hobby Diver Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 65

Christopher F. Amer and Carl Steen

Additional Bibliography for Part I.B .................................... 71

I.C: Ethics and the Great Debate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. TJ

8. The World's Worst Investment:

The Economics of Treasure Hunting with Real-Life Comparisons . . . . . . . . . .. 75
Peter Throckmorton

9. Why Dr. Bass Couldn't Convince Mr. Gumbel:

The Trouble with Treasure Revisited, Again ........................... 85
Wilburn A. Cockrell

10. Archaeology on Trial ............................................... 97

R. Duncan Mathewson, III

11. American Naval Archaeology: Past and Prologue ......................... 105

William S. Dudley

Additional Bibliography for Part I.C .................................... III

PART II: AREAL STUDIES 113

II.A: The Mediterranean: Bibliography 115

II.B: Northern Europe: Bibliography 117

II.C: The Caribbean: Bibliography ................................... 119

11.0: The Pacific Rim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 121

12. Current Status of Underwater Archaeology in Malaysia .................... 123

Adi Haji Taha

13. Philippine Underwater Archaeology:

Present Research Projects and New Developments ...................... 127
Wilfredo P. Ronquillo

Additional Bibliography for Part H.D ................................... 135

II.E: The Arabic World: Bibliography 137

II.F: Africa: Bibliography ........................................... 139

Contents xvii

PART III: RESEARCH DESIGN 141

1I1.A: Hypothesis Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 143

14. The Method of Multiple Working Hypotheses:

With This Method the Dangers of Parental Affection for
a Favorite Theory Can Be Circumvented . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 145
T. C. Chamberlin

15. Strong Inference: Certain Systematic Methods of Scientific Thinking May

Produce Much More Rapid Progress Than Others . . . . . . . . . . . . . . . . . . . . .. 155
John R. Platt

16. The Limitations of Inference in Archaeology . . . . . . . . . . . . . . . . . . . . . . . . . . .. 167

M.A. Smith

17. Archaeological Methods ........................................... 175

Fred T. Plog

18. The Berth 52 Vessel (09CH69I): The Interim Report ..................... 187
L.E. Babits

III.B: Theory ................................................... 203

19. Middle-Range Theory in Archaeology:

A Critical Review of Origins and Applications ......................... 205
L. Mark Raab and Albert C. Goodyear

20. Technology, Theory, and Analysis: Using Remote Sensing as a Tool for
Middle-Range Theory Building in Maritime and Nautical Archaeology . . . . . . . 223
Richard J Anuskiewicz

21. Considerations for Research Designs in Shipwreck Archaeology ............. 233

Daniel J Lenihan and Larry Murphy

Additional Bibliography for Part III ....................................... 241

PART IV: FINDING SITES 243

IV.A: Site Location Factors ..................•...................... 245

22. Sea Level Change as a Variable in Colonial American Archaeology 247

Reynold J Ruppe

23. Site Location Factors .............................................. 253

James F. Muche

Additional Bibliography for Part IV.A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 257

XVIII Contents

N.B: Site Formation Processes ....................................... 259

24. Physical Processes at the CSS Chattahoochee Wreck Site .................. 261
Richard Stephenson

25. The Archaeology of Shipwrecks ...................................... 267

Keith Muckelroy

26. Cultural Resources Magnetometer Survey and Testing ..................... 291

Robert L. Gearhart

27. A Diachronic Study of Some Historical and Natural Factors Linked to

Shipwreck Patterns in the Northern Gulf of Mexico ...................... 303
Ervan G. Garrison

N.C: Survey Methodology ......................................... 317

28. Nautical Archaeology Survey Methods ................................. 319

Kimberly Watson

29. A Developmental Model for Survey and Inventory of Submerged Archaeological

Resources in a Riverine System: The Patuxent River, Maryland ............. 323
Donald G. Shomette and Ralph E. Eshelman

Additional Bibliography for Part IV ........................................ 337

PART V: HIGH TECHNOLOGY 339

V.A: History of Going Underwater ................................... 341

30. History of Diving ................................................. 343

u.s. Navy
V.B: Magnetometer ............................................... 355

31. Considerations of Remote Sensing Limitations to Submerged Historical Site Survey .. 357
Larry E. Murphy and Allen R. Saltus

32. An Airborne Magnetometer Survey for Shipwrecks and

Associated Underwater Test Excavations .............................. 363
J. Barto Arnold, III

33. Magnetic Search and Survey in Shallow Water and Beach Areas ............. 375
Jack Hudson, Kay G. Hudson, and Harry W. Rhodes, III

V.C: Sonar ..................................................... 379

34. New Capabilities for Side-Scan Sonar ................................. 381

Martin Klein
Contents xix

35. Isis - Versatile Sonar Data Acquisition ................................ 387

Laura Jean Penvenne and John Penvenne

V.D: Mapping and Recording ....................................... 395

36. Computer Video Image Digitization on the USS Monitor:

A Research Tool for Underwater Archaeology ........... . . . . . . . . . . . . . . 397
Roderick M. Farb

37. Electronic Mapping of Underwater Sites ............................... 403

Peter J.A. Waddell

V.E: Remote Operating Vehicle (ROy) 411

Additional Bibliography for Part V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 413

PART VI: SITE SIGNIFICANCE 415

V1.A: Significance 417

38. An Envelope Full of Questions That Count in Underwater Archaeology . . . . . .. 419

Parker B. Potter, Jr.

Additional Bibliography for Part VI.A 427

VLB: Specific Criteria ............................................. 429

39. Toward Establishing Research and Significance Criteria for

Civil War Shipwreck Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 43 I
Gordon P Watts. Jr.

40. A Small-Craft Topology: Tool for Archaelogical Research 441

Michael B. Alford

Additional Bibliography for Part VI.B .................................. 447

PART VII: DATA RECOVERY 449

VILA: Pre-Excavation Recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 451

41. Preliminary Investigations of a Shipwreck.

Pumpata Cahuita National Park. Costa Rica ........................... 453
Stephen J. Gluckman

V1LB: Sampling .................................................. 469

42. The Analysis of Sea-Bed Distributions: Discontinuous Sites 471

Keith Muckelroy
xx Contents

VII.C: Extreme Data Recovery Conditions .............................. 491

43. Mapping and Site Characterization in Zero Visibility: The C.S.S. Georgia ...... 493
James G. Baker, Richard J. Anuskiewicz, and Ervan G. Garrison

44. Bridging the Experience Gap:

Techniques for Reducing the Stress of Zero Visibility Training ............... 507
Steven H. Sellers

45. ZeroNisibility Diving on the Maple Leaf: The Tricks of the Trade ............ 513
Hans K. Van Tilburg

Additional Bibliography for Part VII ....................................... 519

PART VIII: CONSERVATION 521

46. The Investigation of the Factors That Affect the Preservation

of Underwater Archaeological Sites .................................. 523
Ian Oxley

Bibliography for Part VIII ............................................... 531

PART IX: INTERPRETATION AND EXHIBITION 533

IX.A: Interpretation ............................................... 535

47. 16th-Century Spanish Basque Coopering Technology ...................... 537

Lester A. Ross

IX.B: Exhibition ................................................. 551

48. Shipwreck! - The 1554 Flota Exhibit ................................ 553

J. Barto Arnold, III
A Children's Museum Exhibit on the 1554 Flota Wrecks ................... 558
Becky Alsup

Additional Bibliography for Part IX ........................................ 561

Conclusion ......................................................... 563

Appendix A. Archaeological Organizations and Journals ....................... 565

Appendix B. Supplemental Bibliographical Sources for 17th through 19th Centuries .. 567

About the Contributors ................................................ 571

Index .............................................................. 575

 '.~~.'
~
Introduction
o i' .
q;,O_" .': .
.j'~l

It has been over thirty years since the first calls for a "new archaeology" were made at
the 1963 Society for American Archaeology meetings in Boulder, Colorado. Building
upon the work of James B. Griffin, Albert Spaulding, Julian Steward, W. W. Taylor, and
Leslie White, Lewis Binford and others began trying to explain change (Binford 1972:
2-13). The "new archaeologists" utilized an explicitly evolutionary approach, and they
proposed scientific methodology for examining their data and its interpretation.
At the same time, Historical Archaeology and Maritime Archaeology were in
their beginning phases. Both were initially seen as illegitimate children of "real"
archaeology, which dealt with classic old world sites or new world prehistoric sites.
The work of Ivor Noel Hume at Colonial Williamsburg (1963; 1969; 1970) and J.e.
Harrington at National Park Service fortifications (1957) set standards which still guide
historical archaeology. In underwater archaeology, George Bass demonstrated that ter-
restrial standards could be applied to underwater work (Bass 1967), something already
accomplished by Beaucaire (1964) but not generally recognized (Diole 1954:96-98).
A systematically organized body of knowledge used in a field to predict, analyze,
explain, and then guide interpretations is called theory. In many ways, theory is a col-
lection of the assumptions with which a researcher enters an arena to collect data and
find answers. Thomas Kuhn described this system of beliefs as a paradigm, the "univer-
sally recognized scientific achievements that for a time provide model problems and
solutions to a community. of practitioners" (Kuhn 1970:viii).
Binford articulated the scientific paradigm in his classic paper "Smudge Pits and
Hide Smoking: The Use of Analogy in Archaeological Reasoning" (1967). Binford
used historical sources to propose alternative interpretations about small firepits found
on prehistoric sites. Ten years later, Stanley South published both Method and Theory
in Historical Archaeology (1977) and Research Strategies in Historical Archaeology
2 Introduction

(1977a). Not coincidentally, Binford provided an introductory foreword to the theory

book. South presented a graphic view of theory in his first chapter and then, in a well-
reasoned second chapter, articulated the importance of quantitative analysis to detect
patterning in archaeological deposits. Neither Binford nor South cited either Chamber-
lin (1965) or Platt (1964) on the use of multiple working hypotheses and best fit with
the data but they clearly espoused their views.
Within a year of South's seminal publications, Keith Muckelroy began to define
and formulate a theoretical position for what he called Maritime Archaeology, a holis-
tic view of people, sites, and things related to a lifestyle based on exploiting a water-
oriented environment. As an entity, Muckelroy, Throckmorton, and Bass set basic
standards followed today. Like their colleagues in Historical Archaeology, they cited
neither Chamberlin nor Platt although they certainly followed their guidelines.
At the present time there is no text for teaching a course in the History and The-
ory of Nautical Archaeology. Muckelroy's Maritime Archaeology is out of print. There
are two standard texts on "how to do" underwater archaeology: Jeremy Green's Mari-
time Archaeology (1990), is good for clear water; the Nautical Archaeology Society's
Archaeology Underwater (1990) is basically a guide for "doing" underwater archaeol-
ogy in Britain. Neither discusses theoretical underpinnings of what they advocate.
Instructors in Maritime Archaeology must search through the literature and
assign readings related to the specific classroom presentation of problems under discus-
sion, a time-consuming, costly, and frustrating task, especially for the typically impov-
erished graduate student. Writing a textbook on theory would be time consuming and,
for students, the work could be the quintessentially boring text of all time. The situation
was once true of Historical Archaeology.
Robert L. Schuyler solved that problem by editing Historical Archaeology: A
Guide to Substantive and Theoretical Contributions (1978). We have emulated
Schuyler's work in length and content but without duplicating his selections. We have
drawn from a variety of worldwide sources to avoid tedium and keep the approach as
practical and down to earth as possible.
We recognize every site is unique and calls for different techniques and interpre-
tations, but there must be an integrating whole that places the site in time and space be-
yond its existence as a locus of past human activity. Following the lead of northern
European medievalists, maritime historians, and archaeologists, we advocate a three-
stage approach to looking at maritime sites. First, they are sites and must be exploited
to the fullest informational extent possible; second, sites of a given locality or type
must be interrelated with each other to provide both interpretations and predictive mod-
eling; finally, sites must be presented within interdisciplinary and regional, if not
global, perspectives to allow determinations of importance and provide better under-
standing of each individual site.
A caveat should be mentioned. The readings are organized topically. Chapters
show what various writers were thinking at a specific point in time. Many articles are
dated or out-ol-date. Many readings represent neither current thinking nor techniques;
instead they provide a background for recent work. Outdated and out-ol-print readings
are included to show the history of Maritime Archaeology. This is deliberate and de-
signed to provide students with some sense of time, depth and change in the field. More
recent materials can be found in current literature whereas the older works provide
access to our past via their citations.
Introduction 3

This text is a starting point for thinking beyond the "how to" and into the inter-
pretive "how" and "why." For student and teacher alike, we wish you luck in formulat-
ing your research designs, analyzing your sites and samples, and interpreting "what it
all means."

REFERENCES
Bass, George F., 1967, Cape Gelidonya: A Bronze Age Shipwreck. Transactions o/the American Philosophical
Society Volume 57 (8), Philadelphia.
Binfor?, Lewis H., 1967, Smudge Pits and Hide Smoking: The Use of Analogy in Archaeological Reasoning.
American Antiquity 32 (I): 1-12.
Binford, Lewis H., 1972, An Archaeological Perspective. Seminar Press, New York.
Binford, Lewis H. and Sally R. Binford (eds), 1968, New Perspectives in Archaeology. Aldine, Chicago.
Chamberlin, T.e., 1965, The Method of Multiple Working Hypotheses. Science 148:754-759.
Dean, Martin, et al (eds)., 1990, Archaeology Underwater: the NAS Guide to Principles and Practice. Nautical
Archaeology Society, London.
Diole, Philippe, 1954, 4.000 Years under the Sea. Julian Messner, New York.
Greene, Jeremy, 1990, Maritime Archaeology. Academic Press, London.
Harrington, J.e., 1957, New Light on Washington s Fort Necessity. National Park Service, Government Printing
Office, Washington, D.e.
Kuhn, Thomas, 1970, The Structure o/Scientific Revolutions. University of Chicago Press, Chicago, Illinois.
Muckelroy, Keith, 1978, Maritime Archaeology. Cambridge University Press, New York.
Noel Hume, Ivor, 1963, Here Lies Virginia. Alfred A. Knopf, New York.
Noel Hume, Ivor, 1969, Historical Archaeology. Alfred A. Knopf, New York.
Noel Hume, Ivor, 1969, Guide to Artifacts o/Colonial America. Alfred A. Knopf, New York.
Platt, John R., 1964, Strong Inference. Science 146:347-353.
Schuyler, Robert L. (ed.), 1978, Historical Archaeology: A Guide to Substantive and Theoretical Contributions.
Baywood Publishing Company Inc., New York.
South, Stanley, 1977, Method and Theory in Historical Archaeology. Academic Press, New York.
South, Stanley, I 977a, Research Strategies in Historical Archaeology. Academic Press, New York.
 Part I

The Field

Maritime archaeology draws from a variety of major source fields, including terrestrial
archaeology, anthropology, and history. The interdisciplinary theoretical grounding has
meant that some investigators and places were far more sophisticated in their treatment
of maritime sites than others because of their more holistic approach. As the need for
obtaining more information from a site grew and research questions became more com-
plex, archaeologists continued to draw inspiration for their sites from any fields that
seemed relevant.
Drawing from diverse fields created problems because the theoretical basis
underlying specialized research was often not borrowed and, in many cases, was poorly
understood. The borrowed ideas became inspirational models to guide underwater and
maritime research without critical assessment of their applicability. This was not neces-
sarily bad, as long as researchers understood and stated that their borrowed guidelines
were just that, something to help organize data so that it could be better examined. This
borrowing was particularly true in terms of investigating ship construction, site forma-
tion processes, and artifact analysis.
As the field and the data base grew larger, more attention has been paid to inter-
pretation and to understanding how sites fit into a larger universe. This shift reflects a
change from shipwreck reporting to maritime interpretation. As this occurred, models
derived from interdisciplinary studies on land and sea began to make their appearance.

5
 Part I.A

Underwater Archaeology

In this section, the first (1946) archaeological work on an underwater site is reported by
Beaucaire. The first reading is followed by Baker's discussion of what must be re-
corded and why, especially in the light of missing documentary material about vessels.
The multidisciplinary approach is spelled out in detail by J. Bas Kist, who also articu-
lates a theoretical grounding for research and interpretation of the Amsterdam. Keith
Muckelroy's Maritime Archaeology is the source for the next reading, which uses sites
as case studies while presenting an overview of the field and its theoretical background.
Finally, Paul Fontenoy discusses the differences in approach between researchers in
northwestern Europe and those in the United States.
Beaucaire describes the first time terrestrial archaeologists moved their activities
into the water as part of a regional study of the area around Fos-sur-Mer in Provence.
Their work is important not simply because it was done in 1946, but because it was not
concerned with a wrecked ship. Instead, they concentrated on a drowned Roman villa
in heavily silted water at Saint-Gervais. Their work is little known but provides insights
into how archaeology went underwater some time before better known scuba divers
began made headlines exploring sunken ships in clear water. The conditions under
which they worked are also typical of more recent, and less spectacular, investigations
in poor visibility.
Baker offers a good introduction to what is, and what is not, important to people
who study ship remains. In his view, anything previous to A.D. 1650 gains additional
importance due to the lack of documentary evidence.
An excellent overview and introduction to the field by one of the more theoreti-
cally minded early practitioners, Muckelroy gives examples of major sites and accom-
plishments of the new science, but never strays far from a theoretical grounding.
Kist focuses on the importance of archaeological evidence as one tool for the
multidisciplinary group of historians, archivists, archaeologists, and museum people to

7
8 I.A. Underwater Archaeology

use in the studying the Dutch East India Company. Of 50 wrecks inspected since 1965,
the Amsterdam is chosen as the main example. From the start of the excavation in
1984, historical hypotheses served as a guide to the research. Due to the extensive
physical remains and the abundance of archival material, the combined historical-
archaeological approach proves especially productive.
Fontenoy prepared his work on how Europeans approach medieval ships and
shipwrecks for a paper in a course on Method and Theory in Nautical Archaeology. His
basic premise is that an interdisciplinary approach provides a great deal of regional
information despite very little archaeological data because the material culture is seen
in a broader perspective drawn from problem oriented research.
 Chapter 1

The Fos Underwater

Excavations

RENE BAUCAIRE

Before presenting the report, quite incomplete, on the underwater excavations at Fos
and its gulf, I see it as an obligation to sincerely thank Mr. Fernand Benoit, member of
the Institut and director of the XII Archaeological Circumscription, for the honor of
designating me director of the underwater excavations in our region. I also wish to
express my thanks for his moral and financial help as well as his numerous tokens of
esteem and interest, which he has always shown me. I beg him to find an expression of
my sincere regard here.
I shall also be careful not to forget the excavation teams and the participants who
deserve congratulations for the magnificent work they accomplished at Fos (Messrs. V.
Coni I, Michon, Gouverneur, Terras, Paniagua, Sicardi, Puggioni, L. Arnaud, H.
Clement, Cerdan, Reynaud, Aquaron, Illaret, Agnel, Diot, Delange, Michaud, Thomas,
Flevet, Aillaud, Vial, Chauliac, Laforest, Lanscotte, J.-Cl. Meo, Bousset, Malavielle,
Pugniere, Gimet, la Ste H.A.M. , l'Inscription Maritime, les Ponts et Chaussees, l'Ad-
ministration des Douanes, and, more recently, the important Club Culture et Sportif
from-de-Bouc and its 40 members).
In addition, because of the real difficulties and the. very particular characteristics
that they represent, I wish to mention that these excavations were made possible only
through the constancy of this teamwork, the efforts of my faithful collaborators, to-
gether with their scientific probity and their disinterestedness.

Source: Bulletin de fa Societe des Amis du Vieillstres. Tome Quat9rzieme 1964, La Pen see Universitaire, Aix-
En-Provence. Translated by Thomas A. Babits. By permission of Societe des Amis du Vieillstres.

9
 10 R. Baucaire

Let me say that we are proud of undertaking a veritable rescue operation of the
antique site of Fos, of patiently collecting the participants' accounts and of assembling
them at the Regional Archaeological Center, the Musee du Vieil-Istres, where the pub-
lic can examine and study them and gain knowledge at long length.
The Fos excavations started in early 1946, following the removal of mines from
the beaches and before, like today, their invasion by summer residents. The excavations
were, stricto sensu, the first underwater excavations in France.
Since the results obtained went much further than the most optimistic of fore-
casts, the teams quickly increased, in particular, when the laws about protection of un-
derwater archaeological sites were applied. In our opinion, they are still, nevertheless,
insufficient to disarm the clandestine divers. Let us be frank, though, and admit that,
for more than a year, a very fruitful effort has been undertaken to counter archaeol-
ogy's wound, pilfering.
What dreadful statements have been said and even written about the underwater
site in the Fos gulf, without even thinking an instant that a sunken town, under 8-10 m
of water, might appear unusual. If nothing else, just from a purely geographical point of
view, the legend of the Vieux-Fos, a simplified myth of a Provencal Ys, enviously ped-
dled, especially by those who have never dove, appears to have a very hard life to me.
lt is sufficient to situate this antique village on the Saint-Gervais rock, adjacent to its
extra-muros cemetery, of which we at least know the epitaph, to credit the Expentiana,
on display at the Gap museum. The dwellings descended in tiers all the way to the
shore of the small gulf of Saint-Gervais, where a rudimentary pier and a dock were dis-
covered next to fallen columns and huge squared blocks, all of which are under an
average of 2 m of water. Professor Hafemann of the University of Mayence studied the
geological phenomena of Fos. lt is similar to many others in the Mediterranean basin.
He confirms the consistent sea level since protohistory, but allows for a slow sinking of
the Fos gulf region shortly after the Christian era. What are the reasons that he gives to
explain this local phenomenon? Perhaps a telluric shock, but probably the incredible
weight of the Rhone alluvium weighing on a diaclase is a probable cause. In any case,
and the proof exists, the sinking, whether slow or rapid, has hardly gone over 2 m. lt is
precisely at this depth that the "sunken Fos" can be found. As we can see, there is a
great distance separating legend and reality.
Objects found at this depth are protected by a thick layer of light colored clay,
very sticky and compact, certainly of alluvium origin, from which they have been
extracted in a surprising state of conservation. The Fos stratigraphy is well known. We
can find this layer of archaeological clay at a maximum depth of 2 m and with layer of
sand and gravel about 40 centimeters thick. The village stretched out over 100 minto
the small gulf. All of this area was, or is, in the process of being methodically exca-
vated by shovel and undermining, with or without the help of a diving suit, by teams
very specialized in this particularly tiring work. I Nevertheless, I make a very precise
distinction between underwater excavations that I have just described, real archaeologi-
cal excavations transposed to an underwater environment, and the search for sunken
wrecks in the gulf. The latter is certainly equally as difficult, although, it is, in reality,
only prospecting and recovery.

I Team led by Maitre Ch. Michon, Gouverneur, Meo, Gimet.

The Fos Underwater Excavations 11

The Peutinger table, the first geographical map that we know of, designates Ostie
and Fos ports by the same symbol: a sort of semicircular monument. Alas, we are very
far from seeing of this resemblance in actuality. In truth, nothing is comparable be-
tween the two ports. We have never, and this is truly unfortunate, discovered the least
indication at Fos of monumental construction or storehouses. As for streets or sunken
buildings, no one has ever seen them offshore, even on a favorable day! We are not
afraid of saying that they are a figment of the imagination, a fantasy or fiction.
Now, nevertheless, even if we are so poor on this aspect of a port, we are much
richer and well informed on the daily life of a Gallo-Roman Fossa Maritima, simple
fisherman, sailor, merchant, or small artisan.
If we can judge by the abundance, or the rarity of such and such a pottery type,
we see at Fos, after a long indigenous and episodic period, there is an occupation of av-
erage importance at the time of Marius and the digging of the Fosses mariennes canal,
characterized by the Campanien fragments of various types. These are followed by
Italic and Arezzo pottery which show an active trade from the time of Caesar through
Tiberius. Because of the numerous Graufesenque vases we can attest to an important
occupation until the end of the second century. From there on artifacts become more
rare. The Constantinien fragments and those of light sigillate earth are not very abun-
dant, and Christian fragments are hardly represented. After the fourth century, the exca-
vations no longer render artifacts and we are obliged to admit that, after this date, Fos
was put to sleep.
Thus, this transit port is clearly very prosperous by the end of the Republic and
the beginning of the Empire, thanks to its location with respect to the RhOne river delta
and the Fosses mariennes canal. It was designated on the Peutinger Table by the same
symbol as Os tie and probably suffered the same end as the Marienne canal.
The latter, already difficult to navigate at the time of Augustus, became almost
completely impracticable before the Bas-Empire and the "Itint!raire Maritime" (or
d'Antonin), dating from the third and fourth centuries, only shows Fos as a simple stop-
over between Port-de-Bouc and the Grand RhOne.
The dwellings at Fos appear, in our present state of knowledge, to have been very
simple, perhaps even primitive. The ground floor is hard earth in spite of the presence
of a few rare mosiac and marble coverings indicating a more elaborate dwelling. The
walls are uncovered, of low height and have traces of monochrome mortar. The beams
and rafters are roughly squared and support a roof covered with tegulae and imbrices
decorated with a few rare footings. Nothing allows us to imagine classic domus (of the
Pompeian style), or a vast house, even an insula or house of several stories, so very
common in Rome and Ostie. Confronting these poor remains, we think of makeshift
constructions using light material, even very simple "huts."
By excavating these structures, we find genuine kitchen middens whose methodi-
cal exploitation is of the greatest interest. Pots are in abundance with their umbilicus
lids and their Campanien type fish platters. There are circular frying pans with striped
bases, still blackened by successive cooking fires, basins, bowls, cups, various plates,
platters with and without feet, pitchers, jugs of various types, ceramic and marble mor-
tars, door balances belonging to weavers and, lastly, a great quantity of amphoras.
Among all of this great variety of kitchenware can be found a few perforated cylindri-
cal vases giving the impression of a crude strainer or cheese molds, a feeding bottle, or
 12 R. Baucaire

a funnel. There is also a branch of a long pincer in bronze, marked with a dotted name
of Silvinus and many other articles whose significance is still unknown to us.
Vases from the Ruthenes workshop (/a Graufesenque, Lezoux, Banassac, etc.) are
very abundant and identifiable by both shape and decoration. Numerous fragments are
signed (potter's marks) either on the bottom of the vase or within the decoration. For
Fos, they make up an almost inexhaustible materiel and each excavation session brings
up new ones. The description and identification of the potter's marks shall soon be the
result of the work of many.
A very beautiful medallion can be mentioned (only 533 known in the world) of a
applied relief vase from "ceramiste a la rosace" representing an Anadynomene Venus
accompanied with her attributes: Cupid, shell, and dove (third century A.D.) and a few
particularly well-preserved fragments of green ceramic or yellow lead glazed vases,
decorated with leaves of myrtle, wreaths, and pine cones. They imitate metallic vases
from Asia Minor (first century). A great variety of vases "with subtle sides" are still
found and have a decoration that is smooth, combed, incised, and sanded with a crust
of clay. Others have the shape of a pine cone (first century).
In the corner of a very simple dwelling, we found a small altar of carved stone, a
family sanctuary still in place. Unfortunately it is without any inscription. It is sur-
rounded by lamps, one of which is from the Campanien period, and the second is deco-
rated with a flame and signed Mento.Dioni. A third lamp is very curious and is shaped
like a boat having several lights and decorated with a very realistic scene of a fisher-
man and a siren bird. Other lamps have winged creatures, a monkey, a scorpion, flow-
ers, a Cupid, Pegasus, various divinities, and a very fine reproduction of the myth of
Actaeon devoured by dogs. There also are many lamps without decoration, a few mod-
els of the "Po valley" type (first century), and a few rare lamps of the Christian type,
however, without any Christian figuration.
Excavations over the past 2 years have revealed a wood lathe-worker's shop. The
methodical examination of this area gave us astonishing surprises. All of the artifacts
are in a remarkable state of conservation. Thanks to the efforts of Mr. Bouis, head of
the Department's [Culture] laboratory at the Borely museum in Marseilles, they have
been well preserved and stabilized, thus minimizing the terrible effects of desiccation.
The following objects have been recovered: part of rigging (e.g., pulleys, reels), house-
hold utensils, various shaped boxes with covers, knife and other instrument handles,
fragments of scrubbing brushes, part of a cage (probably a birdcage), a large quantity
of combs and hairpins, a cup with game tokens, a bridge from an eight-cord lyre and its
plectrum, a fragment of a wicker basket, scraps of leather, a leg from a piece of furni-
ture (bed or chair), an enigmatic article with holes and a sliding part (an identical arti-
cle was found in a Roman sepulcher near Tarascon) and, lastly, a large tonne, a utensil
of Gaulish invention, which, unfortunately, could not be uncovered quickly enough and
was dismembered by pillagers within a very short time.
Several objects in bone or ivory illustrate a variety of uses, sewing needles, a
handle for a mirror and stylet, combs, tokens, playing dice, hinges for jewel boxes, a
small slab (tessere), and a great quantity of spatulas and spoons for rough, a stock of
balls containing a blue coloring (rough? cf. National Museum of Naples), statuettes in
baked earth [rooster head, female torso (a divinity with spear and shield: Minerva)].
The last article is a very beautiful artistic find, the famous head of Aphrodite on ivory
(an applique to a chair or chest), a Hellenistic work of remarkable sureness and fine-
The Fos Underwater Excavations 13

ness in its workmanship which attests to its execution by an experienced artist of the
first century B.C.
Examination of the "kitchen waste" reveals that the most frequently consumed
animal is the pig or the wild boar (which agrees with Caesar's De Bello Gallieo). We
then find goats, sheep, beef, horse, and tuna. We have also found pits from peaches and
olives, hazelnut and almond shells as well as an egg.
Some animal horns, more or less sculptured, used for prophylactic or apotropaic pur-
poses, were also discovered in the excavations. Whether they were attached to the bow or
the stem of a barque, as gunwale or aplist heads, or nailed to door lintels, or to a roof's
edge is unknown. Both explanations appear valid. In fact, similar horns coming from
wrecks have been found at Hereulanum (Casa dei cervi), at Monaco, and at Albenga. Even
today, in the region of Naples, in Calabria, and in Sicily, they decorate doors to some
houses to keep evil spirits away. We have three of these horns (steer, ram, and stag), used
for this purpose, which have fastening holes as well as traces of decoration.
As for glass fragments, their abundance and variety are surprising. There is
everything including blocks of glass or raw glass pulp (was there a glassworks at Fos?),
ribbed vases (vials) in raw multicolored glass (miriajiore), a blown relief (geometric
motifs, gladiator combats), decorated and engraved, various bottles, balsam aires, gut-
tule glass, a funnel and a wishing glass having the Greek inscription "Bum and rejoice
yourself." A big, Constantinien type, vase, almost intact, with decoration carved by a
small engraver which can be compared, even though it is in baked earth, to the wishing
glass mentioned above. It has the following engraved big capital letters around the
edge: ALEXANDER AVE. Last of all, there is a very nice cameo representing the
goodness Rome (third century), surrounded by gold debris (team of l-Cl. Meo and
Gimet) that was found in a small jug.
The amphorae alone merit a special chapter. They are what strikes the public imagi-
nation the most. They are much sought after by amateurs and they have been promoted to
garden decoration, thus a lucrative traffic against which the government (at least today's)
has started to deal severely with. Except for small flat-bottomed amphorae, jugs or single
handled "olpes," big amphorae have never been discovered during our excavations of
houses. They have been found, however, in the gulf. The tendency is to admit that these
recipients, or the major part at least, are either cast up by the sea or from accidents during
transshipment. The types are very varied with a few rare shapes (Punic and Augst types).
Most are from the first century. The ones from the Bas-Empire are much less numerous.
However, to be precise, a greater portion seems to appear from the time of the port's maxi-
mum activity, that is, from 50 B.C. to 150 A.D. A great teamwork on the Fos amphorae is in
progress and we will, at last, give precise statements, not only on their various origins, but
also on the commercial itineraries and the contents of these utensils.
Hence, one of our amphorae (from Foumon's team's wreck) contained a great
quantity of fish bones mixed with herbs (probably aromatic). The state of conservation
of some fragments collected from our excavations of dwellings has afforded us a re-
markable series of necks and bellies with inscriptions made with a paintbrush and black
paint, in Greek or Latin, capital and cursive characters. Equivalents to these Fos frag-
ments can only be found among those of the Testaeci [ceramic] middens in Rome. As
of today, several Pica tum Vetus (old resin, cf. resin used by Greeks today), Garum,
Cortixariu (bark wine), Phoebum (wine from Rhodes), etc., have been deciphered. Nu-
merous other inscriptions are still being studied since (as Mr. F. Benoit thinks) some of
 14 R. Baucaire

these amphorae have palimpsest inscriptions. This could demonstrate their use for sev-
eral trips with different contents.
The gulf still yields many more finds, basalt millstones (catilus et meta), pitchers,
jugs, lead anchor stocks, mooring stones, a fragment of a cancel (fourth century) and
Christian lamps, roofing tiles marked APRI and HER(ennius) 0 (ptatus). One very dis-
persed wreck, having pegged and copper nailed framework, had a load of the big-sized,
Republican-type amphorae, stamped MAHE(s) or MAHE(tis). There were also common
unvarnished Companien-type potteries, engraved with MS (including Fournon, Illaret,
and Agnel). Near this wreck, a lead anchor stock weighing 84 kg and marked with
C(aius) AC (ilius) ASICI(us) and the number XVI! was brought up. A lead salmon
weighing 66 kg, probably of British origin, stamped within a frame with the mark SO-
CIORUM PLUMB (ariorum) GER (mani) was also brought up. To our knowledge, this
is the heaviest antique ingot known today.
Finally, in August 1959, a team from the Orange air base, under the leadership of
Chief-Sergeant Ardouin-Dumazet and Sergeant Vauthier, came to Fos to undertake me-
thodical research. The seriousness, willingness, and technology of these divers were re-
warded by the discovery of a decorative bow or gunwale figure representing a wild
boar's head. It is in bronze, weighs l3 kg, and was found so very close to a wreck that
it was probably a part of. It was deeply buried in the sand and rocks at a depth of 12 m.
The head from the Hellenistic period (first century B.C.) can be compared to the beam
heads (lion and wolf) of Caligula's galley recovered from Lake Nemi. However, their
aspect is quite different and their destination even more so. This absolutely unique ob-
ject, displayed at the Louvre's International Exposition in 1963, entitled "Art in Roman
Occident," is, without question, one of the very best pieces of antique sculpture that the
Musee du Vieil-Istres, a regional archaeological center, is proud to possess.
The dredging of the narrows of Port-de-Bouc, to a depth of 8 feet, brought up a
Corinthian cornice and a small sarcophagus made of marble from Carrera (about 50
B.C.). It probably comes from a wreck (framework, lead anchor stock) and is of the Hel-
lenistic type, showing a very ornate cresset, sided with panther's paws in high relief.
About the small monetary treasure of Fos, over a very restricted area on the archae-
ological level, a great quantity of bronze coins (large and small), a few silver pieces
(Vespasian, 69-79, and Trajan, 97-117), and a gold coin from Philippe (244-249) have
been uncovered. The latter, accidentally discovered before our excavations started, was
deposited at the museum of Cavaillon. These coins, numbering 167, are, for the greater
part, difficult to identify because of a very poor conservation. Most of them are from the
later periods. For example, the silver "saucee" [a coin of common metal covered with a
layer of precious metal] coins from Julia Domma (217), Galienus (253-263), Gordien
(238-244) and Maximien (285-305) are easy to date. Unfortunately, shameful pillaging
has raged and before we could intervene several hundreds pieces (it is said that there
were enough to fill a bucket) disappeared in the hands of these strangers.
Thus, Fos does not escape the troubles so frequently noticed in Gaul toward the
end of the third century, the time of the Barbaric invasions and periods of military anar-
chy, as burying of monetary treasure at sites on the coast indicates. 2

2 Cf. A. Blanchet, Les tresors de monnaies romaines et les invasions germaniques en Gaule (1900), p. 52, and
Dechelette, Archeologie gallo-romaine, Vol. I, p. 101, and Vol. 6, p. 935.
The Fos Underwater Excavations 15

This monetary treasure, along with the traces of a fire still visible on the ceiling
beams, rafters, certain pottery, and frequently encountered melted lead debris, allow us
to believe that the village ofFos, after the fourth century, was completely ruined and its
inhabitants dispersed. Moreover, it is certain that, just as Mr. Henri Rolland, director of
the Antiquites de Haute Provence, mentioned, the inhabitants grouped themselves later
on neighboring heights, in particular Castel-Veyre (Sainte-Blaise), where another page
of their history opened.
Thanks to the preservation efforts of our diggers, their dynamism, and their team
spirit, I have the conviction that other discoveries will come, adding to our knowledge,
and shedding greater light on the simple and laborious life of the small population of
Fossa Maritima. As for the exceptional collections enclosed in our museum, they con-
stitute a valuable illustration of this work.
Let all those who have given us some of their time and work or their support, no
matter what the reason, be thanked as in the pure tradition. But, I can assure them that
the discovery of a far-off past, that we thought buried or forgotten forever, shall, for all
of them, as for myself, be the most flattering compensation.
 Chapter 2

The Technical Importance

of Shipwreck Archaeology

WILLIAM A. BAKER

"Is this vessel important?" is a question often received in connection with underwater
ship finds. An underwater ship find may be of importance to a number of persons
depending on their particular interests. A technical historian might want to study the
vessel's form and construction; an archaeologist will seek to retrieve artifacts pertain-
ing to the cargo, passengers, if any, and crew; a salvager would look for items that
could easily be disposed of to collectors and scrap dealers; and an oil company might
simply consider the find an obstruction to a proposed pipeline. The current questions
concerning the importance of underwater ship finds may stem from the recent studies
of the cultural resources of the continental shelf.
Underwater ship finds are popularly called shipwrecks but the term wreck implies
destruction. A real shipwreck, a vessel driven on shore by a storm, more often than not
is broken up and its contents scattered. A study of the remains of such a vessel and its
contents is rarely fruitful; items retrieved may be of interest only as curiosities from the
sea. There are, of course, exceptions to this and many valuable objects have been
brought up from the seabottom but their context is lost forever.
Vessels have been and occasionally still are overwhelmed by storms at sea and
founder relatively intact; the same condition applies to victims of collisions. The new
schooner John F. Leavitt, battered by a 3-day North Atlantic gale on her maiden pas-
sage south with cargo, went down about 150 miles southeast of Cape Cod on December
27, 1979. After foundering, much then depended on what happened on the seabottom.
Source: Underwater Archaeology; The Proceedings of the Eleventh Conference on Underwater Archaeology,
edited by Calvin R. Cummings (1982), pp. 141-144. Fathom Eight. San Marino, California. Reprinted by per-
mission.

17
 18 W.A. Baker

In the case of a wooden vessel, it was a race between gribbles, shipworms, and other
destructive forces, and a protective covering of mud and sand. Iron and steel vessels
have long underwater lives but they, too, ultimately will disintegrate. It is such foun-
dered vessels, however, that offer the best opportunities to archaeologists and technical
historians.
In the majority of cases, what remains on the seabottom after a century or more
of exposure is a collection of the heavier objects that were in the foundered vessel-
anchors, chains, guns, coins, ballast, and the like - better examples of which may
already be displayed in museums. Perhaps buried in the mud or sand, there may also be
the bottom structure of the vessel looking like the backbone and ribs of a fish lying on
a beach.
Identification by name of a chance ship find in this condition is nearly impossi-
ble. The find then assumes an importance in the categories outlined earlier for what it
is, not because of name and associations with events which may only have produced a
certain notoriety.
It also approaches the impossible even to determine the type of vessel, for in spite
of what has been printed in nautical dictionaries, nautical terminology has never been
static. Sailing vessels whose type names were based on rig are impossible to identify
from bottom structure as are those whose names were derived from use. An example of
the former might be a ketch, and of the latter, a pinnace.
Clues concerning a vessel's origin at least may sometimes be obtained from cast-
ing marks on guns; these can provide information as to date and place of manufacture.
Guns, however, often were transferred from ship to ship, particularly in the cases of
armed merchant vessels, and sometimes from country to country. Anchors are even
more difficult to identify but the shanks of Spanish anchors were said to have been
longer and of smaller diameter than for comparable weight anchors from other nations.
The nationality of an anchor, however, is not always that of the vessel, for anchors
could be salvaged from wrecks or the bottom of a harbor and sold to any ship that
needed them.
In spite of the lack of identification - there may be enough small artifacts to
allow the guessing of a date and nationality - some ship finds are well enough pre-
served that a technical historian may be able to demonstrate the differences between
how a ship should have been built based on the evidence of models, scantling tables,
and textbooks, and how it actually was put together. Every vessel is likely to show mi-
nor peculiarities that are its builder's practices. On the other hand, some peculiarities
may be chance indications of prudence, for in the days when timber and plank were
hewn and sawn by hand a piece of wood was not discarded simply because it was not a
mirror duplicate of the one used on the other side of the vessel. The concept of symme-
try in shipbuilding apparently came relatively late in some area.
It may be that the lowest level of importance in underwater work - and one
wonders just what may be the ultimate solution - involves the planned or attempted
recovery of material from vessels stripped and deliberately scuttled. Dry archaeologists
find things of interest in abandoned land sites; the same may prove true underwater.
Except as obstructions and scrap value, a large percentage of the relatively mod-
ern unidentified ship finds are of no real importance to anyone. When in service the
vessels were ordinary carriers transporting prosaic cargoes; their only importance lies
in what may be obtained for objects brought to the surface and sold to collectors. In
The T echnicallmportance of Shipwreck Archaeology 19

these days of energy problems, however, a cargo of coal might be an attractive discov-
ery and there may be retrievable oil in the tanks of sunken oil carriers.
The point to emphasize here is the importance of the shape, structure, and fittings
of a ship versus its contents which pertain to its use, how the crew and passengers
lived, and what cargoes were carried. We have relatively few examples of ships of the
past, but there are museums full of the common artifacts of earlier eras. There is no
point in retrieving and conserving items from the seabottom when equal or better
examples already exist.
The technically important ship finds in general are those that can be dated before
say 1650, about the time that printed books on ships and shipbuilding began appearing
in greater numbers. Recent years have seen a number of such finds, but detailed reports
have not always followed the preliminary accounts; most of these finds are well known
to underwater archaeologists.
The so-called Kyrenia ship of the late fourth century B.C. found off the north
coast of Cyprus is important because of its shape and construction details. In it we have
a type of construction that even in its day was old - a shell of smooth edge-fastened
planking to which framing was added. The edge fastenings were the familiar mortise
and tenon. Apparently an ordinary cargo carrier, it is likely that the Kyrenia ship was
one of a kind, a vessel of a given size built to suit a merchant's requirements.
The Punic vessel now being assembled at Marsala, Sicily, considered to be a war-
ship sunk in a battle in 241 B.C. has details similar to the Kyrenia ship; it is considered
to have been relatively new when sunk. Its importance, however, may be more in the
line of implication than in actual features.
It is known that in 261 B.C. the Romans produced 100 quinqueremes and 20
lighter triremes in 60 days "from the tree." The vessel at Marsala shows from markings
along the keel and on the planking that such feats of construction had to be accom-
plished by standardization. We can only conjecture at this time how such stand-
ardization was achieved - by models, by plans of some sort, or by taking an existing
vessel apart and using each piece as a pattern. The latter apparently was the Roman
procedure for the 100 quinqueremes which were copies of a captured Punic warship.
Patterns have been used in modern times in small boat construction which, for exam-
ple, enabled four men to complete a 28-foot whaleboat in 28 hours.
The Serce Liman (Sparrow Harbor) vessel found off the coast of Turkey and
excavated primarily because of the possibility of its being an evolutionary step in ship
construction has been dated about the II th century A.D. It is - was might be better, for
relatively little of the hull remains - a small flat-bottomed double ender about 52 feet
long with a breadth of between 17 and 20 feet. At present it is the earliest known vessel
in which the planking was fastened to a preerected skeleton frame. A first century B.C.
Roman vessel now being excavated off the southern coast of France seems to be an
intermediate stage between shell and skeletal construction.
The five Viking-age vessels circa 1000 A.D. excavated from Denmark's Roskilde
Fjord in the early 1960s carried a bit further in time the northern European type of shell
building first seen in the Nydam find of about 300 A.D. and found fully developed in
the ninth-century vessels from Oseberg and Gokstad in Norway. This type of building
used the lapped edge-fastened planking commonly called clinker. One of the five ves-
sels was the first knorr or cargo vessel to be found; all earlier Viking-age ships had
been warships or what might be called "yacht" versions of them, vessels not suitable
20 W.A. Baker

for a voyage to Iceland. Some surprisingly advanced engineering details were found in
these vessels - what might be called webbed angle clips carved from solid chunks of
wood, and floor timbers that were thin where they were deep across the keel but which
gradually thickened as they became shallower over the planking.
For years maritime researchers argued about the features of two strange-looking
types of vessels portrayed on the seals of various medieval towns, in manuscripts, and
in church paintings, the cog and the hulk. During dredging operations in 1962 in the
harbor of Bremen, West Germany, there was the important find of a cog which ended
most of the speculations about that type. This cog appears to have been sunk before
completion, and it is now believed that it was swept by a flood, tide, or otherwise, from
the building yard into the Weser river. Having been sunk in a river and probably fairly
quickly covered by silt, the Bremen cog, now dated about 1380 A.D., is very well pre-
served. With a length of about 77 feet and a breadth of a bit over 24 feet, it is estimated
to have been able to carry about 125 tons of cargo.
The Bremen cog's construction is unlike any previous type and its ancestry is still
questionable. Its bottom, nearly flat amidships, is formed of three flush-seamed non-
edge-fastened planks on each side of a shallow keel, the type of planking usually called
"caravel" because, as the Portuguese put it, caravels were usually built that way. The
remainder of the cog's planking is standard clinker. Several heavy tie beams that sup-
port the cog's single deck protrude through the side planking; these beam ends are the
characteristic lumps along the sides of the various portrayals of the type. The Bremen
cog has provided an explanation for a 1943 find in Denmark's Kolding Fjord, others in
the new polders created from the Zuider Zee, and a vessel found in 1976 near Elsinore
north of Copenhagen.
The Mary Rose, sunk off Portsmouth, England, in the 16th century, will, when
and if raised, fill a considerable gap in maritime knowledge. It accidentally capsized in
the Solent on July 19, 1545, while going into action against the French. Immediate
attempts to right the Mary Rose and lift it by stages into shallow water failed but vari-
ous attempts to recover some of its armament continued until 1549. Its masts broke off,
it settled into the mud, and was forgotten.
Built in 1509-1510 as a 500- to 600- ton carrack for the fleet of Henry VIII, the
Mary Rose was rebuilt as a 700 tonner in 1536. The method of calculating these ton-
nages is not known. This carrack is important because it was the first English warship
to carry complete batteries of siege artillery as main armament on complete gun decks.
Rediscovered in the 1830s by the first helmeted divers, artifacts were retrieved from
the wreck in 1836 and 1840. Forgotten again until the 1960s, a concerted effort relo-
cated the wreck in 1965, and since then a considerable amount of survey and excava-
tion work has been accomplished.
By contrast, near the Mary Rose are the remains of two vessels that are relatively
uninteresting as ships, the Royal George which capsized at anchor in 1782 and the
Boyne which sank in 1795; both were line-of-battle ships from a reasonably well-
known period. Nelson's Victory that is preserved at Portsmouth was built in 1765.
Interest in old ships has changed considerably since the 1920s when one, said to
have been as well preserved as the Swedish royal ship Wasa, was blown up in Stock-
holm harbor to provide wood for the manufacture of furniture. No one can deny the
importance of the Wasa which capsized on its firSt trial under sail in 1628; it is impor-
tant from many points of view, but primarily because it is an almost complete hull of
The T echnicallmportance of Shipwreck Archaeology 21

1628. It is far from the situation where the probable appearance of a ship is deduced
from a few bottom planks and a rib or two. There are some practical limits to the
shapes of wooden ships.
It is unlikely that anything as important as the Kyrenia ship, the Bremean cog, or
the Wasa ever will be found in the waters of the United States, but there may still be a
few surprises lurking here and there. The vessel raised on August 28, 1976, from the
Black River at Brown's Ferry, South Carolina, is one such surprise. A double-ended
round-bilged, flat-bottomed craft without a keel and built of iocal species of timber, it
is about 50 feet long with a breadth of 14 feet. Dated about 1740, it is unlike anything
previously known in the United States, but museum curators in the Low Countries find
little unusual about her. The vessel's European ancestry may be explained by the fact
that colonists of Germanic origin moved from Pennsylvania to South Carolina early in
the 18th century. Although found in a freshwater river with a cargo of brick on board,
the Brown's Ferry vessel had seen saltwater service, for its planking shows teredo dam-
age and among the artifacts on board was a nearly complete Davis quadrant.
Still being excavated in the harbor of Stockton, Maine, are the remains of an
American privateer identified as the brigantine Defence that was part of the disastrous
colonial expedition in August 1779 against the British in fortified Castine. When a
British fleet appeared at the entrance of Penobscot Bay, about 40 colonial vessels fled
up the river where they were scuttled or run aground and burned. The Defence was in
the latter category; its stern was blown off when the fire reached its magazine. It sank
and settled into the mud where it is today.
Although from a period for which standard shipbuilding techniques are fairly
well known, the Defence is of interest because its structural details indicate hurried
construction with perhaps the feeling that it was expendable. Many of its frames and a
large breasthook still have bark on them, and its structure is considerably lighter than
that of what is thought to be the wreck of another privateer of the same period. Because
of its size and shape, the Defence could have had but limited postwar commercial use.
Other examples could be cited but the foregoing will suffice. Without ships there
would be no artifacts pertaining to crews, passengers, and cargo. In conclusion, let this
be a plea to concentrate on the excavation of underwater ship finds that offer some
promise of information about the ships themselves.
 Chapter 3

Introducing Maritime
Archaeology

KEITH MUCKELROY

GENERAL INTRODUCTION AND DEFINITIONS

In any preindustrial society, from the Upper Paleolithic to the 19th century A.D., a boat
or (later) a ship was the largest and most complex machine produced. At Star Carr, the
Mesolithic site in Yorkshire excavated by Professor Grahame Clark, none of the
artifacts discussed in the report would have rivaled in terms of size, variety of materi-
als, or construction time the skin-craft whose existence the excavator has postulated
(Clark, 1954:23). At the other end of that time span, the 18th-century First-Rate naval
ship, with its IOO-plus guns and crew of over 800, exceeded several times over, in num-
bers of constituent artifacts and in quantity of power harnessed, the largest machines
used on land for transport, manufacture, or mining. Even the Roman Empire, with its
development of large-scale systems in military, mining, and food-processing technol-
ogy, is not exempt, as these operations were paralleled by a gigantism in shipbuilding
that reached its peak with the grain ships running between Egypt and Rome (Casson,
1971: 184-189). But such a dominating position for maritime activities has not been
limited to the technical sphere; in many societies it has pervaded every aspect of social
organization. The political importance of these same grain . ships, in giving the ruling
emperor the whip hand over the Roman populace, constituted an important part of his
power base (Lewis and Reinhold, 1955:138-142). In 5th-century B.C. Athens, the
political power of the Demos owed a great deal to its role as the motive force for the
Athenian galleys, on which in tum the security of the state was thought to depend
Source: Maritime Archaeology. New York: Cambridge University Press, 1978. Copyright 1978 by Cambridge
University Press. Reprinted with the permission of Cambridge University Press.

23
24 K. Muckelroy

(Ehrenberg, 1967:216). And in 18th-century England, the Admiralty was the biggest
single employer of labor in manufacturing, and played no small role in determining the
level of economic activity, and stimulating industrial innovation. At a different level, in
many societies past and present, seafaring and fishing folk have formed a distinct sub-
culture, alongside the more generally recognized urban and rural groups (HassIOf,
1972: 15-17). In these ways, and countless others besides, the course of human history
has owed not a little to maritime activities, and their study must constitute an important
element in the search for a greater understanding of man's past.
When considering any aspect of the past, there are several different approaches that
may be used, the principal distinction between them being the type of evidence they are
designed to utilize. The longest established and most highly developed of these disci-
plines, in the study of seafaring as in the study of most other activities, is the historical
one, in which the primary concern is with the uncovering and interpretation of surviving
documentary evidence for past events, and by which the researcher seeks to understand
not only the precise course of events but also the reasons, causes, or motives behind
them. Another approach, the development of which has proceeded furthest in the Scandi-
navian countries, is ethnological- the systematic study of surviving indigenous prac-
tices, traditions, and customs, in this case within specialized fishing and seafaring
communities. Finally, one can study the objects that have survived from past activities on
and around the sea, and from them derive insights into the men and societies that pro-
duced them; this is essentially an archaeological study. The information and ideas con-
tributed by these various approaches sometimes duplicate and sometimes contradict each
other, but above all they should be viewed as complementary in the overall field of mari-
time studies. In the present work, attention is focused on the special characteristics of the
last of these, maritime archaeology, which can be defined as "the scientific study of the
material remains of man and his activities on the sea."
Some of the ideas implicit within this statement are worth elaborating. Above all, it
should be noted that the primary object of study is man, as asserted in the first half of the
formula, and not the ships, cargoes, fittings, or instruments with which the researcher is
immediately confronted. Archaeology is not the study of objects simply for themselves,
but rather for the insight they give into the people who made or used them, a sentiment
summed up in Sir Mortimer Wheeler's trenchant dictum "the archaeologist is digging up,
not things, but people" (Wheeler, 1954: 13). Thus, the first part of this definition simply
defines archaeology, while the second part accounts for the qualification "maritime."
With respect to this latter phrase, it is worth noting that there is no mention of boats or
ships, but rather of everything that is connected with seafaring in its broadest sense. As
considered in this book, maritime archaeology is concerned with all aspects of maritime
culture; not just technical matters, but also social, economic, political, religious, and a
host of other aspects. It is this fact that distinguishes the subdiscipline from the closely
allied subject of nautical archaeology, which is here taken to mean the specialized study
of maritime technology - in other words, ships, boats, and other craft, together with the
ancillary equipment necessary to operate them. It is thus a specialty within maritime
archaeology, in just the same way as, for example, the study of town houses can be
regarded as a specialty within urban archaeology.
The adjective "scientific" has been inserted at the start of the definition to show
that this study is aimed at generating new insights and ideas through systematic research,
and not as a tendentious contribution to the worn-out debate as to whether archaeology is
Introducing Maritime Archaeology 25

a science or not. The reference here is to science in its widest sense, the disciplined
search for knowledge (cf. Latin scire: "to know") as opposed to the aimless delight in
"curiosities." It carries with it the implication that archaeological research in any field
must be problem oriented; in other words, to ensure the maximum return from the avail-
able material, the researcher must always have in mind the questions outstanding in the
current state of his discipline toward which that evidence might be expected to contribute
some of the answers. It is only by this steady accretion of data within a systematic frame-
work that any real advances in knowledge or understanding can be made; without it, each
worker is essentially starting from scratch, and it is as if all previous workers had not
existed, the same basic questions being considered over and over again. It can thus truly
be said that the sign of a really successful piece of research in any discipline is a state-
ment in the worker's conclusions to the effect that his studies "have raised more ques-
tions than they have solved." Real progress has been made, since his successors can
study new material from a more advanced viewpoint. From this approach flows the
implication that a discipline can be most effectively described by considering the prob-
lems toward which research is currently directed, the specific questions being raised, and
the ways in which workers are seeking to answer them. Hence, the main body of this
book is concerned with problems rather than with artifacts, with questions rather than
with treasure.
Turning now to what this definition does not say, some of the principal implica-
tions of this "scientific" approach will be made explicit. Some readers will be surprised
that it gives no time limits either before or after which the study of the material remains
ceases to be archaeology. In fact, the requirement that the principal concern of any
study must be man effectively defines a starting date, at the point at which the first
hominids can be recognized. However, no terminal date is specified, even by implica-
tion, and none is intended. It is necessary though that scientific research should be con-
tributing new knowledge, so that there is an effective closing date at the moment when
other sources of information give the required data more readily and directly than an
archaeological approach. This date will vary according to the question concerned, and
for certain topics may be only a few years ago, while for others it may be several thou-
sand of years back; any attempt to name a general closing date for the whole of mari-
time archaeology would thus be impossible, and contrary to its scientific nature.
The other outstanding omission from the above definition is the fact that it does not
include the further phrase "together with related objects on shore." Concern with coastal
communities that derive their livelihoods predominantly from the sea is excluded since,
being primarily terrestrial settlements, they will be more closely related to surrounding
communities in their material culture, and will display their maritime connections only
marginally. Many of the objects used in seafaring are rarely brought ashore, and any arti-
fact collection made there will represent very poorly the seafaring community itself. This
exclusion is made in full consciousness that maritime ethnologists and anthropologists
have included such communities in their brief; their case is somewhat different since
they rely very heavily on such communities for their material, and can readily sort it into
maritime and regionally oriented elements, according to the contexts within which it was
gathered. With excavated finds, the distinction is often less clear and there will always be
a gray area of uncertainty, defeating any attempt to separate out the two elements. Where
there is no doubt, as with installations directly servicing ships and seafarers, such as har-
bors or wharves, then their study can properly be said to come within the scope of this
26 K. Muckelroy

subdiscipline. To sum up, therefore, maritime archaeology is the scientific study, through
the surviving material evidence, of all aspects of seafaring: ships, boats, and their equip-
ment; cargoes, catches, or passengers carried on them, and the economic systems within
which they were operating; their officers and crew, especially utensils and other posses-
sions reflecting their specialized lifestyle. Reference to current work in maritime history
will show that this definition of the field of research mirrors the concerns of that disci-
pline as well; only the sources of information tapped are different.
The relationship between these two disciplines, the historical and the archaeologi-
cal, is a complex one, and must be explored a little further. "Archaeology, is archaeol-
ogy, is archaeology," wrote Dr. David Clarke (1968:13), and this applies in the present
field as in any other; the aim is not simply to produce "counterfeit" maritime history. The
danger of accepting a role little better than that of an academic maidservant, collecting
interesting facts for historians to interpret, is a particularly insidious one, since it has the
superficial attraction of association with an established and reputable discipline. As indi-
cated above, both specialists have their own sets of evidence, and their own questions to
answer, and with both disciplines becoming ever more sophisticated and specialized, it is
increasingly difficult for one man to be an expert in both of them. He will either be a
good historian and a bad archaeologist, or vice versa. This is not to say, of course, that
either should ignore the results of the other's work, where it contributes toward a topic
under discussion, but rather that each should consider critically the conclusions of the
other's work, integrate them within their own conclusions where possible, and indicate to
their colleagues where a dichotomy exists. Thus, problems can be tossed backwards and
forwards between specialists in different disciplines to their mutual advantage, without
anyone of them asserting a general primacy. Unfortunately, because of the way aca-
demic research has developed in recent centuries, there is still a general tendency to
assume a priority for the results of historical studies where conclusions conflict; this is
certainly true at present with maritime history, and will probably continue to be so until
maritime archaeology can evolve an ordered and coherent structure.
There is a similar confusion of aims and ideas between this subject and maritime
ethnology, based on the fact that both proceed, at least in part, by the investigation of
material evidence. However, the essence of ethnology is to view this evidence in the con-
text of social forms, economic systems, etc., which themselves can also be recorded,
while archaeology has only the material evidence to study. Thus, from the point of view
of the archaeologist, ethnological studies are just a part of the wide range of sources from
which inferences and parallels can be gathered when seeking to interpret the remains on
a par with historical evidence, the results of experimental archaeology, or theories pro-
duced by logical processes. For various reasons to be considered further in Part Two,
resort to ethnological parallels has been more popular in maritime archaeology than in
almost any other archaeological subdiscipline, with many exceedingly fruitful results.
However, the point to be emphasized at present is that, despite this close relationship, the
two disciplines are essentially distinct, and any fudging of the boundary between them
can only lead to both bad ethnology and bad archaeology.
So far, the subject has been defined in terms of the problems to be investigated, and
the types of evidence used, thus distinguishing it from other branches of archaeology or
allied specialties. However, it possesses one further attribute that fundamentally affects
the procedures used, and that separates it markedly from all other archaeological subdis-
ciplines - almost all of the fieldwork in maritime archaeology is carried out underwater.
Introducing Maritime Archaeology 27

As this is a direct consequence of the type of material studied, it is not properly part of
the definition of the subject; nevertheless, it is probably its most striking characteristic. A
ship undertaking a voyage leaves absolutely no imprint on the archaeological record,
and, if all goes well, the evidence will be effectively dispersed at the end of the voyage,
when the cargo is sold, the crew go to their homes, and the ship is taken on for a new en-
terprise or broken up. It is only if disaster strikes during the voyage, and the whole unit
- ship, cargo, and shipboard community - is deposited on the seabed, that there is any
chance of a permanent material record which is archaeologically recoverable. The actual
quality of the remains will, of course, depend on a number of factors, such as the manner
in which the vessel was wrecked, or the nature of the seabed on which it landed. Thus,
the scope of this subdiscipline as actually practiced is largely determined by the potential
and limitations of the underwater environment, both as a medium for the preservation of
remains, and as the situation in which fieldwork has to be undertaken. Similarly, the
modes of analysis appropriate to the material evidence are closely associated with an un-
derstanding of the process of a shipwreck.
An objection to this assertion of the dominance of underwater remains in these
studies might be raised on the grounds that a considerable number of boats and ships
have been excavated above water, ranging in date from the Ferriby boats (ca. 2000 B.C.)
through the Nydam boat (fourth century A.D.) or the Graveney boat (9th century A.D.) to
the Amsterdam (1749). However, while not denying the value or validity of the evidence
gleaned from these sources, it should be recognized that they all represent special cases.
Probably the largest group among these terrestrial finds consists of maritime equipment,
above all boats, which has been deposited in graves; for example, Muller-Wille has iden-
tified over 420 boat-burials of the Iron Age and early medieval period in northern Europe
(Muller-Wille, 1974). At the moment of deposition these craft were not engaged in mari-
time activities; they were usually stripped of their sailing gear, and often otherwise struc-
turally modified. For example, the Sutton Hoo boat (6th century A.D.) had a substantial
wooden chamber erected amidships to house the grave treasure (Bruce-Mitford,
1975: 176-180). While such finds can obviously contribute considerable evidence, their
total interpretation involves many considerations apart from the purely maritime; so
these sites must be regarded as peripheral in the present context, and their special de-
mands and features will not be considered further. Another terrestrial group is the re-
mains of craft that have been deliberately beached and abandoned, usually after
everything useful had been stripped from them; notable examples of this situation in-
clude the Graveney boat (Fenwick, 1972) and the Bursledon ship (15th century A.D.;
Prynne, 1968). While undoubtedly falling in the mainstream of this subdiscipline, these
remains are a special category because they too have been preserved at a time when not
involved in a maritime activity, so that they betray little concerning their original eco-
nomic and social roles. Finally, a substantial group within this category is in fact only
terrestrial in a limited sense; it comprises sites originally underwater that have been arti-
ficially drained, or otherwise removed from a marine environment. Occasionally, this re-
moval has been at the instance of the archaeologists, as with the Roskilde Viking boats
(Olsen and Crumlin-Pedersen, 1967), but more usually it is the consequence of some
other human activity, as with the Blackfriars Roman boat (Marsden, 1966) or the hun-
dreds of vessels discovered on the reclaimed lands of the Zuider Zee (van der Heide,
1976). In these instances, a considerable number of the special considerations imposed
by the underwater environment still apply, and it is principally in the techniques of field-
28 K. Muckelroy

Nautical
archaeology

Archaeology
under water

Area of concern of
maritime archaeology

Figure I. A diagram illustrating the scope

of maritime archaeology and its relation-
ship to the allied topics of nautical archae-
ology and archaeology underwater.

work that distinctions must be made. Ultimately, it is only at sea that seafaring disasters
can occur, so that it is under the surface of the sea that the bulk of the evidence must lie.
To clarify and summarize the import of this discussion, the relationship between
maritime archaeology and the two allied topics of nautical archaeology and archaeol-
ogy underwater is represented diagrammatically in Figure 1. The area of concern in this
book is outlined by the double circle, and thus excludes small sections of each of the
other topics. In the case of nautical archaeology it is those boat and ship finds that are
in a totally nonmaritime context, notably grave finds (area A). In the case of archaeol-
ogy under water, it is those sites that are not concerned directly with maritime activi-
ties, notably submerged ancient land surfaces (area F). But this covers only a very
small section of underwater work; the bulk of it is concerned with maritime affairs,
being relevant both to the study of maritime technology (area D) and to the many other
aspects of seafaring (area E). In addition to these, however, maritime archaeology
involves those sites that are not submerged and that contain evidence either about
ancient shipping alone, e.g., beached craft (area B), or about the whole range of past
maritime activities, e.g., drained sites (areas B and C). But since these last were origi-
nally underwater, it remains true to say that nearly all of the evidence must come from
submerged sites, so that the constraints of the underwater environment can reasonably
be said to be one of the main characteristics of this subdiscipline.
Another outstanding feature, as things stand in the late 1970s, is a remarkable
lack of development or systematization, when compared with most other archaeologi-
Introducing Maritime Archaeology 29

cal subdisciplines. This arises directly from the fact that it is a relatively new study, and
is only now reaching the position where the database is sufficiently extensive to allow
some tentative steps in defining the discipline. In order to understand this academic
immaturity, it is necessary to have a brief look at the history of the subject, an exercise
that will also place in context the specific examples of research problems that are
described later in this book.

THE DEVELOPMENT OF MARITIME ARCHAEOLOGY

Ever since ships first voyaged on the sea, there have been shipwrecks, and these
in turn have always attracted the attentions of potential salvors; only wrecks in deep
water or off totally uninhabited coasts will have completely escaped salvage. On many
of the more accessible or valuable sites, this work has continued intermittently through
the centuries until the present, in some instances accompanied by increasing antiquar-
ian curiosity, and sometimes merging in the recent past with genuine archaeological
investigations. For many thousands of years, the only tools available for such work
were nets, grabs, or grappling hooks, aided in warmer, clearer waters by the services of
free divers. In recent centuries, these operations have been made more efficient by the
development of means of getting men onto the seafloor: first in bells (17th century),
then in enclosed barrels (18th century), later with 'hard-hat' standard diving gear (19th
century), and finally in the past 30 years with self-contained breathing apparatus
(Davis, 1955). Over the same period, archaeology has emerged from the unrestrained
speculations of antiquarians into a systematic and disciplined study, with the aims and
objectives indicated above. However, in relation to many other sciences, archaeology
was a late developer, and can be recognized in its modern form only in the later 19th
century, with the work of such pioneers at c.J. Thomson (1788-1865), O. Montelius
(1843-1921), and General Pitt-Rivers (1827-1900) (Daniel, 1967; Clarke, 1968:4-11).
It is thus not surprising that little in the way of maritime archaeology can be detected
before the opening years of the present century.
Glancing at these earlier times, one can see some enquiring minds that were fasci-
nated by the possibilities of such remains. An early reference to such interest comes from
the 11 th century A.D., when Abbot Ealdred of st. Albans sent his men to the ruins of
Roman Verulamium to collect stones for his new abbey, during which operations they
found "oak timbers with nails sticking inside and smeared with naval pitch," an event
deemed worthy of note in his Life (Ellmers, 1973). A few centuries later, as a result of
the interest of Cardinal Colonna in the tradition of large Roman ships said to lie within
Lake Nemi in Italy, an attempt was made to salvage one by the architect Leon Battista
Alberti (1446). Continued interest in this site led, a century later, to one of the earliest
recorded examples of diving, when a crude suit was used in a reconnaissance by a certain
Franchesco Demarchi (1535). Moving on several centuries, and into colder waters, it was
still antiquarian curiosity that inspired the fine watercolors made for the Deane Brothers
of finds recovered by them from various historic wreck sites around Britain. At about the
same time, the first really scientific consideration of the potential for the preservation of
human artifacts within marine sediments was published, although by one of the pioneers
of geology, Charles Lyell, in his Principles of Geology (1st edition, 1832). Chapter 16 of
Volume 2 is titled "On the imbedding of the remains of man and his works in subaqueous
strata," and includes a summary of recent shipping losses to show the extent of the mate-
30 K. Muckelroy

rial being deposited, as well as accounts of contemporary submarine finds that exhibit
high degrees of preservation. He concluded that "it is probable that a greater number of
monuments of the skill and industry of man will in the course of ages be collected
together in the bed of the ocean, than will exist at anyone time on the surface of the Con-
tinents" (ibid., 258). However, more general 19th-century attitudes to such antiquities are
nicely illustrated by the case of an old boat found at Rye (Sussex) in 1822, which was put
on display in London for a time, but broken up when public interest flagged (Rice, 1824).
Naturally, the earliest archaeological studies on this subject resulted from boat
finds on land, beginning with the great series of early medieval craft found in Scandina-
via, the first systematic excavation being that in 1863 by Conrad Engelhardt of the 4th-
century A.D. boat from Nydam (Denmark). However, while modern archaeology was
developing on land, there seemed to be no archaeologists adventurous enough to go un-
derwater in the standard diving equipment of the day. Even in 1907, when the Society of
Antiquaries of London wished to investigate a site from which much Roman pottery had
been dredged up north of Herne Bay in Kent, they employed a "Certificated Diver," Mr.
Hugh Pollard, to go and have a look (Smith, 1909). However, a year later an amateur ar-
chaeologist took the plunge, in the unlikely person of a Benedictine priest, the Reverend
Odo Blundell, of Fort Augustus in Scotland. He was interested in the history and con-
struction of a crannog (lake dwelling) in Loch Ness a couple of kilometers from his ab-
bey, and he soon concluded that the only way to find out how it was built was to have a
close look at the bottom of it. He persuaded a diving crew from the Caledonian Canal
Company to provide him with a suit and assistance, and went down without any major
problems on August 7, 1908 (Blundell, 1909) .... In the following summer this remark-
able character dived on a number of other crannog sites (Blundell, 1910) - work that
initiated a British Association research project on Scottish man-made islands.
During this same decade, the attention of all those interested in classical antiquity
was being drawn by a series of spectacular finds of ancient works of art from the Medi-
terranean. The first of these was off the island of Antikythera, midway between Crete
and the Greek mainland, where, in 1900, Greek sponge divers discovered a pile of mar-
ble and bronze statues in 60 m of water. On learning of this find, the Greek government
organized recovery operations, using naval craft, with operations being directed from
the surface by the Director of Antiquities, Professor George Byzantinos. Over the next
year, an impressive collection of statuary and other objects was recovered, including
the famous Antikythera youth and an early mechanical calculator; the find apparently
represented the remains of a Roman vessel carrying Greek treasures to Rome after the
victories of Sulla in 86 B.c. (Weinberg et al., 1965). Seven years later, an almost identi-
cal sequence of events occurred off the Tunisian coast at Mahdia, where a similar cargo
of loot was also found by sponge divers. In this case, the Tunisian Department of
Antiquities continued to support salvage operations until 1913 (Frondeville, 1965). The
value of these chance finds, in both artistic and archaeological terms, was such that the
great potential of underwater sites was widely appreciated among classical archaeolo-
gists from the early years of this century, a fact that had great consequences when the
invention of the aqualung finally allowed that potential to be exploited. Salomon Rei-
nach summarized this realization when he wrote that "the richest museum of antiquities
in the whole world is still inaccessible. I mean the seabed of the Mediterranean."
But the exploitation had to wait until after the Second World War. The decades
between the wars saw advances only in a few limited areas, along with a few spectacular
Introducing Maritime Archaeology 31

chance finds, such as that at Cape Artemision. Undoubtedly the most publicized opera-
tion in this period was the draining, on Mussolini's orders, of Lake Nemi in order to
reveal and raise the famous Roman craft known to lie there (Ucelli, 1950). The project
was a success, providing both a spectacular museum display, and much new and detailed
information about the construction and sheathing of Roman ships. However, the extent to
which they could be regarded as typical was in doubt, since, with their marble columns,
heated baths, and decks paved with mosaics, they were evidently very special craft.
Much less widely known was the fundamental work done by the Jesuit, Father Andre
Poidebard, on the harbor remains of the Palestine coast, beginning at Tyre in 1934. Him-
self a pioneer of both aerial and underwater photography for archaeological purposes, he
brought both skills to bear, with the assistance of French naval personnel and local
sponge divers, on the problems of identifying and mapping these extensive remains. His
criteria for recognizing, and to some extent dating, these huge structures have remained
valid into the aqualung age (Poidebard, 1939; Frost, 1963:65-114).
Although various attempts at producing a self-regulating underwater breathing
apparatus had been made previously, it was the work of a French Navy officer, Jac-
ques-Yves Cousteau, and an engineer, Emile Gagnan, in 1942, that finally achieved an
aqualung that could be used by ordinary people. The cost of the old standard "hard-hat"
gear, the considerable strength and endurance required to use it, the amount of training
and experience necessary to work effectively within the suits, the large number of sur-
face assistants required, and above all the cumbersomeness of the diver on the bottom
had all ruled out the possibility of genuine archaeological work in the old equipment:
on all of these scores the appearance of the aqualung represented a revolution. Of
course, it also represented an advance for those interested in wreck sites as a source of
souvenirs or profitable antiquities, so that the early years after the War saw consider-
able depredation of undersea sites, a loss that still continues, despite the strict laws
regarding seabed antiquities enacted by most coastal states.
One of the first serious attempts to investigate a classical wreck site systemati-
cally was undertaken by Captain Cousteau himself, when he led his Undersea Research
Group in an excavation of a large amphora mound off the island of Grand Congloue,
near Marseilles. Although a number of techniques and tools were first developed on
this site, the archaeological standards appear, in retrospect, to have been unacceptably
low - for example, no plan of the wreck site was ever produced - and there is still
considerable controversy as to whether there were one or two wrecks on this site
(Benoit, 1961). Nevertheless, the practicability of disciplined excavation underwater
using aqualung divers had been demonstrated, although it was several years before any
further progress was made. In the meantime, it was left to concerned individuals, such
as Frederic Dumas, one of Cousteau's associates, to record what they could between
the looting. Another activist was Commander Philippe Taillez, who, while chief of the
French Navy's Diving School, organized the excavation of a 1st-century B.c. wreck on
the Titan reef off the French coast, and who was particularly conscious of the inadequa-
cies in the archaeological control of his work. As he wrote in the conclusion to his
report: "We have tried sincerely, to the best of our ability, but I know how many mis-
takes were made .... If we had been assisted in the beginning by an archaeologist, he
would surely have noted with much greater accuracy the position of each object; by
personal inspection he would have drawn more information from the slightest indica-
tions" (Taillez, 1965:91).
32 K. Muckelroy

Unfortunately, in France no diving archaeologist appeared until a great deal had

been lost. It was a U.S. team from the Museum of the University of Pennsylvania,
working off the Turkish coast, who first demonstrated that it was possible for archae-
ologists to work on the seabed, even in 30 m of water. In 1960, George Bass led a team
including Peter Throckmorton, Joan du Plat Taylor, and Frederic Dumas in excavating
a site, off Cape Gelidonya, of a ship that sank around 1200 S.c. The result was a major
archaeological triumph, both in terms of techniques, which allowed few if any conces-
sions to the fact of being under water, and in terms of a substantial contribution to stud-
ies on late Bronze Age trade in the eastern Mediterranean (Bass 1967). This project
initiated a great series of U.S.-sponsored underwater excavations in Turkish waters,
notably on the 4th- and 7th-century A.D. wrecks off Yassi Ada. Members of these teams
were also responsible for the extensive excavations undertaken on a 4th-century B.C.
wreck near Kyrenia in Cyprus directed by Michael Katzev. This project concluded with
the raising and conserving of the surviving ship's structure, which is now on display in
Kyrenia Castle (Swiny and Katzev, 1973; Katzev, 1974).
In touching on only the more outstanding milestones in the development of mari-
time archaeology in the Mediterranean, a great deal of important work has been over-
looked, especially in countries other than France or Turkey. The last decade has seen a
general increase in positive government support for such work; for example, consider-
able progress has been made in Yugoslavian waters since 1969, when a well-regulated
system was established based on "working teams" in every major coastal town by the
Office for the Protection of Monuments (Zagreb) (Vrsalovic, 1974). Another hopeful
sign for the future has been the success of French archaeologists in excavating sites
from which many of the surface finds (notably amphoras) have been looted, but on
which extensive buried remains have been found, sometimes including areas of ship's
structure: examples include the sites of Dramont D (Joncheray, 1975), Grand Ribaud A
(Carraze, 1975), La Roche Fouras (Joncheray, 1976), and La Tradeliere (Fiori and Jon-
cheray, 1975). At the same time, the standards, ideas, and procedures laid down by
George Bass and his team over 15 years ago remain the model that present-day workers
seek to emulate.
However, over the past couple of decades maritime archaeology has not been
limited to Mediterranean waters. Possibly the most spectacular project of all has been
carried out in Sweden, where the warship Wasa has been raised from Stockholm harbor
and conserved intact. Built in 1628, this ship appears to have possessed a fundamental
design fault, for she capsized on her maiden voyage. Her remains were relocated in
1956, and after several years of preparatory work under water, she finally resurfaced in
May 1961 (Franzen, 1966). From a technical point of view, this underwater work,
which involved tunnelling under the wreck in order to pass cables around her, was a
major feat in itself. Similarly, the excavation of the seabed around the wreck site after
the vessel had been lifted, in order to recover elements of the ship's decorations and
other items that had fallen from her, was itself one of the most intensive undersea exca-
vations undertaken up until that time. But Swedish maritime archaeology has not
stopped with the Wasa, and fieldwork has continued on many other spectacular wrecks
preserved within the worm-free waters of the Baltic (see, for example, Cederlund and
Ingelman-Sundberg, 1973; Cederlund, 1977).
Elsewhere in northern Europe, probably the most extensive work has been that
undertaken by Dutch archaeologists, notably G. van der Heide, on the drained lands of
Introducing Maritime Archaeology 33

the polders, which, with their moving shoals and sudden violent storms, had been a
graveyard of ships since seafaring commenced in that area (van der Heide, 1976).
Unfortunately, the sheer volume of this material, along with other unfavorable circum-
stances, has meant that little detailed information has yet emerged from these sites.
However, the evidence to be gleaned from this material, concerning North Sea vessels
between the 12th and 19th centuries A.D., will undoubtedly prompt a considerable revi-
sion of established ideas. Elsewhere in the Netherlands, the discovery and excavation
of occasional finds of riverboats, such as those at Zwammerdam (de Weerd and Haale-
bos, 1973), has contributed to knowledge of such craft in the early centuries A.D.
Across the Channel in Britain, similar chance finds have complemented those in
the Low Countries, and revealed new aspects of Romano-British shipbuilding; a nota-
ble example was the discovery and excavation of the Blackfriar's Roman ship (Mars-
den, 1966). Other vessels of later dates have since been found in the same vicinity
(Marsden, 1971). For specialists interested in the early medieval period the focus of
attention has remained the dramatic discoveries made at Sutton Hoo (Suffolk) in the
summer of 1939, where a royal ship-burial preserved, along with outstanding treasures,
the impression of a 6th-century A.D. English ship. Between 1965 and 1967 this site was
reexcavated at the instance of the British Museum. Another important medieval boat
find occurred in 1970 when a 9th-century vessel was discovered at Graveney in Kent.
The rescue excavation of this craft by the National Maritime Museum, Greenwich, in-
itiated a major research program into the archaeology of boats in northwest Europe
(Greenhill, 1976:221-233), which is still continuing.
Archaeological research in British waters developed slowly, and has been con-
centrated almost exclusively on remains of the po'st-1500 period. The first historic
wreck site to be positively identified was a Dutch East Indiaman, De Lie/de (1711), off
the Out Skerries (Shetland Isles) (Bax and Martin, 1974). That was in 1965. The events
that followed were initially more related to commercial salvage than archaeology, but
the activities of bodies such as the Council for Nautical Archaeology (founded 1964)
and others began to improve standards and a few archaeologists have taken up diving.
The vessels of the Dutch East India Company, several of which were lost off British
shores, have continued to attract attention, perhaps the most spectacular site being that
of the Amsterdam where the ship herself has sunk into beach sands near Hastings
(Marsden, 1972; 1974). Another subject of continuing interest has been the wrecks of
the Spanish Armada of 1588, beginning with the location in 1968 of the remains of the
Santa Maria de fa Rosa (Martin, 1975) and the Girona (Stenuit, 1972). The growing
number of club divers since the mid-1960s has also increased the range of historic
wreck sites being discovered, sometimes by chance, and sometimes is the result of a
methodical search; the positive identification of the site of the remains of the Mary
Rose (1545) in the Solent between 1967 and 1971 is perhaps the most outstanding
example of the latter (McKee, 1973). The new field of research has achieved limited
success in gaining acceptance in academic circles, and it was as late as 1973 that Brit-
ain's first Institute of Maritime Archaeology was established at St. Andrews Univer-
sity. The same year also saw the first steps in legal recognition of the special problems
posed by these developments, with the Protection of Wrecks Act, 1973.
Outside Europe, the most active area for underwater work on historic wrecks has
undoubtedly been North and Central America, where the great treasures in gold and sil-
ver contained in the wrecks of the Spanish plate fleets have put a special edge on their
34 K. Muckelroy

exploitation. Undoubtedly much of this has been guided by commercial rather than
archaeological considerations, but recent years have seen various attempts to control the
situation. Many maritime states of the United States have imposed legal controls on
underwater sites, and some have also established state underwater archaeological units to
supervise operations: some of the systematic surveys and excavations undertaken by
these teams in recent years have been of a very high calibre. Less spectacular, but
equally informative, finds of wrecks of the War ofIndependence have been made along
the eastern coast of the United States, as at Yorktown (Bass, 1966:123) or in the Penob-
scot River with the site of the Defense (Mayhew, 1974). Similar sites have also received
attention off the coasts of Canada, some at the instigation of the National Historic Sites
Service, as with the excavations on the wreck of the French warship Machault (1760)
(Zacharchuk, 1972), and some from groups of amateur enthusiasts, as with the Sapphire
(1695) (Barber, 1977). Also of note have been the studies undertaken on the remains at
porterage sites along rivers in the midwest, revealing a fascinating cross section of the
trade goods sent into the interior in the 18th and 19th centuries from east coast settle-
ments (Wheeler and van Gemert, 1972).
So far as the rest of the world is concerned, systematic work is only just begin-
ning, and there is less to report. Israeli workers have studied wrecks in the Red Sea as
well as the Mediterranean, notably in the Sharmal-Sheikh area (Linder and Raban,
1975:44-47). An interesting wreck of a 17th-century Portuguese vessel has been
known off Mombasa, Kenya, for many years (Kirkman, 1972; Wheeler et al., 1975),
and a major investigation of this site by an international team began early in 1977.
Recently, work has started on an investigation of a 14th- or 15th-century A.D. wreck
near Sattahip in Thailand (Weier 1974), and other important sites are reported to lie in
this area. The few reports available concerning fieldwork in Japan show an awareness
of the subject there that may yield important results in the future (Osaki, 1973). A nota-
ble hive of activity in maritime archaeology since 1971 has been Western Australia,
where, under the aegis of an excellent series of legal provisions, the State Museum has
established a specialist unit, directed by Mr. Jeremy Green, and staffed by a full range
of appropriate specialists (Pearson, 1976; Green and Henderson, 1977). Their principal
concern has been with the sites of some of the Dutch East Indiamen wrecked on their
shores, notably the Batavia of 1629 (Green, 1975), the Vergulde Draeck of 1656
(Green, 1973), and the Zeewijk of 1727 (Ingelman-Sundberg, 1977). Equally notewor-
thy has been their application of similar techniques and standards to the wrecks of the
colonial period, i.e., the early 19th century (Henderson, 1976).
This brief history of the subject has obviously been far from exhaustive, and many
important research projects have been omitted; some will appear in the appropriate sec-
tions of Chapter 3. However, in the course of this survey, certain important attributes of
this subdiscipline have emerged. Above all, its extreme youth is apparent; its modern
development was dependent on the invention of the aqualung, which occurred only just
over 30 years ago. Furthermore, the whole business of separating archaeology from pure
salvage or treasure hunting, and building up a new profession of diving archaeologists,
meant that the first 20 years of this period saw only faltering steps, and that in most parts
of the world genuine archaeological excavations have only been attempted within the
past decade. A related feature is the wide variety of backgrounds from which the practi-
tioners of this subject have been drawn, including naval diving (e.g., Captain Cousteau
and his team), commercial salvage operations (e.g., much of the work in the United
Introducing Maritime Archaeology 35

States), as well as conventional archaeology (a succession with Professor Bass at its

head). That this was inevitable, in view of the nature of the new technology and the other
skills required in such work, must be accepted, but it has given the whole subject a flavor
that has set it apart from more conventional archaeology. These considerations have also
influenced the types of problems that have been tackled in recent years; the early threat
to amphorae wrecks in the Mediterranean has focused attention there on merchantmen of
the classical period, while the lure of treasure and the prominence of cannon elsewhere
has concentrated work outside that sea on the centuries after 1500 A.D. This short sum-
mary of the checkered development of the subject should at least explain, if not excuse,
the uneven range of topics discussed later. Finally, it should be apparent why maritime
archaeology has lacked a general statement of its theories and concerns to date, and why
such a synthesis should now be attempted.

REFERENCES
Barker, P., 1977, Techniques ofArcheological Excavation. Batsford, London.
Bass, G.F., 1966, Archaeology Under Water. Thames and Hudson, London.
Bass, G.F, 1967, Cape Gelidonya: A Bronze Age Shipwreck. Transactions of the Philosophical Society 57(8).
Bax, A., and Martin, C.J.M., 1974, De Liefde, a Dutch East Indiaman Lost on the Out Skerries, Shetland, in
1711. International Journal of Nautical Archaeology and Underwater Exploration 3: 81-90.
a a
Benoit, F., 1961, L 'epave du Grand Congloue Marseilles. XIVe supplement Gallia, Paris.
Blundell, Rev. 0., 1909, The Crannog of Eilean Muireach. Proceedings of the Society ofAntiquaries ofScotland
43: 159-164.
Blundell, Rev. 0., 1910, Further Examination of Artificial Islands. Proceedings of the Society ofAntiquaries of
Scotland 44: 12-33.
Bruce-Mitford, R., 1975, The Sutton Hoo Ship-burial, Vol. I. British Museum Publications, London.
Carraze, F., 1975, L'epave 'Grand Ribaud A'. Cahiers d 'Archeologie Subaquatique 4: 9-58.
Casson, L., 1971, Ships and Seamanship in the Ancient World. Princeton University Press, Princeton, NJ.
Cederlund, e.0., 1977, Preliminary Report on Recording Methods Used for the Investigation of 87 Merchant
Shipwrecks at Jutholmen and Alvsnabben in 1973--74. International Journal of Nautical Archaeology
and Underwater Exploration 6: 87-99.
Cederlund, e.0., and Ingelman-Sundberg, C., 1973, The Excavation of the Jutholmen Wreck, 1970--71. Inter-
national Journal of Nautical Archaeology and Underwater Exploration 2: 30--327.
Clark, J.G.D., 1939, Archaeology and Society. Methuen, London.
Clark, J.G.D. 1954, Excavations at Star Carr. Cambridge University Press, London.
Clarke, D.L., 1968, Analytical Archaeology. Methuen, London.
Daniel, G., 1967, The Origin and Growth ofArchaeology. Penguin, London.
Davis, R.H., 1955 Deep Diving and Submarine Operations, 6th ed. Siebe Gorman, London.
de Weerd, M., and Haalebos, J.K., 1973, Schepen voor het opscheppen. Spiegel Historiael (Bussum) 8:
386-397.
Ehrenberg, V., 1967, From Solon to Socrates. Methuen, London.
Ellmers, D., 1973, The Earliest Report on an Excavated Ship in Europe. International Journal of Nautical
Archaeology and Underwater Exploration 2: 177-179.
Fenwick, V.H., 1972, The Graveney Boat. A Pre-conquest Discovery in Kent. International Journal of Nautical
A rchaeology and Underwater Exploration I: 119-129.
Fiori, P,. and Joncheray, J.-P., 1975, L'epave de la tradeliere. Cahiers d'Archeologie Subaquatique 4: 59-70.
Franzen, A., 1966, The Warship' Wasa. 'Norstedts, Stockholm.
Frondeville, G. de, 1965, Mahdia. Marine Archaeology, edited by J. du P. Taylor, pp. 39-52. Hutchinson,
London.
Frost, H. 1963, Under the Mediterranean. Routledge and Kegan Paul, London.
Green, J.N., 1973, The Wreck of the Vergulde Draeck, 1656. International Journal of Nautical Archaeology
and Underwater Exploration 2: 267-289.
36 K. Muckelroy

Green, J.N., 1975, The VOC Ship Batavia Wrecked in 1629 on the Houtman Abrolhos, Western Australia.
International Journal of Nautical Archaeology and Underwater Exploration 4: 43-63.
Green, J.N., and Henderson, G., 1977, Maritime Archaeology and Legislation in Western Australia. International
Journal ofNautical Archaeology and Underwater Exploration 6: 245-248.
Greenhill, 8., 1976, The Archaeology of the Boat. A. and C. Black, London.
HasslOf, 0., 1972, Maritime Ethnology and Its Associated Disciplines. In Ships and Shipyards. Sailors and
fishermen, edited by O. HasslOf, H. Henningsen, and A.E. Christensen, pp. 9--19. Copenhagen Univer-
sity Press. Copenhagen.
Henderson, G., 1976, James Matthews Excavation, Summer 1974; an Interim Report. 5: 245-251.
Ingelman-Sundberg, C., 1977, The VOC Ship Zeewijk Lost off the Western Australian Coast in 1727. Interna-
tional Journal of Nautical Archaeology and Underwater Exploration 6: 225-231.
Joncheray, J.-P., 1975, Etude de d'epave Dramont D; les objets metaIliques. Cahiers d'Archeologie Subaquatique
4: 5-18.
Joncheray, J.-P., 1976, Le Roche Fouras. International Journal of Nautical Archaeology and Underwater
Exploration 5: 107-114.
Katzev, M.L., 1974, Last Harbor for the Oldest Ship National Geographic 146: 618-025.
Kirkman, J., 1972, A Portuguese Wreck off Mombasa, Kenya. International Journal of Nautical Archaeology
and Underwater Exploration I: 153-157.
Lewis, N., and Reinhold, M., 1955, Roman Civilisation, 2 vols. Columbia University Press, New York.
Linder, E., and Raban, A., 1975, Marine Archaelogy. Cassel, London.
LyeIl, C., 1832, Principles of Geology, 1st ed., 3 vols. London.
McKee, A., 1973, King Henry VIII's 'Mary Rose. ' Souvenir Press, London.
Marsden, P., 1966, A Roman Ship from Blackfriars. Guildhall Museum, London.
Marsden, P., 1971, A Seventeenth Century Boat Found in London. Post-Medieval Archaeology 5: 88-98.
Marsden, P., 1972 The Wreck of the Amsterdam near Hastings, 179. International Journal of Nautical Archae-
ology and Underwater Exploration I: 73-96.
Marsden, P., 1974, The Wreck of the 'Amsterdam.' Hutchinson, London.
Martin, CJ.M., 1975, Full Fathom Five: The Wrecks of the Spanish Armada. Chatto and Windus, London.
Mayhew, D.R., 1974, The Defense; Search and Recovery, 1972-3. International Journal of Nautical Archaeology
and Underwater Exploration 3: 312-313.
MiiIler-WiIle, M., 1974, Boat Graves in Northern Europe. International Journal of Nautical Archaeology and
Underwater Exploration 187-204.
Olsen, 0., and Crumlin-Pedersen, 0., 1967, Skuldelev Ships, II. Acta Archaeologica 38: 73-174.
Osaki, E., 1973, Seventeenth Century Japanese Harbour Works. In Science Diving International, edited by
N.C. Flemming, pp. 66-69. British Sub-Aqua Club. London.
Pearson, C., 1976, Legislation for the Protection of Shipwrecks in Western Australia. International Journal of
Nautical Archaeology and Underwater Exploration 5: 171-173.
Poidebard, A., 1939, Un grand port disparu, Tyr. Bibliotheque archeologique et historique, Vol. 29, Paris.
Prynne, M.W., 1968, Henry V's Grace Dieu. Mariner's Mirror 54: 115-128.
Rice, WMcP., 1824, An Ancient Vessel Recently Found under the Old Bed of the River Rother in 1822.
Archaeologica 20: 553-565.
Smith, R.A., 1909, The Pudding Pan Rock, Heme Bay, Kent. Proceedings of the Society of Antiquaries 22:
395-414.
Stenuit, R., 1972, Treasures of the Armada. David and Charles, Newton Abbot.
Swiny, H.W, and Katzev, M.L., 1973, The Kyrenia Shipwreck; a Fourth Century B.c. Merchant Ship. In
Marine Archaeology, edited by DJ. Blackman, pp. 339--359. Butterworths, London.
Taillez, P., 1965, Titan. In Marine Archaeology, edited by J. du P. Taylor, pp. 76-92. Hutchinson, London.
Ucelli, G., 1950, Le navi di Nemi. La Liberia deIlo Stato, Rome.
van der Heide, G., 1976, Archaeological Research in the Zuider Zee. National Maritime Museum, London.
Vrsalovic, D., 1974, Istrazivanja i zastita podmorskih archeoloskih spomenika u SR Hrvatskoj. Republicki
Zavod za zastitu spomenika kulcture, Zagreb.
Weier, L.E., 1974, The Deterioration of Inorganic Materials under the Sea. Institute of Archaeology Bulletin
II: 131-163.
Weinberg, G.D., Grace, Y.R., Edwards, G.R., Robinson, H.S., Throckmorton, P. and Ralph, E.K., 1965, The
Antikythera Shipwreck Reconsidered. Transactions of the American Philosophical Society 55(3).
Introducing Maritime Archaeology 37

Wheeler, R.C., and van Gernert, R.C., 1972, Waterways Open the New World. In A History of Seafaring.
Based on Underwater Archaeology. edited by G.F. Bass, pp. 282-304. Thames and Hudson, London.
Wheeler, R.C., Kenyon, W.A. Woolworth, A.R., and Birk, D.A., 1975, Voices from the Rapids. Minnesota His-
torical Archaeology Series, No.3. Minnesota Historical Society, St. Paul.
Wheeler, R.E.M., 1954, Archaeology from the Earth. Oxford University Press, Oxford.
Zacharchuk, w., 1972, The Restigouche Excavation. International Journal of Nautical Archaeology and
Underwater Exploration 1: 157-163.
 Chapter 4

'.~~.' Integrating Archaeological

~

o i' .
and Historical Records
q;,O_" .': .
.j'~l in Dutch East India
Company Research

J. BAS KIST

The Dutch East India Company, vac, was founded in 1602 by the amalgamation,
under government pressure, of various overseas trade companies. It was a decentralized
company with branches in six important cities in the Netherlands. Each of these
branches had its own administration, warehouses and shipyards. The central direction
of the company consisted of a board of 17 directors, 8 of whom were appointed by the
City of Amsterdam.
Marketing of cream ware wreaked havoc in the pottery industries of England and
the continent.
The capital of the company was provided by private investors. The financial
structure of the enterprise was as complicated as its organization. Separate accounts
were kept in the different branches and in Asia. These accounts were kept in 4-year
cycles, while each year fleets were equipped and sent out as if each was a separate
undertaking. The returning ships and their cargoes were again dealt with separately by
each local branch of the company.
The 17 directors and also the directors of the local branches were expected to par-
ticipate with a sum of £6000 in the company. Later this sum was reduced to £3000 for
the directors of the smaller branches. Many private investors participated in the com-
Source: Underwater Archaeology Proceedings from the Society for Historical Archaeology Conference. edited
by Donald H. Keith and Toni L. Carrell (1982), pp. 5~57. Reprinted by permission of the Society for Historical
Archaeology, Tucson, Arizona.

39
40 J.B. Kist

pany, but they had very little real influence in the running of the company. Their acqui-
escence was assured by a yearly 12.5 percent dividend throughout the two centuries of
existence of the company.
The VOC built a trade network through large parts of Asia, based on ca. 250 trad-
ing stations. Many of these were just local offices where a representative of the com-
pany, aided by local personnel, engaged in import and export trade. In some cases, e.g.,
Ceylon, Malaya, and various places in the Indonesian archipelago, the company's set-
tlements included fortifications, garrisons, and territorial control. Personnel in Asia
numbered about 35,000.
The connections between Europe and the Asian network were maintained by a
fleet of oceangoing armed merchantmen, called retourschepen, while the inter-Asian
routes were plied with a great variety of Asian craft under the company's flag and also
with the company's large European ships. Between Holland and Asia, a total of 5000
ship movements have been counted during two centuries, while 3300 ships undertook
the return voyage. On the route between Europe and Asia and in Asia the company lost
about 250 ships in the two centuries of its existence.
The history of the Dutch East India Company has been written under the influ-
ence of different prevailing ideologies. Throughout the 19th and in the 20th century the
image of the Dutch East India Company was stamped by an imperialistic belief, a
belief in the value of colonies and dependencies. This ideology in Holland was mainly
inspired by the British example. Within this imperialistic attitude there was little appre-
ciation of the merits of the overseas trade company. In the second half of the 18th cen-
tury, when the English East India Company was changing from a trade company into a
colonial government, British propaganda helped to create a negative image of the
Dutch East India Company which had remained mainly a trade company. About 80 per-
cent of its turnover was derived from trade and no more than 20 percent came from
exploitation of rents, taxes, and such.
The negative appreciation of the role of the VOC as a trade company in Asia per-
sisted into the anticolonialist vision that developed after the Second World War. In the
eyes of the colonialists, the imperfections of the VOC had led straight to the estab-
lishment of colonial rule, which represented, in their view, a higher stage in develop-
ment. Although anticolonialists reversed this judgment, the linear view of the East
India Company and colonial history persisted for some time.
Recently, however, with the healing of the postcolonial trauma, there is a ten-
dency to look at the Dutch East India Company as an early example of an interconti-
nental manufacturing, trade, and transport company. Parallels on a management and a
technical level with modem business are obvious, and the question of survival of such a
large organization in changing technological, economic, and political circumstances
remains as relevant as ever.
In this climate of renewed interest the study of the Dutch East India Company
was given added impetus by the development of the underwater archaeology and in
particular the archaeology of shipwrecks of Dutch East Indiamen. Since 1965, about 50
wrecks of Dutch East Indiamen have been localized and excavated in Europe, Africa,
and Asia. Whatever the scientific standards in these various excavations may have
been, and they certainly varied from excellent to near zero, they all produced prodi-
gious amounts of hitherto unknown material. This material added something entirely
new to the records that had been available so far: unsorted material sources of informa-
Integrating Archaeological and Historical Records in Research 41

L Y S T E, 'VIII 'c gene tat

d'Equipagie behorende. en voort aan Amonitie van

Oorlog is melle gegeyea. UD 't Scbip gmumt

daar Schipper op is
de Ncdcrlandli:be Geo&oyeerdc OoA:-lndifchc Compgnic torhe-
hoorcndc, en dat om pertinenle aanlcykening gehouden Ie worden
van aile 'I geae dar YlD iadienft, YlodegcmellcComt-gnicword
geconfumccrt, "crbruykt, geRctea, ook wei op de rcyze komI te "cr-
pn, of ~ d'cea. of d'ander gclegenlheit weg IC naken, en dlc
volgc", de ordrc co InltruCtic daar van zyndc. bier '001'
lID gedrultt.

Te A M S T E R DAM.
By N leo L A A S 8 Y L, Drukker nn d.
0011: -Indifche Compagnie.
Mel PrwileKie. 178B.
Figure 1. Equipage list for the Dutch East Indiaman (Amsterdam) from the company archives.

tion found within the strict context of a particular ship. This material was produced in
great quantities and could be identified and classified with the aid of other historical
records in the archives of the Dutch East India Company. Thus, object-oriented
research in the archives helped to identify artifacts, but it also helped to select records
with a particularly heavy load of material information, such as inventory lists (equi-
page/ijs!) (Figure 1), lists of victuals and bookkeeping records, which previously had
only been studied on a more abstract level by economic historians.
The further study of the administration of the company gave a better insight into
the production side of the enterprise. The company possessed some of the largest prein-
dustrial complexes in the Netherlands in its shipyards in Amsterdam and Middelburg. It
also employed a multitude of subcontractors for the supply of things that could be
bought cheaper on the open market than produced. The complex relationships of the
company with its suppliers, but also with its work force, have become an important
new subject for study.
A multidisciplinary group of historians, archivists, archaeologist, and museum has
been engaged in the study of the Dutch East India Company's shipyards in Amsterdam
42 J.B. Kist

... - _... _...

............ ....,..... .............

................ _... 10 ao ..

~ C ..._ ltt) ~ ' ~I.I ... ~wwy

A'
-- --
~
1. -....-
...
........ ..........
...... . Cit __ ... , J ""'" . . . . . ""'-Mr.
--r-.
. ._AnIslw ....
.-.
Figure 2. Plan map of the Amsterdam wreck site.

and Batavia, the present-day Jakarta. The integration of the historical and archaeological
record has led to some publications but also to the construction of a literal model.
The renewed interest in the ships of the Dutch East India Company led to two
full-scale replica projects in Holland, one of a 17th-century East Indiaman, Batavia, in
Lelystad, and one of an 18th-century East Indiaman, Amsterdam, in Amsterdam. In
.both cases, the actual building of the ship had already started before thorough research
in historical and archaeological records became possible.
The very special situation existing in present-day Dutch East India Company
research, with the availability of very detailed archival records, the ever increasing
amount of archaeological records and finally the availability of material records, such
as contemporary ship models, drafts, and nontechnical images, allows a detailed scru-
tiny of historical material reality.
An example of such detailed scrutiny is the study of Hollandia 's artifacts. Hol-
landia was wrecked in 1743 off the Isles of Scilly. After initial efforts by the Dutch
East India Company to retrieve at least the treasure aboard, the wreck was forgotten
until it was rediscovered by Rex Cowan in 1973. Over the years, a considerable number
of objects were raised from the wreck site and studied and reported on by archaeolo-
Integrating Archaeological and Historical Records in Research 43

L Y S T E
V.n d.

VICTUALIEN, E.a onlrc op de

RANTSOENEN.
Ter vergaderinge van de Heeren Sevcnticne, rcprefcncercnde de
Gcneiale Nederlandfchc Geofuoyecrde Ool1:·lndifche Com-
. Amllerdam gcarrc_
pagnie , op den 1 ~d•• December ,6, I . tor
lIeerr, dXlc lI1ne volgcns opgevolgde Refolurien, vall den 1 ')
Juny 1 ;0.', •• July '706, 1 811p~~L~12J.' It.~ July '1:.' ••
7 Augulh '7 ' 3. en 1 + Tuny '7. +,. ccOlge ver:uiueringen,.zyn
gcmJ.2kr, volgens dewc"lke voorwn ;tile de K.uneren vm de
geOldte Compagnie op cen cenparige vocr de
JchtpeJJ 'zo groote al; k/eyne, Flllykn,
JachIm, (fJ kleynder Schepcn.
NJ,. Ooil·/ndicn uyrgcrull: lIferdende, zullen Provid""reii; en
;"JJr nlJr de Opperhoolilen en Ollicieren v~n de fclve, in
r uyrdcylen, en dtl!nbucren Vln de RandlOc:nen, haar ltdlen
!.ebbell to rc,;ulecren, en dar om in de voorlZ. Schepen een
ecnplrige vO(r en oedr. tc houden.

Figure 3.. List of victuals and regulations

for their distribution from the Dutch East ,'A M S T E R 0 AM.
India Company Archives. 8y R. en G. \V ETf T.',.. Drukkers ..n-d·~.. E. Heeren newinuhtb.
be.en .Ie. Ooll·lndllch. Comp.gni•.

gists Peter Marsden and Howard Pell, as well as by Rex and Zelide Cowan (Cowan and
Cowan, 1975).
Many of the recovered artifacts were sold at auction and bought by museums and
private collectors. What remained, about 3000 objects, were acquired by the Rijks-
museum in Amsterdam together with the archaeological records. Because the objects
were spread over a large surface of the seabottom by the sheer violence of the disaster,
little could be deduced from their context (Figure 2). It proved useful to develop
another method to bring order in the raised objects.
After analysis of the material and manufacturers of the objects, they were ordered
in a functional system which was then compared with a similar functional system
derived from historical East India Company records, in this case, the equipage/ijs! (list
of equipment issued to each vessel of the Dutch East India Company), the Lyste van de
victualien en ordre op de rantsoenen (list of victuals and .regulations for the distribu-
tion of victuals) (Figure 3) and the Ordre en Instructie voor de Chirurgijns (orders and
instructions for the ship's surgeons). The historical records also provide the nomencla-
ture of the company.
44 J.B. Kist

These two functional systems could be compared and the attribution of historical
terminology allowed further reference to the company's bookkeeping providing infor-
mation about the manufacturers, quantities and prices of equipment, and also the differ-
ence between equipment goods and cargo.
This two-pronged historical-archaeological method was applied on a much larger
scale in the Amsterdam project. The East Indiaman Amsterdam was beached during a
strong southwesterly gale in 1749 near Hastings; the ship settled in deep mud in a very
short time and remained more or less inaccessible until 1969. In that year the wreck
was damaged by a contractor engaged in the construction of a sewage system. After an
exploratory excavation by Peter Marsden in 1969 showed the great promise of the
wreck, nothing much happened until 1984 when the VOC ship Amsterdam Foundation
succeeded in mounting a combined Anglo-Dutch underwater excavation. Two more
seasons of excavation followed in 1985 and 1986.
From the start in 1984, the Amsterdam research project was conceived as a his-
torical-archaeological project. Historical hypotheses concerning shipbuilding, equip-
ment, and functional arrangements on board the ship served as a starting point for
archaeological research. The differences between the model built from historical
records and the real-life data from the archaeological situation were further investi-
gated to arrive at a more detailed and comprehensive reconstruction. This involved par-
ticipation of a variety of specialists who helped to solve many problems but who also
asked many new questions. In this way, problems were dealt with again and again from
different angles, creating very detailed reconstructions. These reconstructions can first
be used to write the material history of the Dutch East India Company and then,
because the ships were the principal tool of the company, it is through them that proce-
dures and practices of the company can best be studied.
In the case of the Amsterdam, a considerable amount of technical documentation
was found in the East India Company archives because the company went through a
process of technical and procedural innovation in the period 1740-1750. Not only was
the design of the standardized East Indiamen completely renewed by ordering a set of
drafts from an English shipwright, which incorporated the latest ideas from Holland,
England and France, but the sailing regulations and operation procedures were all
rewritten in the light of contemporary expert opinion.
The archival material comprised drafts, reports, and resolutions and could be ampli-
fied with some very detailed technical models which were made for the company in the
course of the design process. Thus, it was possible to start the excavation of the East
Indiaman Amsterdam, working from a highly detailed theoretical model derived from a
wide variety of sources. Almost immediately the excavation of the wreck showed discrep-
ancies between the historical and the archaeological data. In some cases entirely new ques-
tions were raised about the way the ship was actually built, questions that would never
have been asked working from one category of sources alone. On the level of functional
and procedural matters the interaction between historical and archaeological records
proved particularly fruitful. In this case, iconographic and written sources, such as orders
and instructions for ship's officers, could be combined to create a hypothetical image of a
situation or a procedure on the ship, such as the treatment of the sick, or the arrangements
in a given compartment of the ship for the distribution of food. Then these hypotheses
could be tested in the archaeological situation and later often expanded because the com-
parison of the various records nearly always led to a much more detailed picture.
Integrating Archaeological and Historical Records in Research 45

REFERENCES
Blusse, L., and Gaastra, F.S. (eds). 1981, Companies and Trade. Leiden.
Bruijn, J.R., Gaastra, F.S., and SchOffer, I., 1979, Dutch Asiatic Shipping (3 vols.). The Hague.
Cowan, R., and Cowan, Z., 1975, The Dutch East Indiaman Wrecked on the Isles of Scilly in 1743. Interna-
tional Journal of Nautical Archaeology 4 (2): 267-300.
Gaastra, F.S., 1982. De Geschiedenis van de VOe. Bussum.
Gawronski, J., 1992, Hollandia Compendium: A contribution to the history. archaeology. classification. and
lexicography of a 150ft. Dutch East Indiaman. Elsevier, Amsterdam.
Gawronski, J. (ed.), 1984, Annual Report VOC Ship Amsterdam Foundation. Amsterdam.
Gawronski, J. (ed.), Annual Report VOC Ship Amsterdam Foundation. Amsterdam.
Gawronski, J. (ed), 1986, Annual Report VOC Ship Amsterdam Foundation. Amsterdam.
Van Rooij, H.H., and Gawronski, J., 1989, East Indiaman Amsterdam. Amsterdam.
 Chapter 5

A Discussion of
".~i).:
~

Maritime Archaeology
-r/: ..
':.i
PAUL FONTENOY

The advent of practical scuba technology in the 1940s profoundly impacted

archaeology. Free diving techniques opened up a vast range of sites to excavation with
a degree of precision that had previously been impossible. At the same time the possi-
bilities of this new technology posed fundamental questions - methodological, ethical,
and philosophical - for archaeology, academia, the private and public administration
of the arts, and, ultimately, for public opinion to answer. These problems were pre-
sented in their starkest forms during the first 20 years after the availability of scuba
technology, the community at large sidestepped the challenge, and, as a consequence,
even today few of these issues have generated entirely satisfactory solutions.
Sites pertinent to the subdiscipline of maritime archaeology exist both on land
and underwater. Clearly, excavation of land sites - vessels themselves and the struc-
tures and manufactories associated with maritime activities - requires no significantly
different techniques from those generally applicable to terrestrial archaeology as a
whole. Work on underwater sites requires both the use of diving technologies of one
type or another and the development of techniques to permit the use of general contem-
porary archaeological methodology and produce comparably satisfactory results. A
third category of sites exists, best classified as submerged. These are either partially or
wholly underwater, and either underwater or terrestrial techniques may be employed;
the latter often employing cofferdams and pumps to render the site dry. It is important
to realize that the distinction between various categories of sites is between the appro-
Source: Ms on file, Ruppe Library (1994), Program in Maritime History and Nautical Archaeology, East Caro-
lina University, Greenville, North Carolina. Reprinted by permission of the author.

47
48 P. Fontenoy

priate techniques to employ for access, rather than between the actual methods to apply
to their excavation.
Maritime archaeology did not emerge as a separate subdiscipline without drawing
from preexisting theoretical paradigms. Its development, however, was far from
methodical and depended, in part, on the creative legacies of a group of powerfully
influential individuals during the course of its early evolution.
Technological developments were crucial to the evolution of maritime archaeol-
ogy. Theoretically, there were no serious barriers to the application of standard contem-
porary archaeological techniques to every maritime site prior to the 1940s. In practice,
two elements militated against the successful development of the field. The limitations
of available technologies, such as diving bells and helmet diving, for accomplishing
underwater excavations were a serious impediment, but the restrictions these very limi-
tations imposed on the available range of sites were even more significant. Without a
sufficiently wide-ranging universe of maritime sites there could be no valid inde-
pendence for maritime archaeology.
In the mid-1940s Jacques-Yves Cousteau and his associates developed the equip-
ment that first permitted successful free diving. During the succeeding decade they
concentrated their efforts toward expanding the operational envelope of their equip-
ment. One of the arenas within which this was accomplished was underwater archaeol-
ogy. Underwater excavations, even when carried out under the supervision of qualified
professional archaeologists, as at the Grand Congloue site, were viewed as tests, not for
archaeological techniques, but for the requisite diving technologies. Cousteau's contri-
bution to the subdiscipline's evolution was an advance in technology rather than a dis-
ciplinary development (Cousteau, 1954; Diole, 1954:4~56).
One of the most remarkable early underwater archaeological excavations was that
undertaken at Fos-sur-Mer by Dr. Rene Beaucaire from 1948 onwards. The site itself
was a Roman port of the second century B.C. which was partially inundated. Beaucaire
carried out excavations on land and underwater, diving himself and working with a
team of volunteers who had previously been engaged in digging at prehistoric sites in
Provence under his supervision, and had learned free diving techniques to carry out the
work at Fos-sur-Mer. All of the techniques of terrestrial archaeology - a preliminary
survey followed by detailed mapping of the site, planned methodical selection of areas
to be uncovered, attention to stratigraphy, meticulous conservation of precisely located
and identified artifacts, effective exhibition of suitable finds, and careful documenta-
tion and publication - were features of this extraordinary undertaking, which appears
to have been almost entirely ignored outside France. Although not nautical in nature,
this was maritime archaeology carried out to the highest standards, and it is surprising
that Beaucaire's accomplishment has been consigned to obscurity (Beaucaire, 1964;
Diole, 1954:94-109).
Peter Throckmorton's contribution to maritime archaeology was twofold: an
appreciation of the multidisciplinary nature of the field itself, and application of his
professional journalistic skills to mobilize effective support for work on significant
sites which otherwise were likely to be ignored. Throckmorton, perhaps because of his
amateur status, was among the first practitioners to appreciate the importance of sur-
veying a wide geographic range of vessels of similar type, and his professional training
allowed him to comprehend the powerful influence of governmental support for such
epochal excavations as that at Kyrenia (Throckmorton, 1977).
A Discussion of Maritime Archaeology 49

George Bass's most significant contribution to maritime archaeology was his

accomplishment at Cape Gelidonya. This site, off the coast of Turkey, was the first
widely publicized demonstration that maritime archaeological excavation required no
compromises from the professional standards set by terrestrial archaeologists. This
excavation defined the parameters of much subsequent work; it was rigorous and com-
prehensive, but, at the same time, it was viewed as a single event disconnected from
other undertakings (Bass, 1961).
Cousteau, Throckmorton, and Bass were all pioneers in the development of a new
field. Cousteau's contribution remains significant in the technology he developed and
refined, Throckmorton's legacy is his excavations and the commitment of eastern
Mediterranean governments to the preservation of their underwater heritage, and from
Bass comes our modem dedication to an uncompromising stance for rigorous archae-
ological methodology. The particular legacy of Cousteau and Bass, however, also
includes maritime archaeology'S obsession with the techniques of underwater excava-
tion, and its tendency toward viewing sites in isolation.
There is good reason to believe that most of the U.S. maritime archaeological
effort has been devoted to undertaking thorough site-specific investigations whose
findings have not been integrated into a broader context but have been assessed as iso-
lated events. A cursory survey of thesis topics presented at East Carolina University
and Texas A&M during the past decade probably forms the most valid basis for reach-
ing this conclusion, although more recent titles indicate that a more rigorous theoretical
paradigm has come to the fore.
While this narrow view has been characteristic of maritime archaeology as prac-
ticed in the United States, it is also apparent that similar limitations are present
throughout the subdiscipline at a global level. There are a number of notable exceptions
in a very few areas, most notably medieval and postmedieval maritime archaeology in
Europe, and the Dutch East India Company on a worldwide basis (see particularly
McGrail, 1987, and Gawronski et aI., 1992).
There are a number of factors that could explain this phenomenon. One signifi-
cant element has been the dominance of technology, engendered by the perceived diffi-
culties associated with access to underwater sites for excavation. This has generated
intellectual confusion within maritime archaeology so that its unique field of interest
has been defined by the underwater location of its sites rather than their place within a
past maritime culture. Maritime archaeologists need to appreciate that it is the subject
matter of sites that is distinctive, not that diving is necessary to reach them.
Associated with this confusion is a reluctance to undertake the creation of a rigor-
ous theoretical framework for the field itself. Some notable efforts in this direction
exist, particularly Keith Muckelroy's Maritime Archaeology, published in 1978, but
this is exceptional as a comparison with Jeremy Green's 1990 work with the same title
clearly reveals. Muckelroy devotes approximately 25 pages of a 250-page book to the
technology of maritime archaeology; Green employs 5 pages of a work of similar
length for his entire discussion of the field's theoretical paradigm. Neither of these
authors is American: the problem is near universal.
A third element may well be maritime archaeology'S relative academic stature.
The subdiscipline has needed to acquire the respectability conferred by scientific meth-
odology, uncomprising application of professional techniques, and academic publica-
tion. This has been a particularly difficult problem in the United States, related to
50 P. Fontenoy

maritime archaeology's early links with searchers for historical relics, which ruined its
reputation within archaeology for many years (Goggin, 1960; Peterson, 1962).
The field's weakness in creating a rigorous theoretical foundation has been exac-
erbated by its limited input from other related disciplines. History and anthropology
have contributed substantially to its mentality but few maritime archaeologists have
drawn on the resources of other fields. It is noticeable that the two areas that have pur-
sued a broader theoretical foundation for their researches have also embraced multi dis-
ciplinarianism most widely.
Recent trends indicate that maritime archaeologists are becoming aware of many
of these limitations, and addressing and correcting them. The success of these moves is
essential- the field's academic standing is relatively unimportant, but its contribu-
tions assuredly will be diminished if they fail.
One area in which a broad multidisciplinary approach using a wide-ranging com-
parative methodology incorporating rigorous inclusive survey of similar sites at its core is
the archaeology of the European medieval maritime world. Very distinctive features char-
acterize medieval maritime archaeology's physical and intellectual frames of reference.
The seas, rivers, and littoral of northwestern Europe and the Baltic form the
field's preponderant geographic parameters. The medieval maritime experience of the
Mediterranean has received far less attention.
Medieval maritime archaeological methodology strongly emphasizes broad sur-
vey techniques - much stress is placed on the location, excavation, and analysis of
vessels of similar type.
Experimental archaeology is regarded as valuable and benefits from concentrated
attention. Reproductions, using scale models, computer modeling, and full-size repli-
cas, are considered tests to be supplanted by more refined models incorporating the les-
sons of earlier generations. This experimental paradigm encourages ready reassessment
and reevaluation of previous work in the light of late knowledge.
The product of the study of the medieval maritime experience is thoroughly
multi-disciplinary, incorporating the contributions of ethnographers, archaeologists,
naval architects, anthropologists, and economic, social administrative, political, and
military historians. This interaction facilitates efforts to identify and demonstrate con-
tinuing themes in design development, construction techniques, and technological evo-
lution - placing them within a rigorous contextual theoretical framework.
The physical and intellectual frames of reference of practitioners of medieval
maritime archaeology are highly distinctive. Although some, ifnot all, features are pre-
sent in varying degrees among other branches of the field, this area's particular combi-
nations and emphases are unique.
One immediately noticeable feature is the branch's geographical concentration.
The focus of interest, and virtually all work is concentrated in northwestern Europe and
the Baltic, in this area's seas, rivers, and littoral. In recent years this area of concentra-
tion has been expanded because of the breakup of the Soviet Union and its satellite sys-
tem, resulting in much fresh work being undertaken in Poland and the Baltic states,
which has expanded the existing strength of the field in Sweden, Denmark, and north-
ern Germany.
Little similar serious work has been undertaken in the Mediterranean apart from
some concentrated efforts focused on Byzantine sites to the east. These are viewed as a
limited part of the Classical continuum, which tends toward a historical emphasis.
A Discussion of Maritime Archaeology 51

The major distinctive characteristic of medieval maritime archaeology is its focus

on broad survey. Much effort is concentrated on discovering the extent of the field's
universe through uncovering a multitude of sites relatively closely related to each other
in type or date. Even by 1982 there were over 40 sites known and at least partially
excavated dating to the period from the 8th to 12th centuries, and another 20 from the
succeeding 200 years (McGrail, 1984:30-31). The last decade has seen a veritable
explosion of discoveries which has tripled these numbers.
The significance of this phenomenon lies in the validity of results generated
through the systematic analysis of large numbers of similar sites. Only through such
analysis is it possible to ascertain the characteristics that define normality, and to com-
prehend deviations from such norms. This distinctive broad-gauge approach is in
marked contrast to most others visible in maritime archaeology, and allows its practi-
tioners to postulate valid generalizations.
A second major emphasis of medieval maritime archaeology is its truly multidis-
ciplinary embrace. The most rigorous participants demand the inclusion of ethnogra-
phy, all branches of history, anthropology, naval architecture, and conservation, wood,
engineering, dating, environmental, and computer science within the range of disci-
plines contributing to the archaeology of medieval vessels. The vessels themselves are
viewed contextually: medieval maritime archaeology embraces not simply the con-
struction of ships but their entire operational environment (McGrail, 1984:37-38).
Finally, medieval maritime archaeology is distinctive for its reliance on experi-
mental methods for validation; "floating hypotheses" in Sean McGrail's terms. Here
again, a rigorous methodology has been developed, starting from creating a workable
model, and progressing to second-generation reconstructions that are built using repli-
cations of contemporary technologies and techniques. Closely allied to this genera-
tional approach is the unusual willingness of archaeologists in this particular field to
readily reevaluate previous initial conclusions in the light of subsequent discoveries or
analyses.
It is difficult to attribute this distinctive approach to any single factor. The likeli-
est explanation for its development is the wide base of support engendered by its mul-
tidisciplinary nature and its close connection to ethnography and folkways which have
substantial nationalistic or patriotic appeal. The end result, however, is the contextual
appreciation of medieval maritime life.
A parallel trend toward the rigorous inclusive search for related sites in order to
ascertain the normal through comparative study has emerged in two other areas of
maritime archaeology: postmedieval archaeology, particularly in the Baltic, and the
archaeology of the Dutch East India Company. In both of these areas one major con-
centration of effort has been the identification of the relevant population of sites
through rigorous analysis of archival data and wide-ranging survey of the physical en-
vironment. Consequently, those working in these areas will be able to draw more valid
and useful conclusions from their discoveries, since they will have solid baselines for
comparative analysis (see particularly Gawronski et ai., 1992, and Cederlund, 1983).
Maritime archaeologists' continued fascination with individual sites as phenomena
to be analyzed in isolation generates two undesirable consequences: it condemns them to
produce incomplete and inadequate results because they are unable to incorporate their
findings into a contextual continuum, and it perpetuates the image of the subdiscipline's
practitioners as lacking professionalism and rigor, and virtually indistinguishable from
52 P. Fontenoy

antiquarian treasure seekers. Nautical archaeology has the potential to transform the his-
torical study of maritime culture - the value of its findings will be in proportion to the
rigor of its theoretical paradigm.

REFERENCES

Bass, G.F., 1961, The Cape Gelidonya wreck: preliminary report. American Journal 0/ Archaeology 65:
267-276.
Beaucaire, R., 1964, Les fouilles sous-marines de Fos. Provence Historique XIV: 16-25.
Cederlund, c.-O., 1983, The Old Wrecks o/the Baltic Sea: Archaeological Recording o/the Wrecks o/Carvel-
built Ships. BAR, Oxford.
Cousteau, J.-Y., 1954, Fish men discover a 2,200 years-old Greek ship. National Geographic Magazine 105:
1-36.
Diole, P., 4,000 Years under the Sea. Julian Messner, New York.
Gawronski, J., Kist, B., and Stokvis-van Boetzelaar, 0., 1992, Hollandia Compendium. Elsevier, Amsterdam.
Goggin, J.M., 1960, Underwater archaeology: its nature and limitations. American Antiquity 25: 348-354.
Green, J., 1990, Maritime Archaeology. Academic Press, San Diego.
McGrail, S., (ed.), 1984, Aspects 0/ Maritime Archaeology and Ethnography. National Maritime Museum,
London.
McGrail, S., 1987, Ancient Boats in N. W. Europe. Longman, London.
Muckelroy, K., 1978, Maritime Archaeology. Cambridge University Press, Cambridge.
Peterson, M., 1955, The last cruise of H.M.S. Loo. Smithsonian Miscellaneous Collections 131.
Peterson, M., 1962 History under the Sea. Smithsonian Institution, Washington, DC.
Rule, M., 1982, The Mary Rose. Conway Maritime Press, London.
Throckmorton, P., 1977, Diving/or Treasure. Viking Press, New York.
Throckmorton, P., (ed.), 1991, The Sea Remembers. Smithmark, New York.
 Additional Bibliography
".~i).:
~

for Part I.A

-r/: ..
':.i

Bass, George F., 1967, Cape Gelidonya: A Bronze Age Shipwreck. Transactions ofthe American Philosophical
Society Volume 57 (8), Philadelphia.
de Borhegyi, Suzanne, 1961, Ships, Shoals and Amphoras. Holt, Rinehart and Winston, New York.
Burgess, Robert F., 1970, Sinkings. Salvages, and Shipwrecks. American Heritage Press, New York.
Cotter, lohn L., 1978, Symposium on Role of Archaeology in Historical Research, Summary and Analysis.
Historical Archaeology: a Guide to Substantive and Theoretical Contributions, edited by Robert L.
Schuyler, pp. 18--19, Baywood Publishing Company Inc., New York.
Diole, Philippe, 1954, 4. 000 Years under the Sea. luI ian Messner, New York.
Falcon-Barker, Ted, 1963, 1600 Years under the Sea. David McKay Company, New York.
Finley, M. I., 1971, Archaeology and History. Daedalus 100: 168--179.
Fish, Carl Russell, 1978, Relation of Archaeology and History. Historical Archaeology: a Guide to Substantive
and Theoretical Contributions, edited by Robert L. Schuyler, pp. 8--10. Baywood Publishing Company
Inc., New York.
Goggin, lohn M., 1960, Underwater Archaeology: its nature and limitations. American Antiquity 25: 348-354.
Grosset, Harry, 1954, Down to the Ships in the Sea. 1.B. Lippincot, New York.
lones, Tom 8., n.d., Archaeology and History. Contributions to Aegean Archaeology: Studies in Honor of William
A. McDonald. Center for Ancient Studies, University of Minriesota, Minneapolis.
Kuhn, Thomas, 1970, The Structure of Scientific Revolutions. University of Chicago Press, Chicago, Illinois.
Lipke, Paul, Peter Spectre and Benjamin A.G. Fuller (eds.), \993, Boats A Manualfor their Documentation.
American Association for State and Local History, Nashville, Tennessee.
Marx, Robert F., 1990, The Underwater Dig. Introduction to Marine Archaeology. Pisces Books, Houston,
Texas.
Schuyler, Robert L., 1978, Historical and Historic Sites Archaeology as Anthropology: Basic Definitions and
Relationships. Historical Archaeology: a Guide to Substantive and Theoretical Contributions, edited by
Robert L. Schuyler, pp. 27-31. Baxwood Publishing Company Inc., New York, 1978.
South, Stanley, 1977, History and Theory in Historical Archaeology, Academic Press, New York.
South, Stanley, I 977a, Research Strategies in Historical Archaeology. Academic Press, New York.
Taylor,w.w., 1968, A Study of Archaeology, Southern Illinois University Press, Carbondale, Illinois.
Thomas, David H., 1990, Archaeology, Holt, Rinehart and Winston, Chicago, Illinois.
Trigger, Bruce G., 1989, A History of Archaeological Thought. Cambridge University Press, New York.

53
 Part I.B

'.~~.' The Secular Debate

~

o i' .
q;,O_" .': .
.j'~l

This section presents differing viewpoints about who should conduct archaeological
investigations in the maritime environment. While academic professionals wish to pre-
serve a nonrenewable resource, amateurs feel they have a right to investigate sites they
have found, and salvors point out that they have rights to recover materials for com-
mercial purposes. These interest groups reflect the diversity of underwater workers but
do not solve the problem of who, ultimately, is allowed to recover information. As tech-
nology for underwater work becomes more sophisticated, previously untouched sites
are subjected to investigation, often at great cost. Who is entitled to fund, conduct and
report on this work?
In Chapter 6, Shiner offers an excellent short informal look at the origins of the
amateur underwater archaeologist movement in the United States. He suggests sport
diving associations (PAD!, NAUI) be designed to funnel trained workers to underwater
projects along the same lines as Europe's British Sub-Aqua Club. He clearly states that
the divers would be under supervision of a senior archaeologist, and that they would
comprise an important and interested resource base for labor.
Amer and Steen, in Chapter 7, present the most successful state program incorpo-
rating sport divers within a state program for recovering underwater archaeological
data. This represents a first for the more conservative of the academic community.

55
 Chapter 6

Underwater Archaeology,
".~i).:
~

European versus American

':.i
-r/: ..
JOEL L. SHINER

There do exist some differences between the general methods and techniques of marine
archaeology as practiced in Europe and the United States. I am more familiar with the
current scene in Britain and the Republic of Ireland, so I will use that area as a point of
departure. The apparent, but not terribly important, preference for the use of metal
detectors instead of magnetometers, of solo divers instead of the buddy system are well
known. It would serve nothing to pursue these trivia. My own impression is that there
are three important differences that merit some discussion in depth. It is obvious that
Americans manage to acquire and use more and better equipment. On the other hand,
the British have the National Maritime Museum at Greenwich with its models, docu-
ments, paintings, and relics. Nothing in the New World can touch it. The British also
have the Sub-Aqua Club(s) which have no counterpart here. Our organizations such as
NAUI, PADI, and YMCA stress, in addition to safe diving, some concern for clean
water, coral, and fish. As regards our cultural heritage in wrecks or submerged sites
there is a profound ignorance at all levels and ranks. The British have a strong tendency
toward scholarly research on the part of amateur scientists, and a reverent respect for
the past. American amateurs, generally, are interested in hobbies.
A similar situation existed prior to 1930 in U.S. terrestrial archaeology. Arrowhead
collectors and "pothunters" threatened to devastate all of the prehistoric sites in North
America. This disaster was averted by the formation of state and local societies dedicated
to serious studies rather than souvenir collecting. The active participation of a few pro-
Source: Beneath the Waters of Time; The Proceedings of the Ninth Conference on Underwater Archaeology.
edited by 1. Barto Arnold III (1978), pp. 199-204. Texas Antiquities Commission, Publication no. 6, Austin,
Texas. Reprinted by permission of the Texas Antiquities Commission.

57
58 J.L. Shiner

fessionals within these societies improved the opportunities for education and conserva-
tion. There were many alarmed scientists who believed that we were simply arming col-
lectors with better weapons for looting sites, but this did not occur at all. Today these
societies are making real contributions in salvage, research, and conservation.
The danger to underwater sites obviously comes from two sources: industrial
needs and sports divers looking for treasure. Again, it is a problem for widespread edu-
cation and organization. My belief is that the solution lies in offering advanced diver
education in the form of a course in underwater archaeology for amateurs. Two aims
would be served by such a program. First, the diver would be made acutely aware of
the scientific values of the wreck and the odds against being able to find a treasure
trove of gold, silver, and gems. Hopefully he might come to look on an old wreck as he
would a whale, a redwood tree, or a living reef. Second, he would acquire introductory
training in the tools of underwater archaeology, that is, the artifacts, the air lifts, map-
ping, cataloging, preliminary preservation, and the like. He would thus be eligible for
employment as a working diver in an underwater excavation. I think that this would be
the most exciting reward for the newly trained archaeology diver: the opportunity to
participate in a scientific investigation of an important historic ship.
At this point I would like to present the major message of this paper. I propose to
ask one or more of the national associations such as NAUI or YMCA to create a new
certification. Just what the precise name might be is yet to be decided, but it will be
something like "Archaeological Diver.". More important than the name is the set of
standards. They require that the candidate be instructed in the essential aims and pur-
pose of archaeology, especially as regards the value of all data, not just the whole
pieces, not the gold, not the souvenirs. The ethics of recovery are the most important
factors; everything belongs to the public and nothing to any individual.
The other areas are subordinate. The diver should be taught how to carry out the
everyday tasks of a crew member on an archaeological excavation. These include alter-
native techniques of removing overburden, of setting grid systems, of tagging artifacts,
of keeping field catalogs, of the nature and appearance of prehistoric and historic arti-
facts. A less intensive introduction to remote sensing, strobe photography, initial arti-
fact preservation would be presented. In addition he would learn to use these tools and
this information under conditions of poor visibility. He would become acquainted with
the procedure and results of several significant underwater projects, both good and bad.
This training would not be intended to turn out an underwater archaeologist capa-
ble of working alone or in charge of a group. It is designed to produce a working diver
who could fit into a team which, in turn, was led by a senior archaeologist. Even if the
graduate did not seek employment in archaeology, he would be invaluable in crusading
for conservation among his fellow divers, students, or friends.
QUESTION: I just thought that I'd point out that the Association of Canadian
Underwater Councils has an archaeological guide book.
ANSWER: Do they have a constitution and so forth that we could copy in this
country?
QUESTION: Yes.
ANSWER: Tremendous! I'm very glad to hear it. Perhaps that could lead directly
to communication with YMCA and with NAUI.
QUESTION: PAD I also has a course that is set up to be taught like that. It's never
been taught that I know of.
Underwater Archaeology, European versus American 59

ANSWER: Well, during my long long association with PADI I've never heard
them mention the things that I felt were most important. Their wreck diving course and
their other things - perhaps it could be just the individuals I've talked to, but all the
way up through the instructors and what they call their "super-instructors" or master
instructors I heard really very little mention of the ideals of pure science. Perhaps they
were there, but they certainly were not talked about.
QUESTION: Why do you expect to find them in NAUI or YMCA?
ANSWER: Because NAUI and YMCA are noncommercial, nonprofit.
QUESTION: It's not the organization. So much depends on the individual instructor.
ANSWER: Exactly, it depends very much on the individual instructor, but the lit-
erature you get, the training you get, in one program stresses certain things above oth-
ers, and I really don't...
QUESTION: ... the training instructors to be ...
ANSWER: No not the instructor. I would prefer to draw if possible instructors
from the universities, trained in underwater archaeology, who would in turn train the
advanced divers not to be leaders of expeditions, because I think those people would
need to come out of anthropology, but to be crew member divers. That's what I'm sug-
gesting and that recognition be given by a national certifying group.
QUESTION: Your hope is that the normal scuba diver and the sport diver are moti-
vated simply because they like to be a good guy and that they would enroll in a course
like that and pay money for it?
ANSWER: Right and I think it would work because after some five years of teach-
ing underwater archaeology at a university, I suggested that such a course be taught for
nonuniversity people, and I was swamped. There were more than 50 applications out of
the Dallas area alone. I think they were all serious.
QUESTION: You don't suspect that they are just trying to learn some techniques to
improve their looting?
ANSWER: I hoped you would ask that because the same thing was asked of the
organizations in 1930 when the amateurs decided to get together and teach land archae-
ological techniques to each other, and it simply did not happen.
QUESTION: Well, it sure happened in Wisconsin.
ANSWER: Well, I'm sure we could find exceptions but over the entire nation the
organization of the amateurs into state and local groups did not result in looting - not
nationally - it resulted in amateurs becoming involved in science.
QUESTION: Dr. Shiner, what is the European method you seem to stress?
ANSWER: Perhaps it is the same thing that has motivated Europeans to become
tremendously interested amateurs in such things as gemology, astronomy, history, and
things. Whereas American amateurs rarely go that far. I don't know what the reason is.
I do not know the psychology of it, but it seems to happen.
QUESTION: I declare my interests as I am the chairman elect of the British Sub-
Aqua Club starting in March of this year. I'm also a member of the Committee on Nau-
tical Archaeology in Britain and the question asked by the last speaker as "What is the
European method?" can be answered very briefly. I'm not really answering for archae-
ology but simply about the organization of divers, and I'm not stressing [for] one
minute that it is actually superior or that it could be transplanted in anyway to the USA
or Canada. I'm simply describing it as natural history. The point is that we have diving
organized in most European countries on a consumer basis. The basic structure is div-
60 J.L. Shiner

ing clubs, clubs federated nationally, and I don't mean instructors. I mean divers are a
stronger political force than the diving instructors or equipment manufacturers or retail-
ers or commercial diving schools. This puts the whole thing in a very different balance
compared with most of the Americans and the result is when you get looting it is quite
easy to put local dive club organizations in contact with local museums, the people who
are responsible for salvage enforcement, that kind of thing, police if necessary, and
very considerable social and moral pressure can be brought to prevent looting. This is
for casual looting - not really professional salvage or treasure hunters in it for the
money. The thing that the casual discoverer wants - the guy who just finds the bones,
spear head, or wreck - what he really wants is involvement. If the law wipes him out
as soon as he reports the site, then he will never report the site. What you want is in-
volvement so that you report the thing to a museum and instead of getting a spear head
to put on your mantelpiece you get told you are a good guy, you get involved, you get
drawn into the museum or university world. You become a member of the team. You
get the whole damn wreck up and you see it coming up - you and the local press. This
is what the amateurs want. They want to feel if they find something it gets carried right
through and this happens, as you say, when you described the course you gave. People
do tum up for the courses. They do pay for it. They do want to be involved and this will
work when you've got an amateur structure and magazines to report it from. It doesn't
work, or course, for people who are commercially involved in salvage. If you've got a
firm which has to raise money from its backers or from its bank or whatever and has to
get up so much gold or copper or whatever, then this kind of inducement doesn't work.
If we assume, therefore, different areas of lawmaking where you have to devise penal-
ties and percentages they can take and so on so that there is an incentive for the treas-
ure hunter or salvager to report a find and to split the thing down the middle so that by
slowing his work up a bit, reporting the material and doing a little bit of archaeology he
still gets enough money off it. They are two far separate problems.
ANSWER: I thank you. You state it exactly as I felt it, and thank God you were
here to state it so well. This was the precise impression I got through my association
with the Derry Sub-Aqua Club. One of them kicked that cannon with his flipper and
through that they discovered the Trinidad Valencera. They immediately went ashore
and discussed it and said, "O.K., we will not touch one single thing until we can organ-
ize a complete, thorough, scientific dig of this thing," and that's exactly what they did.
It took them years before they even started but it is that attitude which is so common in
Europe and which we have not had that much success in doing here in the States that I
would like to see encouraged.
QUESTION: I think that point about public pressure is important. I know I can't
say anything for the rest of the country but in California we have a lot of organized
clubs that organize pot hunters in the water, if you will. We've started a public aware-
ness program where we'll show up at gun shows, antique shows, diving shows - div-
ing shows particularly because they are popular - and other groups. They usually have
a very large press exhibit and returns have been small, but they are picking up. Many of
the divers would have originally picked something up, brought it home, and that would
have been the end of it. We are starting to be contacted continually now by people who
have found things, and in many cases now they are leaving them in situ so that we can
go out and check it out in that way. I think this has to be done on a large scale also
before you have even a very large success with a class. It is important to get them used
Underwater Archaeology, European versus American 61

to the idea because let's face it, the treasure hunters have a lot better public relations
than we do. They have the press and the public imagination and the rest of it to go
along with it.
ANSWER: Let's face it, about 99 percent of our divers are trained through
NASDS, PADI, NAUI, YMCA, and a couple of others. If they were infected with the
idea, even in basic scuba courses, and had it further impressed on them in more
advanced training, it wouldn't be such a strange idea by the time that they were regular
divers and out among the shipwrecks.
QUESTION: Dr. Shiner, at this point I would like to interject an idea that I have
noted in many of the slides here and talks in other places having to do both with sport
diving and diving archaeology. I would suggest that if you are going to pursue a course
of this sort that you contact a former or active professional salvage master as opposed
to treasure hunters and learn something about recovery safety.
ANSWER: I couldn't agree more.
QUESTION: Particularly in the lifting of heavy objects because I have seen slides
that positively horrify me as a retired professional salvage master, not as a treasure
hunter.
ANSWER: My dive master was just such a person and he opened my eyes.
QUESTION: As a retired salvage master I'm being included with "pot hunters." I
won't accept the title right now. I may in the future.
ANSWER: Thank you sir. [laughter]
QUESTION: I just want to make a point about follow-up. If you inject this feeling
into the course at NAUI or PADI level, they still haye a lifetime of diving with only a
few weeks' course. They get the right idea put in their heads, but over the next 20 years
they are going to have it knocked out. I mean from the first year onwards, it's follow-up
at the local level we're going to need. We have our friend from California saying they've
got clubs and so on but it's through university groups, clubs, local museums, or whatever
that you've got to get the spirit into the divers, where they are actually practicing.
ANSWER: I think this will come about with about one more real cold winter.
We'll either throw out the Executive and all of Congress and get us a new energy pro-
gram or we'll do something and we'll be back into the search for oil and gas and other
stuff. However you may feel about it, we've got to go after it. This will provide
employment for these people at the diver level, and an opportunity to practice their art
and I'm sure that this will be back with us within a year. In fact, I have a little inside in-
formation that about a year from now we're going to be pretty busy out there.
QUESTION: There's only one thing about not really having a law that will allow
the hobby diver or an amateur to go out and collect artifacts. If we go out and train all
these people so that they can do it, and we're going to teach morality, won't we be
teaching that it's against the law but here's how to do it anyway?
ANSWER: I have an idea on my list for the course which is the legal aspects of the
situation both state and federal. And there is - in fact, Texas is the only state that
absolutely forbids any kind of split between the discoverer and the state. In Texas the
state gets 100% of everything.
QUESTION: Changing the subject, that is the salvager you are talking about.
We're talking about a hobby man, a guy that goes out on the weekend. You're saying
Texas is the only state that doesn't allow a split. The rest of the states allow a salvager
to go out - a guy that's out to make money.
62 J.L. Shiner

ANSWER: Right. He promises to dig it right and he splits according to whatever

law the state has. In Texas there is no splitting, no nothing, so there is no encourage-
ment for anybody but a university-trained person to even go look at a ship. So he goes
and steals it.
QUESTION: We have the additional problem that not only are you legally not pre-
pared for this kind of act of cooperation and participation by amateurs, but also there is
no state that I know of that has the staff to put together a program like this. I worked in
North Carolina for several years as an amateur under contract with the state and our
problem there was that we had plenty of manpower and the state didn't have anybody
who could help us get mobilized to do any work. And it's a problem that's almost
insurmountable at this point from the way I see it. You can't turn people loose without
any supervision and you don't have the personnel to supervise.
ANSWER: We have the supervisory personnel to probably handle 1000 such
divers at this time because my university alone is turning out 10 every other year and
there are at least a half a dozen other universities who are turning out very well trained
individuals and who are getting experience during the summer, particularly at such
operations such as Barto Arnold is running on Padre Island. I'm sure there are many
things in favor and many reasons to say this program might not work.
QUESTION: I agree completely that the professionals and the amateurs should
cooperate and work together. We haven't met, my name is George Bass.
ANSWER: Yes, we've corresponded, I believe.
QUESTION: You have worked with the amateurs in your town, for example, and it
worked out very well. I have little knowledge of the things you are saying. I don't have
an answer for this, but I'm concerned about it now since most of my work has been in
Europe. I can see that underwater archaeology has gone through four distinct stages. I
don't know what the fifth is going to be. The first of course was the sponge divers and
fishermen. The second was the early work by the people with the scuba equipment after
it was designed where they attempted excavations. During that phase archaeologists
weren't diving and there were no plans made and no proper publications. Then the third
stage was when we archaeologists began to dive ourselves, and we knew what the
archaeology was all about, and we then developed the techniques for mapping and div-
ing and working on wrecks which are now standard throughout the Mediterranean. But
people like me - I was trained as a classical archaeologist. I was interested mainly in
the cargo. I didn't know the difference between a garboard strake and a stern post, and
I barely do now. It's a joke that I wrote this book, The History of Seafaring, because I
know that I would drown if I was put in a sailboat one mile off the coast. [laughter]
So we brought in a ship expert, Dick Steffy, who knew about the hulls. This is
what Bill Baker was talking about. We archaeologists don't know enough, and so it
bothers me that people think they might be trained archaeological divers when they are
not trained in the theory of wooden hull construction. Very, very few people are. I have
no business ever excavating a wreck again, as long as I live, unless I have somebody
like Dick Steffy or one of the people he has trained, working with me. Otherwise it's
dumb, it's wrong, it's immoral. We did it, and the stage that I was a leader in, I think,
that was stage three, is just as backwards now as stage two was back when Cousteau
was sending down professional divers to raise a lot of amphoras. So I would say people
have to get trained in the ship's history, and that's something you're not going to get in
any anthropology course that I know of.
Underwater Archaeology, European versus American 63

ANSWER: You're right, George, but we've got to start somewhere, and these peo-
ple are going to dive. We can't stop them from that, and they are going to dive on
wrecks because wrecks are fun. Unless we contact them in some way, they are going to
continue to remove pieces of those wrecks by wrench or by dynamite, and if we can
contact them in some way and give them what we know at this point, and hold onto
them until we can get that further information on ships, let's do it as soon as possible, I
feel. That's all I have to say, Mr. Chairman. Thank you very much. I hope that during
the succeeding days that I can talk to some of you individually and map out a feasible
program to send on up to NAUI and YMCA and perhaps PAD!.
 Chapter 7

The South Carolina

".~i).:
~

Hobby Diver Program

-r/: ..
':.i
CHRISTOPHER F. AMER
and CARL STEEN

South Carolina, as we all know, has a long coast, punctuated by numerous rivers and
bays. There are over 12,000 linear miles of water courses in the state, and the state
claims title to all constantly inundated land beneath these water courses, and to the
ocean bottom out to the 3-mile limit. With the introduction of modern scuba equipment
and the general growth of the leisure industries, a new frontier was opened. With the
opening of this frontier a new problem was created; one of what to do about the impact
of divers on underwater resources. South Carolina's waters contain, according to hobby
divers, everything from fossilized paleontological remains to stolen bicycles.
Prior to the passage of the state's first legislation dealing with underwater antiq-
uities in 1968 there was no law regulating salvage operations. The 1968 law, entitled
Control of Certain Salvage Operations, was written by a lawyer, a shrimper, and a diver.
Their goal in writing the legislation was to have their rights as the discoverers of an
underwater resource - a blockade runner - protected from other potential salvors. In
other words they wrote the law to keep the site to themselves, not to preserve it for pos-
terity, but to preserve it for their own use. This was not an especially auspicious base for
a law that evolved into a cultural resource protection act, but it was a beginning.
An innovative feature of the law, however, was the establishment of the Hobby
Diver Program (now the Sport Diver Archaeology Management Program). The intent
of this feature was to allow recreational divers to collect fossils and artifacts from state
"land" without fear of prosecution or harassment. Because the state had declared that it
was the legal owner of all abandoned property located below the mean low water mark,
Source: South Carolina Antiquities (1988), 20 (1):41-44. Reprinted by pennission of the Institute of Archaeo-
logy and Anthropology, University of South Carolina, Columbia, South Carolina.

65
66 C.F. Amer and C. Steen

and thus gave itself the power to regulate who did and did not have the right to recover
property from these lands, it had to make provisions for the various forms that recovery
might take. The protection of the rights of the common citizen, as well as the salvor,
had to be considered for the law to pass.
The law, revised to its present form in 1982, is entitled the South Carolina
Underwater Antiquities Act (South Carolina Code of Laws, 1976, as amended, Section
54-700 et seq.). The South Carolina Institute of Archaeology and Anthropology
(SCIAA), Columbia, South Carolina, administers the law which provides for three
types of licenses for underwater search and recovery: (l) search license for conducting
underwater search operations using, for example, remote sensing, or other methods; (2)
salvage license which is required for conducting a major underwater salvage operation,
all too often with a profit motive from the sale of artifacts; and (3) hobby license.
The hobby license was designed to allow sport divers to perform intermittent,
noncommercial search and salvage operations of a recreational nature. Here is how it
works, or should work:
For a five dollar annual fee the licensed hobby diver can comb the State bottoms
to his or her heart's content collecting bottles, ceramics, shark teeth, and other souve-
nirs of the dive. To fulfill the licensing agreement the finder is required to report the
recovery of the artifacts, as well as the location and circumstances of recovery, to
SCIAA, on a monthly basis.
Once reported, a representative of SCIAA determines from the report if the arti-
facts and/or fossils recovered display any potential historical or scientific significance.
If so, the representative will contact the diver and arrange to see the collection to draw
and photograph it as well as record any further information about the site.
The law states that, in return for their cooperation in reporting both their finds
and the location of their finds, legal title be given to the diver. If, within 60 days of
receiving the report the state representative does not contact the diver, then again the
ownership of the material reverts to the diver.
There is also provision under this licensing system for the State to make a divi-
sion with the diver. This, to our knowledge, has never been done. Cultural material has
either been freely given to the State (SCIAA, museum, etc.), or ownership has been
given over to the diver after the artifacts have been recorded.
Hobby reports should always be reviewed and significant information verified.
The information should then be plugged into a master plan for studying the underwater
resources of the State. We should then be able to assess the importance of the artifacts
and sites reported in relation to how they further our knowledge about past lifeways
along the waterways of South Carolina.
Now let us look at how the Hobby Diver Program has actually fulfilled its
intended role and at future directions for the program.
The intentions of the Hobby Diver Program are threefold:
1. To establish a working trust relationship between professional archaeologists
and sport divers.
2. To encourage divers to report their activities and finds, to verify reports and,
thereby, gain new information about potential sites.
3. To develop among the sport divers an awareness of some sort of archaeologi-
cal and historical responsibility.
The South Carolina Hobby Diver Program 67
It has been argued that the law is useless because divers do not report what they
find, or that they just ignore the law. The Hobby Diver Program is based on trust. The
divers voluntarily subscribe to the program and make their reports voluntarily. If a
diver should omit items from his report or misrepresent the number or location of items
he has collected, the State will never know. The State trusts that divers will not collect
cultural material from the State waters without a hobby license. This, of course, is not
always the case.
People, however, should remember that unless their finds are reported legally and
properly, they are not the legal owners of the property and they are technically in pos-
session of stolen property, which according to the law can be confiscated at any time.
We should hasten to add that we have no real interest in prosecuting the average diver
who goes out two or three times a summer and finds a handful of shark teeth, or a cou-
ple of 19th-century bottles. People who think that they can make a living stealing the
heritage of the people of our state are the ones we want to stop. We would be happier if
no one was collecting artifacts, but in the real world we know that is not the case and
will not be the case.
The divers, in turn, trust the State to provide encouragement, consultation, and
advice, and to show an interest in what they are bringing up. They also want to know
that the information they provide is being utilized.
The best means to develop the trust, and therefore the cooperation of the diving
public, is through dialogue and education. During 1985 and 1986, in the wake of a loss
of trust on the part of the divers in the State, a series of Diver Contact Surveys were
conducted which laid the groundwork for improving the Hobby Diver Program. These
surveys involved talking to the diving public through dive shops, dive clubs and to
individual divers. The dialogue went a long way toward renewing the divers' trust in
the State. By explaining the true value inherent in the artifacts as tangible remnants of
cultural history, and the need to know their exact location or provenience, and how the
information fit into the cultural process, divers began to appreciate the need for accu-
rate and honest reporting.
In order to utilize new information about potential sites, that information must be
assessed as to its reliability. Hobby diver reports have been mounting up over the years and
a tremendous source of virtually untouched data exists therein. Hobby divers report ship-
wrecks, dugouts, plantation landings, fossil beds, lost tackle boxes, Civil War munition
trains, discarded murder weapons and stolen goods, and, well, you name it. Anything used
on land or water can be found in the water. The problem right now is that the reports are
mounting up, but there is only a small staff to assess, collate, and cross-reference them.
The situation, it is hoped, will be remedied in the near future with the anticipated
addition of more Underwater Archaeology Division staff. In the interim, staff from the
Underwater Antiquities Management Program assist the Division whenever feasible.
Many of the divers' reports do not require immediate verification on our part.
However, occasionally a report of a major find occurs which necessitates a timely
response. For instance, a hobby diver recently reported a dugout canoe washing out of
a creek bank. SCIAA staff visited the site and determined that the craft should be
moved to a safer location. The hobby diver, a contractor by trade, expressed a willing-
ness to provide his expertise and all materials necessary to raise the vessel.
The dugout will be scientifically excavated by professional archaeologists and
stored in a stabilized environment pending conservation of its structure. I mention this
68 C.F. Amer and C. Steen

case to illustrate the value of the state hobby diver cooperation, and to introduce the
third intent of the Hobby Diver Program, namely, to develop public awareness of our
submerged cultural resources.
The sport diving community of South Carolina, on the whole, has a good sense of
its history. Many divers, though, fail to make the necessary connection between the
cultural material they find beneath the State's water and that history. The artifacts are
too often viewed as interesting objects, not for what they can tell you about how people
lived (archaeology) or what light they can shed on major events in the past (history),
what new questions can be posed by their presence (science), or what they can tell us
about ourselves (anthropology). Much of that information can only be determined by
studying the artifacts in relation to other artifacts on a site and to their environment, or
the context.
One potentially valuable contribution that the Hobby Diver Program can make is
to the study of plantations and slavery. Already one of the biggest discoveries of low-
country plantations has come from data recovered by hobby divers; namely, the discov-
ery of pottery apparently made by, but surely used by, slaves. This discovery has
opened up a new perspective on the lives of slaves, suggesting the retention of African
material culture as well as folklore, herbal healing, speech, and other better-docu-
mented aspects of their culture.
A second interesting direction of study on plantation sites also comes from the
Hobby Diver Program. When we excavated the "Two Cannon Wreck," a possible
Revolutionary War-period vessel found offshore from a Cooper River plantation, the
vast majority of artifacts that we recovered were associated not with the wreck, but
with the plantation.
Looking back on the hobby reports of years gone by, it was evident that a pattern
of such site characteristics existed and that the "Two Cannon Wreck" site was not
exceptional, but one of many such sites. One might not expect to find a Revolutionary
War ship off of each one, but one can expect to find rice barges, dugouts, and other
vessels, as well as patterned deposits of plantation artifacts. In most cases these arti-
facts are in much better condition than those found on land, if only because they did
not break into as many pieces when they hit the ground! This is excellent in terms of
the study of artifacts, but it can also be used to extract much more detailed information
from the sites regarding the occupational structure of the plantation slaves, their subsis-
tence, the goods being brought into and out of each plantation, activities being carried
on at the riverside, and numerous other topics. To top it off, this is before we even
begin to consider the characteristics of the watercraft found at the sites!
In 1986, it was noted that the Hobby Diver Program was thought, by many pro-
fessionals, to be the best answer to the national problem, because it took a potential
threat and turned it into an asset. Today, cooperation between the sport diving commu-
nity and professional archaeologists and cultural resource managers is widely recog-
nized as the most viable strategy to the management of submerged resources.
After all, who finds submerged sites anyway? Most of the underwater sites being
worked in Europe and North America were found by amateurs and commercial divers,
not through the planned efforts on the part of professional archaeologists. Where the
contribution of amateurs has been acknowledged and as strategy of cooperation and
involvement of the sport diving community with professional archaeologists under-
taken, the results have usually been beneficial. Examples of this cooperation abound,
The South Carolina Hobby Diver Program 69

for example, in Britain with the Sub-Aqua Society, in Canada with Ontario's Save Our
Shipwrecks and the Underwater Archaeological Society of British Columbia, and in the
United States with the numerous state programs that have followed in the wake of
South Carolina's Hobby Diver Program. The locating and raising of South Carolina's
Brown's Ferry vessel, the oldest colonial vessel excavated to date in North America, is
testimony to the sport diver and professional cooperation (Steffy, 1979).
While a strategy of cooperation and involvement develops trust, it is the role of
education to develop a sense of ethics and respect for our submerged history. Such an
education program includes symposia, diver education handbooks, and training in
underwater archaeology with a regional emphasis on South Carolina. The program is
all geared toward developing these values and thereby increasing the quality of the
hobby diver reports, and emphasizing the fragility of our finite and nonrenewable
underwater resources.
What about the State's responsibility? SCIAA. is currently working on a plan to
better manage our underwater cultural resources and to more effectively utilize the
hobby diver reports. Statistical analyses of information contained in the reports will
help to define areas of high archaeological potential and to provide information on set-
tlement patterns and trade routes as well as military activities associated with the
waterways of the state. Based on the locations of known sites in the state, and by
extrapolation, hitherto unknown sites may be found.
The Hobby Diver Program has proven itself to be a valuable tool for cultural
resource management. Instead of fighting pointless battles against human nature, our
efforts have been directed toward educating the public in the value of their finds in
informational terms, and encouraging their cooperation. This policy has, to be sure,
resulted in the loss of both information and artifacts, and has provided grounds for
debate within the archaeological community, but we believe the positive results far out-
weigh the negative. Before the potential of the Hobby Diver Program is realized, how-
ever, we have a long way to go.

ACKNOWLEDGMENTS

This paper has drawn on a number of sources, both published and unpublished, to
fulfill its goal. Our appreciation goes out to the following persons: Mr. Alan Albright
who was head of the Underwater Archaeology Division of SCIAA for 14Y2 years, and
to whom credit goes for the formation and development of the Hobby Diver Program.
His papers, published between 1983 and 1986, formed the backbone of this paper. Mr.
David Brewer, during 1985 and 1986, was the Assistant Underwater Archaeologist as
well as the manager and field director of the Hobby Diver Program at SCIAA. Thanks
go to Mr. Brewer for the information provided in the papers of the Hobby Diver Pro-
gram, written during his short tenure with the Underwater Division, as well as his per-
sonal communications thereafter.

REFERENCES

Steffy, J.R., 1979, Preliminary Report: Hull Construction Features of the Brown's Ferry Vessel. The Notebook
10: 1-29. South Carolina Institute of Archaeology and Anthropology, University of South Carolina,
Columbia.
 '.~~.' Additional Bibliography
~

o i' .
for Part I.B
q;,O_' .': .
.j'~l

Douglass, Darren, 1991, NA U/ Wreck Diving Speciality Instructor Guide. NAUI Press, Montclair, California.
Douglass, Darren, 1991, NA Ullntroduction to Underwater Archaeology. NAUI Press, Montclair, California.
Lockery, Andy, 1985, Marine Archaeology and the Diver. Atlantis, Gronton, Ontario.
Smith, William C. and Grebmeier, Green, Duskin., 1981, Puget Sound: a Progress Report from the Center for
Marine Archaeology. Underwater Archaeology: the Challenge before Us. edited by Gordon P. Watts,
pp. 348-367. Fathom Eight, San Marino, California.

71
 Part I.e

Ill!. Ethics and the Great Debate

'ii
o...t
-r/: ..

The discussion about who investigates underwater resources continues with specific
details about ethical considerations relating to the archaeological profession. These
questions center on profit and site loss.
In Chapter 8, Throckmorton covers public issues - the salvage boom, ethics, and
profits - in one of the few articles to deal with measurable facts concerning the vola-
tile arena of treasure hunting and salvage of artifacts. Cockrell, in Chapter 9, discusses
ethical questions based on his own experience as Florida's underwater archaeologist.
In Chapter 10, Mathewson, an archaeologist who worked with salvors, presents
the Atocha salvage project in an interesting perspective. By showing the project's
achievements in the light of stricter archaeological standards, he argues that the ethi-
cally shaky initial project ultimately proved successful in terms of recovering data. The
"which comes first - the dollar or the data?" question is not answered but very
thought provoking questions about funding for data recovery are raised.
Dudley, in Chapter 10, offers the enlightening perspective of a United States Navy
historian, emphasizing the difficulty of protecting underwater resources, even when they
are still the property of the U.S. Navy. The Naval Historical Center faces the challenge of
educating the Navy Department while at the same time having its funds restricted. Par-
ticular mention is made of the CSS Alabama and its international implications.
These divergent views represent ongoing differences between professional histo-
rians, archaeologists, and salvors. The editors do not advocate all these views nor pro-
mote them; however, students must be aware of the divergent views and the possible
controversies they may generate.

73
 Chapter 8

The World's Worst

Investment
The Economics of
Treasure Hunting with
Real .. Life Comparisons

PETER THROCKMORTON

INTRODUCTION

The cost of undersea treasure hunts is double that of projects carried out by competent
scientists. Only 1 in 20 salvage companies has any chance of making money. However,
there is a way for investors to profit from shipwrecks.

HISTORIC PRESERVATION AS A PUBLIC ISSUE

The American public has been exposed to a storm of rhetoric arising from the con-
flict between salvors and historic conservationists. Treasure hunters argue that their trade
is good old American enterprise at its best and history and archaeology are boring and
unnecessary pursuits carried out by a "bunch of bureaucrats feeding at the public trough"
(Mel Fisher's attorney Paul Horan as quoted in Time magazine in 1985). Extremist
archaeologists say that compromise between business and science is impossible and
seem to want to hide in their ivory towers while the salvors smash what they like.
Source: Underwater Archaeology Proceedings from the Society for Historical Archaeology Co,!!erence, edited
by Toni L. Carrell (1990), pp. 6-10. Reprinted by permission of the Society for Historical Archaeology, Tucson,
Arizona.

75
76 P. Throckmorton

The real issue is that American treasure hunting is destroying scant resources of
desperately poor emerging Caribbean nations. Historic preservation in the Caribbean is
an economic issue. Traditional sources of income for many Caribbean islands have
declined in the past 10 years. For example, the value of sugar exports from the Domini-
can Republic in 1987 was one-third of what they had been in 1977, and in the smaller
islands they were about one-half. Jamaica's bauxite exports were down by over one-
half in the same period. In contrast, tourism, two-thirds of it North American, has
increased in all of the islands. In Jamaica, the Dominican Republic, St. Kitts and Nevis,
St. Lucia, Antigua, and the Cayman Islands it has doubled since 1980, and in St. Martin
it has tripled. Today, tourism is rapidly becoming the treasure of the Caribbean.

ARCHAEOLOGY AND TOURISM

A recent study of the economic impact of the arts in Britain, by Mr. John Myer-
scough of the Policy Studies Institute, analyzed the economic impact of the arts, includ-
ing museums, on Britain's economy. Foreign spending accounts for 37 percent of the
turnover of the arts, compared to 27 percent of British manufacturing as a whole. The
arts are the fourth biggest earner in Britain. Twenty-seven percent of tourism earnings
are attributable to cultural attractions. A study by the Association of American Muse-
ums has shown that each tourist who visits a museum leaves about ten dollars in the
immediate local community's tax structure.
Archaeological resources are an asset to communities that create museums,
because museums attract tourists. The ruined crusader castle in Bodrum, Turkey was
converted into an archaeological museum in 1959. At that time the town's population
was around 5000 and there was almost no tourism. Since then, George Bass and his
group have been excavating shipwrecks and creating exhibit material. Today the
museum is the second most visited in Turkey, after the National Museum in Ankara.
The population has also tripled and local businesses are thriving. There are perhaps half
a million visitors per year.
In 1967 Bass's group started excavating a 4th-century B.C. ship off Kyrenia, a vil-
lage well off the main tourist track at the southeastern end of the island of Cyprus. In
1967 Kyrenia Castle was the sixth most-visited archaeological site in Cyprus. By 1974,
when the Turks invaded Kyrenia and put a stop to tourism on that end of the Island,
there were over 100,000 visitors per year, and Kyrenia castle, with its ship exhibit, had
risen to second in popularity.
Table 1 (courtesy of Michael Katsev) illustrates the process.
A significant part of the Caribbean's historical heritage is its dramatic history of
buccaneering, privateering, slave trading, and naval warfare. The islands were a cock-
pit for the conflicts of Europe for four centuries. The relics of all this are scattered over
thousands of reefs and cays and they are being mindlessly destroyed by treasure hunt-
ers. Americans are paying for this destruction twice: first, in the tax deductions granted
to investors in salvage companies, and second in aid money given to impoverished gov-
ernments. Tourist dollars paid to taxi drivers, hotel keepers, restauranteurs, store own-
ers, and the like go directly into the local economy. Economies supported by tourism
don't need our aid money.
Shipwrecks and the material in them, properly excavated, curated, and exhibited
in a museum, are as valuable as historic houses, moldering castles, or pre-Columbian
The World's Worst Investment 77

Table I. Visitors to Kyrenia Castle

1966 11,867
1967 27,206 October survey of site
1968 29, 791 First summer of excavation
1969 47, 739 Second summer of excavation
1970 69,405 Begin preservation; temporary museum opened
1971 93,025 Continue preservation
1972 93,025 Begin hull reassembly
1973 113,500 Continue reassembly
1974 62, 137 Visitors through mid-July; Turkish invasion

ruins. When Caribbean countries barter away their historic shipwrecks in return for a
percentage of an imaginary take, they lose tourist attractions that could produce income
for all foreseeable time.
Opportunities like the one that existed in Bodrum 30 years ago are scattered all
over the Caribbean. There are even suitable castles: St. Juan de Ulloa in Veracruz, Eng-
lish Harbor in Antigua, Sans Suci in Haiti, and Brimstone Hill in St. Kitts, to name a
few. The treasures of the Conception, housed today in the old Governor's Palace in
Santo Domingo, are the centerpieces of a spectacular historic restoration project and a
thriving local boom, which brings the Dominicans much more money than if the collec-
tion had been sold. If the Conception had been professionally excavated, the collection
would be even more valuable.

MUSEUM ECONOMICS

Digging into historic sites whether on land or under the sea usually produces
material that has no value unless it has been intelligently excavated, conserved, and
exhibited. Shipwrecks are an extreme example. A few farseeing investors are beginning
to understand that museums can pay. Several companies are now investigating financ-
ing archaeological projects that will result in museums, and building resorts around
them, so that they acquire the long-term profit from the visitors attracted by the
museum. If the museum is nonprofit, the excavations that produce the material can be
done as 50l(C)(3) projects through universities; this cuts costs in half. Investors are
beginning to understand that archaeology is valuable, not because it produces "treas-
ure," but because it brings in tourists.
A successful museum that might serve as a model for museums in emerging Car-
ibbean countries is the Maine State Museum in Augusta, Maine, which was in part
inspired by the National Museum of Mexico. The museum is free. Its attractive exhibits
illustrate Maine's natural history and changing seasons, the disappearing world of 19th-
century Maine, and the 1O,000-year history of the state. It attracted 180,000 visitors in
1978. The state legislature funds the museum according to annual visitation, figured at
a rate of $10 per visitor per year. Half of this is spent for day-to-day operations, the
other half for statewide educational programs and development. Its director, Paul
Rivard, says that people don't come to Maine specifically to see the museum, but stay
longer in the state because of it.
Maine, with a population of 1.2 million, has about 4 million out-of-state visitors
per year who spend $1.7 billion. Like the Caribbean islands mentioned above, Maine's
78 P. Throckmorton

tourism is increasing at the rate of about 10 percent per year. Not so spectacular as
some Caribbean islands, but still impressive. Maine's per capita income of $16,000 a
year is not much higher than that of the most successful Caribbean tourist islands: the
Cayman Islands per capita income is $12,000, the Bahamas' is $9000.
Jim Thompson, of Maine's Department of Economic and Community Develop-
ment, says that his office is investigating the relationship between amenities such as
museums, theaters, concerts, fairs, and festivals in order to establish how much such
amenities affect the length of stay of visitors; the idea being that the longer you keep
tourists in the state, the more money they will spend. Another aspect of providing cul-
tural amenities is that their clientele will be more prosperous than, say, visitors to
beaches or rock concerts.
Vasa, a battleship of 1628 salvaged intact from Stockholm Harbor in 1961, fur-
ther illustrates the point. Conservation proved very expensive; there was an outcry
against the project. Today, according to the Swedish tourist board, one million tourists
spend an extra day in Sweden because of the Vasa. At about $300 per day per tourist,
this amounts to several hundred million dollars per year added to Sweden's economy.
It appears then, that the best way to make real money out of old shipwrecks is to
use them to create museums and then profit from the added revenue that follows the
increase in tourist length of stay.

THE SALVAGE BOOM

Treasure hunting in Florida boomed in the early 1960s, with the discovery of the
1715 plate fleet off Vero Beach. What had been a relatively gentle weekend hobby as
practiced in the Keys in the early days of scuba diving, became a gold rush, with its
attendant cast of profit-minded adventures. The best known, Mel Fisher, came in from
California to work on the 1715 plate fleet, then moved to Key West for the famous
search for the Atocha which he eventually found, to the acclaim of the U.S. public.
Florida's policy toward its underwater antiquities has cost the state millions. The
state's 25 percent share from the treasure grubbing of the past 20 years is a collection
worth only about $5 million today. The Florida state museum has in its possession
approximately 1500 gold coins worth on the market about $2000 each, and about
20,000 silver ones, worth $80 to $150 each. This represents the state's 25 percent of all
treasure recovered in Florida pre-1982.
The collection has cost more than its value to maintain, especially if one includes
the cost of the continual legal cases that have resulted from the state's policy. If Florida
had used state money, and invested $10 million in two great maritime museums back in
the 1960s, instead of giving leases to salvors, the state would be nearly half a billion
dollars richer each year, if the Swedish example applies to Florida. The state's share of
that sum in taxes would be not less than $25 million per year. If the state was maintain-
ing two museums at the Maine rate of $10 per visitor, there would still be a direct profit
in tax money of $5 million per year. It could even be argued that individuals and corpo-
rations who wanted to do legitimate work on searching for and excavating shipwrecks
in collaboration with the state, would have profited. As things stand today in Florida,
the state has set up such efficient barriers to stop treasure hunters that even legitimate
archaeology is inhibited.
The World's Worst Investment 79

The emerging economies of most small Caribbean islands can't even afford
decent education systems and healthcare, much less archaeology and museums. As a
result, they are wide open to the blandishments of salvors. As of the summer of 1988
there were five salvage companies working in the Bahamas and three or four in the
Dominican Republic. A big one has a monopoly in the Turks and Caicos. Another,
smaller, has rights to the British Virgin Islands. Mel Fisher has managed to get conces-
sions from the Antiguan government for the reefs of Barbuda. Treasure hunters with
millions of dollars behind them are working in the Marinas, the Philippines, and the
China Sea. In most cases the treasure hunters' contracts give them exclusive rights,
thus keeping out archaeologists. In 1988 I flew in a light plane over the east coast of
Florida and spotted seven different salvage vessels industriously blowing sand with
their mailboxes in the Vero Beach area where the 1715 fleet went ashore. These are
said to be working under concessions from the Fisher organization, which controls the
leases. The hunt for antiquities has been reduced to the level of strip mining.
Florida treasure hunting sprang from a booming get-rich-quick society that has
little historical past. Salvors tended to be working-class Midwesterners. Their divers
were mostly typical products of the failed education system in this country, where one-
quarter of the population can't read the directions on a can of soup and three-quarters
don't read newspapers as often as once a week. According to a National Geographic
Survey released in July 1988, one in four Americans can't identify the Pacific Ocean or
the Soviet Union on a map, and 75 percent can't locate the Persian Gulf.
Today's salvors are no more aware of the cultural material they destroy than the
peasant farmers who rob tombs for a living in Sicily or Colombia. Treasure investors
are not well informed. Individuals knowing nothing about history or the ocean, with
their sources limited to the national slicks and pulp "Treasure" magazines, are fair
game for promoters. As P.T. Barnum said, "a sucker is born every minute."
When Mel Fisher won the Cobb Coin Case entrepreneurs turned treasure hunting
into a nationally financed industry. The passage of PL 828 and the public confronta-
tions that preceded it drew attention to the operations of salvors from environmental
agencies, state and federal prosecutors, and the Securities and Exchange Commission.
The promoters moved to the Caribbean. Like the drug trade that began with a bunch of
happy-go-lucky hippies smuggling marijuana with sailboats, the treasure salvage indus-
try has grown.
Today's treasure hunts are promoted on Wall Street and the Vancouver Stock
Exchange. Their investors include some of the wealthiest men in the world. What we
are seeing today is an assault on antiquity by an industry, not by a bunch of small-time
adventurers. In scale it is larger and better financed than any assault on antiquities in
history. About 25 treasure hunting companies are touted every year. About half get
financed, for a total of up to $100 million. This industry is equipped with big ships that
have attached prop blowers that can blast away 500 tons of sand in 15 minutes, and
open a hole in the seabottom that is 15 feet around at the bottom and 50 feet across at
the top. Sophisticated instruments can detect a cannon 100 yards away. Treasure sal-
vage is now an industrial process.
While we deplore Lord Elgin's removal of the marbles from the Parthenon, we
must admit that Europe's willingness to pay Turkish pashas good money for pagan stat-
ues saved a lot of statues from being smashed by Moslem fanatics. When dealers and
governments paid adventurers to rob Egyptian tombs, they wanted the goodies because
80 P. Throckmorton

they cared about them. Today's investors are paying salvagers to take the valuables and
smash the rest. The difference is one of scale. The modern salvagers of the Geldermar-
sten (the Nanking ship) are said to have dynamited the almost intact wreck after salvag-
ing the Chinese blue or white porcelain so its location would remain unknown and the
government from which it was stolen could not prove ownership.
The government of the Turks and Caicos gave an exclusive contract to a profes-
sional industrial salvage company called TACMAR to rummage the shipwrecks in their
territory for gold and silver in 1987. This is the moral equivalent of the Egyptian gov-
ernment giving Morrison Knudson a contract to bulldoze the Valley of the Kings in
order to increase Egypt's gold reserves.
In the case of the Egyptian tombs, the salvaged material is now mostly in muse-
ums and the tombs themselves are mostly intact. While one can deplore the destruction,
a large percentage of their historical value survives. This is not so in the case of a
looted shipwreck, where 90 percent of the interest lies in the ship itself and the artifacts
that are ignored by the salvors, and destroyed, because they have been uncovered. One
shipwreck like the Geldermarsten equals, in quantity of material destroyed, over 3000
Etruscan tombs.
Now, in 1990 TACMAR seems to be going out of business. Nothing of value has
been found and millions of dollars have been spent. The company's Wall Street inves-
tors have, perhaps, wearied of supporting a profitless project.

EVALUATION OF SOME UNDERSEA

ARCHAEOLOGICAL PROJECTS

Note that valuation of material recovered is based on today's estimated market

value of the collection as a whole. In the case of expeditions that recovered material of
important scientific but little market value, the value of recovered material is given as
$1 x $1. Value of the collection to the host country is counted as $10 per visitor to the
museum per year in the years since the exhibit became public. Money spent is the
amount spent on the actual excavation calculated in todays dollars, calculated as
1960-1967 x 3.4, 1977-1987 x 1.7. All expeditions listed in Table 2 are over 100 on
the rating scale given below:

EVALUATION OF A GROUP OF SALVAGE OPERATIONS

Key to numbered items graded on a I to 10 scale, 10 being excellent. The items

below are broken down into ten questions for each heading. In the case that an item is
not required, ten credit points are given.
I. Magnetometer
2. Research
3. The expedition
4. The ship
S. Equipment shallow water
Sa. Equipment deep water
6. Archaeology
7. The crew
The World's Worst Investment 81

Table 2. Evaluation of selected undersea archaelogical projects

ExpeditionlYear Cost Return % return

Cape Gelidonya 1960 $1,000,000 Museum x II
Pelagos 1969 $50,000 $210,000 Market x 4
Torre Sgarratta 1967--68 $100,000 $100,000 Market xl
Kyrenia 1968 $120,000 $240,000 Museum x 2
Yassi Ada 1961--63 $370,000 $1,000,000 Market x 3
Ulu Burun $600,000 $1,000,000 Market x 2
Totals $1,330,000 $3,550,000 Mixed x 2.6

8. The company
9. Financing and accountability
10. Political
This is a gross return of perhaps $35 million, of which investors got less than $10
million. However, the gargantuan Atocha project skews the figures. If one subtracts the
approximately $13 million spent on the Atocha, we are left with 15 projects that cost a
bit over $17 million, which returned to their investors a total of between $3 and $4 mil-
lion. Only one project, Conception II, returned any profit to the investors. This is a good
deal less than donors got by supporting the 501(3)(C) nonprofit tax organizations than
did the nonprofit projects listed, i.e., 33 percent as opposed to 50 percent plus. Cost of
the nonprofit operations averaged about $5000 per month, while salvage operations aver-
aged $9000. The most inefficient treasure hunt spent $500,000 for 16 days at sea.

Table 3. Evaluation of a group of salvage operations

Rating in each category

Total
Project 2 3 4 5 6 7 8 9 10 rating Cost return for $1
Conception 1\ 1987 10 \0 7 10 9 0 8 10 9 9 92 $130,000 x $18
Phoenix * 10 \0 10 10 \0 6 8 9 8 9 90 $2,000,000 x $0
White Squall 1984 10 9 9 6 7 7 7 5 7 6 73 $100,000 x $1
Accuro 1979 5 6 9 9 7 4 8 7 8 8 71 $375,000 x $0
Seaview 1987-88 3 2 8 6 8 9 9 9 3 9 66 $1,250,000 x $0
World Explor 1984 8 8 8 8 8 3 6 6 5 6 66 $250,000 x $0
Concept 1978 2 10 5 7 9 8 8 9 2 6 66 $3,000,000 x $1.5
Steed 1982 3 5 6 8 6 5 7 8 7 7 62 $210,000 x $0
Rex II 1982-83 5 4 6 \0 10 2 7 7 6 4 61 $275,000 x $0
TACHAR 1986-88 0 5 4 7 8 3 5 4 9 5 50 $4,000,000 x $0
Atocha 1972-86 3 9 5 5 6 6 7 2 3 2 49 $13,000,000 x $2.5
MAVL 1984-85 1 5 5 \0 10 0 5 3 4 6 49 $350,000 x $0
Baltimore 1982 0 3 5 6 6 4 6 6 5 5 46 $175,000 x $0
Sea Search 1987-81 0 4 8 7 6 0 6 3 1 3 39 $750,000 x $0
James Bay 1978 3 0 3 5 8 0 5 3 5 4 36 $800,000 x $0
NCR 1985-88 0 0 2 8 5 0 2 2 2 5 26 $500,000 x $0
Total $27,165,000 x $0
·Ongoing project.
Notes: The grading system derives from experience with about 50 expeditions over the past 30 years in eight different coun-
tries. It is too lengthy to detail here. Amounts are approximate and are translated to 1988 dollars. Return is difficult to cal-
culate due to the secretive nature of the salvage business. Returns of less than ten cents on the dollar have not been listed.
82 P. Throckmorton

TRENDS

Pre-1985 Spanish treasure projects are relatively small time. The large industrial-
type projects, i.e., TACMAR, Phoenix, and two large projects in the Philippines (not
listed), have appeared since the 1985 Atocha discovery and its attendant publicity.
Treasure hunting seems to be something for which you can raise millions even if
your organization is incompetent. The old-style investors didn't seem to care. When the
Miami Herald put an investigative reporter onto Mel Fisher's operation, she found that
a majority of Fisher's investors were satisfied because it had been an exciting adven-
ture, and a much smaller percentage were suing because they felt conned. No one had
any criticism of the inefficiencies of the project.
A grade of over 70 seems necessary for success. Conception I, although it
returned $2 gross for $1 invested, went bankrupt, because the "take" was split with the
Dominicans, which returned investors less than 50 cents on the dollar.
The Atocha project is the subject of much speculation. Based on the Miami Her-
ald report, the luckiest Atocha investors probably didn't break even. The 3: I return
does not include investors. A low-rated project such as Atocha can achieve something,
if good money is continually thrown after bad, but is unlikely to return a profit.
Competent archaeologists are excavating shipwrecks and conserving what's
worth saving for less money than the salvors spend ripping them up by the roots.
Nearly all of the salvage projects being floated around the world today are doomed to
failure because of the incompetence and ignorance of their principals, or because they
are scams, designed only to benefit the promoters. No one seems to notice.
If investors understood that they cannot profit unless they employ professionals
and professional societies could guarantee that the archaeologists are competent, and
persuade host governments that they need cultural resource surveys, and need to protect
shipwreck sites that can stock museums, there would be a better future.
It is not to the advantage of any salvor who wants to make money to spend time
wrecking sites that contain no treasure. It is not to the advantage of governments to allow
cultural resources that might eventually benefit tourism, to be destroyed. It is not to the
advantage of investors to be victims of the incompetence and chicanery that are charac-
teristic of the treasure salvage industry today. Only archaeologists can prevent this.

CONCLUSION

Even a profitable barbarity, like the Geldermarsten project, would have paid its
investors many times what it did if the project had been legitimate. Because the mate-
rial was probably stolen, there was a chance that the country that owned it might sue.
This caused a fast auction in Amsterdam. The porcelain sold is said to be bringing sev-
eral times its auction price today. Even if the salvage company had gotten only half the
porcelain, the investors would have made more money.
The only groups that have consistently performed in the shipwreck salvage business
are a few salvors, and archaeologists using the systems developed by Penn Museum and
their followers in the 1960s. If investors want to make money, they need to tum away from
the incompetents who infest the salvage business, and deal with professionals.
Governments of countries being exploited by salvors need to take a second look
at what they are permitting and what it will cost the next generation. Archaeologists
The World's Worst Investment 83

need to fulfill their professional responsibilities better. All of us who stand to benefit
from reform of the salvage business - salvors, archaeologists, investors, and the citi-
zens of the countries being exploited - should ask the people who control our foreign
aid investment in the third world to take a hard look at the economics of museums and
the tourist trade.
As things stand, investment in the salvage industry only benefits promoters and
lawyers.
 Chapter 9

Why Dr. Bass Couldn't

".~i).:
~

Convince Mr. Gumbel

-r/: ..
':.i
The Trouble with Treasure
Revisited, Again

WILBURN A. COCKRELL

INTRODUCTION

This past summer I received a telephone call from Greg Stemm, a very nice gentleman
who is the president of the Seahawk organization, out of Tampa, Florida. Mr. Stemm is
interested in pursuing undersea exploration, and is specifically interested in ship-
wrecks. He called to tell me that his corporation had just discovered a wooden ship-
wreck, apparently from the early 17th century, in 1400 feet of water off the Florida
Keys. He told me that the shipwreck was in a remarkable state of preservation, and
asked if I would be interested in coming down with him to view the wreck with remote
sensing equipment on their second trip to the site. I told him I was intrigued with it, and
would discuss it further.
Over the ensuing weeks, I was interested yet perplexed about my possible role.
During my 11 years as Florida State Underwater Archaeologist, I had seen the terrible
depredations and damage caused by the treasure hunters. I resolved that I had two con-
ditions that I felt professionals should adhere to in dealing with any archaeological site,
whether it be a land site or shipwreck. The first condition would be that no archaeologi-
cal site should be destroyed for fun and/or profit. The second condition would be that
Source: Underwater Archaeology Proceedings. Conference on Underwater and Historical Archaeology, edited
by Toni L. Carrell (1990), pp. IJ-18, Reprinted by permission of the Society for Historical Archaeology, Tuc-
son, Arizona.

85
86 W.A. Cockrell

no publicly owned materials should be converted to private ownership. This, of course,

has as its presumption that archaeology is a destructive process, whether done by
archaeologists or as treasure hunting. As Mr. Stemm's wreck was apparently located
outside of Florida and U.S.-controlled waters, the wreck did not fall within the legal
jurisdiction of Florida or the United States. There would be no problem with the legal-
ity of Mr. Stemm's proposed explorations, as the wreck was in international waters.
Mr. Stemm indicated that he planned on converting some or all of the materials on dis-
play, and perhaps some at a museum. He indicated that they had very sophisticated
technology, an adequate crew, and financial backing to pursue this exploration and
excavation. This was a perplexing situation for me. I was fascinated by the possibility
of being one of the first archaeologists to view what could perhaps be an immensely
significant time capsule: an undisturbed 17th-century shipwreck. However, I felt very
apprehensive about associating myself with a treasure hunting operation, even if it
meant having the opportunity to see such a valuable shipwreck. I mentally reviewed all
of the old self-serving justifying arguments I'd heard from all archaeologists who'd
ever needed or wanted to work with treasure hunters, and concluded that I could not
allow myself, within my own self-imposed ethical framework, to participate in such an
expedition. In that frame of mind, I called my close friend and colleague, Larry Mur-
phy. I discussed my dilemma: my interest in seeing the wreck, yet my concern with
maintaining my thus far long-standing ethical purity. We discussed the nature of the
ultimate damage to the resource; I gave much thought to the situation, and ultimately
did not pursue going out on the second expedition.
As my work at Warm Mineral Springs began to preoccupy me again, I soon let
the debate slip to the back of my mind; then one morning last summer I was having
coffee while watching the NBC Today Show and I heard an announcement about treas-
ure hunting in Florida. I immediately began to pay attention to it, and saw host Bryant
Gumbel interviewing Dr. George Bass of the Institute of Nautical Archaeology on the
issue of treasure hunting. The network had a field hookup with Greg Stemm on board
his vessel in the Keys. I was fascinated as I watched Dr. Bass eloquently detail to Mr.
Gumbel and the viewers the reasons archaeologists are opposed to treasure hunting,
and watched him fail to persuade Mr. Gumbel. All involved were personable; George
was very amiable, Greg was very amiable, and Bryant amiably acted as moderator. At
every disputed point, I, as an archaeologist, understood Dr. Bass's point, but as a
viewer, I kept seeing that Dr. Bass was losing the battle. Ultimately, the program was
over, and had I not been a person with an archaeological background, I would have had
to clearly agree with Bryant Gumbel that Greg Stemm was indeed doing a service. This
caused me to recall my earlier discussions with my colleagues, and my self-review, and
I attempted to clarify for myself why it was that archaeologists, myself included, have
had such a difficult time in making our case to the public. I have no new conclusions,
nor do I have any new ideas to support any new strong recommendations. I will share
some thoughts and reflections on the reasons we fail to convince the treasure hunters
and the general public of the correctness of our convictions.

WHAT IS ACCEPTABLE?

We have a continuing inability to convince the public, the media, or even our-
selves of the great threat of treasure hunting and the need to stop the destruction of
Why Dr. Bass Couldn't Convince Mr. Gumbel 87

shipwrecks. The consistent question that we're faced with when we as archaeologists
attempt to stop or contain treasure hunting is, "What is the harm?" We seem to not be
able to come up with a concise answer. If we continue to fail to agree among ourselves
about the nature of the harm, or about the significance of the resource, then I feel that
we will never convince the public.
The question then remains, what is acceptable to professional archaeologists?
What approach to shipwreck recovery or shipwreck preservation meets our standards?
What are our standards? These are some of the questions I asked myself. In struggling
to comprehend, I examined some situations that some archaeologists find unacceptable
and that others find acceptable. In this manner, I was hoping to clarify in my own mind
the nature of the problem. I had a basic feeling that I could distinguish that which was
right from that which was wrong. I felt, as did many of my colleagues, that I could dis-
tinguish between unacceptable and acceptable behavior, yet there were too many situ-
ations that fell in the gray areas, rather than in the polar black or white areas. It seemed
to me that there must be some way of determining the basic criteria for acceptability or
unacceptibility. I then attempted to isolate determining factors for acceptability, that is,
to factor out our key concerns. Following are some key criteria I was able to isolate.

1. Age. One obvious key criterion is age, that is, it seems that we have no par-
ticular problem among ourselves as historic preservationists in allowing a salvager to
salvage a vessel that went down a year ago, or even 5 years ago. In reviewing the age
criterion, it is instructive to view the way that we archaeologists react to treasure hunt-
ing on vessels of varying ages. For example, the Andrea Doria went down in 1956, and
people had many diving expeditions on it. I have heard no outcry from the profession
about the Andrea Doria work. The USS Arizona went down on December 7, 1941 in
Pearl Harbor. As this is a U.S. warship and contains' the remains of U.S. servicemen, I
think there would predictably be a strong feeling that the USS Arizona should not be
disturbed by treasure hunters: this strength of feeling, of course, would derive probably
most largely from the fact that the USS Arizona is a memorial, rather than an archae-
ological site. The situation becomes grayer as we move to the HMS Titanic, which went
down in 1912. You'll recall the uproar that occurred after the discoverer, Dr. Robert
Ballard, had examined the vessel and left it, only to find foreign treasure hunting inter-
ests were planning, and in fact, did execute, a recovery attempt on the Titanic. Then
there is the Maple Leaf, a Civil War vessel that went down in 1864 in Florida's St.
Johns River, containing a wealth of artifacts. That vessel has been legally worked by
treasure hunters, or "a dentist turned archaeologist," according to newspaper accounts.
The State of Florida Division of Historical Resources is actively supporting state fund-
ing for the Maple Leaf operations. Virtually no professional has objected publicly to the
continued looting of this wreck. Another Civil War-period vessel is the Central Amer-
ica, which went down off the coast of the Carolinas. It was recently found, and there
has not been a great outcry about protecting the archaeological integrity of the Central
America. Moving back in time, there is the 1622 wreck, Atocha, worked by the Mel
Fisher organization: as archaeologists, we have generally held that the treasure hunting
on the Atocha is unacceptable. Earlier still, the oldest known shipwreck in the Western
Hemisphere is the Santa Maria, wrecked in 1492 off the northern coast of Hispaniola,
near modem-day Cap Haitien. We would have universal agreement that this vessel was
important. In reviewing our responses, and in examining our personal and professional
88 W.A. Cockrell

feelings about the preservation and the protection of these vessels from treasure hunt-
ing, it can be seen that we're probably in universal agreement about protecting the old-
est sites, and raise very little fuss about treasure hunters coming into the newest sites. It
can be seen that there is no very distinctive age criterion. We know that various laws in
various states, and now the Abandoned Shipwreck Act, attempt to define legally the
temporal window of legal concern. Any of us involved in the creation of those laws,
however, knows the necessarily arbitrary nature of those temporal boundaries.

2. Provenience. Another key criterion that some find significant is provenience.

The Bertrand, which went down in the Missouri River in the 19th century, at the time
of its excavation would probably have been protected by federal law had it been on the
river bottom. Yet given that the river had changed course and the wreck was on land,
the Bertrand artifacts, albeit against the wishes of some archaeologists, were divided
up between the federal government and the private treasure hunters. As this vessel was
buried in the dirt, some felt that it no longer had the protection afforded an underwater
shipwreck.

3. Type. That leads to the next key criterion, type. Perhaps we find ourselves less
concerned about historic underwater sites if the sites are planes instead of ships.
Another interesting case is that of the Liberty Bell VII, the space capsule that is being
recovered. This could fall under the category of age, because it's new, but should also
be considered in the "type" category; it isn't a ship in the nautical sense, even though it
is a historic craft in the water. Yet we're not concerned about it, apparently, as an en-
dangered cultural resource being recovered by those who are not archaeologists. A fur-
ther example, in Florida, is the ship Valentine off Hutchinson Island. This is a lumber
vessel that went down in the early 1900s. Not many archaeologists have expressed a
strong concern about protecting the Valentine, yet in Florida I attempted to protect it
from looters because it has historic significance, particularly because documents of the
rescue of its crew still exist. Portions of tl}e wreck can be seen from the House of Ref-
uge on Hutchinson Island. Another case of a particular type being perhaps not so sig-
nificant to us is the Civil War supply boat in South Carolina that was worked by a
treasure hunter in conjunction with the federal government and the State of South Caro-
lina. In reviewing these instances it can be seen that there are types of vessels to which
we apparently accord less protection.

4. Legality. Another key criterion would be in the area of legality. In foreign

waters, the Dominican Republic for example, or off the coast of Haiti where the Santa
Maria lies (or is purported to lie), it can be seen that professional archaeologists, with
very few exceptions, tend to express little or no concern about shipwrecks being looted,
as long as that looting is done according to local law. In international waters, where
legal protection is disputed or nonexistent, destruction is legal; only lack of technologi-
cal access has protected many of the most historic shipwrecks known. The Titanic and
the Central America are in international waters, as is the so-called Merced being sought
by the Seahawk group: their exploitation has elicited very little professional outcry.
Ships in U.S. federal jurisdiction have also been allowed to be looted. Vessels under
NOAA or National Park Service control are very strenuously protected, although, the
federal government and the State of South Carolina have allowed a Florida treasure
Why Dr. Bass Couldn't Convince Mr. Gumbel 89

hunter to recover artifacts in a treasure hunting operation with legal professional super-
vision. Another case would be that of the Atocha which the U.S. courts have allowed to
be destroyed by treasure hunters.
Some archaeologists are willing to accept the fact that states such as Florida, South
Carolina, Massachusetts, and soon California have active treasure hunting operations
proceeding with state sanction, state supervision, and as a matter of fact, the state
allowing the treasure hunters to keep artifacts that the state originally had ownership of
as publicly owned antiquities.

5. Professionalism. A further criterion is whether or not professionals are involved.

For example, it is publicly claimed by Seahawk, Inc., that their Merced operation off the
Florida Keys has the support of professionals from the State of Florida and the Institute
of Nautical Archaeology. The purpose of these claims is to convince the profession that
treasure hunting is acceptable as professionals are involved. In this case, this professional
support may not be as strong or as clear as is claimed by the treasure hunters, but they
did make contact with the State of Florida, and the State of Florida assisted them in treat-
ment and analysis of the materials they recovered. The involvement of professional
archaeologists is a key factor on the Maple Lea/in the St. Johns River. The State of Flor-
ida's archaeologists are publicly approving of the treasure hunters work, and a recent
newspaper article states that the treasure hunters' claim the support of National Park
Service personnel; indeed the dentist is professionally qualified, according to the news-
paper, as a "dentist turned archaeologist." Another example is that of the previously men-
tioned South Carolina case in which the federal government and South Carolina are
giving professional archaeological support to the operation. In the State of Massachusetts
it is well known that the Whydah involves professional supervision by the government,
as well as professional involvement by archaeologists and conservators.
Aside from whether professionals are involved is the issue of just who qualifies as
a professional, qualified to work archaeological sites. In examining this confusing crite-
rion, it becomes necessary to address the very delicate question of what and who are
professionals. Are professionals only those who are SOPA certified or the equivalent?
Are professionals those who are educated and trained as archaeologists? If so, in which
field? Would the field be anthropology, or history, or would it be engineering? Would it
be conservation? In all of those fields and many others are those who claim to be pro-
fessionals. Are professionals those who, in addition to the above, are ethical? This
becomes a very delicate question, as ethics are rather difficult and vague to define; and
once defined, it seems that in the area of treasure hunting, a good many professionals
skirt or fall over the edge of what has commonly and traditionally been understood as
the limit of professional and ethical behavior.

6. Artifact Disposition. Some archaeologists find that they focus on artifacts

when they consider the damages from, or the prosperity of, treasure hunting. A situation
that would be acceptable to most archaeologists would be that in which all artifacts are
kept in permanent public ownership, such as is the case with George Bass and his INA
operations in Turkey. This could be the case in Florida and California law, should the
state administrations wish. Of course, the State of Florida, and possibly the State of
California, would be required under their laws and regulations to pay the treasure hunt-
ers for anything they kept. Another variation wquld be that in which some, but not all,
90 W.A. Cockrell

artifacts are in permanent public ownership. This is currently the status in the State of
Florida and has been for over 20 years, and could probably be the case in California.
At the other extreme would be the case in which all artifacts remain in private
ownership, and this is, of course, the most common case when artifacts are looted from
wrecks without any kind of government control. An innovative suggestion, to say the
least, would be that of the Seahawk group in which artifacts would first be in public
possession but then go into private ownership after the public had time to view them,
and the archaeologists had time to study them.
A final variation would be the case in which only the worthless artifacts are given
to the public; this is primarily the case with the Atocha, in which the treasure hunters
gave the Florida State Museum in Gainesville monetarily valueless artifacts that they
couldn't sell, in an attempt to lend professional legitimacy to the treasure hunting op-
eration; this move also significantly increased the value of the stock and the market
value of the remaining artifacts held by the treasure hunters.

7. Profit Motive. Most treasure hunting operations are run on stock sales. As
demonstrated by extensive court testimony, the Atocha is a classic case of stock sales
gone mad. More recent is Seahawk, the operation that found the so-called Merced; they
ran toll-free numbers in Florida newspapers attempting to sell stock. When I called the
number, identified myself, and told them I wanted to know more about it, they frankly
told me that it was a risky venture and I might not get any return. In the 1970s' case of
the Atocha venture, the archaeologist for the Mel Fisher organization was paid in stock.
His stock would be valuable only insofar as he legitimized the operation; should the
operation have lost legitimacy, his stock, which was his principal payment for all of the
years he worked, would have been worthless. In this case, the archaeologist's conclu-
sions must be suspect. If he draws conclusions that make the stock more valuable, his
pay goes up; if he draws conclusions that make the stock less valuable, his pay is less
or nothing.
A further variation on the profitability category would be one in which a state and
its public museums are paid in artifacts. This is the case with the Museum of Florida
History in Tallahassee, and the Florida Museum of Natural History in Gainesville. The
former took artifacts from treasure hunters as payment to the state for allowing the
treasure hunters to work publicly owned shipwrecks; the latter, while not officially per-
mitting the initial destruction, profited (by garnering artifacts) from the activity another
state agency permitted.

8. Transferring Legitimacy. Another area where variable opinions can be found

regarding acceptability would be that in which private archaeologists are paid to over-
see treasure hunters, as in the case of the Whydah in Massachusetts. A variation would
be one that government archaeologists are paid to oversee the treasure hunters. This has
remained clearly the case in Florida and South Carolina, and has been the case in Lou-
isiana and other states and would apparently be the case in California once their pro-
grams are under way. The state is paid in artifacts: in legal fact, however, the state pays
in artifacts, as the case is in Florida.

9. Artifacts as Currency. Another useful analytic category is that of the artifacts

as currency, or valued goods. The State of Florida, under Florida law, owns all materi-
Why Dr. Bass Couldn't Convince Mr. Gumbel 91

als abandoned on the seabed. When a treasure hunter contracts with the State of Florida
to bring them up, the treasure hunter is legally acting as a subcontractor or employee of
the State of Florida, and is paid for services. The treasure hunter can be paid in cash, or
paid in kind, that is, artifacts. The State of Florida has never paid cash; it has always
paid in artifacts. In this case, artifacts become currency, or valued goods. This could
also come to be the case in the State of California.

10. Practicality. The common claim is that if archaeologists don't work with sport
divers or treasure hunters, the data will be lost. This is most often made to justify
archaeologists working with the Atocha looting expedition. This is currently the State
of Florida's justification for sanctioning the Maple Leaf operations. Another pragmati-
cally based claim is that if we don't work with the sport divers or treasure hunters, we
will lose our laws. This was the case in Florida over the years. As a matter of fact, Flor-
ida ultimately did effectively lose its law; it has essentially lost all effectiveness in its
ability to prevent most of its shipwrecks from being worked by treasure hunters. With
rare exceptions, if an organization identifies a wreck it wants, and has the necessary
capital, the state is effectively unable to deny a permit, or "contract." In 1983, Florida's
Secretary of State, the head of its preservation program, chose to drop out of an ongo-
ing court proceeding and allow the law to operate weakly rather than have the law
tested in the federal appeals court, where it seemed destined to be upheld and greatly
strengthened.
Another pragmatic concern is heard in the claim that if archaeologists don't work
with treasure hunters or sport divers, they will lose public support. In the case of Flor-
ida, I was partially responsible to an advisory council that was made up state agency
representatives, professionals, and members of the general public. A well-respected and
prominent history professor on our panel said, "Let the public play with their history,"
advocating continued treasure hunting, as a response to my request to stop the destruc-
tion. This led the agency head to say to me, when I again asked to stop treasure hunt-
ing, that "there is no public support; even your own colleagues don't agree with you."
He was correct.
One last pragmatic concern is exemplified by the Central America, an operation
that claims legitimacy for an interesting reason. According to the chief scientist, cited in
an October 1, 1989, UPI article, they have an estimated $450 million in gold on the Cen-
tral America, which went down in the 19th century; also, however, they have found that
there are a number of beer and soft drink cans on the seabed. This led the chief scientist
to state that "the project's main mission is a scientific one," after saying that he wants to
check the cans to see if they provide habitat for some unique marine form of life that
might contain some chemical they can synthesize to kill human cancer cells. With this
possibility, he lost interest in the treasure, we're told. It strikes me that here the pragmatic
aspect is clearly that the treasure hunter is not a treasure hunter; here he has undergone a
transmogrification and has become both humanitarian and environmentalist. So, treasure
hunting in this case, at least, is a boon to humankind, pragmatically speaking.

THE ULTIMATE CONCERN

The question thus arises, "How do we determine our ultimate concern, that which
is really important to us?" To clarify my thinking, I asked several colleagues what is
92 W.A. Cockrell

the most important thing we have to do as archaeologists in protecting shipwrecks?

What are our key concerns?

I. Cooperation. One colleague told me that his key concern was to see coopera-
tion between archaeologists and sport divers and treasure hunters. I thought this was
particularly self-serving, because in his position the administration would not allow
him to be hard-line about preservation, and had let him clearly understand that his job
depended on making sport divers and treasure hunters happy. I might add that he is
doing a good job of making sport divers and treasure hunters happy.

2. Trade in Artifacts. Another key concern expressed by a colleague was that the
most significant aspect of shipwrecks as sites would be the artifacts, and that stopping
of the trade in artifacts was a primary means of protecting the resource. This a very
important concern, because the artifact market is a vacuum into which the sites flows.
The site may be conceptualized through the use of a three-part model, as I have
noted in earlier publications (1981, 1982). Using this heuristic device, the site can be
perceived as data, as structure, and as artifacts (Figure 1). Once we've extracted the
artifacts from the site, all aspects of the site are destroyed, that is, the entire site is
destroyed. As long as there is a market, there are unscrupulous individuals who will
continue to destroy irreplaceable resources to feed the market. The way to stop the
destruction of the resource is to stop the consumption of these endangered resources. If
the demand stops, the destruction stops. If the demand ceases, one large aspect of the
destruction will stop, leaving natural causes and development as the main threats to the
preservation of the archaeological site.

3. Conservation of Artifacts. Some colleagues felt that the most significant ele-
ment of concern was the conservation of the artifacts. Some feel that treasure hunting is

Figure 1. Aspects of the archaeological site. Adapted from Cockrell (1983),

Why Dr. Bass Couldn't Convince Mr. Gumbel 93
acceptable if the artifacts are properly preserved, even though they go into private
hands. One state that allows treasure hunting also allows its conservation lab to con-
serve shipwreck artifacts regardless of whether they were recovered under state super-
vision or not, and allows its employees to work privately for treasure hunters at home
and in other states. This clearly lends legitimacy to the treasure hunters who see artifact
preservation as the "professional" and ethical way to deal with the destruction of
archaeological sites. Additionally, by hiring the state's expert, they assume a mantle of
respectability in their dealings with other governments, the media, and the public.

4. Ownership of Artifacts. Some archaeologists, as noted, feel that the primarily

ethical question is the ownership of artifacts, that is, whether they are publicly or pri-
vately owned. As long as the artifacts are publicly owned, some archaeologists would
not strenuously object to having the artifacts recovered by treasure hunters. Most
archaeologists today work with operations that have as their end result the private own-
ership of the artifacts.

5. Preservation of the Site. Some colleagues felt that the total preservation of the
site is very significant, with excavation being done only by professionals under control-
led conditions. This would be the case with sites owned and controlled by NOAA, the
National Park Service, or the State of Texas. The most common model, however, is the
conservation of the site, that is, utilization of the site as a resource to be mined under
more or less government supervision. Again, the states of Florida, Massachusetts, Lou-
isiana, California, and South Carolina are prominent.

6. Only Professionals Dig. Some colleagues felt that the key to controlling dam-
age to sites would be that only professionals dig, but it is very, very unclear as to who
or what is a professional, once we attempt to start defining criteria. It was hoped that
establishing SOPA would resolve the problem, but it did not.

7. Public Approval. Some colleagues felt that excavating shipwrecks would be

acceptable as long as the media, the public, and Congress or legislatures approved. Some
feel that the most important aspect here would be the legal standing; however, some
nations don't have legal protection, and the degree oflegal protection varies from nation
to nation. Indeed, legal protection varies within our own federal government. We find a
bizarre situation in which one branch of the federal government, the judicial, allows the
Atocha depredations to continue, after the executive branch attempted to stop it, while
another section of the executive branch assists the South Carolina treasure hunting group.
The guidelines for the Abandoned Shipwreck Act make some very strong statements
about the desirability of retaining public control of publicly owned antiquities, but the
Act itself allows the states to continue the destruction of the resource.

8. Colleague Approval. Some colleagues clearly felt that colleague approval was
the key, but our colleagues have vacillated during the 20 years I've been attending to
the problem. The Society for Historic Archaeology has wavered somewhat constantly
until very recently. This current conference (SHA-CUA 1990) is apparently the first
one in which treasure hunting papers have been effectively barred. The Conference on
Underwater Archaeology has been even more forgiving of treasure hunting than has the
94 W.A. Cockrell

SHA. The Advisory Council on Underwater Archaeology has probably been wavering
more than either SHA or CUA. When I was a member of the ACUA, I requested that
the ACUA censure the State of Florida, and me, as the State Underwater Archaeologist,
for allowing treasure hunting to go on. I did this because I felt that ifI had strong back-
ing from the profession for shutting treasure hunting down, I would be more able to
effectively stop treasure hunting in the State of Florida. I was never able to get that
from the ACUA; I am pleased that they are taking stronger stands recently, but some
issues are still ignored.
When the National Park Service, working with the School of American Research,
held the Shipwreck Anthropology Advanced Seminar, which ultimately resulted in the
publication of the book Shipwreck Anthropology (Gould, 1983), one of the primary
intentions was to attempt to call attention from the anthropological profession at large
to the double standard that existed with reference to shipwrecks and terrestrial sites. It
was constantly pointed out that they would not allow the mining of Mimbres pots from
a Southwestern site, or bricks from Independence Hall, yet these same colleagues
expressed no concern at all about our underwater sites being mined.

SUMMARY
In summary, we have seen that opinions and definitions vary on who should do it,
what should be done, how it should be done, when it should be done, and where it
should be done. Opinions vary on ages or types of significant sites. Opinions vary on
the laws and legal interpretations. Opinions vary on the definition of what is profes-
sional. Is a professional one who proclaims oneself to be? Is a professional one who has
had an adequate level of training, or one whose ethics reach a certain level? Or one
who is a government employee, as opposed to a private contractor? Opinions vary on
what field is professional, whether it be anthropology, history, conservation, or engi-
neering. Can we define ethics and ethical behavior? I would imagine that we probably
could. Is the real threat the traffic in artifacts? I feel that the real threat is apathy or
ignorance. A colleague said to me recently that I should not be so distressed about the
destruction because the situation is getting better. I thought this was a particularly lim-
ited and naive view, as the situation is probably not getting better at all. But even if it
were to slowly get better, the sites are nonrenewable. Every time we lose a site, we
have lost something that simply cannot be replaced. We should mourn the passing of
each site as the loss of an irretrievable part of the record of who we are, where we've
been, and a clue to what we might become. We have lost a page in the manuscript of
history, one that cannot be retrieved. Even if the situation slowly gets better, it may not
get to be acceptable before we've lost essentially all of the resources.
I feel that ultimately the real threat is to the data base, that is, to the site as a
whole. I constructed the three-part heuristic device (Figure 1) as an attempt to clarify
my own thinking and to create operational definitions to clarify communication. I find
that I will frequently be talking about preserving the site as a structure, or the site as a
whole, when others are talking about artifacts, or I might be talking about preserving
the data base, while others are thinking about the structure. The heuristic device was
created as a teaching and a communication tool. Utilization of the device allows us to
operationally define our concerns, and see more clearly what we're talking about. The
structure is the concern most often of managers, who want to keep a site intact for the
Why Dr. Bass Couldn't Convince Mr. Gumbel 95

public or students to view and visit. The site can be seen as artifacts by the archaeolo-
gist, or as goodies by the treasure hunter, or collections by the museum. The site can
also be viewed as a data base; this is how professional archaeologists most often see a
site. This model, when used to deal with the profession, or with the public, can assist in
explaining what we as preservationists or managers are talking about. It is usually the
public perception that archaeologists recover things, and that the interest of archaeolo-
gists is to recover things. I find it is frequently very difficult to convey that the ultimate
goal of archaeology is not to find "things," but to gather information, and that a very
important goal of managers is to preserve the structure of the site, as well as the arti-
facts within it. If we see the data base as the ultimate concern of the anthropological
archaeologists, it can be seen that damage to the data base is damage to the artifacts,
and that what we must preserve is the site as data base. When there is a market in arti-
facts, the data base flows into the vacuum of the artifact market.
Finally, to return to the title "Why Dr. Bass Couldn't Convince Mr. Gumbel" the
answer may be that it can't be done at present. We may not be able to convince the
media or the public. A large part of the reason is that we can't articulate our concerns
adequately, or agree among ourselves. How, then, do we try to convince Mr. Gumbel,
the public, and even ourselves? I suggest the three-part heuristic device is a very useful
tool in this process. This must follow the clear explanation to ourselves, as well as the
public, that archaeology is a destructive process. We as archaeologists destroy the con-
textual relationship of the materials with their surroundings in order to get at the data,
and to get at the artifact. Although it may be that some day we may be able to do non-
destructive remote sensing investigations that will produce a large data base, currently
archaeology is very destructive. It needs to be clearly understood by ourselves and by
the public that archaeologists and treasure hunters alike destroy the site. Then, we must
explain how we justify, or we try to justify, that destruction, and distinguish the ulti-
mate results of that destruction by us from the destruction by them. I feel that we may
not succeed, because this concept seems to be too abstract even for our colleagues. As
noted, there has been, and there continues to be, a double standard. Our own colleagues
refuse to demand the protection for underwater sites that land sites have. There is cur-
rently an anti looting campaign by the Society for American Archaeology that can be
contrasted with an almost bovine acceptance of the ineffectiveness of the Abandoned
Shipwreck Act. However, I knew, as well as did some others, that all the Abandoned
Shipwreck Act would do would be to put the states back in control, after the treasure
hunters' lawyers had weakened the states' control. The ultimate need is to accord ship-
wrecks, as archaeological sites on public property, the same protection as any other
archaeological site on public property. Archaeologists who would blanch at the thought
of commercially mining artifacts of Chaco Canyon think nothing of the State of Florida
allowing treasure hunting boats on both the Atlantic and Gulf coasts to work all sum-
mer long mining artifacts from archaeological sites.
I feel that, in the end, we all need to understand clearly the nature of the resource,
that is, the site as data base. We need to clearly define our goals. We need to agree on
professional and ethical standards. And once that agreement is reached, we must
clearly label all departures from professional or ethical behavior as such, and treat
those who make the departures accordingly. I feel that we have all failed to meet purist
standards, either through self-exemption or self-justification. Some of us exempt our-
selves out of apathy or ignorance; some of us say that we've lost many sites, but the
96 W.A. Cockrell

situation is getting better. Ifwe perceive the situation as getting better, and the sites are
being lost, it is not getting better at all; we are perceiving the situation incorrectly.
Some exempt themselves, as noted, by justifying their lack of concern for the type, age,
or the provenience of sites. Some exempt or justify their behavior based on legal or
technical reasons. Some exempt themselves in the area of artifact disposition. Some are
just pragmatic. Given this situation, with all of these self-exemptions and self-justifica-
tions, can we ever define the ethical position?
So who will cast the first stone? I can't, and you don't seem to be able to: maybe
none of us can. Perhaps that is why it's so difficult to convince Mr. Gumbel. We can't
convince ourselves.

REFERENCES
Cockrell, W.A., 1980, The Trouble with Treasure: A Preservationist View of the View of the Controversy.
American Antiquity (45:) 333-339.
Cockrell, W.A., 1981, Some Moral, Ethical and Legal Considerations in Archaeology. In Realms of Gold: Pro-
ceedings of the Tenth Conference on Underwater Archaeology, edited by W.A. Cockrell, pp. 215-220.
Fathom Eight San Marino, California.
Cockrell, W.A., 1983, A Trial Classificatory Model for the Analysis of Shipwrecks. In Shipwreck Anthropol-
ogy, edited by R.A. Gould, pp. 207-217. University of New Mexico, Albuquerque, New Mexico.
Gould, R.A., ed., 1983, Shipwreck Antropology, University of New Mexico Press, Alburquerque, New Mexico.
Scott, N., 1981, The Central America. In The Realms of Gold: Proceedings of the Tenth Conference on Under-
water Archaeology, edited by W.A., Cockrell, pp. 95-99. Fathom Eight:, San Marino.
 Chapter 10

'.~~.' Archaeology on Trial

~

o i' .
q;,O_' .': .
.j'~l R. DUNCAN MATHEWSON III

Finding the hull of the Atocha took 16 years. Recovering the cultural wealth of the site
could take even longer. The wrecks of the 1715 treasure fleet, first located in the 1960s,
are still yielding valuable archaeological information and quite a bit of treasure. While
Kane's dramatic discovery of the main pile ended Mel Fisher's quest, for the archae-
ological team this is only the first phase of a long, arduous process. I can't predict how
long it will take to raise, clean, catalog, preserve, and then study and report on the
300,000 or so artifacts we eventually expect to recover from the Atocha and Margarita.
Mel has won his battle with history, but for me, the archaeology of the project is still
very much on trial.
When I joined Treasure Salvors on Independence Day, 1973, I wasn't just the only
archaeologist working with Mel - I was the only archaeologist in the country who
would work with a commercial salvage operation. I believed then, and still believe, that
archaeologists and commercial salvors can work together on historical sites. To my pro-
fessional colleagues, that's a radical notion. They believe that the interests of archaeolo-
gists and those of shipwreck salvors are diametrically and unalterably opposed.
As a result, for most of the time I've been associated with Treasure Salvors, I've
been blackballed in the archaeological community. On several occasions, I was pre-
vented from giving papers at archaeological conferences and have been discouraged
from submitting reports on the Atocha site to professional journals.
When I joined Mel, I had three major goals: to map the site and describe the find-
ings, to interpret the artifacts, and to tell the public and the academic community every-
thing I could about the 1622 ships.
Source: Treasure of the Atocha, pp. 116--123. New York: Dutton, 1986. Reprinted by permission of the author.

97
$cattered ]irtifacts-----------

,..... ......, .
,I................ 'e, ••
...... ... .
,.' ...... . f..~"". "',\",.......::.....:
,
-: '"
'. ",......... -.....:.
.~. " "
"........
.~~:: .,' _ ....
.......
THE QUICkSANDS ,.1 "
..... ,. 1 ., •
,1-"

..... .'
'

.'

...... .. ....":::,
.. ..... ....
:.,

,.... . ......
••••• "... ••• ........ • Upper Sterncastle Strud~~'~"
..•...••..••. " _.' ••.••.•:.
•••••• I, •
........ ''''''', ............................
,..................... ,.. SANTA r-::'-'
... ,.
~

MARGARITA: ....:t..!. :
2 gal/eon anchors ----- ."
,...... "
b
'. ./' 1 ronze gun
d:• ~ •
:
• ... • • ...... ':
" X. gal/eon anchor
",,"''''... Ear/ier A tocha Finds: :
,:.'
•• ~ ••••••••• , . . . . . . .
........ ...
.... ............................... ...................... .
" ... ~ "'
. . . . . .- .
":'• I

....... •••• ~
galleon anchor
" ; 9 bronze guns
I.·.. .·t·;i:· _. ·4···· ..""
.. _........
......
"X.f..' ':-
" ••••••
Margarita" f'
a ••••• ~ l

••••••••
••',..
•••
.... ••••
• •••••••••••••••.••••• ' ...,.•••• ,,'
Bank of Spain 1",,";"---
..........
~
cattered
Impact
t
• • ••••••• ••••• Patch Reef art/£.
aCts

M,~,;,I W~k'nd I/nd~/~ iron swivel gun

HAWk CHANNEL 1
3 Margarita
Galleon
Anchors
... _........
....
,.. ........ .I::
.......... .-':.
". --.-
......
OUTER REEF

A schematic map of the sinking and break-up of the Atocha and Margarita.
 '. ..' .'
.:
.'

.
.,,-
",
,,'

..... ....
" ......... .
,"

.... ,......... ."

,'_.

."'...... .
• ,t ..... .f' ••••;:
• • • _ . . . . . . : ••• ~..... ,t ,t "

f... , .. ,..... .'

,.-". I,. • • .. •••••• ." .1'

,. "
." .. , .......... .
.'
I

.'
.' .'

.............:
• ,t"
....... ,f

.~ .... -', -,
,"

,
• ' ••••••

.... If... ................... ,,' ."........

. . . . . . . . _. " . . . . ••••• ..... I f . . . . .. "

_t"
• If

to . . . . . , . . . . . .
: : :•••• ".f • ::. .. :::...... .

... .. ....:....... .
............

....... ...',............. .
• I, •••••••• • ......... : ••••
:
,
NUESTRA
SENORA
. . . . .,

... .:........
• I , ••••

...•... .,,'
-................
•

DE A TOCHA .........

~
+ Lower Hull Structure

................ .................... ...... ....... '0 ............................

",
••••
.......... .
• •••••••••

:,.,
_
"',. J' tt,

~...;.
'"
.j~:.'~-'. . . ". '. ....... .
..... a.....
. _.a ...........
............. ..: ...:::.....•.........._....
f . ' aO . /• • • • •: ,'... . .. . . .

•••. .•... -N- ...•..

~ Not to scale. Bottom topography only generally outlined.

 100 R.D. Mathewson III

To map and record our finds, I worked to weld the treasure divers into an archae-
ological team and to train other company personnel in archaeological mapping, and
curating and drawing artifacts.
In the early years, I was the only one in the company with archaeological experi-
ence. I did everything from mapping the site to cataloging the artifacts. As I trained the
staff to do more of this work themselves, I was able to spend more time on the second
goal- interpreting the finds.
This required more archaeological knowledge than I was able to convey to the
divers as on-the-job training. It therefore entailed tapping the specialized knowledge of
other professionals. Slowly, people with academic backgrounds like Claudia Linzee,
Jim Sinclair, and Austin Fowles, joined our in-house staff. At the same time, outside
professionals from universities and museums were brought in to help improve our
expertise and to make their own studies of our material. The one-person archaeological
staff has now grown to a team of 15, and at present, more than 20 outside scholars are
involved with the interpretation of the Atocha and its artifacts.
My third goal is to educate the general public about the meaning of our finds.
This has already begun. We have plans to expand the number of publications, as well
as museum exhibits, a traveling display, films, and video programs already developed.
A permanent museum has been built in Key West to house the core of the artifact col-
lection. Two films have been produced for TV by National Geographic about the
search for the Atocha. Millions of people all over the country will see for themselves
some of the wondrous things Mel and his divers discovered. And although it will take
years to fully study the vast number of artifacts, a preliminary scientific description of
the site and the artifacts is scheduled for publication in 1988.
Treasure Salvors has made an enormous investment in treating, conserving, curat-
ing, and making available for study the artifacts and knowledge recovered from this
site. The salaries of the professional staff, the conservation lab, and the expense of a
custom-designed computer installation cost Mel- who once couldn't even afford to
pay his divers - hundreds of thousands of dollars annually. Despite these expenditures
and the sincere efforts of the Treasure Salvors staff and other scholars brought in as
consultants, I am certain that all of us will be roundly criticized by the academic and
government archaeologists. "Not precise enough," they'll say. "Not careful enough.
Not slow enough. It was still a treasure hunt."
There is merit to the basic argument behind some objections. Archaeology is a de-
structive science. Once a site has been excavated, it can never be put back together. The
contextual data, the spatial relationships that are so critical to proper interpretation of a
site, are destroyed. Yet, historical sites are part of our common heritage, and the informa-
tion they contain belongs to everyone. In the process of excavating a site, the archaeolo-
gist assumes a responsibility to the public to proceed carefully, extracting every morsel
of data from the site as it is dismantled. Because so much is unknown about everyday life
aboard ship, every artifact, no matter how small or seemingly insignificant, might be
important. Each must be mapped as it is removed, then cleaned, preserved, studied, and
curated. Treasure hunters usually can't be bothered with digging slowly, with careful
mapping, or with tagging and conserving every splinter of wood, every shred of pottery.
The treasure hunter is only concerned with the valuable artifacts on a site.
For these reasons, the academic archaeologists would prefer that historic ship-
wrecks be left on the bottom, intact, until a nonprofit institution can fund their excava-
Archaeology on Trial 101

tion. In effect, if they had their way, they would shut the doors to anyone who doesn't
have an advanced degree in anthropology or archaeology. They've promoted federal
legislation that would do just that. If the wrecks could remain safe indefinitely on the
bottom, this approach might make sense. But they won't. As our work on the 1622 gal-
leon shows, shallow-water shipwrecks on exposed reefs and shoals are continuously
deteriorating as their artifacts are shifted around on the bottom. Not only have the Ato-
cha artifacts in the Quicksands been uncovered and exposed repeatedly to destruction
in the ocean for over 350 years, but the Margarita's hull was almost certainly exposed
to periodic deterioration before we found it.
I believe our work on these sites will eventually vindicate my conviction that
good archaeology is possible during a commercial salvage effort. There certainly are
pressures to work quickly, but archaeologists have successfully excavated land sites for
years while staying just one step ahead of the land developers and highway contractors.
We did not move as slowly as we would have in an ideal archaeological expedi-
tion, but we were not on a treasure hunt either. Treasure hunting never involves serious
record keeping, and we have an extensive record of our work. But a full-blown scien-
tific expedition, a comprehensive interdisciplinary project, was impossible on this site.
Somewhere between these two extremes, I believe I developed a sound archaeological
salvage program. We did the best we could under the circumstances, and a lot more
than any government or academic agency would have done to find and recover the
1622 fleet. Mel said it in 1977: "We go through all kinds of hell out on that rough
ocean. We're risking our lives and doing a lot of dangerous things. The state officials
would not do that. The people they hire don't have the incentive, the will; they don't
have that feeling for the hunt. They aren't going to be rewarded even if they do find it.
To them, it's just another paycheck and I don't think they would ever succeed."
If Treasure Salvors had not found and recovered the ships, they would still be sitting
down there, deteriorating. There isn't a museum in the country that has the endowment
money to gamble $8 million on the chance of finding a historic shipwreck. Mel Fisher did.
Within recent years the federal government has sponsored only one historic ship-
wreck project - a survey of the ironclad USS Monitor. The State of Florida, with its
Spanish heritage and thousands of important wrecks, has never paid for the excavation
of a single historic vessel. If, as they claim, the state is so eager to do good ship archae-
ology, they could map the 1715 and 1733 sites or they could locate and excavate the
fleet of the French Huguenot Admiral Jean Ribault. Ribault sailed to Florida in 1564 to
relieve the beleaguered French colony at Fort Caroline, on the site of what is now Jack-
sonville, but his flotilla was ambushed by the Spaniards. Their ships captured or
burned, the French Protestant sailors were taken ashore at Matanzas Inlet and asked,
one by one, to convert to the Roman Catholic faith. When they refused, each was
beheaded. Their defeat effectively ended the French challenge to Spanish power in
southeastern North America. It's doubtful that Ribault's ships will ever be recovered by
commercial salvors, yet their excavation would add immensely to our knowledge of
Florida's early history and of French seafaring technology.
The issue facing all of us is not whether commercial salvors or nonprofit groups
actually do the work on these sites. Our primary concern should be to preserve the
eroding archaeological information. Commercial salvors are the only ones who have
the money to recover these precious time capsules before they dissolve in the seawater
that surrounds them.
 102 R.D. Mathewson III

Since my pioneering work with Treasure Salvors began, more archaeologists

have gotten involved with the private salvage companies as university jobs and govern-
ment grants have dried up. Now, there are more than a dozen archaeologists working as
consultants for salvage companies. The trend is accelerating.
The genie is out of the bottle. Archaeologists no longer have the legal right to bar
private salvage companies from working shipwrecks. The long court battles fought over
the Atocha have crystallized a whole new body of laws dealing with historic shipwrecks,
affording new options to salvors everywhere in the country. The years of haggling have
left us with a much simpler set of guidelines, as defined by federal judges. First, ship-
wrecks more than 3 miles from land can be placed under federal jurisdiction and fall
under the provisions of admiralty law and what is known as the "American Rule." The
American Rule, essentially, amounts to "finders-keepers" when it comes to abandoned
shipwrecks. A more recent Florida ruling stipulates that admiralty law has precedence
over state laws even on shipwrecks within the 3-mile state territorial limits. The courts
have established a tradition that if admiralty law is to be applied to a historic wreck, the
salvor must apply acceptable archaeological controls during its recovery.
There is a danger that uncontrolled salvage and depredation by hobby divers
could result in the destruction of a great number of shipwrecks. I believe the way to
prevent this is through education. A former adversary has joined me in this effort.
Charles M. McKinney III, an archaeologist for the U.S. Department of the Interior who
fought me on the Atocha project, has helped form The Atlantic Alliance for Maritime
Heritage Conservation. This national nonprofit group headquartered in Washington,
D.C., brings together divers, shipwreck salvors, archaeologists, historians, and hobby-
ists from all over the country to work together to preserve our maritime heritage. The
Alliance sponsors workshops across the country that teach recreational divers the
basics of underwater archaeology and gives them respect for the historic value of ship-
wrecks. Prohibitions rarely work. If recreational divers want to find historic ship-
wrecks, they will find them. If government policy is to keep sport divers off of wrecks,
when the divers find wrecks, they'll keep the locations secret and salvage them slowly
and incompletely, destroying the sites in the process. In the Alliance program, they are
rewarded for reporting the wrecks to professional archaeologists by being encouraged
to assist in their study. The alliance seeks to bring people into the process instead of
locking them out. Paraprofessionals, such as sport divers with some basic training, can
do valuable work on historic shipwrecks. So can commercial salvors.
Three years ago, a number of wrecks from the 1715 fleet were salvaged by com-
mercial groups working with Treasure Salvors. Using archaeological guidelines devel-
oped on the Atocha site, a 12-member team worked with six different groups of salvors
along a 50-mile stretch of Florida's east coast. This was the largest shipwreck project in
the country. It directly involved over 75 people, including salvors, archaeologists, state
officials, and graduate students. In 3 months, the project amassed more archaeological
data and reports on these sites than state archaeologists have been able to generate in
the past 20 years. This was possible largely because we were working under the juris-
diction of federal admiralty law instead of under state statutes. The salvors were very
helpful, because they were treated as an essential part of the project instead of being
harassed by the archaeologists.
All of the shipwreck research, artifact conservation, museum exhibitions, and
publications sponsored by Treasure Salvors, Inc. have been produced without a single
Archaeology on Trial 103

taxpayer's dollar. The Atocha project vividly demonstrates that archaeologists should
learn to use the profit motive to preserve the integrity of wreck sites being legally sal-
vaged by commercial companies.
John S. Potter's The Treasure Diver's Guide, in 1960, laid the foundation for the
treasure hunters of the following two decades. Now, the study of the 1622 sites - com-
bining both historical and archaeological research - is laying the foundation for ship-
wreck salvage over the next decade. A scholarly approach to the study of the Atocha and
Margarita has shown that there is no longer a place for treasure hunting in the 1980s.
Concern for the cultural value of the 1622 sites turned a treasure hunt into a project with
profound implications for the future of shipwreck studies throughout the Americas.
Nothing dies that is remembered. The salvage of the vessels of the 1622 fleet has
jogged our collective memory by bringing to light the property of men and women who
colonized the Americas. It is these memories that Treasure Salvors, Inc., Mel Fisher,
Gene Lyon, and I have worked so long and so hard to preserve. Forget the jewels, for-
get the gold, forget the silver. These memories are the real treasure of the Atocha.

SELECTED BIBLIOGRAPHY
Lyon, E., n.d., 1622 Fleet Loss Narrative. Ms.
Lyon, E., n.d., List of Gear and Equipment and Some Construction Details ofNuestra Senora de Atocha 1622.
Ms.
Lyon, E., 1975, Data on the Identification of Shipwreck Site 8MOl41 in the Marquesas Keys. Florida. Ms.
Lyon, E., 1976, The Identification of a 17th-Century Spanish Galleon Nuestra Senora de Atocha. Paper presented
at the St. Augustine Historical Society.
Lyon, E., 1977, Spanish Cultures in Colonial Florida and Their Connection with Historic Shipwrecks. Paper
presented at the Conference on Florida Historic Shipwreck Archaeology.
Lyon, E., 1980, A Historian's Thoughts on Some Shipwreck Models. Paper presented at the conference on
Maritime Cultural Heritage of the Florida Keys: How Can It Be Preserved? Florida Endowment for the
Humanities, Key West.
Lyon, E., and Mathewson, R.D., III, 1975, The Historical and Archaeological Meaning of the 1622 Shipwrecks
Off the Marquesas Keys, Florida. Paper presented at the Florida Historical Society Conference, Gaines-
ville.
Lyon, E., and Mathewson, R.D., III, 1975, An Introduction to the Ethnohistory of the Lower Florida Keys. Paper
presented at the American Society for Ethnohistory Conference, Gainesville, Florida.
Lyon, E., and Purdy, B., 1982, Contraband in Spanish Colonial Ships. Itinerario: Journal of the Institute of
European Expansion. University of Leiden.
Mathewson, R.D., III, 1975, Historical Shipwreck Archaeology: New Developments fr(lm the Lower Florida
Keys. The Conference on Historic Site Archaeology Papers, 1973, 8: 121-128.
Mathewson, R.D., III, 1975, A New Methodological Approach to Shipwreck Archaeology. Paper presented at
the Society for Historical Archaeology and International Conference on Underwater Archaeology,
Charleston, South Carolina.
Mathewson, R.D., III. 1975, Historical Shipwreck Ceramics: A Preliminary Analysis of Olive Jar Data from the
Wreck Site of Nuestra Senora de Atocha (ms.), Department of Anthropology, Florida Atlantic University,
Boca Raton.
Mathewson, R.D., III, 1975, Archaeological Recovery: Its Potential and Limitations on New World Shallow
Water Sites (ms.).
Mathewson, R.D., III, 1976, Introductory Notes on Operationalizing a Procedural Model for the Conservation
of Archaeological Data from the Wreck Site of Nuestra Senora de Atocha (ms.).
Mathewson, R.D., III, 1976, An Introduction to the Numismatic Assemblage from the Nuestra Seliora de Atocha
(ms.).
Mathewson, R.D., III, 1977, Archaeological Research on the Wreck Site of the Nuestra Senora de Atocha: A
General Overview of the Mapping and Survey Procedures. Paper presented at the Society for Historical
Archaeology and International Conference on Underwater Archaeology, Ottawa, Canada.
104 R.D. Mathewson III

Mathewson, R.D., III, 1977, An Introduction to the Use of Aerial Photographic Imagery for Locating and
Interpreting Shallow Water Shipwreck Sites Off the Florida Coast (ms.). Department of Geography,
Florida Atlantic University, Boca Raton.
Mathewson, R.D., III, and Clyne, P., 1977, Digging Procedures Utilized in the Search for the Lower Hull Section
of the Nuestra Seiiora de Atocha. Paper presented at the Society for Historical Archaeology and Interna-
tional Conference on Underwater Archaeology, Ottawa, Canada.
Mathewson, R.D., III, Dorwing, J.T., Sinclair, U., and Marken, M.W., 1986, Atocha Archaeology: A Look at
the Anatomy of a Spanish Treasure Galleon. Paper presented at the Society for Historical Archaeology
and Conferences on Underwater Archaeology, Sacramento, California.
Mathewson, R.D., III, and Lyon, E., 1976, The Guns of the Nuestra Seiiora de Atocha (ms.).
Mathewson, R.D., III, Lyon, E., and McAlister, R.F., 1976, Atocha Data Evaluation Report (ms.).
Mathewson, R.D., III, Murphy, L., and Spencer, B., 1975, New Concepts in Marine Archaeology: Shallow Water
Historic Archaeology in the Lower Florida keys. Conference on Historic Site Archaeology Papers,
9:141-151.
 Chapter 11

American Naval
Archaeology
Past and Prologue

WILLIAM S. DUDLEYI

During recent years, several U.S. Navy shipwrecks have been found, rediscovered,
explored, and exploited under varying conditions and for different reasons. The ships
involved were the USS Somers, USS Kearsarge, USS Tecumseh, and USS Cumberland.
The French discovery and exploration of the wreck of the Confederate States Navy
steam sloop of war Alabama in the English Channel off Cherbourg has become a sub-
ject of international contention. The facts differ but the issues are similar. These valu-
able historical artifacts are in danger of being exploited, looted, and ruined because of
the increasing numbers of human scavengers (sport divers, commercial fishermen, and
treasure hunters) who have begun to strip the wrecks.
The U.S. Navy's policy toward wrecked and sunken ships until now has been
guided by admiralty law. The United States does not lose its title to any vessel as a
result of wrecking or sinking, regardless of the passage of time. Unless the United
States legally abandons title in the manner directed by an act of Congress, the ship-
wrecks and their contents, including human remains, are U.S. property and are consid-

I The author is a civil service employee of the Naval Historical Center. but the opinions expressed herein are
his own and do not necessarily represent those of the Center or the Navy Department.

Source: Underwater Archaeology Proceedings from the Conference on Underwater and Historical Archaeol-
ogy. edited by Toni L. Carrell (1990). pp. 8~90. Reprinted by permission of the Society for Historical Archaeol-
ogy. Tucson. Arizona.

105
106 W.S. Dudley

ered water graves. Any person attempting to salvage a U.S. government vessel does so
at his or her own risk and is subject to the prior rights of the U.S. government. This
legal doctrine was set forth in the Hatteras, Inc. vs. the USS Hatteras case by the U.S.
District Court for the Southern District of Texas in 1980. The judge stated that the
common law doctrine and abandonment had no application in the case. Only Congress
has the power to divest public property under the property clause of the Constitution.
The executive departments of the government must fall in line with this ruling.
Thus, the policy of the Navy has been not to encourage private salvage operations
with respect to sunken or wrecked naval vessels, particularly when such operations
may disturb the final resting place or desecrate the remains of crew members, or when
other hazards such as unexploded ordnance may be present on board the vessel or in its
vicinity. This policy is occasionally relaxed when historical or other interests are
deemed to outweigh these policy considerations and when there is no conflict with the
law, both statutory and regulatory, property of historical, architectural, or archaeologi-
cal significance.
Presumably, this is what led Secretaries Middendorf and Claytor to acquiesce in
the transfer of ownership of the War of 1812 schooners Hamilton and Scourge to
Canadian authorities during the period 197&-1979. Secretary John Lehman came to
regret this act of his predecessors but was unable to reverse the decision in 1984.
Despite such precedents and prohibitions, problems continue to arise:
1. It has become increasingly difficult to prevent exploitation. The relatively in-
expensive technology and easily attained skills of scuba diving make that sport
attractive to many people. The development and production of remote-contro1-
led robots has not only improved the direction of underwater photography to
pinpoint accuracy, it has also made possible the retrieval of artifacts of increas-
ing size and weight at far greater depths than those that limit individual divers.
2. U.S. Navy shipwrecks are historical artifacts of a special nature. The ships
usually have detailed histories that make them part of the Navy's traditions.
Before it foundered during the Mexican War, the brig Somers was the scene
of an infamous mutiny and execution that raised a storm of criticism in the
United States and led to various reforms including the establishment of the
U.S. Naval Academy. George Belcher, a private citizen, located Somers off
Veracruz in 1986 and solicited U.S. government assistance to protect the
remains. Unfortunately, before anything could be done, unauthorized persons
visited the site and took away several artifacts including human remains. For-
tunately, the State Department, last November, negotiated an agreement with
Mexico for protection of the wreck site. The first dive should take place dur-
ing the summer of 1990.
3. The USS Cumberland, now beneath the waters of Hampton Roads, was an
important frigate of the American sailing navy before it became one of the
earlier victims of the armored, steam-powered CSS Virginia. Cumberland
went down fighting with great loss of life in 1862. Both salvors and archae-
ologists have investigated the wreck. Since 1983, U.S. Navy divers have vis-
ited the site, and so have those who have exploited the site. The most recent
proof of this was seen in a trade magazine advertising for sale belt buckles
and other items manufactured from metal extracted from the Cumberland and
American Naval Archaeology 107

CSS Florida. Several maritime archaeologists have contacted us about this

illegal activity. The Naval Investigative Service and Admiralty Branch of the
Navy Judge Advocate General's Office are looking into the matter.
4. The USS Kearsarge now rests at the bottom of the Caribbean, off the. Colom-
bian coast on Roncador reef. During a long and active life it gained fame as
the Union warship that hunted down the CSS Alabama and sank it in battle
off the coast of France in 1864. Two years ago, Kearsarge was the object of a
souvenir hunt financed and filmed by Hollywood Adventure Films. Two or
three years ago, the Naval Historical Center was asked to do historical
research to ascertain if any ordnance remained in or on the hull. According to
the Navy's records there was none in the area. It is likely that an early
attempt at salvage about 100 years ago may have lifted such items but the
ship could not be moved. Still the ship is within reach and could attract oth-
ers. Sovereignty in these waters is claimed by Colombia. It is foreseeable that
another "international incident" could occur requiring negotiations should
Americans see fit to assert sovereignty over the wreck.
5. Since 1984, when the French naval minesweeper Circe rediscovered the
remains of the Alabama in 180 feet of water in the English Channel off Cher-
bourg in French territorial waters, there has been a widening and increasingly
intense international debate as to how the wreck site would be protected, who
would be allowed access, and whether the ship or any of its artifacts would be
raised. The U.S. Navy, of course, never owned or operated this vessel and has
no technical right, title, or interest in Alabama. Vessels owned by the Confed-
erate States of America became the property of the United States on the surren-
der of the Confederate government. As an internal U.S. government matter, the
Alabama probably falls within the jurisdiction of the Administrator of the Gen-
eral Services Administration. However, we have learned that the GSA does not
claim jurisdiction because of the location of the wreck in foreign territorial wa-
ters. The really interested parties include: the French ministry of culture, Brit-
ish enthusiasts from Birkenhead because of the construction of the ship in the
Laird shipyard and service of British seamen in the Alabama crew, and Ameri-
can Civil War buffs who see both the French and British claims as unpleasant
interference with the United States's naval heritage. The domestic interest in
this country has now led to the introduction of two bills in Congress calling for
the return of the Alabama and its artifacts to the United States. The Senate bill,
sponsored by Howell Heflin of Alabama, specifies that state as the eventual
site of the artifacts. The House bill, cosponsored by Walter Jones of North
Carolina, is more general in its approach, implying that Alabama relics are the
possession of all of the people of the United States.
The Navy's position on this has been one of narrow construction, simply
to assert U.S. sovereignty of the site on the basis of international law. The
Navy does not "own" the wreck because although it was sunk by gunfire in
battle and struck its flag, the Union Navy did not board and capture the ves-
sel. As of October 3, 1989, the State Department negotiated an agreement
with the French government establishing a bilateral "scientific committee" to
oversee any future study of the vessel. It was this agreement that formed a
precedent for the subsequent U.S.-Mexican conversations in late November
108 W.S. Dudley

1989. In a separate communication to the State Department, the French con-

ceded U.S. ownership. The Navy will be reevaluating its position on the Ala-
bama in the light of these developments.
6. The monitor USS Tecumseh provides another example of confused and over-
lapping federal, state, and local jurisdictions with respect to a famous navy
shipwreck. As Wilson West has pointed out, the GSA asserted the Navy aban-
doned the vessel, and the Navy Judge Advocate General's admiralty office
confirmed recently that it was abandoned in 1901. For a while it was thought
that the National Oceanic and Atmospheric Administration might include
Tecumseh in its Marine Sanctuary Program, but NOAA's resources and per-
sonnel have been cut, and it is unlikely that there is a future for any other
shipwreck sanctuary program aside from the USS Monitor. Even the Monitor
Sanctuary seems devoid of archaeological activity for the time being. At the
same time, Tecumseh comes under the local jurisdiction of the Alabama His-
torical Commission, off Fort Morgan State Park. Private attempts to salvage
the vessel have come to naught, although there is evidence of tampering with
the hulL
7. The Navy presumably would be within its legal boundaries if it were to
embark in naval archaeological projects, but the prevailing attitude has been
that this was an inappropriate use of public resources in a federal department
authorized to defend the country in a maritime environment.
Yet, the Navy has discovered land archaeology and the Naval Facilities Engineer-
ing Command has the responsibility to develop policy in this area. Additionally, some
components of the Navy, such as the diving personnel under the Director of Ocean En-
gineering (Supervisor of Salvage) and the research vessels of the Deep Submergence
community, have become involved. Individual Navy divers have participated in under-
water archaeological projects on invitation from the National Park Service. But as
skilled as these people are as salvage experts, they are not archaeologists.
Although the Naval Historical Center is the most logical agency within the Navy to
monitor shipwreck sites, the Center has never been charged with the responsibility of
protecting, surveying, or retrieving artifacts of interest to the archaeological community.
Yet, a mechanism does exist. The Center's director has the title Curator for the
Navy Department and the Center has a Curatorial Branch, whose function is to receive,
record, and take custody of naval artifacts. From this inventory, the Center sends many
artifacts out on temporary or indefinite loan to schools, churches, and patriotic and vet-
erans organizations. This mission could be expanded to include more of the active
functions of naval underwater archaeology.
Recently, a reserve officer in the Naval Historical Center's Volunteer Training Unit
has undertaken to catalog the extant Civil War wrecks still underwater in varying condi-
tions of decay. He has discovered about 30, and there may be more yet undiscovered. This
list was published in an issue ofthe Confederate Naval History Society newsletter.

CONCLUSIONS

This discussion indicates both hope and fate are equally operative in maritime
and naval archaeology. Although many other examples of U.S. Navy shipwrecks could
American Naval Archaeology 109

be cited, these should suffice to show the existence of potential danger to U.S. naval
archaeology. On the other hand, some recent developments indicate that, with a show
of goodwill and respect, we may be able to make progress in protecting and studying
these historic maritime artifacts.
There are many obstacles to be overcome in managing domestic U.S. shipwreck
sites as the Cumberland and Tecumseh situations demonstrate. Overlapping state and
federal jurisdictions have created confusion and friction where, in one example, the
Virginia state underwater archaeologist wanting to protect the Cumberland could only
stand by and watch unauthorized, untrained individuals dive on the wreck because the
Navy's legal authorities were either unwilling or unable to prevent it. The application
of the new legislation on underwater shipwrecks has yet to be tested in the courts. Cer-
tainly, the states' enforcement attitudes vary greatly, from zeal to indifference, in the
way they view the exploitation of underwater wrecks and artifacts within their jurisdic-
tions. No one wants to see another HMS DeBraak fiasco develop where the Delaware
state authorities stood by while a salvage firm devoured a historically valuable wreck
on the chance it carried a payroll. It is conceivable that a U.S. Navy ship could receive
the same treatment.
At the Naval Historical Center, we are attempting to create an awareness within
the Navy Department of the abuses of historic underwater sites for commissioned air-
craft as well as ships and to establish a new mission for the Center. Unfortunately, this
comes at a time when we have all been warned of greater austerity in future Defense
budgets. Thus, our ability to hire additional personnel to undertake this new mission is
in doubt. For a while, at least, we will have to be content to monitor naval wreck sites,
to collect information systematically, to cooperate with authorities who have more
access to the sites than we do, and finally to alert those in positions to affect govern-
ment policy in Congress that the Navy has a newfound but abiding interest in historical
maritime archaeology. We will be relying on established information networks and
would be very interested in sharing our information with you and would welcome your
communicating your concerns to us on any situation regarding underwater navy wrecks
and artifacts.
The Navy's objectives in undertaking this new initiative are the following: moni-
tor U.S. Navy underwater wreck sites; help to protect the Navy's legal interests; pre-
vent destruction of historic Navy ship wrecks; discourage looting of naval artifacts;
provide sanctity of water graves of U.S. seamen; encourage naval archaeological/his-
torical studies; encourage nonintrusive surveys of wreck sites; encourage preservation
and display of artifacts retrieved; and encourage historical understanding of related
naval events.
 Additional Bibliography
for Part I.e

Cockrell, Wilburn A., 1981, Some Moral, Ethical and Legal Considerations in Underwater Archaeology. In the
Realms ofGold: The Proceedings ofthe Tenth Conference on Underwater Archaeology. edited by Wilburn
A. Cockrell, pp. 215-222. Fathom Eight Publications, San Marino, California.
Cockrell, Wilburn A., 1981 a, Panel Discussion: Crises in Underwater Archaeology. In the Realms of Gold: The
Proceedings of the Tenth Conference on Undelwater Archaeology. edited by Wilburn A. Cockrell, pp.
243-255. Fathom Eight Publications, San Marino, California.
Hopkins, Fred, 1985, The Involvement of Sports Divers in Underwater Archaeology in Maryland. Proceedings of
the Sixteenth Conference on Underwater Archaeology. edited by Paul F. Johnston, pp. 39--40. Society for
Historical Archaeology, Tucson, Arizona.
Jeffries, Roger and Kendall McDonald, 1966, The Wreck Hunters. London.
Kynott, Mark 1. and Alison Wylie (eds.), 1995, Ethics in American Archaeology: Challenges for the I 990s.
Society for American Archaeology, Washington, DC.
Marx, Robert F., 1990, The Underwater Dig. Pisces Books, Houston, Texas.
Potter, John S., 1972, The Treasure Diver s Guide. Bonanza Books, New York.
Tucker, Teddy, 1966, Treasure Diving with Teddy Tucker. Bermuda.

III
 Part II

Areal Studies

Dividing the world into component parts for specific study has general utility for
organizing a research effort. The concept of areal studies is derived from anthropologi-
cal efforts to organize research with respect to area, time period, and people. Here sites
and ship types are similarly consolidated into spatial contexts relevant to societies and
areas which produced watercraft and maritime sites.
Creating a context for sites is important but it can lead to difficulties. Where, for
example, should the ships of Columbus, Magellan, Cheng Ho, and other early explorers
be placed? They impacted the world but their historical and cultural context fits more
readily into the Iberian, western Mediterranean, or southern Chinese areas. In the new
world, study of ships of discovery seems to represent an "oldest is best" fixation, but
focusing on these vessels is worthwhile from a funding standpoint, as well as an educa-
tional approach to documenting the spread of European civilization. People are interested
in these periods and therefore support to learn about them is more readily available.
In this text, those areas which are well known, and for which a large body of pub-
lished material exists, are presented in bibliographic format only. Lesser known areas
are given additional space because they need more effort and a preliminary reporting of
material provides a baseline for future research.

113
 Part II.A

The Mediterranean:
Bibliography

The Mediterranean Sea is especially well known, especially those sites which were
readily accessible to sport divers. No presentations are made for this area because the
sites are so well known and some sites are dealt with in sections dealing with the his-
tory of underwater archaeology.

Basch. Lucien, 1969, Phoenician Oared Ships. Mariners Mirror 55 (3):227-234.

Bass, George F., 1967, Cape Gelidonya: A Bronze Age Shipwreck. Transactions o/the American Philosophical
Society 57, Part 8, Philadelphia, Pennsylvania.
Bass, George F., 1976, A Byzantine Trading Venture. Avenues to Antiquity Readings/rom Scientific American.
pp. 277-288. W. H. Freeman and Company, San Francisco, California. (originally pub!. in 1971).
Bass, George F., 1987, Oldest Known Shipwreck. National Geographic 172 (6):693-734.
Bass, George F. and Frederick H. van Doomick Jr., 1982, Yassi Ada: a Seventh-Century Byzantine Shipwreck.
vol. I, Texas A & M University Press, College Station, Texas.
Benoit, Femand, 1961, L 'Epave du Grand Congloue a Marseille. Supplement to Gallia. Centre National de la
Recherche Scientifique, Paris.
Casson, Lionel, 1971, Ships and Seamanship in the Ancient World. Princeton University Press, Princeton, New
Jersey.
Casson, Lionel, 1991, The Ancient Mariners . Princeton University Press, Princeton, New Jersey.
Frost, Honor, 1963, Under the Mediterranean. Prentice-Hall, New York.
Geanette, Mark, 1987, Resurrector of Wrecks. Oceans 20:36--41.
Greene, Jeremy, 1986, The Archaeology o/the Roman Economy. University of California Press, Berkeley.
Groenewegen-Frankfurt, H. A. and Bernard Ashmole, 1972, Art o/the Ancient World. Prentice Hall and Harry
Abrams, Englewood Cliffs, New Jersey.
Hamblin, Dora Jane, 1988, The Trade and Treasure of Phoenicia Come Alive. Smithsonian 19:62-72.
Harrison, Richard J., 1988, Spain at the Dawn 0/ History: Iberians, Phoenicians. and Greeks, Thames and
Hudson, New York.
Hocker, Frederick, 1995, Lead Hull Sheathing in Antiquity. Tropis 11/ Third International Symposium on Ship
Construction in Antiquity. edited by Harry Tzalas, pp. 197-206. Greek Ministry of Culture, Athens.

115
116 The Mediterranean: Bibliography

Jenkins, Nancy, 1980, The Boat Beneath the Pyramid. Holt, Rinehart and Winston, New York.
Johnston, Paul, 1980, The Sea-Craft of Prehistory. Harvard University Press, Cambridge, Massachusetts.
Johnston, Thomas C., 1965, 'Did the Phoenicians Discover America? St. Thomas Press, Houston, Texas.
Landstrom, Bjorn, 1961, The Ship. Translated by Michael Phillips, Allen & Unwin, Den Hague, The Netherlands.
Landstrom, Bjorn, 1970, Ships of the Pharaohs: 4000 Years of Egyptian Shipbuilding. Doubleday, Garden
City, New York.
Linder, Elisha, 1992, Ma'agan Micha'el Shipwreck Excavating an Ancient Merchantman. Biblical Archaeology
Review 18 (6):24-35.
Lipke, Paul, 1984, The Royal Ship ofCheops. British Archaeological Report International Series 225, National
Maritime Museum, Greenwich, Archaeological Series Number 9, Oxford.
Lloyd, A. B., 1975, Were Necho's Triremes Phoenician? The Journal of Hellenic Studies XCv.
Markoe, Glenn, 1990, A Nation of Artisans. Archaeology 43 (2):3 I-55.
Morrison, J.S. and J.F. Coates., 1986, The Athenian Trireme. Cambridge University Press, Cambridge.
Marx, Robert and Jennifer Marx, 1979, The Phoenician Candidate Transoceanic Explorations of the First
Maritime Traders. Oceans 12(4):16-20.
Parker, A. J., 1990, Classical Antiquity: The Maritime Dimension. Antiquity 64:335-46.
Parker, A. J., 1992, Ancient Shipwrecks of the Mediterranean and the Roman Provinces. British Archaeological
Reports, International Series 580, Oxford, England.
Stein, John, 1991, An Interview with George Bass. Omni 14:70-116.
Taylor, Joan du Plat, ed., 1966, Marine Archaeology. Crowell, New York.
 Part I1.B

Northern Europe:
".~i).:
~

Bibliography
-r/: ..
':.i

No presentations are made for this area because the topic is well known and repre-
sented by large number of publications. Some information is presented in other parts of
the text.

Adams, Jonathan, 1990, The Oskarshamn Cog, Part II: Excavation, Underwater Recording and Salvage. Inter-
national Journal of Nautical Archaeology 19:207-220.
Brogger, A. W. and Haakon Shetelig., 1971, The Viking Ships: Their Ancestry and Evolution, Dreyers Forlag,
Oslo, Norway.
Cederlund, Carl Olaf (ed.), 1985, Postmedieval Boat and Ship Archaeology. Swedish National Maritime
Museum Report Number 20, Stockholm, Sweden., 1990, The Oskarshamn Cog, Part I: Development of
Investigations and Current Research. International Journal of Nautical Archaeology 19: 193-206.
Fenwick, Valerie, 1978, The Graveney Boat, British Archaeological Reports, International Series, Oxford,
England.
Greenhill, Basil, 1976, Archaeology of the Boat. A. & C. Black, London.
Kiedel , Klaus-Peter, and Uwe Schnall (eds.), 1985, The Hanse Cog of 1380. Forderverein Deutsches Schif-
fahrtsmuseum e. V., Bremerhaven, Germany.
McGrail, Sean and Eric McKee, 1985, Sewn Plank Boats, National Maritime Museum, Greenwich Archae-
ological Series no. 10, British Archaeological Reports, International Series, Oxford, England.
McGrail, Sean (editor), 1987, Ancient Boats in N. W Europe, Longman, New York.
Rule, Margaret, 1982, The Mary Rose, Conway Maritime Press, London.
Thorvildsen, Karl, 1967, The Viking Ships of Ladby, National Museum, Copenhagen, Denmark.
Tuck, James A. , Excavations at Red Bay, Labrador, 1977-1984. Proceedings of the Sixteenth Conference on
Underwater Archaeology. edited by Paul F. Johnson, pp. 102-104. Special publication series no. 4,
Society for Historical Archaeology, Michigan.
Unger, R. w., \980, The Ship in the Medieval Economy, 600-1600, McGill-Queen's University Press, Montreal,
Quebec, Canada.

117
 Part II.C

The Caribbean:
".~i).:
~

Bibliography
-r/: ..
':.i

The Caribbean has been subjected to a great deal of reporting on underwater sites for a
variety of reasons. One of the driving elements has been the search for vessels used in
the earliest exploration of the area by Europeans. Another has been the national interest
of small states seeking to learn more about their history. The presence of artifacts, often
of saleable quality has led salvors into the area as well. The Caribbean area is presented
in bibliographic form because it is well known and reporting is fairly good.

Edwards, Clinton., 1992, Design and Construction of Fifteenth-century Iberian Ships: a review. Mariner sMirror.
78 (4 ):419-432.
Geddes, Donald G. III., 1992, Archival Research: the Search for the Columbus Caravels at St. Ann's Bay,
Jamaica. Underwater Archaeology Proceedings from the Society for Historical Archaeology Confer-
ence. edited by Donald Keith and Toni Carrell, pp. 148-152. Society for Historical Archaeology , Tuc-
son, Arizona.
Keith, D. H. and J. J. Simmons, 1985, Analysis of Hull Remains" Ballast and Artifact Distribution of a 16th-
Century Shipwreck, Molassas Reef, British West Indies. Journal of Field Archaeology 12:411-424.
McDonald, David (trans.), 1979, Documentary Sources for the Wreck o{the New Spain Fleet of 1554. Texas
Antiquities Committee, Austin, Texas.
Smith, R. C. and D. H. Keith., 1986, The Archaeology of Ships of Discovery, Archaeology 39 (2):30-35.
Weddle, R. S., 1985, Spanish Sea: the GulfofMexico in North American Discovery. 1500-1685, College Station,
Texas.

119
 Part II.D

The Pacific Rim

".~i).:
~

-r/: ..
':.i

Similar to many fields of academic pursuit, history reflects the perspectives of those
involved in its study. For nautical archaeologists and maritime historians, this means that
(for better or worse) most of the focus has been on the expansion of western Europeans,
from the ancient roots of Mediterranean cultures to the initial voyages of discovery
across the Atlantic. This maritime expansion, as exemplified in Parry's The Discovery of
the Sea (1981), certainly deserves the attention it has received. From a global perspec-
tive, however, large gaps exist in the historical and archaeological fabric.
Regions of the Far East, in particular coastal China and Southeast Asia, have expe-
rienced some of the most phenomenal phases in maritime history. The Chinese fleets of
the early Ming dynasty (15th century) roamed from the western Pacific to the east Afri-
can coast, uniting tributary kingdoms and pacifying crucial sea routes. The size of some
of these wooden sailing junks (over 400 ft) remains unconfirmed; except for a rudder
buried at the Imperial shipyards in Nanking, no hull remains have yet been found. The
endemic sea trade of Southeast Asia during the same period flourished in many styles,
alongside Chinese and Arabian trading entrepots. Trading ports along the Straits of
Malacca, Java, and Borneo saw the constant flow of worldwide commodities for hun-
dreds of years. Chinese, Burmese, Javanese, Thai, and Ryukyuan ships all plied the busy
waters. Further back in time, the area represents the birthplace of prehistoric sailing tra-
ditions, from which the early navigators spread throughout the South Pacific. Compared
to the work accomplished elsewhere, non-European maritime study of the Pacific ap-
pears to the western student to be a black hole. The work, however, has begun.
Nautical archaeology programs already exist in the Far East. In the People's
Republic of China, Xiamen University's Department of Anthropology, the Dalian
Maritime College, and the Quanzhou Museum of Overseas Communications History
represent some of the major research institutions. Thailand, Malaysia, the Philippines,
and Korea all maintain specialized units in underwater archaeology, though sometimes

121
122 11.0. The Pacific Rim

under the guise of various other names. Unfortunately, many obstacles exist to deter the
transmission of these findings to the West. Conditions limit the amount of archaeology
accomplished. The cultural priorities, scientific approach, and even definition of his-
tory within each country often differ. Finally, the language barrier imposes a constant
challenge to the western student new to these regions. The challenges are large, but the
rewards to research concerning the far side of the Pacific are rich in centuries of inten-
sive maritime history and related material remains. Until these "other" earlier seagoing
cultures are understood, there can be no real sense of a global maritime history.
Two brief articles provide only an introduction to the larger scope of maritime
archaeology in the Pacific. In Chapter 12, Taha provides a look at the beginnings and
accomplishments of underwater archaeology. Funding and training constrictions can be
taken as representative of larger scale realities in the region. In Chapter 13, an excellent
summary of the historical background to important past and present projects, Ronquillo
discusses the position of the Philippine's National Museum, and includes map of site
locations.
 Chapter 12

Current Status of
Underwater Archaeology
in Malaysia

ADI HAJI TAHA

INTRODUCTION

Underwater archaeology is a new subject for most Malaysians. There was little or no
interest in it until quite recently when a Dutch East Indiaman, the Risdam, was located
and looted by a Singapore-based company off the Mersing Coast in lohore 1984. The
discovery had considerable publicity and created interest among Malaysians.
Recently Malaysians were amazed by news of the recovery ofa hefty 7 billion Ma-
laysian ringgit Portuguese wreck, the Flor de la Mar, which is said to have been laden
with gifts for the king and queen of Portugal, and sank in 1512. The ship was wrecked
off the coast of Sumatra, thus putting the Malaysian authority in a difficult situation to
recover and excavate the wreck, unless it has the cooperation of the Indonesian authority.
Malaysia has a long seacoast for its land area, and has tremendous potential for
the development of maritime archaeology which will not only generate pride for the
Malaysians of their past, but also has potential for the development of tourist industries
by way of maritime museums and shipwreck exhibits. The Straits of Malacca and the
South China Sea have been the world's busiest shipping lanes for many centuries until
the present. Hundreds of wrecks of various types of sea- and ocean-going vessels rang-
ing from barges, sampans, perahus, and sailboats may be lying within the territorial
waters of Malaysia.
Source: Bulletin o/the Australian Institute/or Maritime Archaeology (1989), 13 (2):3--4. Reprinted by permis-
sion of the Australian Institute for Maritime Archaeology, Inc.

123
 124 AH. Taha

The abundant untouched treasures said to be lying buried in the waters of the Malay
archipelago, the center of the spice trade, and the graveyards of fabulous Portuguese ships,
Spanish galleons, Dutch merchantmen, and the English petty traders have nevertheless
attracted many treasure hunters with sophisticated equipment, daring local marine police
searching for vessels carrying gold and other cargo which can yield high profit.
It is clear from the foregoing that there is a wealth of marine archaeological mate-
rial waiting to be discovered in Malaysian waters. The authority is certain that most of
the sites have been pilfered and vandalized, and the government is very concerned
since this would mean loss of revenue to them.

PERSONNEL AND TRAINING

The Museums Department has been given the authority to deal with antiquities
found in West Malaysia whether on land or in the bed of a river, lake, or from the sea
as provided in the Antiquities Act of 1976. Unfortunately, there is no special unit in the
department to deal with the subject. The task is placed with the Archaeology Unit, and
for this purpose, an Assistant Curator has been trained under an SPAF A training pro-
gram and recently, with a UNESCO fellowship award, attended a special course in
Scotland. The Sabah State Museum has qualified divers, also being trained under the
SPAFA training program. The foregoing personnel, however, are still far from ade-
quate to undertake underwater archaeological operations. The Royal Malaysian Navy
divers have been engaged in the rescue operation of the Risdam. While they are capa-
ble of doing a good job, they still need experienced archaeologists to guide and direct
them in the systematic surveys and excavations of wrecks. We hope that with new
interest in underwater archaeology in Malaysia, an Underwater Archaeology unit will
be established within the department in the near future.

PAST RESEARCH

In 1980, an attempt was made to locate a Japanese ship wrecked off the Gender-
ing coast of Terengganu by the State Government and the Museums Department with
assistance from the Royal Malaysian Navy. The ship, believed to be laden with treas-
ures, was found at a depth of about 30 m below the surface. It is an iron ship and was
badly hit by a torpedo. Nothing was recovered from the wreck, but the divers were able
to produce the profile drawing of the ship. Because of a strong underwater current, the
operation was stopped after four dives.
In 1984, just before the start of the Risdam saga, the Museums Department, with
assistance from a private diving firm, conducted a preliminary survey 1 km from the
Malacca River estuary. Visibility was zero at the time of the operation, but through
sensing and touching, the divers claimed that every inch of the seabed contained arti-
facts. They were able to collect some interesting items such as Ballarmine stoneware
dated to around the late 17th century.
The most significant underwater archaeological work ever carried out in West
Malaysia was the rescue operation on the Risdam, a Dutch East Indiaman of the fluit
type wrecked in 1727 off the coast of Mersing in Johore. The wreck was first traced in
archival records in Leiden through the effort of H.C. Besancon, a member of the Dutch
East Indian Wreck Research Foundation. With Michael Hatcher, they surveyed the
Current Status of Underwater Archaeology in Malaysia 125

wreck in 1983, and urged the Dutch government to negotiate with the Malaysian
authorities to salvage the wreck. Apparently, a member of the team leaked the secret
and worked with another partner to loot the wreck. they were in the midst of looting
the cargo when their operation were exposed.
Following that, the Malaysian authorities, under the care of the Museums Depart-
ment and with the Royal Malaysian Navy's divers, rescued 120 tin ingots, 29 elephant
tusks, cannon balls, and sampan wood from a hole cut open by the looters.
The Malaysian government then requested assistance from the Australian and
Dutch governments which were represented by Mr. Jeremy Green and Professor J.R.
Bruijn, respectively. Since then there has been no follow-up to the salvage operation
though at one stage Michael Hatcher offered his services to excavate the wreck.
It has been decided that the site will be reserved for training purposes since the
wreck is situated close to the shore and part of the wreck is still intact, as had been sug-
gested by Mr. Jeremy Green.

CURRENT DEVELOPMENT

With new interest and development in underwater archaeology, many quarters of

the Malaysian public have called for a more serious approach in handling the subject.
The first of this group was the Department of History of the University of Malaya, who
immediately, after the news first broke of the Risdam, organized a I-day seminar on
"Ships and Sunken Treasures." It was then that the Portuguese ship Flor de fa Mar and
her treasures were first mentioned in public in Malaysia.
The Malaysian authorities are aware of the activities of underwater treasure hunt-
ing and have taken steps to curb such activities through the marine and naval forces.
The outcome of such a situation led to many interested parties applying for licenses to
operate survey and salvage work in Malaysian waters. Many came with generous
offers, willing to bear all cost of the operation, including sponsoring officers stationed
on the ship for the entire operation.
To process such applications, the government has set up a committee chaired by
the Director of the Contract and Purchase Division of the Ministry of Finance with
members comprising representatives of various relevant agencies such as the Museums
Department, the Marine Department, Marine Police, Royal Malaysian Navy, and the
State Goverment where the said wreck is found.
The committee prefers to offer licenses to operators who have identified a wreck to
work on. However, in the case of companies who have no definite wreck to work on, a
guideline has been drawn by the committee in which a 10-km grid is drawn perpendicu-
lar to the coast from the sea boundary. The operator is thus given the choice to select any
three grids to survey within a period of 6 months. If a wreck is located, the operator will
lay claim to salvage rights to the wreck to the Marine Department. A formal agreement
will then be signed. Before a permit is issued, the committee places significance on the
fact that the operation should be done scientifically, meaning a maritime archaeologist
must be a member of the salvage party and recording of the operation has to be in video,
photographs, and written form.
 126 A.H. Taha

LEGAL ASPECTS

At present, there is no specific act pertaining to shipwreck and sunken treasures.

The Antiquities Act of 1976 provides for the control and preservation of, and research
into, antiquities found in West Malaysia, while the states of Sabah and Sarawak have
their own enactments almost similar in nature to the West Malaysian one. The act is
largely based on land antiquities, but also covers any antiquities thought "to be part of
the soil or of the bed ofa river or lake or of the sea."
Although the phrase "in the sea" was thought to be sufficient then, with new
developments and advances in marine technology the present jurisprudence concerning
historical artifacts in marine wrecks is inadequate. The act is now under revision in the
Attorney General's chambers, and in the revision, the fundamental concepts of the "his-
toric shipwreck" will be given special attention, such as those being used by Australia
and the United Kingdom.
Malaysia is a federation and with the state governments showing great interest in
underwater antiquities to enrich their historical collection, it seems that the Australian
shipwreck laws (Commonwealth and State) are applicable and practical to Malaysia.
The Merchant Shipping Ordinance of 1952, which follows the British tradition,
has a section on wrecks where the harbor master is the principal receiver of wrecks.
Unfortunately, their main concern is more on wrecks that hinder or are hazardous to
present-day navigation. The only real test of ownership of antiquities found in Peninsu-
lar Malaysia was when a disposal inquiry was held in a Singapore district court over
the confiscated tin ingots, elephant tusks, and a porcelain bowl believed to have been
looted from the Risdam. Based on the Merchant Shipping Ordinance of 1952 and the
Antiquities Act of 1976, the judge ruled that the antiquities rightly belong to the Gov-
ernment of Malaysia.

CONCLUSION

The Malaysian government recognizes two types of wrecks lying in her seabed.
The first are those that have no direct relationship to the Malaysian cultural signifi-
cance which include ships passing through Malaysian waters from east to west and vice
versa. The second are local Malay ships or Chinese junks plying within the Malay
archipelago which hold direct significance to Malaysian culture and history. Underwa-
ter salvage and scientific excavation requires large financial expenditure which devel-
oping countries like Malaysia cannot afford to undertake. In the instance of the first
type of wreck, the government is willing to contract out with private firms who qualify
to undertake scientific survey and excavations. The revenue from such operations will
be used then for undertaking research of the second type of wreck, which in financial
terms is not attractive to them.
To safeguard the historical wrecks from being plundered by unauthorized people,
a new or a revised act will include a specific article or schedule for historical ship-
wrecks.
Efforts toward establishing a regional cooperation in underwater archaeological
research and training center should be established in Southeast Asia for the benefit of
safeguarding our common heritage.
 Chapter 13

Philippine Underwater
Archaeology
Present Research Projects and
New Developments

WILFREDO P. RONQUILLO

INTRODUCTION

Underwater archaeological activities in the Philippines have been documented from the
earliest recorded work in 1967 (Lopez, 1967) to the more recent undertakings in the
1980s (Conese, 1981, 1983; Alba, 1984; Philippine Delegation, 1984; Cuevas, 1986,
1988; Nicolas, 1986). The discovery and excavation of lashed-lug planked boats in
northeastern Mindanao and the results of the archaeological research at sites having a
maritime orientation, likewise, have been published (Peralta, 1976, 1980; Scott, 1981;
Ronquillo, 1987).
In more recent years underwater archaeological research in the archipelago has
been given its share of importance and attention for a number of reasons, foremost of
which are:
1. The potential of underwater archaeology in the country as a result of the pre-
historic, protohistoric, and historic events that occurred in the archipelago
which are directly connected with maritime activities.
Source: Bulletin of the A ustralian Institute for Maritime Archaeology (1990), 14 (1 ):21-24. Reprinted by per-
mission of the Australian Institute for Maritime Archaeology, Inc.

127
128 W.P. Ronquillo

~
. , g

."
~
c;)
N
D
, t:> ~

~NERCEDES
~ CAM. NORTE
LUBANG IS.
acc.MINDOR~aniia
eo;)·
PUERTO GALERA
OR. MINDORO
TRES "EYES
\).,....,
eN' ~~
e----___ STD. DOMINGO
MARINDuaUE • ; ~ ALlAY

aUINILUSAN IS.
PALA"'AN
~ 0''0 ~'2

ROYAL ~

.0 · O~ALANGHAI
Cebu

SITES
BUTUAN CITY
Q D

••

Figure I. Map of the Philippines showing location of sites.

Philippine Underwater Archaeology 129

2. The discovery, excavation, conservation, and exhibition oflashed-lug planked

boats from northeastern Mindanao which serve as direct evidence of the pre-
historic boat-building and -using traditions in the archipelago.
3. The entry of foreign private entities interested in joint underwater archae-
ological projects with the National Museum.
This paper discusses the most recent research projects in underwater archaeology
in the country and presents the new developments in the area of legislation for the pres-
ervation and protection of the country's maritime cultural heritage. A summary of ear-
lier underwater archaeological activities in the Philippines is also presented.

BRIEF HISTORICAL BACKGROUND

A chronological presentation of the underwater archaeological activities under-

taken in the country is summarized as follows.

1967: Ajoint project activity between the National Museum and the Times-Mir-
ror-Taliba (a defunct newspaper outfit) at Nabasagan Point, Sto. Domingo, Albay, about
500 Km south of Manila. The shipwreck was believed to be a Spanish galleon found
40--65 m below the surface. Recovered were two huge anchors each weighing about 3
tons and indicating a forging date of 1649, "a bronze butterfly hinge, pottery sherds,
copper plates, nails, chain links, capstans, plankings, and the centre bolt of an auxiliary
mast" (Lopez, 1967).

1982: A joint venture between the National Museum and a private diving outfit
- Marine Archaeology Unlimited, Incorporated (MAUl) - on the southeast coast of
Marinduque Island about 150 km south of Manila. Located 39.65 m below the surface
the sunken vessel is believed to be a merchant boat carrying a cargo of "stoneware jars
and covers, porcelain plates, dishes, saucers, and bowls." The porcelain dates to the
Ming dynasty and the stoneware jars recovered are brown-glazed with relief dragon
designs (Conese, 1981, 1983).

1983: A joint venture between the National Museum and Mr. Brian Homan, an
Australian sport diver, at Puerto Galera, Mindoro, an island immediately west of
Marinduque Island. Under 15.25-24.4 m of water, the site was partially looted prior to
the work of the Museum. The finds include "porcelain plates, dishes, and bowls, glazed
jarlets, a blue-and-white kendi, a porcelain wine jar, two covers of porcelain powder
boxes, several celadon incense burners, and a brown-glazed stoneware jar believed to
be of Siamese origin" (Alba, 1984).

1985: A joint venture between the National Museum and a private entity called
Scientific Survey and Location Limited which entailed the survey and exploration for
sunken vessels in the waters off northern Palawan, The exploration venture is for 5
years. Aside from one site that looked promising, no other sites were found.

1985: A joint venture between the National Museum and World Wide First,
Incorporated, a French private entity at the Royal Captain Shoal, a coral reef west of
130 W.P. Ronquillo

Palawan Island. Initially in search of the 18th-century East India company merchant-
man the Royal Captain which sank in 1773, the team encountered archaeological mate-
rials that are 200 years older. The site was called Wreck 2 of the Royal Captain Shoal.
The archaeological materials recovered from this site include porcelain plates, saucers,
bowls, cups, boxes and box covers, blue-and-white and terra-cotta pear-shaped bottles,
jarlets, jars, over 200 glass beads, 33 identical gongs, bronze, iron, and copper objects.
The tradewares recovered from the wreck indicate that the nature of the paste as well as
the execution of the decorative patterns "very clearly point to the Wan Li Period-
1620)" (Goddio and Jay, 1988).

1985: A joint venture between the National Museum and World Wide First to
excavate and study the remnants of an East India Company vessel, the Griffin, north-
west of Basilan Island in the southern Philippines (Cuevas, 1986; Nicolas, 1986). The
archaeological materials recovered from this site include ceramics, mostl)G porcelain,
which make up the bulk of the cargo recovered; octagonal and oblong dishes, a group
of tea or coffee cup sets, an underglazed blue and celadon-glaze semidetached model of
a couple without heads, clay pipes, ivory in the form of fan sticks, glass in the form of
different shaped bottles, and a few metal objects which include "iron ingots used as
ballast, iron tool such as adzes, cannon balls, lead sheets used to line the wooden tea
crates, lead musket balls, tea pots, a Chinese coin of copper alloy, shoes and belt buck-
les of copper alloy and gilt bronze, and a few other objects used for daily life on board
ship" (Goddio and Jay, 1988).

1986: A joint venture between the National Museum and World Wide First, to
excavate the wreck of the galleon San Jose off the waters of Lubang Islands, Mindoro.
Undertaking archaeological excavation work only during the northeast monsoon period
(January-May) has resulted in the discovery ofa portion of the ship's planks, numerous
shards of blue-and-white chocolate cups, and fragments of bronze, iron, and copper
materials. Archaeological activities at this site will continue in January 1990 (Cuevas,
1988).

PRESENT RESEARCH PROJECTS

Aside from the archaeological work continuing at the San Jose wreck site off
Lubang Islands from January to May 1989, three other underwater archaeological
research activities were undertaken.

The Reexcavation of the Marinduque Island Wreck Site

The National Museum and Scientific Survey Limited reexcavated the Marin-
duque Island wreck site for a period of 2 months. This site was worked on by the
National Museum and MAUl in 1982. A statistical analysis has recently been done on
the ceramics recovered from the excavation of this site (Henson, 1989). The recent
archaeological activities at the site were undertaken to verify if more archaeological
data may be generated from the site as a result of the early termination of the project by
MAUL The recent joint venture in Marinduque is also envisioned to result in a publish-
ed report on the project which will include results of terrestrial archaeology in the
Philippine Underwater Archaeology 131

island of Marinduque and the role of the island in the early maritime trade network in
the archipelago (Alba, 1989).

Articulation of the Butuan Boat

One of the lashed-lug planked boats recovered at Libertad, Butuan City, Agusan
del Norte, northeastern Mindanao is at present being articulated for public exhibition at
the National Museum in Manila. To be placed in the Philippine Maritime Heritage Gal-
lery, the renovation of which was funded by the Duty-Free Philippines, Incorporated,
of the Department of Tourism, research on this 13th-century indigenous Southeast
Asian boat, one of eight discovered, indicated that it has a well-balanced hull which is
rounded amidships and gracefully tapering sharply toward the prow and stern. This
type of boat construction does not entail the use of outriggers. Technologically and
functionally, the Butuan boat is ideal for coastal and up-river trading activities (Santi-
ago, 1989, personal communication).
The Butuan boat possibly evolved from an Indo-Pacific boat-building technology
which employs end blocks to fasten the ends of the planks and the keel (Horridge,
1982; Green, 1986).

A Reported Shipwreck at Camarines Norte, Southern Luzon

In July 1989 a shipwreck was reported to the Archaeology Division by Dr. Fran-
cisco Pano!. The wreck is located at one of the tributaries near his 45-hectare prawn
pond in Mercedes, Camarines Norte. The site is about 375 km southwest of Manila.
Visited in August by personnel from the Archaeology Division, it was noted that the

6'"=.i.....1....

Figure 2. Brass nail and tack from the shipwreck site at Mercedes, Camarines Norte.
132 W.P. Ronquillo

wreck is situated in a silty-mud area. The preliminary exploration activities indicated

that a portion of the wreck is visible during low tide, specifically, parts ofthe ribs. The
ribs exposed are noted to be 10 cm thick and 45-50 cm apart. It also exhibited a tenon
and mortise type of joining. Most of the parts of the ribs are under the silt and four ribs
were exposed during low tide. A thin strip of brass and a number of 2.0-cm-Iong brass
tacks were noted to be still intact at the bottom part of the wreck. It was also noted that
a number of these tacks were retrieved from the site earlier and are now in the posses-
sion of a resident of the area. A large brass nail, square-shaped and with barbs on two
comers of the shaft, was donated by Dr. Panol to the Museum. This nail is II cm long.
From the materials found it is likely that the wreck is an 18th-century European ship. A
full-scale excavation of the site is scheduled for next year and is envisioned to result in
detailed accounts of the wreck and the associated materials found at the wreck site
(Henson et aI., 1989).

NEW DEVELOPMENT

The National Museum is the lead agency in the Philippines that is tasked to
undertake archaeological research work, both terrestrial and underwater. It is also
tasked to protect the cultural heritage of the country including its maritime cultural
heritage.
Although underwater archaeological research in the archipelago is slowly build-
ing up its capabilities in this specific field in terms of improvements of personnel
through various underwater archaeology training programs, the lack of basic underwa-
ter archaeology equipment remains the most pressing problem of the Archaeology
Division. Joint ventures with private entities characterized the National Museum's
entry into underwater archaeological activities in 1967 and have continued to be the
normal practice up to the present. In some cases this arrangement has worked well
while in others problems arose due mainly to the absence of specific policies by the
National Museum on the matter. Each joint venture had its specific contract different
from any other.
Since 1986, a "Policy guidelines for underwater archaeology applications to
explore and to excavate in Philippine waters" has been drafted. It has, in the succeeding
years, undergone several improvements and revisions. The latest revised version is
called the "Rules and regulations for underwater archaeology exploration and excava-
tion in Philippine waters." This attempt to formalize the policy ofthe National Museum
regarding underwater archaeological activities may soon be paying off.
This work became the main source of materials for a bill, No. S-695, which has
been introduced in the Philippine Senate by the Chairman of the Committee on Educa-
tion, Senator Edgardo Angara. Known as the Protection of Underwater Cultural Heri-
tage Act of 1988, the bill provides a permit system for underwater exploration and
excavation aside from establishing definite archaeological and scientific procedures to
be followed. The bill addresses the urgent problem of the massive loss of the country's
cultural heritage as a result of indiscriminate plunder, exploitation, and vandalism. The
bill categorically states that the "ownership of all objects of underwater cultural heri-
tage found in territorial waters, regardless of origin, is vested directly in the state." The
National Museum is tasked by the bill to be the lead agency to implement the provi-
sions of the proposed Act.
Philippine Underwater Archaeology 133

The enactment of this bill into law, with its allocated financial appropriation, will
ensure that underwater archaeological activities may be undertaken in Philippine terri-
torial waters without unduly compromising the preservation and protection of the coun-
try's maritime cultural resources.

REFERENCES

Alba, L. A., 1984, The Genesis of Underwater Archaeology in the Philippines. Ms. National Museum, Manila.
Alba, L.A., 1989, Status Report on the 1989 NM-SSL Marinduque Underwater Archaeological Project. Ms.
National Museum, Manila.
Conese, E.T., 1981, Report on Marine Archaeological Exploration Conducted in Gaspar Island. Marinduque
Province. Ms. National Museum, Manila.
Conese, E.T., 1983, Second Preliminary Report on Marinduque Archaeology Project. Ms. National Museum,
Manila.
Cuevas, M.A., 1986, Survey of the Griffin Wreck Site. Ms. National Museum, Manila.
Cuevas, M.A., 1988, Report on the Underwater Archaeological Activities at Tagbac Cove. Lubang Island. Ms.
National Museum, Manila.
Gatbonton, E.B., 1989, Report on the State of Marine Archaeology and Prevailing Conditions. In SEAMED
Member Countries. Ms. SPAFA Regional Centre, Bangkok.
Goddion, F., and Jay, E., 1988, 18th Century Relics of the Griffin Shipwreck. Makati, Metro Manila.
Green, J., 1986, Eastern Shipbuilding Traditions: A Review of the Evidence. Bulletin Australian Institutefor
Maritime Archaeology. Proceedings from the Australian Institute for Maritime Archaeology Seminar on
Asiatic Shipbuilding techniques, held in Launceston, 24--26 September 1985, 10.20: 1-6.
Henson, F.G., 1989, Spatial Analysis of Ceramics from a Philippine Underwater Site. Paper presented at the
National Conference ofindonesian Archaeologists, Jojakarta, Indonesia, July, 1989.
Henson, F.G., Alonto, A., and Action, M., Jr., 1989, Report on the Mercedes Wreck Site Investigation at Cama-
rines Norte. Philippines. Ms. National Museum, Manila.
Horridge, G.A., 1982, The Lashed-Lug Boat of the Eastern Archipelagoes. the Alcina MS and the Lomblen
Whaling Boats. Maritime Monographs and Reports No. 54-1982. National Maritime Museum, Green-
wich, London.
Lopez, P.T., Jr., 1967, Marine Archaeology in the Philippines. Ms. National Museum, Manila.
Nicolas, N.C., 1986, Preliminary Report: Griffin Underwater Archaeological Excavation Project. Ms. National
Museum, Manila.
Peralta, J.T., 1976, Calanghai: Ancient Asian Boat Unearthed. Ms. National Museum, Manila.
Peralta, J.T., 1980, Ancient Mariners of the Philippines. Archaeology 33.5:41-48.
Philippine Delegation, 1984, Prehistoric Maritime Trade in the Philippines. In SPAFA final report on the consult-
ative workshop on research on maritime shipping and trade networks in Southeast Asia (I-W7). Cisarua,
West Java, Indonesia, November 20-27,1984.
Ronquillo, W.P., 1987, The Butuan Archaeological Finds: Profound Implications for Philippines and Southeast
Asian Prehistory. Man and Culture in Oceania 3(Speciallssue): 71-78.
Scott, W.H., 1981, Boat Building and Seamanship in Classic Philippine Society. Anthropological PaperNo. 9.
National Museum, Manila.
 Additional Bibliography
for Part fl.O

Atkinson-Millar, Karen, 1990, The First Stage of a Maritime Archaeological Training Programme in the Peo-
ple's Republic of China: a Brief Summary of the Aims, Procedures and Results. International Journal
of Nautical Archaeology 19 (3):255-258.
Blake, Warren and Michael Flecker, 1994, A Preliminary Survey of a South-East Asian Wreck, Phu Quoc Island,
Vietnam. International Journal ofNautical Archaeology 23 (2):73-91.
Carrell, Toni L (ed.), 1991, Micronesia: Submerged Cultural Resources Assessment, National Park Service,
Southwest Cultural Resources Center, Sante Fe, New Mexico.
Clark, Paul, Eduardo Conese, Norman Nicholas, and Jeremy Green, 1989, Philippines Archaeological Sites
Survey, February 1988. International Journal of Nautical Archaeology 18 (3):255-262.
Delgado, James P. et aI., 1991, The Archaeology of the Atomic Bomb: a Submerged Cultural Resources Assess-
ment of the Sunken Fleet of Operation Crossroads at Bikini and Kwajalein Atoll Lagoons, National
Park Service, Southwest Cultural Resources Center, Sante Fe, New Mexico.
Emmerson, Donald K., 1980, The Case for a Maritime Perspective on Southeast Asia. Journal of Southeast
Asian Studies II (1):139-145.
Flecker, Michael, 1992, Excavation of an Oriental Vessel of ca 1690 off Con Dao, Vietnam. International
Journal of Nautical Archaeology, 21 (3):221-244.
Goddio, Franck, 1994, San Diego: An Account of Adventure, Deceit, and Intrigue. National Geographic
186:34-57.
Green, Jeremy, 1983, The Song Dynasty shipwreck at Quanzhou, Fujian Province, People's Republic of China.
International Journal of Nautical Archaeology 12 (3):253-26 \.
Green, Jeremy, 1983a, The Shinan Excavation, Korea: an interim report on the hUll structure. International
Journal of Nautical Archaeology 12.
Green, Jeremy, 1986, Eastern Shipbuilding Traditions: a review of the evidence. Bulletin of the Australian
Institute for Maritime Archaeology 10 (2): 1-6.
Green, Jeremy and Zae Guen Kim, 1989, The Shinan and Wando Sites, Korea: further Information. International
Journal of Nautical Archaeology 18 (1):33-4\.

135
136 11.0. Bibliography

Green, Jeremy and Rosemary Harper and Vidya Intakosi, 1987, The Maritime Archaeology of Shipwrecks and
Ceramics in Southeast Asia; [and} the Ko Si Chang 3 Shipwreck Excavation. Australian Institute for
Maritime Archaeology, special publication no. 4. Victoria NSW, Australia.
Guy, John S., 1987, Ceramic Excavation Sites in Southeast Asia: A Preliminary Gazetteer. Research Centre for
Southeast Asian Ceramics Papers Number 3, Art Gallery of South Australia, Adelaide.
Guy, John S., 1992, China and the Maritime Silk Route Conference - Quanzhou, South China. International
Journal ofNautical Archaeology 21 (1):70--73.
Henderson, Graeme, 1986, Maritime Archaeology in Australia, University of West em Australia Press, Nedlands,
W.A.
Horridge, Adrian, 1981, The Prahu: Traditional Sailing Boat of Indonesia. Oxford University Press, New York.
Inoue, Takahito, 1991, A Nautical Archaeological Study of Kublai Khan's Fleets. Master's thesis, Texas A & M
University, College Station, Texas.
Keith, Donald H. and Christian 1. Buys, 1981, New Light on Medieval Chinese Seagoing Ship Construction.
International Journal of Nautical Archaeology 10 (2): 119--\32.
Li Shitao, 1984, The Boat-coffins of the Wuyi Mountains, in Ancient Discoveries in Chinese Archaeology,
Foreign Language Press, Beijing.
Lyon, Eugene, 1994, Track of the Manila Galleons. National Geographic 178:4-37.
Manguin, Pierre-Yves, 1980, The Southeast Asian Ship: An Historical Approach. Journal of Southeast Asian
Studies. II (2):266-276.
Mathers, William, 1990, Neustra Senora de la Concepcion. National Geographic 178:38-53.
Mathers, William M., Henry S. Parker III and Kathleen A. Copus (eds). 1990, Archaeological Report: the Recov-
ery of the Manila Galleon Nuestra Senora de la Concepcion, Pacific Sea Resources, Sutton, Vermont.
Merwin, Douglas (translator), 1977, Selections from Wen Wu on the excavation of Sung Dynasty vessel in
Quanzhou. Chinese Sociology and Anthropology 9 (3):3-106.
Parry, John H., 1974, The Discovery of the Sea, Dial Press, New York.
Scott, William Henry, 1982, Boat-building and Seamanship in Classic Philippine Society. Philippine Studies
30 (3):335--375.
Strachan, Shirley M., 1986, Indian and Southeast Asian Shipbuilding for the European Market: A Survey
1790--1815. Bulletin of the Australian Institute for Maritime Archaeology 12 (2): 13-16.
Strachan, Shirley M., 1986a, A Research Design for European Influenced Shipbuilding in India and Southeast
Asia: The Sydney Cove (1797). Bulletin of the Australian Institute for Maritime Archaeology 12
(2):37--43.
Throckmorton, Peter, 1987, The Sea Remembers. Smithmark, New York. (See pages 152-167).
 Part II.E

'.~~.' The Arabic World:

~

o i' .
Bibliography
q;,O_' .': .
.j'~l

No presentations are made regarding the Arabic World as an areal study. The field work
at present is preliminary and often unavailable in English as can be seen in the thinness
of the bibliographic section.

Bound, Mensun, 1990, Conference on Maritime Archaeology of the Indian Ocean Countries. International
Journal of Nautical Archaeology 19 (2): 159-161.
Chittick, Neville, 1980, Stone anchor-shanks in the Western Indian Ocean. International Journal of Nautical
Archaeology 9 (I ):73--76.
Choudhuri, K. N., 1985, Trade and Civilization in the Indian Ocean. Cambridge University Press, London.
Christides, Vassilios, 1989, Some remarks on the Mediterranean and Red Sea ships in Ancient and Medieval
times. Tropis I First International Symposium on Ship Construction in Antiquity, edited by Harry
Tzalas, Greek Ministry of Culture, Piraeus.
Christi des, Vassilios, 1995, Byzantine Dromon and Arab Shini. Tropis III Third International Symposium on
Ship Constructio~ in Antiquity. edited by Harry Tzalas, pp. 111-122. Greek Ministry of Culture, Athens
Hornell, J., 1942, A Tentative Classification of Arab Sea-craft. The Mariner s Mirror January.
Hourani, George Fadlo, 1951, Arab Seafaring in the Indian Ocean in Ancient and Medieval Times. Princeton
University Press, Princeton, New Jersey.
Howarth, David, 1977, Dhows. Quartet, New York.
Kuppuram, G. and K. Kumudamani (eds.), 1996, Marine Archaeology: The Global Perspectives. Sundeep
Prakashan, Delhi.
Lewis, A., 1978, Maritime Skills in the Indian Ocean. The Sea and Medieval Civilizations, pp. 239-64. Variorum
Publications, London.
Li Hui-Iin, 1961, Mu-Ian-p'i: a Case for Pre-Columbian Transatlantic Travel by Arab Ships. Harvard Journal
ofAsiatic Studies 23: 114--126.
Martin, E. B. and C. P. Martin, 1978, Cargoes of the East: The Ports. Trade and Culture of the Arabian Seas
and Western Indian Ocean. Elm Tree Books, London.
Mathew, K. S. (ed.), 1995, Mariners. Merchants and Oceans: Studies in Maritime History. Manohar Publishers,
New Delhi.

137
138 The Arabic World: Bibliography

Moreland, W. H., 1039, Ships of the Arabian Sea about 1500 AD. Journal of the Royal Asiatic Society 63.
Paula, Dona, 1987, A Note on Geophysical Explorations for Marine Archaeology off Tamilnadu Coast India.
International Journal of Nautical Archaeology 16 (2): 159-164.
Prados, Edward F., 1993, The Development of Arabian Shipping in Classical Antiquity, M.A. Thesis, East
Carolina University, Greenville, North Carolina.
Raban, Avner, 1990, Medieval Anchors from the Red Sea. International Journal of Nautical Archaeology 19
(4):299-306.
Rajamanickam, G. V., V. S. Arul Raj, 1994, Maritime History ofSouth India: Indigenous Traditions ofNavigation
in the Indian Ocean. Tamil University, Thanjavur, India.
Rao, S. R., 1987, Submerged City and Shipwreck at Dwarka. International Journal of Nautical Archaeology
16 (3):252-254.
Rao, S. R., 1994, Indian Conference on Marine Archaeology of Indian Ocean Countries. Society for Marine
Archaeology, National Institute of Oceanography, Dona Pawa, Goa.
Rao, S. R., I 994a, New Frontiers of Archaeology. Popular Prakashan, Bombay.
Ray, Himanshu Prabha, Jean-Francois Salles (eds).
Rao, S. R., 1996, Traditions and Archaeology: Early Maritime Contacts in the Indian Ocean. Manohar Pub-
lishers, New Delhi.
Severin, Tim, 1982, The Sinbad Voyage. G. P. Putnam's Sons, New York.
Tibbets, G. R., 1971, Arab Navigation in the Indian Ocean before the Coming of the Portuguese, London.
 Part II.F

Africa: Bibliography

Subsaharan Africa is very little known in terms of maritime archaeology. Virtually all
readily available published material on subsaharan African sites falls into three catego-
ries, European vessels, Arabic/Mediterranean vessels or discussions of African-built
vessels as models for other areas. This is the result of the same forces which worked to
obscure oriental maritime sites and a similar effort must be made to recover older, Afri-
can-built vessels from south of the Sahara.

Burman, K., 1968, Strange Shipwrecks of the Southern Seas. Struik, Cape Town.
Harris, Lynn 8., 1993, The Seaport, Table Bay - An Archaeological and Historical Perspective. Program in
Maritime History and Nautical Archaeology, East Carolina University, Greenville, North Carolina.
Marsden, P., 1976, The Meresteyn, wrecked in 1702, near Cape Town, South Africa. International Journal of
Nautical Archaeology 5 (3): 201-219.
Martin, E. B. and C. P. Martin, 1978, Cargoes of the East: The Ports, Trade and Culture of the Arabian Seas
and Western Indian Ocean. Elm Tree Books, London.
Piercy, R. C.M., 1978, Mombasa Wreck Excavation. International Journal of Nautical Archaeology 7 (4):
301-319.
Rawe, J. and A. Crabtree, 1978, Shipwrecks of the Southern Cape. Atlantic Underwater Club, Cape Town.

139
 Part III

Research Design
".~i).:
~

-r/: ..
':.i

Archaeology is the scientific study of past cultures through the material remains those
cultures left behind. The archaeologist asks questions of a site even before excavating.
Whether these questions are inductive or deductive, they guide the fieldwork. By
including articles that spell out the scientific method as used in other sciences, archae-
ological students can better formulate their questions of the data.
The modern paradigm for guiding a scientific research effort involves the testing
of assumptions (hypotheses) about what is expected. The predictive hypotheses are
tested against the actuality of what is found during field work. Ultimately, the re-
searcher must discuss what it would take to disprove the final interpretation which is
arrived at. Asking this question completes the effort by posing a set of hypotheses
which can not be refuted by the information available.
This methodology is put forth in two seminal articles. The first, by Chamberlin,
was originally written in the late 19th century and has been since reprinted two more
times in 1932 and 1965. The second, by Platt, provides additional explanation about
how testing of multiple working hypotheses occurs and what the ultimate questions
should be. The other two articles amplify research frameworks. Finally, a paper by two
underwater archaeologists shows that, while archaeological research has paid lip serv-
ice to the notion of theory, few presentations about theory have been made.

141
 Part III.A

Hypothesis Building

One of the best articles for laying the foundation of a healthy skepticism needed in
research planning, Chamberlin's contribution in Chapter 14 needs little else as intro-
duction. It is well written and just as pertinent today as 100 years ago when originally
written. In a companion article to Chamberlin's, Platt (Chapter 15) eloquently ties in-
itial research phases to larger issues of scientific methodology. As with Chamberlin, the
work speaks for itself. Platt stresses the necessity of keeping the study simple and to
the point.
Smith (Chapter 16) makes several points about how far can one go in the inter-
pretative phase of archaeological investigation. She addresses the distinctions between
arguments, assertions, and inferences and finishes by referring to Bishop Berkeley,
stating that the aim of interpretation must be limited to the potentialities of the evi-
dence; one can only infer so much from any given assemblage.
In Chapter 17, Plog lays out methodology and goals of archaeological inquiry.
From the construction of arguments and statement of multiple hypotheses to maintain-
ing the focus on observable variation, the article is clear cut and readable.
Babits prepared Chapter 18 to show students how mUltiple working hypotheses
were used in a field investigation. The hypotheses were incorporated into a brief field
report given to engineers and lawyers working for a company controlling land access to
the site. Their granting permission for future access depended on an understanding of
what was being done and why. In many ways, this shows an overly academic applica-
tion to field work but it was created precisely to show students how to organize their
own research problems.

143
 Chapter 14

The Method of Multiple

Working Hypotheses
With This Method the Dangers of
Parental Affection for a Favorite
Theory Can Be Circumvented

T C. CHAMBERLIN

As methods of study constitute the leading theme of our session, I have chosen as a
subject in measurable consonance the method of multiple working hypotheses in its
application to its investigation, instruction, and citizenship.
There are two fundamental classes of study. The one consists in attempting to fol-
low by close imitation the processes of previous thinkers, or to acquire by memorizing
the results of their investigations. It is merely secondary, imitative, or acquisitive study.
The other class is primary or creative study. In it the effort is to think independently, or
at least individually, in the endeavor to discover new truth, or to make new combina-
tions of truth, or at least to develop an individualized aggression of truth. The endeavor
is to think for one's self; whether the thinking lies wholly in the fields of previous
thought or not. It is not necessary to this habit of study that the subject-material should
be new; but the processes of thought and its results must be individual and independent,
not the mere following of previous lines of thought ending in predetermined results.
The demonstration of a problem in Euclid precisely as laid down is an illustration of
the former; the demonstration of the same proposition by a method of one's own or in a
Source: Science (1965), 148:754-759. This paper was originally presented in 1890. It was reprinted by Science
about 1932, and again in 1965 after John R. Platt called attention to it in his 1964 article "Strong Inference" (see
Chapter 15). Copyright 1965 by the American Association for the Advancement of Science and reprinted by per-
mission.

145
 146 T.C. Chamberlin

manner distinctly individual is an illustration of the latter; both lying entirely in the
realm of the known and the old.
Creative study, however, finds its largest application in those subjects in which,
while much is known, more remains to be known. Such are the fields which we, as
naturalists, cultivate; and we are gathered for the purpose of developing improved
methods lying largely in the creative phase of study, though not wholly so.
Intellectual methods have taken three phases in the history of progress thus far.
What may be the evolutions of the future may not be prudent to forecast. Naturally the
methods we now urge seem the highest attainable. These three methods may be desig-
nated, first, the method of the ruling theory; second, the method of the working
hypothesis; and, third, the method of multiple working hypotheses.
In the earlier days of intellectual development the sphere of knowledge was lim-
ited, and was more nearly within the compass of a single individual; and those who
assumed to be wise men, or aspired to be thought so, felt the need of knowing, or at
least seeming to know, all that was known as a justification of their claims. So, also,
there grew an expectancy on the part of the multitude that the wise and the learned
would explain whatever new thing presented itself. Thus, pride and ambition on the one
hand, and expectancy on the other, developed the putative wise man whose knowledge
boxed the compass, and whose acumen found an explanation for every new puzzle that
presented itself: This disposition has propagated itself; and has come down to our time
as an intellectual predilection, though the compassing of the entire horizon of knowl-
edge has long since been an abandoned affectation. As in the earlier days, so still, it is
the habit of some to hastily conjure up an explanation for every new phenomenon that
presents itself. Interpretation rushes to the forefront as the chief obligation pressing
upon the putative wise man. Laudable as the effort at explanation is in itself, it is to be
condemned when it runs before a serious inquiry into the phenomenon itself. A domi-
nant disposition to find out what is, should precede and crowd aside the question, com-
mendable at a later stage, "How came this so?" First full facts, then interpretations.

PREMATURE THEORIES
The habit of precipitate explanation leads rapidly on to the development of tenta-
tive theories. The explanation offered for a given phenomenon is naturally, under the
impulse of self-consistency, offered for like phenomena as they present themselves,
and there is soon developed a general theory explanatory of a large class of phenomena
similar to the original one. This general theory may not be supported by any further
considerations than those which were involved in the first hasty inspection. For a time
it is likely to be held in a tentative way with a measure of candor. With this tentative
spirit and measurable candor, the mind satisfies its moral sense, and deceives itself with
the thought that it is preceding cautiously and impartially toward the goal of ultimate
truth. It fails to recognize that no amount of provisional holding of a theory, so long as
the view is limited and the investigation partial, justifies an ultimate conviction. It is
not the slowness with which conclusions are arrived at that should give satisfaction to
the moral sense, but the thoroughness, the completeness, the all-sidedness, the imparti-
ality, of the investigation.
It is in this tentative stage that the affections enter with their blinding influence.
Love was long since represented as bend, and what is true in the personal realm is meas-
The Method of Multiple Working Hypotheses 147

urably true in the intellectual realm. Important as the intellectual affections are as stimuli
and as reward, they are nevertheless dangerous factors, which menace the integrity of the
intellectual processes. The moment one has offered an original explanation for a phe-
nomenon which seems satisfactory, that moment affection for his intellectual child
springs into existence; and as the explanation grows into a definite theory, his parental
affections cluster about his intellectual offspring, and it grows more and more dear to
him, so that, while he holds it seemingly tentative, it is still lovingly tentative, and not
impartially tentative. So soon as this parental affection takes possession of the mind,
there is a rapid passage to the adoption of the theory. There is an unconscious selection
and magnifying of the phenomena that fall into harmony with the theory and support it,
and an unconscious neglect of those that fail of coincidence. The mind lingers with
pleasure upon the facts that fall happily into the embrace of the theory, and feels a natu-
ral coldness toward those that seem refractory. Instinctively there is a special searching-
out of phenomena that support it, for the mind is led by its desires. There springs up,
also, an unconscious pressing of the theory to make it fit the facts, and a pressing of the
facts to make them fit the theory. When these biasing tendencies set in, the mind rapidly
degenerates into the partiality of paternalism. The search for facts, the observation of
phenomena and their interpretation, are all dominated by affection for the favored theory
until it appears to its author or its advocate to have been overwhelmingly established.
The theory then rapidly rises to the ruling position, and investigation, observation, and
interpretation are controlled and directed by it. From an unduly favored child, it readily
becomes master, and leads its author withersoever it will. The subsequent history of that
mind in respect to the theme is but the progressive dominance of a ruling idea.
Briefly summed up, the evolution is this: A premature explanation passes into a
tentative theory, then into an adopted theory, and then into a ruling theory.
When the last stage has been reached, unless the theory happens, perchance, to be
the true one, all hope of the best results is gone. To be sure, truth may be brought forth
by an investigator dominated by a false ruling idea. His very errors may indeed stimu-
late investigation on the part of others. But the condition is an unfortunate one. Dust
and chaff are mingled with the grain in what should be a winnowing process.

RULING THEORIES LINGER

As previously implied, the method of the ruling theory occupied a chief place
during the infancy of investigation. It is an expression of the natural infantile tenden-
cies of the mind, though in this case applied to its higher activities, for in the earlier
stages of development the feelings are relatively greater than in later stages.
Unfortunately it did not wholly pass away with the infancy of investigation, but
has lingered along in individual instances to the present day, and finds illustration in
universally learned men and pseudoscientists of our time.
The defects of the method are obvious, and its errors great. If I were to name the
central psychological fault, I should say that it was the admission of intellectual affec-
tion to the place that should be dominated by impartial intellectual rectitude.
So long as intellectual interest dealt chiefly with the intangible, so long it was
possible for this habit of thought to survive, and to maintain its dominance, because the
phenomena themselves, being largely subjective, were plastic in the hands of the ruling
idea, but soon as investigation turned itself earnestly to an inquiry into natural phenom-
 148 T.C. Chamberlin

ena, whose manifestations are tangible, whose properties are rigid, whose laws are rig-
orous, the defects of the method became manifest, and an effort at reformation ensued.
The first great endeavor was repressive. The advocates of reform insisted that theoriz-
ing should be restrained, and efforts directed to the simple determination of facts. The
effort was to make scientific study factitious instead of causal. Because theorizing in
narrow lines had led to manifest evils, theorizing was to be condemned. The reforma-
tion urged was not the proper control and utilization of theoretical effort, but its sup-
pression. We do not need to go backward more than 20 years to find ourselves in the
midst of this attempted reformation. Its weakness lay in its narrowness and its restric-
tiveness. There is no nobler aspiration of the human intellect than desire to compass the
cause of things. The disposition to find explanations and to develop theories is laudable
in itself. It is only its ill use that is reprehensible. The vitality of study quickly disap-
pears when the object sought is a mere collocation of dead unmeaning facts.
The inefficiency of this simply repressive reformation becoming apparent,
improvement was sought in the method of the working hypothesis. This is affirmed to
be the scientific method of the day, but to this I take exception. The working hypothesis
differs from the ruling theory in that it is used as a means of determining facts, and has
for its chief function the suggestion of lines of inquiry; the inquiry being made, not for
the sake of the hypothesis, but for the sake of facts. Under the method of the ruling the-
ory, the stimulus was directed to the finding of facts for the support of the theory.
Under the working hypothesis, the facts are sought for the purpose of ultimate induc-
tion and demonstration, the hypothesis being but a means for the more ready develop-
ment of facts and of their relations, and the arrangement and preservation of material
for the final induction.
It will be observed that the distinction is not a sharp one, and that a working
hypothesis may with the utmost ease degenerate into a ruling theory. Affection may as
easily cling about a hypothesis as about theory, and the demonstration of the one may
become a ruling passion as much as of the other.

A FAMILY OF HYPOTHESES
Conscientiously followed, the method of the working hypothesis is a marked
improvement upon the method of the ruling theory; but it has its defects - defects
which are perhaps best expressed by the ease with which the hypothesis becomes a
controlling idea. To guard against this, the method of multiple working hypotheses is
urged. It differs from the former method in the multiple character of its genetic concep-
tion and of its tentative interpretation. It is directed against the radical defect of the two
other methods; namely, the partiality of intellectual parentage. The effort is to bring up
into view every rational explanation of new phenomena, and to develop every tenable
hypothesis respecting their cause and history. The investigator thus becomes the parent
of a family of hypotheses: and, by his parental relation to all, he is forbidden to fasten
his affections unduly on anyone. In the nature .of the case, the danger that springs from
affection is counteracted, and therein is a radical difference between this method and
the two preceding. The investigator at the outset puts himself in cordial sympathy and
in parental relations (of adoption, if not of authorship) with every hypothesis that is at
all applicable to the case under investigation. Having thus neutralized the partialities of
his emotional nature, he proceeds with a certain natural and enforced erectness of men-
The Method of Multiple Working Hypotheses 149

tal attitude to the investigation, knowing well that some of his intellectual children will
die before maturity, yet feeling that several of them may survive the results of final
investigation, since it is often the outcome of inquiry that several causes are found to be
involved instead of a single one. In following a single hypothesis, the mind is presum-
ably led to a single explanatory conception. But an adequate explanation often involves
the coordination of several agencies, which enter into the combined result in varying
proportions. The true explanation is therefore necessarily complex. Such complex
explanations of phenomena are specially encouraged by the method of multiple
hypotheses, and constitute one of its chief merits. We are so prone to attribute a phe-
nomenon to a single cause, that, when we find an agency present, we are liable to rest
satisfied therewith, and fail to recognize that it is but one factor, and perchance a minor
factor, in the accomplishment of the total result. Take for illustration the mooted ques-
tion of the origin of the Great Lake basins. We have this, that, and the other hypothesis
urged by different students as the cause of these great excavations; and all of these are
urged with force and with fact, urged justly to a certain degree. It is practically demon-
strable that these basins were river-valleys antecedent to the glacial incursion and that
they owe their origin in part to the preexistence of those valleys and to the blocking-up
of their outlets. And so this view of their origin is urged with a certain truthfulness. So,
again, it is demonstrable that they were occupied by great lobes of ice, which excavated
them to a marked degree, and therefore the theory of glacial excavation finds support in
fact. I think it is furthermore demonstrable that the earth's crust beneath these basins
was flexed downward, and that they owe a part of their origin to crust deformation. But
to my judgment neither the one nor the other, nor the third, constitutes an adequate
explanation of the phenomena. All these must be taken together, and possibly they
must be supplemented by other agencies. The problem, therefore, is the determination
not only of the participation, but of the measure and the extent, of each of these agen-
cies in the production of the complex result. This is not likely to be accomplished by
one whose working hypothesis is preglacial erosion, or glacial erosion, or crust defor-
mation, but by one whose staff of working hypotheses embraces all of these and any
other agency which can be rationally conceived to have taken part in the phenomena.
A special merit of the method is that by its very nature it promotes thoroughness.
The value of a working hypothesis lies largely in its suggestiveness of lines of inquiry
that might otherwise be overlooked. Facts that are trivial in themselves are brought into
significance by their bearings upon the hypothesis, and by their causal indications. As an
illustration, it is only necessary to cite the phenomenal influence which the Darwinian
hypothesis has exerted upon the investigations of the past two decades. But a single
working hypothesis may lead investigation along a given line to the neglect of others
equally important: and thus, while inquiry is promoted in certain quarters, the investiga-
tion lacks in completeness. But if all rational hypotheses relating to a subject are worked
coequally, thoroughness is the presumptive result, in the very nature of the case.
In the use of the multiple method, the reaction of one hypothesis on another
stands to amplify the recognized scope of each, and their mutual conflicts whet the dis-
criminative edge of each. The analytic process, the development and demonstration of
criteria, and the sharpening of discrimination, receive powerful impulse from the coor-
dinate working of several hypotheses.
Fertility in processes is also the natural outcome of the method. Each hypothesis
suggests its own criteria, its own means of proof, its own methods of developing the
 150 T.C. Chamberlin

truth; and if a group of hypotheses encompass the subject on all sides, the total out-
come of means and of methods is full and rich.
The use of the method leads to certain peculiar habits of mind which deserve passing
notice, since as a factor of education its disciplinary value is one of importance. When
faithfully pursued for a period of years, it develops a habit of thought analogous to the
method itself, which may be designated a habit of parallel or complex thought. Instead of a
simple succession of thoughts in linear order, the procedure is complex, and the mind ap-
pears to become possessed of the power of simultaneous vision from different standpoints.
Phenomena appear to become capable of being viewed analytically and synthetically at
once. It is not altogether unlike the study of a landscape, from which there comes into the
mind myriads oflines of intelligence, which are received and coordinated simultaneously,
producing a complex impression which is recorded and studied directly in its complexity.
My description of this process is confessedly inadequate, and the affirmation of it as a fact
would doubtless challenge dispute at the hands of psychologists of the old school; but I ad-
dress myself to naturalists who I think can respond to its verity from their own experience.

DRAWBACKS OF THE METHOD

The method has, however, its disadvantages. No good thing is without its draw-
backs; and this very habit of mind, while an invaluable acquisition for purposes of
investigation, introduces difficulties in expression. It is obvious, upon consideration,
that this method of thought is impossible of verbal expression. We cannot put into
words more than a single line of thought at the same time; and even in that the order of
expression must be conformed to the idiosyncracies of the language, and the rate must
be relatively slow. When the habit of complex thought is not highly developed, there is
usually a leading line to which others are subordinate, and the difficulty of expression
does not rise to serious proportion; but when the method of simultaneous vision along
different lines is developed so that the thoughts running in different channels are nearly
equivalent, there is an obvious embarrassment in selection and a disinclination to make
the attempt. Furthermore, the impossibility of expressing the mental operation in words
leads to their disuse in the silent process of thought, and hence words and thoughts lose
that close association which they are accustomed to maintain with those whose silent as
well as spoken thoughts run in linear verbal courses. There is therefore a certain predis-
position on the part of the practitioner of this method to taciturnity.
We encounter an analogous difficulty in the use of the method with young stu-
dents. It is far easier, and I think in general more interesting, for them to argue a theory
or accept a simple interpretation than to recognize and evaluate the several factors
which the true elucidian may require. To illustrate: it is more to their taste to be taught
that the Great Lakes basins were scooped out by glaciers than to be urged to conceive
of three or more great agencies working successively or simultaneously, and to esti-
mate how much was accomplished by each of these agencies. The complex and the
quantitative do not fascinate the young student as they do the veteran investigator.

MULTIPLE HYPOTHESES AND PRACTICAL AFFAIRS

It has not been our custom to think of the method of working hypotheses as appli-
cable to instruction or to the practical affairs of life. We have usually regarded it as but
The Method of Multiple Working Hypotheses 151

a method of science. But I believe its application to practical affairs has a value coordi-
nate with the importance of the affairs themselves. I refer especially to those inquiries
and inspections that precede the coming-out of an enterprise rather than to its actual
execution. The. methods that are superior in scientific investigation should likewise be
superior in those investigations that are the necessary antecedents to an intelligent con-
duct of affairs. But I can dwell only briefly on this phase of the subject.
In education, as in investigation, it has been much the practice to work a theory.
The search for instructional methods has often proceeded on the presumption that there
is a definite patent process through which all students might be put and come out with
results of maximum excellence; and hence pedagogical inquiry in the past has very
largely concerned itself with the inquiry, "What is the best method?" rather than with
the inquiry, "What are the special values of different methods, and what are their sev-
eral advantageous applicabilities in the varied work of instruction?" The past doctrine
has been largely the doctrine of pedagogical uniformitarianism. But the faculties and
functions of the mind are almost, if not quite, as varied as the properties and functions
of matter: and it is perhaps not less absurd to assume that any specific method of
instructional procedure is more effective than all others, under any and all circum-
stances, than to assume that one principle of interpretation is equally applicable to all
phenomena of nature. As there is an endless variety of mental processes and combina-
tions and an indefinite number of orders of procedure, the advantage of different meth-
ods under different conditions is almost axiomatic. This being granted, there is
presented to the teacher the problem of selection and of adaptation to meet the needs of
any specific issue that may present itself. It is important, therefore, that the teacher
shall have in mind a full array of possible conditions and states of mind which may be
presented, in order that, when anyone of these shall become an actual case, he may
recognize it, and be ready for the emergency.
Just as the investigator armed with many working hypotheses is more likely to
see the true nature and significance of phenomena when they present themselves, so the
instructor equipped with a full panoply of hypotheses ready for application more read-
ily recognizes the actuality of the situation, more accurately measures its significance,
and more appropriately applies the methods which the case calls for.
The application of the method of multiple hypotheses to the varied affairs of life
is almost as protean as the phases of that life itself, but certain general aspects may be
taken as typical of the whole. What I have just said respecting the application of the
method to instruction may apply, with a simple change of terms, to almost any other
endeavor which we are called upon to undertake. We enter upon an enterprise in most
cases without full knowledge of all the factors that will enter into it, or all of the possi-
ble phases which it may develop. It is therefore of the utmost importance to be pre-
pared to rightly comprehend the nature, bearing, and influence of such unforeseen
elements when they shall definitely present themselves as actualties. If our vision is
narrowed by a preconceived theory as to what will happen, we are almost certain to
misinterpret the facts and to misjudge the issue. If, on the other hand, we have in mind
hypothetical forecasts of the various contingencies that may arise, we shall be the more
likely to recognize the true facts when they do present themselves. Instead of being
biased by the anticipation of a given phase, the mind is rendered open and alert by the
anticipation of anyone of many phases, and is free not only, but is predisposed, to rec-
ognize correctly the one which does appear. The method has a further good effect. The
 152 T.C. Chamberlin

mind, having anticipated the possible phases which may arise, has prepared itself for
action under anyone that may come up, and it is therefore readily armed, and is predis-
posed to act in the line appropriate to the event. It has not set itself rigidly in a fixed
purpose, which it is predisposed to follow without regard to contingencjes. It has not
nailed down the helm and predetermined run a specific course, whether rocks lie in the
path or not: but, with the helm in hand, it is ready to veer the ship according to whether
danger or advantage discovers itself.
It is true, there are often advantages in pursuing a fixed predetermined course
without regard to obstacles or adverse condition. Simple dogged resolution is some-
times the salvation of an enterprise; but, while glorious successes have been thus
snatched from the very brink of disaster, overwhelming calamity has in other cases fol-
lowed it on this course, when a reasonable regard for the unanticipated elements of suc-
cess. So there is to be set over against the great achievements that follow on dogged
adherence great disasters which are equally its result.

DANGER OF VACILLATION

The tendency of the mind, accustomed to work through multiple hypotheses, is to

say to one line of policy or another, according as the balance of evidence shall incline.
This is the soul and essence of the method. It is in general the true method. Neverthe-
less there is a danger that this yielding to evidence may degenerate into unwarranted
vacillation. It is not always possible for the mind to balance evidence with exact equi-
poise, and to determine, in the midst of the execution of an enterprise, what is the
measure of probability on the one side or the other; and as difficulties present them-
selves, there is a danger of being biased by them and of swerving from the course that
was really the true one. Certain limitations are therefore to be placed on the application
of the method, for it must be remembered that a poorer line of policy consistently
adhered to may bring better results than a vacillation between better policies.
There is another and closely allied danger in the application of the method. In its
highest development it presumes a mind supremely sensitive to every grain of evi-
dence. Like a pair of delicately poised scales, every added particle on the one side or
the other produces its effect in oscillation. But such a pair of scales may be altogether
too sensitive to be of practical value in the rough affairs of life. The balances of the
exact chemist are too delicate for the weighing-out of coarse commodities. Dispatch
may be more important than accuracy. So it is possible for the mind to be too much
concerned with the nice balancings of evidence, and to oscillate too much and too long
in the endeavor to reach exact results. It may be better, in the gross affairs of life, to be
less precise and more prompt. Quick decisions, though they may contain a grain of
error, are oftentimes better than precise decisions at the expense of time.
The method has a special beneficent application to our social and civic relations.
Into these relations there enter, as great factors, our judgment of others, our discern-
ment of the nature of their acts, and our interpretation of their motives and purposes.
The method of multiple hypotheses, in its application here, stands in decided contrast to
the method of the ruling theory or of the simple working hypothesis. The primitive
habit is to interpret the acts of others on the basis of a theory. Childhood's unconscious
theory is that the good are good and the bad are bad. From the good the child expects
nothing but good; from the bad, nothing but bad. To expect a good act from the bad, or
The Method of Multiple Working Hypotheses 153

a bad act from the good, is radically at variance with childhood's mental methods.
Unfortunately in our social and civic affairs too many of our fellow citizens have never
outgrown the ruling theory of their childhood.
Many have advanced a step further, and employ a method analogous to that of the
working hypothesis. A certain presumption is made to attach to the acts of their fellow-
beings, and that which they see is seen in the light of that presumption, and that which
they construe is construed in the light of that presumption. They do not go to the
lengths of childhood's method by assuming positively that the good are wholly good,
and the bad wholly bad; but there is a strong presumption in their minds that that he
concerning whom they have an ill opinion will act from corresponding motives. It
requires positive evidence to overthrow the influence of the working hypothesis.
The method of multiple hypotheses assumes broadly that the acts of a fellow
being may be diverse in their nature, their moves, their purposes, and hence in their
whole moral character; that they may be good though the dominant character be bad;
that they may be bad though the dominant character be good; that they may be partly
good and partly bad, as is the fact in the greater number of the complex activities of a
human being. Under the method of multiple hypotheses, it is the first effort of the mind
to see truly what the act is, unbeclouded by the presumption that this or that has been
done because it accords with our ruling theory or our working hypothesis. Assuming
that acts of similar general aspect may readily take anyone of several different phases,
the mind is freer to see accurately what has actually been done. So, again, in our inter-
pretations of motives and purposes, the method assumes that these may have been any
one of many, and the first duty is to ascertain which of possible motives and purposes
actually prompted this individual action. Going with this effort there is a predisposition
to balance all evidence fairly, and to accept that interpretation to which the weight of
evidence inclines, not that which simply fits our working hypothesis or our dominant
theory. The outcome, therefore, is better and truer observation and juster and more
righteous interpretation.

IMPERFECTIONS OF KNOWLEDGE

There is a third result of great importance. The imperfections of our knowledge are
more likely to be detected, for there will be less confidence in its completeness in propor-
tion as there is a broad comprehension of the possibilities of varied action, under similar
circumstances and with similar appearances. So, also, the imperfections of evidence as to
the motives and purposes inspiring the action will become more discernible in proportion
to the fullness of our conception of what the evidence should be to distinguish between
action from the one or the other of possible motives. The necessary result will be a lesser
disposition to reach conclusions on imperfect grounds. So, also, there will be a lesser
inclination to misapply evidence; for, several constructions being definitely in mind, the
indices of the one motive are less liable to be mistaken for the indices of another.
The total outcome is greater care in ascertaining the facts, and greater discrimina-
tion and caution in drawing conclusions. I am confident, therefore, that the general
application of this method to the affairs of social and civic life would go far to remove
those misunderstandings, misjudgments, and misrepresentations which constitute so
pervasive an evil in our social and our political atmospheres, the source of immeasur-
able suffering to the best and most sensitive souls. The misobservations, the misstate-
154 T.C. Chamberlin

ments, the misinterpretations, of life may cause less gross suffering than some other
evils; but they being more universal and more subtle, pain. The remedy lies, indeed,
partly in charity, but more largely in correct intellectual habits, in a predominant, ever-
present disposition to see things as they are, and to judge them in the full light of an
unbiased weighing of evidence applied to all possible constructions, accompanied by a
withholding of judgment when the evidence is insufficient to justify conclusions.
I believe that one of the greatest moral reforms that lies immediately before us
consists in general introduction into social and civic life of that habit of mental proce-
dure which is known in investigation as the method of multiple working hypotheses.

REFERENCE

Chamberlin, T.e., 1965, The Method of Multiple Working Hypotheses. Science 148:754-759.
 Chapter 15

Strong Inference
Certain Systematic Methods
of Scientific Thinking May
Produce Much More Rapid
Progress Than Others

JOHN R. PLATT

Scientists these days tend to keep up a polite fiction that all science is equal. Except for
the work of the misguided opponent whose arguments we happen to be refuting at the
time, we speak as though every scientist's field and methods of study are as good as
every other scientist's, and perhaps a little better. This keeps us all cordial when it
comes to recommending each other for government grants.
But I think anyone who looks at the matter closely will agree that some fields of
science are moving forward very much faster than others, perhaps by an order of mag-
nitude, if numbers could be put on such estimates. The discoveries leap from the head-
lines - and they are real advances in complex and difficult subjects, like molecular
biology and high-energy physics. As Alvin Weinberg (1964), says "Hardly a month
goes by without a stunning success in molecular biology being reported in the Proceed-
ing of the National Academy of Science."
Why should there be such rapid advances in some fields and not in others? I think
the usual explanations that we tend to think of-such as the tractability of the subject,
or the quality or education of the men drawn into it, or the size of research contracts -
are important but inadequate. I have begun to believe that the primary factor in scien-
Source: Science (1965), 146:347-353. Copyright 1965 by the American Association for the Advancement of
Science and reprinted by permission.

155
 156 J.R. Platt

tific advance is an intellectual one. These rapidly moving fields are fields where a par-
ticular method of doing scientific research is systematically used and taught, an accu-
mulative method of inductive inference that is so effective that I think it should be
given the name of "strong inference." I believe it is important to examine this method,
its use and history and rationale, and to see whether other groups and individuals might
learn to adopt it profitably in their own scientific and intellectual work.
In its separate elements, strong inference is just the simple and old-fashioned
method of inductive inference that goes back to Francis Bacon. The steps are familiar
to every college student and are practiced, off and on, by every scientist. The difference
comes in their systematic application. Strong inference consists of applying the follow-
ing steps to every problem in science, formally and explicitly and regularly:
1. Devising alternative hypotheses
2. Devising a crucial experiment (or several of them), with alternative possible
outcomes, each of which will, as nearly as possible, exclude one or more of
the hypotheses
3. Carrying out the experiment so as to get a clean result
4. Recycling the procedure, making subhypotheses or sequential hypotheses to
refine the possibilities that remain, and so on
It is like climbing a tree. At the first fork, we choose or, in this case, "nature" or
the experimental outcome chooses to go to the right branch or the left; at the next fork,
to go left or right; and so on. There are similar branch points in a "conditional com-
puter program," where the next move depends on the result of the last calculation. And
there is a "conditional inductive tree" or "logical tree" of this kind written out in detail
in many first-year chemistry books, in the table of steps for qualitative analysis of an
unknown sample, where the student is led through a real problem of consecutive infer-
ence: Add reagent A; if you get a red precipitate, it is subgroup alpha and you filter and
add reagent B'; if not, you add the other reagent, B; and so on.
On any new problem, of course, inductive inference is not as simple and certain
as deduction, because it involves reaching out into the unknown. Steps 1 and 2 require
intellectual inventions, which must be cleverly chosen so that hypothesis, experiment,
outcome, and exclusion will be related in a rigorous syllogism; and the question of how
to generate such inventions is one which has been extensively discussed elsewhere
(Polya, 1954; Platt, 1962). What the formal schema reminds us to do is to try to make
these inventions, to take the next step, to proceed to the next fork, without dawdling or
getting tied up in irrelevancies.
It is clear why this makes for rapid and powerful progress. For exploring the
unknown, there is no faster method: this is the minimum sequence of steps. Any con-
clusion that is not an exclusion is insecure and must be rechecked. Any delay in recy-
cling to the next set of hypotheses is only a delay. Strong inference, and the logical tree
it generates, are to inductive reasoning what the syllogism is to deductive reasoning, in
that it offers a regular method for reaching firm inductive conclusions one after the
other as rapidly as possible.
"But what is so novel about this?" someone will say. This is the method of sci-
ence and always has been; why give it a special name? The reason is that many of us
have almost forgotten it. Science is now an everyday business. Equipment, calcula-
tions, lectures become ends in themselves. How many of us write down our alternatives
Strong Inference 157

and crucial experiments every day, focusing on the exclusion of a hypothesis? We may
write our scientific papers so that it looks as if we had steps I, 2, and 3 in mind all
along. But in between we do busywork. We become "method-oriented," rather then
"problem-oriented." We say we prefer to "feel our way" toward generalizations. We
fail to teach our students how to sharpen up their inductive inferences. And we do not
realize the added power that the regular and explicit use of alternative hypotheses and
sharp exclusions could give us at every step of our research.
The difference between the average scientist's informal methods and the methods
of the strong-inference users is somewhat like the difference between a gasoline engine
that fires occasionally and one that fires in steady sequence. If our motorboat engines
were as erratic as our deliberate, intellectual effort, most of us would not get home for
supper.

MOLECULAR BIOLOGY

The new molecular biology is a field where I think this systematic method of
inference has become widespread and effective. It is a complex field; yet a succession
of crucial experiments over the past decade has given us a surprisingly detailed under-
standing of hereditary mechanisms and the control of enzyme formation and protein
synthesis.
The logical structure shows in every experiment. In 1953 James Watson and
Francis Crick proposed that the DNA molecule - the hereditary substance in a cell-
is a long two-stranded helical molecule. This suggested a number of alternatives for
crucial testing. Do the two strands of the helix stay together when a cell divides, or do
they separate? Matthew Meselson and Franklin Stahr (1958) used an ingenious isotope-
density-labeling technique which showed that they separate. Does the DNA helix al-
ways have two strands, or can it have three, as atomic models suggest? Alexander Rich
(1959) showed it can have either depending on the ionic concentration. These are the
kinds of experiments John Dalton would have liked, where the combining entities are
not atoms but long macromolecular strands.
Or take a different sort of question: Is the "genetic map" - showing the statisti-
cal relationship of different genetic characteristics in recombination experiments - a
one-dimensional map like the DNA molecule (that is, a linear map), as T. H. Morgan
proposed in 1911, or does it have two-dimensional loops or branches? Seymour Benzer
(1959) showed that his hundreds of fine microgenetic experiments on bacteria would fit
only the mathematical matrix for the one-dimensional case.
But of course, selected crucial experiments of this kind can be found in every
field. The real difference in molecular biology is that formal inductive inference is so
systematically practiced and taught. On any given morning at the Laboratory of
Molecular Biology in Cambridge, England, the blackboards of Francis Crick or Sidney
Brenner will commonly be found covered with logical trees. On the top line will be the
hot new result just up from the laboratory or just in by letter or rumor. On the next line
will be two or three alternative explanations, or a little list of "What he did wrong."
Underneath will be a series of suggested experiments or controls that can reduce the
number of possibilities. And so on. The tree grows during the day as one man or
another comes in and argues about why one of the experiments wouldn't work, or how
it should be changed.
 158 J.R. Platt

The strong-inference attitude is evident just in the style and language in which
the papers are written. For example, in analyzing theories of antibody formation,
Joshua Lederberg (1959) gives a list of nine propositions "'subject to denial," discuss-
ing which ones would be "most vulnerable to experimental test."
The papers of the French leaders Francois Jacob and Jacques Monod are also
celebrated for their high "logical density," with paragraph after paragraph of linked
"inductive syllogisms." But the style is widespread. Start with the first paper in the
Journal of Molecular Biology for 1964 (Davison et al., 1964), and you immediately
find: "our conclusions ... might be invalid if...(i) ... (ii) ... or (iii) .... We shall describe
experiments which eliminate these alternatives." The average physicist or chemist or
scientist in any field accustomed to less closely reasoned articles and less sharply stated
inferences will find it a salutary experience to dip into that journal almost at random.

RESISTANCE TO ANALYTICAL METHODOLOGY

This analytical approach to biology has sometimes become almost a crusade,

because it arouses so much resistance in many scientists who have grown up in a more
relaxed and diffuse tradition. At the 1958 Conference on Biophysics, at Boulder, there
was a dramatic confrontation between the two points of view. Leo Szilard said: "The
problem of how enzymes are induced, of how proteins are synthesized, of how antibod-
ies are formed, are closer to solution than is generally believed. If you do stupid experi-
ments and finish one a year, it can take 50 years. But if you stop doing experiments for
a little while and think how proteins can possibly be synthesized, there are only about 5
different ways, not 50! And it will take only a few experiments to distinguish these."
One of the young men added: "It is essentially the old question: How small and
elegant an experiment can you perform?"
These comments upset a number of those present. An electron microscopist said,
"Gentlemen, this is off the track. This is philosophy of science."
Szilard retorted, "I was not quarreling with third-rate scientists: I was quarreling
with first-rate scientists."
A physical chemist hurriedly asked, "Are we going to take the official photo-
graph before lunch or after lunch?"
But this did not deflect the dispute. A distinguished cell biologist rose and said,
"No two cells give the same properties. Biology is the science of heterogeneous sys-
tems." And he added privately, "you know there are scientists; and there are people in
science who are just working with these oversimplified model systems - DNA chains
and in vitro systems - who are not doing science at all. We need their auxiliary work:
they build apparatus, they make minor studies, but they are not scientists."
To which Cy Levinthal replied: "Well, there are two kinds of biologists, those
who are looking to see if there is one thing that can be understood, and those who keep
saying it is very complicated and that nothing can be understood .... You must study
the simplest system you think has the properties you are interested in."
As they were leaving the meeting, one man could be heard muttering, "What does
Szilard expect me to do - shoot myself?"
Any criticism or challenge to consider changing our methods strikes of course at
all our ego-defenses. But in this case the analytical method offers the possibility of
such great increases in effectiveness that it is unfortunate that it cannot be regarded
Strong Inference 159

more often as a challenge to learning rather than as a challenge to combat. Many of the
recent triumphs in molecular biology have in fact been achieved on just such "oversim-
plified model systems," very much along the analytical lines laid down in the 1958 dis-
cussion. They have not fallen to the kind of men who justify themselves by saying, "No
two cells are alike," regardless of how true that may ultimately be. The triumphs are in
fact triumphs of a new way of thinking.

HIGH-ENERGY PHYSICS
This analytical thinking is rare, but it is by no means restricted to the new biol-
ogy. High-energy physics is another field where the logic of exclusions is obvious,
even in the newspaper accounts. For example, in the famous discovery of C. N. Yang
and T. D. Lee, the question that was asked was: Do the fundamental particles conserve
mirror symmetry or "parity" in certain reactions, or do they not? The crucial experi-
ments were suggested; within a few months they were done, and conservation of parity
was found to be excluded. Richard Garwin, Leon Lederman, and Mareel Weinrich
(1957) did one of the crucial experiments. It was thought of one evening at suppertime;
by midnight they had arraigned the apparatus for it; and by 4 a.m. they had picked up
the predicted pulses showing the nonconservation of parity. The phenomena had just
been waiting, so to speak, for the explicit formulation of the alternative hypotheses.
The theorists in this field take pride in trying to predict new properties or new
particles explicitly enough so that if they are not found the theories will fall. As biolo-
gists W.A.H. Rushton (personal communication) has said, "A theory which cannot be
mortally endangered cannot be alive." Murray Gell-Mann and Yuval Ne'eman recently
used the particle grouping which they call "the Eightfold Way" to predict a missing
particle, the omega-minus, which was then looked for and found (Chew et al., 1964).
But one alternative branch of the theory would predict a particle with one-third the
usual electronic charge, and it was not found in the experiments, so this branch must be
rejected.
The logical tree is so much a part of high-energy physics that some stages of it
are commonly built, in fact, into the electronic coincidence circuits that detect the parti-
cles and trigger the bubble-chamber photographs. Each kind of particle should give a
different kind of pattern in the electronic counters, and the circuits can be set to
exclude or include whatever types of events are desired. If the distinguishing criteria
are sequential, they may even run through a complete logical tree in a microsecond or
so. This electronic preliminary analysis, like human preliminary analysis of alternative
outcomes, speeds up progress by sharpening the criteria. It eliminates hundreds of thou-
sands of the irrelevant pictures that formerly had to be scanned, and when it is carried
to its limit, a few output pulses, hours apart, may be enough to signal the existence of
the antiproton or the fall of a theory.
I think the emphasis on strong inference in the two fields I have mentioned has
been partly the result of personal leadership, such as that of the classical geneticists in
molecular biology, or of Szilard with his "Midwest Chowder and Bacteria Society" at
Chicago in 1948-50, or of Max Delbruck with his summer courses in phage genetics at
Cold Spring Harbor. But it is also partly due to the nature of the fields themselves.
Biology, with its vast informational detail and complexity, is a "high-information"
field, where years and decades can easily be wasted on the usual type of "low-informa-
 160 J.R. Platt

tion" observations or experiments if one does not think carefully in advance about what
the most important and conclusive experiments would be. And in high-energy physics,
both the "information flux" of particles from the new accelerators and the million-dol-
lar costs of operation have forced a similar analytical approach. It pays to have a top-
notch group debate every experiment ahead of time; and the habit spreads throughout
the field.

INDUCTION AND MULTIPLE HYPOTHESES

Historically, I think, there have been two main contributions to the development
of a satisfactory strong-inference method. The first is that of Francis Bacon (1960). He
wanted a "surer method" of "finding out nature" than either the logic-chopping of all-
inclusive theories of the time or the laudable but crude attempts to make inductions "by
simple enumeration." He did not merely urge experiments, as some suppose; he
showed the fruitfulness of interconnecting theory and experiment so that the one
checked the other. Of the many inductive procedures he suggested, the most important
I think was the conditional inductive tree, which proceeded from alternative hypotheses
(possible "causes," as he calls them), through crucial experiments ("Instances of the
Fingerpost"), to exclusion of some alternatives and adoption of what is left ("estab-
lishing axioms"). His Instances of the Fingerpost are explicitly at the forks in the logi-
cal tree, the term being borrowed "from the fingerposts which are set up where roads
part, to indicate the several directions."
Many of his crucial experiments proposed in Book II of The New Organon are
still fascinating. For example, in order to decide whether the weight of a body is due to
its "inherent nature," as some had said, or is due to the attraction of the earth, which
would decrease with distance, he proposes comparing the rate of a pendulum clock and
a spring clock and then lifting them from the earth to the top of a tall steeple. He con-
cludes that if the pendulum clock on the steeple "goes more slowly than it did on
account of the diminished virtue of its weights ... we may take the attraction of the mass
of the earth as the cause of weight."
Here was a method that could separate off the empty!
Bacon said the inductive method could be learned by anybody, just like learning
to "draw a straighter line or more perfect circle ... with the help of a ruler or a pair of
compasses." "My way of discovering sciences goes far to level men's wit and leaves
but little to individual excellence, because it performs everything by the surest rules
and demonstrations." Even occasional mistakes would not be fatal. "Truth will sooner
come out from error than from confusion."
It is easy to see why young minds leaped to try it.
Nevertheless there is a difficulty with this method. As Bacon emphasizes, it is
necessary to make "exclusions." He says, "The induction which is to be available for
the discovery and demonstration of sciences and arts, must analyze nature by proper
rejections and exclusions; and then, after a sufficient number of negatives, come to a
conclusion on the affirmative instances." "[To man] it is granted only to proceed at first
by negatives, and at last to end in affirmatives after exclusion has been exhausted."
Or, as the philosopher Karl Popper (1959) says today, there is no such thing as
proof in science - because some later alternative explanation may be as good or better
- so that science advances only by disproofs. There is no point in making hypotheses
Strong Inference 161

that are not falsifiable, because such hypotheses do not say anything: "it must be possi-
ble for an empirical scientific system to be refuted by experience."
The difficulty is that disproof is a hard doctrine. If you have a hypothesis and I
have another hypothesis, evidently one of them must be eliminated. The scientist seems
to have no chance but to be either soft-hearted or disputatious. Perhaps this is why so
many tend to resist the strong analytical approach - and why some great scientists are
so disputatious.
Fortunately, it seems to me, this difficulty can be removed by the use of a second
great intellectual invention, the "method of multiple hypotheses," which is what was
needed to round out the Baconian scheme. This is a method that was put forward by T.
C. Chamberlin (1897), a geologist at Chicago at the tum of the century, who is best
known for his contribution to the Chamberlin-Moulton hypothesis of the origin of the
solar system.
Chamberlin says our trouble is that when we make a single hypothesis, we become
attached to it.

The moment one has offered an original explanation for a phenomenon which seems
satisfactory, that moment affection for his intellectual child springs into existence,
and as the explanation grows into existence, and as the explanation grows into a defi-
nite theory, his parental affections cluster about his offspring and it grows more and
more dear to him .... There springs up also unwittingly a pressing of the theory to
make it fit the facts and a pressing of the facts to make them fit the theory ...

To avoid this grave danger, the method of multiple working hypotheses is urged. It
differs from the simple working hypothesis in that it distributes the effort and divides
the affections .... Each hypothesis suggests its own criteria, its own means of proof,
its own method of developing the truth, and if a group of hypotheses encompass the
subject on all sides, the total outcome of means and of methods is full and rich.

Chamberlin thinks the method "leads to certain distinctive habits of mind" and is
of prime value in education. "When faithfully followed for a sufficient time, it devel-
ops a mode of thought of its own kind which may be designated the habit of complex
thought."
This charming paper deserves to be reprinted in some more accessible journal
today, where it could be required reading of every graduate student - and for every
professor.
It seems to me that Chamberlin has hit on the explanation - and the cure - for
many of our problems in the sciences. The conflict and exclusion of alternatives that is
necessary to sharp inductive inference has been all too often a conflict between men,
each with his single Ruling Theory. But whenever each man begins to have multiple
working hypotheses, it becomes purely a conflict between ideas. It becomes much easier
then for each of us to aim every day at conclusive disproofs - at strong inference-
without either reluctance or combativeness. In fact, when there are multiple hypotheses
which are not anyone's "personal property" and when there are crucial experiments to
test them, the daily life in the laboratory takes on an interest and excitement it never had
and the students can hardly wait to get to work to see how the detective story will come
out. It seems to me that this is the reason for the development of those "distinctive habits
of mind" and the "complex thought" that Chamberlin described, the reason for the sharp-
 162 J.R. Platt

ness, the excitement, the zeal, the teamwork - yes, even international teamwork - in
molecular biology and high-energy physics today. What else could be so effective?
When multiple hypotheses become coupled to strong inference, the scientific
search becomes an emotional powerhouse as well as an intellectual one.
Unfortunately, I believe, there are other areas of science today that are sick by
comparison, because they have forgotten the necessity for alternative hypotheses and
disproof. Each man has only one branch - or none - on the logical tree, and it twists
at random without ever coming to the need for a crucial decision at any point. We can
see from the external symptoms that there is something scientifically wrong. The Fro-
zen Method. The Eternal Surveyor. The Never Finished. The Great Man With a Single
Hypothesis. The Little Club of Dependents. The Vendetta. The All-Encompassing The-
ory Which Can Never Be Falsified.
Some cynics tell a story, which may be apocryphal, about the theoretical chemist
who explained to his class.
"And thus we see that the C-Cl bond is longer in the first compound than in the
second because the percent of ionic character is smaller."
A voice from the back of the room said, "But Professor X, according to the
Table, the C-Cl bond is shorter in the first compound."
"Oh, is it?" said the professor, "Well, that's still easy to understand, because the
double-bond character is higher in that compound."
To the extent that this kind of story is accurate, a "theory" of this sort is not a the-
ory at all, because it does not exclude anything. It predicts everything, and therefore
does not predict anything. It becomes simply a verbal formula which the graduate stu-
dent repeats and believes because the professor has said it so often. This is not science,
but faith; not theory, but theology. Whether it is hand-waving or number-waving or
equation-waving, a theory is not a theory unless it can be disproved. That is, unless it
can be falsified by some possible experimental outcome.
In chemistry, the resonance theorists will of course suppose that I am criticizing
them, while the molecular-orbital theorists will suppose I am criticizing them. But their
actions - our actions, for I include myself among them - speak for themselves. A
failure to agree for 30 years is public advertisement of a failure to disprove.
My purpose here, however, is not to call names but rather to say that we are all
sinners, and that in every field and in every laboratory we need to try to formulate mul-
tiple alternative hypotheses sharp enough to be capable of disproof.

SYSTEMATIC APPLICATION

I think the work methods of a number of scientists have been testimony to the
power of strong inference. Is success not due in many cases to systematic use of Ba-
con's "surest rules and demonstration" as much as to rare and unattainable intellectual
power? Faraday's famous diary (Faraday, 1932-1936), or Fermi's notebooks (Platt,
1962; Anderson and Allison, 1955), show how these men believed in the effectiveness
of daily steps in applying formal inductive methods to one problem after another.
Within 8 weeks after the discovery of X rays, Roentgen had identified 17 of their
major properties. Every student should read his first paper (Watson, 1945). Each dem-
onstration in it is a little jewel of inductive inference. How else could the proofs have
gone so fast, except by a method of maximum effectiveness?
Strong Inference 163

Organic chemistry has been the spiritual home of strong inference from the be-
ginning. Do the bonds alternate in benzene or are they equivalent? If the first, there
should be five disubstituted derivatives; if the second, three. And three it is (Wheland,
1949). This is a strong-inference test - not a matter of measurement, of whether there
are grams or milligrams of the products, but a matter of logical alternatives. How else
could the tetrahedral carbon atom or the hexagonal symmetry of benzene have been in-
ferred 50 years before the inferences could be confirmed by X-ray and infrared meas-
urement?
We realize that it was out of this kind of atmosphere that Pasteur came to the
field of biology. Can anyone doubt that he brought with him a completely different
method of reasoning? Every 2 or 3 years he moved to one biological problem after an-
other, from optical activity to the fermentation of beet sugar, to the "diseases" of wine
and beer, to the disease of silkworms, to the problem of "spontaneous generation," to
the anthrax disease of sheep, to rabies. In each of these fields there were experts in
Europe who knew a hundred times as much as Pasteur, yet each time he solved prob-
lems in a few months that they had not been able to solve. Obviously it was not ency-
clopedic knowledge that produced his success, and obviously it was not simply luck,
when its repeated over and over again: it can only have been the systematic power of a
special method of exploration. Are bacteria falling in? Make the necks of the flasks S-
shaped. Are bacteria sucked in by the partial vacuum? Put in a cotton plug. Week after
week his crucial experiments built up the logical tree of exclusions. The drama of
strong inference in molecular biology today is only a repetition of Pasteur's story.
The grand scientific syntheses, like those of Newton and Maxwell, are rare and
individual achievements that stand outside any rule or method. Nevertheless it is inter-
esting to note that several of the great synthesizers have also shown the strong-infer-
ence habit of thought in their other work, as Newton did in the inductive proofs of his
Opticks and Maxwell did in his experimental proof that three and only three colors are
needed in color vision.

A YARDSTICK OF EFFECTIVENESS
I think the evident effectiveness of the systematic use of strong inference suddenly
gives us a yardstick for thinking about the effectiveness of scientific methods in general.
Surveys, taxonomy, design of equipment, systematic measurements and tables, theoreti-
cal computations - all have their proper and honored place, provided they are parts of a
chain of precise induction of how nature works. Unfortunately, all too often they become
ends in themselves, mere time-serving from the point of view of real scientific advance, a
hypertrophied methodology that justifies itself as a lore of respectability.
We praise the "lifetime of study," but in dozens of cases, in every field, what was
needed was not a lifetime but rather a few short months or weeks of analytical induc-
tive inference. In any new area we should try, like Roentgen, to see how fast we can
pass from the general survey to analytical inferences. We should try, like Pasteur, to
see whether we can reach strong inferences that encyClopedism could not discern.
We speak piously of taking measurements and making small studies that will
"add another brick to the temple of science." Most such bricks just lie around the brick-
yard (Forse her, 1963). Tables of constants have their place and value, but the study of
one spectrum after another, if not frequently reevaluated, may become a substitute for
 164 J.R. Platt

thinking, a sad waste of intelligence in a research laboratory, and a mistraining whose

crippling effects may last a lifetime.
To paraphrase an old saying, beware of the man of one method or one instrument,
either experimental or theoretical. He tends to become method-oriented rather than
problem-oriented. The method-oriented man is shackled: the problem-oriented man is
at least reaching freely toward what is most important. Strong inference redirects a man
to problem orientation, but it requires him to be willing repeatedly to put aside his last
methods and teach himself new ones.
On the other hand, I think that anyone who asks the question about scientific ef-
fectiveness will also conclude that much of the mathematicizing in physics and chemis-
try today is irrelevant ifnot misleading.
The great value of mathematical formulation is that when an experiment agrees
with a calculation to five decimal places, a great many alternative hypotheses are pretty
well excluded (though the Bohr theory and the Schrodinger theory both predict exactly
the same Rydberg constant!). But when the fit is only to two decimal places, or one, it
may be a trap for the unwary; it may be no better than any rule-of-thumb extrapolation,
and some other kind of qualitative exclusion might be more rigorous for testing the as-
sumptions and more important to scientific understanding than the quantitative fit.
I know that this is like saying that the emperor has no clothes. Today we preach
that science is not science unless it is quantitative. We substitute correlations for causal
studies, and physical equations for organic reasoning. Measurements and equations are
supposed to sharpen thinking, but, in my observation, they more often tend to make the
thinking noncausal and fuzzy. They tend to become the object of scientific manipula-
tion instead of auxiliary tests of crucial inferences.
Many - perhaps most - of the great issues of science are qualitative, not quan-
titative, even in physics and chemistry. Equations and measurements are useful when
and only when they are related to proof; but proof or disproof comes first, and is in fact
strongest when it is absolutely convincing without any quantitative measurement.
Or to say it another way, you can catch phenomena in a logical box or in a mathe-
matical box. The logical box is coarse but strong. The mathematical box is fine-grained
but flimsy. The mathematical box is a beautiful way of wrapping up a problem, but it
will not hold the phenomena unless they have been caught in a logical box to begin with.
What I am saying is that, in numerous areas that we call science, we have come
to like our habitual ways, and our studies that can be continued indefinitely. We meas-
ure, we define, we compute, we analyze, but we do not exclude. And this is not the way
to use our minds most effectively or to make the fastest progress in solving scientific
questions.
Of course it is easy - and all too common - for one scientist to call the others
unscientific. My point is not that my particular conclusions here are necessarily correct,
but that we have long needed some absolute standard of possible scientific effective-
ness by which to measure how well we are succeeding in various areas - a standard
that many could agree on and one that would be undistorted by the scientific pressures
and fashions of the times and the vested interests and busywork that they develop. It is
not public evaluation I am interested in so much as a private measure by which to com-
pare one's own scientific performance with what it might be. I believe that strong infer-
ence provides this kind of standard of what the maximum possible scientific
effectiveness could be - as well as a recipe for reaching it.
Strong Inference 165

AIDS TO STRONG INFERENCE

How can we learn the method and teach it? It is not difficult. The most important
thing is to keep in mind that this kind of thinking is not a lucky knack but a system that
can be taught and learned. The molecular biologists today are living proof of it. The
second thing is to be explicit and normal and regular about it, to devote a half hour or
an hour to analytical thinking every day, writing out the logical tree the alternatives and
crucial experiments explicitly in a permanent notebook. I have discussed elsewhere
(Platt, 1963) the value of Fermi's notebook method, the effect it had on his colleagues
and students, and the testimony that it "can be adopted by anyone with profit."
It is true that it takes great courtesy to teach the method, especially to one's peers
- or their students. The strong-inference point of view is so resolutely critical of meth-
ods of work and values in science that any attempt to compare specific cases is likely to
sound both smug and destructive. Mainly one should try to teach it by example and by
exhorting to self-analysis and self-improvement only in general terms, as I am doing
here.
But I will mention one severe but useful private test - a touchstone of strong in-
ference - that removes the necessity for third-person criticism, because it is a test that
anyone can learn to carry with him for use as needed. It is our old friend the Baconian
"exclusion," but I call it "The Question." Obviously it should be applied as much to
one's own thinking as to others'. It consists of asking in your own mind, on hearing any
scientific explanation or theory put forward, "But sir, what experiment could disprove
your hypothesis?"; or, on hearing a scientific experiment described, "But sir, what hy-
pothesis does your experiment disprove?"
This goes straight to the heart of the matter. It forces everyone to refocus on the
central question of whether there is or is not a testable scientific step forward.
If such a question were asked aloud, many a supposedly great scientist would
sputter and turn livid and would want to throw the questioner out, as a hostile witness!
Such a man is less than he appears, for he is obviously not accustomed to think in terms
of alternative hypotheses and crucial experiments for himself; and one might also won-
der about the state of science in the field he is in. But who knows? - the question
might educate him, and his field too!
On the other hand, I think that throughout most of molecular biology and nuclear
physics the response to The Question would be to outline immediately not one but sev-
eral tests to disprove the hypothesis - and it would turn out that the speaker already
had two or three graduate students working on them!
I almost think that government agencies could make use of this kind of touch-
stone. It is not true that all science is equal, or that we cannot justly compare the effec-
tiveness of scientists by any method other than a mutual-recommendation system. The
man to watch, the man to put your money on, is not the man who wants to make "a sur-
vey" or a "more detailed study" but the man with the notebook, the man with the alter-
native hypotheses and the crucial experiments, the man who knows how to answer your
Question of disproof and is already working on it.
There are some really hard problems, some high-information problems, ahead of
us in several fields, problems of photosynthesis, of cellular organization, of the molecu-
lar structure and organization of the nervous system, not to mention some of our social
and international problems. It seems to me that the method of most rapid progress in
 166 J.R. Platt

such complex areas, the most effective way of using our brains, is going to be to set
down explicitly at each step just what the question is, and what all of the alternatives
are, and then to set up crucial experiments to try to disprove some. Problems of this
complexity, if they can be solved at all, can be solved only by men generating and ex-
cluding possibilities with maximum effectiveness, to obtain a high degree of informa-
tion per unit time - men willing to work a little bit at thinking.
When whole groups of us begin to concentrate like that, I believe we may see the
molecular-biology phenomenon repeated over and over again, with order-of-magnitude
increases in the rate of scientific understanding in almost every field.

REFERENCES AND NOTES

Anderson, H.L., and Allison, S.K., 1955, Reviews 0/ Modern Physics 27:273.
Bacon, F., 1960, The New Organon and Related Writings. Liberal Arts Press, New York. Especially pp. 98,
112,151,156,196.
Benzer, S., 1959, Proceedings o/the National Academy o/Sciences USA 45: 1607.
Chamberlin, T.e., 1897,5:837. Journal o/Geology I am indebted to Professors Preston Cloud and Bryce Craw-
ford, Jr., of the University of Minnesota for correspondence on this article and a classroom reprint of it.
Chew, G.F., Gell-Mann, M, and Rosenfeld, A.H., 1964, Scientific American 210 (Feb.):74, 210, (Apr.):60. 210,
(June): 54.
Davison, P.F., Freifelder, D .. and Holloway, B.W., 1964, Journal 0/ Molecular Biology 8: I.
Faraday, M., 1932-36, Faraday's Diary 1820-62. Bell, London.
Forscher, B.K., 1963, Science 142:339.
Garwin, R.L., Lederman, L.M., and Weinrich, M., 1957, Physical Review 105: 1415.
Lederberg, 1., 1959, Science 139: 1649.
Meselson, M., and Stahl, F., 1958, Proceedings o/the National Academy o/Sciences USA 44:671.
Platt, 1.R., 1962, The Excitement 0/ Science. Houghton Mifflin, Boston. See especially Chapters 7 and 8.
Polya, G., 1954, Mathematics and Plausible Reasoning, Vols. I and 2. Princeton University Press, Princeton, NJ.
Popper, K.R., 1959, The Logic o/Scientific Discovery, p. 41 Basic Books, New York. A modified view is given
by Kuhn, T.S., 1962, The Structure 0/ Scientific Revolutions, p. 146. University of Chicago Press, Chi-
cago. It does not, I believe, invalidate any of these conclusions.
Rich, A., 1959, in Biophysical Science: A Study Program, edited by J.L. Onc1ey et al.,Wi1ey, New York. p. 191.
Watson, E.C., 1945, American Journal 0/ Physics 13:281. Includes an English translation of both of Roent-
gen's first papers on X rays.
Watson, 1.0., and Crick, F.H.C., 1953, Nature 171 :737.
Weinberg, A.M., 1963, Minerva winter: 159; Physics Today 17:42.
Wheland, G,W" 1949, Advanced Organic Chemistry. Wiley, New York. See Chapter 4 for numerous such
examples.
 Chapter 16

'.~~.' The Limitations of

~

o i' .
Inference in Archaeology
q;,O_' .': .
.j'~l

M.A. SMITH

Most of the time of this Conference is being devoted to assessments by experts of dif-
ferent types, or aspects, of archaeological evidence. From their intimate knowledge of
the particular fields they are able to point out to us both the shortcomings and the
potentialities of the data that they study. An archaeologist, in interpreting his material,
takes into account the results of all these, and of other particular studies that are rele-
vant to his problem. What I want to do, alongside these specialist accounts, is to try to
consider, in general, what sort of inferences an archaeologist can expect to make when
he utilizes all this evidence to the full. We are here to discuss the reliability of archae-
ological evidence, but obviously before we can decide how far any type of evidence is
reliable we have to know what it is supposed to be reliable for; in other words, we have
to be clear what sort of problems archaeological evidence can legitimately be expected
to answer.
I shall not need, here, to give an account of the history of archaeological research
in this country. Our progress in fieldwork dates from the general adoption, in the pre-
sent generation, of standards that General Pitt-Rivers was already applying, as a pio-
neer, in his excavations at the end of the 19th century. Accurate and exhaustive
recording of sites raised archaeological excavation to the status of a science. Then, in
the 20th century, in the period between the two wars, there was the further assimilation
of archaeological to scientific practice, with the increasingly active cooperation of
Source: The Archaeological Newsletter (1955), 6:3--7. A paper read by Miss M.A. Smith at the Conference of
the Prehistoric Society, held at the University of London Institute of Archaeology in April, 1955. Reprinted by
permission.

167
 168 M.A. Smith

natural scientists themselves in archaeological problems. Scientists began to provide

the archaeologist with precise determinations, not only of his finds, but also of their
contexts. Quite rightly today we stress the value of this cooperation, which helps the
archaeologist to make use of every potentially significant fact about his material.
But, on the other hand, the very precision of scientific analyses, and the elegance
of a well-laid-out excavation, do tend to have a rather hypnotic effect on the mind, and
some people have thought that there was a danger that archaeology might come to be
equated simply with a competence in handling a set of techniques. The result has been
that when archaeologists in recent years have had occasion to stand back from their
detailed studies and review the subject as a whole, what they have tended to do is to
reassert its essentially human aspect against a too persuasive "scientific" approach. We
are reminded that the relics we study are the result of human activities. As archaeolo-
gists we classify these remains, and the recurrent combinations of material we term
"cultures," in a technical sense - the Maglemose culture, the Windmill Hill culture,
the Deverel-Rimbury culture, and so on. But, if I may quote from Professor Childe,
prehistoric artifacts, being made by men, have "their significant features ... determined,
not by mechanical processes, but by social traditions and purposes. Accordingly," he
concludes, "to be significant archaeological classification must embrace the cultures,
that is, the societies, whose aims and needs determine the artifacts' characters.,,1 Pro-
fessor Clark, in his inaugural lecture, declared: "As prehistorians we deal not with indi-
viduals ... but with societies, including their internal stratification and their local
organisation, and their relations to one another and to the world of nature .... Prehistory
is essentially a social study and as such it has obvious affinities to social anthropol-
ogy.,,2 Finally, an eloquent expression of the humanistic approach from the pen of Sir
Mortimer Wheeler: "the archaeological excavator is not digging up things, he is dig-
ging up people; however much he may analyze and desiccate his discoveries in the
laboratory, the ultimate appeal across the ages ... is from mind to intelligent mind, from
man to sentient man."3 "The only thing that really matters in our work," as he said on
another occasion, "is the re-creation of the past. .. archaeology and history are alike
frustrated unless they contribute to a vital reconstruction of man's past achievement, in
other words aspire to interpretation as well as to mere transliteration.,,4
We have then here assertions that an archaeologist should try to re-create the
past, or at least that in his cultures he is dealing with societies, and therefore claims to
apprehend societies that existed before the times of written record from their material
remains. If this is correct, prehistory must in some sense approximate to history. But
alongside this view of the nature of the subjects we have to keep in mind the caution
that Sir Mortimer Wheeler himself appended to the passage I have just quoted: "The
archaeologist may find the tub," he said, "but altogether miss Diogenes."s
It is admittedly often the case that the conclusions of an archaeological study can
be expressed in the form of a narrative about past events. The statement "Gold lunulae
were exported from Ireland to the Continent in the Early Bronze Age" may be taken as
an example. Now the essential character of this conclusion is that it translates back into
the observations that gave rise to it. We map the distribution of known lunulae, and
find a marked concentration in Ireland, a country rich in mineral gold. The evidence of
other metal types found in that country suggests that the Irish were in fact skilled in
metal work in this period, so that it is plausible that the lunulae were made there. Fur-
thermore, other distributions show that a connection with the Continent across England
The Limitations of Inference in Archaeology 169

or Scotland did exist. The conclusion that lunulae also were exported along this route
seems to follow these observations, and it is the sum of these relevant considerations
that we have in mind in asserting it.
We should, I think, all agree that in such instances the conclusions are legitimate
inferences from the archaeological material. All of the evidence used in them can be
empirically verified, and nothing has been added to it.
If, then, we allow that a certain sort of narrative about undocumented events does
emerge from archaeological studies, we can now go'on to consider how far this corre-
sponds to, or is different from, the re-creation of the past, which is said to be our proper
aim as archaeologists. It is often point out that the nonperishable material remains of
prehistoric peoples imply fairly clearly how their economic life was carried on, and
how it was a adjusted to fit the environment in which they lived. Certain aspects of
religious and social practices also are reflected in what remains for us to study. Among
the megalithic tombs of the Marne, in France, for example, some are found to contain
only a few skeletons, and rich grave-goods, while in others of a similar plan there are
many skeletons and only poor offerings. Professor Childe concluded from this that the
former "must have been reserved for chiefs," and that the others were "presumably
graves of commoners." Of course, it might be another distinction that is represented
here, say between priests and laity, but in its main lines this conclusion is acceptable,
because it is easy to see how it is related to its evidence. It is very similar in sort to the
conclusion about the Iris lunulae because, like that, it is in fact virtually only a para-
phrase of empirical observations.
Arguments to the social and religious practices of prehistoric peoples, when there
is no direct or indirect written evidence, however, are not always so simple. Let us con-
sider an example that Professor Hawkes uses to illustrate this: "If you excavate a settle-
ment in which one hut is bigger than all the others, is it a chiefs hut, so that you can
infer chieftainship, or is it really a medicine lodge, or a meeting hut for initiates, or a
temple?,,6 Obviously it will be important to know which of these interpretations is cor-
rect if in fact we are engaged in re-creating the past, or in studying societies "including
their internal stratification." as Professor Clark said. How should such problems be
tackled? Clearly, the conclusions of prehistory cannot even approximate to history
unless there is some way of deciding between the various possible interpretations in
Professor Hawkes's example. We do not get very far in re-creating the past, or in
understanding a prehistoric society, if we cannot be sure whether the largest hut in a
village is a temple or a storehouse.
R.G. Collingwood used to say that the way to interpret a piece of archaeological
evidence was to try to appreciate the problem of the man who was responsible for cre-
ating it. This method is all very well when one already knows, or can guess with rea-
sonable certainty, what that problem was likely to be: the explanation of multi vallate
hill-forts, for example, follows from our knowing that a defense was needed against
sling-stones. But even if we decide to allow that prehistoric peoples were as intelligent
as ourselves - if not more astute - when faced with problems that we know con-
fronted them, we must not make the mistake of assuming that they could have been
aiming only at things that seem justifiable to important, and that an interchange of vis-
its for mutual admiration was a constant feature of village life. If we imagine a per-
fectly conducted excavation of an extinct Trobriand group, supposing that conditions
were at their most favorable for the preservation of evidence, and that the potential sig-
170 M.A. Smith

nificance of the remaining material was fully appreciated, how far could we expect to
get towards re-creating Trobriand society? How far could the archaeological classifica-
tion of the remains incorporate the "aims and needs" of the society, which is what Pro-
fessor Childe requires if the classifications are to be significant? The size and layout of
the yam garden would confuse us, if we imagined that an increase in crop production
was the only aim of work undertaken. Calculations from the estimated yield of yams
would suggest a far larger population than in fact existed. We could not know that the
structures built in the gardens were intended only for temporary display of the har-
vested crop, or that the coral walls round the gardens had magical significance only.
And if yams from village A turned up in village B, how could it be guessed that the
transaction was not commercial? Even if it were suspected that a token offering was
represented, we should be more likely to think of it in terms of a tribute to political
authority, than the expression of a kinship tie.
The complexity that was Trobriand society, the sort of considerations that would
enter into a reconstruction of it which was in any way comparable with the study of the
anthropologist, simply would not follow from the observations an archaeologist would
be able to make. I have not chosen to quote the Trobriand society because the princi-
ples that govern it are especially aberrant from European values, so that it would be a
particularly difficult instance to infer to its life and institutions from the material
remains. That would be special pleading. I hope, rather, that it will illustrate and bring
home a general point.
I am thinking of the problem of Diogenes and his tub, which will have to be met
by anyone who maintains that archaeology aims to reconstruct man's past achievement,
and of the comparable problems involved in asserting that archaeological cultures rep-
resent societies, i.e., the essential social divisions of prehistoric peoples. It would obvi-
ously be impossible to understand the relics of the Trobrianders from us. The Western
European standards of value that condition our own thought are not absolute. It used to
be thought that studies of surviving primitive peoples would provide the necessary
analogies for interpreting prehistoric societies; but in the event the extension of ethno-
logical studies has only served to show what an incredible variety of codes of behavior
in fact actuate human conduct.
Malinowski was the first to undertake a thorough investigation of the life of a
primitive people, and his study of the Trobriand Islanders of New Guinea is the obvi-
0us one from which to quote. 7 He found that the growing of the staple yam crop was
the central activity of the community, but that it was an activity shot through with
magical, religious, and social implications, which cannot, in Trobriand society, be
thought of as features additional to the economic aspect of the work, which is never
considered there in abstraction. To take only one example: at harvest the crop is
divided according to the rule that a man must give his best yams to his married sister
and her family, often carrying them over long distances to where she lives in her hus-
band's village. Thus, the distribution of the staple crop, which we should think of, and
tend to interpret, in purely economic terms, is there entirely governed by the require-
ments of a complicated kinship system.
Moreover, to have any understanding of the central activity of Trobriand society
we must take into account what Malinowski called the "aesthetics of gardening." He
reported that considerable energy is spent on purely aesthetic effects. "The ground
before planting is cleared with a meticulousness far beyond what would be necessary
The Limitations of Inference in Archaeology 171

on technical grounds. The cleared soil is divided into neat rectangles about 4-10 metres
long and 2-5 metres wide by means of sticks laid on the ground. These rectangles have
little practical purpose, but much value is attached to their proportions and to the qual-
ity of the sticks which mark their boundaries."
Again, far more yams are grown than are needed for sustenance. During the har-
vest, work is twice interrupted while the yams are scr:tIpulously cleaned and arranged on
show, with temporary structures built over them. In short, in order to understand what the
people were doing when engaged in growing their subsistence crop, it would be essential
to realize that at all stages aesthetic considerations were the evidence of the material
remains alone: they don't even act rationally, on standards we can comprehend, when
engaged in growing their subsistence crop. By considering why even the closest possible
attention to the material remains reveals so little of what is significant, I think we can get
a clearer idea of the sort of questions we should expect archaeology to answer.
If, as in Professor Hawkes's example, you were to infer from the single large hut
to the institution of chieftainship, this conclusion would not be comparable with the
one about the export of Irish lunulae, which we examined earlier. This conclusion will
not translate back exactly into empirically verifiable statements that are supposed to
imply it; it is not merely a paraphrase of a set of observations. It contains an element of
conjecture, which cannot be tested. If you decide to call the large hut a chief's hut, and
not a meeting house, or a temple, this is an assertion, not an argument. You can't really
say that you know that it is, and if someone criticises your assertion, it is impossible to
produce sufficient evidence to convince him you are necessarily right. This is not the
result of any fortuitous incompleteness in the archaeological record: the position
couldn't be improved by better excavation, by finding a more favorable site, or by the
invention of a new technique for analysis.
It arises from the sort of relation that holds between the human activities we
should like to know about, and the visible results that survive from them; between these
two things there is logically no necessary link, so that it is a hopeless task to try to get
from what remains to the activities by argument. This is what I mean by the limitation
of inference in archaeology. A code of behavior, or the idea of chieftainship that pre-
vailed in a particular tribe, cannot be expressed adequately by things like the layout of
yam gardens, or large huts, so that, conversely, the garden and hut plans, from which
the archaeologist has to work, cannot imply them. It seems to me important to realize
that trying to establish the existence of, say, chieftainship in the prehistoric past is
something quite different from reaching the conclusion that Irish lunulae were exported
in the Bronze Age. It is not related to its evidence in the same way, and can never attain
to be same status in knowledge. You can infer, i.e., proceed by argument, from archae-
ological remains to some aspects of prehistoric economics, and reasoning will take you
to a limited range of social practices, like the distinction of rite perceptible in the
Marne tombs. But to expect an archaeologist to infer from a hut to cheiftainship, or
from the tub to Diogenes, is nothing less than a demand for logical alchemy.
This means that the accounts of the past we can offer in prehistory are necessarily
different in kind from the accounts of history, where at least something of what was
truly significant in the period is recorded. There is, of course, no distinction in the sub-
ject matter: human experience runs back unchanged from historic to prehistoric times.
The difference arises from the nature of archaeological evidence, that is, from the
means we have of coming to know about this human experience and activity in undocu-
 172 MA Smith

men ted phases of the past. This in tum requires some caution in assertion about how far
archaeological interpretation can take us in knowing about the past, and what the
proper aim of prehistory should be. It is often said that archaeological cultures repre-
sent prehistoric societies, and by way of example I propose to examine this one asser-
tion about what is revealed about the past by archaeology. Now the minimum definition
of a society in anthropology or sociology is a group of people acknowledging a single
political authority, obedient to a single system of law, and in some degree organized to
resist attack from other such societies. How is an archaeologist to be able to judge
when these conditions hold? Are we to think that the Windmill Hill culture was a soci-
ety in this sense? Was there one system of law and a single political authority through-
out, or perhaps a single legal system common to all users of plain Western Neolithic
pottery, and a separate political authority for each of the variant styles of ceramic deco-
ration? A Hembury, an Abingdon, a Whitehawk, and an East Anglian authority? These
questions are soon seen to be meaningless, and indeed how could it be otherwise? Rec-
ognition of political authority and of systems of law, like Diogenes, do not find expres-
sion in material things, and so evade the prehistorian. If we ask ourselves what
archaeological material we should accept as an indisputable indication that a single
political authority was in force, it becomes obvious that, when we are working from
material remains alone, there could never be the sort of evidence necessary to substan-
tiate a claim that the Windmill Hill culture represents one, or alternatively several, of
the communal groups that an anthropologist or sociologist calls a society. We may
wonder whether it does, but we can never know. Similarly, we find ourselves speculat-
ing about the relation of the succession of cultures we establish in archaeology to the
historical realities of the period. But once again, we can never know whether the events
that were really decisive in the history of the times we are studying got reflected in
material equipment at all, or in activities that left material results. It looks very much as
though questions like these, how far archaeological cultures correspond to actual socie-
ties, or what is the relation of arachaeologically recorded events to historical realities,
are not only meaningless but irrelevant to the archaeologist, because there is no means
for him even to begin to answer them.
I began this evening by recalling how much the natural sciences have contributed
to the progress of archaeology, By providing minute analyses they have recovered
much significant information, which in the days of haphazard excavation and less criti-
cal studies was destroyed or overlooked. We may say that science has taught the
archaeologist a due respect for his material and for his sites.
It has to be acknowledged that there is no logical relation between human activity
in some of its aspects and the evidence left for the archaeologist. Accordingly, there are
real and insuperable limits to what can legitimately be inferred from archaeological
material. Some time ago, a competition in the periodical The Spectator asked for defi-
nitions of "prehistory." One of the suggestions commended by the adjudicator was "the
cradle of mankind rocked by the mother of invention." Now if as archaeologists we
purport to be trying to re-create the past, or to be able to apprehend prehistoric socie-
ties, this type of comment becomes embarrassingly pertinent. I want therefore to end by
suggesting that, having been taught by science a due respect for the material we study,
we might equally learn from its methods a respect for our subject itself.
I am not suggesting that the archaeologist has anything useful to learn from the
intricate conceptual apparatus in which current scientific research is conducted. But it
The Limitations of Inference in Archaeology 173

is well worthwhile to turn back to the 18th century, to the discussions leading to the
establishment of the method that has been basic to the scientific progress of the last two
centuries.
Bishop Berkeley is the classic commentator on the unsuccessful efforts of earlier
science to comprehend the essence of material things in the way of Greek philosophy;
in posing themselves this problem, he said, natural philosophers were first "raising a
dust, and then complaining that they could not see." Now in many respects the problem
that Berkeley was dealing with here is, methodologically, analogous to the one of
deciding what we can expect archaeological evidence to be reliable for. Of course we
are quite conscious that there must have been rich historical realities behind the mate-
rial relics we study, whereas Berkeley denied that material substance existed at all. But
nonetheless his recommendations to his contemporary scientists are directly relevant to
us. He urged that, since it was proving impossible to come to any knowledge of the
essence of material things and real causes from the data available, this could only mean
that it was wrong for science to aim at finding out about them. Scientists should content
themselves with arranging the information that they could come by, by working corre-
lations between events, and so on; by establishing "observable regularities" as he put it.
Unobtainable ends cannot be the proper ends for any subject. What any subject aims to
do must be appropriate to the means at its disposal.
The subsequent acceptance of Berkeley's teaching was to provide the philosophi-
cal basis for the natural sciences in the progress they have since made. If we apply his
axiom to archaeology, it follows that, since historical events and the essential social
divisions of prehistoric peoples don't find an adequate expression in material remains, it
cannot be right to try to arrive at a knowledge of them in archaeological interpretation. A
recognition that archaeological evidence, when it is confined to material remains,
demonstrably supports only a limited range of conclusions about human activity, is in-
compatible with too ambitious a program for archaeology. It is incompatible, as I see it,
with an attempt to "re-create the past" in any real sense, or with a claim to recognize pre-
historic societies from their surviving relics, so that the subject could be compared either
to history or to social anthropology. Moreover, if it appears that archaeologists do make
such a claim, or are engaged in such an attempt, it may seem to an outsider, conscious of
the weak logic this involves, that the subject has no sound intellectual basis at all.
I shall therefore end by suggesting that a respect for the subject requires that we
should consciously adopt Berkeley'd method of defining the aim of our interpretations
by strict reference to the potentialities of the evidence. It is only the problems to which
the evidence can provide answers which can legitimately be asked. The fact that we
know that human societies and their histories were responsible for the remains we
study tends to obscure this, but it should not affect a realization that attempts to estab-
lish prehistoric societies, and events, in the historical sense, must rest on conjecture,
not on argument. I should prefer not to say that archaeology "re-creates," or "recon-
structs" at all; it merely recovers what it can. That of itself is a sufficient program of
research. And we should not be tempted to clothe the results of this research with more
significance than they can legitimately carry. There is no point in asserting that an
archaeological culture that we have established represents a prehistoric society if we
can never know that it does.
When Bishop Berkeley repudiated the search for material substance in natural
science, in favor of the "observable regularities" he could validly establish, he said he
174 M.A. Smith

determined to know "what I know, and not some other thing." And, in face of any
exaggerated claims about what archaeology can tell us about the past, I think this
would be a good motto for the archaeologist.

REFERENCES
I. Archaeology as a Social Science. Inaugural Lecture. 3rd Annual Report. 1946. London University
Institute ofArchaeology, p. 58, 1947.
2. The Study of Prehistory. An Inaugural Lecture, p. 13. 1954.
3. Archaeology from the Earth, pp. 2-3, 1954.
4. What Matters in Archaeology. Antiquity, XXIV, Sept. 1950, pp. 128 and 130.
5. Ibid, p. 130.
6. American Anthropologist, Vol. 56. No.2, Pt. I, April 1954, pp. 161-162.
7. Coral Gardens and Their Magic, 1935.
 Chapter 17

Archaeological Methods

FREDT PLOG

The goals, techniques of inquiry, and data that characterize a discipline should be deter-
mined by conscious and explicit decisions. There must be harmony among the three.
Moreover, decisions about goals, techniques of inquiry, and appropriate data are inter-
nal to a discipline.
Decisions about method are only partially internal to a discipline. Rules of evi-
dence and standards of verification are good or bad not for a particular discipline, but
for science as a whole. This point is made clear by Rudner (1966:5) in his Philosophy
of Social Science:

To become aware that various scientific disciplines employ differing techniques of

investigation is not to become aware of anything significant about the nature of social
science .. .. To claim that there is a difference in the methodology between two disci-
plines or two types of disciplines is, by contrast, to make a radical claim. For the
methodology of a scientific discipline is not a matter of transient techniques but of its
logic of justification. The method of a science is, indeed, the rationale on which it
bases its acceptance or rejection of hypotheses or theories. Accordingly, to hold that
the social sciences are methodologically distinct from the nonsocial sciences is to
hold not merely (or perhaps not at all) the banal view that the social sciences employ
different techniques of inquiry, but rather the startling view of the social sciences
require a different logic of inquiry. To hold such a view moreover, is to deny that all
of science is characterized by a common logic of justification in its acceptance or
rejection of hypotheses or theories.

Source: The Study of Prehistoric Change. edited by Fred T. PJog, pp. 12-25. New York: Academic Press, 1974.
Copyright 1974 by Academic Press and reprinted by permission.

175
 176 F.T. Plog

Archaeologists have spent as much time trying to justify remaining outside an

ongoing discussion of scientific method as attempting to see whether or not our disci-
pline could live by the same standard employed in others. How else can one explain the
fact that doctrinal Baconian inductivism finds its most vigorous, if not its only modern
proponents among archaeologists? How else can one explain the major role that anal-
ogy, a form of argument universally labeled as the weakest, still plays in archaeology?
By discussing the nature of explanation and the acquisition and use of laws, I intend to
show how archaeology can carry on research within the established boundaries of good
scientific method.

THE NATURE OF EXPLANATION

The subject of archaeology is behavioral and cultural variability - the similari-
ties and differences in behavior at specific temporal and spatial loci. Its object is to
explain these similarities and differences. Explanation is a term that is used frequently
by anthropologists in discussing research objectives. But, one rarely finds a consistent
meaning in such usages. Moreover, this problem is not unique to anthropology.
Our disagreement over the meaning of explanation is a reflection of similar dis-
agreement among epistemologists and philosophers of science. Viewed from different
philosophical perspectives, explanation can have very different meanings, Yet, all defi-
nitions of explanation are attempts to provide models for knowing and knowing that
one knows, Much of the variation in definitions of explanation reflects the varying
emphases that different authors place on three components of the knowledge process:
(1) the problem of constructing valid arguments, (2) the problem of constructing argu-
ments that account for observed variability, and (3) the problem of ensuring that the
observed variability in both topic and causal variables is indeed observed and not a
product of bias in techniques of data collection of analysis. (I have not included discov-
ery in this list. While this topic is critically important, I know of no really successful at-
tempts to explain or even define this phenomenon.) Within the context of these varying
emphases, it is possible to define a number of diametrically opposed schools: deduc-
tivists versus inductivists; formalists versus substantivists; argumentivists versus corre-
lationists; and so on. It will be argued that explanation that fails to place equal
emphasis on all three components of explanation, explanation that fails to include
deductive and inductive arguments, formal and correlational techniques, is incomplete.

THEFOruMALCOMPONEN~
CONSTRUCTING VALID ARGUMENTS
Philosophers concerned with the formal component of explanation are defining
the classes of statements of which an argument is composed and analyzing the logical
relationships that must exist between these classes of statements in order that an argu-
ment be deemed valid.
The model of explanation in this formal context that I have found most useful is
the Hempel-Oppenheim, or deductive--nomological, model. In this model, an explana-
tion is an argument that fits a phenomenon "into a pattern of uniformities and shows
that its occurrence was to be expected given specified laws and pertinent particular cir-
cumstances" (Hempel, 1966:50). Three classes of statements are used in constructing
Archaeological Methods 177

the arguments: (1) statements about the phenomenon under investigation, (2) state-
ments about pertinent particular circumstances, and (3) statements of laws. The last two
types of statements are called the explanans, " ... the class of those sentences that are
adduced to account for the phenomenon" (1966:51). The explanandum is a sentence
describing the phenomenon under investigation.
An explanation in this model looks like this:
LI' L 2, Ln General Laws
CI' C 2, C n Statements of Antecedent Conditions
Logical Deduction
Description of the Empirical Phenomenon To Be Explained
Having built such a skeleton, one turns to linkages, relationships that must exist
between the propositions if they are to be considered valid. Philosophers have specified
that the propositions must share certain terms. Moreover, the order of the terms in the
proposition is critical- denying the first term (antecedent) or affirming the second
(consequent) produces invalid arguments.
A primary question that arises from this model of explanation concerns the term
law. It seems pointless to engage in the old debate over the existence of anthropologi-
cal (behavioral and cultural) laws - behavioral scientists use such laws explicity and
implicitly. A more important question concerns the sources of such laws, or at least
statements that are worth testing because they might be laws. Miller's (1965) "Living
Systems: Cross-Level Hypotheses" and Human Behavior: An Inventory of Scientific
Finding by Berelson and Steiner (1964) are excellent sources of such propositions. The
ultimate source is of course the mind of the investigator attempting to create the knowl-
edge. Hanson (1965) has written a provocative discussion of the discovery process.

THE SUBSTANTIVE COMPONENT:

ACCOUNTING FOR OBSERVED VARIATION

Explanation in substantive terms means decomposing and accounting for variabil-

ity in some phenomenon under investigation. Anthropologists deal with variations in be-
havior and cultural processes at different spatial loci at a single point in time, variations
in culture and behavior at different points in time at a single spatial locus, and variations
in culture and behavior at different points (usually parallel) at multiple spatial loci. In
dealing with spatial phenomena, it is argued that the phenomenon under study has some
patterned spatial distribution. In dealing with temporal phenomena, the variable forms a
trace, or trajectory, over time. Explaining variability means developing the ability to pre-
dict or retrodict the spatial pattern or the shape of the trace, whichever is relevant.
The process of substantive explanation involves placing a phenomenon in a system
of antecedent and intervening phenomena, placing a variable in a system of antecedent or
intervening variables. These act and interact to cause the topic phenomenon to vary as it
does. One might, for example, be interested in explaining changes in the popUlation of a
community over time. The topic variable might be the actual population at some series of
points in time, or increments of popUlation per unit of time, or any of the various meas-
ures of rates of population change. Having chosen on~ or a number of measures as topic
variables, a list of phenomena in the social and natural environments of the community
that are viewed as potential determinants of population change is compiled and measure-
178 F.T. Plog

ment instruments are developed for each of them. Holding all other variables constant,
the relationship between each antecedent or intervening variable and the topic variable is
considered. So are interaction effects. Statistics such as the partial correlation coefficient
are used to tell the investigation in a reasonably precise fashion the percentage of the
variability accounted for by a particular antecedent variable. The square of the multi-
ple correlation coefficient indicates the percentage of variation in a topic variable that
the entire system of variables under investigation accounts for. It is rarely possible to
account for 100% of the variability in behavioral phenomena, but correlational tech-
niques provide an investigator with a measure of his success in reaching this goal.
Simulation models provide an even more significant solution to the problem. The
simplest of these is the linear regression model, in which the equation y = a + bx is
used to define the relationship between a topic and antecedent variable. Multiple
regression techniques allow for the analysis of a number of variables and permit the
investigator to control for intervening variables and interaction effects. By specifying a
series of variables and defining the relationship between them, in a statistical or formal
simulation model, the investigator should be able to generate a trace for population
change that very closely resembles the one that was observed.

THE OPERATIONAL COMPONENT:

FINDING VARIATION AND KEEPING IT OBSERVED

The operational component of explanation goes further in coming to grips with

the whole activity or process of explaining, of doing research. Therefore, somewhat
more attention will be devoted to it. Operational explanation is testing particular argu-
ments or models. Operational definitions of explanation specify the classes of activities
that constitute research and the way in which these activities should be ordered if an
explanation is the desired product of the research. On the other hand, they attempt to
ensure that the variability with which an investigator works is in fact observable and
not a product of bias in the techniques ·of observation or analysis that he uses. The
operational definition of explanation that I have found most useful specifies the follow-
ing activities: formulation of a problem; formulation of hypotheses; operationalization
of the hypotheses; collection of data; analysis; testing; and evaluation.
Such a research design must be followed in the order presented. Some people
argue one can begin at any point in the list and still do valid research as long as each of
the steps is ultimately carried out. This viewpoint is probably not useful because the
single most important factor in formulating a research strategy and in acquiring and
analyzing data is a hypothesis to which both are adapted; we do not collect or analyze
data in a vacuum. We collect those data we see as relevant to the solution of a particu-
lar problem and analyze those aspects of data we see as relevant to the solution of a
particular problem. Hempel (1966) makes this point very clear:

What particular sorts of data it is reasonable to collect is not determined by the prob-
lem under study, but by a tentative answer to it that the investigator entertains in the
form of a conjecture or hypothesis. (p. 12)

If a particular way of analyzing and classifying empirical findings is to lead to an

explanation of the phenomena concerned, then it must be based on hypotheses about
Archaeological Methods 179

how those phenomena are connected; without such hypotheses, analysis and classifi-
cation are blind. (p. 13)

It is sometimes argued that one need not worry about the absence of explicit
hypotheses because all research has implicit ones. Kluckholn (1939) asserted many
years ago that no one collects data in the absence of some theoretical orientation that
causes him to collect some data and ignore others. But, a fundamental identity usually
exists between good research and research that has been done according to a well-
planned design. When part of the research process is implicit, explicit consideration of
the viability of the research design is unlikely.
Having made clear that the order of steps in the research is not arbitrary, I now
want to discuss it in detail.

Problem Formulation
At this first stage in the research process, the problem to be investigated is formu-
lated in such a way that data and/or solutions to the problem are suggested. That is,
statements like "I am interested in culture change" or "I am interested in Southwestern
prehistory" are replaced by at least relatively precise questions: "Why did horticultur-
ists living on the Colorado Plateau at about A.D. 700 begin to adopt water control tech-
niques?" Formulating the problem should always involve an extensive review of the
literature pertaining to the problem. It is often the case, and has too often been the case
in the past, that the problem centered on a set of data: a site that an archaeologist had
been told to dig, a region he has been asked to survey. I assume that archaeology is
beyond the point where problems focused primarily on data collection are likely to be
productive. None of anthropology's subdisciplines are lacking in either available data
or research situations. To select a research site and to claim to be interested in "all of
the variability that can be observed there" is naive and results in the rambling mono-
graphs that have been the bane of the discipline. Data sets and research settings must be
selected with a problem in mind.

Formulation of Hypotheses
Hypotheses are statements of relationship between two or more variables. They usu-
ally take the form "ifx. theny," and may be regarded as potential- but unproven -laws.
At a minimum, they are potential solutions to the problem that has been formulated. The
acquisition of hypotheses may occur within the context of an ongoing attempt to solve a
particular problem or it may occur in the context of a fresh start on an essentially new
problem. In the first case, the hypotheses with which one will be dealing are likely to be
suggested conclusions of past research. In the second case, potential solutions to an essen-
tially new problem may be original abductions or may be found in the literature of archae-
ology, of anthropology, of the social sciences, or of general systems theory.
When the hypotheses on which a given piece of research will be based are the
suggested conclusions of past research, they are outgrowths of the sixth step in the
research design: reformulation and evaluation of research. In this sense, the research
design is part of an ongoing process. Nevertheless, it is still an integral unit within that
research process, a unit that cannot be arbitrarily divided.
180 F.T. Plog

When hypotheses are acquired in a situation of approaching a new problem, they

general1y come from three sources: abduction, the literature of social sciences, or gen-
eral systems theory.
Abduction is the perception that two or more variables may be related and that
this relationship may explain the occurrence of the phenomenon under examination. It
is the process of perceiving patterns in data (Hanson, 1965:86---S7; Hempel, 1966:15).
The quality of a given abduction is explained by the creative ability of an individual
scientist.
Abduction is not induction. Good abductions do not depend on the quantity of
data a given investigator has examined. As Hempel (1966: 15) points out,

The transition from data to theory requires creative imagination. Scientific hypothe-
ses are not derived from observed facts, but invented in order to account for them.
They constitute guesses at the connections that might obtain between the phenomena
under study, at uniformities and patterns that might underlie this occurrence.

Familiarity with data is an essential condition for carrying out any research, for
operationalizing any hypothesis. But, while familiarity is essential, the probable valid-
ity of a hypothesis cannot be evaluated on the basis of the amount of data an investiga-
tor treated in creating it. It is a function of his creative ability as a scientist, and no
more. One often suspects that those who wish to explain hypothesis creation as a linear
function of data examination are trying either to shift responsibility for hypotheses they
could not create or to protect themselves lest they be wrong. The courage to risk being
wrong is the essence of innovation. And, to claim that explanations are derived from
data rather than the minds of scientists is intel1ectual cowardice.
Hypotheses may be indicated conclusions of the research of other archaeologists,
anthropologists, or social scientists: That is, one may be using the abductions of others.
In trying to solve a particular problem, one approach is seeing what conclusions one's
coresearchers have reached about that problem or about related problems. An important
resource in this activity is Berelson and Steiner's (1964) Human Behavior: An Inven-
tory of Scientific Findings. This book presents 1000 hypotheses concerning problems
with which social scientists frequently deal and summarizes the research that has been
done in attempting to solve these problems. It treats no problem in sufficient detail to
be a final authority, but it suggests some testable hypotheses representing potential
solutions to specific problems and also provides references to research conducted on
these problems.

Operationalizing the Hypotheses

The kinds of operational procedures that an investigator will use will vary with
the problem and field setting. There are, however, important steps that must be taken in
operationalizing any set of arguments.
(1) Test implications of the arguments must be identified. The investigator should
specify particular states and conditions of, or patterns in, col1ectable data that he should
find if the arguments he is proposing are valid. Any worthwhile argument should have
observable implications, and it is to these implications that the investigator should
direct his attention. (2) The data that must be col1ected in the field to determine
Archaeological Methods 181

whether the predicted states, conditions, and patterns are present should be specified.
(3) Measurement devices for converting raw data to variables should be identified, and
their efficacy demonstrated. Raw field data are bundles of attributes some of which are
relevant to the investigator and some of which are not. Devices for observing and
measuring the specific attributes of behavior that are relevant to the test implications
must be specified. (4) A technique for ensuring that there is no bias, or at least that
there is a statable bias, in the data that investigator collects must be developed. The
investigator must know the relationship between the sample of behavior that he
observes and the behavioral universe. This concern need not imply random sampling. It
is my impression that for most anthropological problems, random sampling is useful
only within a factorial design, a design stratified by antecedent variables.
All social scientists, archaeologists no exception, make conclusions about human
behavior and cultural processes on the basis of samples. Few archaeologists who do sur-
vey work can legitimately claim to have found all of the sites in the surveyed area and
few digs attempt the complete excavation of even a single site. Generalizations are made
about the area surveyed on the basis of the sample and about the site on the basis of a
partial excavation. Yet, these generalizations are oft~n made without adequate concern
for the nature of the sample taken. Archaeologists have begun to give more attention to
sampling (see Heizer and Graham, 1967; Hill, 1965). An excellent discussion of various
techniques of sampling appears in Loeational Analysis in Human Geography by Haggett
(1966). I do not want to describe here the various kinds of mathematical sampling tech-
niques as these have been discussed in the volumes cited above. However, I do intend to
summarize why such techniques of sampling should be used in testing. Recall that by
this point in the research design hypotheses have already been generated.
When an investigator is not concerned with sampling, there is no guarantee his
results are based on reliable data. If he selects units he believes are typical, "typical" is
subjectively defined when it should be defined on the basis of objective mathematics. if
the universe is not defined, if the sample size is unknown, and if sites are not selected
in a rigorous manner, the conclusions may warrant little confidence. If sites are
selected because they are easily accessible or because they have some special charac-
teristic, or if a survey is limited to a small portion of the area about which generaliza-
tions will be made, the presumption must be that the results are biased. If, on other
hand, the survey is designed so as to ensure that the total range of areas used by a cul-
ture will be investigated and the study area will be regularly sampled, the chances of
bias are minimized. More important, when techniques of mathematical sampling are
employed, the probability that the conclusions might be based on sampling error exists,
but bias can be stated.
Second, sampling saves time. When the universe is defined, the proportion of it
that must be sampled in order to collect enough cases to support statistically valid con-
clusions can be calculated. When the universe is not known, the investigator has no
basis for limiting his data collection and may waste time collecting and processing
unneeded data.
Finally, when an investigator is not concerned with sampling, he may limit the
ability of his discipline to improve itself. One way in which any discipline improves its
understanding of the phenomena it studies is by using new techniques and new theories
on old problems. When unneeded data are collected, data are taken away from future
investigators who might be interested in reworking a problem. By sampling, and
 182 F.T. Plog

thereby leaving some information in the ground, an investigator ensures the availability
of data in its original context for use by future colleagues.

Acquisition of Data
Acquisition is the mechanical process of deriving from the field the data that will
be used to test a hypothesis. What data are to be collected and how they are to be col-
lected have already been specified. The specific techniques of data collection will vary
with the problem under investigation.
There has been a point of view in archaeology suggesting that absolute standards
are appropriate to all situations of data collection. The inaccuracy of this view is indi-
cated by the following comments on what have been archaeological bibles: Wheeler's
(1956) Archaeology from the Earth and Heizer and Graham's (1967) A Guide to
Archaeological Field Methods. Chang (1967: 129-130) comments on Wheeler:

If a piece of archaeological work is "aimed at recovering information bearing on

man's cultural antecedents (Ackerman and Carpenter, 1963:13)," it should then be
evaluated on that basis rather than on the neat or sloppy appearance of a site during
the excavations. I do not share in Mortimer Wheeler's advice that "on approaching an
excavation, the trained observer can at a glance evaluate its efficiency. It is an axiom
that an untidy excavation is a bad one" (1956:80). Neatness is preferable to untidi-
ness, not a priori on aesthetic grounds but only if it enhances the effectiveness of the
information-retrieval processes. Things have an order of significance, and a tidy
excavation is not necessarily a good one.

In a recent review of Heizer and Graham's updated version of A Guide to Archae-

ological Field Methods, Binford (1968:807) argued

It approaches archaeological field work from the perspective of excavation procedure

with little attention to the problem of scientific data collection as such. It is fre-
quently asserted that certain types of observations are "important" yet there is little
discussion of why they might be important or what relevance various types of infor-
mation may have in the investigation of the past.

Excavating beautifully for the sake of beautiful excavations is no longer a justifi-

able goal. In the absence of a complete work that approacbes the task of excavation
from the point of view of valid techniques of data collecting, it is probably safest to
take the position that we should employ techniques of data collection most appropriate
to the particular sorts of data we are trying to collect. This does not mean data in which
we are not interested should be ignored or discarded; we have an obligation to save the
data we would destroy or [whose] context we would destroy. However, the Archives of
Archaeology and similar repositories offer a mechanism for saving unused data without
having to study it carefully and to analyze it in the same manner as relevant data.

Analysis of Data

Analysis is the process whereby data from the field are put into the form in which
they will be used in testing hypotheses. The analysis undertaken will relate to the defi-
Archaeological Methods 183

nition of variables provided in the first step of the research design. Analysis may
involve counting or factor analysis. One might need to do no more than count the
number of sites with kivas. On the other hand, it might be necessary to make and record
a score of measurements for each of several thousand stone tools. There is no set list of
acts one performs in analysis - what occurs is determined by the problem under inves-
tigation.

Testing of Hypotheses
Testing means studying the data to determine whether the predictions as to how
data should look if the hypothesis were valid have been accurate or inaccurate. If the
predictions are accurate, then the test of the hypothesis is considered positive. If the
predictions are inaccurate, the hypothesis is considered to have been negatively tested.
Testing will frequently, but not necessarily, involve the use of statistical techniques to
measure the degree of association between variables.
Testing makes the difference between a hypothesis and a law. A law is a posi-
tively tested hypothesis, One negative test is enough to suggest that a particular
hypothesis is probably not a law. One positive test, however, is not enough to make a
hypothesis a law. A given piece of research is likely to involve one thorough test of a
hypothesis. Therefore, more than one piece of research is necessary before a hypothesis
can be considered a law.
This fact should not lead to the conclusion that the more times a particular
hypothesis is tested, the greater the probability of its validity; the validity of an expla-
nation is not a function of the quantity of data used in testing it. In the first place, such
a view fails to take into account the law of large numbers - the increment to validity
from each new piece of positive data is less than the preceding bit of data until the
increment nears zero. Furthermore, this viewpoint fails to recognize the importance of
variety and diversity.
More important than trying to add to the number of specific cases supporting it
given hypothesis is deducing and testing predictions about different kinds of data that
bear on the hypothesis. The independence of the tests for a given hypothesis is more
important than their number. To test the proposition "if x, then y," enumerating 15
cases in which x and y are associated in the predicted manner would be a form of test-
ing. A better form of testing would be deducing from the position "if x, then y" the test
implications that if the proposition is valid then the following equalities should hold: a
=b, c = d, e =f, g = h, and i =j, where a, ... j are independent classes of data shown to
be related to the proposition "if x, then y" in multiple, independent cases. Testing on in-
dependent evidence does more to establish validity than testing that involves the
repetitive use of a single set of evidence.

Evaluation of the Research

This step in the research design involves a recapitulation of the completed
research. It should include a discussion of the outcome of the tests and of the weak-
nesses of the tests.
The probability that the hypotheses are laws is discussed. In this sense the research
as a case of law discovery is examined. The extent to which the hypotheses offered
 184 F.T. Plog

insight into the phenomenon is also examined, and in this sense the research as a case of
using a statement oflaw is reconsidered. Suggestions for future research on this problem,
as well as suggestions of new problems the research raises should be considered.

SUMMARY
The research design I have discussed consists of the following steps:
1. Formulation of the problem
2. Formulation of hypotheses
3. Operationalization of hypotheses
4. Acquisition of data
5. Analysis of data
6. Testing of hypotheses
7. Evaluation of research
The design involves the use of both deduction and induction. To test a hypothe-
sis, test implications are deduced from it. Induction comes into play when the test
implications are compared with specific sets of data to determine whether or not the
deductions stand the test of data.
The research design sets out to establish and/or use laws by performing experi-
ments with potential laws. The essence of the process is the experiment - making a
prediction about how specific classes of data should look if an explanation is valid, and
then collecting the data to see if the prediction was accurate. One often hears the argu-
ment that experiments in the social sciences do not have the conclusiveness of experi-
ments in the natural sciences because social science experiments lack replicability. I
think this view is wrong and will attempt to demonstrate its lack of validity by perform-
ing such an experiment.
I and a number of other investigators have argued that it is important to follow
these steps in the order listed. Research is a process characterized by feedback - the
reformulation of arguments and models while research is under way, the redefinition of
variables, and so forth. Especially when an investigator works on a single problem for
many years, this retooling will be continual. However, at any stage in the research
process, it is desirable to stick with a basic set of procedures. A precise record of refor-
mulations and modifications of a research design is one of the investigator's most valu-
able tools in seeking and finding new directions.
Research must be formally, substantively, and operationally complete if it is to
result in explanations of observed variation. It is possible to formulate elegant argu-
ments that are substantively meaningless. It is possible to operationalize trivial
hypotheses. It is possible to account for 98 percent of the variation in some phenome-
non without understanding its behavior a whit. It is impossible to explain rigorously
and to acquire understanding without being concerned with all three components of
explanation.

REFERENCES
Berelson, 8., and G.A. Steiner, 1964 Human Behavior: An Inventory o/Scientific Findings. New York: Har-
court Brace and World.
Archaeological Methods 185

Binford, L.R., 1968, Review of a guide to field methods in archaeology. Robert F. Heizer and John A. Graham.
American Anthropologist 70:806--808.
Chang, K.C., 1967, Rethinking Archaeology. New York: Random House.
Haggett, P., 1966, Locational Analysis In Human Geography. New York: St. Martin's Press.
Hanson, N., 1965, Patterns of Discovery. Cambridge: Cambridge University Press.
Heizer, R.F., and lA. Graham, 1967, A Guide To Field Methods in Archaeology. Palo Alto, CA: The National
Press.
Hempel, C.G., 1966, Philosophy of Natural Science. Englewood Cliffs, NJ: Prentice-Hall.
Hill, J.K., 1965, Broken K.: A Prehistoric Society In Eastern Arizona. Unpublished Ph.D. dissertation. Univer-
sity of Chicago.
Kluckholn, C., 1939, The place of theory in anthropological studies. Philosophy of Science 6:328-344.
Miller, J.G., 1965, Living systems: Cross-level hypotheses. Behavioral Science 10:380-411.
Plog, F.T., 1974, The Study ofPrehistoric Change. NY: Academic Press.
Rudner, R.S., 1966, Philosophy ofSocial Science. Englewood Cliffs, NJ: Prentice-Hall.
Wheeler, M., 1956, Archaeology from the Earth. Baltimore, MD: Pelican.
 Chapter 18

The Berth 52 Vessel

(09CH691)
The Interim Report (1982)

L. E. BABITS

INTRODUCTION

Hutchinson Island forms the northern bank of the main shipping channel of the Savan-
nah River opposite the Georgia Port Authority terminal at Garden City. In actuality, this
modem island is the result of the merging of King Island with Fig Island by the deposi-
tion of dredged spoil from the river. This junction apparently occurred after 1934 when
the Frazer station was placed on the riverbank at a point some 100 m downstream from
the site (Barney Putziger, personal communication, November 3, 1982).
The wreck currently lies buried in some 3 to 5 feet of gray clay overlaid by
eroded spoil. It rests at an angle to the river with its bow well up on the shore and the
stern exposed to erosive action by the current and tidal flow. Some additional erosion
also occurs as a result of the commercial traffic on the channel. Despite the effects of
the river, the wreck is in a very good state of preservation.
The wreck was first noted in the fall of 1981 by a diver from the U.S. Army
Corps of Engineers. In the spring of 1982, a Corps survey during low water officially
recorded the vessel as a derelict with some archaeological potential. At this time it was

Source: Ms. on file, Georgia Historical Society, Savannah, Georgia.

187
 188 L.E. Babits

photographed in color and black and white (R. Anuskiewicz, personal communication,
May 1982; 1. Wood, personal communication, June 1982). The Corps photographs
reveal the stem portions of a relatively intact small vessel that is almost totally covered
by mud. A few frames on the starboard side were exposed, allowing some determina-
tion of its construction. As part of a field methods course in historical archaeology at
Armstrong State College, students were given the opportunity to work on the wreck in
an effort to assess its importance and integrity. Prior to working on the wreck, permis-
sion was obtained from the U.S. Army Corps of Engineers, the State of Georgia,
Chatham County, and Union Camp. These permits were necessary in order to comply
with federal, state, and local regulations as well as to provide a means of access
through the privately owned property of Union Camp.

METHODOLOGY

The condition of the wreck suggested the vessel itself could serve as a guide to
excavations.
Accordingly, the downstream side of the keelson became the baseline from which
the site was oriented for measuring. Measuring consisted of recording exposed struc-
tural features in relation to the vessel when the water was low enough to permit this.
Timbers were measured as to width, thickness, and length as well as their relationship
to the keelson and keel.
As water receded on the site, excavation proceeded with the removal of the clay
overburden. This proved to be a difficult task as the gluelike consistency of the clay
made standing and shoveling very difficult. Nevertheless, a logical method of excavation
was arrived at. Excavation was by shovel down to the extant timbers which were then
cleaned with trowels. Excavators sliced through the clay overburden and threw it to
either side of the wreck. Eventually, as work concentrated on the starboard (downstream)
side, the clay was thrown into the river alongside the outer edge of the vessel to stabilize
the planking. Before work was terminated, some 27 ft of keelson were exposed as well as
portions of the starboard side. Some of the port side was also uncovered (Fig. I).
Elevation mapping was accomplished by using the top of the keelson's deadwood
as a datum and measuring from this point to the timbers with line and line level to
obtain an idea of how high the frames extended at the present time. A total of 24
frames and floor timbers were noted on the starboard side and lion the port side.
Strakes were also plotted in on both sides of the vessel as well as the internal planking
on the starboard side. Loose ribs and other timbers found in the clay matrix were
mapped in relation to the stem. On the last day of excavation, the site was secured by
driving fence staples into the timbers and connecting them with nylon line. This was
done in order to prevent structural elements from washing away. Erosion of the spoil
was rapid on the site and this indicated that the side would be buried within a short
time which would also act to secure structural elements as well as protect the site from
drying out or other damage. Subsequent examination of the site has confirmed this as
the vessel is now buried except for the sternmost sections.
Excavation continued for 2 weeks as tide permitted. Since the river is a "black
water" environment (Garrison et aI., 1980: 36), with a strong current, and the students
were neither trained in underwater archaeology nor certified as divers, excavation was
limited to those periods when the tide was low. Time on site was further restricted by
 lI' Metal Nail
_f • Treenail
o Charred Area

I
~"j
~JII."lltJ,\"'I"!I"'I!""!'!"'\"I""'III"",,
U
_....
• I I I I I I I
•
-Q
': . I
. I:
, , :,-....
I. t I r I : -r J I -;:
I
"-
\.,.
I : I !: • : . ~ : : : : ! I
I ... :
..... "'1,';1:-;--.• '

II IJ-;;rr~ o Scale
.....---.
1 Foot
Figure I. Plan view of Berth 52 site (09CH69 I).
 190 L.E. Babits

the requirement that work not interfere with the regular operations of Union Camp.
This second requirement meant that work had to terminate with sufficient time to allow
passage through a gate which was locked at 3:30 P.M.
A total of some 120 student hours was spent on the site. Most of this time was
spent in removing overburden and mapping exposed timbers. A great many photo-
graphs were taken of this stage of work and a videotape was also made showing the dif-
ficulty of working in the clay overburden. After the field course ended, work continued
with the principal investigator working alone to identify internal features which might
yield additional information about the wreck. This additional work also consisted
largely of removing overburden and mapping along the keelson as well as investigating
specific portions again as questions were raised in completing the site map.
Two additional visits have been made to the site. One was made in late Septem-
ber with the State Archaeologist, Louis Larson. This visit was intended as a site inspec-
tion and familiarization. At this time, a determination was made by Dr. Larson that the
site was eligible for inclusion in the National Register of Historic Places. A second
visit was made to the site with Barney Putziger on November 3, 1982 as part of a
recovery attempt related to the operations of the U.S. Coast and Geodetic Survey. At
this time a brief inspection of the site was made.
The following report relates to the activities on the site, the findings and interpre-
tations of excavation results. It is intended as a preliminary report and the conclusions
must be considered as subject to those errors often found in interim reports. Additional
excavation is planned for late 1982 and the spring.
Prior to actually starting work on the wreck a number of testable hypotheses were
designed to guide excavation and recording procedures. The use of multiple, working
hypotheses has been shown to be a useful means of arriving at conclusions that will aid
in interpretation (Chamberlain, 1965; Platt, 1964). Accordingly, hypotheses related to
the age, use, and propulsion systems of the vessel, and the creation of the site, were
prepared on the basis of prior experience, other published work, and discussion with
fellow professionals. The hypotheses were as follows:

Hypothesis O. The wreck does not exist. The null hypothesis was drawn up in
order that all eventualities would be covered. For this hypothesis to be supported by the
evidence, it would be necessary to show that the materials noted during the Corps of
Engineers' survey were not the remains of a vessel.

Hypothesis 1. The wreck is a cargo vessel. For this hypothesis to be supported,

evidence of use as a carrying vessel would have to be found. This evidence could be
either structural or artifactual. Structural evidence would be configurations used in
other, documented examples of small cargo vessels. These would include interior
planking, storage areas, and damage resulting from use. Artifactual evidence would
consist primarily of nonship material interpreted as cargo.

Hypothesis 2. The wreck is a military vessel. For this hypothesis to be supported,

clear evidence of military use would have to be present. This evidence could be either
structural or artifactual. Structural evidence would consist of supporting structures
needed to bear the additional weight of armament, gun-related elements such as gun-
ports or magazines, or specially reinforced planking. Artifactual evidence would be a
The Berth 52 Vessel (09CH69l) 191

statistically significant number of military items such as buttons, accoutrements, or

other items.

Hypothesis 3. The wreck is a civil, noncargo vessel. This hypothesis is difficult

to deny and the evidence to support it would be largely negative. The lack of any mili-
tary-related materials would be good negative evidence. The lack of cargo-related fea-
tures (remains of cargo, damage resulting from cargo lading) would constitute the
principal evidence for supporting this hypothesis. An example of a vessel of this type
might be seen as a ferryboat or a fishing vessel.

Hypothesis 4. The wreck dates to a particular time period. This is really a series
of related subhypotheses. These have been consolidated under one hypothesis for ease
in handling.
H4a: The wreck dates to the 18th century
H4b: The wreck dates to the antebellum period
H4c: The wreck dates to the postbellum period
H4d: The wreck dates to the 20th century
H4e: The wreck dates to the Civil War era
These subhypotheses were purposely generalized in order to allow some flexibil-
ity in providing a date for the vessel. Construction techniques, certain structural ele-
ments, materials utilized, and the presence or absence of nonship artifactual evidence
would all be evaluated to determine a date which could support, or deny, one, or some,
of the subhypotheses.

Hypothesis 5. The wreck was a certain type of vessel. This hypothesis has been
subdivided as above to allow for multiple interpretations under one rubric.
HSa: The vessel was a rice boat
HSb: The vessel was a river "poleboat"
HSc: The vessel was a barge
H5d: The vessel was oceangoing
HSe: The vessel was a canal boat
H5f: The vessel was a gunboat
HSg: The vessel was a pleasure craft (nonworking)
HSh: The vessel was a fishing boat
Several subhypotheses under Hypothesis S can be further subdivided as informa-
tion becomes available to refine them. In particular, the category "oceangoing" can be
seen as covering a very wide group of vessels. Identification of oceangoing features
would result in the generation of a larger, most specific group of hypotheses under the
healing "oceangoing."
A large number of other related hypotheses periodically suggested themselves,
but in the interest of brevity, these are omitted in this interim report. The evidence to
support the subhypotheses would be similar to dealing with earlier hypotheses and
would be related to structural features, dating, artifactual materials, and the like. In
some cases, the vessel could easily be seen as fulfilling the criteria for more than one
vessel type, thus forcing more explicit analysis.
 192 L.E. Babits

Hypothesis 6. The vessel had a propulsion system. As with H4 and H5, this
hypothesis can be subdivided on the basis of different propulsion systems. If the vessel
had no means of propulsion, it would have been built on the site.
H6a: The vessel was propelled by oars
H6b: The vessel was propelled by poling
H6c: The vessel was propelled by towing
H6d: The vessel was propelled by sail
H6e: The vessel was propelled by inboard steam power (propeller)
H6f: The vessel was propelled by outboard steam power (paddle wheel[s])
H6g: The vessel was propelled by inboard gasoline power
H6h: The vessel was propelled by outboard gasoline power
Evidence supporting any of the propulsion subhypotheses would be largely struc-
tural although the possibility that artifactural evidence might be found within the vessel
could not be discounted. Artifactual evidence would consist of finding the oars, poles,
oarlocks, or tholepins, coal or other fuel, and the like. Structural evidence could be sup-
porting structures for the engines or oars, evidence for the mounting of engines, places
where power was transferred from engine to propulsion devices, or the presence of a
"walkway" for the polemen.

Hypothesis 7. The site was created by placing the vessel in its present location.
This hypothesis was similarly subdivided into various categories relating to site forma-
tion. The nature of the subhypotheses is related to Hypothesis 6, since much of the
same evidence would apply under Hypothesis 7, but to serve a different purpose.
H7a: The vessel was accidentally wrecked while in use
H7b: The vessel was abandoned after stripping it of material
H7c: The vessel saw multiple use prior to abandonment
H7d: The vessel was burned
H7e: The vessel was deliberately sunk, driven aground
Evidence supporting any of these hypotheses could be structural or artifactual. The
damage resulting from accidental wrecking while in use would be quite different from
that associated with deliberate scuttling or destruction. The lack of materials would also
constitute strong negative evidence for lost while in use. Evidence of burning would be
easy to detect structurally as charring of a significant portion of the vessel.
While other hypotheses were generated concerning crew size and carrying capac-
ity, these are not germane to the interim report and will be presented in detail after the
project is completed. The few hypotheses presented here were designed to guide the
retrieval of information from the site. As such, a great many were disproven as the
excavation proceeded but it was only by formulating the hypotheses that conscious
effort could be directed toward obtaining the information necessary to refute them.

FINDINGS

Inspection of exposed structural elements and the subsequent excavation of other

portions of the vessel revealed a number of things. These observations shall be dis-
cussed in the following order: hull planking, interior planking, and artifacts.
The Berth 52 Vessel (09CH69I) 193

Hull Planking

The exterior planks making up the casing of the hull were of varied thickness,
joined together in carvel fashion and fastened to the interior timbers with both metal
nails and treenails. On the starboard side, five strakes were still in place at the stern
although there was some separation from the keel and from each other (Figure 1). The
garboard strake immediately adjacent to the keel was missing at the stern but the sec-
ond plank making up this strake was noted 7.7 ft from the stern. The strakes ranged in
width from 1.08 ft to 0.458 ft with a thickness ranging from 0.145 to 0.16 ft. On the
port side, less of the vessel was uncovered but it was possible to note that five strakes
were present at the stern and that these were seemingly less disturbed than on the star-
board (downstream) side. The strakes on the port side ranged in width from 0.5 to 0.91
ft. The thickness again varied from 0.145 to 0.11 ft. As with the starboard side, the
strakes were fastened to the frames and floor timbers with a combination of nails and
treenails. The treenails on the port side had, in some cases, been spread wider by the
use of four-sided wedges to provide a better grip. The relatively intact nature of the
port side allowed the observation that spaces between at least three of the strakes had
been filled with small, flat wedges, apparently in an attempt to prolong the useful life
of the vessel CR. Fleetwood, personal communication, August 18, 1982).
The exterior planking on both sides did not extended to the tops of the frames. On
both sides, strakes were no longer attached to the frames outboard of the chine and the
upper strakes and wale were missing from the excavated portions of the vessel. This
made it impossible to tell how high the gunwale had been although it could not have
extended much past the top of the frames. No evidence of teredo damage was noted on
the interior of the hull planking.
The planking was not inspected on the underside since this was resting on the
river bottom.
Nevertheless, it was possible to tell that it had not been damaged by teredo or
other marine organisms. It also did not show any evidence of burning or severe ero-
sions such as might have occurred if it were run aground frequently. The strakes were
all cut off even with the stern, presenting a straight line across the stern section of the
vessel. No evidence of caulking beyond the wedges mentioned previously was noted.

Interior Supporting Structures

The interior structures consisted of floor timbers and frames. The floor timbers
were generally rougher and less well finished than the frames. They were often rounded
and all appeared to be in a poor state of preservation. The floor timbers ranged in size
from 0.41 x 0.58 ft high to 0.27 x 0.33 ft wide. The length of these timbers varied from
2 to 4 ft. All observed floor timbers were in advances stages of deterioration compared
with the ribs. No notches, grooves, or other gaps appeared on the underside of these
timbers suggesting that their poor state of preservation may have been the result of
standing bilge water during the life of the vessel.
The frames were much better preserved. The ribs were sawn out of larger timbers
as the curves did not appear to be natural. All frames were well squared up and most
were 0.33 ft wide by 0.41 ft high. At the outer, upper end some rounding had occurred
which seemed related more to water action than human activity. No notches or grooves
 194 L.E. Babits

were noted on the underside of the frames but they were in much better condition than
the floor timbers.
The frames and floors timbers on the starboard side were pulled away from the
keel and keelson near the stem. Forward of a point about 8 ft from the stem, they were
still mounted between the keel and keelson. The frames did not exhibit any nail or
treenail damage on their inboard ends indicating that they were held in place between
the keel and keelson by the pressure of those two timbers being drawn together by
treenails and nails. The frames on the port side were separated from the keel a short
distance but forward of a point 8.4 ft forward of the stem, they were still in place.
The gap between ribs and keel made it impossible to see precisely how these
members had been fastened together and further barred observation of the manner in
which the garboard strake abutted the keel. Nevertheless, it was possible to see that it
was not inlet or rabbeted into the keel near the stem.
The frames and floor timbers had a standard distance along the length of the vessel.
The frames were separated from each other by a space ranging from 1.2 to 1.4 ft. The
floor timbers were separated from each other by a space ranging from 1.2 to 1.4 ft as
well. The distance from rib to floor timber varied from 0.4 to 0.6 ft.
No evidence offuttocks or hanging knees was observed. The only abnormality in the
patterning of the ribs and floor timbers occurred 19.5 ft from the stem where the internal
planking ended on the starboard side. At this point a timber was noted in a tilted position
some 45 degrees from the normal and abutting a frame. Since this portion of the vessel was
not completely cleared of overburden, the function of this timber could not be ascertained.

Keel and Keelsons

The chief structural members of the vessel identified to date are the keel and
keelson.
The keelson consists of at least three different pieces of wood. The two bottom
pieces are 0.66 ft wide by 0.33 ft thick. These two pieces form a single main keelson
and are joined together with a common scarf joint (Smith, 1965: 90-92) beginning 8.9
ft from the stem and continuing to a point 10.25 ft from the stem. This scarfed joint is
fastened with two 3/4-in. iron nails located about 2 ft apart.
The final portion of the keelson is the spur, or deadwood. This piece of wood is
slightly narrower than the keelson proper on which it rests. It is fastened to the keelson
with three iron nails. It consists of a thin section forward, resting on the keelson at a
point 8.5 ft from the stem which becomes thicker toward the stem. It rises rapidly and
then forms the inner post, or spur, which extends out past the keelson 0.66 ft. This is
made of one piece of wood cut from a single log.
The keelson spur is penetrated by one treenail hole which no longer contains a
treenail. This is located on the spur proper. No use is suggested for the hole beyond
commenting that it was for a fastening. At the base of the spur, there are a pair of rough
notches which are of differing depths (Fig. 2). The importance of these notches is un-
known at present but may be related to the interior planking. The keel is, to date, a sin-
gle piece of wood, 0.75 ft wide by 0.375 ft thick which extends the length of the
excavated portion of the vessel. At this stage it has been impossible to determine if the
keel is made up of multiple timbers jointed together like the keelson because the keel is
hidden from view by intact planking and the mud which rests in the vessel interior.
 t
t:i:I
nI
S.
\Jt
1 I 1 foot ~

11 paired treenails ~
i.
II! triple treenails

~ metal nail
~J:
--~

Rib Profile
(Rib P2)

0\ =
Figure 2. Cross section of keel and keelson.
-D
-
\Jt
 196 L.E. Babits

The keel and keelson are fastened together by iron nails 3/4 in. in diameter and
about 1.125 ft long. A single square yellow metal nail is located at the stem of the keel-
son where it protrudes from the keelson in an upward and sternward direction. There is
no information as to what this nail was attached to although it probably relates to the
sternpost or rudder assembly. This nail is 3/4 in. square and extends above the keel
0.83 ft.
No evidence has been found that would indicate that the keel is deeper than the
dimensions given above. There appears to be no deep keel such as would be found on a
sailing vessel. The keel seems to have protruded only 0.208 ft below the bottom plank-
ing of the vessel. No damage to the keel was visible although some minor damage, pos-
sibly from teredo, was visible on the keelson spur.

Interior Planking
The interior planking of the vessel was visible only on the starboard side until
excavations were made on the port side some 8 ft forward of the stem. The starboard
planking consisted of two planks which were not securely fastened to the ribs on which
they rested. The outer plank was 0.729 ft wide and 0.08 ft thick. The inner plank was
1.06 ft wide and 0.10 ft thick. Both planks extended forward to a point 19.5 ft forward
to the stem.
At a point 17 ft from the stem, damage from burning was noticed on both planks.
This damage consisted of charred wood, which had either burned through the planks, or
the charring had allowed rotting which penetrated through the planks. No other area of
the vessel exhibited any evidence of burning. The only other modification to the inte-
rior planks consisted of two mortises cut into the inboard plank.
The two alterations were located on either end of the inner plank (Fig. 1). The larg-
est cut was a trapezoidal cut near the stem. This was 0.75 ft long on the edge of the plank
and 0.525 ft long on the inner edge of the cut. It extended to a varying depth of 0.25 to
0.34 ft into the board. This may have been related to notches cut in the keelson spur but
this has not yet been confirmed. The second alteration consisted of a cut in the forward
end of the same plank which was 0.1 ft deep and extended from a point 0.7 ft from the
bow end to the end of the plank. No reason for this second mortise has been discovered.
A third interior plank was noted on the port side 7.9 ft from the stem. This was
found during excavation in the interior of the vessel to learn how wide it might be. This
interior plank was still attached to the ribs above the chine. It was noted as work was
finishing. More important details claimed the crew's attention and it was not further
examined.

Artifacts
The cultural material recovered from the site falls into two categories. These are
those materials associated with the wreck and those that are found amid or on top of the
overburden. Without fail, those artifacts associated with the strakes on the port side. No
other materials related to the wreck were recovered prior to November 3, 1982, when
port frame number 2 was removed for measuring. Artifacts not associated with the wreck
are either river debris which would wash onto the site over the tide changes and then,
usually, wash out on the next tide, or aboriginal materials. The river debris was simply
The Berth 52 Vessel (09CH69I ) 197

ignored. The aboriginal materials were catalogued although they are clearly related to
dredging activities which deposited the spoil on top of the clay matrix in which the ves-
sel lies. The ceramics consist of Savannah and Irene series pottery which is plain, cord
marked, or complicated stamped (Williams, 1977: 116-134). One Savannah River point
was also found. This was made of Briar Creek chert (L. Larson, personal communication,
September 14, 1982). Since the Irene Mound was located approximately across the river
from the Berth 52 Site, the occurrence of aboriginal material is not unexpected.

INTERPRETATION

It seems best to deal with interpretation within the framework of the hypotheses.
As hypotheses are denied or supported, the evidence presented will be given in an
interpretive fashion. A word of caution is necessary here because the vessel has not
been completely excavated and a large portion of the site still needs examination prior
to firm conclusions. Nevertheless, it is possible to reject several hypotheses at this
stage and provide some firm support for others.
The null hypothesis that the wreck did not exist can be refuted. The remains at
the Berth 52 Site are clearly those of a boat. They are intact enough to show a large
number of constructional details and thus demonstrate that the remains are not those of
a washed-out bridge or dock. Hypothesis 0 is rejected. Hypothesis I stated that the
wreck is a cargo vessel. This hypothesis received considerable support because of the
presence of interior planking and some possible damage to the planking as a result of
loads being placed on it. No evidence of cargo was noted, however, suggesting that fur-
ther work is necessary prior to confirmation of the" hypothesis. Hypothesis I is sup-
ported but not confirmed by the evidence recovered.
Hypothesis 2 stated that the wreck was a military vessel. No structural evidence
to support this hypothesis was found. There were no reinforcing timbers of the sort
found on gunboats, nor were there any features such as gunports to indicate a military
use. The lack of any artifactual material is negative evidence that the vessel was not
military but this is not conclusive. While structural evidence argues against a military
use, the lack of artifactual material does not deny the possibility. Hypothesis 2 is nei-
ther rejected nor supported.
Hypothesis 3 stated that the wreck was a civilian, noncargo vessel. This hypothe-
sis is not supported by the evidence found at the site. The negative evidence of no mili-
tary artifacts suggests it might have been civilian. Structural evidence, however,
indicates that the vessel was probably engaged in cargo hauling. This evidence is lim-
ited to the interior planking on the port and starboard sides. Interior planking would be
necessary to keep cargo out of the bilges and to provide a stable platform for it to rest
on while undergoing transportation. No artifactual evidence was found to suggest what
that cargo might have been. On the basis of the damaged interior planking, Hypothesis
3 is rejected.
Hypothesis 4 stated that the wreck dated to a certain time period. This time
period was indicated in a number of subhypotheses relating to the age of the vessel.
Hypothesis H4d seems to be rejected on the basis of fasteners used in the vessel. The
nails were hand-forged iron which had not been galvanized. No Munz metal nails were
found indicating that the builder was either not willing, or unable, to use better-quality
late 19th- or 20th-century materials in the fastenings of the vessel, or that the vessel
 198 L.E. Babits

was built prior to the time these materials became generally available. None of the
other temporal hypotheses can be rejected or confirmed although some additional dis-
cussion is germane.
If Hypothesis H5e, that the vessel was a canal boat, is correct, then it might be pos-
sible to ascribe the vessel to either antebellum or postbellum temporal provenience. The
Dundee Canal was licensed after the Civil War (T. Dees, personal communication,
November 7, 1982). Since the Dundee Canal is within 500 m of the site upstream, it
might prove to be the source for the vessel. At the same time, it might be possible to as-
cribe the vessel to the earlier Savannah and Ogeechee Canal which is approximately 1
mile downstream. Either interpretation depends on a better identification of the vessel.
Hypothesis 5 suggested that the type of vessel could be identified. At the present
time it is possible to reject some of the subhypotheses. H5d stated that the vessel was
oceangoing. The absence of a deep keel, the shallow draft, and the low gunwales sug-
gest that this vessel was not oceangoing. There is little or no evidence on the outer hull
to show teredo damage. This lack of teredo damage argues for a freshwater use.
Accordingly, H5d is rejected.
Hypothesis H5f stated that the vessel was a gunboat. This is hardly likely since
there are no supporting structures to allow heavy weaponry on board. There are no gun
mounting features and the narrow size of the vessel also argues against this sort of use.
Accordingly, H5f is rejected. Hypothesis H5g stated that the vessel was a pleasure
craft. The lack of any accommodations for seating, the damage suggestive of cargo
handling, and the lack of any fine appurtenances such as finer wood, brass fittings and
the like argue against acceptance of this hypothesis. Accordingly, H5g is rejected.
Hypothesis H5h stated that the vessel was a fishing boat. This is not supported by the
evidence found to date, but this is not sufficient grounds for rejection. H5h is neither
rejected nor confirmed.
The rejection of three of the eight vessel-use subhypotheses indicates that some
progress is being made. Narrowing down the possibilities is a helpful sign. The struc-
tural similarities between a rice boat, riverboat, and a barge or canal boat preclude
making any assessment at this time although there is a growing feeling that a canal boat
might be the best fit with the evidence. A rice boat is generally thought to be somewhat
less sturdily built than the vessel at the site.
Riverboats which were poled have been used on the Savannah River since the
18th century but most are of a size much larger than the Berth 52 vessel (Fleetwood,
1982: 88-89; Rahn, 1968: 5-58). Commercial poled vessels appear in the documentary
sources from circa 1740-1836 (Fleetwood, 1892:89-90; Rahn, 1968:12,18). The Berth
52 vessel is thought to be a little narrow for consideration as a barge; the latter are
thought to be much wider than the approximately 9 ft credited to the Berth 52 vessel.
All of this negative evidence points to the idea of a canal boat having the best fit
with what information is available at this time.
A canal boat would have additional support based on the geographical location of
the site. Less than 500 m upstream was the exit into the Savannah River of the Dundee
Canal. This canal was built by the Dundee Railroad and Canal Company after the Civil
War (T. Dees, personal communication, November 7, 1982). Similarly, the Savannah
and Ogeechee Canal, to which the Dundee Canal had some connections, enters the
Savannah River about I mile downstream. The Savannah and Ogeechee Canal was in
use from before 1831 to circa 1865 (Rahn, 1968;35-36). One scenario for placing the
The Berth 52 Vessel (09CH69l) 199

vessel on the site is to see the vessel as a canal boat which was used, repaired, and
finally abandoned in the canal after being stripped of its useful furnishings, A flood
could have flushed it out into the river where it then drifted until it washed up down-
stream from the canal. At the present time, this is largely a subjective matter as the por-
tion of the vessel that might shed the most light on the question, the bow, has still not
been inspected.
While the low freeboard, lack of propulsion system (see Hypothesis 6), and cargo-
related structural features suggest use in calm waterways, this is not conclusive evidence
to support the provisions of Hypothesis 5e. Nevertheless, the idea of a canal boat does
agree with the available information presented. The possibility that the vessel is a rice
boat, a poled boat, or a barge is still under consideration and information will be sought
to make this determination when the site is inspected later this year.
The sixth hypothesis was related to the propulsion system. Of the eight subhy-
potheses, four can be rejected at this time. There was no evidence for steam or gasoline
power, either as to machinery remaining in the vessel or in indications remaining on the
vessel showing where such machinery had been attached. There were no mountings for
gasoline engines, nor were there any structural indications of steam power. Since these
elements would most likely have been mounted in the stern, or midships, section, they
should have been found and identified by now because the effort concentrated on that
area. Accordingly, Hypotheses 6e, 6f, 6g, and 6h are rejected.
Hypothesis 6d stated that the vessel was propelled by sail power. This is not sup-
ported by the evidence recovered. No mountings for supporting lines or rigging for
sails have been found although this is negative evidence since the gunwales are not in-
tact in those areas that have been inspected. Similarly, no mast step has been found.
The lack of a mast step is not conclusive since the full extent of the keelson has not
been examined. Nevertheless, it is unlikely that a mast would be located so far forward
on a vessel of this size. Another mitigating factor relating to the lack of sail propulsion
is the absence of a deep keel. The keel on this vessel is not designed for providing
resistance against the thrust of the wind into the sails. While Hypothesis 6d cannot be
conclusively rejected, it has little supporting evidence.
The remaining three subhypotheses are interrelated. They suggest that the vessel
was propelled by oars, by poling, or by towing. No evidence of oarlocks or T hole pins
has been found on the vessel, but the damage to the gunwale area is sufficient to have
caused the loss of this information. Poling and towing the vessel would not require spe-
cialized structural features, except perhaps a supported walkway for poling and rein-
forcement of the bow for towing. The three subhypotheses relating to rowing, towing,
or poling all have equal validity at this stage.
Hypothesis 7 related to the creation of the site. Hypothesis 7a stated that the ves-
sel was wrecked while in use. This is not likely because there is no evidence of artifi-
cial material such as tools, equipment, or cargo which would have been on a vessel in
use. Accordingly, this hypothesis is rejected.
Hypothesis 7e stated that the vessel was deliberately sunk or driven aground. This
is not supported by the evidence recovered to date. The vessel seems to have been
stripped of all furnishing prior to its eventual arrival on the site. No structural evidence
relates to deliberate sinking or running aground. All damage seems to be the result of
use, age, and river action, rather than deliberate human destruction.
Consequently, Hypothesis 7e is rejected.
200 L.E. Babits

Hypothesis 7d stated that the vessel was burned. This can be rejected because
there was no evidence for burning except a rather small area on the two starboard inte-
rior planks. This fire damage seems more likely to have occurred during use, probably
as a result of crew activities such as cooking, than from burning of the vessel. There-
fore, Hypothesis 7d is rejected.
The last two subhypotheses under Hypothesis 7 are that the vessel saw multiple
use prior to abandonment (7c) and that it was abandoned after being stripped (7b).
Until such time as the vessel is completely excavated and the analysis is completed, it
is difficult to deny or support 7c. The evidence for 7b seems conclusive. The vessel
clearly seems to have been stripped of useful material. No artifacts have been found
except for structural elements such as nails and wedges. All hardware, if there was any,
has been removed. Stripping might also account for the missing gunwales and interior
planking. Hypothesis 7b has the best fit with the evidence noted.
The summation of the results suggested that it is most likely that the vessel was a
canal or river boat which was propelled by oars, poling, or towing. This vessel saw
extended hard use and there were efforts to extend its life by driving wedges into
treenails and between strakes to plug leaks. Despite these alterations, the vessel was
eventually seen as unfit for further use and was then stripped of materials that might be
utilized elsewhere. The vessel was then abandoned to the river where it washed, or was
pulled, up into its present location.
This last stage of the vessel's history occurred before dredging activities in the
Savannah River piled large amounts of spoil on Hutchinson Island. It occurred far
enough in the past to permit the burial of the vessel under up to 4 ft of thick clay which
has acted as a preservative ensuring that the vessel would be largely intact. Without argu-
ing for any particular rate of clay deposition along the Savannah River, a date in the late
19th century seems most likely for the disposal of the vessel at its present location.

CONCLUSIONS

If the vessel at the Berth 52 site is, indeed, a canal boat, then it is almost certainly
eligible for inclusion on the National Register of Historical Places. No known Georgia
canal boats survive although canals were heavily used in the 19th century. As a one-of-
a-kind item dealing with a poorly documented period of history, this vessel is entitled
to the protection of the law and its preservation should be encouraged and supported by
all agencies concerned. Even if the vessel is not a canal boat, the lack of knowledge
about southeastern watercraft (Fleetwood, 1982;vi) argues for the continued study of
the vessel and what information it can yield about ship construction techniques and
uses in the area.

RECOMMENDATIONS
It is recommended that the vessel be moved as soon as excavation can be com-
pleted. Movement to a more protected environment is essential since dredging to widen
the channel is planned for the area. Even if the site were not directly impacted, the
resulting increase in water flow and the agitation created by deep draft vessels would
hasten the ongoing erosion of the site and soon destroy the material evidence that
remains of the site.
The Berth 52 Vessel (09CH691) 201

Excavation of the site is hampered by several factors. The first of these is the
restriction placed on work by the tides. Working is limited to about 4 or 5 hours a day
every other week. This time is further restricted by a second factor. Union Camp does
not desire that persons utilize their property for access. Consequently, they have
erected a gate which is open only between 8:30 A.M. and 3:30 P.M. across the one
access road to the site. Further restrictions, understandably created by concern for
liability, have been made as well. This combination of natural and cultural factors
means that work can only proceed for short times during alternate weeks.
Work is similarly constrained by academic demands during the school year. If
access to the site could be granted for one or two weekends when favorable tide and
weather conditions exist, then it is possible that a larger work crew could successfully
excavate the vessef and remove it to a more protected environment. Such an effort
would be better based on the land rather than operating from the river for logistical rea-
sons. The removal of the vessel by water may be the only means to transport it without
heavily damaging it or dismantling it because no vehicles could operate along the river-
bank where the site is located.
Preservation of the vessel can be seen as forming three options. The first is to do
nothing, in which case the vessel will be lost to the action of the river within the next 3
or 4 years. The second option is to remove the vessel and place it into a similar envi-
ronment in a more protected location. This could be done by placing it in mud in a wet
area away from the actions of the river. The third option is to remove the vessel, stabi-
lize and preserve it with the use of petrochemical products, and then make it available
for public display.
The first option is not considered valid since the idea is to learn something about
a little-known aspect of southeastern history. The second option is the one initially con-
sidered. Under this option, the vessel would be removed to Fort Jackson, a historic site
with an agency working to preserve southeastern maritime materials. It would be bur-
ied in the moat of the fort, thus utilizing presently wasted space for preservation pur-
poses As part of the burial activities, a section of the moat would be excavated with
archaeological techniques, thus revealing additional information about Fort Jackson.
This option would be cheaper than option three but would basically be a holding action
with regard to preservation. The vessel would not be available for further study without
reexcavation of its new resting place.
The third option is the most expensive. Preservation of the vessel by immersion
into a tank of petrochemicals until the wood is saturated and stabilized would later
allow display. Unfortunately, this process would be expensive and it is not certain that
funding is available. The only nearby facility for this process is at the University of
South Carolina in Columbia and it is presently being used to stabilize the Brown's
Ferry Wreck (Albright and Steffy, 1979). The time allotted for this preservation is
unknown.
Since the first option is unacceptable and the third too expensive, only the second
option can be seen as valid. Consequently, it is recommended that a work crew be mus-
tered on the site as soon as possible and that this crew remove the vessel as rapidly as
possible within the constraints of acceptable archaeological practice. The vessel would
then be removed to Fort Jackson where the requisite portion of the moat will be exca-
vated in conjunction with additional measuring of the vessel during periods when dig-
ging is not possible, then the vessel will be available for final stabilization treatment.
202 L.E. Babits

ACKNOWLEDGMENTS

In a project of this sort, involving so many different governmental agencies, there

are necessarily a large number of people to thank for their participation. First, of all the
agencies, the u.s. Corps of Engineers, Savannah District, must be credited with finding
the site and calling it to the attention of the principal investigator. The Corps also
issued permission to work on the site and continued cooperation with the staff archae-
ologists has ensured that the project would go on. The Corps personnel who have
assisted in this project are Richard Anuskiewicz, Paul Rubenstein, and Judy Wood.
Other permits were issued by the Georgia Department of Natural Resources,
through the offices of the State Historic Preservation Officer and the Real Estate Unit.
Encouragement and assistance in coping with the details of permission were readily
provided by these agencies. Similarly, a considerable amount of interest was shown by
the State Archaeologist, Lewis Larson, who visited the site soon after the main work
was terminated due to academic commitments.
The Chatham County Commissioners provided permission to work on their prop-
erty and Union Camp allowed access to the site across their dike. Mr. James M. Piette of
Union Camp and Thomas A. Smith, Jr., Director of Public Works in Chatham County
provided permission to complete the project.
Finally, the crew members who worked on the site in August and September are
to be given a great deal of thanks for their toleration of poor working conditions and
the care with which they excavated the vessel under the most trying situations. Without
them, the project would never have come as far as it has. Claire Livingston, April Scott,
and William Maddox endured mud, heat, and insect pests with equal fortitude while
moving overburden from the site. They then meticulously recorded the necessary meas-
urements to ensure that data would be available when needed.

REFERENCES
Albright, A.B., and J.R., SteffY, 1979, The Brown's Ferry Vessel, South Carolina. International Journal oj
Nautical Archaeology 8(2): 121-142.
Chamberlain, T.e., 1965, The Method of Multiple Working Hypotheses. Science 148: 754-759.
Fleetwood, R., 1982, TIdecraft. Coastal Heritage Society, Savannah.
Garrison, E.G., R.M. Holcombe, and L. L. Lowery, Jr., 1980, A rchaeological and Engineering Study of the C. S. S.
Georgia: Final Report. Report on file U.S. Army Corps of Engineers, Savannah District, Savannah.
Platt, J.R., 1964, Strong Inference. Science 146: 347-353.
Rahn, R.A., 1968, River Highway Jor Trade. The Savannah. U.S. Army Corps of Engineers, Savannah. District.
Smith, H.G., 1965, Boat Carpentry. Van Nostrand-Reinhold, New York.
Williams, S. ed., 1977, The Waring Papers. Papers oJthe Peabody Museum oj Archaeology and Ethnology 58.
Cambridge, Mass.
 Part III.B

Theory

Incorporating concepts from Merton and Binford, Raab and Goodyear (Chapter 19) lay
out the definitive work on middle-range theory. Most importantly to the nautical
archaeologist, middle-range theory has applications for site formation processes and
site interpretation. However, their article goes beyond these applications and into the
realms of sociology and epistemology, with a corresponding thickening of vocabulary.
Archaeologists must rely on middle-range theories to provide the context for
observations. Anuskiewicz, in Chapter 20, provides a good example of how the middle-
range potential of remote sensing devices can be incorporated into research design. His
"expected archaeological indices" are followed by "expected instrumental indices." His
plan, centered around 8t. Catherine's Island, Georgia, also utilizes Gearhart's buoyant
hull scheme.
In Chapter 21, Lenihan and Murphy review the basic literature of shipwreck
archaeology for explicated research designs. They concentrated primarily on the Inter-
national Journal of Nautical Archaeology and Cultural Resource Management docu-
mentation, looking for indications of research design and found few examples.
Theoretical approaches contained within general works by Bass, Muckelroy, and others
have not been explicitly translated into the level of site reports. The field needs further
progress in theory building, especially in light of the destructive nature of underwater
archaeological investigation.

203
 Chapter 19

Middle . . Range Theory

".~i).:
~

in Archaeology
-r/: ..
':.i
A Critical Review of
Origins and Applications

L. MARK RAAB and

ALBERT C. GOODYEAR

The concept of middle-range theory, arising over three decades ago in sociology, is
reviewed. The concept was proposed as an approach to theorizing, urging consolida-
tion of high-order theories with low-order empirical studies. The critical elements in
such hierarchies are theories of a middle-range of abstraction. However, most current
conceptions of "middle-range theory" in archaeology are far more narrowly con-
ceived. Derived primarily from Binford's work, they continue the New Archaeology's
attempt to develop a materialist epistemology for archaeology. In this view, principles
of site formation processes are nearly synonymous with "middle-range theory." The
dangers to theory building of this approach are outlined. Examples of middle-range
theory that expand our capacity for explanation of cultural behavior are presented.

The concept of middle-range theory in archaeology is an increasingly influential one.

For example, in their widely read volume, A History of American Archaeology, Willey
and Sabloff (1980) suggest that middle-range theory may play a major role in the
development of theory within our discipline in coming decades. Such well-publicized
predictions have a way of becoming self-fulfilling prophecies. At this point, critical dis-
cussion of the topic is badly needed.
Source: American Antiquity (1984), 49 (2):255-268. Reprinted by pennission of the Society for American
Archaeology.

205
206 L.M. Raab and A.C. Goodyear

Having apparently introduced the concept of middle-range theory in archaeology

several years ago (see Schiffer, 1980:377), we have followed subsequent developments
with considerable interest. We proposed (Raab and Goodyear, 1973) that middle-range
theory might help bring order to theory building under the spirited but eclectic
onslaught of the New Archaeology. In the same vein, we treated the topic briefly
(Goodyear et al., 1978:161-162) in connection with difficulties of theory building in
conservation archaeology. In the meantime, several discussions of the subject have
appeared, which, in our opinions, are producing considerable confusion.
There is little evidence that archaeologists understand the concept of middle-
range theory as originally conceived by social science theorists. Those theorists
advanced the concept, not as a substantive theory of any particular phenomenon, but
rather as middle-range theorizing. The essential point was to develop a strategy for in-
tegrating research problems and data into cumulative bodies of scientific knowledge in
which theories of limited scope, arrayed at different levels of generality, could be sub-
sumed under domains of increasingly general principles. The concept of middle-range
theory was thus a part of a larger theory building enterprise. Archaeological interpreta-
tions of middle-range theory have so far been much more narrowly focused. These
interpretations result largely from the distinctive logical problems involved in attempt-
ing to infer behavior from material traces. At the same time, confusion is compounded
by current uses of the term "theory" in archaeology. Any hypothetical statement,
regardless of its topic or degree of abstraction, seems to be a candidate for theory and
exists on the same plane of importance as other ideas. One result of this "flat" view of
theory is a tendency to confuse "data language" (cf. Price 1982:711), involved in the
operational or methodological detection of data patterning, with ideas about the causes
of such patterning contained in a "theory language." Current understandings of "mid-
dle-range theory" in archaeology encourage such confusion.
This paper's first objective is to examine briefly middle-range theory within its
original sociological context. Archaeologists have overlooked this background, thinking
perhaps that it would tell them only something about sociological theories. The facts are
quite different. One consequence of this oversight is that confusion has followed at-
tempts to assimilate into archaeology what has been called "middle-range theory."
The second objective of the paper is to examine the problems and prospects of such
attempts at assimilation. Some of the confusion here is probably semantic, owing to an
unfortunate borrowing of the term middle-range theory from sociology. Nevertheless,
the difficulty goes beyond semantics. At issue is the ability to discriminate between dif-
fering research strategies; i.e., whether to emphasize the development of general method-
ology or cultural theory of human behavior, be it middle-range or otherwise.

MIDDLE~RANGE THEORY IN SOCIOLOGY

Middle-range theory was integral to major changes in sociological theory follow-

ing the Second World War. Several works by Parsons (e.g., 1948, 1950, 1951) were of
crucial importance. Where previous theory had focused on the individual or groups of
individuals, Parsons advanced the structural-functionalist concept of the "social sys-
tem." This highly abstract entity was conceived as being comprehensible through
analysis of statuses, roles, norms, institutions, and other societal "functions." More-
over, Parsons took the position that this research program could proceed from an a pri-
Middle-Range Theory in Archaeology 207

ori and overarching theory of the social system and social action. The strategic objec-
tive was to bring social phenomena within the scope of a unified body of theory. If the
desired inclusiveness of this scheme necessarily involved construction of highly ab-
stract and generalized concepts that were sometimes difficult to relate to actual
research, Parsons and his followers were willing to accept this condition as the price of
a unitary theoretical approach (Larson, 1973:12~136; Mullins, 1973:57-59).
One sociologist, R.K. Merton, was unwilling to adopt such a position. Beginning
with an essay in 1948, Merton began development of a theoretical program with the
concept of middle-range theory at its core (e.g., Merton, [1949] and subsequently in
revised editions [1957 and 1968]). A major part of this program was an attempt to
counter perceived dangers in Parsons's approach.
Merton sees an all-too-easy separation between theory and empirical studies.
Although motivated in part by a concern that grand theorizing may come to little if it
cannot be applied to actual research, middle-range theorizing is equally a critique of
blind empiricism. On the one hand, Merton notes a tendency to erect grand-scale "mod-
els," highly abstract "approaches," and other broadly conceived conceptual schemes
that have little hope of actually being tested:

A large part of what is now described as ... theory consists of general orientations
toward data, suggesting types of variables which theories must somehow take into
account, rather than clearly formulated, verifiable statements of relationships
between specified variables. We have many concepts but fewer confirmed theories;
many points of view, but few theorems; many "approaches" but few arrivals. (Mer-
ton, 1968:52)

Broad theoretical principles are not entirely rejected. Such schemes can provide
antidotes to uncontrolled eclecticism. And yet Merton calls for recognition that abstract
ideas do not necessarily lead to tested, or even testable, theories.
Conversely, Merton attacks empiricism that fails as "directed," i.e., theoretically
guided, work. Merton (1968: 149-150) argues that "the notion of directed research implies
that, in part, empirical inquiry is so organized that if and when empirical uniformities are
discovered, they have direct consequences for a theoretic system." This stricture in no way
negates the value of empirically based work. An empirical discovery may be the stroke of
serendipity that compels a new theoretical direction (Merton, 1968: 157). In the long run,
however, shotgun empiricism is seen to offer a poor return for systematic theory.
Merton recognized this dilemma between empiricism and theorizing as pervasive
in human behavioral research generally. Accordingly, Merton's (1968:38) stated objec-
tive for middle-range theory has been welcome to many:

Middle-range theory is principally used ... to guide empirical inquiry. It is intermedi-

ate to general theories of social systems which are too remote from particular classes
of social behavior, organization and change to account for what is observed and to
those detailed orderly descriptions of particulars that are not organized at all. Middle-
range theory involves abstractions, of course, but they are close enough to observed
data to be incorporated in propositions that permit empirical testing.

Middle-range theory is intended to vary in levels of abstraction, to be flexible in

seeking sources of working hypotheses, and to be aimed at accumulating a body of the-
208 L.M. Raab and A.C. Goodyear

ory. The substantive content of research problems obviously varies from one area of
inquiry to another, the specificity of these problems varying according to existing data
and hypotheses. Despite the fact that particular research problems are more or less
broadly conceived, middle-range theory is seen as providing a logical link between
relatively low-order empirical generalizations and comparatively high-order theories.
One outcome of middle-range theorizing can be the creation of a logical structure
in which low-order working hypotheses tend to confirm or negate propositions in a
middle stratum and the latter in tum reflect on the validity of yet more generalized
theories. From an inductive perspective, one can enter this hierarchy by means of
"grounded theory" (cf. Glaser and Strauss 1967) based on empirical findings. On the
other hand, a series of testable propositions can be derived deductively from existing
theories in ways suggested by Hempel (1965), Popper (1959), and others. In the Mer-
toni an scheme middle-range theory is the critical bridge between theory and data that
allows both kinds of operations to be effective. Debates about whether an inductive or
deductive approach is "best" are rendered pointless.
Merton's own work as a sociologist reflects the tenets of middle-range theory.
One example is his theory of social structure and anomie (Merton, 1968:185-214).
Briefly, Merton proposed that contemporary U.S. society be conceived as a social sys-
tem within which there are strongly held beliefs that individuals should be able to attain
wealth, power, and prestige on the basis of their own abilities and accomplishments
(upward mobility within a meritocracy). He also suggested that society in fact does not
provide the structural means for all or even most people to actually attain such "suc-
cess." The fundamental consequence is that anomie (normlessness, strain toward social
conflict) emerges, caused by a disjunction between culturally valued ends and available
means. Merton went on to develop a matrix of predicted responses to this conflict that
included such outcomes as criminal behavior, ritualistic conformity to the rules of
bureaucracies, rebellion, suicide, and others.
The important point is that Merton attempted to show that a variety of specific social
behaviors can be understood in relation to a more general principle. From the middle-range
theory perspective, the theory of anomie is a critical link between many specific forms of
behavior (e.g., criminal acts, suicide) on the one hand, and certain fundamental social
structures on the other hand, of which contemporary U.S. society provides one case.
Where the theory of anomie is too generalized to be tested directly, it can be evaluated by
seeing how well a series of derived predictions about different and specific behaviors hold
up to empirical testing. This is of course a great simplification of this work, but it does
show perhaps the logico-empirical objectives of middle-range theory in practice.
While one need not tum to sociology to learn that research should be both theo-
retically directed and empirically grounded, a look at the history of middle-range the-
ory in sociology shows that the concept has generated both praise and criticism. Even
after 30 years of debate about middle-range theory, uncertainty persists about what it is
and what its value might be. Merton's assertion that research should be guided by mid-
dle-level theories, and such examples of this approach as his theory of anomie, have
struck critics (e.g., Larson, 1973:138) as too vague. Critics (e.g., Freese) ask how, for
example, one obtains such propositions in the first place? At what level of abstraction
ought these propositions be aimed? How might insight into various middle-level con-
structs eventually be melded into unified theory (Freese, 1980a, 1980b:206-207)? Pro-
ponents of alternative theoretical schools also point out that Merton's structural-
Middle-Range Theory in Archaeology 209

functionalist approach represents an ideological bias that supports the study of systems
as static entities, and ignores fundamentally different theoretical approaches to behav-
ior such as exchange theory, symbolic interactionism, Marxian conflict, and others
(Freese, 1980b:206; Larson, 1973:140-141). The truly large body of literature com-
menting on Merton's work (see Coser, 1975:516-522, for references) makes evident
that his ideas have received both widespread support and criticism.
Nevertheless, Merton's concept of middle-range theory has been a fundamental
influence on thinking about social theory. Freese (l980b:206) summarizes the current
view of the Parsons-Merton debate, including the middle-range theorizing that is an
integral part of it. He suggests that "this issue is still embedded in the sociological con-
sciousness, though now it has the character of old wine in new bottles." Freese
(1980b:206) further suggests that

The proposal for a comprehensive, unified system oftheory for sociology has usually
equated the systematic growth of scientific knowledge with consolidation of theory
and empirical research. Merton's classic essays on the relation of theory to re-
search ... supplied the conventional view: The problem was to consolidate sociological
knowledge so that theories and data become integrated and special theories converge.

MIDDLE,RANGE THEORY IN ARCHAEOLOGY

How did the term middle-range theory find its way into archaeology, and how do
its archaeological applications compare with the original sociological objectives?
Given the decades of use that the term received in sociology prior to its appearance in
archaeology, one might well expect archaeologists to have acknowledged the sociologi-
cal connection. Curiously, however, such a connection rarely appears in current archae-
ological literature. Apart from a brief notation of the term's origin in sociology by the
present authors (Goodyear et al., 1978:161), we can find no other publication to date
that deals with the intellectual history of this term, despite the fact that "middle-range
theory" is referenced in several archaeological publications, including a textbook, as
well as book-length research reports (see below). For the reader of current literature on
"middle-range theory," the term seems to have arisen de novo in archaeology.
To a large extent, then, we are forced to provide our own analysis of the proc-
esses through which "middle-range theory" has been introduced to archaeology. When
compared with the intellectual program advanced by Merton, such an analysis produces
revealing results. It should be clear that Merton was urging a certain view of the entire
research enterprise. Middle-range theory emphasized careful formulation and testing of
theories that could link empirical data with higher-order conceptual schemes. All of
this work addressed explanations of variability in social behavior as the immediate and
central purpose of theory building. However, study of "middle-range theory" in many
current archaeological publications reveals quite different objectives.
Instead of looking at the fit between theories at various levels of application in
archaeology or at the adequacy of such theories as explanations of cultural behavior,
"middle-range theory" has tended to become narrowly methodological in character.
This is the result of efforts to deal with the archaeological problem of material data.
Principles of site formation processes have become virtually synonymous with
"middle-range theory." In our view there are dangers here. One danger is simply that
210 L.M. Raab and AC. Goodyear

confusion arises once the term middle-range theory has been transplanted into archae-
ology without adequate attention to intellectual history. A more serious problem is that
some may believe that pursuit of methodological problems alone necessarily consti-
tutes an exercise in building "theory." That belief is unwarranted, if we mean by theory
the conceptual devices by which we seek explanations of cultural behavior.
In order to understand how "middle-range theory" in archaeology has developed
such a narrow orientation, it will be necessary to look briefly at aspects of the recent
intellectual history of the discipline. We take up that task next. We will then consider
current uses of "middle-range theory" in the archaeological literature, and follow this
with an examination of dangers that those uses involve for theory building. Finally, we
will present examples of theory building in archaeology that we believe more nearly
conform to the original concept of middle-range theory, and point out why we believe
these strategies better serve the needs of archaeology.

The "Archaeological Theory" Precursor to "Middle, Range Theory"

Issues concerning adequate scientific interpretation of the archaeological record

constituted a major focus of the New Archaeology in the late 1960s. Although the term
New Archaeology may conjure up such issues as improving the role of archaeology as
anthropology, the merits of deductive versus inductive reasoning, or the desire to
employ more precise and sophisticated analytical techniques, these concerns had been
voiced before (Caldwell, 1966; Kluckhohn, 1940; MacWhite, 1956; Taylor, 1948; see
Willey and Sabloff, 1980: 185-188). What was fundamentally new and significant in
this movement was the questioning of the entire conceptual structure of archaeology as
a science. By this we mean the metaphysical, logical, and procedural assumptions on
which depended conventional means for basing knowledge claims about the past on the
archaeological record.
Binford's work was perhaps the single most influential element of this move-
ment. Binford (1968a,b) was as much concerned with the scientific adequacy of archae-
ology as he was with archaeology's contribution to anthropology's broader goals.
Indeed, the Binfords argued (1968:2) that there would be no significant contributions to
anthropology's goals until the scientific adequacy of archaeology was improved. The
methodological issue of adequately warranted inferences from the archaeological
record, and the connection between this logic and the higher goal of explaining cultural
processes, were variously described by Binford and his students as "arguments of rele-
vance," "bridging arguments," and "archaeological theory" (Binford, 1968a; Binford
and Binford, 1968:2; Fritz, 1968; Fritz and Plog, 1970). A clear statement of the nature
of what came to be called archaeological theory is given by the Binfords and well
expresses their concerns about epistemological adequacy:

Archaeological theory consists of propositions and assumptions regarding the archae-

ological record itself - its origins, its sources of variability, the determinants of dif-
ferences and similarities in the formal, spatial, and temporal characteristics of
artifacts and features and their interrelationships. (Binford and Binford, 1968:2)

It is important to realize the great extent to which the practitioners of the New
Archaeology named above were concerned with the issue of scientific adequacy with
Middle-Range Theory in Archaeology 211

regard to material remains and the development of data languages for dealing with
them. That concern may well be one of the most important intellectual legacies of the
New Archaeology. The tenets of this approach may be summarized as follows: The
archaeological record is a contemporary phenomenon in which it is not possible to
experience the past directly but only indirectly by means of appropriate instruments
(Binford, 1968b, 1981 b; Fritz, 1972). Consequently, archaeological remains entail no
inherent or objective meanings, but receive only those meanings supplied by contempo-
rary observers (Binford, 1968a; Hill, 1972). One must, therefore, differentiate a past
dynamic or systemic context (Schiffer, 1972) of events from the presently observable,
or static, archaeological context. Once this distinction has been made, concepts for
accurately translating statics into dynamics can be identified. To effect such transla-
tions, the behavioral and natural processes responsible for the material record must be
securely identified in order to build a structure of inference (Binford and Binford,
1968:2; Fritz, 1972; Reid et al., 1975; Schiffer, 1972). Moreover, any such principles
of translation, in order to provide reliable knowledge, must be covered by lawlike
propositions and be based on uniformitarian assumptions (Binford, 1968a; Watson et
al., 1971; Watson, 1976; see Sullivan, 1978, for a review of these ideas).
The importance of these developments is that they are very much a prologue to
present problems with "middle-range theory." Turning to the present literature, it
becomes evident that what Binford currently describes as "middle-range theory" is no
more than the methodological domain formerly described as "archaeological theory";
the modifier "archaeological" again refers to the intellectual program intended to deal
with the problem of material data. In our view, however, nothing has been gained by a
change of labels, except perhaps confusion.

"Middle-Range Theory" in the Current Literature

To the best of our knowledge, the phrase middle-range theory was initially intro-
duced into the published literature of archaeology by Binford in his introduction to For
Theory Building in Archaeology (1977a: 1-10). Although it was never formally defined
by him in that or subsequent works, several passages may be quoted that present a good
idea of what Binford intended by "middle-range theory":
There are urgent needs for theory building on at least two levels. One level is
what I refer to as middle-range theory. If one accepts observations made on the archae-
ological record as contemporary facts along with the idea that such facts are static, then
clearly basic problems for the archaeologist include (a) how we get from contemporary
facts to statements about the past, and (b) how we convert the observationally static
facts of the archaeological record to statements of dynamics (Binford, 1977a:6).
From these circumstances, Binford (1977a:7) concludes that "we must develop
ideas and theories (middle-range theory) regarding the formation processes of the
archaeological record. Only through an accurate understanding of such processes can we
reliably give meaning to the facts that appear, from the past, in the contemporary era."
The concept of archaeological "middle-range theory" was further developed by
Binford (l981a) in a book-length treatment of the subject, which specifically addressed
problems in the general methodology of faunal analysis. There, he refers to this form of
analysis as "middle-range research" and "middle-range theory building." In a footnote
(Binford 1981 a:25), he indicates that these expressions are, in essence, what David
212 L.M. Raab and AC. Goodyear

Clarke (1973:8) referred to as "interpretative theory," and they apparently also equate
with "behavioral archaeology" as proposed by Michael Schiffer (1976). Again,
although he does not provide formal definitions, Binford (1981 a:25) specifies what is
desired from the conduct of "middle-range research":

... accurate means of identification, and good instruments for measuring specified
properties of past cultural systems. We are seeking reliable cognitive devices; we are
looking for "Rosetta stones" that permit the accurate conversion from observation on
statics to statements about dynamics. We are seeking to build a paradigmatic frame of
reference for giving meaning to selected characteristics of the archaeological record
through a theoretically grounded body of research, rather than accepting folk knowl-
edge ~ let alone implicit folk knowledge ~ as the basis for describing the past.

Note that Binford (l977a, 1981a) is careful to distinguish "middle-range theory"

from "general theory." Although he has never discussed the structure of general theory
in detail in any of his writings, he uses it to refer to concepts about why cultural sys-
tems were organized as they were (or are), and why they changed from one organiza-
tional state to another. For Binford, middle-range theory is apparently intended to
provide logico-empirical bridges between the static phenomena evident in the contem-
porary archaeological record and the behavioral dynamics that are inferred to have pro-
duced those phenomena. But those dynamics do not encompass larger questions about
the causes of change or stability in cultural systems. The objective in the 1981 volume,
Binford's most extended treatment to date of "middle-range theory," is to create a kind
of data language for determining in which cases faunal remains may accurately be said
to be the product of natural biophysical forces or the result of human manipulation.
When set against the larger goal of explaining cultural dynamics, Binford's promulga-
tion of "middle-range theory" is clearly methodological in character because it allows
archaeologists to deal with material records but not necessarily with problems of cul-
tural dynamism. Binford is not at all confused about this distinction. He views his work
with faunal materials as a necessary step toward testing anthropological, i.e., cultural,
theories. Unfortunately, the matter does not end there. Others have apparently gained
the impression that "middle-range theory" is equatable with principles of site formation
processes, and that such an equation can somehow constitute an adequate "theoretical"
program for archaeology.
Thomas (1979), for instance, following Binford's (l977a) notion of middle-range
theory, soon introduced this phrase in several places in a textbook on archaeology.
There, "mid-range theory" was presented as bridging arguments between the static
properties of the archaeological record and the interpretations of past dynamics by
archaeologists. Thomas writes that "the function of mid-range theory or bridging argu-
ments is to bridge the gap between the known, observable archaeological contexts and
the unknown, unobservable systemic context. This is why mid-range theory is neces-
sary to provide relevance and meaning to archaeological objects" (Thomas, 1979:398).
The idea that "middle-range theory" deals with statics, dynamics, and site forma-
tion processes, all points made by Binford, opened the door to further ambiguity.
Indeed, the connection of "middle-range theory" with the concept of site formation
processes probably created an all but irresistable tendency to see the two concepts as
synonymous. Concern with developing an adequate materialist epistemology for
Middle-Range Theory in Archaeology 213

archaeology is widely identified in the current literature with Schiffer's (1972, 1976,
1977) concept of site formation processes. Recast in the role of Schifferian "behavioral
archaeology," "mid-range theory" as conceived by Thomas may strike some as a plau-
sible and appealing link between the static record and once-dynamic events. We should
note, however, that Schiffer (1976) clearly intended behavioral archaeology to be a
general methodological program, and in none of his writings has he seen fit to call it
"middle-range theory." The idea of site formation processes was for Schiffer
(1972:156) a branch of archaeological theory, but he has continued to view the former
as a methodological domain (Schiffer, 1983). There seems to be little advantage in
offering "mid-range theory" as a new designation for principles of site formation proc-
esses, and thereby creating the impression that something new in the way of theory
building is at hand.
Later, in a discussion of "middle-range theory," Willey and Sabloff (1980:249-
254) proposed arguments that are compatible to some extent with those of both Thomas
and Binford. Willey and Sabloff conceive of archaeological reasoning as consisting of
lower, middle, and upper levels of theory (Willey and Sabloff, 1980:249). Upper-level
theory is to them apparently the same as "general theory," i.e., highly abstract explana-
tions of the behavior of whole cultural systems. The various "c-" and "e-transforms" of
Schiffer's (1976) behavioral archaeology are regarded as "lower-level theory." Exam-
ples are inferences that rely on lawlike statements concerning the possible relations that
shape, size, and so forth may have to the spatial distributions of objects within an ar-
chaeological site. A specific illustration of such a relationship is provided by the "size
effect" (Baker, 1978), which shows how the size of an artifact affects the probability of
its being exposed, and hence recognized, on the surface of the ground. General argu-
ments of this sort, usually rooted in physical and biological principles, partially explain
the form, content, and distribution of the archaeological record.
For their "middle-level theory," Willey and Sabloff adopt Binford's (1977a)
view that such theory serves as a bridge between the static facts of the archaeological
record, and the behavior that produced them: "In other words, not only must archae-
ologists learn how the archaeological record was formed (an exercise on the lower
theoretical level of transformation processes), but they must be able to explain why a
dynamic system of the past produced the static archaeological record of today (the
middle-range theoretical exercise in the assignment of meaning)" (Willey and Sabloff,
1980:250-251). They go on to say that "middle-range" refers only to this bridging
function and that nothing is implied about the scale or scope of such "theories"
(Willey and Sabloff 1980:251): "Middle-range theories can vary from very specific,
particularistic statements to quite broad ones with general significance, depending on
the kinds of building blocks the archaeologist needs to formulate and test general
archaeological hypotheses."
Despite the attempt by Willey and Sabloff to define a coherent hierarchy of the-
ory building, their scheme contains problems. It is difficult to see significant differ-
ences between so-called lower-level theory and middle-level constructs. They adopt
Binford's view that middle-level constructs ought to provide bridges between statics
and dynamics. The implication is that these constructs somehow exist on a higher theo-
retical level than the lower-level site formation principles. As we have seen, however,
Binford's work embraces the "lower-level" constructs as methodological devices for
transforming statics to dynamics. In reality, Willey and Sabloffs "lower-level theory"
214 L.M. Raab and A.C. Goodyear

and "middle-level theory" both seem to refer to the central methodological problem of
dealing with material traces. It is not clear, however, by what means theory enters their
hierarchy in order to answer questions about cultural behavior, except perhaps at the
most highly abstract levels of "general theory." On close inspection their scheme is
actually polarized between limited and concrete questions of formation principles at
one end and highly abstract theories at the other.
Finally, our published discussion .of "middle-range theory" (Goodyear et aI.,
1978:161) was meant to be Mertonian in character, and stands in contrast to all of the
previous treatments. In our discussion we tried to emphasize development of an axio-
matic body of theory, i.e., hierarchically linked theories of differing scales and compre-
hensiveness that could be made eventually to converge with more general principles.
Our intent was to invite thinking about theory in this "vertical" mode. We were less
concerned with operational or methodological considerations, believing that beyond
methodology, archaeologists face a challenge in developing a more ordered universe of
theoretical constructs if they are to move away from the "flat" view of theory we men-
tioned earlier. Presented only sketchily, and with brief mention of intellectual antece-
dents, our discussion of "middle-range theory" probably shares in any blame to be
assigned for creating ambiguity about this concept.

Potential Problems of "Middle, Range Theory" in Archaeology

We hope that the differences between the original conception of middle-range
theorizing and the conception of "middle-range theory" in most current archaeological
discourses are now apparent, at least in broad outline. We have attempted to show that
while middle-range theory was a heuristic principle that sought to organize a complete
and dynamic program of theory building, archaeological uses of the term have, for the
most part, focused on the methodological problems of dealing with material data. Our
objection to this application in archaeology is not at all founded on rejection of the
search for methodological adequacy. Archaeologists do face distinctive, if not unique,
problems in advancing behavioral research on the basis of material traces. It is abso-
lutely vital that we deal with this problem. The search for solutions to it has been under
way for nearly two decades, however, under a number of headings, such as "archae-
ological theory" and "principles of site formation processes." Again, we see no advan-
tage in now calling such efforts "middle-range theory." But there is a much more
serious problem involved in restricting the concept of middle-range theory to methodo-
logical concerns. Archaeologists need both a more expansive and more organized view
of theory building. A narrow focus on methodology will do little to encourage such
development if archaeologists become convinced that "middle-range theory," as pres-
ently construed, constitutes an adequate approach to theory building in its own right.
This problem is particularly acute if principles of site formation processes are held to
be synonymous with "middle-range theory." One of us (Goodyear, 1977:670) noted the
potentially deadening influence of formation principles when he pointed out that "it is
physically and mentally impossible to study all relationships, since most of them are
probably trivial and theoretically uninteresting. Without significant questions and prob-
lems, behavioral archaeology has a tendency to turn into mechanical archaeology."
More recently, Price (1982:714) makes a similar point, but about the logical depend-
ence of methodological tools on true theory:
Middle,Range Theory in Archaeology 215

In ethnography ... behavior can be observed directly, while in archaeology much

behavior must be reconstructed, indirectly, from its still observable consequences. If
such a step mandates consistent procedures of its own, it does not follow that these
operations constitute theory above the very lowest level, if that. The principle ofhier-
archy in theory building indicates that any such procedures are themselves directed
and their applications guided by middle-level and higher-level theory, however
implicit; the former do not substitute for the latter.

What middle-range theory might look like in archaeology has been obscured by
problems with theory in general. There seems to be confusion over the concepts of gen-
erality and comprehensiveness in developing theories. One can readily appreciate that a
theory is general by definition when it specifies a set of conditions under which some
phenomenon will occur. At the same time, one is interested in estimating the comprehen-
siveness of the theory in question - a factor reflected in the number of phenomena it is
able to predict. This is where we return to our earlier" comment on Willey and Sabloffs
(1980:250-251) three levels of theory building. We argued that their classification, in
effect, dichotomized types of theories into quite high-order ones on the one hand and
ideas about dealing with material records on the other. The examples they give of "mid-
dle-level" and "lower-level" theory in fact both deal with the latter. Willey and Sabloffs
scheme is instructive because it may reflect the underdeveloped sense of theory as con-
sisting of partitive constructs that serve to link methodology with high- order principles.
There is a kind of all-or-nothing view of theory at work here. Both the so-called
low-level and high-level theories are made to carry enormous burdens of generality and
comprehensiveness. In the former, for instance, the artifact size-effect principle men-
tioned earlier is intended to be lawlike and applicable to a great many, if not all,
archaeological cases. Similarly, theories of the rise of the state, an example of high-
level theory, are thought to be highly general and comprehensive within the domain of
their applicability. What strikes one here is the theoretical void between these
extremes. In a sense this is the "flat" view of theory mentioned earlier. We tend to see
all conceptual ideas as having the same burdens of generality and comprehensiveness,
and this view of theory exacts a price. The extreme separation of methodological ideas
from ideas about variability in cultural systems, when coupled with the tacit expecta-
tion that ideas at both ends of the spectrum can carry the whole burden of establishing
scientific principles, results in research where methodology cannot be connected very
well with the discipline'S most fundamental theories. There is, in addition, the frustra-
tion of attempting to test ideas that are so generalized that we joke about them as being
"Mickey Mouse" or lost in the "ozone."
We recognize that in actual research method and theory are intimately linked.
Separation between the two, however, is an interesting gray area thus far only hinted at
in the literature (Goodyear, 1977; Willey and Sabloff, 1980:250-252). There is an
interesting difficulty here in our conflicting knowledge claims about how and why the
archaeological record comes to exist. Willey and Sabloff (1980:250) liken questions of
how the record was formed to "lower-level theory," for instance. Their idea is that, by
means of appropriate methodological tools, we can understand the formation processes
of the record, and can eventually assign certain types of behavior securely to certain
physical remains. This seems to be a process of identification. At some point, however,
behavioral scientists would like to know why the behaviors in question came to be. This
216 L.M. Raab and A.C. Goodyear

constitutes a search for explanations of cultural behavior. It is not always clear, how-
ever, at what point questions about behavior leave the realm of formation processes and
assume the role of cultural theory. In actual research the transition from one to the
other is likely to be seamless, for the process of creative research consists of intuitive
transitions back and forth between methodology and theory. This observation is worth
remembering, because certain research strategies are unlikely to sustain such creativity.
Roughly speaking, when we cease to ask merely what kinds of behaviors can be
linked to certain records and start to ask why the behaviors in question came into exist-
ence, changed, or remained stable, we approach meaningful theory building. To date,
most treatments of site formation processes have been carried out at a mechanical level
and easily related to principles already extant in biology and the physical sciences (e.g.,
Binford and Bertram, 1977; Wood and Johnson, 1978). Such studies have the great
value of stripping away both patterning and "noise" in the archaeological record that
are not the result of human behavior, and perhaps even implicating behaviors that do
correlate with some kinds of records (e.g., Binford, 1979, 1981b). But if we equate for-
mation principles with "middle-range theory," then we must agree with Binford
(1981 a:29) that "middle-range theory plays no role in the explanations offered for the
variability of the subject of interest."
The point here concerns the possible relationship between methodological ade-
quacy - what we have referred to interchangeably as formation principles and archae-
ological theory - and true theory building. It would be a misunderstanding to think
that the two are in some way incompatible. It seems more likely that little progress will
be made toward archaeological explanation until both areas are considerably more
advanced than at present. Even more importantly, progress toward developing adequate
forms of explanation depends on a close integration of method and theory. At the
moment, discussions of "middle-range theory" seem to be obscuring that point, owing
to the semantic confusion surrounding the term in archaeology.

MIDDLE-RANGE THEORY AS EXPLANATION

OF CULTURAL SYSTEMS
Despite the problems outlined above, there should not be undue pessimism about
prospects for developing middle-range theory in archaeology. Brief examples drawn
from current research may illustrate not only some plausible approaches to middle-
range theorizing but also applications of such theory to quite different subject matters.
Good examples are the alternative theoretical models of hunter-gatherer settle-
ment behavior offered by Wiessner and Binford, respectively. Both seek to explain
organizational variability in modem and prehistoric groups. Binford (1977b, 1978,
1980) has chosen to order variation in settlement behavior by emphasizing the avail-
ability in time and space of target natural resources that are ultimately determined by
geographic and climatic variables. He describes this as a theory of adaptation (Binford,
1980). Wiessner, on the other hand, argues that this approach is insufficient to account
for "social relations of production" (Wiessner, 1982). She offers an alternative frame-
work called "risk theory," which is said to hold greater potential for explaining more
kinds of human behavior, such as camp layout, exchange, style, and burial programs.
Binford (1980) is able to explain some of the organizational patterning based on en-
vironmental variables, while Wiessner is potentially able to explain some of the same data,
Middle-Range Theory in Archaeology 217

as well as other patterns, based on risk theory analysis of social strategies. At this point,
neither approach is capable of explaining all aspects of hunter-gatherer behavior that are of
interest to anthropologists_.; Both examples can be thought of as middle-range theoretical
approaches because they possess causal or potentially causal statements about aspects
(e.g., exchange, social organization, logistics systems) of hunter-gatherer cultural systems
that can be explored empirically, using archaeological remains. Yet, from these provisional
and partial treatments we may eventually expect an amalgamation of "special" theories
into a yet more comprehensive theory that explains all of these behaviors and perhaps goes
beyond them. This is the process of theory building as envisioned by Merton in his concept
of middle-range theory. It is provisional, testable, relative in the scale of phenomena to be
explained, axiomatic in that one hypothesis, principle, or model can be subsumed under
another, aimed at explaining cultural behavior and, above all, dynamic.
Another promising area of middle-range theory development takes up the ques-
tion of the origin of social complexity, or socioeconomically stratified societies. Unlike
the theory building in the case of hunter-gatherers, however, which seems to have
developed from a "grounded" perspective in archaeological data, the problem of social
complexity has been approached from a more "deductive" position. The impetus
behind this work seems to lie in highly abstract models derived from engineering and
physics. Shortly after the Second World War, physical scientists became interested in
the theoretical and practical problems involved in storage, and transmission of informa-
tion within communication systems. Some of that work, for instance, was concerned
with artificial intelligence and cybernetic systems (Wiener, 1967), while other parts
focused on the mathematical properties of information transmission (Shannon and
Weaver, 1949). One of the more important consequences of this work was widespread
appreciation of a "cybernetic" or "systems" approach, in which explanations of phe-
nomena in various scientific fields were couched in terms of understanding how a sys-
tem functions by means of communication between its constituent parts. Highly
abstract, yet offering a useful holistic perspective, this basic concept eventually dif-
fused to many areas of research, including anthropology.
Social scientists were quick to realize the potential applications of this construct
to problems of human behavior. Of particular interest were possible relationships
between the complexity of social organization and the structure of communications or
interactions within parts of social systems. More specifically, investigators in several
disciplines became convinced that social organizations tend to be structured by the
fashion in which information is exchanged. In other words, the complexity of social or-
ganization attainable by a social system is dependent on its ability to maintain orderly
communication between its constituent parts (e.g., Beer, 1967; Dubin, 1959).
These brief statements scarcely survey the extent of this pan-social-science body
of theory, but they have provided a broad theoretical foundation for recent archaeologi-
cal efforts to explain the rise of complex societies. Although these efforts have several
labels, perhaps the most useful one is "hierarchy theory" (following Peebles and Kus,
1977, and Johnson, 1982). These archaeologists argue that some cultures eventually
encounter both environmental and social conditions that reward a shift from loosely
structured "horizontal" systems of communication and control to more efficient "hier-
archical" ones (cf. Flannery, 1972).
Peebles and Kus (1977), for example, present a "cybernetics model of chie-
fdoms" aimed at demonstrating that Mississippian sites in the Moundville, Alabama,
218 L.M. Raab and A.C. Goodyear

area are the expression of ranked, or hierarchically organized, cultures. Although part
of this discussion questions the linkage of ethnographic models of chiefdoms with eco-
nomic redistribution mechanisms, a major part of their conceptual base (l977:428-429)
draws on cybernetics and information theory. The thrust of these arguments is that the
power and ritual offices of the chief can be viewed as a controlling node in a two-tiered
hierarchical system. The adaptive advantage of such a system, according to cybernetics
and information theory, would be the greater efficiency of information processing and
decision making over that of an egalitarian society.
Wright and Johnson (1975) provide another informative case. These authors are
concerned with early state formation in southwestern Iran, during which a formal
administrative hierarchy developed. What is not known is how and why administrative
controls shifted from local communities to semiautonomous offices in a developmental
progression that ultimately culminated in the emergence of a bureaucratic hierarchy
within a state-level society. They hold the general theory that a shift toward a hierarchi-
cally organized system was encouraged by gains in information processing and deci-
sion making (Wright and Johnson, 1975:28 5). These are some of the same theoretical
points made by Peebles and Kus (1977).
Wright and Johnson combine information theoretics with concern for other vari-
ables operating in Greater Mesopotamia such as trade, war, and economy and they
implicate the increased efficiency of control hierarchies in the creation of new organ-
izational relationships among those variables. Incidentally, they note that traditional
attempts to account for the rise of the state by single "prime mover" theories can be
avoided in favor of more productive multivariate models.
What is interesting about these cases is that a general theoretical argument has
been converted into a series of derivative propositions regarding a variety of archae-
ological phenomena, including mortuary practices, trade, and political offices. These
works create a body of middle-range constructs in archaeology that serve as stimulus to
the development of new theories, while posing the question whether a more abstract
principle may not subsume all of these efforts. This is nothing less than an invitation to
think in terms of axiomatic theory, i.e., about the logical and empirical consequences of
a hierarchic system of theories dealing with many otherwise disparate data.

CONCLUSIONS

In its original sociological context, middle-range theory was advanced as a basis

for theorizing about the causes of human social behavior. It was advanced by Merton to
counter a tendency for social science research to split into high-level but untestable
theorizing on the one hand, and low-level empirical studies detached from theory on
the other. Productive research was conceived as being empirically based, but moving
through a hierarchy of propositions, which existed at a middle range of abstraction and
provided a crucial linkage between data collection and higher-order theories. Merton's
theory of social-structural anomie provided an example.
In sharp contrast, most current usages of "middle-range theory" in archaeology
are far more narrowly focused on the methodological issue of site formation processes.
This emphasis in fact continues the development of a materialist epistemology for
archaeology begun by certain practitioners of the New Archaeology, most notably Bin-
ford. The fundamental objective of such an epistemology is to ground inferences about
Middle-Range Theory in Archaeology 219

past human behaviors by developing a reliable methodology for differentiating the

effects of behavior from the mimy other causes of the material record.
Objection was raised to the characterization of operations of this sort as "theory."
Statements about site formation processes are methodological in that they allow detec-
tion of patterning in human behavior. Such tools may work well in conjunction with
ideas about the causes of behavioral patterning. Principles of site formation processes,
taken by themselves, often lead to "explanations" that are trivial or easily reduced to
simple biophysical principles. However successful in accounting for the form of the
archaeological record, site formation principles do not tell us how the record came to
exist as a result of the behavior of cultural systems. Explanations of the latter kind will
require formulation and testing of propositions aimed at explaining cultural dynamism.
In a scientific field operating within a largely unorganized universe of theoretical
ideas, development of middle-range theory may provide a useful perspective for look-
ing at the perennial problem of theory building. We argued that some theory construc-
tion of a middle-range sort is already under way in archaeology. The examples
presented, ranging topically from hunter settlement-subsistence to the rise of complex
societies, may show that middle-range theorizing is an approach that is not bound by
any particular subject matter. Equally important, the cases cited attempt to explain not
merely the archaeological record, but the cultural dynamism responsible for that record.

ACKNOWLEDGMENTS

To the many reviewers of this paper we owe a large debt of gratitude. To

unnamed reviewers for American Antiquity we extend our thanks. Reviewers we can
thank by name include Dena Dincauze and her editorial assistant Joseph W. Martin,
Patty Jo Watson, Fred Limp, Lewis Binford, Mike Schiffer, Robin Robertson, James
Bruseth, Randy Moir, Joe Saunders, Stanton Green, Stanley South and Antonio Gil-
man. Though reviewers might not individually agree with every interpretation in the
paper, each offered valuable advice and comment. Of course, any errors of fact or inter-
pretation are ours alone.

REFERENCES
Baker, C.M., 1978, The Size Effect: An Explanation of Variability in Surface Artifact Assemblage Content.
American Antiquity 43:288-293.
Beer, S., 1967, Cybernetics and Management. The English Universities Press, London.
Binford, L.R., 1968a, Archaeological Perspectives. In New Perspectives in Archaeology, edited by S.R. Bin-
ford and L.R. Binford, pp. 5-33. Aldine, Chicago.
Binford, L.R., 1968b, Some Comments on Historical Versus Processual Archaeology. Southwestern Journal of
Anthropology 24:267-275.
Binford, L.R., 1977a, Introduction. In For Theory Building in Archaeology, edited by L.R. Binford, pp. 1-10.
Academic Press, New York.
Binford, L.R., 1977b, Forty-seven Trips. In Stone Tools as Cultural Markers, edited by R.V.S. Wright, pp.
24-36. Australian Institute of Aboriginal Studies, Canberra, Australia.
Binford, L.R., 1978, Nunamiut Ethnoarchaeology. Academic Press, New York.
Binford, L.R., 1979, Organization and Formation Processes: Looking at Curated Technologies. Journal of
Anthropological Research 35(3):255-273.
Binford, L.R., 1980, Willow Smoke and Dogs' Tails: Hunter-Gatherer Settlement Systems and Archaeological
Site Formation. American Antiquity 45:4-20.
220 L.M. Raab and A.C. Goodyear

Binford, L.R., 1981a, Bones: Ancient Men. and Modern Myths. Academic Press, New York.
Binford, L.R., 1981b, Behavioral Archaeology and the "Pompeii Premise." Journal of Anthropological
Research 37:195-208.
Binford, L.R. and Bertram, lB., 1977, Bone Frequencies-and Attritional Processes. In For Theory Building
in Archaeology, edited by L.R. Binford, pp. 77-156. Academic Press, New York.
Binford, S.R. and Binford, L.R., 1968, Archaeological Theory and Method. In New Perspectives in Archaeol-
ogy, edited by S.R. and L.R. Binford, pp. 1-3, Aldine, Chicago.
Caldwell, lR. (ed.), 1966, New Roads to Yesterday: Essays in Archaeology. Basic Books, New York.
Clarke, D.L., 1973, Archaeology: The Loss of Innocence. Antiquity 47:6-18.
Coser, L.A., 1975, Commentaries on Merton's Work. In The Idea of Social Structure. Papers in Honor of
Robert K. Merton, edited by L.A. Coser, pp. 516-522. Harcourt, Brace, 10vanovich, New York.
Dubin, R., 1959, Stability of Human Organizations. In Modern Organization Theory, edited by M. Hains, pp.
218-248. Wiley, New York.
Flannery, K.V., 1972, The Cultural Evolution of Civilizations. Annual Review of Ecology and Systematics
3:399-426.
Freese, L., I 980a, The Problem of Cumulative Knowledge. In Theoretical Methods in Sociology, edited by L.
Freese, pp. 13-69. University of Pittsburgh Press, Pittsburgh.
Flannery, K. V., 1980b, Formal Theorizing. Annual Review of Sociology 6: 187-212.
Fritz, 1.M., 1968, Archaeological Epistemology: Two Views. Unpublished master's thesis, Department of
Anthropology, University of Chicago.
Fritz, 1.M., 1972, Archaeological Systems for Indirect Observation of the Past. In Contemporary Archaeology,
edited by M.P. Leone, pp. 135-157. Southern 111inois University Press, Carbondale.
Fritz, 1.M. and Plog, F.T., 1970, The Nature of Archaeological Reasoning. American Antiquity 35:405-412.
Glaser, B.G. and Strauss, A.L., 1967, The Discovery of Grounded Theory. Aldine, Chicago.
Goodyear, A.C., 1977, Review of Behavioral Archaeology, by M.B. Schiffer. American Antiquity 42:668-671.
Goodyear, A.e., Raab, L.M., and Klinger, T.e., 1978, The Status of Archaeological Research Design in Cul-
tural Resource Management. American Antiquity 43: 159-173.
Hempel, C.G., 1965, Aspects of Scientific Explanation. Free Press, New York.
Hill, IN., 1972, The Methodological Debate in Contemporary Archaeology: A Model. In Models in Archaeology,
edited by D.L. Clark, pp. 61-\07. Methuen. London.
10hnson, G., 1982, Organizational Structure and Scalar Stress. In Theory and Explanation in Archaeology. The
Southampton Colljerence, edited by e. Renfrew, M.J. Rowlands, and B.A. Segraves-Whallon, pp.
389-421. Academic Press, New York.
Kluckhohn, C., 1940, The Conceptual Structure in Middle American Studies. In The Maya and Their Neighbors,
edited by e.L. Hay et al., pp. 41-51. Appleton-Century, New York.
Larson, C.1., 1973, Major Themes in Sociological Theory. David McKay, New York.
MacWhite, E., 1956, On the Interpretation of Archaeological Evidence in Historical and Sociological Terms.
American Anthropologist 58:3-25.
Merton, R.K., 1948, Discussion of Parsons. American Sociological Review 13: 164-168.
Merton, R.K., 1949, Social TheOlY and Social Structure. Free Press, New York.
Merton, R.K., 1967, On Theoretical Sociology. Free Press, New York.
Merton, R.K., 1968, Social Theory and Social Structure (3rd ed.). Free Press, New York.
Morse, D.F., 1977, Dalton Settlement Systems: Reply to Schiffer (2). Plains Anthropologist 22: 149-158.
Mullins, N.C., 1973, Theories and Theory Groups in Contemporary American Sociology. Harper and Row,
New York.
Parsons, T., 1948, The Position of Sociological Theory. American Sociological Review 13: 156-164.
Parsons, T., 1950, The Prospects of Sociological Theory. American Sociological Review 16:3-16.
Parsons, T., 1951, Toward a General Theory of Action. Harvard University Press, Cambridge.
Peebles, C.S. and Kus, S.M., 1977, Some Archaeological Correlates of Ranked Societies. American Antiquity
42:421-448.
Popper, K.R., 1959, The Logic of Scientific Discovery. Basic Books, New York.
Price, B.1., 1982, Cultural Materialism: A Theoretical Review. American Antiquity 47:709-741.
Raab, L.M. and Goodyear, A.C., 1973, On the Value of Middle Range Theory in Archaeological Research
Strategies. Ms. in the possession of the authors.
Reid, J.J., Schiffer, M.B., and Rathje, W.L., 1975, Behavioral Archaeology: Four Strategies. American Anthro-
pologist 77:864-869.
Middle,Range Theory in Archaeology 221

Schiffer, M.B., 1972, Archaeological Context and Systemic Context. American Antiquity 37:156-165.
Schiffer, M.B., 1976, Behavioral Archaeology. Academic Press, New York.
Schiffer, M.B., 1977, Toward a Unified Science of the Cultural Past. In Research Strategies in Historical
Archaeology, edited by S. South, pp. 13-40. Academic Press, New York.
Schiffer, M.B., 1980, Review of For Theory Building in Archaeology: Essays on Faunal Remains, Aquatic
Resources, Spatial Analysis, and Systemic Modeling, edited by L.R. Binford. American Antiquity 45:377.
Schiffer, M.B., 1983, Toward the Identification of Formation Processes. American Antiquity 48:675--706.
Schiffer, M.B., and Rathje, w.L., 1973, Efficient Exploitation of the Archaeological Record: Penetrating Prob-
lems. In Research and Theory in Current Archaeology, edited by C.L. Redman, pp. 169--179. Wiley-
Interscience, New York.
Shannon, C.E. and Weaver, W., 1949, The Mathematical Theory of Communication. University of Illinois
Press, Urbana.
Sullivan, A.P., 1978, Inference and Evidence in Archaeology: A Discussion of the Conceptual Problems. In
Advances in Archaeological Method and Theory, Vol. I, edited by M.B. Schiffer, pp. 183-222. Aca-
demic Press, New York.
Taylor, Walter W., Jr., 1948, A Study of Archaeology, AAA Memoir No. 69, American Anthropological Asso-
ciation, Menasha.
Thomas, D.H., 1979, Archaeology. Holt, Rinehart, and Winston, New York.
Watson, PJ., LeBlanc, S.A., and Redman, C.R., 1971, Explanation in Archeology: An Explicitly Scientific
Approach. Columbia University Press, New York.
Watson, R.A., 1976, Inference in Archaeology. American Antiquity 41 :58-66.
Wiener, N., 1967, The Human Use of Human Beings. Avon, New York.
Wiessner, P., 1982, Beyond Willow Smoke and Dogs' Tails: A Comment on Binford's Analysis of Hunter-
Gatherer Settlement Systems. American Antiquity 47: 171-178.
Willey, G.R., and Sabloff, J.A., 1980, A History of American Archaeology (2nd ed.). W.H. Freeman, San Fran-
cisco.
Wood, W.R., and Johnson, D.L., 1978, A Survey of Disturbance Processes in Archaeological Site Formation. In
Advances in Archaeological Method and Theory, Vol. I, edited by M.B. Schiffer, pp. 315--381. Aca-
demic Press, New York.
Wright, H.T., and Johnson, G.A., 1975, Population, Exchange and Early State Formation in Southwestern Iran.
American Anthropologist 77:267-289.
 Chapter 20

Technology, Theory,
and Analysis
Using Remote Sensing
as a Tool for Middle.-Range
Theory Building in Maritime
and Nautical Archaeology

RICHARD J. ANUSKIEWICZ

INTRODUCTION

This paper is presented to provide some guidance to the archaeologists working under-
water contemplating using geophysical-prospecting data as a tool for middle-range the-
ory building in maritime and nautical archaeology. The methodology and data presented
herein are based on archaeological research conducted on and around St. Catherines Is-
land, Georgia, and focus on the use of magnetometer remote sensing. The research objec-
tives were basically threefold: first, to develop a historic maritime model for St.
Catherines Island; second, to test this model by conducting comprehensive maritime and
nautical archaeological studies of the waterways adjacent to and contiguous to the island;
and third, to develop a correlation between remote sensing signatures and the archae-
ological context for middle-range theory building. This presentation will specifically dis-
cuss the methodological approach used to develop a maritime model and how the model
used nautical archaeology in middle-range theory building for St. Catherines Island.
Source: Underwater Archaeology Proceedings from the Conference on Underwater and Historical Archaeol-
ogy, edited by Donald H. Keith and Toni L. Carrell (1992), pp. 92-99. Reprinted by perrnisison of the Society
for Historical Archaeology, Tucson, Arizona.

223
224 R.J. Anuskiewicz

MIDDLE-RANGE THEORY AND HOW IT WORKS

An example of how middle-range theory building works in historical archaeology

can be drawn from a brief discussion of David Hurst Thomas's recent work (1987:67)
at the Mission Santa Catalina de Guale on St. Catherines Island. Thomas was able to
define linkages between the traditional archaeological concepts of walls, structures, and
features and the way they are perceived remotely by sensors of geophysical prospect-
ing, such as magnetometers (Anuskiewicz, 1989:6).
Further, Thomas defines archaeological concepts as typically abstract categories
employed by the archaeologist. In his research Thomas explored the archaeological
context of 16th and 17th-century Spanish Florida, such as buildings, pits, graves, pali-
sades, bastions, wells, and the like, on St. Catherines Island. Therefore, effective mid-
dle-range theory relates these concepts to an unambiguously defined class, empirically
observed phenomena; in remote sensing these phenomena are the battery of signals and
signatures that derive from nondestructive geophysical prospecting (Thomas, 1987:66;
Anuskiewicz, 1989:7).
Constructing a correlation between remote sensing signatures and the archae-
ological context must be viewed as middle-range theory building in archaeology. This
is simply another way of assigning meaning to our empirical observations (Schiffer,
1976; Garrison and Bray, 1976; Binford, 1977; Thomas and Larsen, 1979; Hayden and
Cannon, 1984; Thomas, 1986:238; Anuskiewicz, 1989:7). Middle-range theory is how
we perceive the past and is quite different from how we explain the past (Binford,
1981 :29; Thomas, 1983a,b).

MARITIME ARCHAEOLOGY

The study of sunken watercraft on St. Catherines Island and their associated eco-
nomic and cultural activities were subsumed under the general headings of historical
and maritime archaeology. Muckelroy specifically defines maritime archaeology as:

The scientific study, through the surviving material evidence, of all aspects of seafar-
ing: Ships, boats, and their equipment; cargoes, or passengers carried on them, and
the economic systems within which they were operating; their officers and crew,
especially utensils and other possessions reflecting their specialized lifestyles.
(Muckelroy, 1978:6)

Maritime activity sites and sunken watercraft are a part of the archaeological
resources. They were part of an active cultural landscape wherein maritime activities,
processes, and the people who participated in them were part of a larger historical cul-
tural context.
In my research, the use of nautical archaeology remote sensing technology to
evaluate the St. Catherines Island maritime landscape and waterways made it possible
to discover and examine specific types of archaeological sites and materials of the his-
toric period. By examining the physical characteristics and recent geological history of
the island's landscape and waterways, this study was able to specify which waterways
were navigable and to what size of vessel, and therefore predict the archaeological
record for shipwrecks (Anuskiewicz, 1989: 11).
Technology, Theory, and Analysis 225

THE ST. CATHERINES ISLAND MARITIME MODEL

The marine model for St. Catherines Island was developed to conceptualize
archaeological expectations and to formulate and test a set of verifiable hypotheses.
The model is represented by six major descriptive and analytical categories for data
input, interpretation, and analysis (Table 1). The categories described in this model
were developed from archaeological information initially derived by Thomas (1987,
1988) from the discovery and issuing studies of the Mission Santa Catalina de Guale.

MARITIME SITE TYPOLOGY

This category identifies the types of specific maritime sites expected to occur on
St. Catherines Island. These included a Spanish mission, a careening site associated
with the mission complex, shipwrecks, and a ballast pile or marine dump site associ-
ated with the Spanish, British, early American, and modern periods of occupation.

CULTURAL PERIODS

This category is pretty much straightforward and represents individual cultural pe-
riods considered in this model. Each cultural period is matched with a maritime site type
to provide specific site-type correlates for each period of the island's maritime history.

SITE FACTOR LOCATIONAL INDICES

This category describes the expected geographic locations of maritime sites

within the physiographic landscape of the island. These indices are specifically corre-
lated with the Maritime Site and Cultural Periods categories to determine the most
probable geographic location at which to search for a specific maritime site type.
The expected Site Locational Indices for shipwrecks on the island consist of the
specific concept of "loss traps" as described by Schiffer (1976). These are specific
areas where vessels are lost as a result of natural phenomena of storms, currents, and
shoals. These loss traps are expected to be concentrated along open and unprotected ar-
eas of the eastern Atlantic coast and beaches of St. Catherines Island. Heavy shoaling
areas located near the northeastern tip of the island and an inlet near the center of the
island are also expected to be additional high-probability areas for loss traps.
It is also expected that there is a direct correlation between the size of the vessel
lost and the size of the loss trap or waterway in which it was lost. For example, the
smaller, meandering creeks found on the island have historically been navigable only to
smaller-type vessels such as canoes, launches, sloops, skiffs, and smaller motor-pow-
ered recreational and sport fishing watercraft.

SITE FORMATION PROCESSES

This category describes how each type of maritime site was formed. For example,
Thomas's recent work (1987) suggests that the remains of the Mission Santa Catalina
de Gaule were formed as the result of the construction, destruction, and reconstruction
sequence of the mission during the Spanish occupation of the island.
Table I. The St. Catherines Island Maritime Model: Chronological periods - Spanish. British. Early American. and Modern ~
0\
Expected site factor Expected site Expected instrumental
Maritime site typology Cultural periods locational indices formation processes Expected archaeological indices indices
3a. Shipwreck, wooden 3. Spanish, British, 3. "Loss traps," harbors, 3. Poor navigation, natural 3. Wooden debris and some 3a. Bouvant hull pattern.
hull (continuous) Early American shoals, ocean-side foundering, storms or metal fittings, nautical and A linear distribution of
beaches, shallows of hurricane, accidental fire, personal implements, marine mUltiple anomaly
navigable rivers and economic abandonment, hardware, ship's cannons, peaks. A dipolar
creeks warfare scuttling, or battle sheathing, and ballast stones signature pattern
damage oriented along the
same heading as the
long axis ofthe hull.
Low to medium
gamma intensity
(10--80 gamma).
3b,c. Shipwreck, wooden 3b. Bouvant hull fracture
hull (discontinuous) pattern. Multiple
dipolar anomalies
radiating upslope and
down cruent from a
more tightly clustered,
high-intensity
anomalies. Low
gamma intensity (5-65
gamma).
3c. Bouvant structure
pattern. Nonclustered
dipolar anomalies of
varying intensities,
scattered unevenly ;:0
c..
across the beach. Low
gamma intensity (5-45 S>
gamma). C
II>

[
r:l
Technology, Theory, and Analysis 227

Site formation for shipwrecks is expected to be caused by poor navigation, natu-

ral foundering, accidental fire, economic abandonment, or mutiny, warfare, scuttling,
or battle damage. The subsequent examination of the archaeological indices of individ-
ual wreck sites is expected to substantiate the site-specific shipwreck formation proc-
ess. If a wreck is located in a "loss trap" and shows evidence of burning, one can
assume that the vessel caught fire and ran aground.

EXPECTED ARCHAEOLOGICAL INDICES

This category represents specific archaeological features and material culture

remains expected to be found in association with a particular maritime site type
(Anuskiewicz, 1982) identified in this model.
Wooden-hulled shipwreck sites associated with the Spanish, British, and early
American period are expected to have the following archaeological indices: wooden
debris, some metal fittings and fasteners, ballast rock, cannons, nautical implements,
marine hardware, and personal items of the crew.
Wrecks of the later American and modern periods are assumed to contain more
metal components associated with later construction techniques and the presence of the
debris from these motor-powered vessels. Modern wrecks are expected to be con-
structed of materials such as steel, aluminum, and fiberglass and to be powered by die-
sel or gasoline engines (Garrison, 1989).

EXPECTED INSTRUMENTAL INDICES

This category is expected to produce correlative remote sensing signatures for

specific maritime features located during this study. These signatures and their verified
archaeological correlates will form the foundation of middle-range theory building for
maritime sites associated with the island.
The Expected Instrumental Indices for a shipwreck associated with the island
should vary with the particular historic period. For example, sailing vessels of the 16th,
17th, and 18th centuries were constructed mainly of wood and had relatively few asso-
ciated ferrous metal fasteners and fittings. Some of these vessels are expected to have
associated cannon, and all vessels should have associated anchor, ground tackle, and
the crew's personal items as part of the ship's archaeological context. It is expected
that historic period shipwrecks reflect specific wreck patterning and correlative mag-
netic signatures. In general, the magnetometer signature should reflect the lack of large
quantities of ferrous components and produce low- to medium-amplitude dipolar
anomalies.
Nineteenth- and twentieth-century ships were, and modern ships are, constructed
of more ferrous and steel components. These wrecks, and their specific wreck pattern-
ing, are expected to produce multi-point-source, dipolar anomalies that are larger,
sharper, and broader at a medium to high amplitude. These signatures would reflect the
amount of iron or steel in the vessel's construction and the associated metal in the
steam, diesel, or gasoline power train components.
It must be noted that the Expected Instrumental Indices represent only a general
range of magnetometer readings for the periods identified in the St. Catherines Island
maritime model. There are multiple variations of these instrumental indices for ship-
228 R.J. Anuskiewicz

wrecks, variations caused by the wreck distribution pattern and the amount of ferrous
material associated with the wreck.

THE ARCHAEOLOGICAL AND MATERIAL CULTURAL

EXPECTATIONS OF SHIPWRECK SITES

Muckelroy wrote extensively on the expectations for shipwreck distributions and

the preservation of specific elements of these sites (1978: 157-225). His fundamental
taxonomy divided shipwrecks into continuous and discontinuous types.
The continuous sites represent shipwrecks that, while undergoing varying levels
of wrecking processes, are still relatively localized in their remains of the hull and any
cargo or ship's fittings. The artifact distributions associated with these wreck have not
been interrupted by sterile areas which do not have to be taken into account during the
interpretation (Muckelroy, 1978:182).
Discontinuous sites are those with elements of the ship widely scattered, with no
single specific locus of the wreck site. These sites have been disturbed by the wrecking
process. There is a total absence of any defining framework, making the reconstruction
of such sites extremely difficult (Muckelroy 1978: 196).
Clausen (1966) and Clausen and Arnold (1975) further discuss the discontinuous
shipwreck patterns for shallow coastal wrecks:

In the majority of cases, vessels of wooden construction lost on active, exposed

coasts tend to break up and disintegrate under the influence of storm-generated
waves and currents. Later, they may also be destroyed by intense attacks of various
marine organisms and the effects of succeeding storms, scattering their components,
ballast, and cargo over an area much larger than the dimensions of the original ship.
(Clausen and Arnold, 1975:80)

Recent research in maritime and nautical archaeology has classified shipwreck

patterning and developed Expected Instrumental Indices for specific wreck patterns,
indices based on studies of the wreck's physical remains. Delgado et al. (1984) and
Gearhart (I 988a,b ) have further refined shipwreck patterning by developing distinctive
site patterns using correlative magnetic signatures. They have designated these specific
site patterns as buoyant hull, buoyant hull fracture, and buoyant structure. These wreck-
type patterns and their correlative magnetic signatures were used as a basis to predict
and develop shipwreck instrumental indices in this maritime model.

DEVELOPMENT OF SHIPWRECK
INSTRUMENTAL INDICES EXPECTATIONS

The Buoyant Hull Site is defined as a continuous wreck site in which the vessel
comes ashore and settles in the sand relatively intact. Gearhart (1988:40-43) reports
that buoyant hull wrecks may differ from one wreck to the next because of materials
used in their construction (e.g., wooden versus steel hulls). His expectations for this
site type are characterized by two important magnetic patterns. First is a linear distribu-
tion of multiple anomaly peaks within the overall pattern produced by the remains of
the intact hull. For a wooden-hulled vessel, one expects the anomaly patterns to exhibit
Technology, Theory, and Analysis 229
a complex, elongated anomaly containing areas of high and low magnetic intensity
within its boundaries. Further, the expectation is that the long axis of the anomaly pat-
tern will be oriented along the same heading as the long axis of the hull. Finally,
Gearhart suggests that the long axis of the anomaly pattern should be oriented parallel
to the surf line because of the tendency ofa drifting hull to turn broadside to the waves.
Buoyant Hull Fracture Sites are discontinuous wreck sites that occur when the
hull of the ship comes ashore intact but breaks up on the beach and is dispersed by the
surf. Therefore, the expected anomaly pattern for this wreck type would consist of mul-
tiple anomalies (i.e., wreck scatter) radiating upslope and downcurrent from an area of
more tightly clustered, high-intensity anomalies (i.e., the area of hull breakup). This
magnetic signature is produced as a result of the distribution of wreck parts (e.g., iron
fittings or magnetic ballast material) that become scattered away from the main body of
the wreck as a result of storms and wave action.
Buoyant Structure Sites are also discontinuous wrecks, formed when a vessel
breaks apart offshore and washes onto the beach in pieces (Gearhart, 1988:40). This
wreck type could leave a trail of wreckage scattered for miles along the beach. The
magnetic signature would depend on the size and quantity of associated ferrous debris
that remained with the floatable materials that came ashore and the areal extent of their
dispersal onto the beach.
This is a very complex wreck type because of the many variables to consider
(e.g., distributional length of the wreck site, construction materials of the ship) when
deriving expectations as to the magnetic signature pattern. Gearhart (1988:43) expects
such sites to consist of nonclustered anomalies· of varying intensities, scattered
unevenly across the beach.

DEVELOPMENT OF SPECIFIC HYPOTHESES

The specific information presented above has provided the necessary archival
data and theoretical concepts to formulate working hypotheses to test the maritime, and
nautical model for St. Catherines Island. From the maritime model, six working
hypotheses were generated with respect to locating maritime sites and shipwrecks asso-
ciated with the island. The hypotheses concerning shipwreck instrumental indices were
easily evaluated using the St. Catherines Island data. The wrecks encountered, and their
magnetic signatures, provided exhaustive data on the variety of expected site types dis-
cussed above. Certainly the data allow us to broadly classify sites based on the instru-
mental data. Evaluation of these hypotheses has led to the recognition of ancillary
hypotheses. For example, the high correlation of wrecks with the "loss traps" of shoals
and bars leads one to pose hypotheses concerning vessel type and size for other areas
and to project probabilities for losses in those areas (Ervan G. Garrison, 1992, personal
communication). Even though several of these hypotheses are germane to this discus-
sion, only one is presented below.

RESULTS OF TESTING THE SPECIFIC HYPOTHESES

What must be noted here is that state-of-the-art proton magnetometer instrumen-
tation and underwater and terrestrial search techniques were used to test the specific
hypothesis.
230 R.J. Anuskiewicz

HYPOTHESIS: Shipwreck sites will be concentrated at "loss traps."

This statement is true. The six shipwrecks inventoried during this research sup-
port the maritime model categories of the Expected Site Factor Locational Indices,
Expected Site Formation Processes, Expected Archaeological Indices, and Expected
Instrumental Indices developed for S1. Catherines Island.
The Expected Site Factor Locational Indices category for shipwrecks in the
model projected that wreck sites would be located in loss traps (Schiffer, 1976). This
study envisioned St. Catherines Island's loss traps at shoals and ocean-side beaches.
All of the shipwreck sites inventoried during this study were located in these areas. The
wrecks exhibited various types of vessel damage prior to, or as a result of, the wrecking
process. From the vessel damage, one could postulate the wrecking process and com-
pare it with the Expected Site Formation Processes identified in the model. The debris
observed at the wreck sites supported the Expected Archaeological Indices for modern-
period shipwrecks. The magnetic signatures recorded for these wrecks also supported
the Expected Instrumental Indices for modem wrecks as described in the model.

MIDDLE.RANGE THEORY BUILDING FOR

ST. CATHERINES ISLAND USING THE MARITIME MODEL

Maritime model building for S1. Catherines Island through the use of archival
research has developed sets of perceived archaeological indices for anticipated mari-
time sites and assigned correlative magnetic signatures to these expected sites. The test-
ing of the maritime model through remote sensing fieldwork has developed sets of
remote sensing signatures that can be used as baseline reference information. These
signatures have produced a framework for middle-range theory building for maritime
sites associated with St. Catherines Island.
The shipwrecks studied and analyzed during the maritime study of St. Catherines
Island are certainly specific to the island. The model building and testing by scientific
inquiry for this study have provided sets of verifiable magnetic signatures. Therefore,
this part of the research has provided the foundation for baseline geophysical signatures
and the foundation for middle-range theory building for modem shipwreck sites associ-
ated with St. Catherines Island and similar physiographic sites throughout the south-
eastern United States.

CONCLUSION

The intent of this paper was to provide some guidance to archaeologists working
underwater contemplating using geophysical-prospecting data, in particular from the
use of the proton magnetometer, as a tool for middle-range theory building in nautical
archaeology. This has been accomplished by presenting the methodological approach
to building a maritime model. Further, this paper has shown that the systematic applica-
tion of the scientific method and state-of-the-art instrumentation, along with a theoreti-
cal model, a sound methodological approach, and systematic field techniques, has
provided the desired results in locating modem shipwreck sites associated with the
island.
Technology, Theory, and Analysis 231

Using instrumental survey techniques in the service of well-defined theoretical

expectations has eliminated many of the areas where many shipwreck sites could not
occur. At a basic level of archaeological inquiry, this study has increased the discovery
probability of locating these particular nautical sites with a continued application of
this methodology.
This statement is particularly true if the specific theoretical expectations are mani-
fest in discrete, archaeological indices. The archaeological indices are either the features
and assemblages themselves or the observable instrumental correlates of these indices
established by the application of middle-range theory building (Anuskiewicz, 1989:228).

REFERENCES
Anuskiewicz, RJ., 1982, Site Reconstruction and Survey Methodology in a Blackwater Environment (ms. on
file). U.S. Army Engineer District, Savannah, Georgia.
Anuskiewicz, R.J., 1985, Three Dimensional Magnetometer Characterization and Marine Magnetic Distribu-
tional Analysis of the Hillsboro Beach Shipwreck: The Search for the Gil Bias, a 19th-Century Spanish
Brig. Paper presented to the Broward County Historical Commission, Marine Archaeology Advisory
Council, Ft. Lauderdale.
Anuskiewicz, R.J. 1989, A Study of Maritime and Nautical Sites Associated with St. Catherines Island, Georgia.
Ph. D. dissertation, University of Tennessee. University Microfilms, Ann Arbor.
Binford, L.R. (ed.), 1977, For Theory Building in Archaeology. Academic Press, New York.
Binford, L.R. (ed.), Bones: Ancient Men and Modern Myths. Academic Press, New York.
Clausen, CJ., and Arnold, 1.B., III, 1975, A Magnetometer Survey with Electronic Positioning Control and
Calculator-plotter System. International Journal of Nautical Archaeology 4(2): 1-88.
Delgado, J.P., Murphy, L., and Kelly, R., 1984, Shipwreck Survey of Portion of Ocean Beach, Golden State
National Recreation Area, San Francisco, California
Garrison, E.G., 1989, A Diachronic Study of Some Historical and Natural Factors Links to Shipwreck Patterns
in the Northern Gulf of Mexico. Paper presented at the First Joint Archaeological Congress, Baltimore.
Garrison, E.G., and Bray, R.T., 1976, Archaeological Investigations at George Washington Carver National
Monument. Report prepared by the American Archaeology Division, Department of Anthropology, Uni-
versity of Missouri. Submitted to National Park Service, Midwest Region.
Gearhart, R.L., 1988a, Cultural Resources Magnetometer Survey and Great Highway/Ocean Beach Seawall
Project, San Francisco, California (ms. on file). San Francisco Clean Water Program, San Francisco.
Gearhart, R.L., 1988b, Marine Magnetometer Survey of a Proposed Sand Borrow and Sand Transfer Site In-
dian Rocks Beach Nourishment Project, Pinellas County, Florida (ms. on file). Pinellas County Board
of Commissioners, Clearwater, Florida.
Hayden, B., and Cannon, A., 1984, The Structure of Material Systems: Ethnoarchaeology in the Maya High-
lands. Society ofAmerican Archaeology Papers No.3.
Muckelroy, K., 1978, Maritime Archaeology. Cambridge University Press, Cambridge.
Schiffer, M.B., 1976, Behavioral Archaeology. Academic Press, New York.
Thomas, D.H., 1983a, The Archaeology of Monitor Valley: I. Epistemology. Anthropology Papers American
Museum of Natural History 58( I).
Thomas, D.H., 1983b, The Archaeology of Monitor Valley: 2. Gatecliff Shelter. Anthropology Papers Ameri-
can Museum of Natural History 59( I).
Thomas, D.H., 1986, Contemporary Hunter-Gatherer Archaeology in America In edited by DJ. Meltzer, D.D.
Fowler, and J.A. Sabloff, American Archaeology Past and Future: A Celebration of the Society for
American Archaeology 1935-1985, pp. 237-276. Smithsonian Institution Press, Washington, DC.
Thomas, D.H., 1987, The Archaeology of Mission Santa Catalina de Guale: I. Search and Discovery. Anthro-
pology Papers American Museum of Natural History 63(2).
Thomas, D.H., 1988, St. Catherines: An Island in Time. Georgia History and Culture Series. Georgia Endowment
for the Humanities, Emory University, Atlanta.
Thomas, D.H., and Larsen, C.S., 1979, The Anthropology of St. Catherines Islands: The Refuge-Deptford
Mortuary Complex. Anthropology Papers American Museum of Natural History 1(56).
 Chapter 21

Considerations for
Research Designs in
Shipwreck Archaeology

DANIEL J. LENIHAN
and LARRY MURPHY

INTRODUCTION

There are two basic questions addressed in this paper: (I) To what degree is the activity
of shipwreck archaeology meaningfully contributing information to the overall disci-
pline of anthropology, and (2) to what degree is the scientific methodology being
employed in shipwreck archaeology as evidenced by explicit designs either inductive
or deductive in nature.
It should be understood that no value judgment is necessarily implied in framing
these questions, viz., negative answers to the above would not de facto mean that
underwater archeology is less valuable than standard terrestrial archaeological pursuits.
There is much justification for carrying out nonanthropologically oriented archaeologi-
cal research for historical purposes, and to whatever degree underwater archaeology is
not towing the line in regards to scientific research design and implementation one may
assume that it is not too much athwart of its terrestrial counterpart. Several years of
reviewing research reports from many institutions in this country will make one realize
that most archaeological research is carried out with the scantiest of explicit research
designs. In most cases, they are afterthoughts tacked on after the fieldwork has been
Source: Underwater Archaeology: The Challenge before Us; The Proceedings of the Twe(fih Conference on
UndelwaterArchaeology, edited by Gordon P. Watts, Jr. (198\), pp. 69--75. Fathom Eight, San Marino, Califor-
nia. Reprinted by permission.

233
2.34 D.J. Lenihan and L. Murphy

conducted and analysis completed. Our land-bound colleagues, however, have been
much more introspective and self-critical on this point and it would well behoove us to
follow their example in that respect.
Our focus in this discussion will be limited to shipwrecks, primarily because this
is a type of archaeological resource that is almost exclusively dealt with by underwater
archaeologists. Often the work conducted on drowned sites, i.e., areas of former dry-
land human activity, is done by individuals who have more traditional archaeological
backgrounds. The degree to which their work does not reflect an explicit problem-ori-
ented research approach is usually not related to the fact that the site is underwater.
Shipwreck archaeology, however, has traditionally been the domain of marine histori-
ans and classical archaeologists who have a much different perspective on the types of
data they wish to collect from material remains in archaeological sites.
After assessing to what degree explicit research designs are used in shipwreck ar-
chaeology and to what degree anthropological problems are addressed, we will present
a brief evaluation of the implications of our conclusions and suggest some ways that
the conceptual approaches to shipwreck archaeology may be strengthened.

THE LITERATURE

The best source for gaining an understanding of how underwater archaeologists

conceptualize shipwrecks as their database is to examine the end product of their
research - published reports, articles, and monographs. The major sources of finished
reports on shipwreck survey and excavation can be found in the International Journal
of Nautical Archaeology, environmental impact assessments, and the mitigation docu-
ments associated with federal construction projects. The hardcovered monographs and
anthologies that exist dealing with underwater archaeological activities are almost
always popularized and rarely deal with more than descriptions of methods and tech-
niques and catalogues of finds. These publications are interesting in their own right and
some also meaningfully relate underwater archaeological data returns to maritime his-
tory, e.g., A History of Seafaring edited by George Bass (1972). It is interesting to note
that the publications that specifically address the issue of "what is underwater archaeol-
ogy" categorically avoid theorc;!tical issues. Much of this situation can be explained by
the fact that underwater archaeologists are confronting an environment that demands
equipment-intensive, technologically novel approaches. Consequently, the tendency is
to fixate on narratives of how we came, saw, and conquered. The one exception to an
almost complete avoidance of theoretical issues in published volumes on shipwreck
archeology is Maritime Archaeology by the late Keith Muckelroy (1978). Muckelroy
developed models for addressing what he felt were the basic conceptual concerns in the
archaeology of shipwrecks, the first being the series of events that took place between
the ship's existence as a functioning entity and the discovery of shipwreck remains by
the archaeologist. He also addresses the problem of what a shipwreck is in terms of an
archaeological entity but recognizes the irony in the fact that the process of deposition,
which is only important in so far as it contributes to our ability to interpret the remains
themselves, is much easier to generalize about.
Muckelroy, using a David Clark-type (1968) analytical approach to considering
the nature of archaeological remains, has modeled a strategy for dealing with what
Schiffer (1976) would term N transforms and C transforms and which Muckelroy
Considerations for Research Designs in Shipwreck Archaeology 235

called extractive filters, i.e., how do you distinguish the effects of natural processes
from cultural ones which act on the final distribution of shipwreck remains? What sort
of effect did movement of the seabed, the process of wrecking, and prior salvage opera-
tions have on the final entity confronted by the archaeologists? Was the lack of certain
perishable' items the result of cultural variables or was it dependent on the rate of
mechanical or biochemical deterioration in the particular site? His models for dealing
with these questions are well thought out and skillfully operationalized. His treatment
of the nature of the "ship" itself includes a breakdown wherein it can be seen as a
"machine" (or perhaps in our words "ship as artifact"), ship as a "basic element of a
mitigating or economic system," and ship as "closed community" which provide some
intriguing possibilities for theory building, but they haven't been as well developed as
his environmental models.
The authors of this paper wish to note at this point that we were not personally
acquainted with Keith Muckelroy before his recent untimely death and do not, there-
fore, share in the sense of personal loss that many who knew him now feel, but as
archaeologists who see the development of a coherent theoretical framework for deal-
ing with shipwrecks of critical importance, we can't help but feel the weight of this
tragedy for our discipline.

IJNA
To get a recent representative sample of the site reports and article in The Inter-
national Journal of Nautical Archaeology and Underwater Exploration, we reviewed
the eight issues in Volumes 7 and 8 covering the years 1978 and 1989. In the 19 "main
articles" as classified in the "notes to authors" and the many other news briefs, notes,
and reviews, the tenn research design was never prominently mentioned. The style of
presentation almost invariably includes a section on historical background material, a
history of the wreck if known or, if unknown, a discussion aimed at trying to name the
wreck or at least a description of the artifacts collected and possibly a discussion of the
ship's architectural attributes. An involved section on methodology and techniques
employed in testing or excavation will either preface or follow the artifacts collected
section. Much of the descriptive material on method and recovered artifacts in UNA
seems to be well organized and interpretation of individual items is often well
researched. What is noticeably missing is even the most basic framework of a research
design to give the data meaning beyond the documentation of one unique historic
event. The presence of a brief section on aims and objectives which occurs in some
reports does not fill this bill. The aims are usually to test or excavate a particular wreck
using a particular technology to confirm its name or origin and collect material wares
in a good state of preservation. This is not a research design and does not develop any
type of explicit reasoning for using any particular research approach.
How the data are collected, it should be remembered, heavily influences what can
be done with them. There is no accountability for the excavation/destruction of a ship-
wreck in these reports to the rest of the profession or to the public that most often col-
lectively owns the resource. It has become apparent that it is a questionable activity
indeed when even the most competent nautical archaeologist disturbs a site that is part
of a rapidly diminishing nonrenewable resource base without fully developing a
research design beforehand.
236 D.J. Lenihan and L. Murphy

In fairness, we should also note in regards to the JJNA that some general analyti-
cal articles on materials recovered from wreck sites are appearing that have very direct
application to the overall discipline of anthropology. An excellent example of this type
of contribution is Colin Martin's analysis of Spanish Armada ceramics (1979).

CULTURAL RESOURCES MANAGEMENT - CRM REPORTS

In the course of reviewing some 15 CRM reports, it was apparent that work to
date generally follows a three-phase approach: the first phase is remote sensing and lit-
erature search; the second phase consists of subsurface testing on anomaly areas and
significance evaluation; and the third phase is mitigation of impacts to the threatened
sites. The majority of the work has been done under the first phase; there have been a
few second-phase surveys in rivers, but very little work done involving the mitigation
phase. No explicit research designs were presented in any of the material reviewed. The
one volume in which research design formulation was attempted was Gagliano's "Cul-
tural Resource Evaluation of the Northern Gulf of Mexico Continental Shelf' (1977).
This effort was a major regional overview and more directed toward isolating sections
within the general region as high-probability areas for site location. It did not offer a
research design for operational applications on a site-specific field level. Although this
volume does offer a framework for developing project-specific research designs, we
have not seen the bulk of the survey work (oil lease block surveys) for which the
Gagliano report was developed. However, it should be noted that in an outline of a
study of these lease block surveys (Ruppe, 1979) commissioned by BLM -OCS to pro-
vide objective professional comment on quality and results of these surveys, no men-
tion was made of research design, questions, or problems - the report was simply a
methodological critique.
Without exception, the CRM surveys speak not to questions of research, but to
concerns of clients, viz., are there any cultural resources in the area, what are they, and
what is the minimum we have to do about them? Gumerman in "The Reconciliation of
Theory and Method in Archaeology" (1977) acknowledges the possibility and desir-
ability of problem-oriented research done within contract requirements. He further
points out that archaeological work done as a result of cultural resource management
concerns may facilitate the utilization and testing of general theoretical frameworks.
It is readily apparent that for the realization of the potential of CRM archaeologi-
cal work to occur, general theoretical and research concerns must be addressed at the
earliest possible point in the CRM survey process. The phased structure of the CRM
work necessitates the statement of these considerations early so as to be available in en-
suring phases to provide guidelines for location models, significance evaluation, and
data collection, as well as for analysis of material and integration of data.

WHAT ARE SOME OF THE ANTHROPOLOGICAL ISSUES

THAT MAY BE ADDRESSED IN SHIPWRECK RESEARCH?

We can see shipwrecks as artifacts much in the sense of Muckelroy's concept

"ship as machine," i.e., a complex creation of humans for the purpose of transportation
over water. In this sense it becomes a stylistic phenomenon what can be seen as a prod-
Considerations for Research Designs in Shipwreck Archaeology 2.37

uct of a particular culture at a particular point in time. Much of the architectural data
accrued by nautical archaeologists in traditional shipwreck work could be used to
develop models for understanding universals in ship design as they relate to various
stimuli, i.e., needs for hydrodynamic efficiency, aesthetic preference, environmental
considerations, economic utility, and so forth. The degree to which maritime social
traits such as highly stratified shipboard status hierarchies manifest themselves in
design and use is another important status hierarchies manifest themselves in design
and use is another important question. How is the above also evidenced in the personal
items of the crew, officers, and passengers? Also, is the heavy tendency toward "male-
ness" in most crews reflected in either ship construction or contents?
Are the location of wrecks anthropologically significant? Basch (1972) and
Muckelroy (1978) both point out that shipwrecks are accidental occurrences and do not
indicate cultural selection. It is our contention, however, that the occurrence of a
number of wrecks in a particular area can tell much about trade patterns and other cul-
ture-specific factors oriented around preference for certain shipping routes, selection of
less seaworthy craft as a result of economic pressures, navigational practices, and
strategies of war. The question of regionality or relevance to regional anthropological
research becomes especially interesting when a relatively contained shipping universe
is considered, e.g., the Great Lakes. Great Lakes shipwrecks are intimately related to
the regional subsistence and communication patterns whereas this may be more diffi-
cult to establish for vessel populations off coastal states. In the latter case, perhaps the
concept of region is less significant than the concept of trade routes as a socioeconomic
entity.

ORGANIZATIONAL PROBLEMS

In the real world of shipwreck archaeology, the commitment to excavation is

developed before the conceptualization of a significant rationale for doing it. This is
sometimes understandable in a CRM milieu, i.e., some sort of mitigation must be carried
out on a site threatened by dredging or other bottom-disturbing construction activities.
This, however, is actually rarely the case; usually an institutional researcher has obtained
money to excavate a shipwreck, then he mayor may not feel obliged to develop a com-
prehensive statement on why he is going to excavate it - but usually not.

WHOLE AS OPPOSED TO PARTIAL SITE EXCAVATIONS

It is important to remember that the complete archaeological excavation of a

wreck site is a completely destructive process. What we have in exchange for the intact
resource is the product of the excavation which includes the artifacts themselves, no
longer in situ, and hopefully some sort of report. It appears that two conditions must
exist to justify such a complete excavation. In regard to the material remains, there
must be readily programmed funds for stabilization, analysis, interpretation, and dis-
play for the benefit of the public. In regard to the data recovery, a report should be gen-
erated that explicitly documents a research approach which was designed to address a
full range of historical and anthropological questions and which can show sufficient
rationale for the decision to fully, rather than partially and selectively, excavate the
site. Using "salvage" or "rescue" as a motive is very shaky grounds unless it can be
238 D.J. Lenihan and L. Murphy

demonstrated that a very real and devastating unavoidable impact was imminent. This
is actually a very rare case in shipwreck archaeology.

SUMMARY AND CONCLUSIONS

Our assessment of the present state of theoretical grounding in shipwreck archae-

ology leaves us with several overriding impressions:
1. There is a major need for an emphasis on theory building by nautical archae-
ologists. Anthropological questions are rarely being asked of the data in ship-
wreck research, and it's well nigh time that they received the attention they
deserve or principal investigators explicitly justify why these issues are not
addressed in their site reports.
2. The key element to enhancing the conceptual integrity of planning in under-
water archaeological activities is explicitness of research design. There is lit-
tle excuse for disturbing sites that are part of a rapidly eroding resource base
without benefit of comprehensive and incisive research designs developed
before, not after, the fact. It is also important, however, that time is not
wasted with a fixation on often specious arguments about new as opposed to
old archaeology and relative merits of inductive as opposed to deductive
approaches. These are appropriate issues for discussion and decision making
by shipwreck archaeologists, but we can't afford the luxury of letting them
turn into grist for the mill of self-righteousness and academic bloodletting
that has characterized dialogue on these points among terrestrial archaeolo-
gists. It must be remembered these problems are peripheral and secondary to
the major problem, which is the fact that research is being conducted without
explicit theoretical and methodological grounding.
If the methodology and theory are almost wholly beneath the level of
consciousness, it is axiomatic that they are inadequate. All aspects of intellec-
tual procedure must be made explicit so that they may be subject to criticism
and empirical testing (Kluckhohn, 1940).
3. A problem orientation (whether inductive or deductive in nature) should be
demonstrated in the explicit research designs that precede fieldwork and
postfield analysis and report writing. These explicit research problems can be
reformulated or new ones generated during any phase in the process, and they
should be clearly reflected in the methodology of testing or excavation. par-
tial site excavation oriented toward answering these specific problems should
always be opted for in lieu of whole site excavation unless there is some le-
gitimate nonscientific rationale for doing so, i.g., an archaeological salvage
situation.
4. It is critical that a conservation ethic be more meaningfully integrated into the
planning and conducting of underwater archaeological investigations. There
should be an increased awareness of the destructive nature of archaeological
testing and excavations, and shipwreck research designs should develop prob-
lem-oriented foci that emphasize maximum data retrieval with minimal site
disturbance. Techniques for removing overburden from major structural com-
ponents of wrecks than recovering or "backfilling" should be perfected.
Considerations for Research Designs in Shipwreck Archaeology 239

REFERENCES
Basch, 1972, Ancient Wrecks and the Archaeology of Ships. International Journal of Nautical Archaeology
and Underwater Exploration I: I-58.
Bass, G., (ed.), 1972, A HistOlY of Seafaring Based on Underwater Archaeology. Walker and Company, New
York.
Clark, D., 1968, Analytical Archaeology. Methuen, London.
Coastal Environments Incorporated, 1977, Cultural Resources Evaluation ofthe Northern Gulfof Mexico Con-
tinental Shelf. Cultural Resource Management Studies, Office of Archaeology and Historic Preserva-
tion, National Park Service, Washington.
Council for Nautical Archaeology, 1972 to Present, The International Journal of Nautical Archaeology. Aca-
demic Press, London.
Gumerman, G., 1977, The Reconciliation of Theory and Method in Archaeology. In Conservation Archaeology: A
Guidefor Cultural Resource Management Studies. AcademkPress, New York,
Kluckhohn, C., 1940, The Conceptual Structure in Middle America Studies. In Contemporary Archaeology:
A Guide to Theory and Condition, pp. 78-84. Southern Illinois University Press, Carbondale.
Martin, C., 1979, Spanish Armada Pottery. The International Journal of Nautical Archaeology 8 (4): 279-302.
Muckelroy, K., 1978, Maritime Archaeology. Cambridge University Press, Cambridge.
Ruppe, R. S., 1979, Final Outline for the Study of Marine Survey Archaeology Commissioned by BLM-OCS.
Unpublished manuscript on file, Bureau of Land Management.
Schiffer, M., 1976, Behavioral Archaeology. Academic Press, New York.
 '.~~.' Bibliography for Part III
~

o i' .
q;,O_' .': .
.j'~l

Mcintosh, Jane, 1986, The Practical Archaeologist, Facts on File, New York.
Orser, Charles E., Jr., and Brian Fagan, 1995, Historical Archaeology, Harper Collins, New York.
Renfrew, Colin and Paul Bahn, 1991, Archaeology. Thames and Hudson, New York.
Thomas, David H., 1974, Predicting the Past. Holt, Rinehart and Winston, New York.
Thomas, David H., 1990, Archaeology. Holt, Rinehart and Winston, New York.

241
 Part IV

Finding Sites
".~i).:
~

':.i
-r/: ..

The methodology for locating sites depends on a variety of factors including research
into natural and cultural history, anticipated site locations and results, as well as the
methods employed to actually conduct the search. In this section, a series of articles are
used to show how archaeologists have worked out ways to predict site locations and
how they conducted their searches. The research and practical methodologies are not
meant to be all-inclusive, rather they should be seen as starting points for organizing a
new search for sites.
A second aspect of searching for maritime sites depends on a careful analysis of
the site formation process. When a living site or a vessel ends up underwater, certain
things have occurred which affect the remains that will be found. Of particular impor-
tance to understanding what might be found are what Muckelroy called "scrambling"
and "mixing" factors, and differential preservation. During loss, pieces of a ship and its
cargo could float away because they were buoyant; other pieces might be salvaged.
Similarly, certain materials survive better than others; in underwater sites, vegetable fi-
bers often do not survive while animal fibers do.
Equipment used for conducting surveys will affect what is found. A magnetome-
ter might be of little use for locating a sewn Arabian vessel but of great utility for find-
ing a 19th century steam boat. Stream bottoms composed of mud could be searched
better with side scan sonar. Depending on the availability of these tools, any person
seeking a site must consider cost, time and estimates of precisely what is being sought.
These techniques were addressed in the foregoing section on Research Design. In this
section, research design is applied to field situations and research to locate sites.

243
 Part IV.A

'.~~.' Site Location Factors

~

o i' .
q;,O_" .': .
.j'~l

The bulk of Ruppe's article (Chapter 22) concerns itself with the process of ocean-level
change as complicated by tectonic shifts. Both ocean-level and land-level change
throughout geologic time, facts that have significant implications for the present loca-
tions of archaeological sites. Using examples from the East Coast, Ruppe outlines the
complexities involved in analysis, demonstrating the need for careful consideration of
hydrologic changes when studying inundated sites.
Referring to aboriginal materials located in offshore areas, in Chapter 23, Muche
outlines a systematic scheme for selecting habitation areas of high probability. Factors
such as water supply, food, and shelter can be correlated with drowned beaches, stream-
beds, and arroyos. This should be accomplished in an overall and systematic manner, and
then areas of high probability can be selected for survey.

245
 Chapter 22

Sea Level Change as

".~i).:
~

a Variable in Colonial
':.i
-r/: .. American Archaeology

REYNOLD J. RUPPE

The genesis of this paper was a series of observations I made at a few Colonial archae-
ological sites I visited along the eastern coast of the United States. In several cases
there was evidence that sea level had risen since the sites were founded. Those sites
dated from the early 17th century to the time of the American Revolution and in every
case are 30 cm or more below high tide at present. I believe that we will find other
early Colonial sites as well as sites of the preceding Exploration period underwater in
locally favorable situations. Those favorable locations are estuaries, lagoons, and the
tidal portions of rivers. Other sites are reported to occur on the oceanfront but they
probably will be difficult to deal with because of alteration of the beaches by storms,
sedimentation, and erosion by longshore currents.
Before we can discuss the archaeology we must review some marine geology that
has a bearing on the problems of sea level change. From a geological viewpoint we are
dealing with a very recent time period and we can confine our considerations to the
problem of sea level rise in the past several hundred years. The problem is difficult
because of a complex set of variables and relatively little knowledge of the processes
involved. Sea level does not remain stationary. It changes as a result of climatic change
on a worldwide scale and has been most dramatically changed as a result of the growth
and shrinkage of ice sheets in the geologically recent past.
Source: In the Realms of Gold; The Proceedings of the Tenth Conference on UndeIWater Archaeology, edited
by Wilburn A. Cockrell (1981), pp. 195-20 I, Fathom Eight, San Marino, California. Reprinted by permission.

247
248 R.J. Ruppe

Present thinking by marine geologists on the problem of recent sea level change
is made more difficult by the inability of the geologists to date accurately recent minor
sea level oscillations. The prevailing opinion can be termed a paradigm in Kuhn's
sense (1962) and holds that there has been relatively little rise in sea level since 6000
B.C. and the rise is graphed as a smooth curve (Shepard and Curray, 1967; Guilcher,
1969). The use of the smooth curve as a model inhibits our thinking about sea level
change in the recent past. A number of estimates of recent sea level rise have been
made in terms of rate per century. One such estimate is 1 mm a year or 1 inch every 25
years (Morner, 1973). That estimate calculates the rise of sea level to be 40 cm in the
400 years since the beginning of the Colonial period in North America.
A major difficulty in dealing with sea level rise is the fact that two distinctly differ-
ent causal factors are involved in the change of sea level relative to land level. The factor
of most concern to us is eustasy, the rise of sea level caused by melting ice sheets.
Eustatic change is worldwide but its effects will be altered locally by isostatic land
movements. Isostasy is the change in land elevations brought about by several causes, of
which the most important to us is crustal warping resulting from the weight of ice sheets
on the land. Tectonic movements are another cause of isostatic change of land elevations.
The importance of isostatic change to our problem is that it is a difficult process in terms
of regularity of earth movement. Consequently, in areas of considerable tectonic activity
such as the Mediterranean Sea we are unable to identify sea-land changes that are appar-
ent in many classical cities. Flemming illustrates the problem nicely when he describes
the differences in the relationship of land levels and sea levels that are dependent on
local tectonic movements as opposed to sea level changes (Flemming, 1971).
The foregoing discussion suggests that sea level changes occur and that they are
made more complex by tectonic movements in specific locations. In an effort to avoid
the confusion of variable causal factors it seems sensible to test our ideas in areas that
are geologically stable so as to assess the effects of sea level alone. One such area of
geological stability is southern Florida (Brooks, 1973). But not all of the coastline of
the Gulf of Mexico is stable, however, because land subsidence has occurred as a result
of the weight of sediments in the delta of the Mississippi River. There is a local down-
warping of the coast for several hundred kilometers on each side of the delta. The
coastline of the Southeast in general is thought to be relatively stable as well. The
important point is that in areas where crustal stability is suspect, we must be careful in
our formulations of relationships between natural and cultural phenomena.
This paper is planned to accomplish two things. First, I hope that the audience
will be alerted to the possibility of sites of the Colonial and earlier periods along the
coasts; and second, I hope that some of you will be able to give me some clues about
the location of sites under water. It is obvious that if sites are drowned and not evident,
they cannot be known. But some of you may have documentary hints to sites that are
either lost or do not fit the written descriptions of them. A case in point is Jamestown
where the original location is thought to be underwater in the James River. I am guess-
ing a little, but I have a hunch that the rangers suggest erosion of the riverbank as the
reason for the supposed inundation. If the latter cause is correct, it would be useless to
attempt to locate the original remains, but if sea level rise is the causal factor, then one
would expect to find archaeological evidence for the original site.
An early claim for inundation of a historical site in the United States was made
by Elso Barghoorn in 1953 (Science Vol. 117, May 29, 1953) when he reported on the
Sea Level Change as a Variable in Colonial American Archaeology 249

drowned condition of the early Colonial Iron Works at Saugus, Massachusetts. He

states that "well preserved colonial wooden structures have been uncovered at levels
now daily inundated by high tides. The arrangement of these structures, which include
remains of water wheels, a timbered waterway in its original position, and remnants of
a dock and wharf, is such that their intended function would be very inefficient if not
impossible under present tidal conditions." He goes on to say that "the entire area of
early Iron Works development has been affected since 1650 by an increase in the
height of tide in the Saugus estuary of approximately 2 and a half to 3 feet."
The questions posed by the Saugus Iron Works are of two kinds; whether the
indicated sea level change was caused by eustatic or isostatic change and whether or
not errors were made in the assessment of the archaeological and geological contexts.
The article indicates clearly that the archaeological and geological contexts were prop-
erly interpreted and we are left with a need to determine the amount of sea level or land
level change. Because New England was very close to the Wisconsin ice sheets we
must consider the possibility that some of the change of sea level relative to land level
was caused by downwarping of the continental margin resulting from reduction of the
load of ice on land.
The warping of the earth's crust caused by tremendous pressures placed on the
surface by the weight of the ice is a phenomenon well known to Pleistocene geologists.
When one considers that the ice sheets were between 1 and 2 km thick, it is easy to see
how the center of the continent would be depressed under the load. The continental
margins were under less of a load, and either were not depressed or were depressed to a
lesser extent than the interior. When the ice sheet melted, the release of the ice load
caused the center of the continent to rebound upward and, at the same time, the margins
of the continent were warped downward. The result of those actions, for our purposes,
is a downward movement of the coastline which would cause water level to appear to
be rising. Because sea level was also rising at the same time, the effect was to increase
the amount of change. That is one of the reasons why it is difficult to understand the
problems in areas where isostatic changes have occurred. The reason why we must be
concerned with the problem is that in the case of the Saugus Iron Works, there is indi-
cated a greater than expectable sea level rise. As we shall see, the other examples of
Colonial site inundation are farther south and do not show as great a sea level change
as the 2 Yz- to 3-ft change at Saugus.
Another Colonial site that appears to have been inundated is Flowerdew Planta-
tion, about 20 miles upstream on the James River in Virginia. The founding of Flower-
dew dates to 1620, and the early structures were built on the bank of the river.
According to Norman Barka, the project director at Flowerdew, there is documentary
evidence of a dock, stockade, warehouse, and other structures close to the river. In
excavations close to the river, the high tides inundated the archaeologists' trenches,
indicating that there has been a change of water level in the area. The James River is
tidal at that location, consequently, changed water level is the result of sea level
change. It is not yet known what the exact relationship of high tides to the lowest
archaeological remains is, but there must have been at least a 30- to 50-cm sea level
rise. In this regard, the situation at Flowerdew probably is identical with that at
Jamestown, 18 miles downriver.
A third example is Fort Moultrie, at Charleston, South Carolina, which has been
studied by Stanley South, and reported in "Palmetto Parapets" (1974). South reports a
250 R.J. Ruppe

1-ft rise of water table between the time of the building of Fort Moultrie in 1776 and the
present. The fort is located on the shore of the outer portion of the Charleston harbor and
the water table should be almost the same, if not the same, as sea level. South found a
large number of archaeological features beneath the water table and states, "As the water
table was reached, from three to four feet in depth below present surface, pumps were re-
quired to lower the water table enough to allow for reading of archaeological layers and
features." Many features, such as a large foundation timber, could not have been placed
deliberately below the water table and we are left with the conclusion that the original
construction was accomplished before water level had risen. Our conclusion, therefore, is
that sea level has risen approximately 1 ft since 1776 on the South Carolina coast.
A fourth example is the site of Londontown, a Colonial harbor town on South
River in Maryland. In a brief reconnaissance, Shomette (1978) discovered two exam-
ples of brick foundations under water off the present shoreline. He does not report the
foundation depths, but they are close to the shore and must be in shallow water. The
site was founded in 1650, and by the 1740s was second only to Annapolis as a mari-
time and commercial center; but by the end of the Revolution, was all but abandoned.
The major building activity had apparently been completed by 1690, and it is possible
that the underwater brick foundations date from that time. We need more information
on the site before we can determine whether or not the underwater features are the
product of recent erosion or of sea level rise. The time span would suggest that sea
level rise is responsible for the inundation, but conclusions must await further study.
Another possible example of historic site inundation is Fort St. Marks on the St.
Marks River in northern Florida (Fairbanks, 1964). Salvage work in preparation for
dredging uncovered "two series of wooden posts below waterline" which were identi-
fied and mapped. Fairbanks thinks that "one line of posts, just upstream from the point
where the Walkulla enters, is possibly part of the southern bastion of the first Ft. St.
Marks. The posts are very small and irregularly set and could well form part of a rather
hastily erected wooden fort. The second series of posts is somewhat further upstream
on the St. Marks side and represents posts of several sizes, some round, some squared.
This is the approximate location of a dock extending out in the stream." The fact that
Fairbanks draws a distinction between the two sets of posts suggests that his interpreta-
tion could be correct.
Except for Barghoorn's observations of the stratigraphic situation at the Saugus
Iron Works and South's stratigraphic sections at Fort Moultrie, we are dealing with sets
of observations that are not supported by stratigraphic evidence. While inundation has
protected the sites, it also has effectively masked the context of the archaeological and
geological material. The problem is how best to investigate the phenomena under
water. Subbottom pro filers will not work in such shallow water and while magnetome-
ters and metal detectors could indicate metals, they tell us nothing about stratigraphy. It
is possible to excavate under water and to determine both cultural and natural levels
with the same precision as on land (Ruppe, 1979). It would also make sense to con-
struct cofferdams around inundated sites and pump out the water. While expensive,
cofferdams are a practical solution particularly in the case of sites as valuable to
American history as Jamestown. But before we can invest in cofferdams we need a
technique that will allow us to identify and interpret the archaeological contexts.
From the above examples, it should be possible to support the notion that in the
coastal areas of the United States, many archaeological sites of the historic period known
Sea Level Change as a Variable in Colonial American Archaeology 251

presently only from the literature, probably are drowned in 30-50 cm of water. Given the
large number of Spanish, French, and English sites listed as being founded in the 16th and
early 17th centuries in the Southeast, it would be strange indeed ifthere were no evidences
of the occupations. It remains, however, to locate those archaeological evidences. It is
hoped that the ideas presented in this paper will stimulate workers who are doing research
on historical sites on seacoasts to pay special attention to documentary evidence that might
describe town areas that are under water at the present time. It might also be possible that
what appear to be vague shadows in shallow water on aerial photographs may be drowned
buildings or dock areas. Shomette found that to be the case at Londontown.
Care must be exercised in the investigation of underwater features, however,
because it is possible that an archaeological structure that appears to have been inun-
dated by rising sea level might have been built on dry land. Later erosion of the shore-
line could have the effect of washing the sands or sediments away, leaving a site
surrounded and underwater. If the site originally had been excavated down to the water
table, later erosion might have created what appears at the present time to be an exam-
ple of inundation. Extreme care must be exercised in the observation of archaeological
levels, features, old surfaces and the hydraulic history of the local area lest the archae-
ologist make incorrect interpretations of his material.
Despite the disclaimer in the preceding paragraph, I am convinced that it is possi-
ble to locate site features that have been hidden by rising sea level. There are a number
of techniques currently used in underwater archaeology that can aid in the location of
underwater remains. A magnetometer is capable of detecting relatively small magnetic
anomalies caused by buried articles, and the instrument can be used in shallow water. It
is no more difficult to pass the sensing unit of the magnetometer over the bottom of a
body of water than it is to do it on dry land. Another instrument that may some day be
useful in shallow water is the subbottom profiler. Unfortunately, at the present time the
instrument cannot operate in water depths less than about 10 ft. Until last year, the
depth limit was 20 ft; but this past year, Martin Klein, a designer and manufacturer of
the instrument, has succeeded in reducing that depth of operation to lOft.
An instrument that sees through the ground must be high on the wishlist of every
archaeologist. Magnetometers and subbottom profilers, to say nothing of plain, old
metal detectors, offer the possibility of site and feature locations both on dry land and
under water. Those instruments, documents, and good archaeological field skills should
be able to pay dividends in the study of drowned Colonial sites.

REFERENCES
Brooks, H.K., 1973, Geological Oceanography. In A Summary of Knowledge of the Eastern Gulf of Mexico, II
E I-II E 27, University System of Florida, Institute of Oceanography.
Fairbanks, C.H., 1964, Underwater Historic Sites on St. Marks River. Florida Anthropologist, 17(2):44-49.
Flemming, N.C., 1971, Cities in the Sea: An Explanation of Ancient Cities Beneath the Mediterranean. Dou-
bleday, New York.
Guilcher, A., 1969, Pleistocene and Holocene Sea Level Changes. Earth Science Reviews, 5(2):69-97.
Kuhn, T.S., 1962, The Structure ofScientific Revolutions. University of Chicago Press, Phoenix Books, Chicago.
Momer, N.-A., 1973, Eustatic Changes During the Last 300 Years. Palaeogeography, Palaeoclimatology,
Palaeoecology, 13:1-14.
Ruppe, RJ., 1979, The Archaeology of Drowned Terrestrial Sites: A Preliminary Report. Bulletin No.6,
Bureau of Historic Sites and Properties, Department of State, Tallahassee, Florida (in press).
252 R.J. Ruppe

Shepard, F. and Curray, J.R., 1967, Carbon 14 Determination of Sea Level Changes in Stable Areas. Progress
in Oceanography 4:283-291.
Shomette, D.G., 1978, Londontown: The Reconnaissance of a 17th-18th Century Tidewater Riverport Complex.
Beneath the Waters of Time: The Proceedings ofthe Ninth Conference on Underwater Archaeology. Texas
Antiquities Committee Publication No.6, pp. 167-174. Austin.
South, S., 1974, Palmetto Parapets, Anthropological Studies. In Occasional Papers ofthe Institute ofArchaeology
and Anthropology. The University of South Carolina, Columbia.
 Chapter 23

Site Location Factors

JAMES F. MUCHE

Until only recently it was believed that any aboriginal material located in offshore
areas must be the result of random loss, beach erosion, ceremonial deposition, etc.,
rather than the presence of an actual inundated site. No systematic attempts, therefore,
were made to locate sites of this type because it was felt that they simply did not exist.
Discoveries in Florida and California have now disproved this and a systematic
approach to site location is needed.
Unlike shipwreck sites, one cannot just consult historic records or depend solely on
indications provided by remote sensing to locate potentially sensitive areas. Anew set of
interpretive criteria is therefore needed. Certain factors have always governed a people's
choice for living space: availability of fresh water, shelter from the elements, and suffi-
cient food supply. With these in mind, it was felt that an extrapolation of conditions
could be made to preinundation times and sites located in that manner. Several factors
were to be considered, including: (l) known land sites, (2) geographical features, (3) iso-
lated find reports, (4) water depth-time correlation, (5) indications of a preexisting water
supply, (6) early ethnographies and historic documentation (primarily Spanish records in
this case; even though these record postinundation times it was felt that descriptions of
areas prior to urbanization and increased European influence would aid in the extrapola-
tion), (7) paleontological evidence, and (8) estimates of climatological data relating to
prehistoric times. In order to satisfy these criteria, the initial steps consisted of basic map
work. Two areas were chosen for the test: the Santa Monica Bay region, including Point
Dume, and the Palos Verdes Peninsula each exhibiting a high concentration of land sites
and, in the case of Santa Monica Bay, a large existing fresh water supply.
Source: In the Realms of Gold; The Proceedings of the Tenth Conference on Underwater Archaeology, edited
by Wilburn A. Cockrell (1981), pp. 137-139. Fathom Eight, San Marino, California. Reprinted by permission.

253
254 J.F. Muche

The first step was to plot all known coastal sites in these areas and note high den-
sity locales. This was followed by a similar plotting of known submerged finds. These
data were then compared and site specifics examined. Several areas with both land and
marine finds were then assigned a probability factor of one to ten based on their meet-
ing of a set of predetermined factors. To exemplify this the two extremes will be dis-
cussed. A low probability factor, a rating of three or less, would be assigned to areas
where known marine finds occurred less than 100 yards from a known land site in an
area where a high amount of erosion and runoff were known to occur. Also, low prob-
ability would be assigned where marine finds occurred at far too great a depth to be
placed within the water depth-time curve for proper artifact time period correlation to
be made.
A high probability factor, seven or more, would be assigned in areas where ero-
sion was not a significant factor, water depth-time correlation fit well (this being that
offshore material was older than that located in the adjoining land site and exhibited a
date that would correspond to a believed time of lower water level), and sufficient fresh
water and natural resources being present to warrant the view that they would also have
been present prior to inundation. These were, in most instances, recorded in historical
area descriptions. In this manner, 11 out of a possible 34 locations, or 32.4%, were
deemed as high probability areas.
This finding in itself is not conclusive as accidental or ceremonial deposition
could still explain the occurrence of the known finds. Seabottom topography was then
taken into account. This was approached in two manners: (I) the inspection of avail-
able remote sensing data and (2) the preparation of seabottom profiles. The latter were
then broken into date lines using the water depth-time curves.
The remote sensing material provided the location of pertinent data such as sub-
merged streambeds, drowned beaches and arroyos, and possible early dry plateau and
peninsular areas. A correlation of these data produced seven more high probability ar-
eas. These were determined by a comparison of depth versus time versus geologic suit-
ability and then correlated to known settlement patterns.
At this point it was felt that enough static data had been obtained to warrant field
testing of those areas designated as high probability. Short duration (less than I week)
visual surveys were carried out at the 18 indicated locations. At 9 of the initial 11 loca-
tions material was noted. These consisted of 12 mortars or bowls, eight pestles, three
points, one drill, three scrapers, and nine shell beads. Midden material was in evidence
at each of these locations.
Of the seven additional sites, four contained aboriginal material. This included
nine mortars or bowls, 11 pestles, four points, two scrapers, one steatite boat model,
and several of what appear to be bedrock mortars. Midden material was once again in
evidence. Dates for the 13 sites as determined by correlation to land finds and the water
depth-time curves ranged between 3000 and 11,000 years B.P. with all artifactual
remains fitting well within accepted land dates and proper curve dates. Curve dates
were accepted at ±1000 years to account for the unstable geological conditions present
at several points within the test area. (Detailed survey reports of the areas will follow in
subsequent publications.)
This method, while far from perfected, did yield a 72% success factor and would
appear to be a valid method of site location, at least for two areas in Southern Califor-
nia. To be truly considered valid, several more areas must be tested in California and,
Site Location Factors 255

ideally, in other areas of the country. It is also felt that while it is possible and also
much more cost effective to work with existing remote sensing data, the variation in
these data, due to the purposes for which they were originally intended, does present a
handicap. Ideally, the surveys should be carried out as needed to ensure uniformity of
data throughout the test area. It is felt that the utilization of a procedure such as this
might well provide the means for the recovery of data that might otherwise be over-
looked or lost in some manner.
 Additional Bibliography
for Part IV.A

Babits, L.E., Annalies C. Kjorness and Jeffrey Morris, 1995, Survey of the North Shore Pamlico River: Bath
Creek to Wade's Point. Ms. on file, Program in Maritime Studies, East Carolina University, Greenville,
North Carolina.
Babits, L.E., and Annalies C. Kjorness, 1995, Final Report on an archaeological Survey of the Western Shore
of the Pungo River from Wades Point to Woodstock Point. Ms. on file, Program in Maritime Studies,
East Carolina University, Greenville, North Carolina.
Bullen, Ripley P., 1969, Further Comments on Emery and Edwards' Archaeological Potential of the Atlantic
Shelf. American Antiquity 34 (3):331-332.
Carrell, Toni, Sandra Rayl and Daniel Lenihan, 1976, The Effects of Freshwater Inundation ofArchaeological
Sites Through Reservoir Construction: A Literature Search. National Park Service, Cultural Resources
Management Division (Archaeology), Washington, D.C.
Emery, Kenneth 0 ., and R.L. Edwards, 1966, Archaeological Potential of the Atlantic Continental Shelf.
American Antiquity 31 (5)(part 1):733-37.
Goggin, John M, 1960, Underwater Archaeology: Its Nature and Limitations. American Antiquity 35 (3):
348-354.
Jameson, Michael H., 1976, The Excavation of a Drowned Greek Temple. Avenues to A ntiquity Readings from
Scientific American. pp. 289--298. W.H. Freeman and Company, San Francisco, California. (originally
pubJ. in 1974).
Moriarity, James R., 1964, The Use of Oceanography in the Solution of Problems In a Submarine Archaeological
Site. Papers In Marine Geology edited by Robert L. Miller, pp. 511-522. MacMillan, New York.
Ruppe, Reynold 1., 1980, The Archaeology of Drowned Terrestrial Sites: A Preliminary Report. Bureau of His-
toric Sites and Properties Bulletin Number 6, pp. 35-45 . Department of State, Tallahassee Florida.
Salwen, Bert, 1967, A Comment on Emery and Edwards' Archaeological Potential of the Atlantic Continental
Shelf. American Antiquity 32 (4):546-47.
Scholl, David w., and M. Stuiver, 1967, Recent Submergence of Southern Florida: A Comparison with Adjacent
Coasts and Other Eustatic Data. Geological Society ofAmerica 78 (April): 437-545.
Leone, Mark P. (ed.), 1972, Contemporary Archaeology, Southern Illinois University Press, Carbondale, Illinois.

257
 Part IV.B

Site Formation Processes

Stephenson (Chapter 24) applies stream formation geomorphology to a wrecksite, mak-

ing possible some speculation concerning site formation. Particularly important in
Stephenson's view is the comparison of the river flow before and after upstream reser-
voir control. A model of how the ship grounded, moved through time, and settled in
place is related to factors such as stream cutting and episodic flooding.
Muckelroy (Chapter 25) gives the basic outline of the wrecking process, with
examples from several sites. His scheme is important for its development of key con-
cepts, called scrambling and extraction filters, which are elucidated here. Muckelroy's
discussion of the site formation processes is a key to understanding what might be
found in any submerged site.
Gearhart (Chapter 26) outlines a scheme for classification of remains and also
gives an important glimpse at details of the wrecking process as it applies to an
extremely dynamic environment, expanding on Muckelroy's work.
In Chapter 27, quantitative methods, along with a number of factors (shipping
routes, port locations, shoals, reefs, currents, winds, and hurricane paths) are combined
in a systematic study of shipwreck patterns. The whole is examined within distinct time
periods, producing chronological patterns. Spatial and statistical analyses such as this
allows correlation of wreck patterns to changes in economic activity and population.

259
 Chapter 24

Physical Processes at
the CSS Chattahoochee
Wreck Site

RICHARD STEPHENSON

INTRODUCTION

During the summer of 1984, the remains of the wooden gunboat CSS Chattahoochee
were investigated. The research was funded by the Confederate Naval Museum in
Columbus, Georgia; on-site operations were conducted by the East Carolina University
Program in Maritime History .and Underwater Research. The goals of the project were
fourfold: (I) to locate the wreck, (2) to at least partially excavate the remains, (3) to
map the wreck, and (4) to help develop a management strategy for the wreck site. To
accomplish these goals it was necessary to ascertain the physical processes and the
environmental conditions of the site. The purpose here is to describe the physical proc-
esses related to the wreck. First, a description of the historical aspects of the demise of
the CSS Chattahoochee will be reviewed. Then, a description of the geographical set-
ting of the area will be presented. These aspects of the wreck of the gunboat serve as
the basis for the analysis of site formation processes which are presented in three parts:
(1) the early fluvial processes, (2) the more recent anthropo-limnologic processes, and
(3) the observed conditions during the on-site investigation. Finally, some management
alternatives are suggested in view of the physical processes and the environmental set-
ting involved.
Source: Proceedings of the Sixteenth Conference on Underwater Archaeology, edited by Paul F. Johnston
(1985), pp. 97-99. Reprinted by permission of the Society for Historical Archaeology, Tucson, Arizona.

261
262 R. Stephenson

HISTORICAL NOTES

From historical accounts one might believe that the gunboat CSS Chattahoochee
was a jinxed ship. The construction of the gunboat began on October 19,1861, at Saf-
fold, Georgia, 33 miles upstream from Chattahoochee, Florida (Turner, 1975). Saffold
was a small town with boatbuilding facilities which were being consolidated into a
Confederate Navy Yard. The construction served the policies of Confederate Navy sec-
retary Stephen R. Mallory, in that: (1) good stands of live oak and pine were nearby,
(2) the contracting was done with local private industry, and (3) by building and using
fast gunboats, the raising of the blockade of Apalachicola and other southern ports
could be achieved. Apalachicola, the seaport for the Apalachicola River system, which
includes the Chattahoochee River, was blockaded by the Union Navy on June 11, 1861
(Turner, 1975). The blockade was not effective at first with only one screw steamer, the
USS Montgomery, on duty. Several other Union ships were assigned to the area later,
and the Confederates responded with defensive ploys such as obstructions in channels
and fortifications on the high ground. The gunboats, however, were to be the offensive
weapons for the raising of the blockade (Turner, 1975).
Because of delays in gunboat construction and an increase in the Union squadron
operating along the Gulf coast, Apalachicola fell into Union hands early in 1862
(Turner, 1975). Fortunately for the Confederates, the Union flotilla, except for small
ship's boats, could not navigate upstream. The Confederates also consolidated their
defenses at Chattahoochee, 100 miles upstream at the confluence of the Chattahoochee
and Flint Rivers. The Chattahoochee River had numerous plantations and several small
communities along its course which were engaged in the war effort. While the produc-
ers could not export their goods through Apalachicola, they were able to transport their
goods upstream to Columbus located at the fall line. Columbus was an extremely
important industrial center. This was related, in part, to the Columbus Iron Works, later
known as the Columbus Naval Iron Works. Further, Columbus was linked to
Montgomery and Atlanta by rail.
While the Chattahoochee was being built at Saffold, the engines and machinery
were being shipped to or made at Columbus. The gunboat itself was slightly smaller
than a first class gunboat at 130 ft in length, 30 ft at the beam, and presumably a depth
of 5 ft forward and 8 ft aft. It was a three-masted schooner with two propellers driven
by two steam engines having a 36-inch bore and a 24-inch stroke. It was to achieve a
speed of 12 knots. The gunboat's armament included six deck guns; four broadside and
one each fore and aft. The complement consisted of about 120 men. The vessel was
painted the usual color (black) for blockade runners, and the men assembled on board
for duty (Turner, 1975). The Chattahoochee's first service was to aid in the defense of
the former Union arsenal at Chattahoochee. Because the machinery had not been com-
pletely assembled on board, the vessel was towed downstream from Saffold in the fall
of 1862. In late May, with its machinery finally in place and operating, the Chatta-
hoochee proceeded to Blountstown Bar, 28 miles below Chattahoochee, to recapture
the schooner Fashion. They could not cross the bar, so they waited overnight for the
river to rise. When it didn't, they raised stream to return only to have the boilers
explode and the gunboat partially sink. The gunboat was eventually raised and towed to
Columbus for repairs. A year and a half after the accident, the Chattahoochee was still
waiting for the completion of its repairs. In April 1865, with the Union Army approach-
Physical Processes at the CSS Chattahoochee Wreck Site 263

ing from the west, the Chattahoochee was set adrift and burned, never having its
repairs completed in almost 2 years of waiting (Turner, 1975).
There are conflicting reports of the demise of the Chattahoochee. One report
indicates that the gunboat had been towed 15 miles below Columbus and burned
(Turner, 1975). The other report indicates that the gunboat was doused with ten barrels
of kerosene at the Columbus Naval Iron Works. Then, slow fuses were ignited and the
Chattahoochee set adrift. The gunboat supposedly ran aground at Race Pass 12 miles
downstream. The latter report has been more acceptable to most historians (Turner,
1975).
The gunboat never saw action and the ineptness of the Confederate bureaucracy
stopped it from doing so. But, the Chattahoochee was, in a way, an offensive weapon;
that is, the Union Navy knew that the Chattahoochee and other boats were being built,
and had to deploy additional forces in the vicinity of the Apalachicola in the event the
boats came downstream.

GEOMORPHOLOGICAL NOTES

Steamboating on the Chattahoochee River in the 19th century was dangerous, and
naval operations were no different. Lack of river surveys, and too few skippers with
local knowledge were the major concerns of military and naval strategists. The river is
a narrow meandering stream within a rather narrow meandering valley. With its head-
waters in the southern Appalachian Mountains at Unicoi Gap, it has a drainage area of
over 17,000 square miles. The southerly facing headwaters with elevations over 4000 ft
catch up to approximately 80 inches of precipitation annually, with high amounts fall-
ing in the early winter and in the late summer. The precipitation regime is the same
throughout the basin, but the amount of precipitation is about 50 inches annually in the
lower reaches. The response of the stream to seasonally high precipitation and to a
lesser extent localized thunderstorms, is a surge or rise in stream level. Conversely,
seasonally low precipitation relates to minimum flows.
The Chattahoochee River at Columbus and below generally had a depth of about
8 ft. At peak flow the channel is scoured sufficiently to maintain a depth for most
steamers, but during the recession from peak flow the channel fills in and narrows. As
the shoals increase in area and the water over them decreases, the stream becomes more
treacherous for steamers. During the fall and late spring the streamflow is sufficiently
slow to allow shoaling, and steamboating tended to be at its minimum. When the
stream floods from December through March, steamboating reached its high.
At Columbus, which is located just downstream of the inner boundary of the Fall
Line Hills of the Gulf Coastal Plain, the stream datum is 185.14 ft above sea level. At
West Point 39 miles upstream the datum is 551.67 ft, while at Atlanta, 104 miles farther
upstream, the datum is 750.10 ft. This indicates an abrupt change upstream from Colum-
bus with a fall of 9.4 ft per mile, and the location of the Fall Zone. The older geological
structures and materials above Columbus have a relatively thick veneer of saprolitic
soils. These materials are easily eroded. As the materials are transported during high
flow they are moved and washed, creating a considerable sediment load and a reddish
brown color to the stream. At low flow the color is light brown with lesser turbidity.
Below the Fall Zone, in the Fall Line Hills, the Cretaceous formations have a
slight dip toward the gulf consisting of grayish, loosely consolidated, but stratified
264 R. Stephenson

marine deposits. The soils are commonly a yellow sand. It is here that the gunboat
reached its final resting place. While the stream datum at Columbus is about 185 ft, the
hills reach elevations of over 300 ft near major streams. It is common to observe
swamp forests or point bars on one side of the channel and high cut banks on the other.
The physical setting for navy yards and naval operations is at the very least interest-
ingly unique, and well protected from the coastal setting of Apalachicola and the threat
of Union landing parties.

SITE FORMATION PROCESSES

The remains of the CSS Chattahoochee were observed during an engineering sur-
vey of the river in 1870. Its location was at the upstream end of a point bar along the
right bank at Broken Arrow Bend. This is 13.7 miles from the railroad bridge in
Columbus and a considerable difference in distance from the historical accounts.
Apparently the thalweg or deepest portion of the channel was along the right bank. It is
likely that the gunboat drifted downstream and its bow struck the bar. Then, the current
forced the stem toward the bank, tilting the boat slightly to the configuration of the bot-
tom. There it burned near to the water line. In April 1865, the stream was probably
above its normal level. As the streamflow subsided, the gunboat revealed more of its
structure. It was reported that the Union forces had planned to salvage its machinery
when the water level decreased, and possibly salvage the gunboat. Most of the machin-
ery was eventually pulled from the wreck (Turner, 1975).
With the gunboat sanded in at Broken Arrow Bend, subsequent flooding probably
eroded the point bar and scoured out the wreck. It is likely that this happened more
than once. Of note are the known floods of 1886,1901,1907,1919, and 1929 (USGS,
1968). The movement of the wreck under flood conditions is very likely. For example,
the discharge at Columbus on March 15, 1929, was 198,000 cubic feet per second (cfs),
with a water level of 53.2 ft above the datum. Under these extreme conditions it is
likely that the wreck would at least slide along the bottom once the sand was scoured
away. In contrast to the floods, the mean discharge at Columbus is 6620 cfs. As stream-
flow subsided and deposition occurred, the gunboat found another resting place down-
stream. This probably occurred several times. That the wreck ever became buoyant,
however is unlikely but not improbable. The present location of the gunboat is approxi-
mately 13.9 miles downstream from Columbus, or about 1000 ft downstream from its
location on the 1870 map.
The Chattahoochee presently has its bows in an upstream direction. Did the gun-
boat become dislodged and have its bow turned in the upstream direction? Or could
this have occurred when it struck the bar, or at a later time during a flood? Or could it
have drifted downstream backwards? This is unlikely as the stem section was probably
more buoyant and would have cut the water better than the stem. In 1912 the wreck
was at least visible enough to recover two of its deck guns (Turner, 1975). Also, it has
been noted from time to time that at low water, fishing from the wreck was popular
(Holcomb, personal communication, 1984).
Since 1926 the streamflow has been regulated by Bartletts Ferry Dam, 15 miles
northwest of Columbus. In addition, since 1957, with the completion of the dam at
Lake Sidney Lanier above Atlanta, flow has been regulated. Low flow is affected by
milldams, small powerplant dams, diversions for municipal water supplies, and numer-
Physical Processes at the CSS Chattahoochee Wreck Site 265

ous farm ponds and small recreational lakes. These activities tend to subdue the peak
flow and stabilize the low flow. It is supposed that the gunboat has not moved appre-
ciably since the flood of 1929.
To improve steam navigation of the Chattahoochee River, the Woodruff Dam, 1.5
miles northwest of Chattahoochee, was constructed to form Lake Seminole in 1954.
The maximum pool level is approximately 78 ft above sea level. And in 1962, the Wal-
ter F. George Reservoir was completed with the dam about a mile north of Fort Gaines
at a maximum reservoir level of about 190 ft. In 1964, prior to full pool level, a 30- ft
stem section of the gunboat was removed to improve the channel. The stern is now on
display at the Confederate Naval Museum in Columbus. Since 1962 the gunboat has
rested quietly in about 20 ft of water. Most of the time the streamflow is slow, but its
velocity increases when the sluices are opened at Bartletts Ferry Dam. Also, flow as
well as turbidity is increased when local precipitation occurs.

ON,SITE OBSERVATIONS OF STREAM/RESERVOIR PROCESSES

In the summer of 1984, an archaeological investigation was begun on the gun-
boat. Only minor work was accomplished on existing physical processes at that time,
but some observations that were made should be enlightening.
The upper section of the Walter F. George Reservoir reaches Columbus as its full
pool level is 190 ft above sea level. Its minimum pool level is about 184 ft, but it can be
reduced to 164 ft if necessary. Because the pool level varies, the base level of the stream
changes downstream from Columbus for approximately 8 miles. This means that when
the pool is low and the stream is above nominal, streamflow will occur at the wreck site.
Conversely, when the pool is high and the stream is low, streamflow ceases at the site.
The dynamics of base level are important as numerous variables are related to stream-
flow. Channel variables such as width, depth, velocity, suspended sediment load, slope,
and roughness are related to discharge according to numerical constants and exponential
values. The artificial control of flow by the multipurpose dams and reservoirs complicate
the physical processes, as well as how to manage the archaeological site.
In addition to the physical processes, the investigation must consider other envi-
ronmental aspects, particularly water quality (Columbus, 1983). The water quality
parameters for which data are readily available are water temperature, dissolved oxy-
gen, pH, turbidity, conductivity, biochemical oxygen demand, total alkalinity, ammonia
nitrogen, nitrate-nitrite, total phosphorous, total organic carbon, and fecal coliform bac-
teria. Insofar as the integrity of the wreck is concerned, several of these parameters are
important. However, their relationship to the in situ preservation or destruction of
wooden hulls and various metals is not well understood. Of interest here is a slightly
acidic mean pH of 6.5, a low conductivity of 69.1 micromhos, and a very high mean
fecal coliform of 19,912 colonies per 100 ml of water. Is it little wonder that eight out
of ten divers during the operation had ear infections with down time being curtailed
from 3 days to 2 weeks?

SOME MANAGEMENT ALTERNATIVES

Additional research at the wreck site should include, but not be limited to, the fol-
lowing: (1) additional mapping and excavation of the wreck, (2) a study of the substrata
266 R. Stephenson

at the site, (3) a detailed chemical analysis of the substrata and the stream, and (4) a
chemical analysis of the gunboat hull material above and below the stream bottom.
These analyses should provide a better understanding about the integrity of the
vessel and how it is aging. Further, it will allow better management decisions to be
made regarding its future.
Management schemes could involve the following: (1) leave the gunboat as is, (2)
blow it up ifit gets in the way, (3) move it out of the way, (4) move it to the Confederate
Naval Museum for conservation and preservation, or (5) provide controlled dives for rec-
reational and interpretive purposes. At the present time, the final report of the 1984
investigation is being written. Perhaps additional presentations will be made concerning
the CSS Chattahoochee in the future. Hopefully, this presentation shows the need for a
multidisciplinary team, particularly a geomorphologist, in underwater research.

REFERENCES
Turner, M., 1975, Naval Operations on the Apalachicola and Chattahoochee Rivers, 1861-1865. Alabama His-
torical Quarterly, XXXVI.
U.S. Geological Survey, 1968, Water Resources Data for Georgia, Part I, Surface Water Records. U.S. Depart-
ment of the Interior, Atlanta.
 Chapter 25

'.~~.' The Archaeology

~

o i' .
of Shipwrecks
q;,O_" .': .
.j'~l

KEITH MUCKELROY

1. INTRODUCTION

Given that maritime archaeology is concerned with the study of ships and seafaring,
and that its principal sources of data lie in the remains of such activities preserved on
the seashore or seabed, it follows that the interpretation of such data is closely bound
up with an understanding of what is involved in a shipwreck. The shipwreck is the
event by which a highly organized and dynamic assemblage of artifacts is transformed
into a static and disorganized state with long-term stability. While the archaeologist
must observe this final situation, his interest, is centered on the former, whose various
aspects are only indicated indirectly and partially by the surviving material. If the vari-
ous processes that have intervened between the two states can be identified and
described, the researcher can begin to disentangle the evidence he has uncovered.
In the present chapter, an attempt is made to identify the several features common
to any shipwreck, and the ways in which they can be assessed on any particular site.
Just as the nature of a ship involves certain basic concepts that are common to all peri-
ods and places, so the phenomenon of the shipwreck must involve certain regular fea-
tures common to all instances. If these can be described, then their implications for any
Source: Maritime Archaeology. New York: Cambridge University Press, 1978. Copyright 1978 by Cambridge
University Press. Reprinted by permission of Cambridge University Press.

267
268 K. Muckelroy

analysis of seabed remains can be ascertained, and the most appropriate procedures
identified. Furthermore, if the latter can be tested and demonstrated on sites for which
the evidence, both archaeological and documentary, is extensive, then they can be
applied in situations where the evidence is more fragmentary and confused, and the
analysis correspondingly more valuable. The validity of any conclusions reached in
maritime archaeology depends fundamentally on the understanding of these processes,
so that their study must occupy a central place in the subdiscipline. To date, facile and
unrealistic assumptions about what happened during and after a wrecking have been a
feature of too many wreck site reports, with a consequent diminution of their value and
authority. These considerations account for the concentration on site analysis in the rest
of this chapter; it is taken for granted that in any particular project standard archae-
ological artifactual analysis will also be undertaken.
Compared with equivalent depositional and postdepositional processes on land
sites, shipwrecks possess many peculiar features. The environmental factors operating
underwater are different from those found on land, and are outside the range of normal
experience. Furthermore, human interference, undoubtedly the most important destruc-
tive agent in a terrestrial context, is minimal underwater, and limited to a few identifi-
able activities. Finally, the same factors are operative on every site, although in varying
degrees, so that the archaeological evidence is more homogeneous in this subdiscipline
than in most others, an attribute that further strengthens its internal cohesion. Thus, an
investigation into the archaeology of shipwrecks constitutes both the final defining
characteristic of maritime archaeology, and the starting point for constructing its gen-
eral theory.
The flow diagram in Figure I represents the processes through which that organ-
ized assemblage of artifacts comprising a ship and its contents will have passed to pro-
duce the collection of items excavated on the seafloor. This assemblage can be
regarded as a system, defined by the necessary characteristics of a ship, which has un-
dergone a series of transformations through time, emerging as the results of an archae-
ological excavation. At the same time, the whole shipwreck process itself constitutes a
closed system, with the ship as input, and with a number of different outputs, about
which there are varying degrees of knowledge (Clarke, 1968:44). Similarly, the under-
standing of the several elements within the central box of the system is partial; the hope
is that by taking it as a whole the various bits of incomplete knowledge can be inte-
grated in such a way as to increase overall comprehension.
Within the present chapter, consideration of the operation of this system is
divided into three parts. First, there is a general discussion of the natural environment
within which it operates, looking for regularities, and supplementing the general ideas
presented above. Then the elements displayed within the central box in Figure I are
divided into two groups, those that act as filters extracting material from the assem-
blage, and those that operate as scrambling devices, rearranging its patterns. The next
part of the discussion is thus concerned with the filtering effects of the process of
wrecking, salvage operations, and the disintegration of perishables. The third part, deal-
ing with the scrambling devices, is covered in two sections, one dealing with the rear-
ranging aspects of the process of wrecking and one with postdepositional seabed
movement. Having thus described the development of seabed distributions, it is possi-
ble to proceed to a brief discussion of some of the modes of analysis that might prove
valuable in their interpretation (see Chapter 42).
The Archaeology of Shipwrecks 269

- - - -

\
I Process of wrecking Material which
floated away

~ Salvage operations Material

salvaged

1 Disintegration of
perishables

1
i
Material which
disintegrated

...., Sea -bed movement I

Matena·1
subsequently
depo~ited
\
on site

I Characteristics
of excavation
I
Observed
sea-bed
distribution

Figure 1. Flow diagram representing the evolution of a shipwreck.

While the consideration of this system is presented in general terms, its implica-
tions in both interpretation and analysis are illustrated with material derived from a
wreck site with which I have been particularly involved in recent years, and which has
proved particularly suitable for testing these ideas. The site is that of the wreck of the
Dutch East Indiaman Kennemerland. which struck Stoura Stack in the Out Skerries
(Shetland Isles, U.K.) on 20 December 1664, while outward bound from the Nether-
lands to the East Indies; only three men survived by being projected onto the Stack
from the lookout post. In addition to the standard archival material from the company
records in Amsterdam, there are a number of documents in Scotland referring to the
legal disputes over the ownership of material subsequently salvaged, including the
treasure. The site was first identified by a student diving team from Aston University in
1971, and since then there have been three major seasons of excavation involving
members of Manchester and other British universities. A number of interim and other
reports have been issued over the years, from which further information about this
work can be obtained (AUSAC, 1974; Forster and Higgs, 1973; Muckelroy, 1976;
Price and Muckelroy, 1974, 1977).
270 K. Muckelroy

2. WRECK SITES AND THEIR ENVIRONMENTS

From the earliest days of archaeological investigations underwater it was appar-

ent that, as on land, ancient remains were more likely to have been preserved within
soft substrates than within rocky ones. However, it soon became clear that the nature of
the seabed was by no means the only controlling factor, and so more complex explana-
tions were developed, notably by the French underwater pioneer, Frederic Dumas (see
especially Dumas, 1962, 1972). Unfortunately, as nearly all of the early work in this
field concerned wreck sites in the Mediterranean, the theories developed were heavily
biased by the special characteristics of that sea. In his discussions, M. Dumas drew a
fundamental distinction between sandy shores, rocky shores, and shores with sub-
merged cliffs (Dumas, 1972:32-33), stating that it was only in the last circumstance
that coherent archaeo~ogical remains were to be found. This feature of submerged cliffs
close inshore occurs along much of the northern shores of the Mediterranean, but is not
so common elsewhere. Wreck material that had not quickly dropped into deep water
was regarded as beyond recovery: "The sea smashes everything in shallow waters, and
such scattered wreckage is of scant interest to the archaeologist" (ibid., 32). From this
grew the general rule that nothing of significance would ever be found in shallow
water; for example, Miss Frost has written: "ancient wrecks are necessarily in deepish
water" (Frost, 1962:82).
At the same time, the limitations of these ideas were clearly stated. M. Dumas
recognized that his general statements need not be directly applicable outside the Medi-
terranean (Dumas, 1972:34), while W. D. Nesteroff, a geologist writing in the same
volume as Dumas (UNESCO, 1972), noted some aberrant sites, such as that at Spargi
(Lamboglia, 1961). He suggested as additional mitigating factors a heavily fissured
seabed, offshore islets, and the protection afforded by marine plants (Nesteroff,
1972: 176-177). But such instances were seen as exceptions to the general rules, so that
it was in terms of the basic distinction between favorable and unfavorable conditions
that subsequent workers, both within and beyond the Mediterranean, approached their
sites. As will have become clear in the course of chapters 3 and 4, however, this is a
dangerous oversimplification, for there are a large number of sites on which the
remains are neither totally coherent nor totally broken up. The factors producing these
intermediate sites are obviously complex and demand further investigation.
One recent study in underwater environmental archaeology, concerning wreck
sites in British waters, has proceeded by appropriating the methods of marine ecology.
The idea was to measure the degree of correlation between the quality of the archae-
ological remains and a number of possibly relevant environmental attributes. These lat-
ter were selected on the basis of those that had proved significant in parallel studies in
other marine sciences, such as biology (Hiscock, 1974) or coastal geomorphology
(King, 1972), being defined as follows:
1. Maximum offshore fetch, within 30° of the perpendicular to the coast.
2. Sea horizon from the site, i.e., sector within which there is more than 10 km
of open water.
3. Percentage of hours during which there are winds of force 7 or more from
directions within the sea horizon.
4. Maximum speed of tidal streams across site.
The Archaeology of Shipwrecks 271

5. Minimum depth of site.

6. Maximum depth of site.
7. Depth of principal deposit on site.
8. Average slope of the seabed over the whole site.
9. Underwater topography: the proportion of the site over which the seabed con-
sists of geologically recent sedimentary deposits.
10. Nature of the coarsest material within these deposits.
11. Nature of the finest material within them. In ordering sites on this attribute
and the previous one, they were ranked initially according to broad categories
of material, and then according to the relative importance of these deposits
on the different sites.
Twenty wreck sites were identified for which sufficient environmental and archae-
ological data were available; their locations and names are shown in Figure 2. Each
attribute was then ranked across the 20 sites, a procedure that allowed both qualitatively
and numerically scored ones to be used; the resulting matrix is given in Table 1. The

I Kennemerland(1664)
2 De Liejde(171 I)
3 Trinidad Valencera
(15 88)
4 Adelaar (1728)
5 Dartmouth (1690)
6 El Gran Grifon (1588)
7 Lastdrager (1652)
8 Girona (1588)
9 Santa Maria de la Rosa
(15 88)
10 Santo Christo de
Castello (1667)
I I Mary Rose (1545)
12 Pen lee cannon site
( 169 0 ?)
13 Low Lee Ledges ( ?)
14 Mewstone Ledges ( ?)
15 Amsterdam (1749)
16 Hollandia (1743)
17 Wendela (1737)
18 Cura{:ao (1729)
19 Evstafii (1780)
20 Colossus (1798)

Figure 2. A map showing the locations of the 20 listed wreck sites in Britain used in the environ-
mental attribute study.
272 K. Muckelroy

Table I. The matrix in which the archaeological survival characteristics (row A) and the I I
environmental site attributes (rows I to II) are ranked across the 20 sites shown in Figure 2'
2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 20
A 5 13 3 17 4 14 II 12 7 10 2 18 16 20 I 8 19 9 15 6
I II 10 4 13 8 12 14 18 15 3 19 9 20 2 17 5 6 7 16
2 3 7 8 10 I 5 2 14 12 9 4 13 II 16 17 15 18 19 20 6
3 8 6 10 12 5 4 2 13 3 19 7 17 9 18 14 15 16 20 II
4 14 15 12 8 13 20 7 19 3 17 5 10 6 2 18 16 II 9 4
5 12 10 7 19 18 6 14 15 16 4 9 5 17 20 2 II 3 13 8
6 2 6 15 10 13 7 18 16 19 9 17 8 12 20 3 5 4 14 II
7 14 9 II 15 16 7 19 18 I 17 6 12 5 13 20 2 4 3 10 8
8 10 13 8 4 20 18 17 12 3 15 2 16 14 19 9 6 7 5 II
9 6 II 3 18 4 14 16 12 5 7 2 17 9 20 8 19 15 13 10
10 9 15 4 13 5 12 16 19 6 7 2 17 8 18 I 3 20 14 II 10
II 5 15 3 18 6 14 9 10 20 12 16 8 17 2 4 19 II 13 7
"Description of the II attributes is given in the text.

Kendall rank correlation coefficient 't was then used to quantify these relationships
(Siegel, 1956:213-223); Figure 3 shows the values produced when correlating archae-
ological survival against the 11 environmental attributes, together with the 5% and 0.1 %
significance levels. For further details of these studies, together with a full discussion of
the principal implications, reference should be made to the original papers (Muckelroy,
1977a,b).
As is immediately apparent from Figure 3, there can be no doubt that the nature
of the seabed deposit (described by attributes 9, 10, and 11) is the main determining
factor in the survival of archaeological remains underwater. While this fundamentally

10 r

09

0.8 r-

.
'0
'"
:l
0.7

0.6

05
C - -- - - - - - - - - - - - - - - - - - - - -
Ir-r- I
o 1 % significance

iii 0.4
>
0.3 50 % significance

02

I
01

0 I • I
2 3 4 5 6 7 8 9 10 11
Site attributes

Figure 3. The values of tau for the II environmental attributes (for list, see text),
The Archaeology of Shipwrecks 273

confirms previously established ideas, it also serves to demonstrate that intermediate

types of wreck site are closely related to intermediate types of seabed, that there are
many gradations between the totally rocky and the totally sandy substrate. The truth of
this can be appreciated by looking at several of the illustrations given earlier in [the
original text] (Muckelroy, 1978:21,26,29,41). Attribute 8, representing slope, also reg-
istered a fairly strong correlation, but as this is essentially only another measure of bot-
tom topography it simply reinforces the above conclusions. The implications of the
slightly higher figure recorded for attribute 2, representing sea horizon, are more inter-
esting. It suggests that it is the variety of disturbing forces acting on a site, coming
from a number of different directions, which is more significant than their force. This
idea makes sense if one thinks of a deposit in a gully; once concentrated in a sheltered
situation, it can only be disturbed by forces acting in a totally different direction, so
that the fewer the number of directions from which disturbance can come, the better the
chances of survival for those remains. Similar considerations apply with a seabed con-
sisting of sedimentary deposits, in which the substrate is more likely to remain stable if
it is not assailed from a number of directions. In terms of the shipwreck flow diagram
(Figure 1), a restricted sea horizon reduces the significance of the recycling channel on
the left-hand side of the box.
At the same time, the low scores registered with respect to certain other attributes
are equally interesting. This is particularly the case with those relating to depth (attrib-
utes 5 to 7), which markedly contradict previously held convictions. A common
denominator in the remaining attributes seems to be water movement across the site, in
terms of either the size of the swell, which is a function of fetch (number 1), the fre-
quency of stormy weather (number 3), or the strength of tidal currents (number 4). This
would suggest that, once initial deposition has been made, these forces have little
impact on the subsequent history of the remains. It seems likely that the influence of al-
gae cover is relevant here, in that it projects material in just those exposed situations
that might otherwise prove unsatisfactory for survival. In a way, it reduces the differ-
ence in severity between open and protected coasts. It might be felt that water move-
ment is more likely to have affected the distribution of the remains, as opposed to their
survival, but a further study, in which the 11 attributes were correlated against a rank-
ing of the 20 sites according to coherence of their distributions, did not support such an
interpretation In fact, the results of this second test showed no appreciable difference
from those achieved in the first exercise; what differences there were appeared to be
related to special features in the process of the wrecking of some vessels, rather than
any environmental considerations (Muckelroy, 1977b:53-54).
In the light of these studies, certain general conclusions were reached concerning
wreck sites in British waters. On the one hand, the sites could be divided into five
classes, according to their degree of survival, as described in Table 2. This classifica-
tion is more discriminating than the simple good/bad dichotomy which has held such
wide currency in the past, and, while it must still be regarded as provisional, it has
proved useful in a number of contexts, and will reappear later in this chapter. And, sec-
ond, arising from and making use of this classification, a series of general statements
regarding the environmental characteristics appropriate to each class in British waters
could be proposed; this is given in Table 3. It was compiled in such a way that every
site was accommodated within the specifications for the appropriate class under four of
the five headings, and so that the aberrant specification, if there was one, was in one of
274 K. Muckelroy

Table 2. The five main classes of wreck site at present apparent in British waters

Structural Organic
remains remains Other objects Distributions Examples (see Fig. 2)
Mary Rose (11)
Class I Extensive Many Many Coherent
Amsterdam (15)
Dartmouth (5)
Class 2 Elements

Class 3 Fragments
}S~' Many Scattered/ordered
Trinidad V. (3)
Kennemerland ( I )
Colossus (20)
De Lie/de (2)
Class 4 Few Some
Girona (8)
} S"",redld;""',,,d Adelaar (4)
Class 5 Few
Pen lee site (12)

the last three attributes, for which the correlation scores given above were lower.
Whether this table can be used in a predictive sense, in assessing the potential of a site
newly discovered but as yet unexcavated, only further testing will tell.
The serious limitations to this study must be very apparent. The sample of 20 sites,
although the maximum available at the time, was very small and unevenly distributed
around the coasts of Britain, with a strong bias toward the Shetlands and the southwest
peninsula (see Figure 2). Furthermore, it was limited to just one offshore island in Europe,
and is thus no more representative than Mediterranean experience. Similar detailed studies
are required for many different parts of the world before any factors common to them all
can be isolated. It could also only consider sites on which material remains had actually
been found, and thus discounted those situations in which conditions have led to the total
eradication of all traces of wreck. As a result, the conclusions given in Table 3 cannot be
used to predict whether any remains will have survived on a site over which a vessel is
known to have been wrecked; the possibilities of research into this matter are discussed
elsewhere (Muckelroy, 1977b:56). In addition, no account has been taken of the fact that
survival conditions can vary enormously across a single wreck site; in each case, aggregate
measurements and overall descriptions have been considered, within which there may
have been much significant variation. And, finally, the wrecking process has been looked
at as a single event; greater understanding should be possible by considering its constituent
elements, and this is the approach pursued in the next three sections.

Table 3. The relevant environmental attributes for each of the five classes of wreck site described
in Table 2, derived from an analysis of the data from the 20 British sites

Topography Deposit Slope Sea horizon Fetch

% of bottom Range over Average over Sector of open Maximum offshore
sedimentary deposit whole site whole site water for 10+ km distance
Class I 100% Gravel to silt Minimal Less than 90Q Less than 250 km
Class 2 More than 70% Boulders to silt Less than 2Q Less than 90Q Less than 250 km
Class 3 More than 30% Boulders to silt Less than 4Q Less than 150Q More than 250 km
Class 4 More than 10% Boulders to sand Less than 8Q More than 30Q More than 250 km
Class 5 Less than 25% Boulders to gravel More than 6Q More than 120Q More than 750 km
The Archaeology of Shipwrecks 275

3. EXTRACTING FILTERS
In terms of the flow diagram in Figure 1, the three processes that lead to the loss
of material from a wreck site are the process of wrecking, salvage operations, and the
disintegration of perishables, each of which thus generates an output shown to the right
of the diagram. Essentially, any consideration of this aspect of a shipwreck resolves
itself into a matter of simple addition, as the sum total of the outputs must necessarily
equal the input (i.e., the original ship). However, the main thesis presented here is that
because the operation of each of these processes is limited by a series of constraints,
some of which can be identified and described for any particular wreck site, certain ele-
ments in their respective outputs can thus be identified, which in turn amplify the evi-
dence regarding the ship itself.
The process of wrecking, which can act as both an extracting filter and a scram-
bling device, raises the question of which items on board a ship are likely to have floated
away. From first principles one knows that metal objects, for example, cannot float,
whereas most wooden and other organic objects ca~ at least until they have become
waterlogged. Thus, on any site there will be a wide range of objects that cannot have dis-
appeared through this process-those in the former category. However, the greater part
of a wooden ship is made of materials that can float, so that the process of wrecking must
also explain in many cases how these elements came to be held on the seabed long
enough to become waterlogged and/or buried. One extreme situation is the vessel that
sinks intact, dragged down by a combination of ballast, contents, and inflowing water; in
this case, everything trapped between decks has a chance to become waterlogged, sink to
the bottom, and, if favorable circumstances present themselves, become buried. The
other extreme is when a vessel disintegrates totally on the surface, spilling all heavy
items, and allowing little chance for light objects to reach the seafloor; here, little will
have survived in even the most favorable of environments. But most sites reflect proc-
esses intermediate between these extremes, as with the many classical ships pinned down
by their amphora cargoes, or with the wreck of the Dartmouth (1690), on which part of
the hull had been pinned down by the iron and flint ballast (Martin, 1978).
Salvage operations obviously vary widely between sites, and are less susceptible to
generalization. In many cases, the assessment of this factor will depend on the evidence
of the historical record, but if this is silent, there are certain assumptions that can usually
be made. There are some sites on which salvage was inconceivable until modern times,
such as those off uninhabited shores or in very deep water. Elsewhere, however, it will
generally be safe to assume that, other things being equal, the local inhabitants are more
likely to have attempted to raise all they could than leave such a bonanza unexploited.
From very early times, many communities will have had the ability to undertake free div-
ing and dragline operations, and since the 16th century more sophisticated devices have
become available within areas of European influence (Davis, 1955:536-642; Ericsson,
1975). At these later periods, there are records of quite drastic methods being used in sal-
vage, such as trying to drag whole ship sections up the shore with teams of horses. In
each instance, the archaeologist must assess both the problems that would have faced any
contemporary salvors and their ability to overcome them, and decide which parts of his
site are most likely to have been interfered with.
With the disintegration of perishables, the constraints concerning the preservation
characteristics of different materials in a marine environment are again more general in
276 K. Muckelroy

application. Here, maritime archaeology must rely on studies of underwater corrosion

(e.g., Uhlig, 1948), as well as the continuing research specifically concerned with the
preservation, and ultimate conservation, of archaeological materials from marine sites
(e.g., Weier, 1974; Hamilton, 1976; Tylecote, 1977). It has become apparent from a
number of sites that such processes can vary even across quite short distances, depend-
ing on the nature of the seabed, the extent of marine growths, etc., so that ideally
seawater and deposit samples should be taken from each area excavated, to be chemi-
cally and biochemically analyzed. Furthermore, the association of different types of
wreck material will itself locally modify the preservative environment; the most
extreme example is probably the electrolytic protection of base metal by a neighboring
proud metal. Unfortunately, these are still early days in such research, so that, until a
greater understanding has been reached, the maritime archaeologist is probably
restricted to only general principles concerning the scale and scope of disintegration of
material on any particular site.
In the light of these general considerations, it is worth looking at how they oper-
ate in practice on the Kennemerland site; Table 4 shows the system outputs as they are
at present understood. This was a vessel that broke up entirely after striking a rocky
coast, although a considerable range of fragile organic materials were somehow carried
from it onto the seabed (Price and Muckelroy, 1977: 193-198). Under the heading
"Floated away," the designation "none" has only been given on materials that in them-
selves could never have floated. As noted above, a good deal is known about the sal-
vage operations on this site from Scottish Court Records which include long
inventories of goods declared; nevertheless, it is to be presumed that considerable
quantities totally escaped detection, especially easily secreted but valuable items such
as coins, and commodities that were consumed immediately, notably food and drink.
One authority mentions that the Out Skerries folk had a wild Christmas that year!
(Brand, 1701: 140). Within the column headed "Disintegrated," the entry "none" has
been given only against those materials that could not have so disappeared on the basis
of general principles. No specific analysis has yet been conducted on the seabed and
waters of this area from which to arrive at more detailed conclusions; the variety of
conditions across this site is such that consistent trends are unlikely to be present. A
summary of material excavated has been included for comparative purposes and in
order to indicate the full range of knowledge about the ship to be derived via the sys-
tem outputs.
A more detailed discussion of the main conclusions arising from Table 4 has been
published elsewhere (Muckelroy, 1976:283-286), and only a few outstanding features
will be mentioned here. Probably the most striking aspect of Table 4 is that over half of
the commodities listed cannot have floated away; with the obvious exception of the
ship itself, this cannot be the sole cause of the loss of evidence. The position under
"Disintegrated" is less satisfactory, but in this case the areas where this element is par-
ticularly important, notably in the lower half of the table, the evidence from salvage
records is particularly full, and probably of more than average reliability. The evidence
derived from the salvage lists and the excavations themselves is also remarkably com-
plementary, emphasizing the importance of integrating the various sources of evidence
in this manner. As the excavations on this site are still far from complete, little can be
deduced from the absolute quantities involved, although there are already some inter-
esting implications that can be noted. For example, the collection of coins in Table 4 is
The Archaeology of Shipwrecks 277

Table 4. System outputs for the Kennemerland site

Excavated Salvaged Disintegrated Floated away

Ship's structure Fragments
Sails 69
Rigging Pieces rope 7 cables; tackle
Anchors 8 2 None None
Ballast bricks c. 10,000 None
Grindstone I None None
Iron Some None None
Lead 108 ingots None None
Nails c.400 None None
Cannon 8 7 None None
Cannonballs c.100 None None
Musket shot c.2000 I chest None None
Scatter shot c.3000 None None
Munitions accessories 9 items 15 lining sheets
Navigation instruments 3 items
Specie 6 coins 3 chests None None
Jewelry 40 pieces None None
Eating utensils 25 spoons
Personal items 49
Quils etc. I chest
Writing paper 8 chests
Clay pipes c.200 None
Bridle bits Some None None
Stirrup irons Some None None
Tar Some 19 puncheons None
Tallow 9 casks None
Some 15 casks None

t"
Rosin
Mercury 2 flagons I chest None None

~d
Olive oil 1320 gallons None
Brandy 100 bon1" 1604 gallons None
c. 100 flagons:
Wine 1254 gallons None
c. 3 pints each
Vinegar 145 gallons None
Beer 8 casks None
Preserved fruits I jar None
Butter 5 barrels None
Flour 2 half-barrels None
Meat 58 bones 2 pieces bacon None
Shoes 120 pairs
Linen 337 yards
Serge 300 yards
Woolen cloth 116 yards
Other cloth 236 yards

very small for the personal moneys held on such a ship (discounting the three official
treasure chests salvaged), and strongly suggests extensive unreported salvage. Regard-
ing the ship herself, one suggestion that arises from the quantities of drink indicated in
the salvage lists is that it is too much for consumption during the voyage, and must
therefore constitute an element in the ship's cargo.
While such a discussion must inevitably deal on the level of probabilities and
degrees of certainty, it is undoubtedly worthwhile in that it should help to remove inc on-
278 K. Muckelroy

sistencies in the deductions being made from the fragmentary evidence available about
any wreck site, and may inspire new ideas about interrelationships and implications
within that evidence which, taken in its constituent parts, had not been apparent. There
should be very few wreck sites for which something useful cannot be said about each of
these processes, a fact that reinforces the idea that a shipwreck is a particularly worth-
while process to investigate, when compared to other types of archaeological type.

4. SCRAMBLING DEVICES - PART I

The Process of Wrecking

A ship floating or sailing on the surface of the sea is a complex machine contain-
ing a large number of constituent parts arranged in a specific order to ensure seaworthi-
ness, ease of handling, and other desirable qualities. From the moment of impact,
however, that high degree of organization begins to break down, until the remains are
assimilated into the seabed in some degree of disorder, often very extensive. This con-
stitutes the first stage in the rearrangement of the elements of a vessel which interests
the researcher, and is covered by the title "process of wrecking." This definition is thus
a broad one, including the continued breakup of the wreckage on the seabed as well as
the stages by which it got there, up until the time when it becomes part of the seascape;
anything that happens after that can be described as seabed movement, and is reviewed
in the next section. In the present discussion, some of the factors involved are presented
by considering a few examples.
The most straightforward wrecking situation is undoubtedly that in which a ship
simply fills with water, sinks to the bottom intact, and remains there undamaged for
many centuries. In fact, such a sequence of events is very exceptional. Undoubtedly the
best known instance is that of the Swedish warship Wasa, which sailed from her fitting-
out yard across Stockholm harbor on 10 August 1628, heeled over to a squall so that
the water entered her open gunports, and sank within minutes, with over 200 persons
on board (Naish, 1968:5). Injudicious careening of the Royal George at Spithead on 29
August 1782, possibly combined with structural weakness, had similar consequences;
this time leading to the loss of over 900 lives. Sixty years later, divers reported that her
hull remained substantially intact, incidentally showing that the teredo-free waters of
the Baltic are not unique in allowing the long-term preservation of complete wooden
hulls. However, such sinkings without considerable structural damage onto receptive
seabeds are special cases, bearing little relationship to the majority of shipwrecks.
A more typical situation, which has received much attention in the past, is that of
the classical wreck now lying at the foot of a submerged cliff in the Mediterranean. As
explained in an earlier section, this underwater topography is fairly common in the Sea,
especially along its northern shores; so far as the best preserved wrecks are concerned,
it involves a peninsula, islet or offshore reef surrounded by relatively deep water (40+
m), as at Cape Dramont, Ile de Planier, or Yassi Ada (Dumas, 1972: Figure 4). Frederic
Dumas, who first rationalized the sequence of events in such situations, has written:
"When a boat runs aground on a rock and is holed, it sinks without other damage to the
foot of the sea cliff, where it lies sheltered from the sea's violent motions. Often
enough it has avoided the talus and settled on loose sand, in excellent conditions for
later study by archaeologists" (1972:32). Its subsequent slow assimilation within the
The Archaeology of Shipwrecks 279

....,
Qi

flo
;:! .,
J:: '"
~
:;
..t'q//
0,
"iii
<> '0
., E '?q
t:., ;; ; '1',

> « '0
Visible cargo

Figure 4. Aschematised section across an amphora wreck mound .

seabed is illustrated in his accompanying diagram (Figure 4), with the hull accommo-
dating itself to the seabed, the cargo spilling over on one side, and the whole acting as a
trap for waterborne sediments, thus creating its own tumulus. This last is the famous
amphora mound' recognized by modem divers in Mediterranean waters. While the
vessel probably reached the seabed relatively intact, only the lower part of the hull will
be available for study through having been protected by that mound, and the cargo
itself will have been considerably displaced from its original arrangements, although in
a way that can be understood.
That a similar type of situation is possible with later wrecks not bearing protec-
tive amphora cargoes has been demonstrated for Mediterranean waters by the excava-
tions of Peter Throckmorton and his colleagues in the harbor at Porto Longo in Greece
(Throckmorton, 1970:2~27). In two cases, with an Austrian brig of 1860 and the Her-
aclea of 1940, the vessels had landed relatively intact on a sandy substrate, worked
themselves into the deposit, creating shallow depressions within which the keel, floor
timbers, and miscellaneous other pieces had become buried while the rest of the vessels
was broken up or eaten away. A somewhat similar process can be seen to have hap-
pened with the settlement of the Mary Rose into the seabed at Spithead after 1545 (see
Muckleroy, 1978:93, Figure 3.17), even though she was a considerably larger vessel,
and the marine environment very different from that in the southern Peloponnese. In
this instance it seems that a rapid scour took place around the wreck soon after sinking,
and that she heeled over into the pit so created, finally reaching an angle of 60°. Her
tendency to sink into the seabed would also have been assisted by the fact that she was
still intact when reaching the bottom, containing all of her ballast and contents.
A slightly different process of the same basic nature can be seen with the wreck
of the Dartmouth (1690) in the Sound of Mull (Martin, 1978). Having been swept from
her moorings over 3 km away, this warship grounded against a small islet onto a seabed
consisting of a thin (c. 25 cm deep) layer of gravel overlying a hard stable clay. The
ship appears to have settled on her starboard side, and then to have rocked to and fro on
280 K. Muckelroy

\ pori
,,
,
,

,,
,

o 101..,
J~ __L - - L_ _~'~I~__- L__~~__~~__~I
Keeluench
o 3melrll!1o

Figure 5. A section across the structural remains on the Dartmouth (1690) wreck site.

her keel, wearing away a good deal of the latter, and digging a depression in the under-
lying clay into which the rest of the keel, deadwood, etc., could settle (Figure 5). The
discovery of a large number of objects from the ship in fine condition within this trench
showed that it must have been formed quite rapidly. The shallower depression, into
which the starboard planking was to settle, must have been formed by a more gentle
fanning action as the ship moved, for the underside of the hull here remained in
remarkably fine condition, including the thin pine sheathing, and only a finer, silty
deposit containing few artifacts remained below this part of the structure. The vessel
seems to have broken her back amidships on a spine of rock and the forward part,
apparently less encumbered by iron ballast, broke up in situ, leaving a significant pat-
tern of artifacts on the seabed.
However, not all vessels will have reached the seabed structurally intact; in fact,
the majority do not, for usually either the force of initial impact is sufficient to break
the vessel asunder, or it remains stuck fast until it breaks up. In the case of the Kenne-
mer/and the course of the ship's disintegration is clearly indicated by a combination of
local tradition and the spread of the archaeological remains (Figure 6). The former was
enshrined in a short rhyme current in the islands:

The Carmelan frae Amsterdam,

Cam on a Maunmas Day,
On Stoura Stack she broke her back,
And in the Voe she ca.
(Forster and Higgs, 1973:292)

Accordingly, the archaeological evidence for initial impact is concentrated

around Stoura Stack, including over a 100 lead pigs and many thousand building bricks
from the ship's ballast, along with a collection of anchors that had probably been
stowed low in the hold around the mainmast step (Glamann, 1958:23). Together, these
The Archaeology of Shipwrecks 281

¢ Cannon

Anchor
.1

,
0 Area of
unabraded bricks

Area of excavations

Trolsome
if
d
Area of lead Ingots

South
Mouth

o 100 metres
I

Figure 6. Site plan of the Kennemerland (1664) wreck site in the Out Skerries, Shetland Isles.

items suggest that the ship had her bottom tom open. However, there is another local
tradition to the effect that the bows of the Kennemerland broke away and sank next to
the Stack (Bruce, 1907: 127), an interpretation not actually contradicted by the evi-
dence. As has already been shown with the Dartmouth, it is quite conceivable for a ship
to break in two amidships; the mainmast can act as a pile driver and split the keel.
Either way, having lost a good deal of ballast and possibly her bows, the ship
wallowed north for over 150 m before any substantial further breakup took place, as
282 K. Muckelroy

A
North
0-,--_ Excavation
areas
Cannon 3
5

10 Lamm Saee Boulders with

Lamm Saeh gravel
Lamm Hyp
15

20 ~----------~---------------------- __~____________________________
( Metres)
(Vertical scale x 2) 100

Figure 7. A north-south section across the Kennemerland wreck site (see Figure 6).

there is a marked break in the seabed distributions (see Figure 6). The wreck material
reappears at the point where the seabed rises to within 10m of the surface, as shown in
the north-south section across the site (Figure 7). This is where the waves would be
breaking during a southerly gale, such as is known to have been blowing on this occa-
sion. Presumably, the white water tossed the vessel about and hastened its destruction.
However, some structural integrity must have been preserved until it reached the level
of the islet Trolsome, where the most northerly of the cannons was dropped. Beyond
that the position is less clear, as the area has yet to be thoroughly searched, although it
is known that much wreckage was washed up on Bruray.
There is a further tradition, the reliability of which is less certain, that most of the
wreckage was soon washed out to sea again by the succeeding tide (Bruce, 1907: 127),
and this receives a degree of confirmation from the pattern of tidal currents running
through the South Mouth that night. High water on 20 December 1664 was at noon, the
following high being at 01:00 on the 21 st (information courtesy of the Institute of
Oceanographic Sciences, Bidston). There are reasons to believe, on the basis of the docu-
mentary evidence, that the wrecking took place between dusk (say 16:00) and midnight
on the 20th, during which period the currents would have been running weakly north-
wards before 20:00, and strongly southwards from 21 :00. It seems likely, therefore, that
the wrecking took place during the evening of the 20th, and that a quantity of remains
were swept back across the site around midnight. A further important implication of this
interpretation is that the two cannons lying south of Old Man Stack (see Figure 6) are
likely to have come from the Kennemer/and. a fact that, in view of their isolation and
lack of association with other wreckage, has always been in doubt. This conclusion arises
because, during the first half of the 4-hour period of northward-flowing currents within
the South Mouth, a strong current also flows eastwards along the face of Stoura and Old
Man stacks which could have carried a part of the wreckage (most likely, a section of the
bows) in a semisubmerged state in that direction. The seabed in this area is very rocky
and heavily scoured, making it quite likely that any other remains settling there would
not survive. The corollary of this is, of course, that the wrecking took place during the
first part of the 4-hour period, say, between 16:00 and 18:00.
Obviously, the extent of the evidence, both archaeological and documentary,
available for the interpretation of the Kennemer/and wreck site is exceptional; that is
 The Archaeology of Shipwrecks 283

B
South

10
Sand Rock
and Lamm 15
shingle Boulders, little algae Hyp

20
too 300

one of the reasons why it is a particularly useful site on which to test these methods of
analysis. However, other wreckings can be cited in which the vessel has driven on
beyond the point of impact, breaking up as it proceeds, and shedding material in its
wake, so that its disintegration can be studied through the trail of remains stretching for
hundreds of meters across the seabed. Among British wreck sites, those of the Santa
Maria de la Rosa (Martin, 1975:23-135), Hollandia (Cowan et al., 1975), and Lovely
(Butland and Stubbs, 1976) would appear to be of this type (Muckelroy, 1977b:53-54).
It may be that this phenomenon is peculiar to large postmedieval ships, witli their spe-
cial techniques of construction and considerable ballast concentrated in one, particu-
larly vulnerable, part of the vessel. With most ships that break up on the surface this
linear arrangement of the material does not occur, and the result is usually a site of
class 4 or 5, according to seabed conditions (see Table 2) .

.5. SCRAMBLING DEVICES - PART II

Seabed Movement
The study of processes within seabed sediments is itself in a period of rapid
development, spurred on by increasing activity in coastal and continental-shelf areas,
and aided by technological developments associated with that work. However, the main
thrust of current research is into topics that have little or no bearing on archaeological
problems, such as the behavior of deep-water sediments, or the factors operating in
areas of substantial shoreline erosion or accretion. In partLcular, the behavior of mobile
sediments on rocky and irregular seabeds has been subject to virtually no systematic
study, and several workers in this field have expressed the hope that archaeological
data may in time contribute to their studies.
The basic concepts for such investigations can be derived from studies in marine
geomorphology (see especially Guilcher, 1958; Zenkovich, 1967; King, 1972). Sediment
disturbance is obviously primarily the result of water movement, i.e., by tidal currents or
wave action. The former can occur at any depth and with any force, according to circum-
stances, but the latter has more limited application. The swell reaching any particular
location depends on the speed and duration of any onshore wind, together with the dis-
284 K. Muckelroy

tance of open water over which it has traveled (i.e., the fetch) (King, 1972:53-60). But
wave energy decreases rapidly with depth, and at a depth of half the wavelength it will
be only 1I23rd of its surface strength (Guilcher, 1958: 19). Thus, wave-induced water and
sediment motion is principally a feature of shallow-water sites, the exact definition of
"shallow" depending on the maximum wavelength of the swell striking that spot. The
next problem is what size of sediment particles are disturbed by a given water velocity;
from experience underwater, it is evident that a current of several knots is required to
move even a coarse sand, and the forces needed to move gravel and shingle appear gen-
erally to be operative only in the surf zone. However, the movement of such coarser
material seems to be assisted frequently by a phenomenon known as "kelp rafting," by
which marine growths on a stone can act as a kind of sail, enabling it to be swept along
by a more moderate current (Jolliffe and Wallace, 1973).
Within most marine sediments, there is usually an upper layer of deposit, any-
thing between a few centimeters and several meters thick, which is in a state of
semisuspension, being lubricated by the surrounding water. Heavy objects will obvi-
ously tend to drop through this, while less dense items caught in it are likely to be thor-
oughly jumbled and heavily abraded. The analogy of topsoil on a land site is obvious.
One of the few research projects specifically designed to investigate these problems
was concerned with the sorting of objects within such a sediment. The area concerned
involved deep and accumulating silt in a Norwegian harbor, and the initial hypothesis
was that, as this deposit was apparently unconsolidated and presumably in a state of
constant flux, objects would be stratified according to specific gravities, with little or
no discrimination according to age (Keller, 1973). In fact, a test excavation suggested
that, on the contrary, the stratigraphy was directly related to chronology. While this
conclusion was obviously based on only one location, it does suggest that the re-sorting
of objects within such apparently mobile deposits can be overemphasized.
On some sites, it seems that all artifacts simply drop through the upper semimo-
bile level, so that it proves to be archaeologically sterile. This is the case on the Trini-
dad Valencera (1588) site, on which there is a layer of shelly beach sand about 25 cm
thick, often with small waves formed on its surface, within which there are no artifacts.
That objects can fall quickly through it to the stable levels below was shown by the
finding of a modem ring-pull beer can alongside a 16th-century timber at a depth of
about 30 cm. The spread of organic remains shown in Figure 2.12 (p. 51) began only a
few centimeters below the base of this beach sand. The implication of these observa-
tions is, of course, that there is a finite possibility of later contamination within the
uppermost archaeological levels on this site. A deeper semimobile layer has been iden-
tified on the Mary Rose (1545) site in the Solent, where the currents are appreciably
stronger and the sediment particles finer. Judging from the depth range of post-16th
century contamination, later objects could sink to a maximum depth of 70 cm, although
in this case wreck material was also discovered within this horizon, and no level in it
was totally devoid of artifacts. The base of this deposit was marked in some places by a
layer of matted organic material, apparently a growth of some kind of seagrass, which
effectively sealed-in the uncontaminated levels below.
On more shallowly buried sites, the matter of continued sediment movement is
obviously more critical. With the Dartmouth (1690) wreck site only the area beside and
under the hull itself seems to have remained totally undisturbed over the centuries,
because it was actually lying below the level of the surrounding clay substrate (see Fig-
The Archaeology of Shipwrecks 285

ure 5). That the overlying gravel was semimobile was shown clearly by the way it had
ground down the upper surface of the surviving structure. However, the forces acting
on this site must always have been relatively gentle, as the maximum fetch is only 18
km, and it is not scoured by any tidal currents, so that any gravel movement will have
been correspondingly moderate (Martin, 1978). Furthermore, as there are no reasons
for thinking that there has been any long-term gain or loss of sediment, it seems reason-
able to suppose that the net result of all of the movements has been no change, although
there may have been some loss of information in the distributions.
The sediments on the Kennemer/and site are of a similar thickness, and the ques-
tion also arises as to whether all or part of them have been subject to periodic rework-
ing over the centuries. Figure 8 shows some sections across one of the excavation
trenches (Site G), in which the second horizon down is of particular interest. It con-
sisted of a layer of matted wood splinters, organic remains, patches of wood tar, and

A 27 2 o 1 -13 B
I I

~f
021
03
Section
A B
at 13 3 metres

04

C 2ri----r----r--__r -___orl____r-~08 0

~lt
02
03
Secllon
C 0
at 10 8 metres

E14 2 26F
r--,----~--_r--_,----r_---r--~----~

o
01 Section
02 EF
03 at 88 metres
04
05

D Organic deposit G··:.•·.1 Loose gravel

D Loose splinters In silt D Light grey gravel

_ Tarred depOSit- splinters C'~: ~,J Fl(m gravel

_ Iron concrellon ~ Compacted yellow gravel

o Wood f' Cliff x Piece of paper

Figure 8. Three sections through the deposits in Site G, Kennemerland wreck site.
286 K. Muckelroy

other artifacts, which cannot have been substantially disturbed since initial deposition.
The origins of this layer are uncertain, but it is likely that it represented at least in part
the contents of the ship's bilges (Price and Muckelroy, 1977: 198). The implication is
that it, and the horizons below it, have remained undisturbed since the 17th century,
while the top level, of between 10 and 30 cm thickness, was the only one that had been
subject to disturbance. The state of preservation of the artifacts from the various hori-
zons supported this interpretation. A similar situation could be detected on four of the
other areas excavated (Sites B, D, E, and F; see Figure I of Chapter 42), while on two
of the others, which lay in more open situations (Sites A and C), there was simply a
homogeneous gravel, 20 to 40 cm thick, overlying the bedrock. Within this, there was a
tendency for artifacts to be found toward its base, and especially around larger rocks

/ Storm
winds

/ Tidal
Jf currents
/

1\1
Trolsome /

tJ I
lI-
/
/

I
\ I
I
r
I tJ

1:
I
I
I South
t Mouth

f
I

1 o
metres
100

Figure 9. A diagram summarising the principal natural forces acting on the Kennemerland site.
The Archaeology of Shipwrecks 287

and boulders. This last phenomenon has been observed on several sites, and suggests a
greater degree of artifact re-sorting over the centuries, the principal characteristics of
which demand further investigation.
Figure 9 illustrates the directions of the main forces acting over this site. Tidal
currents are indicated running from the north as this is the situation for over 8 hours in
every 12-hour cycle, and the reverse current is usually very weak, rarely exceeding half
a knot. The southward-flowing current can exceed three knots at times. As suggested
above, these currents will be dominant in deeper water, and their main thrust may
account for the southwestern trend of the distribution of bricks and other artifacts off
Stoura Stack. For the shallower deposits, however, the reverse wave-induced trend will
probably have been more significant. This site is sheltered totally from all directions
except the southerly quarter, which is thus the direction marked on Figure 9. Further-
more, meteorological observations indicate that storm winds (force 7 or above) blow
from these directions for 2.6% of the time, a far from negligible proportion (hourly
observations at Lerwick over the decade 1960-69, courtesy of the Meteorological
Office, Edinburgh). The net result of considering these forces is that the distributions
may have been "stretched" in a north-south direction, but not actually scrambled. It is
also possible that the contrary trends in shallow and deeper water may partly account
for the marked 150 m gap in the artifact distribution (see Figure 6).
On some sites this postdepositional re-sorting has affected the whole deposit, and
resulted in a total rearrangement of the remains. A clear example of this is represented
by the wreck site of the Adelaar (Dutch East Indiaman, 1728) on the west coast of
Barra in the Outer Hebrides (Figure 10) (Martin and Long, 1975). Exposed to the full

-
:,.:)
...... - ~:!~

~
-"" ,... ....
-

Figure I O. Site plan of the Adelaar wreck site off Barra, Outer Hebrides . Pdr, pounder.
288 K. Muckelroy

force of Atlantic swells, and with a fissured rocky seafloor, the only artifacts to have
survived were those of metal, and many of them were extremely battered and distorted
(an idea of the seabed conditions can be gained from Muckleroy, 1978:27,30). Experi-
ence during excavation showed that the meter or so of shingle filling the gullies was in
a state of constant flux, and only those items that eventually became embedded in the
layer of concretion below this stood any chance of survival (ibid., Figure 6).
As well as looking at site stratigraphy and general tidal and wave-generated water
movement, additional information concerning possible disturbances of the deposits may
be gleaned from other anciliary observations. For localized water movements on any par-
ticular part of a site, records made by a current meter may be of assistance, so long as it
is left in place for a sufficient length of time. Alternatively, an artificial scatter of spe-
cially selected objects representing a range of materials may be placed on the site after
excavation is completed, and then mapped at regular intervals in order to see how the
distribution evolves. However, both of these procedures suffer from the difficulty that
they will, if carried on for a sufficient length of time, record average on-site conditions,
while there are good reasons to believe that it is the exceptional, once-a-century storm
that achieves the most significant reworking of the seabed. A third approach, which may
partly allow for such eventualities, is to record the flora and fauna in each part of a site,
paying particular attention to those species that are tolerant of only a narrow range of
conditions, and can thus be regarded as indicators of overall environmental circum-
stances. Such observations have been made along the depth profile of the Kennemerland
site (Figure 7), and the occurrence of various types of algae, in particular, has supple-
mented the evidence from other sources about the degree of water movement. However,
maritime archaeologists would be deeply indebted to their colleagues in marine biology
for more detailed lists of indicator species for different parts of the world.
In addition to that caused by water movement, disturbance of the seabed may be
occasioned by the activities of marine creatures. The possible impact of crabs has
already been noted. In Mediterranean waters, the house-building instincts of the octo-
pus can result in accumulations of potsherds in suitable crannies, including amphorae;
this may have been the principal cause of the enigmatic distributions recorded on the
Sheytan Deresi site in southern Turkey (Bass, 1976:295-296). At present, there is no
evidence suggesting that marine worms might have similar archaeological conse-
quences similar to those of earthworms (Atkinson, 1957), and no burrowing creatures
analogous to moles or rabbits have been noted.
These observations have together indicated the importance of geomorphological
study to the interpretation of any wreck site, and have shown that the processes
involved are very different from those operating in a terrestrial context. Incidentally,
they have also shown that an underwater site can possess a complex stratigraphy whose
recording and interpretation is essential to the understanding of these processes, and
which may, as the experimental work in M6vik harbor has shown (Keller, 1973), even
have a chronological significance. A further implication for the analysis of seabed dis-
tributions is that if they have been generated by natural processes, as opposed to the
human-inspired ones usually involved on land sites, then the modes of analysis to be
used should perhaps be borrowed from the natural environmental sciences.
It was mentioned above that marine geomorphologists were looking to archae-
ological research underwater to furnish data on seabed processes, and especially the
behavior of sediments in rocky areas. In this context a great deal of the above discus-
The Archaeology of Shipwrecks 289

sion is of potential value - above all, the indications of sediment stability over several
centuries for some horizons and the suggestion that, in other areas, the net results of
any localized disturbances have been minimal. Another type of contribution is illus-
trated by the late 18th-century wreck located in 1974 in the South Edinburgh Channel
in the Thames estuary. This channel, whose precise location and configuration is
always shifting, has recently begun to move across the wreck site, with the result that
the hull has broken open and begun to disintegrate; its previous pristine state demon-
strates that the channel has not occupied this area at any time over the past 200 years,
knowledge that the sedimentologists could not have gathered in any other way. Simi-
larly, the remains of the Mary Rose (1545) illustrate the geological and biological his-
tory of that part of the Solent over the past 400 years (McKee, 1973). The next few
decades should see an increasingly fruitful exchange of ideas and information between
these disciplines.
In concluding this discussion of the various aspects of Figure I, note must also be
taken of the extracting and scrambling consequences of the excavation strategy adopted
for any particular site, consequent on the considerations outlined in Muckleroy (1978,
ch. 2). Coming between the seabed distribution of remains as it actually exists, and as it
appears in the archaeological record, this factor must be taken into account; it is thus
the responsibility of any archaeologist to indicate honestly in his report the procedures
used, standards of accuracy achieved, precise areas excavated, and so forth, so that
their implications can be generally appreciated. Having considered this, and all of the
other aspects of the wrecking process on a particular site, the researcher is in a position
to consider how the observed remains on the seabed should be interpreted.

REFERENCES
Atkinson, RJ.C., 1957, Worms and Weathering. Antiquity 31: 219-233.
AUSAC (Aston University Sub-Aqua Club), 1974, The Wreck of the 'Kennemerland. 'Aston University, Bir-
mingham.
Bass, G.F., 1976, Sheytan Deresi: Preliminary Report. International Journal of Nautical Archaeology and
Underwater Exploration 5: 293--303.
Brand, Rev. J., 1701, A Brief Description of Orkney. Zetland. Pightland Firth and Caithness. Edinburgh.
Bruce, R.S., 1907, Some Old-Time Shetlandic Wrecks, Part 2. O/f Lore Miscellany I 4: 123--128.
Butland, WE., and Stubbs, I.M., 1976, Survey and Excavation of the Sloop Lovely. Maritime Wales I: 51-62.
Clarke, D.L., 1968, Analytical Archaeology. Methuen, London.
Cowan, R., Cowan, Z., and Marsden, P., 1975, The Dutch East Indiaman Hollandia Wrecked on the Isles ofScilly
in 1973. International Journal ofNautical Archaeology and Underwater Exploration 4: 267-300.
Davis, R.H., 1955, Deep Diving and Submarine Operations, 6th ed. Siebe Gorman, London.
Dumas, F., 1962, Deep- Water Archaeology. Routledge and Kegan Paul, London.
Dumas, F., 1972, Ancient Wrecks. In UNESCO, Underwater Archaeology, a Nascent Discipline, pp. 27-34.
UNESCO, Paris.
Ericsson, C.H., 1975, An Eighteenth Century Diving-Suit from Brahestad in Finland. International Journal of
Nautical Archaeology and Underwater Exploration 4: 130-134.
Forster, WA., and Higgs, K.B., 1973, The Kennemerland, 1971; and Interim Report. International Journal of
Nautical Archaeology and Underwater Exploration 3: 291-300.
Forst, H., 1962, Submarine Archaeology and Mediterranean Wreck Formations. Mariner's Mirror 48: 82-89.
Glamann, K., 1958, Dutch-Asiatic Trade. Danish Science Press, Copenhagen.
Guilcher, A., 1958, Coastal and Submarine Morphology. Translated by B.W and Sparks, R.H.W. Kneese,
Methuen, London.
Hamilton, D.L., 1976, The Conservation of Metal Objectsfrom Underwater Sites; a Study in Methods. Texas Me-
morial Museum, Miscellaneous Paper No.4, and Texas Antiquities Committee, Publication No. I, Austin.
290 K. Muckelroy

Hiscock, K., 1974, Ecological Surveys of Sublittoral Rocky Areas. Underwater Association 8th annual report,
pp.46--65.
Jolliffe, I.P., and Wallace, H., 1973, The Role of Seaweed in Beach Supply, and in Shingle Transport below Low
Tide Level. In Science Diving International, edited by N.C. Flemming, pp. 189--196. British Sub-Aqua
Club, London.
Keller, c., 1973, Stratification Problems in Norwegian Harbours. International Journal of Nautical Archaeology
and Underwater Exploration 2:' 187-189.
King, C.A.M., 1972, Beaches and Coasts, 2nd ed. Edward Arnold, London.
Lamboglia, N., 1961, Le navi roman a di Spargi. Proceedings of the 2nd International Congress on Underwater
Archaeology, pp. 124--131. Bordighera.
McKee, A., 1973, King HenlY VIII's 'Mary Rose.' Souvenir Press, London.
Martin, C.J.M., 1975, Full Fathom Five. The Wrecks of the Spanish Armada. Chatto and Wind us, London.
Martin, C.J.M. 1978, The Dartmouth, a British Frigate Wrecked off Mull, 1960. 5. The Ship. International
Journal of Nautical Archaeology and Underwater Exploration 7: 29--58.
Martin, C.J.M., and Long, A.N., 1975, The Use of Explosives on the Adelaar Site 1974 International Journal
of Nautical Archaeology and Underwater Exploration 4: 345-352.
Muckelroy, K.W., 1976, The Integration of Historical and Archaeological Data Concerning an Historic Wreck
Site: The Kennemerland. World Archaeology 3: 280--290.
Muckelroy, K.W., 1977a, Historic Wreck Sites and Their Environments. In Progress in Underwater Science,
edited by K. Hiscok and A.D. Baume, pp. 111-120. Pentech Press, London.
Muckelroy, K.W., 1977b, Historic Wreck Sites in Britain and Their Environments. International Journal of
Nautical Archaeology and Underwater Exploration 6: 47-57.
Naish, G.P.B., 1968, The 'Wasa', Her Place in History. HMSO, London.
Nesteroff, W.D., 1972, Geological Aspects of Marine Sites. In UNESCO, Underwater Archaeology, a Nascent
Discipline, pp. 175-183. UNESCO, Paris.
Price, R., and Muckelroy, K., 1974, The Second Season of Work on the Kennemerland Site, 1973; an Interna-
tional Report. International Journal of Nautical Archaeology and Underwater Exploration 3: 257-268.
Price, R., and Muckelroy, J., 1977, The Kennmerland Site; the Third and Fourth Seasons, 1974 and 1976. An
Interim Report. 6: 187-218.
Siegel, S., 1956, Nonparametric Statistics for the Behavioural Sciences. McGraw-Hili, New York.
Throckmorton, P., 1970, Shipwrecks and Archaeology; the Unharvested Sea. Victor Gollancz, London.
Tylecote, R.F., 1977, Durable Materials for Sea Water: The Archaeological Evidence. International Journal of
Nautical Archaeology and Underwater Exploration 6: 269--283.
Uhlig, H.H., 1948, The Corrosion Handbook. Electrochemical Society, New York.
UNESCO, 1972, Underwater Archaeology, a Nascent Discipline. UNESCO, Paris.
Weier, L.E., 1974, The Deterioration of inorganic Materials under the Sea. Institute of Archaeology Bulletin
II: 131-163.
Zenkovich, v.P., 1967, Process of Coastal Development. Translated and edited by D.G. Fry, 1.A. Steers, and
C.A.M. King. Oliver and Boyd, Edinburgh.
 Chapter 26

'.~~.' Cultural Resources

~

o i' .
Magnetometer Survey
q;,O_' .': .
.j'~l and Testing

ROBERTL.GEARHART

MAGNETOMETER SURVEY METHODS

The Ocean Beach magnetometer survey was performed by a crew of three EH&A
archaeologists, during late September and early October, 1987. Autumn was chosen as
the optimum time for conducting the survey, as the beach is at its maximum width, and
weather conditions are most favorable at this time of year.
Instrumentation used for the survey included an EG&G Geometrics Model G-856
proton precession magnetometer as a field instrument, an EG&G Geometrics Model
G-866 magnetometer as a base station, and a Toshiba T1100+ laptop personal com-
puter for data storage and processing. The G-856 magnetometer is powered by eight
D-cell batteries and mounts in a shoulder harness for operation by a single individual.
The G-856 sensor is designed to be carried in a pouch at the back of the operator's har-
ness or at the end of an aluminum staff. During the survey, the G-856 sensor was
mounted on a staff at a height of 6 ft. The G-866 base station magnetometer was pow-
ered by two 12-volt, wet-cell batteries. The G-866 sensor was mounted on a wooden
stand at a height of 4 ft.
Source: Cultural Resources Magnetometer Survey and Testing: Great Highway/Ocean Beach Seawall Project,
San Francisco, California (1988), Espey, Huston & Associates, Inc. Austin, Texas. pp. ~12, 39-45. Reprinted
by permission.

291
292 R.L. Gearhart

Base station readings were recorded by the G-866 magnetometer at 5-second inter-
vals throughout the course of each day. By interfacing the G-866 with the Toshiba port-
able computer, base station readings could be automatically stored on floppy disk as an
ASCII data file. The computer was operated from the same power source as the G-866.
Likewise, readings stored in the internal memory of the G-856 survey magnetometer dur-
ing the course of the day were transferred to computer disk each evening, at which time
initial data processing and plotting was performed to ensure the quality of the data.
The survey was conducted on a controlled survey grid with measurements of the
magnetic field recorded at uniformly spaced intervals along parallel transects. Tran-
sects were oriented approximately perpendicular to the line of the surf. Grid spacing
(i.e., the distance between adjacent magnetometer readings) was 6 ft in the vicinity of
the King Philip and Reporter wreck sites, and 12 ft for the remainder of the survey
area, including two small sections of riprap-covered slope which were surveyed (east-
ern margin of survey area). The area suspected to contain wreckage from the Atlantic
was later resurveyed at a 6-ft interval, providing resolution equal to that obtained for
the King Philip and Reporter.
Horizontal controls for the survey grid were tied to the City of San Francisco's
project datum line, referred to as the "C"-line, by surveyors from the City's Survey
Division. A temporary wooden hub, set into the beach at the foot of Ortega Street by
the city's surveyors, was designated as a working datum for the survey. This point,
located 6765.61 ft north along the "C"-line and 202.43 ft west of the "C"-line, was
given the designation of North 900/Uest 100 (N 900/U 100) on an arbitrary grid sys-
tem. As a convenience, each whole-number increment along the survey grid lines was
assigned a value of 6 ft, thus the distance between N 900 and N 800 is 100 grid units or
600 ft. Use of this whole-number system for designating grid coordinates not only sim-
plified recordkeeping, but also greatly reduced the chance for error in conducting the
survey. Appendix A (unattached) contains a complete list of city coordinate system
equivalencies (based on the "C"-line) for each of the survey grid coordinate pairs.
Using these equivalencies, any data point within the survey area can be accurately relo-
cated on the beach by surveying from a permanent city datum.
Two parallel baselines, one located at W 100 and the other at W 120, were estab-
lished on the beach using a Wild T-2 theodolite. All distance measurements were taped
using Keeson fiberglass survey tapes. These baselines, oriented approximately parallel to
the "C"-line, were marked on the beach using 3-ft wooden laths, hammered into the sand
to their full length, at 120-ft (20-unit) intervals. Survey transects were established by
stretching measuring tapes perpendicular to the baselines (approximately east/west),
from the base of the riprap, out into the surf. Points of intersection between survey tran-
sects and the baselines were demarcated by placing 3-ft wooden dowels along the baseli-
nes at 6- or 12-ft intervals between the laths. The dowels served a dual purpose: first, as
points of reference for positioning the transect tapes, and second, as anchor points for the
measuring tapes so that they could be positioned along the survey transects by a single
person. When surveying the surf zone and the area between the W 100 baseline and the
riprap, baseline dowels were also used as sighting points for positioning the western end
of the tapes. Horizontal control and recordkeeping were greatly facilitated by marking
the survey tapes at 6-ft intervals with red and blue spray paint, alternately.
During the course of a typical day's survey, the G-856 magnetometer was oper-
ated by one crew member on a nearly continual basis. Magnetic field intensity was
Cultural Resources Magnetometer Survey and Testing 293

recorded at 6- or 12-ft intervals along each transect (depending on the section of the
beach being surveyed) using the G-856. The other two crew members shared the con-
stant responsibilities of recording the beginning and ending coordinates of each tran-
sect, holding the tape out of the surf at the western end of each transect, repositioning
the tapes for each subsequent transect, and monitoring the base station.
Operation of the base station generally involved little attention during the course
of the day, once it was set up and running at the start of each morning. The location of
the base station was varied throughout the course of the survey for security reasons, as
the project area was nearly 3000 ft long and had a substantial amount of pedestrian traf-
fic. Locations for setting up the base station were, therefore, chosen reasonably close to
where we planned to be surveying on any given day. The base station sensor, as men-
tioned above, was mounted 4 ft above the sand on a wooden stand. The sensor was
positioned as close to the high tide line as possible,- without running the risk of being
knocked over by a large wave.
Prior to initiating the survey, an experiment was conducted to determine what
effects, if any, passing motor vehicles might have on the background magnetic field of
the beach. By varying the distance of the base station sensor from the highway and
marking the pass of each car and truck on the G-866's paper strip chart (Figure I), it
was determined that traffic had no noticeable effect on the local magnetic field at dis-
tances greater than 130 ft from the highway. This is in agreement with Breiner's
(1973 :46) calculation that a I-ton automobile at a distance of 100 ft should only pro-
duce a I-gamma deviation in the magnetic field. To be conservative, the base station
sensor was positioned at a distance of at least 130 ft from the highway throughout the
course of the survey. Test strip charts were also run daily on. the G-866 (Figure 1), in
order to determine whether the level of solar-induced magnetic activity was acceptable
to allow the collection of meaningful survey data.
On completion of the survey at the original 6- and 12-ft intervals, a mosaic mag-
netic contour plot, in variable density dot-matrix format, was pieced together for the
entire project area. A preliminary analysis of the resulting map revealed several inter-
esting anomaly patterns believed to be potentially caused by the presence of buried
wreckage. As mentioned above, selected portions of the 12-ft survey area encompass-
ing these patterns were resurveyed at a 6-ft sample interval, in order to increase the
resolution of the data to that achieved by the survey of the King Philip and Reporter
wreck sites.

DATA PROCESSING AND CONTOUR MAPPING

Initial processing of magnetic data was performed on a daily basis. The MAGPAC
computer software package, developed by the staff of EG&G Geometries (Sunnyvale,
California), was used for all data transfer and processing conducted while in the field. On
completion of each day's survey, the corresponding data were transferred, in ASCII for-
mat, from the internal memory of the G-856 to the 3.5-inch computer floppy disk holding
that particular day's base station data. Because of the large number of files generated
during the course of processing the data, it was found convenient and often necessary to
store each day's data files on a separate floppy disk. The first step performed in prepara-
tion for displaying the data was to remove the diurnal variation from the survey data
using the "Diurnal" mode of the MAGPAC software package. Once this was accom-
 294 R.L. Gearhart

"

A r l l ' of traffic nat . . effect en backQraund
maQnet ic f .. ld ; ..n.or lacated approl-
i matel, 110 , . . t from hiQh_a)'
c : car t • truck
(10 QOm",a diwl.ion between lin,.)
B r,pical Bas. Station Strip Chart
show i nQ normal wariatian in back-
Qround maQnetlc 'ield
14 Qamma diwl,ian between IonlS

Figure I. G-866 base station test charts.

Cultural Resources Magnetometer SUlvey and Testing 295

DATA FILE Bll0-2FLD2.REV

LINE NUMBER 259 reversed.
TiN Sh ",g 50180. 0 50660.0
13:30:25 1835 50562.5
Rip Rap Anom~IL.J-----1
.----+-'[._-----_. ._. .
13:29:34 1830 50333.5

13:26:" 1820 50493.9

13:25: 2 1810 50486.9

13:23:23 1800 50559.3

13:21:43 1191 50441.9

Figure 2. Magnetic profile of transect across King Philip bow anomaly.

plished, survey grid coordinates were inserted into the corrected data file, relying on the
daily field notes for this information. Finally, the sequence of the data from alternate
transects was reversed, as adjacent transects were surveyed in opposite directions. At this
point, the diurnally corrected and properly ordered data file was ready to be displayed
either as magnetic profiles of individual transects (Figure 2) or as a variable density dot-
matrix contour map (Figure 3). Both forms of data display were used as interpretive aids
during the course of fieldwork.
On returning to the offices of EH&A in Austin, Texas, all diurnally corrected
data files were combined and transferred as a single ASCII file to EM&A's Digital
Equipment Corporation VAX, mainframe computer system. Using an Intergraph com-
puter-assisted-drafting work station and a Digital Terrain Modeling (DTM) engineering
software package, magnetic line-contour maps were generated from the combined data
files for the entire survey area. The DTM contouring program selected for use in this
study utilizes a triangle interpolation routine which honors the value of each data point,
sacrificing some degree of contour smoothness for accuracy. In other words, contour
lines are never drawn on the wrong side of a data point, as is allowed by some grid
interpolation routines. A comparison of a sample computer-generated contour map,
produced by this method, with a manually interpolated contour map of the same data
set, hand-drawn by the principal investigator, indicates that the triangle interpolation
routine is extremely well suited to this application, producing a map that is a precise
and accurate reflection of the data.
 ~

0\
'"

I LocOIlon of lronseei Profile

• IlIusl,oled in Fi9ure 14

----2_
KING PHILIP

:;tI
t-
O
(\)

Figure 3. Variable density dot,matrix contour map of King Philip and Reporter magnetic signatures.
3-
~
Cultural Resources Magnetometer Survey and Testing 297

Initial contouring of the data revealed a degree of inconsistency between adjoin-

ing data files that had been created on different survey days. This was especially
noticeable when the location of the base station had changed from one day to the next.
The result was that the magnetic data recorded on any given day contoured normally,
while a sharp bend in the contour lines often occurred at the break between one day's
survey and the next. This problem was solved by adding a unique correction factor to
the magnetic readings within each day's data files, effectively adjusting the value of the
background magnetic field component, so that the total magnetic field intensity of all
data points was comparable. Correction factors were determined manually by interpo-
lating between adjacent transects that had been surveyed on different days. The cor-
rected values are provided in Appendix A, as these correlate with the contoured data.

INTERPRETIVE METHODS

The interpretation of magnetic data from Ocean Beach was conducted with three
objectives in mind. First, an attempt was made to identify all magnetic anomalies
resulting from visible sources within the project area, so that these data could either be
ruled out of consideration for potential significance or be taken into account for the
masking effect that they may have on other anomalies. Second, based on a predictive
model that describes magnetic anomaly patterns expected to result from various types
of wreck sites, areas of potential historic significance were delineated. Third, where
possible, the depth of anomaly sources was estimated, in order to determine whether
they are buried too deeply in the sand to be affected by construction of the seawall or
are too shallow to be considered potentially significant. A description of the methods
employed to accomplish each of these objectives is provided below, including a sum-
mary of the predictive model developed specifically for this project.

VISIBLE ANOMALY SOURCES

Because of the high incidence in the project area of modem surface materials
capable of producing magnetic anomalies, a large amount of data contamination was
expected to occur. Unless accounted for, extraneous anomaly sources could greatly
confuse the interpretation of data. This was especially true along the riprap-covered
slope at the top of the beach. The location and nature of all visible anomaly sources
within the project area were, therefore, recorded during the course of the survey. When
possible, these sources were physically removed prior to surveying an area. Visible
anomaly sources were, by far, most often associated with the riprap occurring on the
slope between the Great Highway and the beach. Anomaly sources located on the slope
generally consisted of concrete rubble (small amounts of which contained steel rein-
forcement bar), discarded granite building stones, and igneous boulders. As it was not
possible to remove these materials prior to conducting the survey, careful note was
made of the relative abundance of each type of material along the entire length of the
slope through the project area. In addition, the face of the slope was photodocumented
in 120-ft sections to allow a visual correlation of magnetic anomalies with the material
occurring as riprap on corresponding sections of the slope. Other features, noted during
the course of the survey and mapped as possible anomaly sources, include: all timbers
lying on or protruding from the sand (note was made in each such case as to whether
298 R.L. Gearhart

iron fastenings were present); the remains of intact beach hearths; and major topo-
graphic features, such as winter high-tide berms and beach cusps. On completion of a
line-contour map, the locations of visible anomaly sources were plotted on a working
copy and visually correlated with magnetic anomalies. A majority of the high gradient
anomalies, located near the slope base, were accounted for in this manner.

ANOMALY PATTERN RECOGNITION

Very few detailed studies, such as this one, have been conducted in which mag-
netic contour maps have been produced of known shipwreck sites, using tightly control-
led and closely spaced survey points. Such studies would be helpful in the interpretation
of the Ocean Beach magnetometer data. Fortunately, two known examples of historic
shipwrecks, the King Philip and the Reporter, are present within the survey area. Infor-
mation regarding these vessels has been invaluable to the interpretation of anomaly pat-
terns occurring elsewhere on the beach. For example, in the case of the King Philip, the
composition, orientation, depth of burial, and general state of preservation of her hull are
known. Less information is available for the Reporter; however, from historical docu-
mentation, we know the composition of her hull, the vessel's location and orientation
prior to disintegrating in the surf, and a few of the details regarding the manner in which
she broke apart. The magnetic anomaly patterns of these two wreck sites have been util-
ized extensively in the interpretation of data from the remainder of the survey area.
In addition to the use of the King Philip and Reporter wreck sites as comparative
examples, the interpretation of data from Ocean Beach has relied heavily on a number
of expectations, or working hypotheses, regarding the nature of anomaly patterning for
wreck sites occurring in a beach environment. Delgado and Murphy (1984), in their
classification scheme for environmentally exposed shipwreck sites (i.e., wrecks occur-
ring and remaining exposed in dynamic environments, such as beach or nearshore set-
tings), recognized three distinctive site patterns which they referred to as buoyant hull
sites, buoyant hull fracture sites, and buoyant structure sites. Buoyant hull sites are
those, such as the King Philip, in which a ship comes ashore and settles into the sand,
relatively intact. Buoyant hull fracture sites occur when the hull of a ship comes ashore
intact, but breaks apart on the beach and is dispersed by the surf. The wrecks of the
Atlantic and the Reporter fall in this category. Buoyant structure sites form when a ship
breaks apart offshore and washes onto the beach in pieces, as did the Aberdeen. Del-
gado and Murphy's terminology is followed throughout this report.
The following sections comprise a preliminary predictive model of anomaly pat-
terns expected from each of Delgado and Murphy's (1984) three types of environmen-
tally exposed shipwreck sites: buoyant hull, buoyant hull fracture, and buoyant
structure wreck sites. Magnetic anomaly patterns, observed within the survey area,
which resemble either the pattern of the King Philip or Reporter wreck sites, or one of
the following hypothesized patterns for environmentally exposed wreck sites, were
interpreted as having a potential for historic significance.

BUOYANT HULL

The anomaly pattern of buoyant hull wrecks, such as the King Philip, may differ
from one wreck to the next, reflecting differences in the materials used for construc-
Cultural Resources Magnetometer Survey and Testing 299

tion. Certainly there would be some dramatic differences between that of a steel-hulled
versus a wooden-hulled vessel. Two important characteristics might be expected, how-
ever, for the majority of buoyant hull anomaly patterns. First, we would expect a linear
distribution of multiple anomaly peaks within the overall pattern produced by the
remains of an intact hull. In the case of a wooden-hulled vessel, the anomaly pattern
may appear as a complex, elongated anomaly, containing areas of high and low mag-
netic intensity within its boundaries. The long axis of the anomaly pattern should be
oriented along the same heading as the long axis of the hull. A classic example of this
type of signature is that of the wooden-hulled steamboat Bertrand, illustrated by Petsch
(1974:Figure 18). Arnold (1987:35), similarly, observed the occurrence of multiple
peaks in the anomaly pattern of a concrete-hulled vessel. As it turns out, the anomaly
pattern of the King Philip also exhibits an extremely definitive linear pattern.
Second, the long axis of the anomaly pattern may often be oriented parallel to the
surfline, reflecting the tendency of a drifting hull to tum broadside to the waves. This,
of course, depends on whether the anchor lines are released when a vessel makes con-
tact with the beach, as was done in the case of the King Philip. Failure to release the
anchor lines may keep a vessel at a right angle to the surf and, by holding the hull in
the most violent portion of the surf zone, increase the chances of it breaking apart.
Similar circumstances apparently contributed to the rapid disintegration of the Atlantic
(San Francisco Daily Examiner, 18 December 1886).

BUOYANT HULL FRACTURE

When a ship's hull breaks apart soon after becoming beached, the potential for
anomaly sources, such as iron fittings or magnetic ballast material, to become scattered
away from the main body of the wreck is much greater than for a buoyant hull. Waves
of sufficient energy to break up a ship might also be expected to distribute vessel ele-
ments high onto the beach. At the same time, littoral drift would tend to transport buoy-
ant elements downcurrent. We would, therefore, expect a buoyant hull fracture to
consist of a pattern of mUltiple anomalies (i.e., wreck scatter) radiating upslope and
downcurrent from an area of more tightly clustered, higher-intensity anomalies (i.e., the
area of hull breakup). Smith et al. (1987), for example, observed a landward scatter of
wreckage from the fractured remains of a Civil War period anti torpedo raft on Mustang
Island in Texas.
Buoyant hull fractures should be distinguishable from buoyant hulls, as a result of
horizontal diffusion of the complex, linear anomaly pattern often produced by intact
hulls, together with a pattern of lower intensity anomalies radiating inland and down-
current from this source. Two buoyant hull fractures are known or suspected to exist
within the project area. Portions of the Reporter wreckage have been previously
located, intermingled with the wreckage of the King Philip, while the Atlantic report-
edly washed ashore 1000 ft south of the King Philip, subsequently scattering debris
over a mile of beach.

BUOYANT STRUCTURE
Ships, such as the Aberdeen, that break apart offshore and then wash onto the
beach, may leave a trail of wreckage scattered along miles of shoreline. While larger
300 R.L. Gearhart

iron components (e.g., the steam engine of the Aberdeen) may tend to remain offshore,
this would depend on whether such elements remained attached to large, buoyant sec-
tions of the vessel. The possibility of magnetically distinguishing a wreck of this nature
from other sources of flotsam would depend on the size and number of vessel elements
that washed ashore and on the area of their dispersal over the beach. Unlike the anom-
aly pattern of a vessel that breaks up after coming ashore, the pattern of a buoyant
structure wreck is expected to consist of nonclustered anomalies, of varying intensities,
scattered unevenly across the beach.
Photographs taken of the Aberdeen wreck site in 1916 illustrate a feature of buoy-
ant structure wreck sites, which may also be common to buoyant fracture sites, that is,
the distribution of smaller wreckage, or flotsam, along a strandline. It is expected that
heavier pieces of wreckage, and presumably larger anomalies, would tend to be located
closer to the surfline, while lower-intensity anomalies may be concentrated higher on
the beach, in association with the high-water mark (strandline). Provided that this mate-
rial was not completely scavenged for lumber and firewood before it could become
buried, strandline wreckage from either form of wreck site may result in a linear pattern
of relatively weak anomalies, oriented parallel to the surfline. This, of course, would
depend on the amount of ferrous material associated with the flotsam.
In comparison with buoyant fracture sites, it is also conceivable that higher inten-
sity anomalies would tend to be located farther out into the surf zone on buoyant struc-
ture sites. The reasoning behind this statement is that elements of a buoyant fracture
wreck come ashore while still part of an intact buoyant hull. For as long as the hull
remains intact, high tides may continue to carry the vessel farther inland. The fractured
remains of a buoyant structure wreck, on the other hand, may be less buoyant to begin
with than an intact hull, and perhaps more susceptible to becoming embedded in the
sand, because of the presence of broken and protruding elements.

ESTIMATION OF ANOMALY SOURCE DEPTH

Two methods available for estimating depth of anomaly sources were utilized
during the course of the data interpretation. These include the vertical gradient method
(Breiner, 1973:52-53) and a "slope" technique (Breiner, 1973:31). Estimation of depth
to an anomaly source using the vertical gradient method requires that two magnetome-
ter readings be taken at varying sensor heights over the point of an anomaly'S highest
intensity. Sensor heights used to perform this exercise during the current study were 6
ft and 2 ft, respectively. By subtracting the two readings obtained in this manner, then
dividing this difference (dT) by the difference in sensor height (dz), the latter being 4 ft
in this case, the vertical gradient (dTldz), or falloff rate, of an anomaly can be deter-
mined. An approximate maximum depth estimate can then be obtained by the formula
for depth of a dipolar anomaly source, z == 3 T/(dTldz), where z is the estimated vertical
distance from the sensor, T is the intensity of the residual anomaly (i.e., an anomaly'S
intensity relative to the local background field intensity), and 3 is the falloff factor for a
dipolar anomaly. Monopolar anomalies have a falloff factor of 2 which results in a
shallower estimate of depth using this method. By subtracting the height of the magne-
tometer sensor, used in the determination of T (in this case, 6 ft), from the value
obtained for z, the approximate maximum depth to an anomaly source, below the sur-
face of the beach, is estimated.
Cultural Resources Magnetometer Survey and Testing 301

Slope techniques for estimating source depth involve measuring the horizontal
extent of an anomaly's gradient (i.e., slope) between points of inflection along the
slope of the anomaly (Breiner, 1973:31). Measurement of the horizontal distance
between slope inflection points (see Breiner, 1973 :31; illustration) can be accom-
plished in two ways. The first involves drawing a profile of the anomaly along its axis
(i.e., the line between its points of highest and lowest magnetic intensity). This is the
equivalent of cross-sectioning a three-dimensional plot of the anomaly along the axis of
its poles and only serves to emphasize the location of inflection points. The second
method simply involves determining the points of inflection, along the polar axis of an
anomaly, from inspection of the contour map, then measuring the distance between
these two points with a ruler. The latter method sacrifices little in accuracy over the
more time-consuming method of drawing an anomaly profile, as the horizontal scale of
any profile is the same as that of the contour map from which the profile was projected.
Once obtained, the horizontal distance between inflection points is multiplied by an
empirically determined factor ranging from 0.5 to 1.5 in order to estimate the depth of
the source. A value of 1.5 was selected for use here, based on a comparison of the King
Philip's actual depth with estimates that were calculated using both the vertical gradi-
ent and slope techniques.

REFERENCES

Arnold, J.B., III, 1987, Marine Magnetometer Survey of Archaeological Materials Found Near Galveston.
Texas. Texas Antiquities Committee, Publication No. 10.
Breiner, S., 1973, Applications Manual for Portable Magnetomers. GeoMetries, Sunnyvale, CA.
Delgado, J.P., and Murphy, L., 1984, Environmentally Exposed Shipwreck Remains: Implications for a Natural
Site Formation Process. Paper read at the Fifteenth Annual Conference on Underwater Archaeology,
Williamsburg, VA.
Petsch, J.E., 1974, The Steamboat Bertrand: History, Excavation, and Architecture. National Park Service.
Publications in Archaeology II. U.S. Government Printing Office, Washington, DC.
Smith, H.A., Arnold, J.B., III, and Oertling, T., 1987, Investigations ofa Civil War Anti-Torpedo Raft on Mustang
Island, Texas. The International Journal of Nautical Archaeology and UndelWater Exploration 16 (2):
149-157.
 Chapter 27

A Diachronic Study
".~i).:
~

of Some Historical and

-r/: ..
':.i
Natural Factors Linked to
Shipwreck Patterns in the
Northern Gulf of Mexico

ERVAN G. GARRISON

INTRODUCTION

The observed distribution or pattern for historic shipwrecks is a product of a complex

of historical and natural factors. These factors include imperialism, commerce, warfare,
and technological change, as well as natural phenomena such as equinoctial storms,
currents, winds, shoals, and reefs. It is the purpose of this study to evaluate some of
these factors over the chronological period ranging from the 16th to 20th centuries rela-
tive to their casual linkage to shipwreck patterns. It is expected that these factors will
differentially influence the location and density of shipwrecks in the northern Gulf of
Mexico.
Beginning with the Coastal Environments, Inc. (CEI) (1977) study, we examined
that compendium of shipwrecks. This led to the same conclusion reached by those
authors, i.e., data for historic shipwrecks developed principally from secondary sources
Source: Underwater Archaeology Proceedings from the Society for Historical Archaeology CO'1ference. edited
by J. Barto Arnold (1989), pp. 12-18. Reprinted by pennission of the Society for Historical Archaeology, Tuc-
son, Arizona.

303
304 E.G. Garrison

have several defects. The principal flaw is validity. The most valid reports on ship-
wrecks are primary sources, such as news accounts, official reports, logs or other direct
observations of the specific shipwreck. The problem with primary sources is their lack
of consolidation. To adequately research all primary source data for historic shipwrecks
was beyond the resources of this study, as it was for the CEI study. We attempted to
examine a collection of primary sources or facsimiles of these materials in a limited
number of archives and libraries. At the outset, we further restricted the study to only
those archives in the United States, with the exception of the National Archives of
Mexico (AGN).
The location of earlier studies of the materials in these foreign archives was suc-
cessfully sought out in several instances. For Spanish shipwrecks, excellent sources
were found in studies by researchers of the National Library of France (Bibliotheque
Nationale, Paris) (Chaunu and Chaunu 1955), the P.K. Younge Library of Florida His-
tory, Gainesville, Florida, and newly printed catalogs of the holdings of AGN. For
French shipwrecks of the colonial period, facsimile microfilm of the correspondence
was found in Archives Nationale, Colonies, Series 13, located at the Howard-Tilton
Library, Tulane University, and Paris.
British losses were sought in similar facsimile data of the London Board of
Trade, Lloyds, Admiralty and Foreign Office reports located principally at the research
library of the Mariners Museum, Newport News, Virginia. American shipwreck data
appeared in a variety of sources found at the U.S. National Archives and its branches;
further copies were located at Mariner.
Also used was a source unavailable to earlier researchers, namely, computer-
based data files. Such files have contributed to the creation of the Hangs and Obstruc-
tions File by the Hydrographic Office (HO), the Automated Wreck Obstruction
Information Service file (A WOIS) of the National Ocean Survey, the Historic Ship-
wreck File of the Texas Antiquities Committee (T AC), and the Florida Shipwreck Fife
of the Division of Archives and History, State of Florida.
The data in the computer fields were merged with those of the primary and secon-
dary materials collected at the various archives, and a master file of historic shipwrecks
of the northern Gulf of Mexico was created. This file has over 4000 entries (Figure I).

,c

29

r- 28
~
~ 21
o
m
26

2-

-;-': 96 94 92 90 88 86 84 82 80
LONGITUDE

Figure 1. All shipwreck positions.

Factors Linked to Shipwreck Patterns in the Northern Gulf of Mexico 305

SPECIFIC FACTORS AND SHIPWRECK PATTERNS

This study examines five principal factors affecting shipwreck locations and sub-
sequent patterning in their distribution. These are:
1. Historic shipping routes
2. Port location
3. Shoal, reef, sand bar, and barrier island locations
4. Ocean currents and winds
5. Historic hurricane paths
These factors do not account for all of the shipwreck locations in the northern
Gulf, but clearly they are important to understanding the distribution of shipwrecks and
to developing explanatory models for shipwreck distribution.

Sailing Route: Open Sea, Channels, and Coastal

In a perfect scheme, all losses would be along sailing routes and at nodes of a net-
work. The actual correlation of shipwreck sites to sailing routes is difficult. The distri-
bution of shipwrecks and the location of sailing routes for a given period are linked
variables in that they can predict the behavior of either to a finite and measurable de-
gree. What this level of predictability is remains to be assessed. Sailing routes were of
particular importance, both in a navigational and a strategic sense. During the Spanish
era of exploration, these routes were defined by trial and error. This information was
gained through the loss of ships. Even so gained, the early navigator of the Spanish
Gulf was restricted to a few principal routes determined by the westerlies outbound to
the New World and the tack against them in return, using the Gulf Loop Current
(Rezak et aI., 1985: 24-26) to reach the Gulf Stream (Hoffman, 1980: 5--7). In the Gulf,
Spanish period and later ships braced the Yucatan Channel and thence the Straits of
Florida, i.e., a great arc from New Spain, to near the mouth of the Mississippi River
and southeast to the Straits. It is this route that has greatest significance for all periods
during the age of sail in the Gulf (Fig. 2).

I
I
I
I
I
'
...
11\
\'. \
1700-1763
\\
" I

Figure 2. Shipping routes 1700-1763.

306 E.G. Garrison

Inspection of Figure 1 reveals a moderate to low frequency of shipwrecks in the

deep open Gulf for all periods of interest. Why? The open sea is the ship's element.
There it can maneuver with the winds, tum to the sea, and, in short, be less endangered
other than in exceptional cases such as rogue waves or freak storms such as the Solano
Hurricane of 1780 (Millas, 1968).
In the case of channels and coastal waters, the hazards are self-evident. The ship
must again approach land. The shipwreck patterns reflect this for all periods. In the Span-
ish period, however, there was no need to come to land in the northern Gulf, as there
were no ports; hence, few losses are seen. Channels are the most dangerous even today.
Improvements in navigation, ship design, and sea-keeping still do not prevent losses in
these areas. Likewise, high-loss probabilities exist for coastal routes where the mariner
may be gulled by the apparent safety of the shore only to fall prey to shoals, shifting
shallows, and hazards such as debris from streams entering the Gulf at many points.

Ports
Ships must come to land at ports. The land is the first enemy of a ship built to ex-
ist in equilibrium with a fluid environment. Ships must enter port for obvious reasons,
such as trade, supplies, and refit, etc. They must do so carefully, as the distribution of
wreck sites at these points indicates. Losses were high for the 16th and 17th centuries,
and for good reason, given the lack of navigational aids, vulnerability to storms, known
piracy and warfare. Losses increased during the 18th century for most of the same rea-
sons, as well as the increase in the number of ports. In the 19th and 20th centuries, even
with improvements in navigational aids and ship design, losses at ports continue to be
higher than in other areas except the Straits of Florida. A peak value occurs for ship-
wrecks in ports in the 19th century. The increase in number of ships lost increases with
the same basic curve, although not at the same rate, as that seen for the increase in
number of ports (c.f. Figs. 3 and 4).

Shoals, Reefs, Sand Bars, and Barrier Islands

There are several obvious natural factors that may be used to predict the locations
of historic shipwrecks. These include the locations of shoals, reefs, and bars, and bar-
rier islands, However, the locations of shipping obstacles like sand bars and barrier is-
lands can change dramatically over time.
Bars such as longshore spits off headlands are factors in explaining the occur-
rence of wrecks in shallow waters. Ships, particularly those of larger tonnage, encoun-
tering these hazards are stranded. The nearshore bar, entrance bar, and the shoal are
generally one and the same. Examples of treacherous shoal areas are those off Cape
San Bias (Florida) and the Mississippi Delta. These shoal areas have claimed a larger
proportion of shipwrecks than that seen for the Gulf as a whole.
The distribution of known shipwrecks points to another key factor in Gulf ship-
wrecks namely the 192-mile reef and shoal complex of the Florida Keys, Marquesas,
and the Dry Tortugas. Throughout all of the time periods of the Gulf navigation, these
hazards destroyed more ships than any other natural obstruction.
Other obstructions came into playas vessels encroached into shoal waters in their
efforts to raise the ports along the Gulf shore. Still, the combination of factors of wind,
Factors Linked to Shipwreck Patterns in the Northern Gulf of Mexico 307

~~------------------------------~~

."
:II
1ft
o
c:
1ft
Z
o
-<

1SSD 1500 1100 lUG

YEAR

Figure 3. Shipwreck frequency by decade.

current, reef, and storm made the Straits of Florida the most hazardous area for ships
attempting this exit from the Gulf. Charlevoix (1763) recognized this, as did those who
preceded and followed him. When using the great Loop Current flowing eastward (Fig.
5), any deviation of a half of a degree north or south placed a ship at the mercy of coun-
tercurrents and the west-blowing trades. The Gulf Stream generally availed them nothing
if a powerful storm passed over the Straits at the same time. To the mariner of the 16th
century, all the Straits lacked were dragons. He and others who followed later may not
have feared the unknown, but they learned all too well the real perils of this passage.

Ocean Currents and Winds

The influence of winds and currents during the ages of sail, the 16th to 19th cen-
turies, was significant. The wind fields created by the patterns of the trades made for
easy westward journeys, but necessitated tacking or sailing off the wind in eastward

2~~-------- ________________________--,
LEGEND
20 -- PORTS
z
c
3:
~ 15
::a
o
." 10
~
o
::a
~ 5

O+------,------~----~~----_r----~
o 1700 1800 1800
PERIODS
Figure 4. Port development-northern Gulf
308 E.G. Garrison

Figure 5. The loop current.

crossings of the Gulf. The pattern for the winds varies from easterly in winter to south-
'southeast in summer. To take advantage of the summer wind regime meant the sailing
vessels from New Spain, Tierra Firme, or the Caribbean sailed northeasterly courses for
much of their journeys before turning southeastward to the Florida Straits.
In particular, vessels from Veracruz followed routes that allowed them to take
advantage of easterly flowing currents, most notably the Loop Current. Precursor to the
Gulf Stream, this deepwater movement aided mariners in their battles to sail into the
winds that impeded them from direct crossing of the Gulflatitudes. Only with the com-
ing of steam-powered vessels and colonization of the northern shore was this pattern
significantly modified.
During the 16th to 18th centuries, traffic from the Caribbean crossed the Gulf as
shown in route maps (Figure 2). As ports developed on the north shore, commerce
grew up between the Gulf and Atlantic coasts (Figs. 6, 7). More vessels beat against the
Gulf Stream in rounding Florida. Always dangerous, the reef passage on westward
journeys required vessels to sail north of the Gulf Stream and take advantage of the
countercurrent in this area. The margin for error was reduced and many vessels learned
this only after wrecking on the reefs and shoals of the Straits area. This continued to be
the case into the early 20th century, taking its toll of the coastal schooners and other
sailing vessels involved in this trade.
This discussion reiterates earlier observations of the overall importance of the
Gulf wind and current regimes in the determination of shipwreck locations. Still, the
patterns observed for Gulf shipwrecks are first the result of economic decisions involv-
ing maritime commerce. The mariners took what they could from the winds and cur-
Factors Linked to Shipwreck Patterns in the Northern Gulf of Mexico 309

1763-1821

Figure 6. Shipping routes 1763-1821.

rents are thus boundary conditions within which vessels operated to accomplish their
primary tasks. This is clearly seen in the change from the shipwreck pattern of earlier
periods when compared to later periods. The Spanish lost ships principally at the
Straits, not because of a poor reading of currents or winds. This ignores the skill of
those navigators. Instead, many vessels fell prey to anomalies of weather, e.g., northers
or hurricanes. Less frequently, they made costly errors in navigation that resulted in
their wrecking. As a determining cause in shipwreck patterns, winds and currents must
be viewed as important, but secondary, factors.

Historic Hurricane Paths

The location of sailing routes and ports may determine the principal locations of
shipwrecks, but these historical factors do not, in themselves, cause shipwrecks. The

Figure 7. Shipping routes 1821-1862.

310 E.G. Garrison

interaction of these factors with the third category of the natural factors, hurricanes, is
responsible for numerous maritime losses. These storms of the Gulf of Mexico's sum-
mer and fall months have historically been a menace to mariners of all periods. This is
true even today, when satellites track and computers predict locations of these great
storms. Vessels that fall afoul of the course of a hurricane do so at great risk.
The early navigators learned the hard way. Columbus experienced a hurricane as
early as his second voyage on June 16, 1494 (Henry et al., 1975: 11). The Spanish
quickly learned to schedule fleet sailings to accommodate the peak season for these
storms. Understanding the seasonal nature of the hurricane did not guarantee that such
recognition would translate into adherence to rational sailing practices. Captains of flo-
tas challenged the odds and offtimes lost. Early examples are the 17th- and 18th-cen-
tury fleets lost in the Florida Keys and Bahamas Channel.
The French and British were aware of the force of hurricanes, with the reports of
the storms and their destruction along the northern Gulf beginning almost as soon as
observers were in residence (Gauld, 1796). As shown by the Chaunus (1955), the per-
centage of vessels involved in the "carrera" system that suffered loss related to storms
was 16.7%. This landmark study evaluaed over 11,000 sailings (over a 2% sample).
This percentage for the first one and one-half centuries in the Gulfs maritime history
should be reliable. The intercorrelation of historic hurricane data and suites of well-
documented shipwreck data (MVUS; Florida) show percentages of storm-related loss,
primarily from hurricanes, to be 16 and 9.1%, respectively.

CHRONOLOGICAL TRENDS: 16TH-20TH CENTURIES SUMMARY

To summarize chronological trends in the shipwreck patterns related to general

historical factors such as flota cycles, colonization, commerce, and shipping routes, the
data are organized in 50-year periods from 1500 to 1899 and 20-year increments there-
after Table 1.

SPATIAL ANALYSIS: ARITHMETIC MEAN CENTERS

A clear trend in the scatter plots of the shipwreck location plots is the aggregation
of the distributions over the northern Gulf with time. In order to examine this distribu-
tion, the arithmetic mean center (AMC) has been calculated for the shipwrecks within
quadrats of 0.5 and 1°. No attempt has been made to look at the variation in the aggre-
gation of AMCs over time. This has been done with other techniques such as factor
analysis. The objective is to examine the presence or absence of aggregation at the
most general level. This technique merely takes the average of the cell (quadrat) fre-
quency and plots this average as a coordinate value. The results are summarized in
Tables 2-3 and Figures 8-11.

FACTOR ANALYSIS: CHRONOLOGICAL AND AREAL VARIABLES

Further analyses such as multivariate-factors studies have partitioned associations

between shipwrecks and chronological and areal variables. The results are summarized
in the following analyses.
Factors Linked to Shipwreck Patterns in the Northern Gulf of Mexico 311

Table I. Chronological trends in gulf shipwreck distributions by 50-years periods

1500-1549 Losses reflect period of Spanish exploration of the northern Gulf of Mexico.
1550-1599 Distribution begins to show pattern oflosses detennined by flota routes. Losses off Texas are
flota vessels wrecked by stonn while on this route. Losses off Florida are likewise. The
Straits area beginning to take its toll.
1600-1649 The principal losses are still Spanish flota vessels. The 1622 hurricane losses in the Keys are
significant portion of the shipwreck pattern for this period.
1650-1699 The pattern reflects the first French losses in the Gulf at Matagorda Bay in 1685. The
remainder are Spanish losses.
1700-1749 The distribution shows the first major change in the northern Gulf's shipwreck pattern. This is
due to the French colonization of Louisiana and the increase in a similar interest by the
Spanish in Pensacola to balance the French presence.
1750-1799 The pattern of shipwrecks in the northeastern Gulf is the result of two basic processes:
colonization and commerce. The French and Spanish have reached the height of their
maritime activities in the northern Gulf of Mexico. The flotas end in the last quarter of this
century.
1800-1849 The shipwreck distribution shows the extension of the colonization process to the
northwestern Gulf of Mexico area. Texas and Louisiana, west of the Delta, have port
development at a significant level after the 1830's, with Galveston, Brownsville, and
Freeport rising in importance.
1850-1899 The continued shift westward in the shipwreck distribution is offset by the principal ports of
New Orleans and Mobile in the North-central Gulf area. The observed pattern is shown by
the extent of the Texas data for the period. Losses in the Straits continue, as it is the major
egress channel for inter-Gulf commerce. Eastern Gulf losses in the Civil War are
underrepresented in the Panhandle region, e.g., Apalachicola and Cedar Key.
1900-1919 The pattern is fully modem with intra-and inter-Gulf commerce developed between all major
ports. The eastern area has Tampa growing as a port, and major fisheries off the Panhandle
and Florida Keys. The distribution of open-Gulf shipwrecks reflects the major commercial
sea route to the Mississippi River and New Orleans.
1920-1939 The pattern for the modem era is the result of 20th-century Gulf commerce in commodity
goods, e.g., oil and agricultural exports.
1940-1959 Two principal factors increase the number of shipwrecks off southwest Florida: fisheries and
Tampa trade. For the northwestern Gulf, it is singularly petroleum production offshore.
Intra-Gulf routes shift westward toward Galveston and Houston.
1960-1979 The major intra-Gulf, inter-Gulf route axes are still east-west, reflecting bulk cargo
movement from central to northwest Gulf ports. Losses increase in the northwestern area
and are related to oil exploration and production on the outer shelf.

Analysis I: Chronological Factors

The matrix to be factored is composed of 7 variables and 26 observations for each
variable (Fig. 12a). A principal components factor extraction method was utilized. The
factors were evaluated for independence and variance accountability using axis rotation
techniques. The program used was STATVIEW 512+ (Abacus Concepts, 1986).
The variables include five that measure shipwreck frequency in the six periods.
Note that data for the 16th century were merged with those of the 17th century because
of the low number of shipwrecks known for these periods. Further, it is assumed that
the processes underlying the patterns were basically the same for both periods. The
data for the 19th century have been partitioned for the opposite reason. Processes
responsible for the observed patterns have changed and, to more adequately see this,
the data have been scaled accordingly. The results of the factor analysis appear in Fig-
ure 12b. Our interpretation of these results is:
312 E.G. Garrison

Table 2. AMC analysis: I-degree quadrats

K=I The plot simply shows quadrats with reported shipwrecks. Little in the way of locational or spatial
trends can be seen. (Not shown.)
K ~ 10 Here the criterion for assigning an AMC is that the quadrat must have ~ 10 shipwrecks. What is
interesting is a two-level spatial distribution of AMCs. The inner row of points correspond to
nearshore shipwrecks, while the second, more seaward distribution, indicates deeper-water
shipwrecks. This distribution collapses with the increase of (n) as seen in the next step (Figure 8).
K ~ 50 The distribution of AMCs follows that of the nearshore shipwrecks with little representation of the
less-numerous offshore losses (Figure 9).

• • •••
•
• • •

•••

Figure 8. K 2! 10 wrecks/I 0 quadrats.

Figure 9. K 2! 50 wrecks/I 0 quadrats.

Factors Linked to Shipwreck Patterns in the Northern Gulf of Mexico 313

Table 3. AMC analysis: O.5-degree quadrats

K <: 10 The distribution differs significantly from the I-degree quadrats results. This reflects the effect of
quadrat size on the analysis; larger size reflects broad-scale patterning, while small quadrats are
sensitive to finer-scale patterning, while small quadrats are. sensitive to finer-scale patterning (c.f.
Hodder and Okell, '1978). Note the way the pattern more nearly approximates -degree quadrat
results of n <: 50. The trend is toward the nearshore at Texas, but more seaward in western
Louisiana and western Florida (Figure 10).
K <: 50 Here the distribution collapses onto all of the major port locations of the northern Gulf, with the
exception of Tampa. This may be an artifact of an underrepresentation of data for the western
Florida ports. Given that the Texas and Florida data bases are the most complete of those of the
Gulf area, this may not be the case (Figure II).

Figure 10. K;;;: 10 wrecks/O.5° quadrats.

K ~ 50 wreckll O.5°quadrall

Figure 11. K;;;: 50 wrecks/O.5° quadrats.

 314 E.G. Garrison

Wrecks 20th C. Wreclcl19thC. Wred:s lath C. Wrecks 17·181h C. Age Oldul Port POrll,Mejor Major Storms

I 13 38 0 0 149 I 13
2 10 51 0 4 88 I 8
1 II 42 0 0 14.2 I 15
•
S
41
61
69
64
0
0
2
0
144
151
I
I
10
9
6 102 117 0 0 167 I 12
7 10 29 0 0 148 I 9
•
9
0
0
0
0
0
0
0
0
0
0
0
0
9
12
10 24 0 0 0 118 I 7
II 126 0 0 270 I 26
12 21 12 0 270 I 26
13 57 42 0 118 3 9
14 19 23 12 288 I 21
IS 30 0 0 168 I 3
16 0 0 0 0 0 12
17 15 II 0 l4 I 9
18 0 0 0 0 0 0
19 0 0 0 0 0 14
20 0 0 0 0 0 13
21 53 0 0 113 I 21
22 0 0 0 0 0 10
21 10 0 0 14' 2 7
24 II 0 0 0 0 6
25 22 14 43 IS 166 I 12
26 156 57 87 29 0 0 IS
a.
Factor 1 Factor 2 Factor 3
Wrecks 20th .. . .716 .511 .414
Wrecks 19th .. . . 387 .777 ·.084
Wrecks 18th .. . .955 ·.001 .089
Wrecks 17· .. . . 956 .023 .016
Age Oldest P .. . ·.07 .617 .71
Ports. Major ·.162 .797 .206
b. Major Storms .188 ·.001 .938

Figure 12. Factor analysis, chronological factors. (a) Chronological variables. (b) Chronological fac-
tors, oblique solution reference structure, orthotran/varimax.

Hurricanes Ports Routes Hazards Energy Wrecks

t 10 1 2 0 3 3
2 10 2 2 0 3 12
3 10 6 2 0 1 27
4 5 2 3 3 2 15
5 IS 4 3 2 2 6
6
7
13
4 •
I
0
3
3
2
0
3
0
•
6
8
• 2 3 0 2 6
9
10
•• 1
0
•
4
5
5
0
0
•
17
a.
Factor 1 Factor 2
Hurricanes -.675 -.067
Ports -.097 .707
Roules .698 '.'52
Hazards .672 .001
Energy -.892 -.39
b. Wrecks .468 .94

Figure 13. Factor analysis, areal factors. (a) Areal variables. (b) Areal factors. oblique solution refer-
ence structure. orthotran/varimax.
Factors Linked to Shipwreck Patterns in the Northern Gulf of Mexico 315

1. Three factors were defined.

2. These factors are largely independent of one another, the orthogonal solution
is fundamentally that of the oblique solution (1.454 versus 1.468).
3. The variance is equally divided between these factors (0.43, 0.31, 0.26, for
factors 1-3).
4. Factor 1 is characterized as an association of 16th, 17th, and 18th versus 19th
and 20th century wreck locations or, conversely, a demographic factor, e.g.,
colonial period locations don't predict other later period locations.
5. Factor 2 is characterized by a moderate association of variables representing
19th century shipwrecks and port development.
6. Factor 3 associates age of port and storms. The linkage is not compelling.
The existence of ports seems to be more strongly associated with wreck fre-
quency than that of the years the port (or ports) has been in existence. This
association may be related to the larger number of vessels lost in ports to
storms. Again, the proportion of the variance explained by this factor is low.

Analysis 2: Areal Factors

This factor study utilized a matrix composed of six variables and ten cases per
variable (Fig 13a). The methodology differs from the previous analysis in that large-
scale areas of the Gulf are examined regarding the presence of hurricanes, ports, traffic
reports, hazards, energy zones, as to shipwreck frequency. The hurricane frequency is
taken from Tannehill (1956: 119) with little alteration. The variable "routes" represents
the number of periods with major inter- or intra-Gulf routes present; "hazards" repre-
sents major reef, shoal, or other hazards.
The results of the factor study are as follows (Figure 13b):
1. Two factors were identified. This was seen when restricting the program to
this number of factors, and then allowing the program to determine the
number of factors independently.
2. The factors are not strongly intercorrelated, although the same cannot be said
of the variables.
3. The orthogonal solution seems a good approximation when compared to the
unrotated or oblique solution. Following the oblique solution (varimax), we see
a proportionate accounting of the variance 0.63 for Factor and 0.37 for Factor 2.
4. Factor I is interpreted as depicting a strong association of shipwrecks to
routes and hazards (0.698, 0.672).
5. Factor 5 associates shipwrecks and ports. Our first inclination is to call this
the "ports" factor. Recalling the results of the previous factor study, the asso-
ciation may reflect higher losses of vessels as a result of storms. Ports tend to
concentrate ships and great hurricanes destroyed significant numbers of ves-
sels while in port (cf. Millas, 1968).

SUMMARY

This study has utilized manipulation of an automated data base to examine pat-
terning in shipwrecks, over time, in the northern Gulf of Mexico. Using spatial and sta-
316 E.G. Garrison

tistical analysis methods along dimensions such as routes, ports, hazards, and storms,
we have seen that patterns follow historic trends in population growth, movement, and
concomitant economic activities. European imperialism, colonialization, and commodi-
ties determine routes and port locations and subsequent shipwrecks. Hazards and
storms sank many vessels along shipping routes, with concentrations seen at major
ports and hazards such as the Florida Reef Complex. Hurricanes have been determined
to be less important as a factor in shipwreck patterning than originally thought. This
lack of determinism reflects the random behavior observed for these storms.

REFERENCES
Abacus Concepts, 1986, STATVIEW 512+. Brain Power, Inc., Calabasas, CA.
Charlevoix, P.F.X. De, 1763, Letters to the Duchess of Lesdiguieres. Goadby, London.
Chaunu, P., and Chaunu, H., Seville et I 'Atlantique. Bibliotheque Nationale, Paris.
Coastal Environments, Inc., 1977, Cultural Resources Evaluation of the Northern Gulf of Mexico Continental
Shelf. Cultural Resources Management Studies, Office of Archaeology and Historic Preservation,
National Park Service, Washington, D. C.
Gauld, G., 1796, Observations on the Florida Keys, Reef, and Gulf; with directions for sailing along the keys
from Jamaica by Grand Cayman and the west end of Cuba; also a description with sailing instructions
of coast of west Florida between the Bay of Spiritu Santo and Cape Sable. W. Faden, London.
Henry, W.K., Driscoll, D.M., and McCormack, J.P., 1975, Hurricanes of the Texas Coast. Centerfor Applied
Geoscience, College of Geosciences, Texas A&M University, College Station.
Hoffman, P.E., 1980, The Spanish Crown and the Defense of the Caribbean, 1535-1585. LSU Press, Baton
Rouge.
MilIas, le., 1968, Hurricanes of the Caribbean and Adjacent Regions, 1492-1800. Academy of the Arts and
Sciences of the Americas, Miami.
Rezak, R., Bright, TJ., and McGrail, D. W., 1985, Reefs and Banks of the Northwestern Gulf of Mexico. Wi ley,
New York.
Tannehill, I.R. 1956, Hurricanes. Princeton University Press, Princeton, NJ.
Wood, P., 1979, The Spanish Main. Time-Life Books, Alexandria, VA.
 Part IV.C

Survey Methodology

Watson's paper (Chapter 28) is an explanation of a flow chart showing how archaeolo-
gists proceed from the initial stages of a survey through the identification of certain
types of sunken vessels. At each stage, hypotheses are tested; those supported or denied
by the observed features lead to certain predictable tracks for wreck confirmation or the
dismissal of the research area.
In Chapter 29, Shomette and Eshelman provide a good example of initial project
planning, taking into account historical background, previous work in the area, funding,
and clearly designated goals for phase I and phase II sections. Both expected and unex-
pected results contributed to the observations. The project represents the culmination of
years of careful work, and stands as a model for a comprehensive investigation of a sin-
gle riverain system. Their study has been outdated, however, by more recent work in
the same drainage system. See, for example, Shomette 1995.

317
 Chapter 28

Nautical Archaeology
Survey Methods

KIMBERLY WATSON

When an underwater archaeological excavation is being considered, it is essential to

establish clearly defined objectives that precisely state the nature of the excavation.
These objectives will generally fall into four categories: necessary surveys of areas that
are to be dredged or otherwise significantly altered because of construction, the desire
to locate a single, identified ship, the desire to locate nonspecific resources, and the
desire to salvage. These four scenarios are the most likely situations that will evolve
into organizing resources for an underwater excavation (Fig. I).
It is essential that the objectives of an excavation be clearly defined, as each of
the four listed above follows a slightly different chain of events leading to an eventual
identification of a site locale. In other words, you have to know what you're looking for
and what you want to accomplish. In this article, I will briefly explain the most com-
monly used methods of following through on the above four excavation objectives that
lead to the identification of a shipwreck site. I will only trace the methodology through
until the determination of the type of site as each site will usually follow a further iden-
tification process specific to that ship.
The first option, a survey related to a proposed change in the seascape, can be initi-
ated in three ways. First, an archival study could be done of the area to determine
whether there are significant resources underwater that merit further examination. If
there are, the survey would either be done randomly to examine the largest possible area,
or systematically, to search specific areas within the survey range. If the equipment and
Source: Ms. on file, Ruppe Library (1994), Program in Maritime History and Nautical Archaeology, East Caro-
lina University, Greenville, North Carolina.

319
 320 K. Watson

Survey of construction/
dredging/change-of-
Desire to locate single
identified ship
Desire to locate non-I
specific resources 1
Desire to
salvage
I
seascape location

~
of suspIcious area on seabed

I Archival research consisting

of search for particular ship
Archival research consisting
of search of resources base of
an area
~
~

'\V ~ ~ r'~v ___'\v~~J~v~

)[,,,,"""'tic..- of on "". ~..........- Random"- of an "'" I

'\/
I Anomaly found V
~ ~'L_ Empty search I

IRecognition of ship remainsJ Recognition of other anomaly

(natural gas, ballast pile,
submerged habitation site)

I Identification of non-engine sitel

J
Identification of engine site r--

I Identification of non-sail site I Identification of steam site I

Identification of diesel site I

IIdentification of sail site I Identification of nuclear sitel

Figure I. Objectives of underwater archaeological E:xcavation.

Nautical Archaeology Survey Methods 321

funds are at hand, the archival research may be dropped and a hands-on random or sys-
tematic survey may become the initial step in the process. For the second option, the
desire to locate a specific ship, circumstances generally differ. Most often, the search for
one ship would begin with archival research dealing specifically with that ship.
This would lead to either a random or systematic survey on an area determined
by research suggestions. However, several ships are also brought to light by pleasure
and sport divers in accidental sightings. This negates the need for a survey, as ships are
already pinpointed, though not yet identified.
The third scenario involves the desire to locate nospecific resources, such as ships
important for contributions to typologies of construction techniques. Again, archival
research of a particular area is a common way to begin, as time periods and nationalities
of ships are generally important features of this type of search. Accidental sightings are
also valuable in this type of search, as they can lead researchers to many ships that would
have been passed over in favor of ships with greater historical significance.
Finally, the fourth objective would be to salvage ships. This can be undertaken in
any of the ways described above, and will vary depending on the type of salvage to be
done. Treasure hunters may use archival research either of a specific ship or of an area,
or they may accidentally discover a site. Salvors would probably use either a random or
systematic survey in order to find large concentrations of metal, the most commonly
salvaged material.
Whatever the methods and the objectives, a search in any of the above instances
will either produce an empty search, or an anomaly will be discovered. If there is an
anomaly with sufficient features to warrant further investigation, it will be recognized
as either ship remains or another anomaly. These could be anything from natural gas
pockets to ballast piles or submerged habitation sites; If it is a ship, even further identi-
fication will be required.
Ship sites will fall into three broad categories: nonsail/nonengine sites, sail sites,
and engine sites. Engine sites are subdivided into steam, diesel, and nuclear sites. At
this point, the identification of the ship found should concur with the initial goals and
objectives of the search. There may be sufficient reason to continue the identification
process, or the initial goals may have changed, but whatever the outcome, a clear path
of defined objectives will have facilitated the search, as it will any research.
 Chapter 29

A Developmental
Model for Survey and
Inventory of Submerged
Archaeological Resources
in a Riverine System
The Patuxent River, Maryland

DONALD G. SHOMETTE
and RALPH E. ESHELMAN

INTRODUCTION AND HISTORICAL BACKGROUND

The Patuxent River is one of the few riverine systems on Maryland's western shore of
the Chesapeake Bay which will retains much of the original character found by the
English in their early settlement of the region over 300 years ago (Shomette, 1979b).
Cultivation of tobacco and tonging for oysters, using techniques basically unchanged
over the centuries, continues even to this day. Many of the early plantation homes still
grace the heights above the river's flood plain.
The Patuxent River, Maryland's largest intrastate river, afforded transportation
deep into the state's interior (Figs. 1,2). Historical navigation was possible to the town
of Queen Anne, 56 miles above its mouth, in colonial days. It was possible to float a
Source: Underwater Archaeology: The Challenge before Us; Proceedings of the Twelfth Conference on
Underwater Archaeology. edited by Gordon P. Watts Jr. (1981), pp. 159--171, Fathom Eight, San Marino, Cali-
fornia. Reprinted by permission.

323
324 D.G. Shomette and R.E. Eshelman

MARYLAND

Figure 1. Maryland map showing study area of Patuxent River.

boat by tidal power alone for 46 miles up the Patuxent (Wilstach, 1931: 18). It is there-
fore easy to understand the importance of this river to the early settlement attempts in
the region.
The first Europeans to see this river were probably the Spanish in 1588 (Shomette,
1979a:4). John Smith's 1608 map clearly shows the river was explored at least 50 miles
above its mouth. Jesuit missionaries followed in 1632 (Wilstach, 1931 :281). The estab-
lishment of the colony'S first government at nearby St. Mary's City brought European
expansion which soon spread along the Patauxent. By the middle of the 17th century the
Patuxent River Basin enjoyed the recognition of being the major tobacco-growing region
of the colony, a distinction it still retains (Brune, 1979:71). By early in the 18th-century
49 plantations and at least nine official ports of entry had been established on the river.
Several of these ports, such as the town of Nottingham, became major slave trading cen-
ters, supplying the manpower necessary to work the tobacco fields of southern Maryland.
That shipping on the river was significant can be gleaned from the fact that between
1748 and 1775 a total of 239 vessels, or 15.7% of the vessels clearing the port of
Annapolis bore Patuxent Naval District registry.
The American Revolution brought a quick halt to this prosperity. The British
blockade of the Bay brought a virtual standstill to overseas trade. British and Loyalist
naval forces destroyed numerous river craft, warehouses, and several plantations.
Despite these depradations the Patuxent River reached a considerable hiatus in shipping
soon after the Revolution. It is during this period that its waterfront communities
appear to have been most prosperous (Brune, 1979).
This prosperity, however, was again halted during the War of 1812 when the
Chesapeake was blockaded by the British who also wreaked havoc throughout southern
Maryland. In June 1814 an American flotilla of 18 vessels and a convoy of merchant-
men were blockaded in the river by a superior British squadron (Shomette, 1979b).
Between June I and June 26 three battles, resulting in the loss of several American ves-
sels, were fought. The final engagement was to become the largest naval contest ever
 Jl 3:
0
0-
!l..
..,0'
til
c:
::!
~
po
:l
0-
S'
<
(\)
:l
...
0
~
s·
po
~
~
:1.
:l
(\)
til
'<
<Il
;
3

Figure 2. Map of Patuxent River showing: (I) Selby-Spyglass Transect, sectors A and B; Lyon's Creek and Nottingham Transects, (2) Middle section of IJ,I
~
river survey, (3) St. Leonard's Transect, sectors A. B, C, and D; and Point Patience Presurvey Test Area. 1.11
326 D.G. Shomette and R.E. Eshelman

fought in Maryland waters. British reinforcements pushed the American flotilla

upstream allowing the British Army to land and march on Washington, D.C. By order
of the Secretary of the Navy the flotilla was scuttled to prevent capture.
Devastation by the British and siltation of the river channel as a result of poor
agricultural and deforestation practices brought on an inevitable decline in the maritime
importance of the river. Yet, the Patuxent played an important role in the age of steam
on the Chesapeake when the first steamer runs were established about 1815. During the
Civil War southern Maryland was swiftly occupied by the North. While Union naval
forces attempted to cope with Confederate blockade running and smuggling in the tide-
water, commercial transport on the Patuxent withered.
After the war, harvesting of seafood, particularly oysters, became a major com-
mercial enterprise. "Patuxent Oysters" were packed and sold as far west as St. Louis.
Such enterprises led to the development of a large shipbuilding industry. By 1880 Solo-
mon Island at the mouth of the river, named after Isaac Solomon who established the
first oyster cannery on the river, could boast of three shipyards and a fleet of over 500
vessels, nearly all of which were built locally.
The mouth of the Patuxent offered the deepest natural harbor on the Chesapeake
and was only half as far up the Bay as was the port of Baltimore. Plans for a railroad to
this area to exploit the potential commerce offered by the natural harbor were first con-
sidered in 1868.
After several changes of ownership and bankruptcy, the final attempt at comple-
tion was foiled by the start of World War I (Williams, 1979). Had it not been for these
financial failures, the mouth of the Patuxent might very well have usurped the port of
Baltimore, by this time already suffering from the effects of siltation of its harbor.
The U.S. Navy recognized the advantages of the area when it tested the Dewey
Drydock in 1905, the largest floating drydock of its day, capable of holding the largest
warships of the age. During World War II, the U.S. Navy established several facilities
at Solomon and at Cedar Point, across the river. The U.S. Marines who landed in the
Solomon Islands in the Pacific were trained at Solomon Island, Maryland. Today sea-
food harvesting remains economically important to the lower river, and there is still
talk of developing a deep-water port here.

PROJECT PURPOSE

Wilstach stated in 1931 that:

... the Patuxent seems never to have come out of the seventeenth century .... There is
no established allure about the Patuxent River. It has neither cult nor a lore. It is the
most sparsely inhabited and least known, probably, of all the rivers of Tidewater
Maryland.

This was basically true in Wilstach's time and even up to the last few decades
because of the remoteness of the region from urban centers. Two peninsular bodies,
made up primarily of Calvert and St. Mary's Counties, approximately 40 miles in
length, are separated by the Patuxent. This peninsular grouping, bordered by the Poto-
mac River on the west and the Chesapeake Bay on the east, remained relatively rural
until the last decade. The few roads into the region were used mostly by the local popu-
Model for Survey and Inventory in a Riverine System 327

lation and the few fishermen willing to travel its length when steamboat operations on
the river ceased during the 1930s. The proximity of the ever-expanding urban triangle
of Washington, D.C., Baltimore, and Annapolis, and increasing suburban development
has put intense developmental pressures on the region. The recently completed lower
Patuxent River Bridge and a soon-to-be-finished dualized highway now afford easy
access into the remotest sections of the area. As a result there has been an ever-increas-
ing destruction of the river's ecology, environment, folkways, and the tangible remains
of its maritime history.
It is for this reason that the Calvert Marine Museum (CMM) of Solomon, Mary-
land, and Nautical Archaeological Associates, Inc. (NAA) of Upper Marlboro, Mary-
land, joined forces in 1977 to investigate the submerged cultural potentials of the
Patuxent. What makes the Patuxent Project so exciting and somewhat unique is that for
the first time in state history an attempt was being made to examine, in a holistic man-
ner, the resource base of an entire river system (Shomette, 1979b). It is hoped that this
approach will serve as a model program for the evaluation of all riverine systems in
Maryland.
Why select the Patuxent? As elaborated above, this river system is richly
entwined with practically all phases of Maryland's maritime heritage. Its resource base
is threatened. But of equal importance is that the entire river basin falls within the
political boundaries of the State of Maryland. Other perhaps more notable systems,
such as the Potomac and Susquehanna, do not. Though such considerations may seem
insignificant, funding, permit requirements, and regulatory formalities are easier con-
trolled and managed under one set of rules and guidelines. And finally, there is prob-
ably no area in the Chesapeake tidewater that has remained so unchanged for so long
which now races toward the impact of urbanization. Wilstach (1931) was aware of its
value when he wrote that the Patuxent "is well worth knowing." It is the purpose of this
project to do just that.

PREVIOUS WORK IN STUDY AREA

Several ventures by various shipwreck enthusiasts and relic hunters have sought to
locate the scuttled Chesapeake Flotilla of 1814, but with little success. The only published
account of such an attempt appeared in the Baltimore Sun Magazine on October 19, 1958.
The account documents an attempt to bring up a wreck and artifacts near the old colonial
port site of Nottingham. A rope was tied to parts of the vessel's hull and to a car that tried
to pull the wreck from the water. Several ribs were dislodged and extensive damage to the
site, later determined during the Patuxent Project to be of 18th-century vintage, was
inflicted. The first professional paper dealing with the recovery of historic maritime arti-
facts from the Patuxent is by Schwartz and Green (1962). This paper recounts the discov-
ery of a wooden stock anchor in 1959 with the letters "Dictator" scrawled across it. It is
now known that this is an "Old Plan" kedge anchor from HMS Dictator, a man-of-war that
served as a troop ship during the War of 1812 on the Patuxent.
In the early 1970s the Maryland Department of Natural Resources initiated the
Maryland Derelict Boat Removal Project (MDBRP), a program funded by the State to
clear its waters of navigational hazards and abandoned vessels. The CMM became con-
cerned when several historic shipwrecks in the Patuxent River were removed by clam-
shell bucket and burned. Firsthand accounts of derelict removals indicated that intact
328 D.G. Shomette and R.E. Eshelman

hulls of numerous sailing ships were literally torn apart and destroyed in burn boxes,
including a number with figureheads and other carvings.
The Museum immediately brought the problem to the attention of Maryland State
Archaeologist Tyler Bastian who requested that we attempt to document any wrecks
still extant. The CMM thus began its first archaeological work on the river by scouting
out remote creeks and coves at low water for signs of possible wreck sites. A program
of oral history interviews with senior watermen and lifelong residents of the Patuxent
River region provided an abundance of valuable data. Often the name of a ship, former
owner, date of sinking or abandonment, or simply a tradition recalling the loss of a ves-
sel was obtained. This effort culminated in the initial documentation of 48 vessels
wrecked in Patuxent waters and included such types as pungy, bugeyes, schooners,
sloops, log canoes, steamboats, and even military vessels from the War of 1812.

THE PATUXENT PROJECT

In the fall of 1977 the CMM and NAA. joined together in an effort to gauge the
extent of the submerged cultural resource base of the Patuxent system.
A modest twofold program was launched. The first part was to conduct a limited
documents survey designed to provide a background of historic data pertaining to the
river's history and known archaeological sites therein. The second objective was to
evaluate limited but representative sites in the river by on-site hands-on investigation to
determine if the suspected resource potential existed. An enlarged total of 100 ship-
wreck or derelict sites and 7 potential drowned terrestrial sites were considered. Of
these, 17 sites were examined beginning in April 1978. Four were shipwreck sites, one
drowned trashpit, a 19th-century wharf site, and one an 18th-century riverport site.
Seven other sites provided no data whatsoever.
As a result of this quite limited reconnaissance of the river system during the
1978 season, it became quite apparent that the resource base of the Patuxent was obvi-
ously extensive. Concurrently with this work, NAA's project director, Donald
Shomette, had been independently considering an overview of the subdiscipline of un-
derwater archaeology in terms of goal objectives in the development of an archaeologi-
cal resource management plan for the State of Maryland. Naturally, the Patuxent work
had some influence on these considerations.
To facilitate the development of the plan it was necessary to define the nature of
the aqueous resource base, not only in the Patuxent, but throughout the state. Three
categories were suggested: shipwreck sites, naturally or artificially inundated sites, and
sites intentionally inserted into the aqueous environment. Shipwreck sites, of course,
speak for themselves as self-contained cultural time capsules. Inundated sites represent
the same wide spectrum of archaeological potentials that are encountered in the aque-
ous environment, such as harbor constructions, landings, fishing weirs, and military
obstructions, represent, perhaps, the most direct cultural association of man with the
submarine universe. These three divisions were then included in a tableau of geo-
graphical divisions that were created as convenient management areas. These geo-
graphical units were: rivers and wetlands, the Chesapeake Bay, the Atlantic Coast, and
enclosed aqueous systems such as lakes and ponds.
In light of these divisions, sufficient testing of representative archaeological cate-
gories in representative geographical units (with appropriate adjustments for environ-
Model for Survey and Inventory in a Riverine System 329

mental variables in each) would produce data necessary for the formulation of a predic-
tive model suitable for employment in the development of a statewide management
plan. Conveniently the Patuxent Project proved an appropriate testing grounds for one
segment of the plan.
With matching funds from the Heritage Conservation and Recreation Service
administered by the Maryland Historical Trust, NAA, and the CMM, a multifaceted
program for the evaluation of the submerged cultural resource base of the Patuxent
River was developed. The objectives of this program were aimed at the development of
a comprehensive overview of the maritime character and riverine-related culture, tech-
nology, and history of the Patuxent region lying between the historic head of naviga-
tion at Queen Anne's Town and the river's mouth on Chesapeake Bay; to develop, test,
and refine a sampling scheme suitable for the parameters established by the sundry
environmental situations encountered; to develop a predictive model for the entire river
system; and to locate and identify sites suitable for nomination to the National Register
of Historic Places. The program was divided into two phases.

PHASE I
Specifically, Phase I consisted of three objectives. The first of these was the
undertaking of an intensive literature search to develop a comprehensive overview of
the river's maritime and riverine-related culture, technology, history, and traditions.
The second objective was the examination of artifact assemblages recovered from the
river and its tributaries and to conduct interviews with local collectors and individuals
knowledgeable about the river and its traditions and resources. The final objective was
the undertaking of a remote sensing survey of four selected transects of the river sys-
tem, each representing a distinct environmental variable in which historic shipwrecks,
inundated terrestrial sites, and intentionally inserted sites were known or conjectured.
The literature search resulted in a 438-page historic overview covering the period
1588-1976 (Shomette, 1979a). Interviews with several dozen local informants and the
examination of about a dozen artifact assemblages were also carried out. Five collec-
tions were loaned or donated to the Museum for study or display.
To facilitate the final objective of Phase I, the services of Daniel Koski-Karrell
were secured to direct the remote sensing operation.
Based on historical documentation and evidence provided by local informants,
several specific sectors of the Patuxent and its tributaries were selected for evaluation.
These were grouped into three general geographic areas, which were environmentally
divided into four transect areas: St. Leonard's, a brackish tidal tributary; Nottingham,
an open river transect near the saltwater-freshwater interface; Lyon's Creek, a freshwa-
ter tidal tributary; and Selby-Spyglass, a freshwater tidal, open, shallow transect.
Within these four transects are represented the environmental extremes of the tidal flow
(St. Leonard's, minimum; Selby-Spyglass, maximum), siltation (St. Leonard's, least;
Selby-Spyglass, greatest), bottom composition (Lyon's Creek, mud; Selby-Spyglass,
sand), bottom profile (St. Leonard's, low profile; Nottingham, high profile). The three
geographic groupings were St. Leonard's Creek, the Nottingham region, and Wayson's
Corner-HilI's Bridge. These transects represent not only environmental considerations,
but key historic and prehistoric areas of significance.
The first area surveyed represented major sectors of St. Leonard's Creek.
330 D.G. Shomette and R.E. Eshelman

Previous investigation during 1978-1979 had determined the location of two

sunken vessels. The first of these was a vessel believed to be a U.S. Navy gunboat lost
during the War of 1812. The second vessel proved to be the 1883-built schooner Hen-
rietta Bach.
The second transect surveyed was a stretch of river opposite and north of the
town of Nottingham. In addition to the vandalized site mentioned earlier, the magne-
tometer revealed the presence of a previously undiscovered sunken vessel buried
beneath the sand and estimated to be between 45 and 75 ft in length. Another potential
wreck site was located but because of heavy silt cover could not be verified.
The third transect surveyed was that of Lyon's Creek. The creek itself was sur-
veyed for a length of I mile, the maximum reach upstream that the draft of the survey
boat would permit. In the lower portion of the creek several targets were located. How-
ever, most turned out to be steel pipe lengths used in the construction of modern float-
ing docks along the creek's banks, or now-sunken mooring facilities. Investigation of a
site believed to date ca. 1680-1720 beneath 10 ft of silt proved inconclusive. Because
of the occurrence of steel pipe in the area, the presence or absence of a sunken vessel
by means of the magnetometer could not be confirmed. A brief examination employing
a water jet also failed to produce positive results. Immediately south of the ancient
steamboat landing at Lyon's Creek Wharf, an interesting and extremely well-preserved
collection of early and middle 19th-century handguns, recovered from the river by sev-
eral youths, were examined. The six pistols are believed to have been deposited for safe
keeping by Confederate sympathizers to prevent confiscation by Federal forces during
the Civil War.
Near the mouth of Lyon's Creek an unidentified wreck was located through the
aid of a local informant. The site was visible only at extremely low water. Evaluation
of the site suggests it may have been a small service vessel, probably a dory-type sail-
ing barge, employed between the reach of Queen Anne's and Hills Bridge by the early
steamboat line operations in the area.
The last survey area investigated was the Selby-Spyglass Transect (SST). In the
lower portion of the transect at the mouth of Mattaponi Creek, the remains of a large
wooden vessel, nearly 110ft in length, were located. Initially believed to be the merchant
coaster Lafayette, lost near Selby's Landing in 1855, in-depth archival inquiry revealed the
vessel to actually be the steam-scow Peter Cooper, burned near Lyon's Creek in 1887.
In the Northern portion of the SST, near Spyglass Island, and extending southward
to a reach below Hills Landing lay the conjectured site of the Chesapeake Flotilla
wrecks. This is an area that has undergone extensive development during the 20th cen-
tury. In certain places, large magnetic anomalies associated with contemporary cultural
features served to obscure the possible presence of those relating to earlier times. Despite
these handicaps, a sunken vessel was discovered and confirmed. Preliminary investiga-
tion of this site by divers using a waterjet to probe beneath the sandy bottom revealed a
wreck lying beneath 4 ft of sediment and in 5 ft of water. The vessel was of wooden con-
struction, fastened with both iron and wood, and featured some evidence of burn damage.
A number of magnetic anomalies were recorded both upstream and downstream
of this target, over a reach of more than a mile and a half.
Several of the anomalies investigated turned out to be iron debris of modern ori-
gin, but a number of others were of undetermined origin. It is of some interest to note
that the vessels lost in the flotilla, and the merchantmen under their protection that
Model for Survey and Inventory in a Riverine System 331

were also lost, when placed bow to stem in a single line, as British Admiralty reports
indicate they were, would form a line nearly 1625 ft long.
In conjunction with the magnetometer survey, a number of related shore sites
were briefly observed or examined. In St. Leonard's, for instance, several fortifications
from the War of 1812 were located. One of these was recently bulldozed down for a
housing development. At Nottingham a half-mile-long earthen work, possibly fortifica-
tions from the same period but undocumented, was mapped. At Queen Anne's a total of
seven sites were briefly examined, as well as evidence of a once extensive shell mid-
den. Numerous sites such as these, in fact, up and down the river from Queen Anne's to
Hellen Creek near the mouth of the river, provided extensive evidence of prehistoric
and early historic occupation.
The results of Phase I conclusively reaffirmed the magnitude of the Patuxent's
archaeological resource base.

PHASE II
Phase II consisted of three related objectives for each of the four transects
selected in Phase I. The first of these was the development of appropriate techniques
for the survey and limited mapping of designated sites. The second was the develop-
ment and refinement of sampling procedures at each designated site. The final objec-
tive was the development of conservation procedures suitable to handle low-level
recovery of artifactual materials from a submarine riverine environment.
Specific sites selected for this study were: (l) the Gunboat Wreck in the st.
Leonard's Transect; (2) Mackall's wharf, an 18th-19th-century marine facility in the St.
Leonard's Transect; (3) the Cannon Wreck, a conjectured 18th-century wreck in the Not-
tingham Transect; (4) the Windson Beach Site, an inundated aboriginal site in the Not-
tingham Transect; (5) the Swhenk Site, a conjectured wreck site ca. 1680-1720 lying in
the Lyon's Creek Transect; (6) the Peter Cooper wreck in the SST; and (7) a conjectured
War of l812-era wreck site, referred to as the Turtle Shell Wreck, in the SST.
Survey of the above sites would be carried out by two separate field teams oper-
ating independently of each other. Five vessels, ranging in size from a 35-ft cruiser
down to a 10-ft pram, and four floating barges, each 25 ft in length, were to be em-
ployed during the survey. A total of nearly 30 paid and volunteer staff were utilized
during the survey.
However, the completion of Phase II was unexpectedly delayed as a result of
extraordinary significance of one of the sites encountered and the majority of resources
of the survey were thus diverted to a single goal, namely, the excavation and survey of
a nearly intact warship belonging to the Chesapeake Flotilla, the Turtle Shell Wreck.
Despite the efforts undertaken at this time, two other sites were eventually completed.
These were Mackall's Wharf and the Swhenk Site.
It is not within the scope of this paper to discuss the Turtle Shell Wreck, for that
will be covered in detail in the following papers.
Unfortunately, because of the resources and time expended in evaluating that site,
the data on the Mackall's and Swhenk sites remain to be distilled.
Indeed, final analysis of Phase II must await completion of the surveys of the four
remaining sites before the comprehensive profile of the resource base of the Patuxent
River can be completed.
332 D.G. Shomette and R.E. Eshelman

PRELIMINARY SURVEY OBSERVATIONS

Though data assemblage has yet to be completed, a number of preliminary obser-
vations regarding the quantity, quality, and spatial distribution of the Patuxent River's
submerged cultural resource base have appeared. Most reflect truisms that were expected
but had never been documented. Others were unexpected. All will be incorporated in the
development of a predictive model which, hopefully, will be included in a comprehen-
sive management plan. Below are listed a number of the salient observations.
1. The depth of the river system and its influence on the shipping that plied over
its reaches is reflected quite readily in the representative samples of vessels
lost in its waters. Of the six vessels examined in the upper reaches of the
river for instance, between Nottingham and Spyglass Island, in waters rang-
ing in depth from 6 ft to 30 ft, all were of flat-bottomed or of extremely shal-
low draft construction suitable for use in shoally waters.
2. The state of site preservation in all three categories - shipwrecks, inundated
sites, and inserted sites - is directly proportional to the distance from salt water,
and to the rates of siltation; i.e., where siltation is high, chlorides low, and ship-
worms minimal, such as in upriver areas, the probability of a site being in a fine
state of preservation is increased. Conversely, the closer to the open sea, the less
likely it becomes that a site will be in a fair state of preservation.
3. Vessel abandonment constitutes the single greatest cause of cultural deposi-
tion in the Patuxent River system. The majority of vessels abandoned were
work boats generally employed in the fishery industry or light freight haul-
ing. Generally, they are vessels of indigenous regional construction and were
owned, sailed, and serviced throughout their lives in the Tidewater. Many,
such as bugeyes, pungys, and log canoes, represent extinct or nearly extinct
types and the ends of evolutionary lines whose origins can be traced to the
earliest European settlers, indeed even to the aboriginal natives. Many repre-
sent undocumented variations of regional types constructed by "wrack of
eye" rather than by plan, and as such constitute a considerable cultural
resource for they have rarely been well documented. Many of the abandoned
vessels whose genealogy is traceable were found to have remained in service
for up to three-quarters of a century or more before they were hauled into a
backwater and left to rot. There is, indeed, some evidence that suggests this
practice is a continuation of early colonial practices, and disposal patterns of
abandoned vessels today may well apply to vessels of earlier times.
4. The greatest concentrations of abandoned vessels appear to be centered
around commercial areas, but not necessarily urban areas. Disposal areas are
largely centered around fisheries near the mouth of the river at Solomon,
though many can be found near tiny maritime communities farther upriver
such as Broomes Island.
S. The distribution of prehistoric settlement areas and early urban settlements
and late towns, for the most part, seem to coincide. The prehistoric settle-
ments noted on John Smith's map appear to have been established on upland
terraces adjacent to the river and also near brooks or freshets running into the
main river. This is contrary to the establishment of sites on peninsular high-
Model for Survey and Inventory in a Riverine System 333

lands which would have been more suited for defense. Such exposure may
reflect the peaceful nature of the Patuxent tribes, which Smith makes note of,
and suggests that they relied heavily on the remoteness of the river to protect
them from the more warlike tribes of the upper Bay and the lower tidewater.
6. It now seems possible that the sinking of the Chesapeake Flotilla in the main
channel of the Patuxent only a few miles below the riverport of Queen
Anne's Town and the subsequent failure to remove many of the wrecks may
have been a factor in sealing the commercial fate of an already dying town.
7. The greatest stress factor contributing to the erosion of the Patuxent River's
submerged cultural resource base, indeed, the submerged resource base of the
entire state, is the Maryland State Government. The MDBRP, well founded in
intent, has resulted in the removal and destruction of many hundreds of aban-
doned vessels in Maryland waters and literally dozens in the waters of the
Patuxent. A number of these have been vessels of considerable cultural or
historic import. Unfortunately there is no mechanism in the program to per-
mit the evaluation of any sites by qualified personnel. Ironically, the Mary-
land Department of Natural Resources, which administers the program, is
also the same agency under which the Maryland Geological Survey and the
State Archaeologist's Office operate.
8. A statistical abstract of the 127+ vessels documented as lost in the Patuxent
between 1707 and 1974 presents some interesting points. A total of 38, or
29.9%, were intentionally abandoned; 24, or 18.9%, were war-related losses;
18, or 14.2%, were burned; 16, or 12.6%, foundered; 11, or 8.7%, were
stranded; 7, or 5.5%, were lost via collision; 2, or 1.6%, capsized; I, or 0.79%,
was crushed by river ice; and 10, or 7.9%, were lost as a result of unknown
causes. A complete breakdown of vessel losses by cause and type is presented
in Table I. Of the total losses, the following percentages represent a break-
down by era: 5 vessels are documented as being lost during the 18th century
(3.9%); 2 vessel sites are suspected to be of 18th-century origin (1.6%); 34
vessels are known to have been lost in the 19th century (26.8%); I vessel is
suspected to be of 19th-century origin (0.79%); I vessel was lost in the 20th
century (34.6%); 36 vessels are probably of 20th-century vintage (28.3%); and
5 vessels are of undetermined age (3.9%).
The above statistics are probably representative figures for most riverine
systems in tidewater Maryland on a percentile rate with certain exceptions.
Because of the concentration of military activity in the Patuxent and the in-
cumbent high vessel casualty rate, war-related losses cannot be employed as
a measurable index. Riverine systems that have not experienced the tender
graces of the MDBRP can expect to enjoy a greater percentage of 20th-cen-
tury vessels still present. And obviously, rivers of lesser Maritime impor-
tance, or along which settlement began at a substantially later period, will
have proportionately fewer wreck sites.

SUMMARY AND CONCLUSIONS

The Patuxent River Submerged Cultural Resource Survey represents the first
comprehensive attempt to evaluate, characterize, and synopsize in a holistic format the
 ~
~
~

Table I. Vessel losses by type and cause

Abandoned Burned Capsized Collision Foundered Stranded War loss Ice Unknown causes Total
Barge (row) 13 I3
Barge (frgt) 2 2 5
Brig 2
Bugeye 7 8
Dory
Dredge
Galley I
Gunboat (wood) 2 2
Log canoe 2 2
Lookout boat
Pink
Pungy 2 3
Schooner 14 2 5 7 8 5+ 42+
Schooner (ram) I
Scow I 0
Screw (gas) 3 9 2 15
Screw (oil) 4 5
P
en
Sloop 5 8 :r
0
Sloop (block) 3(9
Steamboat 3 3
Submarine i
Tugboat ~Q,.
Unknown 8 9
?'
Total 38 18 2 7 16 11 24+ 1 10 127+ !"1
m
en
:r
!.
3
~
Model for Survey and Inventory in a Riverine System 335

archaeological potentials of a single navigable riverine system in the State of Mary-

land, and one of the first projects of its kind to test the fruits of that evaluation in a sys-
tematic fashion.
The project has progressed through three distinct phases over a 3-year period:
preliminary reconnaissance in 1978-1979; intensive literature search, remote sensing,
and hands-on examinations of representative sites in 1979; and intensive survey and
excavation of significant sites in 1980. As a consequence of this research the project
has resulted directly and indirectly in the development of the first substantial recon-
struction of the maritime and maritime-related history of this important river system;
the location, examination, and evaluation of numerous prehistoric and historic archae-
ological sites, both below the water and in related upland areas; the testing of survey
and sampling schemes suitable or adaptable to employment in zero-visibility environ-
ments common to the tidewater; one of the first remote sensing and underwater video
employments on a comprehensive underwater project in the state, and the establishment
of the first conservation facility in the state specifically to handle the cleaning, stabili-
zation, and mending of waterlogged artifacts recovered from a marine environment.
Finally, the project has demonstrated that the submerged archaeological resource
potentials in the State of Maryland, and indeed throughout the tidewater, offer a rich
data base for interpreting our maritime history, folkways, and culture, as perhaps no
other resource can.

REFERENCES
Brune, B,H., 1979, The Changing Spatial Organization of early Tobacco Marketing in the Patuxent River Basin.
Geographical Perspectives on Maryland Past, University of Maryland Occasional Papers Geography No.
4, pp. 71--89.
Schwartz. FJ., and Green, J., 1962, Found: One anchor from HMS Dictator. Maryland Historical Magazine 57
(4):367-370.
Shomette, D.G., 1979a, The Patuxent River Submerged Cultural Resources Survey: Historic Overview. Ms.
438 pp.
Shomette, D.G .• 1979b, The Patuxent River Shipwreck Inventory: A Survey Model for Maryland Submerged
Cultural Resources. Ms. 23 pp. Presented to the 10th Annual Conference on Underwater Archaeology.
Williams, A.W., 1975, Otto Mears Goes East: The Chesapeake Beach Railway. Meridian Sun Press, Alexan-
dria, VA.
Wilstach, P., 1931. lidewaterMaryland. Bobbs-Merrill, Indianapolis, IN.
 '.~~.' Additional Bibliography
~

o i' .
for Part IV
q;,O_" .': .
.j'~l

Dean, Martin et ai, (ed.), 1990, Archaeology Underwater: the NAS Guide to Principles and Practice, Nautical
Archaeology Society, London.
Green, Jeremy, 1990, Maritime Archaeology, Academic Press, London.
Haggett, Peter, 1966, Locational Analysis in Human Geography, St. Martin's Press, New York.
Hodder, Ian and Clive Ortin, 1976, Spatial Analysis in Archaeology, Cambridge University Press, New York.
Kenderdine, Sarah, 1994, Historic Shipping on the River Murray, Australia: A Guide to the Shipwreck
Resource. International Journal of Nautical Archaeology 23(3): 173--88.
Parker, A. J., 1978, Method and Madness: Wreck Hunting in Shallow Water. Progress in Underwater Science
4: 7-26.
Riccardi, Edoardo, 1995, Some Considerations concerning wrecks lying in shallow waters. Tropis III Third
International Symposium on Ship Construction in Antiquity. edited by Harry Tzalas, pp. 367-370.
Greek Ministry of Culture, Athens
Schiffer, Michael 8., 1987, Formation Processes of the Archaeological Record. University of New Mexico
Press, Albuquerque,
Shomette, Donald G., 1995, Tidewater Time Capsule: History Beneath the Patuxent. Tidewater Publishers,
Centreville, Maryland.

337
 Part V

High Technology

Compared to other more comfortable academic disciplines, anything as relatively eso-

teric as underwater archaeology is bound to be divided into some fairly arcane and
challenging subfields. From Civil War ironclads to Polynesian voyaging canoes to
Athenian triremes, the peculiarities of different cultures often demand rather intensive
specialization in ship construction and maritime history, therefore, underwater archae-
ologists can be as different in background as treenails and driftpins. They do, however,
have many things in common. The tools of the trade are similar throughout the world;
and at some point, almost everyone learns to dive, sometimes whether they really want
to or not.
Technically speaking, diving has little to do with archaeology itself; the percep-
tion that it is easier to train archaeologists to dive than train divers to be archaeologists
remains essentially correct. Frequently, though, the tools and methods of working un-
derwater undergo enough innovation to require official recognition, especially when
these innovations serve to make accessible shipwrecks previously thought to be impos-
sible to investigate. The tools of the underwater archaeologist have been and are cur-
rently sophisticated enough to be able to dictate methodology in the field. The
following section on technology and underwater archaeology serves to highlight some
of the modern advances in various tools, and also to suggest rapid advances to be
expected in the future.

339
 Part V.A

History of
".~i).:
~

Going Underwater
-r/: ..
':.i

This text, from which Chapter 30 is taken, was written for navy divers' instruction. As
with most military manuals, it contains a history of the field, which is abstracted here.
This section chronicles how humans have gone underwater and what methods they
used to extend their time below the surface. The illustrations are particularly useful. It
is interesting to note how shipwrtecks may have influenced the development of diving
apparatus.

341
 Chapter 30

History of Diving
".~i).:
~

':.i
-r/: .. U.S. NAVY

The origins of diving are fir~ly rooted in the needs and desires of men to conduct mili-
tary or salvage operations, to engage in underwater commerce, and to expand the fron-
tiers of knowledge through exploration and research.
No one knows when man first discovered that he could hold his breath and go
under water, but the beginnings of diving as a profession can be traced back more than
5000 years. These early diving efforts were confined to relatively shallow waters (less
than 100 ft), with the divers harvesting a variety of materials of commercial value
including food, sponges, coral, and mother-of-pearl.
One of the first records of such diving is found in the writings of the Greek histo-
rian Herodotus. He tells the story of a diver named Scyllis, who was employed by the
Persian King Xerxes to recover sunken treasure in the fifth century B.c.
From the earliest times, divers were active in military operations; and their mis-
sions included cutting anchor cables to set enemy ships adrift, boring or punching holes
in the bottoms of ships, and the building of harbor defenses at home while attempting
to destroy those of the enemy abroad. Historical records indicate that Alexander the
Great not only sent divers down to remove obstacles in the harbor of the city of Tyre,
in what is now Lebanon, which he had taken under siege in 332 B.c., but also went
underwater himself to view the progress of their work.
Other early divers developed an active salvage industry centered around the
major shipping ports of the eastern Mediterranean, By the first century B.C., operations
in one area had become so well organized that a scale of payment for salvage work was
established by law acknowledging the fact that effort and risk increased with depth. In
Source: From the US Navy Diving Manual, Chapter I. San Pedro, CA: Best Publishing Company, 1985. Re-
printed by permission.

343
344 u.s. Navy Diving Manual
24 ft of water, the divers could claim a one-half share of all goods recovered. In 12 ft of
water they were allowed a one-third share; and in 3 ft, only a one-tenth share.

EARLY DEVELOPMENTS
The most obvious and necessary step in broadening the capabilities of a diver was
to provide an air supply that would permit him to stay underwater. The first such efforts
used hollow reeds or tubes extending to the surface. The user could remain submerged
for an extended period of time, but could accomplish little in the way of useful work.
Breathing tubes were employed mainly as a tactic in military operations were they
permitted an undetected approach to an enemy stronghold. Men using these tubes, were
not so much divers as they were soldiers using the water as a cover for their mission.
At first glance, it would seem logical that the only thing needed to further
extended the range of the diver would be a longer tube. In fact, a number of early
designs were created following that concept, namely leather hoods with long flexible
tubes supported at the surface by floats (Figure 1). There is no record that any of these
devices were actually constructed; and, if any were tested under actual conditions, the
result may well have been the drowning of the user. Even at a shallow depth of 3 ft, it
is essentially impossible tQ breathe through a tube using only the body's natural respi-
ratory ability as the weight of the water exerts a total force of almost 200 pounds on the
diver's chest. This force increases steadily with depth and is one of the most important
factors in diving. Any successful diving operation requires that the pressure be over-

Figure 1. Early impractical breathing device. This 1511 design shows the diver's head encased in a
leather bag with a breathing tube to the surface.
History of Diving 345

Figure 2. Assyrian frieze (900 Be.).

come or eliminated. Throughout history, imaginative devices were designed to accom-

plish this, many by some of the greatest minds of the time; but, because the problem of
pressure underwater was not fully understood, the designs were impractical.
An entire series of designs were based on the idea of a breathing bag carried by
the diver. This concept may be quite old as an Assyrian frieze of the ninth century B.C.
shows what appear to be divers using inflated animal skins as air tanks (Figure 2).
However, these men were more probably swimmers using the skins for flotation. It
would be virtually impossible to submerge while holding such an accessory.
A workable diving system may have made a brief appearance in the later Middle
Ages. In 1240, Roger Bacon made reference to "instruments whereby men can walk on
sea or river beds without danger to themselves."
Other writers, in the 16th and 17th centuries, described and published drawings
of equipment that foreshadowed later successful developments.
However, it is doubtful that these designs had any substantial influence, and to
give undue credit to the inventors of that period would be similar to giving Jules Verne
credit for inventing the nuclear submarine. Verne had the imagination and foresight,
but lacked the necessary supporting technology.

Early Successes

During the general period 1500-1800, a device was developed and placed in use
that enabled divers to remain underwater for lengths of time measured in hours rather
than minutes. This was the diving bell.
Literally bell-shaped, with the bottom open to the sea, diving bells were large,
strong tubs weighted so as to sink in a vertical position thereby trapping enough air to
permit a diver to breathe for several hours. The principle of the bell is easily observed
346 U.S. Navy Diving Manual

by pushing an inverted drinking glass into a pan of water. The air inside the glass is
compressed slightly by the water, with the pressure equaling at some point to leave a
reservoir of air.
Diving bells are suspended by a cable from the surface and have no significant
underwater maneuverability beyond that provided by moving the support ship. The
diver can either remain in the bell, if positioned directly over his work, or can venture
outside for short periods of breath-holding activity.
The first reference to an actual practical diving bell was made in 1531; and, for
several hundred years thereafter, rudimentary but effective bells were used with regu-
larity. In the, 1680s, a Massachusetts-born adventurer named William Phipps modified
the diving bell techni,que by supplying his divers with air from a series of weighted,
inverted buckets as they attempted to recover treasure valued at £200,000.
In 1690 the English astronomer Edmund Halley (who predicted the periodic return
of the comet that bears his name) developed a diving bell in which the atmosphere was
replenished by sending weighted barrels of air down from the surface (Figure 3). In an
early demonstration of his system, he and four other people remained at 60 ft under the

Figure 3. More advanced diving bells, such as these designed by Halley, included weighted barrels
of air to replenish their atmosphere.
Hi~tory of Diving 347

Figure 4. Lethbridge's dIving suit.

Thames River for almost I Y2 hours. Nearly 26 years later, using an improved version of
his bell, Halley, then 65 years old, spent more than 4 hours at a depth of 66 ft.
In 1715, another Englishman - John Lethbridge - developed a one-man com-
pletely enclosed diving dress (Figure 4). The Lethbridge equipment was essentially a
reinforced, leather-covered barrel of air, equipped with a glass porthole for viewing,
and having two arm holes with watertight sleeves, thereby permitting the occupant to
accomplish useful work. This apparatus was slung from a ship and maneuvered in the
same manner as a diving bell.
Lethbridge was apparently quite successful with his invention and participated in
the salvage of a number of wrecks in various waters of Europe. In a letter to the editor
of a popular magazine in 1749, the inventor noted that his normal operating depth was
10 fathoms (60 ft) with about 12 fathoms the maximum, and that he could remain
underwater for 34 minutes.
Several other designs similar to Lethbridge's appeared and were used in succeed-
ing years (Figure 5). However, they all suffered from the same basic limitation as the
diving bell: the diver had little freedom because there was no practical way to continu-
ally supply him with air. A true technological breakthrough occurred at the tum of the
19th century with the development of a pump capable of delivering air under pressure.

The Deep..Sea Diver

The art and practice of salvage diving was well developed as England entered the
Industrial Revolution in the middle of the 18th century. Using diving bells, devices simi-
lar to that of Lethbridge, or simple skin diving in shallow waters, salvage operators were
very active recovering everything from lost anchors to gold coins and silver ingots.
While the common man was always fascinated by the search for "sunken treas-
ure," such finds were rare and the day-to-day financial rewards came from more mun-
348 u.s. Navy Diving Manual

Figure 5. Successful enclosed diving dress.

dane but surprisingly valuable scrap. The brass to be recovered just from the ordnance
of a 100-gun ship of the line was worth more than $50,000 in the money of the day.
And, with perhaps 1000 new military and civilian wrecks littering the shores of Great
Britain each year, there was a strong incentive for the development of a diving dress
that would increase the efficiency of salvage operations.
Traditional credit for the development of the first practical diving dress has been
given to Augustus Siebe. Actually he was only one of several men who produced a suc-
cessful apparatus at the same time. John and Charles Deane, two brothers who became
active in the salvage business, obtained patents in 1823 on the basic design for a
"smoke apparatus" to permit firemen to move about in burning buildings (Figure 6). By
1828 this had evolved into "Deane's Patent Diving Dress," consisting of a heavy suit
for protection from the cold, a helmet with viewing ports, and hose connections for
bringing in surface-supplied air. The helmet was not fastened to the suit, but simply
rested on the diver's shoulders, held in place by its own weight and by straps to a waist
belt. The exhausted or surplus air passed out from under the edge of the helmet, and
posed no problem as long as the diver was standing upright. However, should he trip
and fall the helmet could quickly fill with water.
History of Diving 349

Figure 6. Deanes' "smoke apparatus."

The Deanes were well enough organized as salvage operators that in 1836 they even
issued a diver's manual- perhaps the first ever produced. Augustus Siebe's initial contri-
bution to diving was essentially a modification of the Deane outfit. Siebe "sealed" the hel-
met to the dress at the collar by using a short, waist-length suit which permitted the exhaust
air to escape under the hem. By 1840, Siebe had adopted a full-length waterproof suit and
added an exhaust valve to the system. Knpwn as "Siebe's Improved Diving Dress," this ap-
paratus is the direct ancestor of the standard deep-sea diving dress in widespread use today.
By 1840, several other types of diving dress had appeared on the scene and were
being used in actual diving operations. At that time, a unit of the British Royal Engi-
neers was engaged in the sizable project of removing the remains of a sunken warship
(HMS Royal George), which was fouling a major fleet anchorage just outside of
Portsmouth, England. Colonel William Pasley, the officer in charge, decided that his
operation was an ideal opportunity to formally test and evaluate the various types of
apparatus. His testing eliminated one outfit (Bethell's apparatus) from further consid-
eration because the method of fastening the helmet t9 the suit was too cumbersome and
time-consuming. He was wary of the Deane apparatus because of the safety factor, and
therefore formally recommended that the Siebe dress be adopted for any future opera-
tions.
When Colonel Pasley's project had been completed, an official government histo-
rian noted that "of the seasoned divers, not a man escaped the repeated attacks of rheu-
matism and cold." The divers had been working for 6 or 7 hours a day, much of it spent
at depths of 60 to 70 feet. Colonel Pasley and his men did not realize the implications
of this observation. What appeared to be rheumatism was, in fact, a symptom of a far
more serious physiological problem which, within a few years, was to become of great
importance to the profession of diving.
350 u.s. Navy Divin~ Manual

Figure 7. This French caisson could be floated over the work site and lowered to the bottom by
flooding the tanks on either side.

The Mysterious Malady

At the same time that a practical diving dress was being perfected, other inventors
were working to improve the diving bell by increasing its size and adding high-capacity
air pumps capable of delivering sufficient pressure to keep water entirely out of the inte-
rior of the bell. This improved pump capability soon led to the construction of chambers
large enough to permit several men to engage in dry work on the bottom. This was par-
ticularly advantageous in such projects as excavation of bridge footings or the construc-
tion of tunnel sections where long periods of work were required. These dry chambers
came to be known by the French word caisson, which literally means "big box."
Caissons were designed to provide ready access from the surface. Through the
use of an air lock, the pressure inside could be maintained while men, debris, or materi-
als could be passed in and out. The development of the caisson was a major step for-
ward in engineering technology, and their use grew quickly (Figure 7).
Concurrent with the expanding use of caissons, an apparently new and unex-
plained malady began to affect the caisson workers. After completing a shift and
returning to the surface, they would frequently be struck by dizzy spells, difficulty
breathing, or sharp pains in the joints or abdomen. If left alone for a period of time, the
sufferer would usually recover, but he might never be completely free of some of the
symptoms. The caisson workers often noted that they felt better working on the job, but
possibly attributed this to being rested at the beginning of a shift as opposed to being
tired when the workday was over (when they always seemed to feel the worst).
As caisson work was extended to ever larger projects - and to greater operating
pressures - the physiological problems increased in number and became more and
more severe. Fatalities occurred with alarming frequency.
The malady was called, logically enough, the caisson disease. However, workers
on the Brooklyn Bridge project in New York gave the sickness a more descriptive name
that has remained ever since: the "bends." The term may have grown out of the similarity
between the contorted posture of the suffering worker, and an awkward, forward-leaning
stance affected by fashionable ladies of the time and known as "the Grecian bend."
History of Diving 351

Today, the bends is the best-known danger of diving, popularized by generations

of adventure fiction in books, magazines, and television. Even though men had been
diving for thousands of years, until the time of the caisson, few men had spent much
time working under any great pressure. Those individuals such as Pasley, who had
experienced some aspect of the disease, were simply not prepared to look for anything
more involved than indigestion, rheumatism, or arthritis.
The actual cause of caisson disease was first clinically described in 1878 by a
French physiologist, Paul Bert. In his studies on the effect of pressure on human physi-
ology, Bert determined that breathing air under pressure forced quantities of nitrogen
into solution in the blood and tissues of the body. As long as the pressure remained, the
gas was held in solution; but when the pressure was quickly released, as when a worker
left the caisson, the nitrogen returned to a gaseous state too rapidly to pass out of the
body in a natural manner. Bubbles of the gas formed throughout the body causing the
wide range of symptoms associated with the disease. If the flow of blood to a vital
organ was blocked by the bubbles, the result could be paralysis or death.
Bert recommended that caisson workers gradually decompress and that divers
return to the surface slowly. His studies led to an immediate improvement for the caisson
workers as they also discovered that their pain could be relieved merely by returning to
the pressure of the caisson as soon as the symptom appeared. Within a few years, specifi-
cally designed "recompression" chambers were being placed at job sites to provide a
more controlled situation for handling the bends. The pressure in the chambers could be
increased as necessary for the particular worker. One of the first successful uses of a
recompression chamber was in connection with the construction of a tunnel between
New York and New Jersey in 1879, which was the first attempt at a subway tunnel under
the Hudson River. The tunnel construction project was abandoned in 1890 because of a
blowout that flooded the tunnel. Use of the chamber markedly reduced the number of se-
rious cases and, most importantly, reduced the number of fatalities caused by the bends.

Other Physiological Discoveries

Bert's recommendation that divers use a gradual but steady ascent was not a com-
plete success, and some divers continued to suffer from the bends. There was a general
feeling at that time that divers had reached the practical limits of the art and that 120 ft
was about as deep as anyone could ever really work. This was partly because of the
repeated incidence of the bends from deeper dives, and also because of a noted ~neffi­
ciency on the part of the diver beyond that depth. Occasionally, divers would even lose
consciousness working at 120 ft.
An English physiologist, 1.S. Haldane, conducted experiments with Royal Navy
divers from 1905 to 1907 and determined that part of the problem resulted from a rela-
tively simple factor: the divers were not adequately ventilating their helmets and conse-
quently high levels of carbon dioxide were present. The problem was solved by
establishing a standard supply flowrate (1.5 cubic feet of air per minute), measured at
the pressure of the diver, and by providing pumps of sufficient capacity to maintain the
flow. Thus, the helmet was ventilated on a continuous basis.
Haldane also composed a set of diving tables that established a stage method of
decompression. Though they have been restudied and improved over the years, these
tables remain the basis of the accepted method for bringing a diver to the surface.
352 u.s. Navy Diving Manual
An immediate result of Haldane's studies was an extension of the practical oper-
ating depth for air divers to slightly more than 200 ft. At the time, this limit was not
imposed by physiological factors, but only by the capabilities of the hand pumps then
available to provide the air supply.
It was not long before divers were moving into deeper water, and another unex-
plained malady began to appear. The diver would seemingly become intoxicated, some-
times feeling euphoric, and frequently losing his sense of judgment to the point where he
would forget the purpose of his dive. In the 1930s this "rapture of the deep" was linked
to the nitrogen in the air when breathed under higher pressures. Technically known as
nitrogen narcosis, it was shown that nitrogen has anesthetic properties which become
progr.essively more severe with increasing air pressure. To avoid the problem today, spe-
cial breathing mixtures such as helium/oxygen are used instead of air for deep diving.

Armored Diving Suits

Numerous inventors, many having little or no experience underwater, worked to
create an armored diving suit that would free the diver from any and all problems of
pressure. In such a suit, he could breathe air at normal atmospheric pressure and, hope-
fully, could descend to great depths without any ill effects. The barrel diving suit of
John Lethbridge was essentially an armored suit but with a limited operating depth.
The utility of most armored suits was questionable. They were too clumsy for the
diver to be able to accomplish much work, and too complicated to provide the protec-
tion for extreme pressure that was the goal of the inventors. The maximum design
depth of the various suits developed in the 1930s was 700 atm, and was never reached
in actual diving. New pursuits in the area of armored suits, now called one-atmosphere
diving dresses, have demonstrated their capability for specialized underwater tasks.

Development of the "Free" Diver

Deane, Siebe, and the others had given man the ability to remain underwater for
extended periods with enough flexibility of movement to accomplish a significant
amount of work during his stay. However, the diver was still unalterably tied to the sur-
face. Without the vital umbilical of his air hose, he could not survive, let along func-
tion. While the hose gave him life, it also severely limited his range of operation.
Inventors searched for methods to release the diver from the surface hose, look-
ing for a system that would permit increased freedom of movement without increases
hazards. The solution was obvious: to provide the diver with his own, portable air sup-
ply. Today the "self-contained underwater breathing apparatus" (SCUBA) is not only a
reality but has replaced the deep-sea diving outfit as the most familiar and most fre-
quently used type of diving gear: For many years, however, the SCUBA had remained
as only a theoretical possibility. Neither a pump (compressor) of sufficient capacity nor
tanks of adequate strength to handle the high pressures needed to provide a practical air
supply existed.
Development of SCUBA took place gradually, and over the years basic types
evolved: open-circuit, closed-circuit, and semi-closed-circuit.
With the open-circuit apparatus, air is taken from a supply tank, inhaled, and the
exhaust is vented directly to the surrounding water. The basic closed system uses a tank
History of Diving 353

of 10% oxygen which supplies a breathing bag. The oxygen being used by the diver is
recirculated in the apparatus. It passes through a chemical filter which removes carbon
dioxide, and additional oxygen is added from the tank to replace that which is con-
sumed in breathing. For clandestine military operations, the closed-circuit system (oxy-
gen rebreather) has a major advantage over the open-circuit type insofar as it does not
produce a telltale trail of bubbles on the surface.
The most recent innovation in closed-circuit systems employs a mixed gas for
breathing and electronically senses and controls oxygen concentration. This type of
apparatus retains the bubble-free characteristics of 10% oxygen recirculators while sig-
nificantly improving depth capability.
The third basic type, semi closed apparatus, combines features of the other two
systems. Using a mixture of gases for breathing, the apparatus recycles the gas through
a carbon dioxide removal canister and continually adds a small amount of oxygen-rich
mixed gas to the system from a supply tank. The supply gas flow is pre set to satisfy
the body's oxygen demand and part of the recirculating mixed gas stream, equal to the
supply gas flow, is continually exhausted to the water. Because the quantity of make up
is constant regardless of depth, the semiclosed SCUBA provides significantly greater
endurance than open-circuit systems in deep diving.
Historically, the development of self-contained breathing apparatus has involved
many significant inventions which, for the lack of supporting components of equivalent
capability, could not be fully exploited. Two main paths of development were followed:
one pursued open-circuit SCUBA, and the other emphasized its closed-circuit counterpart.
The first and highly necessary component of an open-circuit apparatus was
designed as early as 1866. This was the demand regulator patented by Benoit Rouquay-
rol, which adjusted the flow of air from the thank to meet the breathing and pressure
requirements of the diver. However, because tanks of sufficient strength to contain
air at high pressure could not be built at the time Rouquayrol adapted his regulator to
surface supplied diving equipment, and the technology turned toward closed-circuit
designs. The application of Rouquayrol 's concept of a demand regulator to a successful
open-circuit SCUBA was to wait more than 60 years.
In 1878, the first commercially practical self-contained breathing apparatus was
developed by H.A. Fleuss, This was of the closed-circuit type, and used 100% oxygen
for breathing. Because the system used only oxygen, the quantity of gas in the tank did
not have to be as great as it would with compressed air (which is only about 21 % oxy-
gen). Thus, the need for high-strength tanks was eliminated.
Unfortunately, at the time of his invention, Fleuss was not aware of the serious
problem of oxygen poisoning caused by breathing 100% oxygen under pressure. It was
not until many years later that researchers determined the maximum safe depth for use
of 100% oxygen.
Two years after its invention, the Fleuss SCUBA figured promineQtly in a highly
publicized achievement by an English diver, Alexander Lambert. A tunnel under the
Severn River had flooded in 1880, and Lambert, wearing a Fleuss apparatus, walked
1000 ft along the tunnel, in complete darkness, to reach and close several crucial valves.
As development of the closed-circuit design continued, the Fleuss equipment was
improved by the addition of a demand regulator and tanks capable of holding oxygen at
more than 2000 psi; and, by World War I, the Fleuss SCUBA (with modifications) was
the basis for submarine escape equipment used in the Royal Navy.
354 U.S. Navy Diving Manual

In 1933, the thread of open-circuit development was again picked up when a

French naval officer, Commander LePrieur, constructed an open-circuit SCUBA using
a tank of compressed air. However, LePrieur did not include a demand regulator in his
design; and the diver's main effort was diverted to the constant manual control of his
air supply by manipulating a valve. This fact, coupled with extremely short endurance,
severely limited the practical use of LePrieur's apparatus. The main emphasis contin-
ued to focus on closed-circuit development.
Although the closed-circuit equipment was restricted to shallow-water use, and
carried with it the potential danger of oxygen poisoning, its design soon reached a suit-
able high level of efficiency. By World War II it was in wide use by foreign navies on
both sides of the conflict. British divers, working out of midget submarines, aided in
the placement of explosive charges under the keel of the German battleship Tirpitz.
Italian divers, using closed-circuit gear, rode "chariot" torpedoes fitted with seats and
manual controls in repeated attacks against British shipping; and, in the final stages of
the war, the Japanese employed an underwater equivalent of their kamikaze aerial
attack - the kaiten diver-guided torpedo.
At the same time that actual combat operations were being carried out with
closed-circuit apparatus, two Frenchmen - one a naval officer, the other an engineer
- achieved a significant breakthrough in open-circuit SCUBA design. Working in a
small Mediterranean village under the difficult and restrictive conditions of German-
occupied France, Captain Jacques-Yves Cousteau and Emile Gagnan combined an
improved demand regulator and high-pressure air thanks to create the first truly effi-
cient and safe open-circuit SCUBA. Cousteau and his companions brought the "Aqua-
Lung" to a high state of development as they explored and photographed wrecks,
developing new diving techniques and testing their equipment. Their work was the cul-
mination of literally hundreds of years of progress, and blended the work of Rouquay-
rol, Fleuss, and LePrieur. Cousteau used his gear successfully to a depth of 180 ft
without significant difficulty; and, with the end of the war, the Aqua-Lung quickly
became a commercial success. Today, this apparatus represents the most widely used
and familiar of diving equipment, opening the underwater world to anyone with suit-
able training and the necessary physical abilities.
 Part V.B

'.~~.' Magnetometer
~

o i' .
q;,O_" .': .
.j'~l

In their important article (Chapter 31), Murphy and Saltus outline critical assumptions
involved with magnetic survey techniques. Assumptions such as "magnetic anomalies
equal cultural resources" and "50 meter lane spacing" must continue to be examined
and not taken for granted. The authors suggest a 30 meter lane spacing for accuracy.
The realization that simple magnetic survey does not equal comprehensive cultural
resources survey, but must be preceded by historical, economic, and demographic
research, highlights the need for more intensive investigation style.
Arnold's attention to the more concrete realities of budget and hardware make
Chapter 32 useful to others planning similar investigations. The major problems with
airborne survey seem to be the cost of the helicopter rental and the lack of ability to
accurately control the sensor altitude over the water. This survey from Texas serves
also to highlight the ever present delays and equipment problems which must be taken
into account during research design.
By using technology more traditionally associated with geophysics, Hudson et
al., in Chapter 33, were able to successfully inspect areas as shallow as 18 inches or
less, as well as beaches and sand dunes, with a magnetometer. Mounting the sensor on
a long boom from a skiff worked well for shallow water areas, while an All Terrain
Vehicle (A TV) containing the batteries for a portable unit accompanied the person car-
rying the sensor on land.

355
 Chapter 31

Considerations of Remote
Sensing Limitations to
Submerged Historical
Site Survey

LARRY E. MURPHY
and ALLEN R. SALTUS

Magnetic shipwreck survey has been around more than thirty years. During that time,
much discussion has been directed toward survey methodology and data interpretation.
Many untested assumptions have been generated and integrated into the literature and
eventually institutionalized by researchers and managers. Once again, as the subject of
this symposium, magnetic data generation, presentation and interpretation are being
discussed and debated; this indicates many issues remain unresolved. Perhaps we can
add some clarification by examining operative questions and assumptions in light of
our field work in oceans and rivers.
Magnetic survey is done in rivers and oceans and both will be discussed from our
collective experience in State and Federal positions and private contracting. Current con-
tracting practices, management directions and specifications assumptions, presupposi-
tions and models will be briefly critiqued. Our discussion is focused on magnetic survey.
We distinguish between a survey and a search. A search is conducted for a par-
ticular site or object; a survey is a systematic determination of specific cultural remains
within an area. Most magnetic surveys are generated by land agency managers who are
Source: Underwater Archaeology Proceedings from the Conference on Underwater and Historical Archaeology
Conference. edited by Toni L. Carrell (1990), pp. 93-95. Reprinted by permission of the Society for Historical
Archaeology, Tucson, Arizona.

357
358 L.E. Murphy and AR. Saltus

legally required to conduct cultural resource surveys to determine whether develop-

ment activities will affect significant archeological and historical sites. To a lesser
degree, research questions and baseline inventories prompt shipwreck surveys; how-
ever, this paper focuses on the management framework under which most shipwreck
surveys are conducted in the United States.
Principal land manager questions are "Is anything there?" "Is it significant?" and
"How can threats to significant sites be limited?" Managers inevitably balance legal
requirements, professional responsibility, and insufficient funds. In the conduct of
archeological inquiry, as in most scientific inquiry, the right questions must be asked
and answered in the right order. Methodology is judged on its effectiveness in produc-
ing information and data needed to answer questions. Management additionally uses
cost effectiveness and cost efficiency to evaluate methodology.
Some general characteristics and particular assumptions pervade most current
magnetometer surveys. One common assumption is that magnetic anomalies equal cul-
tural resources. Its corollary is big anomalies equal significant cultural resources. Fed-
eral law requires the assessment of all development zone cultural resources. In most
cases, a few large magnetic anomaly evaluations substitute for comprehensive prehis-
toric and historical cultural resources assessment. Often nothing but large ferrous
objects are located or evaluated. Prehistoric sites are often ignored as being "hard to
find" or "too expensive." A magnetic survey, regardless how well executed is not a
complete cultural resources survey and does not meet Federally-mandated obligations.
Consequently the "Is anything there" question when reliance is on magnetic survey is
more correctly stated as "Is anything ferrous there?"
To determine if anything ferrous is there the magnetometer is hauled around in a
boat to locate detectable objects. Systematic survey fundamentally requires determina-
tion of complete coverage to a specified level, a permanent survey data record and the
ability to accurately relocate anomaly areas. Transects or lanes through the survey area
guided by some sort of electronic navigation are the normal procedure. The "specified
level" has been greatly debated. The sensor distance from the target mass determines
registered anomaly intensity or deflection. General consensus rejects the adequacy of
150-meter lane spacing, and generally accepts 50-meter spacing. We estimate that a
concentrated ship-size vessel has less than a 1 in 4 chance of detection at 150-meter
lane spacing, which can consequently only be considered a sampling procedure not a
comprehensive survey - it would be easy to miss an ironclad in a 150-meter grid.
Acceptance of 50-meter spacing began in 1974 after fieldwork by Clausen and
Arnold on two 1554 shipwrecks of around 200 tons (Arnold and Clausen, 1975;
Clausen and Arnold, 1976). The researchers surmised that colonial period shipwrecks
would not be missed with 45 (Arnold, 77:27) or 50-meter lane spacing (Clausen and
Arnold, 1976: 168). This general statement, widely accepted by researchers and manag-
ers, has rarely been examined or questioned.
The estimate was based on the surviving wreck contents in a depositional envi-
ronment that allowed the development of large concretions - the wreck was concen-
trated, rather than scattered. Large, multiple-element concretions containing anchors
and cannons lay in close proximity. Examining the contoured results of a 6.5-meter
lane site survey indicates that the highest magnetic deflection was 125 gammas, created
by the ship's large sheet anchor and five cannon breech blocks. If the concentration
was positioned between two 50-meter lanes, the anomaly would read around 3 gammas
Remote Sensing Limitations to Submerged Historical Site Survey 359

as estimated from Sheldon Breiner's target mass-to-sensor-distance nomogram (Bre-

iner, 1973 :48). Had this large concretion been separated, the readings would be consid-
erably less.
A single 345 kilogram anchor gave a reading of 30 gammas and would not have
been detected if located directly between two 50-meter runs. At 50 meters, a 909 kilo-
gram (one ton mass) between the lanes would give an estimated 4-5 gamma reading
from the Breiner nomogram. The Breiner nomogram cautions that the mass-distance in-
terpolation can vary between a factor of 2-5, a variation we have observed in fieldwork.
Rather than base lane spacing on a particular wreck in a particular depositional environ-
ment our recommendation is to first consider ferrous mass as the detectable target.
Target mass must be considered in a historical context coupled with depositional
environment. A target mass can be reasonably predicted by considering the vessel
population likely to be present and the related ferrous masses likely to be present. For
example, in areas likely containing colonial vessels in an environment not conducive to
large concretion formation, a mass of 454 kilogramos (1000 pounds) would be reason-
able based on anchors and cannon of the latter half of the 16th century. Cannons from
the 1554 fleet ranged from 100 to 140 kilograms and anchors varied from 104 to 428
kilograms. If the largest mass on the site was lying between two 30-meter lanes, it
would produce about a 10 gamma reading. On 50-meter lane spacing, the expected
reading would be 2-3 gammas, which would likely be lost in the noise levels tolerated
by most magnetometer operators. Thirty meter lane spacing more than doubles the
chance of locating a 454 kilogram mass and greatly increases the chances of locating
more scattered wrecks.
Besides magnetic deflection intensity, duration of deflection is an important fac-
tor. The 1554 site contained about 12,000 kilograms of metal that produced a 10
gamma contour encompassing an area of abut 70 meters by 90 meters. Thirty meter
lane spacing would transect the 10 gamma contour encompassing an area of about 70
meters by 90 meters. Thirty meter lane spacing would transect the 10 gamma contour at
least twice in any direction for a minimum distance of 40 meters along a transect. Fifty
meter transects could produce a 10-gamma anomaly on one lane with a 5-gamma
anomaly on an adjoining lane. If the site were more scattered, it would be more diffi-
cult to detect the model vessel. We consider 50-meter lane spacing an outside maxi-
mum for shipwreck survey.
There are few marine vessels that have been documented to the level of the 1554
sites. However, numerous comparable riverine sites have been evaluated (Murphy and
Saltus, 1981; Saltus, 1987, 1988). The larger of these are coastal and marine vessels
found in rivers. Considering a sample of 12 wrecks located by magnetic survey and
contoured,4 (33 percent) would have likely been undetected on 50-meter lane spacing.
These include an 1840 65-foot long towboat, a 55-foot schooner, a 27-foot long seg-
ment of a modem shrimp boat and a 44 foot coastal sailor. All 12 would have been rep-
resented by a 15 meter long 10-gamma segment of a 30-meter spaced lane, however,
there are many exceptions. Many small boats and hull sections may be missed on a
10-gamma threshold.
Our recommendation is 30-meter lane spacing for exploratory magnetic survey
for shipwrecks with data presented in a manner that allows accurate determination of
intensity and duration (for an example see Murphy, 1984). All anomalies of 10 gammas
for a length duration of 15 meters should be considered possible watercraft. Five to ten
360 L.E. Murphy and A.R. Saltus

gamma anomalies appearing on two or more lanes should be considered important.

From all available data, most watercraft resources should be locatable with this meth-
odology. Small craft (an all nonferrous remains) may be missed.
The only exploratory survey data presentation to answer the question "is there
any relevant ferrous material there" we find useful is scaled symbols representing
intensity and duration. We find 30-meter contoured data of diminished use. Magnetic
contours based on IO-meter or less transects are the most useful for interpretation and
target mass location. This means that a second data generation phase is necessary to
reliably determine magnetic site extension and magnetic feature association. Magnetic
contouring is the most reliable way of predicting target mass location for test excava-
tion, in many cases it is the only way. The target mass may not be in many cases it is
the only way. The target mass may not be in the contour representing the maximum or
minimum reading, it is, however, invariably in the area of the steepest gradient.
Time does not allow full magnetic survey discussion for watercraft. However, in
closing, we would like to summarily present some specific points, observations, and
conclusions:
1. Magnetic survey, even when ground-truthed, is not comprehensive cultural
resources survey as required by Federal law.
2. Magnetic survey designed to locate larger riverine vessels, coastal and marine
vessels will not give reliable data on boats, widely scattered shipwrecks or his-
torical refuse areas, which must be considered along with prehistoric and non-
ferrous historical sites in any responsible comprehensive cultural resources
survey.
3. Contouring on 10-meter or less lane spacing is the most reliable data presen-
tation for magnetic data interpretation.
4. The quest for a "signature" for any particular wooden shipwreck is time ill
spent - we will challenge the producer of such a signature to distinguish a
genuine site from the one we can construct of cable, iron sewer pipe and
spikes.
5. Magnetic data can only provide information on the location, orientation, and
possible size of ferrous materials, any determination beyond these attributes
must be done by a competent archeologist examining material in situ.
6. Any anomaly of 10 gammas for a distance of 15 meters appearing on 30-meter
lane spacing or any 5-10 gamma anomaly appearing on 2 or more lanes, may
be watercraft. There is no current methodology or technology that can differen-
tiate watercraft from modem refuse meeting these criteria, other than on-site
examination. Anomaly prioritization is only imperative because of insufficient
funding, in the typical methods are as good as any and we recommend their
adoption in lieu of some unsubstantiated model based on magnetic data attrib-
utes. Failing to examine all anomalies potentially representing significant
watercraft in an impact area is abrogation of responsibility on the part of land
mangers.
7. Historical, economic, and demographic research should precede magnetic
survey and be utilized in assessing site potential and significance. Vessel use
and loss do not occur in a cultural vacuum, but are the results of patterned
human activity. Patterns and correlations can often be observed, for instance,
Remote Sensing Limitations to Submerged Historical Site Survey 361

river landings in the lower Mississippi River drainage have an average of 3.2
abandoned or sunk watercraft in the immediate vicinity. Magnetic survey for
wrecks should take place as part of regional research designs that provide
context for site interpretation. Site interpretation is currently the weakest
aspect of magnetic shipwreck survey, not magnetic interpretation.

REFERENCES
Arnold, J.B., III, 1977, Site Test Excavations: The Sequel to Magnetometer Survey. International Journal of
Underwater Archaeology and Underwater Exploration 6(1 ):21-36.
Arnold, J.B., III, and C.J. Clausen, 1975, A Magnetometer Survey with Electronic Positioning Control and
Calculator Plotter System. International Journal of Underwater Archaeology and Underwater Explora-
tion 4(2):353-6.
Breiner, S., 1973, Application Manual for Portable Magnetometers. Geometrics, Sunnyvale.
Breiner, S., 1973, Application Manual for Portable Magnetometers. Geometrics, Sunnyvale.
Calusen, C.J. and J.B. Arnold, III., 1976, The Magnetometer and Underwater Archaeology: Magnetic Deline-
ation of Individual Shipwereck Sites, a New Control Technique. International Journal of Underwater
Archaeology and Underwater Exploration 5(2): 159-69.
Murphy, L.E., 1984, Phase I Remote Sensing Survey. In Submerged Cultural Resources Survey: Portions of
Point Reyes-Farallon Islands Marine Sanctuary. Ed. Larry E. Murphy, pp. 85-140.
Murphy, L.E., and A.R. Saltus, 1981, Phase II Identification and Evaluation ofSubmerged Cultural Resources
in the Tombigbee River Multi-Resource District Alabama and Mississippi. Report of Investigations No.
17, Office of Archaeological Research, University of Alabama.
Saltus, A.R., 1987, Submerged Cultural Resources Investigation of the Western Portion of the Maurepas Basin.
Southeastern Louisiana University Center for Regional Studies.
Saltus, A.R., 1988, Submerged Cultural Resources Investigation of Various Waterways of Lake Pontchartrain s
North Shore. Southeastern Louisiana University Center for Regional Studies.
 Chapter 32

An Airborne Magnetometer
Survey for Shipwrecks and
Associated Underwater
Test Excavations

J. BARTO ARNOLD III

INTRODUCTION

The Underwater Archaeological Research Section of the Texas Antiquities Committee

conducted field investigations in Matagorda Bay, Texas, from June through mid-August
1978. The project consisted of a I-month magnetometer survey in June and a 6-week
period of underwater test excavations to determine the causes of magnetic anomalies
located during the immediately preceding magnetometer survey.
The primary objective of the project was to locate the two vessels lost by the
French explorer La Salle in 1685 and 1686 during his abortive attempt to establish a
fort and settlement at the head of Matagorda Bay. The purpose of La Salle's expedition
was to establish a colony at the mouth of the Mississippi River, the length of which he
had previously explored, but navigation being what it was at the time without a way to
accurately measure longitude, he ended up in Texas.
The larger of the two ships, L 'Aimable, variously described in the historical docu-
ments as a flute, pinque, or hooker of 250 to 300 tons burden (Gilmore, 1973:92), was
Source: In the Realms of Gold; The Proceedings of the Tenth Coriference on Underwater Archaeology. edited
by Wilburn A. Cockrell (1981), pp. 10>-116, Fathom Eight, San Marino, California. Reprinted by perrnisison.

363
364 J.B. Arnold III

lost near the mouth of the bay when the French had just arrived and were making their
initial entry in February 1685. Minet (l685?), the expedition's engineer, drew two
excellent maps recovered from French archives showing the location of this wreck in
the outer reaches of the Gulf of Mexico end of Pass Cavallo, the natural entrance to
Matagorda Bay. A certain amount of confusion (Gilmore, 1973:93-94) has resulted
from a Spanish map produced several years later (Cardenas, 1691) by one of numerous
expeditions sent from Mexico to find and expel the French; the map, based on hearsay
evidence, placed the location of the wreck inside the bay proper between DeCrow point
and the modern fishing village of Port O'Connor (Figure 1).
The second ship, the Bel/e, was a smallfrigatta of about 5O-Q0 tons and less than
50 ft long (Gilmore, 1973:92). She was lost in late January or early February 1686
while on an exploratory mission in the bay. The location of this wreck is well docu-
mented by both French and Spanish sources. Her hull was still relatively intact and
above water when the Spanish arrived. She lies four leagues (about 10 miles) up the
bay side of Matagorda Peninsula near Green's Bayou and about a musket shot offshore.
Utilizing the historical information from documentary sources and data on shore-
line changes in both Pass Cavallo and the Green's Bayou area (Morton, 1978), two

Figure 1. Matagorda Bay.

Airborne Magnetometer Survey for Underwater Excavations 365

search areas were defined. A base of operations was established in Port O'Connor. We
expected to find a number of anomalies, some caused by modem ferrous debris and some
by modem and other historic shipwreck sites in addition to the La Salle wrecks. We did
indeed find the former in abundance, but the locations of the latter remain unconfirmed.
Other archaeological investigations in the Matagorda Bay area include excava-
tions at the site of La Salle's Fort St. Louis which were carried out by the Texas Memo-
rial Museum in 1950 and whose artifacts are analyzed by Gilmore (1973). Three other
magnetometer survey projects have been carried out in the area (Institute for Underwa-
ter Research, 1971; Meylach et aI:, 1973; McCormick et al., 1978). These were under-
water cultural resource surveys prior to shell dredging and channel dredging to locate
and avoid magnetic anomalies which could indicate the existence of possibly signifi-
cant cultural resources. Magnetometer surveys and underwater site test excavations
similar to the project reported herein but executed in other areas of the Texas coast are
reported in various Texas Antiquities Committee publications (Arnold, 1975, 1977;
Arnold and Weddle, 1978).
This magnetometer survey was carried out with a complete state-of-the-art instru-
mentation package (described in detail below) including proton precession magnetome-
ter, microwave radar positioning, and microprocessor and X-Y plotter to manipulate
the range data provided by the positioning system and record them on magnetic tape
along with the magnetometer readings for subsequent computer processing. The search
areas were not completely and systematically accessible by boat so another means of
transportation was used on an experimental basis. A helicopter was temporarily pro-
cured to provide an airborne survey platform and the results were later checked from a
surface survey vessel in the areas a boat can reach in both the Green's Bayou and Pass
Cavallo survey areas. The site test excavations were carried out using a blower or prop
wash, standard scuba gear for life support, and typical archaeological provenience con-
trol based on trilateration measurements on the bottom from subbase points which were
shot in with theodolites from shore.
Funding for this project was provided by the State of Texas under the auspices of
the Texas Antiquities Committee, Dr. Fred Wendorf, chairman. A matching survey and
planning grant was apportioned from the Annual Program Grant of the Office of Archae-
ology and Historic Preservation, Heritage, Conservation and Recreation Service, U.S.
Department of the Interior, to the Texas Historical Commission, Mr. Truett Latimer,
executive director. The X-Y plotter utilized in the positioning system was loaned to us
free of charge by the president of Houston Instruments, Mr. George Moore.

MAGNETOMETER SURVEY
The timing of a short-duration magnetometer survey such as this is critical if the
instruments are acquired by lease as in our case. The vagaries of fate involved in get-
ting the various instruments installed and working properly in as little time as possible
are bad enough. Those imposed by the capricious nature of corporate shipping depart-
ments and the air freight services are approaching the limits of tolerability.
In carrying out three of these I-month long (the minimum lease period available
from the instrument manufacturers) magnetometer surveys it has been my experience
that one must count on late instrument arrival and several days effort to install the elec-
tronics and work out the bugs. Being operational after a lapse of only 7 days after
366 J.B. Arnold 111

scheduled instrument arrivals would be an exceptionally fine performance. This season

was our worst experience in this regard with fully 17 days involved in delays related to
shipping and installation problems. Fortunately the situation was alleviated by two fac-
tors. First the helicopter rental was so expensive that we had not planned and indeed
could not afford to use it for anywhere near a month, and also it was a sufficiently
faster method in that we could cover the search areas in 7 or 8 days. Second, the instru-
ment companies were willing to make good the delays involved in shipping foul-ups by
letting us keep the equipment past our lease deadline.
Having experienced the time-consuming nature of the electronics installation
process in the past, we decided that it would be well to avail ourselves of the option of
having the positioning system company send an electronics technician from their staff
to expedite this aspect. The expense involved, while not inconsiderable, was undoubt-
edly justified by both his extensive experience and familiarity with the instruments and
by the personalized instruction in the operation of the instruments he provided for my
crew and myself.
Had everything gone according to the original timetable we would have had
about 10 days after the helicopter survey to install the electronics in the Anomaly, our
agency's 34-ft aluminum-hulled research vessel, and continue surveying. The purpose
of this phase would have been to double check part of the helicopter survey magnetic
data from the water's surface and then to expand the survey into adjacent areas of Pass
Cavallo and the nearby Gulf shoreline in search of numerous other wrecks known to lie
in the general vicinity. As it happened we had just enough time to double-check the
magnetic data in portions of the Green's Bayou search area.
The magnetometer survey crew consisted of seven people: project director-
marine archaeologist, electronics technician, two assistants to set up and take down the
radar transponders on shore, one assistant to man our shore headquarters and run
errands as necessary, and two other assistants occupied setting up shop for the site test-
ing phase during the helicopter survey and used on the survey vessel during the surface
survey. Part of this time and at various other periods during the session we also had the
volunteer assistance of Mr. Bill Wiley, a graduate student in archaeology at Rice Uni-
versity. During the helicopter phase of the survey, Bill Crow, electronics technician-
archaeologist, and I took turns flying. During the surface survey he ran the positioning
and I ran the magnetometer. It would have been better to have had two instrument
operators during the helicopter phase, too, but with the size helicopter we were able to
afford neither payload weight limitations nor space considerations allowed more than
one operator and the pilot who, of course, was provided by the helicopter company.
In selecting the appropriate aircraft for a helicopter survey I can offer some advice.
There were several companies serving the offshore oil industry in the area from whom
one might obtain the services of a helicopter and crew. To avoid added expense it is
advisable to select one that has an operational base near your search areas. A fuel supply
and a ground crew of at least one seems to be the minimum requisite for such a base. We
selected Evergreen Helicopters, Inc. which already had one or two helicopters operating
out of Port O'Connor. When you make your initial contact with such a company, you
may be dealing with a staff person who is in essence a salesman, someone who may be
prone to exaggerate the capabilities or the allowable functional parameters of the heli-
copter under discussion. Establishing contact as soon as possible with the chief of opera-
tions will offer a better idea of the actual physical capabilities of the aircraft involved.
Airborne Magnetometer Survey for Underwater Excavations 367

There are a number of parameters to consider such as the weight involved in instru-
ments and crew, the altitude and speed desired, and the duration of the flights in time and
distance. The helicopter we used was a Bell 206B Jetranger. The maximum gross weight
for this aircraft is about 3200 pounds including the aircraft fuel, pilot, and payload. The
maximum payload was only 400 pounds. We had originally planned on carrying our own
power supply in the form of 12 volt automobile batteries. This would have put us over
the limit of the payload and the wet cell nature of the batteries caused further concern
since if the acid spilled extensive damage to the aircraft could be expected.
Taking power from the helicopter's generators proved to be impossible because
of noise introduced in the magnetometer signal, but this was eventually overcome by
taking our power directly from the helicopter's battery, something they were at first
reluctant to allow.
Speed and altitude are also critical factors in helicopter operations. A certain
amount of forward speed is necessary at any given altitude to allow a safe landing by
autogyration should the engine cease to function. We had a towed sensor that was sus-
pended by hand below the helicopter on a 100-ft cable after we had reached the survey
area. It was necessary to keep the sensor not more than 20 ft above the water in the
maximum depths of about 10ft in most of our survey area in order that the distance
from sensor to target not be too great for practical results. Our operational target alti-
tude was therefore 120 ft. At this height the minimum safe speed for our helicopter is
about 40 mph. In order that the magnetometer samples cycling at a rate of one per sec-
ond not be too far apart, we wanted to fly between 20 and 25 mph. This problem was in
part overcome by using a helicopter with pontoons but in essence we simply had to fly
at speeds below acceptable safety standards. Other aircraft-related aspects to be kept in
mind in planning such a survey are the 100-hr maintenance and safety checks which
can consume several hours.
While flight duration was fairly short at our speed and altitude, a little less than 2
hours per flight, there was a definite advantage over surface vessel in transit time to
and from the survey area at speeds of about 80 mph. While electric winch systems are
available for deploying and retrieving the sensor, we found that letting it out the door
by hand while the helicopter hovered worked fine. The problem of the tendency of the
cable to twist and kink was overcome by the old sailor's trick of coiling it in a figure
eight. To recover the sensor we simply landed on the nearby beach with the sensor still
deployed. It was remarkable how gently our pilot, Mr. John Stone, could set the sensor
down. It was then an easy job to hop out and pick up the sensor and cable, stow them in
the near cargo hatch, and return to base.
The distribution of the instrumentation package and crew in this five-seat heli-
copter was as follows. The instruments were for the most part mounted in a rack which
occupied the left rear and center seats. The operator sat in the right rear behind the pilot
who sat in the right front position. The left front was occupied by the X-Y plotter
which the pilot turned on or off as necessary. The sensor was deployed by the operator
out the right rear door. It was secured via seatbelts with one end secured to the usual
seatbelt fastening point and the other to the tow cable's strain relief. The buckle con-
necting the two sections of the seatbelt provided a quick release capability in the event
of fouling. The antenna of the positioning system receiver/transmitter (RlT) unit was
mounted on a bucket below the fuselage between the pontoons. The R/T unit itself was
in the rear cargo hatch.
368 J.B. Arnold III

As far as guidance along the desired survey track there was a left/right indicator
controlled from the microprocessor mounted on top of the control panel. Tracks were
spaced at 50 m, a spacing deemed appropriate from previous experience with ship-
wrecks on the Texas coast. The quality of the guidance in terms of deviation along the
desired course was as good or better than that achievable from a surface vessel.
In critiquing this airborne magnetometer survey I must say that accurate altitude
control was a problem. I watched the altimeter myself as we flew along and it seemed
to be indicating an acceptable level. However, our ground crews at the transponder sta-
tions would occasionally indicate that it seemed to them we were a little high. When
we rechecked the data from the water surface in the Green's Bayou area in depths of
from 12-13 ft maximum in the bay, to 3 ft, we found many anomalies not located dur-
ing the helicopter survey. The extra anomalies were mainly of a point source nature of
the type generated by modern ferrous debris, however. We did not have this experience
in the Pass Cavallo area. During extensive rechecking with the magnetometer in that
area we found only one anomaly which had not been located by the helicopter. To
solve this problem of accurate altitude control, it will be necessary in future applica-
tions of this technique to be sure the helicopter is equipped with a radar rather than a
barometric altimeter. There is even the possibility of having the readout from the elec-
tronic radar altimeter interfaced and recorded on magnetic tape with the magnetometer
readings much as would be done with bathymetric data.
The basic instrumentation package, while very similar to those of previous sur-
veys (Arnold, 1978), was acquired by lease from two companies instead of three as was
formerly the case. This was accomplished by acquiring the data processor from
Motorola as well as the Mini-Ranger III (MR III) microwave radar ranging device. This
data processor had not been available during our previous efforts and it was thought
that eliminating a separate supplier for this instrument would facilitate procurement
and timely delivery. The Mini-Ranger Data Processor (MRDP) also offered faster cycle
time at once per second than the programmable calculator used previously which had 8
cycle time of once every 2 seconds.
In addition to the MR III and the MRDP, a microprocessor or minicomputer, other
positioning system accessories furnished by Motorola included a Texas Instruments data
terminal (a typewriterlike keyboard used to communicate with MRDP), a Techtronics
cassette tape drive with cassettes (for recording the magnetometer readings and the X-Y
coordinates from the positioning system for later computer processing), a Houston
Instruments X-Y plotter (for preparing preplots and postplots), an MR III track indicator
(providing left-right course correction and percentage completion information for the
pilot), and a 250-watt static inverter to provide AC power for those portions of the elec-
tronics system that required it. The MR III itself came with the instrument console, C
band RlT unit with 6-dB antenna to be installed on the survey vessel, and two transpon-
der units or reference stations with 19-dB antennae for setting up at known locations on
shore. The 24-volt DC transponders are each powered by two automobile batteries. We
set up our reference stations at conveniently located U.S.c. and G.S., United States Geo-
detic Survey (U.S.G.S.), Corps of Engineers, or National Oceanic and Atmospheric
Administration (N.O.A.A.) benchmarks of which there were plenty to choose from and
for which the precise map coordinates were readily obtainable. The Universal Transverse
Mercator grid coordinate system was used for this project. The sites of the reference sta-
tions were chosen so that each survey area could be covered by one pair of transponders.
Airborne Magnetometer Survey for Underwater Excavations 369
The survey guidance software provided with the MRDP proved to be very effec-
tive. The unit also has a useful internal clock feature which allows either the real time
or the elapsed time to be recorded on tape with each magnetometer reading and pair of
x- Y coordinates. Another useful feature is the event mark which likewise is recorded
on the magnetic tape during the survey each time the space bar is pressed down on the
data terminal. Equipment acquisition was indeed facilitated by eliminating one sup-
plier. On the other hand, the postplot software acquired with the MRDP was com-
pletely inadequate for magnetometer data having been designed for bathymetric data.
This software utilizing the X-V plotter was supposed to have provided preliminary data
analysis at the close of each day's surveying. Another aspect to bear in mind is the sub-
stantial cost of the magnetic tape cassettes, about $60 each, of which we used about 42
for a total of 207 linear miles of survey. Transferring the data from this variety of cas-
sette to I-inch tape on the local level for the computer analysis of the data also proved
to be a nontrivial problem.
I highly recommend one other option available with the positioning system. The
spare parts kit gives one the opportunity of repairing malfunctions with the minimum
of downtime. We were plagued with electronics malfunctions in both the positioning
and magnetometer systems this season in contrast to the reliable performance of this
same equipment in previous surveys. The only bright side of the positioning system
problems which included having to replace the RlT unit was that good repeatability
was achieved in spite of changing out numerous components.
The magnetometer system was acquired from Varian of Canada. The same model
V 85 instrument console was used with an airborne sensor for the helicopter survey and
with a marine sensor for the surface survey. In addition to the BCD (binary coded deci-
mal) interface with the MRDP for recording the magnetic data on tape, these data were
displayed in real time as acquired on a Hewlett-Packard dual channel 10 inch strip
chart recorder set on 0-100 and 0-1000 scales. Again supportability proved to be a
problem in the event of malfunction. Perhaps we were just lucky in this aspect in pre-
vious years. However, I have begun to think that instruments locally acquired and most
importantly locally supportable or repairable are the way to go even if it means a
slightly higher lease or selecting a competing manufacturer's instrument which may not
have quite all the same or the desired features.
On preliminary analysis of the magnetometer data the results of the survey were
quite promising. In the Green's Bayou area we had located three very good anomalies
which seemed to be probable shipwreck sites and 22 strong but isolated point source
anomalies ranging from 18 to 210 y which seemed more likely to be caused by modem
ferrous debris. In the Pass Cavallo area we had located three very good anomalies,
probably shipwreck sites, and two isolated point source anomalies. The anomalies
located by the helicopter survey in the Pass Cavallo area illustrate two important
points. In two cases we have the classic shipwreck signatures on 50-m spaced tracks
with small readings on distant tracks and a very large reading in the center. On the last
day we had the helicopter, I extended the survey by switching to 100-m spacing for the
last several tracks. As it happened there were two interesting but small anomalies indi-
cated on neighboring 100-m spaced tracks, one of 3 y and one of 7 y. I hypothesized
that in between should be a major anomaly and indeed there was. In 13 ft of water and
with at least another 30 ft to the target we located an anomaly measuring over 1550 y
when we rechecked from the water's surface later during the site testing phase. The
370 J.B. Arnold III

implications of these hard data on anomaly size and range of detectability for the u.s.
Outer Continental Shelf cultural resource surveys with their 150-m track spacing speak
for themselves.
The next step was to initiate the program of investigation into the causes of these
anomalies through the underwater site test excavation phase of the project.

UNDERWATER SITE TEST EXCAVATIONS

On July I the remaining six crew members joined the project. All 13 were certi-
fied scuba divers and most had land or underwater archaeological experience. Addi-
tional special skills represented on the crew were photographer, illustrator, certified
scuba instructor, emergency medical technician, mechanic, and electronics technician.
The site testing consisted of two phases. First was relocating the anomaly and
further evaluating its true strength and extent with the magnetometer; and second, if the
anomaly continued to look promising after this recheck, digging of test holes to iden-
tify the cause of the anomaly. Relocating the anomaly was carried out using one of two
methods of guiding the survey vessel back to the appropriate point. The electronic posi-
tioning system, when it was functioning, made relocating the anomalies a very easy
matter. So much so that I recommend retaining this instrument for the site testing phase
if the budgetary situation permits. Alternatively we used theodolites, optical survey in-
struments, positioned at two known points to guide the survey vessel via walkie-talkie
radios along axes defined by calculated angles that intersect over the location of the
anomaly. In both cases an agency-owned portable proton procession magnetometer, a
Varian Model M-50, was towed behind the vessel to relocate and evaluate the anomaly.
At the end of the season this IO-year-old instrument finally gave up the ghost and we
rented a Geometrics 806A proton procession magnetometer to finish the project.
The two points used for the theodolite stations were the same as the transponder
stations for the Pass Cavallo area but the transponder stations for the Green's Bayou
area were much too far away. There were two convenient spots nearer the survey area,
but there were no official survey markers with known coordinates. One point we chose
was a platform in the bay which supported a range tower. The other was beside the
breakwater or bulkhead of the small harbor of an abandoned air base. The advantage of
these two spots was that we could station our survey vessel immediately beside each of
them and get their coordinates from our electronic positioning system. In this kind of
project utilizing electronic positioning, internal consistency is of the greatest impor-
tance so these two stations with coordinates provided by our own MR III and MRDP
were actually superior to benchmarks established and surveyed by others.
Provenience control was carried out by the same two methods. The position of
each test hole dug with the blower or prop wash deflector was recorded. The ease of ac-
quiring the position with the electronic system was truly amazing. It required only a push
of the button and the UTM coordinates were printed out. With the optical system two
angles were recorded and the UTM coordinates would be produced later after an exten-
sive series of trigonometric calculations (Arnold, 1975). The advantage of having the
UTM coordinates instantly is that one can plot the positions of the holes as they are dug
rather than having to wait until the end of the day after returning to headquarters. This
eliminates the necessity of using dead reckoning in the moving of the boat from place to
place and makes a systematic pattern of the test excavation much easier to achieve~
 ~3
(\)

3:
Table I. Anomalies investigated by test excavation - 1978 ~
~
Strength on recheck Water No. of holes to
Anomaly no." Survey strength Boat Swimmer depth locate cause Comments I
I. GB, offset +700, event 17 1711 16651 6 0 Oyster shell mound in 12 ft of water. Modem debris recovered. en
c
Possible oil well site. 41MG39.
2. GB, offset -150, events 3 & 4 140y 671 1431 6 2 Multiple peaked anomaly, modem steel plate found. Probably 19th- ~
or 20th-century shipwreck. 41MG36. ()l
.,
3. GB, offset -250, events 7 & 8 631 200y 2441 6 0 Multiple peaked anomaly. 19th-century sailing shipwreck. 41MG37. C
4. GB, offset -250, event 10 210y 1961 190y 10 5 No cause found. Sand overburden too deep. Point source anomaly.
Probable modem debris. ~
5. pc, offset -250 1851 2900y 34151 13 0 19th-century steamship wreck. 41CL56.
6. GB, offset -150, event 2 130y 751 100y 9 3 No cause found. Sand overburden too deep. Point source anomaly. a~
Probable modem debris. ~
7. pc, "stranger" offset -250 50y 1200y lost phase 14 Uncovered railroad rail and large timer. Probable 19th- or
m
)(

20th-century wreck. 41CL57.

8. GB, offset -200, event 5 271 800y 8591 10 Modem steel pipe, boom, or mast. 41MG38. Associated with
~
~.
41MG37. g
9. GB, offset +100, event 10 160y 1831 2131 II Nothing found. Unable to penetrate bay muds.
10. GB, offset +700, event 16 311 200y 2161 13 Nothing found. Unable to penetrate bay muds.
'"
II. PC, offset -750 31,71 0n 1200y 15571 13 7 Nothing found. Unable to dig deep enough in sand to reach cause.
100-m tracks 41CL58.
12. pc, offset +100 250y 3520y 21 o 20th-century steel hulled shipwreck. 41CL55.
", GB, Green's Bayou; PC, Pass Cavallo.

-...)
"'"
372 J.B. Arnold III

CONCLUSIONS

The results of the test excavations are summarized in Table 1. Twelve anomalies
were tested and from these five shipwrecks were identified, two 19th century, two 19th
or 20th century, and one 20th century. One of the anomalies which remained unidenti-
fied because of extreme depth of overburden in the Pass Cavallo area could possibly be
one of the La Salle wrecks, the Aimable. None of the other anomalies tested but
unidentified are promising in this respect. Furthermore, all of the anomalies discovered
during the magnetometer survey which seemed promising in size or, w}th what we
learned during the testing, in location have now been tested with the exception of the
one in Pass Cavallo mentioned above. Heavy equipment in the form of a large commer-
cial dredge will be required to test that anomaly, equipment presently beyond our
agency's financial means. As for the Belle I feel the chances are good that she lies in
the waters shallower than the 3 ft into which we could take our vessel or under what is
now marsh land in the washover fan behind Green's Bayou. Plans for further magne-
tometer work in this area are under consideration.
Tables 2 and 3 present management data of interest to those who might plan to
carry out a similar survey. In these tables survey coverage and costs are presented. The
magnetometer survey came to $37,084 consisting of $6897 in salaries and expenses for
the crew excluding permanent Texas Antiquities Committee staff, $3718 for magne-
tometer lease, $14,145 for the MR III and MRDP etc., $8324 for helicopter rental, and
about $4000 for expendable supplies. The site testing came to $29,290 for a total for
the season of $66,374. By way of comparison with 1978 figures on the airborne magne-
tometer survey, the costs for 2 months of surface vessel survey carried out in 1974 and

Table 2. Management data for magnetometer survey

Linear survey coverage Area survey coverage
Green's Bayou area 126 linear km (78.75 linear miles) 6 km2 (2.34 miles2)
Pass Cavallo area 205 linear km (128.125 linear miles) 11.5 km 2 (4.49 miles 2)
Total 1978 331 linear km (206.875 linear miles) 17.5 km2 (6.83 mi1es2 )
Cost 1978 $112llinear km (4 I 79/1inear mile) $21191km 2 ($5430/mile z)

Table 3. Management data for site test excavations

1975 and 1977 totals and averages 1978 totals and averages
Time Expense Time Expense
Bottom time (hours) 412 $2.59/min 55.47 $8.80/min
Operational hours 1,347 $42.55Ihr 352.5 $83.09/hr
Man-hours 16,522" $3.40/man-hr 4582 h $13.00/man-hr
Salaries and expenses $43,353 $16,765
Equipment and supplies $12,760 $12,525
Total expended $56,113 $29,290
Length of project 5Yz months I Yz months
Anomalies tested 35 12
$/Anomaly tested by excavation $1,603 $2,440
"Average crew 12~, hCrew 13.
Airborne Magnetometer Survey for Undetwater Excavations 373

1975 were an average of $40.37 /linear mile for a total of 1201 linear miles or about
$IIOO/mile for about 44 miles surveyed. The reason the expense is so much greater for
the airborne survey is mainly because of the helicopter rental, but also partly because
of the different goals of the surveys. In 1974 and 1975 our goal was to cover as many
miles and as much area as possible. In 1978 we were in a project oriented situation
searching for two particular ships in definable survey areas. Even at the greatest
expense illustrated, the airborne survey offers significant benefits which justify the
added cost, namely, the ability to systematically survey areas inaccessible to the sur-
face survey vessel.

REFERENCES
Arnold, J.B., III, 1975, An Underwater Archaeological Magnetometer Survey and Site Test Excavation Project
off Padre Island, Texas. Publication 3. Texas Antiquities Committee, Austin.
Arnold, J.8., III, 1977,1977 Underwater Site Test Excavations off Padre 1sland, Texas. Publication 5. Texas
Antiquities Committee, Austin.
Arnold, J.B., III, and Weddle, R.S., 1978, The Nautical Archaeology of Padre 1sland: The Spanish Shipwrecks
of 1554. Academic Press, New York.
Cardenas Map, 1691, Map of the Cardenas-Llanos Expedition. From the J.P. Bryan Collection (24a, 1691).
The University of Texas at Austin.
Gilmore, K., 1973, The Keeran Site: The Probable Site of La Salle s Fort St. Louis in Texas. Office of the State
Archaeologist Report 24. Texas Historical Commission, Austin.
Institute for Underwater Research, Inc., 1971, Archaeological Survey for Shipwreck Sites in Northwestern
Matagorda Bay, June 1-12, 1971. Dallas.
McCormick, O.F., Gilmore, K., and Hays, T.R., 1978, Cultural Resource Survey at the LaSalle Terminal,
Matagorda Bay. Final Report to El Paso LNG Terminal Company. Archaeological Program. Institute of
Applied Sciences. North Texas State University, Denton.
Meylach, M., Clausen, C.J., and Arnold, J.B., III, 1973, 1973 Matagorda Bay Magnetometer Survey. Texas
Antiquities Committee, Technical Report 36. Austin.
Minet, 1685?, Two maps of Pass Cavallo and Matagorda Bay. Retrieved from French Archives and recorded on
microfilm at the Old Spanish Missions Research Library, Mission San Jose, San Antonio. (Sketches taken
from those originals are published in Bolton's Spanish Exploration in the Southwest. 1542-1706,
1959:314-315. A historic copy of both ofMinet's maps was also found in the Archivo General de Indias in
Spain under Mexico, 61--{}-20, and is also on microfilm at the Old Spanish Missions Research Library.)
Morton, R.A., 1978, Approximate Inlet and Shoreline Positions in 1685. Matagorda Bay Area, Texas. Prepared
for the Texas Antiquities Committee. Bureau of Economic Geology, The University of Texas at Austin.
 Chapter 33

Magnetic Search
and Survey in Shallow
Water and Beach Areas

JACK HUDSON, KAY G. HUDSON,

and HARRYW RHODES III

Not all shipwrecks of interest to the nautical archaeologist are found in depths requir-
ing diving techniques. Shipwrecks can be found in beaches, under sand dunes, behind
barrier islands, and in sand flats, bays, and rivers. The use of magnetic survey in deal-
ing with locations in extremely shallow water or on beaches requires a different
approach, as the standard marine magnetometer tow system cannot be used efficiently.
Under certain circumstances, the marine sensor can be lifted by net buoys attached by
lines behind the sensor, but one should be aware of problems encountered with this
technique. Buoys generally cause the sensor to fishtail while under tow. These changes
in orientation cause gradient changes in the data, as the sensor observes the magnetic
field in different directions. If the fishtailing is rapid enough, the data often appear to
show constant anomalies or an unacceptable level of noise. The addition of fins helps
to stabilize the sensor; one can also use a drogue chute or small changes in speed or
length of tow cable to achieve good data.
For extremely shallow water and/or land survey, the archaeologist has again bor-
rowed from the geophysicist and run giggling off with his magnetometer base station and
survey sensor. The small sensors are designed for base station recording oflocally occur-
Source: In the Realms of Gold; The Proceedings of the Tenth Conference on Underwater Archaeology. edited
by Wilburn A. Cockrell (1981), pp. 117-119. Fathom Eight, San Marino, California. Reprinted by permission.

375
376 J. Hudson et aI.

ring 24-hour magnetic cycles. They are also designed to be used as portable survey units
for the location of geological structures. We have used them successfully in the latter
configuration to look for archaeological features in shallow water and on beaches.

RNERS AND BAYS

Our first experiment with shallow water magnetic survey occurred in 1973174 as
part of the Red River survey in Louisiana (from mile 0 at the locks on the Mississippi
River to mile 287 at the Arkansas state line). The survey was funded by the New
Orleans District, Corps of Engineers, and included historical research to obtain general
locations, dates, vessel descriptions, and information on causes of wrecks. This
research, coupled with the magnetometer survey, provided some 180+ possible wreck
locations. It should be pointed out that this survey was a first phase effort to gain base-
line data on a major river system.
This was the first time we used the small base station sensor, and it was simply
rigged on the end of an aluminum pole off the bow ofa 26-ft cabin cruiser. Later, when
the river fell some 13 ft in 3 days, we changed to a flat-bottomed aluminum john boat
to finish the survey.
The same configuration was used in a survey in Mobile Bay for a proposed chan-
nel under a contract with the Mobile District, Corps of Engineers. When our survey
boat's engine gave up the ghost, we simply transferred the magnetometer and sensor to
a local sport fisherman and finished the survey in good order. This simple arrangement
works well in depths up to about 20 ft. As with all directional sensors, it is most impor-
tant to orient the North arrow on the sensor until the signal is maximum.

BEACHES AND SHALLOW WATER

In 1977, we did a beach survey of the east beach area at the mouth of the Colo-
rado River, Matagorda, Texas. During this survey, we used an all-terrain vehicle to
carry the magnetometer console and batteries, with the sensor mounted on a staff and
attached to the magnetometer by 200 ft of cable. This configuration allowed us to map
the beach along precise lines set up by transit, with readings taken at 5-m intervals. By
timing the survey to coincide with low tide, we were able to take the all-terrain vehicle
into the water off the beach. The 200 ft of cable allowed magnetometer measurements
to be made almost to the surf line, as a crew member carried the staff and sensor into
the water. Good data were obtained using this method.
In 1978, we surveyed a proposed pipeline route across Pelican Island in
Galveston Bay. Historical records found by the late Robert Alderdice, a local historian,
mentioned that abandoned ships from the early 19th century were towed behind what
was then known as Pelican Spit and left to rot. The Pelican Island of today was formed
over the years from deposition of fill from channel-dredging operations.
For this survey, we utilized the all-terrain vehicle again, with magnetometer, bat-
teries, sensor, and 200-ft cable. The sensor this time was placed in a harness developed
for land survey. We found that by moving the all-terrain vehicle at a slow pace to
match the speed of the crew member wearing the sensor harness we could obtain excel-
lent data along a survey route. About half the island was surveyed, but work had to be
abandoned for the moment because it was found that the middle of the survey area had
Magnetic Search and Survey in Shallow Water and Beach Areas 377

been used for local dumping over the years. It is anticipated that the dump and trash
piles will be cleared and that we will redo the survey in the same manner.
During the week of October 26 through November I, 1978, we participated in a
joint project with the Texas Antiquities Committee, arranged by J. Barto Arnold III.
Two TAC staff members, Margaret Leshikar and Edward Aiken, met us at Port O'Con-
nor, Texas, bringing with them a 19-ft inflatable boat with motor and survey instru-
ments. We brought our 34-ft survey/research vessel Gamma, and the magnetometer
with marine and land sensors.
The problem we worked on together was to survey an area on the Bay side of the
Matagorda Peninsula. This area had been the subject of historical research and was
identified as the general area in which La Salle lost one of his ships, the L 'Belle. The
area had been covered by an airborne magnetometer survey from a helicopter during
the 1978 field season. Because of the extremely shallow water, it was not possible to
use the TAC survey vessel Anomaly for intensive survey.
After a briefing from the TAC staff members, it was decided to use Gamma as a
mobile base station staying at about the 6-ft contour, while the inflatable boat was rigged
for shallow water survey using a base station sensor on an aluminum pole off the bow.
The survey run inshore of the 6-ft contour using this method yielded several good
anomalies in 2 to 3 ft of water. On the third day we decided to run some tests, placing the
sensor in the interior bow compartment of the inflatable boat. After orienting the sensor
for maximum signal, the survey was continued. This configuration yielded excellent data
and eliminated the need for a pole. Several more anomalies were found in this manner
before the survey work was terminated because of an increase in wind and chop in the
area. Survey by this method yielded excellent data in 18 inches of water.

SUMMARY
Shipwrecks are not confined to offshore areas: they can and do occur in very
shallow water and in beaches and sand dunes.
The standard marine sensor can be used in shallow water with buoys, but there
are problems with this configuration. A more efficient means of magnetic survey in
such areas involves the use of the land survey/base station sensor. This can be hung
from a pole off the bow of a small boat, placed in the interior bow of an inflatable boat,
or, in the case of beach areas, carried in a special harness or on a staff by a crew mem-
ber. The latter method also utilizes an all-terrain vehicle carrying batteries as a power
source and the magnetometer console, attached by 200 ft of cable to the sensor. At low
tide the all-terrain vehicle and the sensor may be taken into the water, extending the
survey to 200 ft beyond the point where the vehicle must stop, and obtaining good data
to the surfline.
 Part V.C

Sonar
".~i).:
~

':.i
-r/: ..

In Chapter 34, Klein, a leading developer and producer of underwater sensing devices,
describes side-scan sonar as a scientific tool more thoroughly than in earlier articles.
Digital storage has become much easier since this article was written. Distortion fac-
tors, digital processors, and improved recording techniques, among other features, rep-
resent the speed at which this particular tool has advanced.
The collection of sonar data is no longer the last step before visual interpretation.
Hardware/software packages specifically developed to aid in sonar data collection and
interpretation now feature many of the same "user friendly" programs familiar from
other computer applications. Interpreting thin gray lines on paper printout has become
far more versatile and flexible through computerized data manipulation. Chapter 35
outlines the latest features of underwater archaeological acoustics.

379
 Chapter 34

New Capabilities for

Side~Scan Sonar

MARTIN KLEIN

INTRODUCTION

Now that side-scan sonar has become a generally accepted remote sensing tool, techni-
cal developments are proceeding at a rapid pace. Until recently, basic systems primarily
used straightforward analog signal processing, and digital techniques were confined to
recorder functions such as stepping motor drivers. Now manufacturers are adapting
methods to use more sophisticated analog and digital techniques to improve the quality
of the sonar images. Continuing improvements are also seen in graphic recorders,
transducers, and towing vehicles.

GRAPHIC RECORDERS

One of the important elements in the quality of the sonar images is the graphic
recording mechanism. There are a variety of methods available including "wet" and
"dry" electrosensitive paper recorders; stylus, moving band, and helical scan elec-
trodes; multistylus electrodes; and fiberoptic recorders. Each technique has its own dis-
advantages and advantages. At Klein Associates, we use a precision helical scan
Source: Underwater Archaeology: The Proceedings of the Eleventh Conference on Underwater Archaeology,
edited by Calvin R. Cummings (1982), pp. 186-189. Fathom Eight, San Marino, California. Reprinted by per-
mission.

381
382 M. Klein

recorder of our own design and manufacture. We find that this recorder seems to have
the best combination of wide dynamic range, simplicity and reliability, low power con-
sumption, versatility, and ease of paper loading and maintenance. It also allows imme-
diate viewing of the records which we find is an important consideration in many
operations at sea. We typically use "wet" paper for the highest quality pictures, al-
though a new patented process allows the use of "dry" paper in the same recorder (U.S.
Patent 4,137,538). We also have transceivers available that allow a customer to use our
side-scan sonar on any graphic recorder. The recorders can operate on 24 volts dc (typi-
cally two standard automobile batteries in series) or on 110 or 220 volts ac.

MAGNETIC TAPE RECORDING

One of the new graphic recorder developments is the ability to record sonar data
on magnetic tape and then play it back again on the graphic recorder. This presented a
number of interesting and difficult technical challenges. The graphic recorder helix
drum is a high inertia load for the helix motor which turns the drum. It is also a load
with widely varying torque and a large spread in the speeds required for the different
sonar ranges. Yet the drum must turn at a precision constant rate in order to keep the
quality of the sonar images. We used a printed circuit motor, an optical tachometer, and
a precision crystal oscillator as a frequency source combined with a highly sophisti-
cated phase-lock loop and a special power driver for the motor. The end result is a sys-
tem that still can be battery operated but that has a maximum sweep rate of 30 scans
per second. This is more than three times faster than any other mechanically scanned
graphic recorder on the market. A unique feature of the system is the automatic slew.
This allows the servo drive to lock on to a trigger pulse which is coded on a magnetic
tape and to automatically position the start of each sonar trace at the edge of the paper.
Slewing may also be manual to provide a fixed offset in the trace. Another unique fea-
ture is the ability to check a magnetic tape immediately as it is being made (read after
write). This allows the operator to make sure proper tapes are recorded.
An interesting ability of the postprocessing of the magnetic tapes is the ability to
"speed up time." The tapes may be made on one recorder speed and then played back
on a faster speed. Of course the data on magnetic tape may be further processed before
redisplay using some of the techniques outlined in this article as well as other computer
processing.

MAGNETIC TAPE STORAGE

The sonar data may be stored on analog tape recorders or a variety of digital mag-
netic tape recorders. As the sonar is basically continuous video data, there is a large
information flow. This means that in order to store the data in digital form, a lot of bits
are required, even to record the "processed" data (which has less dynamic range than
the "raw" data). Typically an analog tape recorder can store around 4 hours of data
while a large reel of digital tape can only store about 20 minutes of data. So at present,
analog tape is typically used mainly for convenience and economic considerations. If
digital conversion is required, an analog-to-digital converter may be used after the ana-
log tape. Improvements in digital tape storage are continuing, however, so we can
expect digital storage to be the best medium in the future.
New Capabilities for Side-Scan Sonar 383

DIGITAL PROCESSOR
One of our new developments is a Digital Processor which is a module that can
connect to a graphic recorder. Units are available that can accept one. two, or three
channels of side-scan and/or subbottom profiler data. The processed sonar signals are
converted to digital form using an 8-bit analog-to-digital converter which runs at a rate
of 30 kHz (per channel). The 8-bit parallel binary data are brought out to a port on the
instrument for possible further processing by computer or for possible storage on a
digital tape recorder. The signals are stored in memory, and then the stored signals may
be played out at a controlled rate. This allows signal delay and expansion, or "A
delayed by B" similar to an oscilloscope delay. In this way a portion of the record may
be blown up to a larger size. This is a kind of electronic "zoom lens."
Two memories are used. One is used to store data, while the other is reading out
data. Then the memories are switched so the first reads out and the second stores. Four
bits are used for storage, since this is typically adequate for the dynamic range of the
graphic recorder. A unique feature of the unit is that a full 1000 points of data are used
regardless of the amount of expansion. This preserves the point density of the record
and allows the full resolution of the side-scan to be utilized.
The unit may also be used to remove the "water column," which is the space on
the sonar record created before the first echoes from the towfish bounce off the closest
bottom and return.
Each channel may be individually adjusted for delay and range. For instance, this
allows a subbottom profiler portion of a record to have a different range than the side-
scan portion.

SIDE-SCAN DISTORTION FACTORS

Side-scan sonar records are subject to a variety of distortions. These are primarily
the result of to altitude error, aspect ratio error, and slant range error. Altitude error is
basically the blank area or "water column" at the beginning of the record. This space
will vary as the tow fish height varies. Aspect ratio error reflects the fact that the dis-
tance measured along one axis of the recorder is typically not the same as the distance
along the other axis. Therefore, the picture will be stretched or squeezed. This situation
will be aggravated as the ship speed is changed. Slant range distortion is created by the
fact that the sonar actually reads the distance from the towfish to the bottom target, not
the true distance along the bottom.
In the past, some clever schemes were devised to correct for some of these distor-
tions. For instance, Westinghouse used a helical scan recorder with a nonlinear sweep
to try to eliminate the water column and correct for slant range distortion. Unfortu-
nately, the system required the towfish to fly an exact height off the bottom. IFP used
an anamorphic nonlinear camera to correct aspect ratio.

K_MAPSTM

Klein Associates recently introduced the K-MAPSTM (Klein Automated Plotting

System), a microprocessor-based system for correcting the side-scan distortions. The
system is modular so that it may plug into a standard HYDROSCAN system.
384 M. Klein

In order to correct for the water column, the system first uses a built-in echo
sounder to sense the height off the bottom (altitude). The sonar signal is stored in mem-
ory, and the output of memory is delayed by a time equivalent of the time for the first
echo to go out to the bottom directly below the fish and return. A thumbwheel switch
also allows the height off the bottom to be manually entered.
Aspect ratio is corrected by sensing the boat speed or by using a navigation sys-
tem to give the parameter of speed to the system. A microprocessor programmed con-
troller then calculates the required paper speed and feeds out the correct pulse rate to
the digitally controlled stepping motor paper advance on the graphic recorder. As
graphic recorders may have different paper widths, this variable must be considered.
Ship speed may also be manually set by a thumbwheel on the unit.
In order to correct for slant range, the towfish altitude must again be measured
(or entered manually by thumbwheel). Then the sonar data are stored and read out in
nonlinear fashion. The microprocessor must basically solve the equation of a right tri-
angle in which the actual measured range is the hypotenuse, the altitude is one leg of
the triangle, and the true range along the bottom is the other leg. The system assumes a
reasonably flat bottom, although it turns out that a fairly high slope is required to create
a major error.
The resultant sonograph after all of the corrections is typically a great improve-
ment. The operator can now often lay traces side by side and create a sonar mosaic of the
seabottom, similar to a photographic mosaic. Unfortunately, the record is still subject to
a variety of distortions, but we can expect that future computer processing of the data
will further improve the image quality. Still, with the major distortions removed the side-
scan can now be used and interpreted with much more efficiency and accuracy.

ALPHANUMERIC RECORD ANNOTATOR

Another development to assist the sonar operator is a new Alphanumeric Record

Annotator. This unit allows a variety of data parameters including Time, Data, Event,
Position information and others to be entered either manually or automatically on the
sonar chart. The instrument is also a modular plug-in to a standard recorder. A micro-
processor controller is used which allows the character size, spacing, and position to be
changed easily. Also, the unit may be programmed so characters can print right to left,
left to right, and also upside down. The unit has an external port so information from an
outside source such as a positioning system may be entered. A keyboard is also avail-
able which permits up to 200 characters per line to be printed instead of the internally
set digits.

AUTO,RANGING

An important feature of the Digital Processor, the Alphanumeric Record Annota-

tor, and the K-MAPSTM is the ability to auto-range. This means that the units will auto-
matically determine what range scale the recorder is set on and will adjust accordingly.
This allows the units to be used with most standard oceanographic graphic recorders. It
also allows the instrument to keep a constant character size even though the recorder
range scale may be changed.
New Capabilities for Side.Scan Sonar 385

VERY HIGH RESOLUTION SONAR

Another important new development is a Very High Resolution Side-Scan Sonar

Towfish. This unit uses a frequency of 500 kHz and a horizontal beam angle of 0.2°.
The unit has a pulse length of approximately 20 ~sec for a range resolution of 3 cm.
This angular and range resolution is five times higher than any previously available
commercial system. We find in field operations that the new unit can resolve subtle
bottom features that we had never seen before on the standard units. Of course there is
a trade-off: the higher-frequency unit has less range than the lower-frequency unit.
Therefore, many customers are now using both units, one for a broad look and one for
closeup target inspection.

FULL OCEAN DEPTH OPERATION

Another interesting development is a new side-scan sonar towfish which can go

to the full known depth of the ocean (12,000 m). The sonar towfish has specially con-
structed transducers as well as a special electronics pressure housing with high strength
stainless steel. The unit has been tested to ultimate depth in a small pressure vessel. It
has also been tested to around 6500 m in the large pressure vessel at the U.S. Naval
Ship Research and Development Center in Annapolis, Maryland. This tank is large
enough so we had enough clear time to look at the actual acoustic output of the towfish
as the pressure increased. No substantial difference in output was observed through the
pressure cycle.
As a related development, we have a new electronic multiplex system that allows
all of the side-scan and subbottom profiler data including power to be transmitted over
a single coaxial cable. This is very useful for deep operations. The time-variable-gain
controls are coded and sent down the cable, so that the operator can use the system as if
it were a normal shallow system.

COMBINED SIDE.SCAN SONAR/SUBBOTTOM PROFILING

Side-scan sonar has also been combined with subbottom profiling in one tow
vehicle, and the data are presented on one recorder. Customers find that using the two
instruments together with a single chart presentation greatly facilitates record interpre-
tation. With use of the Digital Processor, the profiler can be used on a different scale
than the sonar. A new heave compensator is an optional accessory that helps remove
the vertical motions of the tow fish from the sonar record.

CRT DISPLAY

For submersible applications, there is often not enough room for a graphic
recorder. For this application we have developed a Transceiver/CRT Display. This unit
has similar power supplies and controls as a graphic recorder. The data are observed on
the CRT screen instead of the recorder. The unit also has read-after-write capability so
it can be used to make sonar tapes and check their quality at the same time. The tapes
may then be played on a graphic recorder for a hard copy.
386 M. Klein

CONCLUSIONS
The side-scan sonar field has moved into a new, more mature area of develop-
ment. Now that the basic technique has been accepted, customers are looking for fur-
ther improvements in record quality and ease of operation. Many important steps have
been taken to move the technique forward as an operational tool. With increased com-
puter processing capabilities we can expect further significant improvements in the
relatively near future.
 Chapter 35

Isis - Versatile Sonar

Data Acquisition

LAURA JEAN PENVENNE

and JOHN PENVENNE

Side-scan sonar has been successfully employed as a search and survey tool since the
early 1960s. Today, side-scan sonar is used for geophysical, environmental, hydro-
graphic, and limnological studies and mine countermeasures. As sonar interpretation
traditionally has been left to sonar experts, efficient real-time decision-making in situ-
ations where an expert is not available has been impeded.
Processes have been developed that improve the quantitative accuracy of side-
scan sonars and enable interpretation by nonexperts. In 1989, Triton Technology intro-
duced the Q-MIPSTM sonar data acquisition system (F.L. Newton, The Consummate
Search and Survey System - New Developments in Sonar Image Processing, in Pro-
ceedings: USD Conference, pp. 1-10, 1989), which filled a gap in sonar data process-
ing technology by providing data acquisition and image processing for any analog
sidescan sonar. In addition, many of the compensations that ease interpretation could
be applied in real time with Q-MIPS. Budget constraints recently have hindered many
businesses and institutions in acquiring dedicated tools or mission-specific systems. As
a result, multiuse tools have become an essential component in an array of technologi-
cally demanding jobs. In many cases, if a tool is to be used for various tasks, it must
also be portable. Triton's Isis™ modular data acquisition system meets these new
industry requirements for sonar data acquisition.
Source: Sea Technology (1994), 35(6):59--65. Reprinted by permission.

387
388 L.J. Penvenne and J. Penvenne

As of May 1994, Triton has delivered nine Isis systems. Integrated digital signal
processing (DSP) technology, similar to that featured in the Isis system, was first deliv-
ered in a modified Q-MIPS system in 1991. In May 1992, Triton delivered a custom-
modified Q-MIPS/DSPTM system to the U.S. Geological Survey at Woods Hole,
Massachusetts, to be used as a sonar acquisition, storage, and display work station for
the Datasonics SIS-7000™ chirp side-scan and subbottom sonar (Sea Technology, June
1993, pp. 49-55).

ISIS GOALS, SYSTEM OVERVIEW

The first goal was to develop a compact, portable, low-cost shipboard system by
minimizing the packaging while designing for the relatively rugged shipboard environ-
ment. To limit the system to a single display monitor (Q-MIPS has two), an approach
that would allow windowing on a high-resolution monitor was chosen. For the system
to fit in a single rack-mount chassis, the chassis needed front panel access for two full-
height mass storage peripherals.
To simplify sonar record interpretation using data processing and display tech-
niques, an intuitive user interface combined with the availability of compensations for
many of the sources of distortion in side-scan sonar and subbottom profiler imagery
was required.
For versatility, a universal analog sonar interface and a flexible, user-configur-
able serial navigation and telemetry interface would be required (similar to Q-MIPS).
Because, there are now so many digital sonars on the market, a framework was needed
within which custom digital sonar interfaces could be developed.
Isis is configured in a rack-mountable desktop chassis with a single high-resolu-
tion monitor, a keyboard, and a trackball. The system architecture is based on the
PC/XFM-ATISA (industry standard architecture) bus and the MS-DOS® and Win-
dows® operating systems, The basic system handles up to eight channels of analog
sidescan sonar data, eight channels of subbottom data, or some synchronous combina-
tion of the two. Each additional A/D-DSP [analog-to-digital (converter)-digital signal
processing] board adds eight side-scan or eight subbottom channels.
The Isis AID converter samples the incoming analog signals at 170 kHz aggre-
gate - four channels would be sampled at 42 kHz per channel. The system automat-
ically uses the highest possible sample rate based on the ping rate and the number of
channels being used. Thus, the analog signal envelope is generally oversampled by Isis,
preventing aliasing and allowing for the highest resolution of small features.

Storing Sampled Data

Isis can store the sampled data from 1,024 up to 16,384 pixels/ping with the
down-sampling method controlled by the user. Therefore, data can be stored at full
sampled resolution - but for side-scan sonar, this is rarely necessary and somewhat
cumbersome. For example, a sonar transmitting a pulse length of 100 mse has a range
resolution of about 7.5 cm. If 1024 pixels are stored for a range of75 (10 pings/sec), a
single down-sampled pixel is stored for every 7.3 cm of range and the acoustic resolu-
tion of the sonar is matched in the stored data. At 16,384 pixels/ping, storage space
would increase by an order of magnitude and little new information would be gained.
Isis - Versatile Sonar Data Acquisition 389

Isis also preserves the dynamic range of the analog sonar beyond that which can
be displayed on a gray-scale paper recorder. The Isis ND-DSP board digitizes 16 bits
of resolution (96 dB dynamic range) for most analog sonars. This entire dynamic range
can be stored on the system's mass storage devices and any eight contiguous bits can
be viewed in a data display window.
Side-scan sonar and subbottom data normally stored in Triton's Q-MIPS format for
archival and postprocessing may alternatively be read from and stored in true SEG-Y
tape format or in pseudo-SEG-Y modified for random-access media. In addition, Isis can
acquire data through serial ports or through Windows' dynamic data exchange (DDE).
Navigation and towfish telemetry data, for example, are normally accepted by Isis via
one or more serial ports. Other types of data such as bathymetry, gravity, and magnetics
can be acquired by Isis through these interfaces.
All simultaneously acquired data are fused in the Q-MIPS and SEG-Y data formats.
Windows allows independent control of the processing and display of each chan-
nel of data. Isis provides imagery waterfall, wiggle, and voltage windows as well as
text-filled status windows for data display. As the color graphics display has a resolu-
tion of 1280 x 1024 pixels with virtual resolution to 2048 94Symbol'" 1024 pixels, the
user can view a large number of data display windows as well as other Windows appli-
cations and utilities on the single monitor - simultaneously.
Standard processing and display features include: automatic and manual bottom
tracking, contrast stretching, slant-range correction, range gate delay and duration
specification, scale lines overlay, and zoom. Twelve color palettes (made up of 16 or
235 colors, which can be inverted and rotated) are provided for the waterfall display of
side-scan and subbottom imagery. Line and chart colors for wiggle and signal and im-
age processing modules (for target and feature analysis, spatial and transfer domain fil-
ters, multiple suppression for subbottom data, and beam pattern and grazing angle
compensation for side-scan) are available or planned for Isis.

Hardware Innovation

The Isis system is an innovative combination of original hardware and off-the-

shelf components.
Isis comes with a CUP, a graphics processor, an SCSI (small computer system
interface) host adapter, a DSP board with AID converters, and a floppy and hard disk
drive. The hard disk and any other mass storage peripherals are connected via the SCSI
bus and controlled by the SCSI host adapter. Each function board is a plug-in ISA type.
A four-channel, analog signal interface unit (SIU4), designed and manufactured
by Triton in a compact enclosure, simplifies the standard interface to analog sonars. It
detects typical triggers and embeds a conditioned pulse in the first of the four channels.
SIU4s, fitted with ± ¥ volt protection and variable attenuation (zero to 90%), are added
as more channels are needed.
Random-access and sequential mass storage options include higher-capacity hard
drives, magneto-optical drives, and Exabyte 8-mm helical-scan tape drives - each
selected for its high reliability and standard size and function.
The custom CPU chassis fits an ElA 4U 19-inch rack-mount and includes a 300-
watt power supply and an eight-slot, passive ISA backplane. The chassis can house up
to two full-height or four half-height 5.25-inch mass storage devices with front-access
390 L.J. Penvenne and J. Penvenne

and provides a mounting plate for a 3.5-inch form factor hard disk drive (commonly
available in up to 2-gigabyte capacities). The configuration eases hardware expansion
and provides efficient cooling of system components.

Software Advantages
ISIS and Q-MIPS keep pace with real-time acquisition, display, and storage by
utilizing up to four independent processing elements operating simultaneously. Each
element serves a specific purpose toward a unique goal and their interaction is control-
led by the host CUP
A higher level of multitasking has been achieved on Isis through the use of the
Windows' utilities dynamic link libraries (DLL) and dynamic data exchange (DOE).
DOE, one of the ways in which data are shared under Windows, allows multiple applica-
tions to share data and hardware resources through a "client-server" relationship. DLLs
serve as high-level storage devices and enable discrete software modules to be linked
together at run-time rather than during the compile phase. This allows multiple, separate
applications to share a single copy on an executable utility such as a printer driver or a
file format translator. An Isis customer can develop his or her own Windows application
leveraging off existing DLLs and then link it via DOE with the Isis application.
In Isis, DOE allows applications to share data (i.e., real-time sonar returns and
logged contact imagery). Each custom digital sonar interface is developed as a server to
the core application. Triton is actively developing a suite of server applications to en-
hance Isis capabilities, including target analysis and mensuration, beam angle and graz-
ing angle compensation for side-scan sonar, and image enhancement.
The "Target" and feature analysis module run concurrently with Isis and any
other Windows application. The software allows the user to extract (at any time while
running Isis in record or playback mode) a full-resolution image centered on any con-
tact or feature from the imagery stored and displayed by Isis or Q-MIPS. Image size
can be specified by the user. The image, along with navigation, towfish attitude, and
sonar information associated with the pings, is transferred to the Target application.
The length, width, height (based on shadow), altitude (based on shadow separation),
and exact position of the contact or feature can be measured. The contact or feature can
be classified and all of this information can be stored for later recall in Target, Isis, and
other applications running on the Isis platform. The images can be stored in other
standard formats for later use elsewhere; i.e., in word processing, computer-aided
design, or geographic information system applications on other platforms. Many
images can be displayed in Target at one time.
Beam angle and grazing angle compensation (BAC/GAC), a server application
under development, enables operators with minimal training to recognize features in
the sonar imagery without drawing upon sophisticated interpretation skills. Beam angle
and grazing angle compensations can be done in real time in both Q-MIPS software
and Isis.
After sonar imagery has been corrected for slant-range distortion and the water
column has been removed, prominent across-track variations in signal intensity remain
in the imaging - primarily from nonuniform transmit and receive characteristics of the
transducers, absorption, scattering, and spherical spreading. Most sonars include time-
varied gain (TVG) amplifiers in the towfish that correct for spreading and absorption; it
Isis - Versatile Sonar Data Acquisition 391

helps reduce the dynamic range of signals received topside. However, the topside elec-
tronics of some sonars also attempt a single time-varying compensation for beam and
grazing angle nonlinearities. This has always been inadequate because the effects caus-
ing the distortion vary with beam and grazing angles, not with time.
BAC and GAC compensate for sonar beam pattern effects and loss of backscatter
intensity at far range. Features ensonified in the beam nulls near the nadir and those at
far range are normally difficult or impossible to identify. The BAC/GAC module
enables easy detection of such targets by removing across-track variations in intensity
from beam pattern and grazing angle variations.

Demonstration, Post Processing

In March 1994, Isis was demonstrated for Holland's Rijkswaterstaat (RWS) on

the Waal and Rhine rivers. The demonstration was carried out aboard RWS's Onder-
zoeker, a 12-m river survey vessel. A conventional 100 kHz/500 kHz sonar was
mounted on a vertically oriented pole over the vessel's starboard side.
Two channels of side-scan data were collected at a range scale of 56 m and a pulse
length of 100 msec. Global positioning system (GPS) navigation was also acquired by
Isis over a serial port. The bottom was tracked using the Isis bottom tracking algorithm
and the data were displayed with slant-range correction applied. Raw data (along with
the tracked sonar altitude and the position of the vessel) were stored on a rewritable opti-
cal cartridge in the Q-MIPS data format at 1024 pixels per channel per ping.
Data from the RWS demonstration were postprocessed in steps at Triton's Wat-
sonville facility. As data from upstream runs were of higher quality than those from the
noisy downstream runs, only the upstream runs were processed. Only raw data were
stored in the demonstration; therefore, slant-range correction was applied in postproc-
essing - not a normal postprocessing step as both raw and corrected data can be saved
by Isis in real time.
In the next step, a beam angle compensation was calculated using the existing
Q-MIPS sonar image processing software (Isis software module is in development).
The BAC successfully removed the near-range beam pattern from the imagery.
Because the BAC is meant to remove distortions from a smoothly varying, consistent
beam pattern, it has no effect on contacts and features with a high-frequency content.
As a result, a narrow bright reflection off the keel of the vessel remained in the ports ide
imagery. Grazing angle compensation was not necessary as the concave riverbed shape
resulted in a relatively high grazing angle for the entire range.
Finally, the imagery from lines along a section of the Rhine River was digitally
"mosaic ked" using Triton's Vista™ software. This software allows the user to quickly
and easily construct and display mosaics electronically and then to print them on a vari-
ety of printers and plotters. Because of ambient noise and port-side hull interference, the
outer edges of the swaths were omitted. The mosaicked imagery clearly shows the effects
of flow and sedimentation control devices previously positioned in the riverbed by RWS.

ROV Operations
Further versatility was added to the Isis system by the development of the
ROVFlight™ software. Traditionally, survey and ROV operations have been separate and
392 L.J. Penvenne and J. Penvenne

independent because of the dedicated tools and specialized crews associated with each
operational phase. The ROVFlight software, running on the Isis platform, bridges this gap.
ROVFlight was designed to simplify and reduce the time of ROV operations by
allowing direct comparison of features illuminated by the ROV forward-look sonar
with targets and features recorded in mosaicked side-scan sonar imagery. Using ROV
position and attitude information received over the Isis serial port(s), ROVFlight posi-
tions the appropriate Vista mosaic under an oriented ROV model centered on the Isis
display. The mosaic moves in concert with the actual ROV motion. The positions of
geocoded contacts and features detected on the forward-look ROV sonar can be plotted
on the mosaic for comparison with previously detected contacts and features.
ROVFlight, Isis, and Target can run simultaneously on the Isis platform sharing
the same database of logged contacts and features via DDE. From within ROVFlight,
logged features can be recalled in Target. The user can also request that original side-
scan sonar imagery be replayed in Isis. The classifications, measurements, and stored
information for logged features can be modified; changes are immediately reflected in
the ROVFlight display.

Future Development

The Isis system has the potential to acquire multibeam bathymetry and imagery
as well as all types of subbottom information including multichannel seismic and
ground-penetrating radar data. The use of side-scan sonar as a quantitative tool can be
enhanced through the use of swath bathymetry: First, bathymetric information for the
sonar coverage area can be used to eliminate distortions introduced by a flat earth
assumption; second, additional information may be gained from the combination of
side-scan sonar imagery with bathymetric data over the same area. The combination of
side-scan and bathymetry enables the same area. The combination of side-scan and
bathymetry enables the correlation/comparison of variable backscatter strength with
bathymetric features and the identification of the bathymetric and geologic environ-
ment of a specific acoustic feature.
For bathymetric enhancement of side-scan sonar imagery in real time, bathymet-
ric data must preexist or be collected concurrently with it.
Recently, two types of side-scan sonar configurations have enabled the computa-
tion of bathymetry across the sonar configurations have enabled the computation of
bathymetry across the sonar swath - the fringe mode (interferometric telesounding)
and the phase mode (split-beam phase comparison) sonars.
More recently, new multibeam bathymetry systems have been developed with the
goal of extracting backscatter information (in essence, side-scan) from each of the
beams. Isis, with its flexible framework and displays, expendability, and side-scan
processing features, will be the ideal topside system for these sonars.
Multichannel seismic systems with relatively low channel counts may also bene-
fit from the use of the portable Isis system. Isis software processes and stores any
number of data channels. For narrow-bandwidth seismic data, the Isis AID converter
can acquire eight channels at 16 bits and at an aggregate sample rate of 170 kHz. With
three additional DSP boards, the Isis total seismic acquisition capability is 32 channels.
Tape (8 mm) or optical data storage in standard SEG-Y formats will allow for data
transfer to any postprocessing system.
Isis - Versatile Sonar Data Acquisition 393

Future improvements to the hardware performance may include host CUP boards
based on the Intel Pentium®, higher-capacity mass storage peripherals, higher-resolu-
tion video displays, and higher sampling rates and channel counts per plug-in A/D-DSP
board. Enhancements such as these come relatively easily because of the flexibility and
standardization of the basic system architecture. As with Q-MIPS, the Isis system will
follow the developing technology as the reliability of that technology allows.
Fred Newton and Ron Schaaf of Triton provided system design and software
development leadership, respectively, for both the Q-MIPS and Isis products and car-
ried out the XXX demonstration survey in the Netherlands. Rijkswaterstaat provided
the test platform and the chance to demonstrate Isis's real-time capabilities.
 Part V.D

'.~~.' Mapping and Recording

~

o i' .
q;,O_" .': .
.j'~l

In Chapter 36, Farb's presentation of his USS Monitor documentation resulted from
scuba dives on the site rather than lockout diving from submersibles. The presentation
of image processing, equipment, and the results is impressive and suggests that regular
inspection dives are not only possible but should be continued.
In Chapter 37, Waddell outlines the success of Parks Canada's Marine Archaeol-
ogy Unit with the SHARPS system for electronic mapping. Parameters such as total
diving hours and accuracy were greatly improved with the equipment. Several sites fea-
turing scattered sections and high three-dimensionality were successfully recorded.
However, the system has high set-up and calibration requirements. SHARPS has
important ramifications for future zero visibility work.

395
 Chapter 36

'.~~.' Computer Video Image

~

o i' .
Digitization on the
q;,O_" .': .
.j'~l USS Monitor
A Research Tool for
Underwater Archaeology

RODERICK M. FARB

INTRODUCTION

Since the 1973 discovery of the remains of the USS Monitor in 235 ft of water off Cape
Hatteras, North Carolina, and its establishment in 1975 as the nation's first National
Marine Sanctuary, the site has been managed by the National Oceanic and Atmospheric
Administration (NOAA). Under a federal management plan, which has as one of its
overall goals the promotion and coordination of scientific research to expand our knowl-
edge of this historically important cultural resource, the ironclad has been the subject of
several research expeditions involving deep ocean technology (NOAA, 1983; Arnold et
al., 1991). Remotely operated cameras, side-scan sonar, deep sea submersibles, submer-
sibles utilizing lock-out divers, remotely operated vehicles, and other equipment have
descended to the ocean bottom to make a record of the site (Arnold et al., 1991). These
operations, although providing valuable information about the site, had limited access to
the shipwreck (mostly its north side) because of prevailing currents and other weather
Source: Underwater Archaeology Proceedings from the Co'!ference on Underwater and Historical Archaeol-
ogy. edited by Donald H. Keith and Toni L. Carrell (1992), pp. 100-105. Reprinted by pennission of the Society
for Historical Archaeology, Tucson, Arizona.

397
398 R.M. Farb

ISI'EfT

Figure I. Artist's conception of the USS Monitor site. By Rich Volz for the Farb Monitor Expedition.

conditions. In 1977 and 1979, during expeditions supported by the Harbor Branch Foun-
dation, several scientists visited the site as lock-out divers tethered to a deep sea sub-
mersible (NOAA, 1983; Watts, 1979). With the discontinuation of lock-out dives
following the 1979 expedition (NOAA, 1983), researchers have been allowed to work on
the site only remotely via submersible, ROV, or data collection devices placed on the
site. Researchers have not been able to do "hands-on" research on the site in over a dec-
ade. NOAA did not permit researchers to dive to the site using scuba equipment because
of the extreme depth and for safety considerations. However, by late 1980s, advances in
scuba equipment, dive computers, underwater lighting, and camera technology made
diving and filming at depths equivalent to that of the Monitor commonplace among tech-
nical scuba divers. In 1990, after several years of consideration, NOAA allowed scuba
divers to do limited research on the Monitor site for the first time.
In June 1990, the Farb Monitor Expedition became the first group to dive and
document the shipwreck using scuba equipment. Following the historic first dive, 17
divers made 105 dives to the shipwreck during four cruises throughout the summer,
photographing and recording areas of the site that had heretofore been inaccessible to
all but free swimming divers. Computer digitization of video analog images has been in
general use since the mid-1980s (Scheingold, 1986; Tompkins and Webster, 1988).
Computer video image digitization (CVID) was used for the first time on an underwa-
Computer Video Image Digitization on the USS Monitor 399
ter archaeological site to increase our knowledge about the shipwreck and to document
its condition.

COMPUTER VIDEO IMAGE DlGITIZATION

Technological advances in microcomputers, video cameras, video image digitiza-

tion, and data analysis software make it possible to acquire accurate measurements
using high performance personal computers and video camcorders (Molinari and Pre-
ston, 1989). The first application of CVID to shipwreck archaeology was utilized to
document the deep ocean shipwreck, USS Monitor, during the first survey of the site by
scuba divers. More recently, a similar approach was used to document the schooner
Alva Bradley in Lake Michigan's water (Stolman, 1991:11-12).
The reduced work time on the site because of its excessive depth (235 ft), good
visibility because of the presence of clear Gulf Stream water, and the low cost of video
imaging make it the preferred means of acquiring data. Film images from 16-mm film
cameras, though having' at least four times the resolution of video images, are more
expensive to produce; film images may be dubbed to video for CVID.
The recording rate of video camcorders is 30 images per second; film cameras
record at 24 frames per second. The potential for data acquisition per unit time is enor-
mous with either of these two media compared to still photography. Producing images,
although necessary for CVID, is not sufficient for useful analysis unless it is coupled
with a computer equipped witJI high-resolution digitization hardware and powerful
software designed to enhance and quantify elements within each digitized image frame.
Before a computer can manipulate and quantify an image, it must acquire image
data from a video source via a digitization board or frame grabber that converts the
analog video image to a digital computer format. The frame grabber's analog-to-digital
converter has both spatial and gray scale resolution characteristics. The picture on a
video or computer screen is made up of tiny elements called pixels. The resolution of
an image is expressed in terms of the number of pixels that fill the image in a horizon-
tal and vertical direction. Furthermore, each pixel within an image can possess a range
of intensities, or gray levels, which is a measure of the reflectance of light coming from
the part of the subject being represented by the pixel. Thus, the greater number of pix-
els comprising a video or computer image and the greater the gray scale range of each
pixel, the greater the resolution of the image and the more "true" the image will repre-
sent the subject. High resolution frame grabbers possess at least 512 x 512 x 8 spatial
and gray scale resolution, a 512 x 512 pixel display that has a 256 gray-level (so-called
8-bit) resolution for each pixel.
In order to optimize digitization, the manner in which original images are
recorded (the spatial relationship of the camera to the subject, the use of lighting to pro-
vide proper image contrast, and the presence of stadia within each image for calibration
purposes) is critical for good image analysis. Successful CVID depends on the resolu-
tion of the original image, the resolution of the computer digitization board used to
capture the image, and the sophistication of software being used for quantification of
elements within the image. Once the video image has been digitized by the computer's
frame grabber, computer software designed for the purpose can modify the image. Con-
trast and brightness may be enhanced, and elements within the image, length, area, or
angle may be determined.
400 R.M. Farb

THE RESEARCH
During the 1990 USS Monitor research, free-swimming scuba divers made two
20-minute dives per day breathing air, using a 3-hour surface interval between dives. A
total of 105 dives were made during a complete CVID survey of the site. Dive times
and decompression schedules were determined by a U.S. Divers Monitor 2 dive com-
puter worn by each diver. Decompression was done using air up to a depth of 20 ft;
pure oxygen was used from 20 ft to the surface. Decompression times ranged from 90
to 120 minutes.
Sony 8-mm High Band video cameras (450 lines of horizontal resolution) and
Airiflex 16-mm film cameras (1200 lines of resoiution) were used with 600 watts of
light to record images for CVID. During each of the dives to the shipwreck, 36,000
images for CVID analysis were generated by free-swimming divers using 16-mm film.
Film images were dubbed to I-inch videotape. Video images were digitized using a
Data Translation 2953 digitization board (512 x 512 x 8 resolution) installed in an IBM
Model 80 80386 20-MHz computer. Still images were recorded using Nikon cameras
and Ikelite strobes. Still film images were digitized using a Hewlett Packard Scan Jet
Plus flat-bed scanner. Measurements from digitized images were made using BioScan's
Optimas data analysis software. The digitized images were output to a 300-dpi laser
printer for halftone images and to an Agfa Matrix film recorder for 4000-lpi 35-mm
slides. Compressed images were stored on disk.

RESULTS
There have been a number of changes in the structure of the site since 1979
(Watts, 1979; Arnold et ai., 1991). The entire port side of the stern displacement hull,
and all but one of its frame members, have collapsed, creating a "valley" between the
engineering spaces and the port armor belt.
The lower hull on the port side of the fire room has settled significantly as indi-
cated by three severely bent frame members. The lower hull surrounding the boilers
and galley is leaning approximately 15° to starboard. One consequence of the shift is
the displacement of three or four bottom hull plates from the port side of the galley
(forward corner of the engineering space). The shift probably caused the fragile rivets
holding the plates to move freely in the presence of strong currents. One displaced hull
plate is lying on top of the adjacent starboard plate.
The starboard armor belt, which was exposed at the bow and stern in 1979
(Watts, 1979:14, 103), is covered with sand and is not visible. The port armor belt has a
large gap on the inboard side of the belt at the stern end, present in 1979 (Watts,
1979: 16), but is no longer visible; the armor belt appears solid across its width for the
last 20 ft to the stern end. A large vertical crack in the armor belt begins at the top of
the belt (edge closest to the sand) a few feet from its stern end and runs vertically for a
distance of approximately 18 inches. Th~ port armor belt does not touch the sand any-
where; there is an area of scour caused by currents sweeping sand away from the belt.
There are at least three bulges (separations) on the bottom of the armor belt that
may be some of the predicted armor belt hinge points referred to in Bruce Muga's
(1982:57) engineering report. One bulge is located above and slightly forward of the
turret; a minor bulge is located across from the midpoint of the engineering spaces; and
Computer Video Image Digitization on the USS Monitor 401

the largest bulge, quite prominent when viewed from the south side of the armor belt, is
located across the valley from and slightly forward of the amidships bulkhead. The
bulge above the turret and the larger bulge were present in 1977 (NOAA videotapes
provided to author). Bulging may result from the deterioration of the wood portion of
the armor belt under the iron plate allowing the belt to settle downward, producing
buckling of the iron plate covering the bottom of the belt. The extent of buckling may
indicate the extent of deterioration of the armor belt's wood substructure.
The inboard side of the bottom of the armor belt has a large crack running longi-
tudinally for much of its length. This crack may represent an earlier separation of the
lower displacement hull from the bottom of the armor belt.
The large opening inboard of the port armor belt and near the port forward comer
of the galley described by Watts (1979:100) is still present. The opening provides
access to the wreck beneath the armor belt and aft to the turret. Deck plates are still
attached from the opening aft to the turret inboard of the armor belt. On the deck, 4 ft
west of the turret, there is a large opening that exposes some pipes of the engine room.
Adjacent to the southwest side of the turret, deck plates are separating and hanging, the
result of the deck settling where it rests on the turret.
The stem end of the engineering space where the propeller, propeller shaft, and
propeller bracket are located appears to be unchanged from 1979 (Watts, 1979: 103,
115) except for more calcareous growth on the structures. The starboard armor belt is
covered with sand and is no longer visible. There is a large opening under and star-
board of the propeller bracket where a flange and some pipes from the engine room are
exposed. There is also a piece of dark leather or rubber fabric in the wreckage near the
base of the opening.
The boilers, engine, galley area, and machinery of the engineering spaces appear
relatively unchanged since 1979 (Watts, 1979:103). The increased calcareous growth
on all of the structures has made many of the features of these areas unrecognizable
compared to 1979 (Watts, 1979). The fire room has the least support inside the engi-
neering spaces (Peterkin, 1985). The hull surrounding it would be expected to collapse
before other areas of the engineering spaces if lack of internal support contributed to
the collapse of the areas forward of the amidships bulkhead.
Frame members along the starboard side of the engineering spaces remain undis-
torted. There are a few frame members missing, creating gaps in their regular progres-
sion along the starboard side. The anchor well on the bow has lost much of its
symmetry but has not become filled with sand.

CONCLUSIONS

Qualified, properly equipped, and supervised scuba divers should be permitted to

evaluate the USS Monitor site annually for changes in its structure using CVID. Free-
swimming divers have access to every nook and cranny of the Monitor site, unlike sub-
mersibles which are limited to the down current (north or turret) side of the site
because of the presence of strong currents and other environmental factors. Submersi-
bles, operating up current of the site, run the risk of being swept into the site, and even
the smallest of them are too large to enter any standing structure on the site. Although
new video and film technology makes it possible to acquire images with two- to eight-
fold resolution compared to a decade ago, historical video and film images may be dig-
402 R.M. Farb

itized and compared to contemporary images for some types of useful analyses. One
such use was the evaluation of the bulges in the area of the armor belt, previously dis-
cussed. A year-to-year comparison of digitized images acquired by free-swimming
divers can provide a rapid, accurate, low-cost, and reliable means of obtaining valuable
information about structural changes in all areas of the Monitor site. Special attention
should be given to acquiring information about "indicator areas" on the site, fragile
portions of the wreck that we believe will undergo changes in structure more rapidly
than other areas. Indicator areas may provide information about site deterioration proc-
esses and may give advance warning of major structural changes in the site. The fol-
lowing are recommended as indicator areas: (1) the bulges in the bottom of the armor
belt; (2) the deck and openings next to the turret on west side; (3) the frame members
along the port side of the engineering spaces, especially the three severely distorted
ones at the fire room; (4) the single frame member that straddles the valley between the
port armor belt and the amidships bulkhead; (5) inside the engineering spaces, espe-
cially the fire room; (6) propeller bracket and the stem of the engineering space; and
(7) the amidships bulkhead.
Information gathered about indicator areas by CVID will be important in making
future management decisions about the shipwreck and its artifacts. NOAA has permitted
the Farb Expedition to continue CVID research work on the USS Monitor site in 1991
and 1992. This work demonstrates the utility ofCVID as a research tool in archaeology.

ACKNOWLEDGMENTS

I wish to acknowledge and thank the following sponsors who have graciously
provided excellent, reliable equipment and support for this work: U.S. Divers, IBM,
Nikon, Eastman Kodak, Hewlett Packard, BioScan, Data Translation, Zodiac Micro-
grafx, Xerox Imaging, Chronosport, Orca, Helix, Amphibico, Ikelite, Evinrude, and
Jack McKenney Productions.

REFERENCES
Arnold, J.B., III, Eshman, G.M., Hill, D.B., Peterson, C.E., Steward, W.K., Gegg, S., Watts, G.P., Jr., and Wel-
don, C., 1991, The 1987 Expedition to the Monitor National Marine Sanctuary: Data Analysis and
Final Report. Prepared for the Sanctuary and Resources Division, National Oceanic and Atmospheric
Administration.
Molinari, F., and Preston, C., 1989, Frame Grabber Boards Acquire and Process Images in Real Time. Source
Handbook. 8-3. Data Translation, Marlboro.
Muga, B.1., 1982, Engineering Investigation USS Monitor. North Carolina Department of Cultural Resources,
Raleigh, and National Oceanic and Atmospheric Administration, 1983, USS Monitor National Marine
Sanctuary. National Oceanic and Atmospheric Administration, Washington, DC.
Peterkin, E.W., 1985, Drawings of the USS Monitor: A catalog and technical analysis. USS Monitor. Historical
Report Series, I (I). North Carolina Department of Cultural Resources, Raleigh, and National Oceanic
and Atmospheric Administration, Washington, DC.
Scheingold, D.H. (ed.), 1986, Analog Digital Conversion Handbook. Prentice-Hall, Englewood, Cliffs, NJ.
Stoltman, T., 1991, Video-Mosaic Imaging of the Alva Bradley. In Stem to Stern: Newsletter of the Program in
Maritime History and Underwater Research 7(9): 11-12. East Carolina University, Greenville.
Tompkins, W., and Webster, J.G., 1988, Intet:facing Sensors to the IBM-PC. Prentice-Hall, Investigating the
Remains of the USS Monitor. A Final Report on 1979 Site Testing in the Monitor National Marine
Sanctuary. North Carolina Department of Cultural Resources, Raleigh.
 Chapter 37

Electronic Mapping of
Underwater Sites

PETERJ.A.VVADDELL

INTRODUCTION

Electronic mapping of underwater sites has been undertaken with limited success from
at least the mid-1970s. Some of the earlier work involved simple submersible acousti-
cal guns; however, published results on that work have been difficult to find. Recently,
electronic mapping has generated more interest with the use of SHARPS (Sonic High
Accuracy Ranging and Positioning System) on many underwater sites (Shomette,
1988:3; Hamilton, 1988:6; Watts, 1989:156; Marc, 1989:43). Unfortunately, few final
site maps have been published although much usable data have been collected. In the
fall of 1987 the Marine Archaeology Unit of the Canadian Parks Service (CPS) began
experimenting with SHARPS for three-dimensional archaeological mapping. Since that
time CPS has employed the system on three separate wreck sites of increasing com-
plexity. Supplementary records were obtained using conventional techniques. From
these experiences we can suggest the supplementary recording and general methods
required to produce usable archaeological maps. This should aid in determining if elec-
tronic mapping is a suitable tool on specific projects.

SYSTEMS OPERATION AND METHODOLOGY

SHARPS is an underwater ranging and positioning system. Its primary commer-

cial and military application is tracking or piloting underwater vehicles. It allows
Source: Underwater Archaeology Proceedings from the Society for Historical Archaeology Conference. edited
by Toni L. Carrell (1990), pp. 57-62. Reprinted by pennission of the Society for Historical Archaeology, Tuc-
son, Arizona.

403
404 P.J.A. Waddell

movement in and around an object without physical contact with the object and it
records the area covered, particularly important in low- or zero-visibility environments.
A detailed explanation of the SHARPS system, its specifications, and its operation is
available from the manufacturer. Many configurations are possible, but for most ar-
chaeological purposes a simple setup is employed. Three transceivers are positioned
around a site, usually elevated or suspended from support stands or lengths of alumi-
num conduit. A fourth transceiver acts as a "gun" and, when activated or fired, the dif-
ference in arrival time of sound signals is measured at the three stationary transceivers
or "net." The differences in time are directly related to the distances between the gun
and the net, and hence an x, y, and z coordinate is established for each recorded point.
The transceivers are hard-wired to the surface where the system is controlled and data
collected on a personal computer The recorded points, when joined, give an outline of
the objects or feature being mapped.
The mapping diver recorded site features by taking a "shot" approximately every
30 cm. At comers, breaks, terminations, joints, or other significant points, a series of
shots were taken to ensure the point was as accurate as possible. During initial calibra-
tion of the system and throughout the recording process, effective diver-to-surface and
surface-to-diver communication is essential. Our first recording attempts without such
communication were both inefficient and frustrating because divers had to surface fre-
quently to discuss the work. If the diver can communicate while submerged, all
recorded data can be separated into meaningful files by informing the surface computer
operator who in tum makes notations as work progresses. The diver can also dictate
pertinent field notes for later review in conjunction with the computer data. The surface
team can monitor recordings on the computer screen and inform the diver if any data
seem questionable, if the system fails, or provide other advice and direction. The
greater the recording depth, the more critical is communication. In our experience two-
way voice communication is essential on sites deeper than 50 ft where any significant
amount of recording is to be done.
On relatively deep sites it was sometimes more efficient to put down a tethered
single diver because of the bottom-time premium. The only drawback to this was the
potential for the gun cable to become entangled with the structure or bottom features.
On the sites worked to date, a second diver swimming above the recording diver acted
as a cable tender. The increased recording speed offset the fact that two divers were do-
ing the same job. In some cases the diver tending the line could work continuously
above the recording diver and therefore was in a better decompression status. This pro-
cedure should be assessed site by site in terms of maximizing bottom-time recording
efficiency.
It was further evident that the number of divers involved in the actual recording
was best kept to a minimum and those doing the underwater recording should be the
prime data editors on the computer. Editing was done as soon after the dive as practical
so the data would be fresh in the recording diver's mind. Raw data were displayed by
file and the recording diver would edit in consultation with the dictated field notes.
Data collection was usually done by area and structure, e.g., section A starboard aft fut-
tocks, and it was sometimes advantageous to view and edit two or more data files at
once when continuous structure overlapped into the next file or when overlapping files
presented a more comprehensible image for editing. Edited SHARPS files were then
converted for use in a CAD system that allowed lines to be added, deleted, or moved. It
further allowed hard copies of the drawings to be printed.
Electronic Mapping of Underwater Sites 405

FIELD SITES
Quarry Wreck

CPS was first introduced to SHARPS in November 1987 when 3 days were
devoted to recording a river logging boat. This small modern vessel was intentionally
scuttled by sport divers as a recreational resource. It was ideal for testing, being in a
local quarry in 45 ft of relatively clear water and very close to shore. More importantly,
it was intact and had a complete superstructure and clear, regular lines. Total underwa-
ter time was 9 hours 40 minutes, with 6 hours 20 minutes actually devoted to data col-
lection. The system recorded quickly. It was possible to cover the forward gunwale of
the 8-m wreck and much of the forward superstructure with one setup. A second net
was needed to do the after portion of the hull (Figure I).
The basic above-gunwale lines of the wreck are quite simple and were recorded by
hand in about 90 minutes (recording diver's time only). When comparing by-hand results
to those obtained with SHARPS, it was evident that SHARPS was not superior in terms
oftime or accuracy. The computer-generated plan was reasonable (Figure 1), but would
obviously have to be corrected by hand to produce an accurate final plan. Without using
hand measurements, the plan lines of the hull and superstructure could only be produced
through a "best fit" trial. Although this was not done, we felt that it would not prove as
accurate as hand mapping. There was also a problem in using the computer to align
SHARPS data from files recorded at the bow with ones recorded at the stern although a
reasonable compromise was completed by eye. An acceptable, strictly computer-gener-
ated final plan was not possible with the data collected. Nonetheless, the results had dem-
onstrated very distinct potential. The researchers were particularly impressed with the
accuracy of individual SHARPS measurements. Typically, overall length of the vessel
(8.17 m) by SHARPS data was within 2 cm of the figure recorded by hand. This accu-
racy was borne out at several points on the hull. Further SHARPS practice and experi-
mentation both underwater and with the software was merited.

•
•

• •

f - - - - - - - - 8.17 - - - - - - - - - - 1

Figure I. Quarry Wreck, computer-generated plan showing two transceiver nets.

406 P.J.A. Waddell

Charles P. Minch
The next site selected for experimentation with electronic mapping was that of the
Charles P. Minch, a three-masted schooner that sank in 1898 at Cove Island in Georgian
Bay. The wreck site is one of over 20 sites within Fathom Five National Marine Park. It
was selected because of its marked three-dimensionality, relatively shallow depth, rea-
sonable visibility, extensive breakup, and site scatter. The main site consists of a rudder
and seven major pieces of wreckage distributed along a descending rocky shoreline and
silt bottom in depths ranging from 15 to 45 ft. The two main mapping objectives for the
SHARPS on the Charles P. Minch were: (1) to produce a satisfactory computer-gener-
ated drawing of one section of the site and (2) to aid in producing an overall site map by
locating all hand-drawn sections in their three-dimensional coordinates.
For the first objective, a section of railing (section G) in about 20 ft of water was
selected. The idea was to do no hand mapping of the section. A few benchmark meas-
urements were taken to check SHARPS accuracy, however, the primary tool for evalu-
ating the effectiveness of the computer drawing was photomosaics. Section G is
approximately 18 m by 5 m and has an elevational range of about 1.5 m. All recording
was done from one net. Eight files were created and 4671 points were recorded. After
each of the files was edited, they were superimposed on the computer to give a com-
posite point drawing showing major site features (Figure 2). Connecting these points
and then deleting extraneous lines resulted in a "reasonable" computer-generated draw-

Figure 2. Charles P. Minch, section

G. Composite file plan of points after
editing.

Figure 3. Charles P. Minch, sec-

tion G. Computer-edited and
generated line plan.

Figure 4. Charles P. Minch, section G fi-

nal plan drawing.
Electronic Mapping of Underwater Sites 407

ing (Figure 3) that clearly shows all major site features, but lacks some detail, particu-
larly in the diagonal ceiling planking. Other details were readily available through the
plan photomosaic. The computer drawing was then traced by hand and modified
according to the photomosaic to produce the final section G plan (Figure 4). The hand
work was essentially tracing the best fit of the computer lines, correcting the problem
ceiling planking area, and eliminating spurious lines created by the software.
Researchers with more extensive CAD facilities and experience could well have
carried the computer-generated drawing further. In any event, a point is usually reached
where it is quicker and simpler to finalize the drawing by hand. Follow-up measure-
ment error for any given timber did not exceed 2% and was normally much less. Table
1 shows the time taken to produce the drawing for section G.
CPS experience with conventional mapping of shallow sites has generally been
that most of the time spent producing a site plan is put in underwater. SHARPS record-
ing of section G provided just the opposite experience, which would have obvious
advantages on deeper, time-limited sites. The underwater recording time was certainly
less than had we used conventional techniques. That plus its reasonable accuracy left us
with quite a favorable impression of SHARPS as a structural recording tool. For the
first time within our recording experience we produced a final plan using an electronic
system as the primary recording means.
The second SHARPS objective on the Charles P. Minch was to locate all hand-
drawn sections three-dimensionally in order to aid in producing the overall site plan
and illustration. To cover this area of approximately 3500 m2 (94 x 37 m), two nets
were required (Figure 5). A single net did not have direct line of sight to all major
wreckage points, causing several areas to be "acoustically hidden." Fortunately the sec-
ond net was the same one used for the detailed mapping of section G.
The two separate net drawings were "locked in" to one another through the best
fit of common features. The accuracy of this drawing was checked directly with tape
measurements. With only a few exceptions, the difference in measurements did not
exceed 2% and in most instances was much less. Ideally SHARPS should be employed
with an equilateral net surrounding all features shown in order to give most accurate
results. As indicated in Figure 5, the setup employed for the site plan far exceeded this
requirement and may account for some of the error.

Table I. TIme taken to produce the drawing for section G

Underwater time
Transceiver setup 2.0
SHARPS recording 9.2
Photomosaic 6.2
Total diving hours 17.4

Surface time
Data editing 12.0
Computer drawing 8.0
Photomosaic assembly 33.0
Hand tracing and modifications 6.0
Total surface hours 59.0
408 P.J.A. Waddell

").
\' 0j
"'.
J~ •

.)
r
1! .'~.lJX.AIG.a
I
l:

Figure 5. Charles P Minch site plan showing transceiver nets.

Electronic Mapping of Underwater Sites 409

Figure 6. Charles P. Minch final perspective illustration.

Two distinct advantages were noted in using SHARPS to produce the site plan:
First, each of the six sections that were hand drawn were drawn parallel to their axis,
i.e., not in plan. Drawing these sections in plan would have been far more complex and
time consuming as some were on slopes exceeding 35°. Although drawn parallel to
their axis, the standard SHARPS setup permitted all sections to be viewed and posi-
tioned in plan. Drawing parallel to the site axis further allows the mapping diver to bet-
ter understand what is being drawn because he does not have to be concerned with the
complications associated with making a plan drawing of a sloped site. Further, from the
point of view of reconstruction and understanding the hull, parallel drawing is far more
useful. Second, distances and elevation changes between sections are obtained as
quickly as the points can be taken. Because the signal is acoustic, it is not affected by
visibility limits or currents. In this case recording was done at distances up to 70 m on
one net and 40 m on the other. Obtaining these same positions through conventional
tape trilateration would have been time consuming and very likely less accurate.
From the SHARPS data and the hand maps a perspective illustration of the site
was developed (Figure 6). SHARPS data from the topography surrounding the wreck
site are demonstrably invaluable in illustrating the wreck in situ.

Philo Scoville
Encouraged by the above results, CPS undertook SHARPS site mapping of a sec-
ond Fathom Five shipwreck, the Philo Scoville, in June 1989. This 1889 wreck is quite
410 P.J.A. Waddell

broken up, lying on a slope with most of the wreckage between 55 and 100 ft. Thirty-
one underwater hours were devoted to SHARPS recording on the site and 10.5 hours of
photo mosaic recording. This was our first use of SHARPS as the primary recording
tool in a site plan production. Plans and illustrations are currently being prepared, and
analysis of this deeper-water SHARPS project will be made on completion of the sur-
face work. However, it was quite obvious to the recording crew that the progressive use
of SHARPS from a shallow uncomplicated site to a deeper, scattered site is essential in
developing a reasonable degree of recording efficiency.

CONCLUSION

Electronic mapping, and in particular SHARPS, has shown significant potential

in its use as an underwater mapping tool. CPS's brief experience has demonstrated sev-
eral specific strengths associated with such mapping: (I) facility in pulling together dis-
crete, scattered sections into a coherent site plan with reasonable accuracy; (2) efficient
and rapid recording of highly three-dimensional points; (3) reduction of underwater
recording time on specific sites, although, on-surface time for editing and data integra-
tion may increase; and (4) hand-drawn maps may be executed underwater in their most
desirable orientation and then with a few SHARPS points rotated into a site plan or
viewed from a selected perspective. This may equally apply to photomosaics.
CPS is currently working with SHARPS data from a deeper, scattered site that is
three-dimensional and has elevation changes of approximately 15 m. Preliminary indi-
cations are that SHARPS may offer the greatest advantages in these conditions.
SHARPS may not provide any marked recording advantage on sites that could only be
dived once or twice for limited periods. The logistics of setup and system calibration
would normally preclude its use in such situations.
SHARPS has proven itself to be a useful tool that is well suited to supplement
hand-drawn mapping, photomosaic and other underwater recording techniques.
SHARPS has also been used as the primary recording tool with distinct success. Its use
as an exclusive recording tool is doubtful and, in any event, not particularly desirable.
CPS plans to continue experimenting with electronic mapping. Sharing results with
those of other users will greatly assist us all in determining the role these tools can play
in underwater heritage recording.

REFERENCES

Hamilton. D., 1988, I.N.A. Enters the SHARPS Era. I.N.A. Newsletter 15(2), College Station.
Marc, J., 1989, Exploring the Lord Western. The Underwater Archaeological Society of British Columbia. Van-
couver.
Shomette, D.G., 1988, The New Jersey Project: Robots and Ultrasonics. Underwater Archaeological Survey.
In Underwater Archaeology Proceedings from the Society for Historic Archaeology Conference, Reno,
edited by J.P. Delgado. Society for Historical Archaeology, Ann Arbor.
Watts, G.P. 1989, The Sinkentine: A Fiberglass Shipwreck Model to Assist in Teaching Three-Dimensional
Mapping. International Journal of Nautical Archaeology 18(2): 151-156.
 Part V.E

Remote Operating
".~i).:
~

Vehicle (ROV)
':.i
-r/: ..

United States government and individual states mandate reporting on cultural resources
in areas which are subject to development. The requirement that an archaeological sur-
vey is necessary to obtain permits for construction in the maritime environment has
produced a huge number of reports. Unfortunately, these reports are not given wide-
spread distribution and some were limited to fewer than twenty copies. This lack of dis-
tribution makes generalizations about areal studies difficult because, while the data
have, in many cases, been collected, the information is not readily available. The best
sources for obtaining information about areal studies along rivers and in coastal areas
are the state archaeologists of the various states and the district offices of the US Army
Corps of Engineers. Included in this section are some reports as well as materials
related to survey such as geographical analysis and positioning systems.

Ballard, Robert D., 1990, The Discovery of the Bismarck. Werner/Madison Press, New York. (Chapter 3 pro-
vides general infonnation and illustrations on the Argo.)
Geyer, Richard A. (ed.), 1977, Submersibles and Their Use in Oceanography and Ocean Engineering. Elsevier
Scientific Publishing Company, Amsterdam, Holland.
Green, lN, 1987, Remote Sensing in Shipwreck Archaeology: Locating the Mahogany Ship. Proceedings of
the 2nd Australian Symposium on the Mahogany Ship. The Mahogany Ship Committee and Warram-
boo I Press. Victoria.
Green, J.N., M. McCarthy, and J. Penrose, 1984, Site Inspection by Remote Sensing: The HMAS Sydney
Search, a Case Study. Bulletin of the Australian Institute of Maritime Archaeology 8(1 ):2~2.
National Research Council, 1996, Undersea Vehicles and National Needs. National Academy Press, Washing-
ton, D.C. (Especially the bibliographies. They have little to do with archaeology but much to do with
the technology).
Rothberg, Robert H., 1993, Diverless Subsea Systems New Features on ROV's and Control Systems add Flexi-
bility and Cut Costs. Oil and Gas Journal (April 5th, 1993).

411
412 Remote Operating Vehicle (ROV)

Shomette, Donald G., 1988, The New Jersey Project: Robots and Ultrasonics in Underwater Archaeological
Survey. Proceedings o/the Underwater Sessions, edited by James P. Delgado, pp. 1-6, Society for His-
torical Archaeology, Tucson, Arizona.
Travis, John, 1993, Deep-Sea Debate Pits Alvin against Jason. Science 259 (12 March 1993).
 '.~~.' Additional Bibliography
~

o i' .
for Part V
q;,O_' .': .
.j'~l

Ackroyd, Neil, and Robert Lorimer, 1990, Global Navigation. A GPS User's Guide. Lloyd's of London Press,
London.
van Andel, T.H., and N. Lianos, 1983, Prehistoric and Historic Shorelines of the Southern Argolid Peninsula: a
Subbottom Profiler Study. International Journal of Nautical Archaeology 12 (4):303-324.
Arnold, J. Barto, 1976, An Underwater Archaeological Magnetometer Survey and Site Test Excavation Project
Off Padre Island. Texas. Texas Antiquities Commission, Austin, Texas.
Caverly, Robert D., 1988, The Application of SHARPS AND CAD Technology on the Yorktown Shipwreck
Project. Underwater Proceedings of the Society for Historical Archaeology Conference, edited by
James P. Delgado, pp. 18-19 Society for Historical Archaeology, Tucson, AZ.
Clancy, Tom, 1994, Armored Cav. Berkeley Books, New York. (pp. 173-181 deal with GPS in easily understood
terms).
Cole, F.W., 1968, A Familiarization with Lateral or Side-Scanning Sonars, Technical Report No. 159; Hudson
Laboratories, Dobbs Ferry, New York.
Flemming, B.W., M. Klein and B.M. Denbigh, 1982, Recent Developments in Side-Scan Sonar Techniques,
Russel-Cargill, WG.A. ed.; Central Acoustics Laboratory, University of Capetown, Cape Town, South
Africa.
Franzen, Anders, 1981, HMS Kronan: The Search for a Great 17th Century Warship, Royal Institute of Tech-
nology, Stockholm. Hodder, Ian, and Clive Orton, 1976, Spatial Analysis in Archaeology. Cambridge
University Press, Cambridge.
Kyriakopoulou, Y., 1990, Underwater Surveying with the SHARPS in the 1989 Excavation Season at Dokos.
Enalia I: 24--26.
Shomette, Donald G., 1988, New Jersey Project: Robots and Ultrasonics in Underwater Archaeological Sur-
vey. Underwater Proceedings of the Society for Historical Archaeology Conference, edited by James P.
Delgado, Society for Historical Archaeology, Tucson, AZ.
Stone, Lawrence D., 1975, Theory of Optimal Search. Academic Press, New York.
Thomas, R.W, 1977, An Introduction to Quadrat Analysis. Geo Abstracts, Ltd. University of East Anglia, Nor-
wich, Great Britain.
Throckmorton, Peter, editor, 1987, The Sea Remembers: Shipwrecks and Archaeology. Smithmark, New York.
Urick, R.J., 1975, Principles of Underwater Sound. McGraw-Hill, New York.
Watts, Gordon P. Jr., 1975, The Location and Identification of the Ironclad U.S.S. Monitor. NauticalArchaeology
4:301-330.

413
 Part VI

Site Significance

One of the most common questions asked of archaeologists is, "How old is it?" Afol-
low-up question usually relates to site importance. Is the site important, and why, are
questions every archaeologist must answer. This is all the more so when conducting
surveys that must not only find and identify sites, but which must recommend addi-
tional work. In American archaeology, questions of significance have been addressed
on several levels, but the chief governing factor relates to a site's potential for inclusion
in the National Register of Historic Places.
Significance relates to several factors relating to both history and the site itself.
These include association with events and individuals already deemed important,
uniqueness, and integrity. Each factor, in tum, can be assessed on several levels of
importance: local, state, regional, and national. Taken as a whole, a site condition and
potential for providing information about the past, its relationship to other sites in the
same time period, and its uniqueness, serve to guide an assessment of significance.
Ways of evaluating significance in terms of legal requirements and archaeologi-
cal theory as well as establishing context are presented in this section. Obviously, the
more sites that are available for comparative purposes, the better an evaluation will be.
In maritime archaeology, there are often only two or three similar sites and documen-
tary sources do not provide many clues. Consequently, more accurate assessments of
significance must wait until a larger data base is formed.
At present, when a site represents a previously unknown or unreported vessel
type, it must be considered significant. However, this evaluation might change as more
sites are found and evaluated. Acquisition of new information is part of the scientific
process and changes must be expected. The introduction of broadbased systematic sur-
veys to locate all sites within particular regions will have pronounced impacts on ques-
tions of significance, but this type of research is only now starting. Guidance for these
surveys is suggested in the following presentations.

415
 Part VI.A

'.~~.' Significance
~

o i' .
q;,O_" .': .
.j'~l

In the United States, the question of significance revolves around eligibility for inclu-
sion in the National Register. The official word from the federal government on ship-
wrecks is in National Register Bulletin 20, which is listed in the bibliography for this
section. These regulations outline parameters for placing significant historical sites on
the National Register. In the federal permitting process, any site eligible for the Regis-
ter has the same status as if it were on the Register.
Potter's examination (Chapter 38) of shipwrecks falls into two categories: his-
toric preservation and critical theory. Under historic preservation, he draws pertinent
connections between shipwreck protection and the process for eligibility for the
National Register of Historic Places, noting that SHPO-style preservation is "slippery
stuff," and should be clearly understood. The second section, critical theory, deals with
Potter's expectations from underwater archaeology in New Hampshire.

417
 Chapter 38

An Envelope Full of
Questions That Count in
Underwater Archaeology

PARKER B. POTTER Jr.

INTRODUCTION

The obvious and necessary starting point for this paper is an explanation of how I am
qualified to write it. To begin, I am not an underwater archaeologist. Instead, I am the
principal historical archaeologist for the New Hampshire State Historic Preservation
Office, also known as the New Hampshire Division of Historical Resources (DHR). As
the DHR's historical archaeologist I have been assigned several tasks related to under-
water archaeology, and the experience I have gained from these activities is the basis
for much of what I have to say in this paper.
My work with underwater archaeological issues has four main parts. First, I have
managed my office's response to and implementation of the Abandoned Shipwreck Act
of 1987. In the area of response, I prepared testimony for the NPS public meetings held
as a part of the process of drafting shipwreck guidelines. In the area of implementation, I
worked with David Switzer of the Institute for New Hampshire Studies at Plymouth
State College to convene a panel of private citizens and public officials with interests in
underwater historical resources. Secondly, I have written and distributed a short pam-
phlet (Potter, 1988) that details the New Hampshire statutes and regulations that pertain

Source: Underwater Archaeology Proceedings from the Society for Historical Archaeology Conference. edited
by Toni L. Carrell (1990). pp. 34-38. Reprinted by permission of the Society for Historical Archaeology, Tuc-
son, Arizona.

419
420 P.B. Potter Jr.

to historical resources under state waters. Third, on several occasions I have worked with
Dr. Switzer to respond to the accidental discovery of underwater historical resources.
Finally, as the DHR's comprehensive planning coordinator, I have coauthored a planning
document called a "historic context" (Potter and Switzer, 1989), which is the blueprint
for shipwreck preservation in New Hampshire. These activities constitute my qualifica-
tions to write this paper.
As the title of this paper suggests, I want to stretch the envelope by stuffing it full
of questions that count in underwater archaeology. Furthermore, I intend to develop
these questions from two different perspectives. The first is my role as a historic preser-
vationist; the second is my status as an advocate for the application of critical theory to
historical archaeology (Leone et aI., 1987). In both of these areas, I operate under the
assumption that underwater archaeology is more than a set of recovery techniques, but
I will leave for others the debate over whether or not contemporary underwater archae-
ology has achieved an adequate level of theoretical sophistication.

UNDERWATER ARCHAEOLOGY IN HISTORIC PRESERVATION

Turning to my first mode of envelope stuffing. I would like to discuss the place
of underwater historical resources within historic preservation. To those on the outside,
State Historic Preservation Offices can easily appear to be awash in an alphabet soup of
standards, programs, and criteria, everything from RP3 to S&P, CLGs, and 36 CFR 61.
However, much of this acronymic overkill can be held at bay through the powerful
magic of one simple incantation: "planning, identification, evaluation, registration,
treatment." This mantra neatly encapsulates the idealized life history of any historical
resource as it passes through the various programs administered under federal mandate
by each of the 56 State Historic Preservation Offices (SHPO). From time to time the
National Park Service will come up with a new program, or a new direction for an old
program, all intended to facilitate the identification, evaluation, registration, and treat-
ment of historical resources. In the rest of this section I want to talk about two particu-
lar historic preservation tools, the National Register of Historic Places (NR) and
"historic contexts." I have two reasons for discussing these SHPO facts of life. First, I
want to describe for academic archaeologists two of the frameworks through which
preservationsts are obligated to consider the shipwreck sites. And second, I want to dis-
cuss the historic context as a way that SHPO preservationists have enlarged the concep-
tual envelope into which shipwreck sites may be placed.
For better or for worse, the National Register and its criteria for eligibility are at
the heart of most SHPO operations. The goal of the inventory process is the collection of
data that can be used to evaluate NR eligibility, and the treatment options for most his-
torical resources are dictated by their NR status. In the world of SHPO preservation, the
questions that count are questions couched in terms ofNR criteria. In other words, from
the standpoint of an SHPO operative, the only time a shipwreck is a historical resource
that counts is when it is discussed in terms ofNR eligibility. For the record NR eligibility
is composed of a fairly firm 50-year age requirement, an integrity requirement, and four
criteria, of which one must be met for a property to be Register eligible.
The quality of significance in American history, architecture, archeology, and
culture is present in districts, sites, buildings, structures, and objects that possess integ-
rity of location, design, setting, materials, workmanship, feeling, and association and:
An Envelope Full of Questions That Count in Underwater Archaeology 421

A. That are associated with events that have made a significant contribution to
the broad patterns of our history; or
B. That are associated with the lives of persons significant in our past; or
C. That embody the distinctive characteristics of a type, period, or method of con-
struction or that represent the work of a master, or that possess high artistic
values, or that represent a significant and distinguishable entity whose compo-
nents may lack individual distinction; or that have yielded, or may be likely to
yield, information important in prehistory or history (McClelland, 1986: I).
Finally, properties that are old enough, intact enough, and which meet one of the
four criteria, may be significant, and therefore eligible at the local, state, or national
level. These eligibility requirements matter because they compose the frame of refer-
ence that any SHPO operative must use to think about any shipwreck site within his or
her jurisdiction.
Delgado et al. cover all of these issues in National Register Bulletin 20, Nominat-
ing Historic Vessels and Shipwrecks to the National Register of Historic Places (Del-
gado et al., n.d.) and here I want to focus on just one, a shift from Criterion C to
Criterion A, which represents one way of stretching the envelope. Generally speaking,
archaeological sites get on the NR by means of Criterion D and a statement of signifi-
cance that extrapolates from what is already known to what could be learned from fur-
ther excavation. Architectural properties usually enter the Register via Criteria A, B, or
C. Shipwrecks are in some ways architectural and in other ways archaeological, thus
opening the possibility of shipwreck nominations under all four criteria. Criterion D is
the safest and easiest, requiring only that the significance of a particular shipwreck site
be argued in terms of current method and theory in underwater archaeology. If, how-
ever, one wants to be a bit more adventurous, the next safest haven is Criterion C,
which is often used as a "default criterion" in architectural nominations. To nominate a
building under Criterion C, one makes an argument based on architectural history, sug-
gesting, for example, that "the John Doe House is significant because it is the best
example of a Greek Revival Bungaloid in southern New Hampshire." The shipwreck
cognate is easy to see here, "the S.S. Jane Doe, lying at the bottom of Lake Umbagog,
is Register eligible because it is the best example of a paddle shell cargo boat in north-
ern New England." I would argue that shipwreck nominations, just like dry-land nomi-
nations, will be more meaningful if significance is argued under Criterion A rather than
under Criterion C. Any historic property can be used as an example in any number of
storylines or historical narratives. Under Criterion C, properties are used mainly as
props with which to tell a canned history of architecture of shipbuilding, and in many
cases, individual properties are largely independent from the stories they are used to
illustrate. However, under Criterion A, there is a much stronger tie between properties
and the stories they illustrate. The S.S. Jane Doe would be nominated under Criterion
A because "it is the only survivor of a fleet of paddle wheel cargo boats that played a
key role in transporting the area's most important cash crop, moose hides, from the
moose ranches north of Lake Umbagog to the hide markets south of the lake." Further-
more, under Criterion A, the S.S. Jane Doe will not automatically become less signifi-
cant on the discovery of a larger, older, or better preserved paddle wheel cargo boat.
The issues I am discussing here pertain not only to shipwrecks but to the whole
NR program. For years it has been standard operating procedure to nominate archae-
422 P.B. Potter Jr.

ological sites under Criterion D and buildings under Criterion C, but for the last several
years the Register has been actively encouraging more archaeological nominations
under Criteria A, B, and C and more architectural nominations under Criterion A.
One tool for accomplishing both of these shifts is a planning document known as
the "historic context." According to NPS guidance, a historic context is defined by a
time, a place, and a theme, as in "short staple cotton production in the White Mountains
of New Hampshire, 1620-1680." When fully developed, a historic context is a 10- to
200-page discussion of the historic properties that represent a given time, place, and
theme. (Each state is required to have a list of defined contexts and to be making pro-
gress toward their development.) One element of a fully developed historic context is a
section called "known and expected property types." In the case of the White Mountain
cotton context, known and expected property types would include the houses of cotton
farmers, the barns used to cure cotton, and the barges used to transport cotton from the
White Mountains down the Merrimack to mills in Manchester and Lowell. Once the
cotton context is fully developed, the preferred method for nominating sunken cotton
barges to the NR will be a statement of significance that demonstrates the importance
of the barges to 17th-century cotton farming, rather than a statement that shows where
Merrimack River cotton barges fit into the history of American shipbuilding. This
approach is a relatively new one in historic preservation, and it may be seen as a shift
from Criterion C to Criterion A, or as a shift from metaphor to metonymy. That is, the
trend in historic preservation is toward understanding and evaluating resources in terms
of the properties with which they were functionally related, rather than in terms of the
properties to which they were physically similar. At least within the arena of historic
preservation, the questions that count in underwater archaeology are going to be based
increasingly on historic contexts and decreasingly on boat topologies or shipbuilding
technologies. Seen in this light, cotton barges have more in common with cotton barns
than they have in common with coal barges. Whether or not this new approach is a
good thing, and I think it is, it is certainly one way of stretching the envelope.
This first section is more of a preservation lesson than I had intended, but SHPO-
style historic preservation is slippery stuff, and over the next several years offices like
mine will be playing a larger rather than a smaller role in shipwreck preservation, so
perhaps this excursion into Historic Preservation will be useful.

UNDERWATER ARCHAEOLOGY AND CRITICAL THEORY

However, I would now like to turn to a second way of stuffing our envelope full
of questions that count. This second way is based on critical theory-(Geuss, 1981; Held,
1980}, and there are several ways to introduce it. When applied to underwater archaeol-
ogy, historic preservation and critical theory have in common the linkage of underwa-
ter and dry-land archaeological sites. (For the record, historic preservation and critical
theory are not the only approaches that work this way; Richard Gould, who is neither a
historic preservationist nor a critical theorist, introduced me to this concept through
discussions of his research project in Bermuda.) Returning to the issue at hand, while
historic preservation and the critical approach I am about to describe both diminish the
conceptual boundary between wet sites and dry sites, there is an important distinction
between these two approaches. Even at its best, historic preservation is basically aca-
demic and so are the questions posed by preservation-driven research. The big battle
An Envelope Full of Questions That Count in Underwater Archaeology 423

between Criterion C and Criterion A is really a fight between architectural history and
anthropology (or history) over which discipline is a better source of research questions.
The difference between historic preservation and critical theory, at least as I operate in
these two intellectual environments, is that critical theory offers a way of moving
beyond the academy in the search for research questions.
I can make this point most clearly by referring to my recent work in Annapolis,
Maryland (Potter, 1989). When I worked with "Archaeology in Annapolis." I con-
ducted ethnographic and ethnohistorical research in order to isolate areas of contempo-
rary ideology with origins that could be explored archaeologically and interpreted for
the public. The result of this work was a series of archaeological site tours containing a
mild critique of modern capitalism that depended on establishing a conceptual link
between contemporary tourism in Annapolis and archaeological assemblages of 18th-
century ceramic tableware. I do not have the space to present that whole argument here,
but visitor evaluations suggest that we had modest success in using the archaeological
record to encourage some site visitors to think more deeply about aspects of contempo-
rary life that are usually ignored or taken for granted.
The task I want to take up in the remainder of this paper is the creation of a criti-
cal underwater archaeology for New Hampshire. The key move in this enterprise is to
study contemporary society in order to develop questions that count.
The most literal way of doing this would be to set up a research project designed to
test one or another of the myths and legends that seem to be spawned by shipwreck sites.
The critical turn I would give to such a research program is a careful consideration of the
contemporary social value of any myth under archaeological investigation. I can illus-
trate this point by taking another example from Gould's work. His piece on the archaeol-
ogy of war (Gould, 1983) is an archaeologically based comparison of the defeat of the
Spanish Armada in 1588 and the Battle of Britain in World War II. Gould sets these two
historical episodes side by side in order to derive general principles relating to the nature
of defensive warfare, and one element of his argument is an evaluation of three myths
that have grown up around the history of the defeat of the Armada. However, while
Gould is inclined to move from the archaeological testing of these myths to general prin-
ciples, I would be more inclined to focus attention on the myths themselves, asking who
tells these stories to whom, in what situations, and to what ends.
Reading history, though, is not the only way to frame historical questions. In
Annapolis and here in New Hampshire I favor a less literal approach which may, in the
end, lead to questions that are more strongly tied to modem life than questions based
on popular history.
Over the last 10 or IS years a number of historical archaeologists have argued
that historical archaeology is the archaeology of capitalism (Schuyler, 1970) or it is
nothing at all, to steal a phrase, and I think it is reasonable to include underwater
archaeology under this same caveat. Turning specifically to New Hampshire, there are
two different environments for underwater archaeology and, at one level, two different
underwater archaeological agendas, each a critique of capitalism and each leading to a
particular set of questions that count.
In the Lakes Region of New Hampshire there are the submerged remains of a vari-
ety of boats that have served the 19th- and 20th-century tourist trade. A key bit of
mythology here is the claim that New Hampshire's last colonial governor, John
Wentworth, was the region's first summer tourist when, conceptually speaking, he was a
424 P.B. Potter Jr.

visitor to the area but in no way a "tourist" - tourism not having been created as a cul-
tural category until the late 19th century (Leone et al., 1987:290). But rather than search-
ing for John Wentworth's barge in the lake that bears his name, I would try to use the
archaeological record of wrecked tourist boats to measure the nature and degree of tour-
ist industry capitalization - an issue as important today as it was in the 19th century.
In addition to its Lakes Region, New Hampshire has an 18-mile coastline and one
major port city, Portsmouth, which is located at the mouth of the Piscataqua River.
Because much of New England was settled and explored by English merchant interests,
there are a variety of questions about capitalism that could be answered through the
analysis of the underwater archaeological resources of the New Hampshire Seacoast.
The particular issue that interests me is one I first developed as a way of thinking
about a dry-land archaeological site, but I think it can also apply to many underwater
sites off the New Hampshire Seacoast. That issue is the relationship between New
Hampshire and Massachusetts. I first began thinking about this relationship archae-
ologically when I was asked to consider the archaeological potential of the "Bound
House" in Seabrook, New Hampshire. This site contains the cellar hole of a building
that stood from the 1630s until the 1880s, marking one of the several boundaries that
have separated New Hampshire and Massachusetts. I had trouble coming up with a
good reason to dig the Bound House site until one day when I went to visit the site and
had to drive through the shadow of the Seabrook Station Nuclear Power Plant to get
there. As I drove by Seabrook Station, I realized that by doing the archaeology of the
Bound House, I would be doing the archaeology of the Seabrook evacuation plan, one
of the hottest political issues in New Hampshire in the 1980s. As a part of its licensing
process, Seabrook Station has been required to file with the federal government an
emergency evacuation plan that covers a 10-mile radius around the plant. This evacu-
ation plan requires cooperation and participation from all of the local governments
within the evacuation area. This has been a point of contention because a part of the
evacuation area is in Massachusetts, which opposes the plant. A handful of Massachu-
setts towns within the 10-mile radius have not submitted their portions of the evacu-
ation plan, which has helped delay the start of full-power operations at Seabrook that
has, in turn, contributed to the bankruptcy of the Public Service Company of New
Hampshire, Seabrook's lead owner and the state's largest electric company.
The Bound House is a specific 17th-century boundary marker in the same way that
the 10-mile evacuation radius is a 20th-century marker, and the more important point is
that from the beginning of English settlement until today, the boundary between Massa-
chusetts and New Hampshire has been fluid, contested, and composed of many elements
above and beyond the cartographic fact of a stripe on a map. This boundary was pushed
north in the late 19th century by the invention of the refrigerated railroad car and the sub-
sequent expansion of the Boston milk market. Today this boundary ebbs and flows in reac-
tion to the Boston housing market; sometimes Concord is a suburb of Boston, sometimes it
is not. Given all of this, people in New Hampshire invest a great deal of energy in main-
taining the boundary, and vilifying Massachusetts, in what amounts to a struggle over iden-
tity. In many ways New Hampshire strives to be the absolute antithesis of Massachusetts,
and the results of this negative definition are most easily seen in the letters to the editor
section of Manchester's famous Union Leader. One letter writer went so far as to suggest
that the star representing Massachusetts on the American flag should be smaller and din-
gier than the stars for the other 49 states. Another writer suggested that Manuel Noriega
An Envelope Full of Questions That Count in Underwater Archaeology 425

would be a more appropriate governor for Massachusetts than Michael Dukakis. And, of
course, many people in New Hampshire refer to the state down south as "Taxachusetts."
This all pertains to underwater archaeology because many of the wrecks under
New Hampshire's coastal waters were involved in commercial activities that took them
beyond New Hampshire waters and tied them into the regional economy, if not the
world economy. Therefore, there is very likely a great deal of underwater archaeologi-
cal evidence that may be used to illuminate the economic relationships between New
Hampshire and Massachusetts. The piece of critical theory in all this is my feeling that
the history of this relationship needs to be explored because the relationship is ongoing,
unresolved, problematic, and ideologically charged. I need to do further ethnographic
and ethnohistorical research before I understand all of the nuances of this relationship,
but on the basis of my work in Annapolis, I am certain that archaeology, both dry and
wet, can playa role in helping contemporary New Englanders better understand the
world in which they live, and the often unexamined forces that act on them.
Finally, to put an O. Henry kind of twist on things, the major issue I would exam-
ine with underwater archaeology on the Seacoast is the same issue I would examine in
the Lakes Region. After all, the tourist industry that developed in the Lakes Region was
not for New Hampshire folks. It was designed to attract Bostonians and other out-of-
state flatlanders and was sometimes capitalized by these same interests. Thus, underwa-
ter archaeology in the Lakes Region can also serve as an important commentary on the
relationships between New Hampshire and Massachusetts.

CONCLUSIONS

In this paper I have sketched two approaches to stretching the envelope in under-
water archaeology, one stretch already in progress and the other a project for the future.
There are many other ways to push at these edges, including Schmidt and Mrozowski's
work on smuggling (Schmidt and Mrozowski, 1983, 1988) which retrofits rather neatly
under the middle range theory banner that Mark Leone is currently running up the his-
torical archaeology flagpole (Leone and Crosby, 1987; Leone, 1988; Leone and Potter,
1988). Finally, and in conclusion. when considering any kind of envelope stretching,
stuffing, or even shredding, I think we would be wise to follow Dick Gould's lead and
make sure that whatever we do, we do shipwreck anthropology.

ACKNOWLEDGMENTS

In writing this paper I have been assisted by a number of people, and in particular
I thank the following: Suzanne Spencer-Wood for inviting me to stretch the envelope,
Paul Shackel and Barbara Little for getting the paper read, Gary Hume and Nancy
Chabot for commenting on the first draft, and David Switzer for letting me take advan-
tage of his expertise.

REFERENCES
Delgado, J., et al., n.d., Nominating Historic Vessels and Shipwrecks to the National Register of Historic
Places. National Register Bulletin 20. Government Printing Office, Washington, DC.
Geuss, R., 1981, The Idea of a Critical Theory. Cambridge University Press, Cambridge.
426 P.B. Potter Jr.

Gould, R.A., 1983, The Archaeology of War: Wrecks of the Spanish Armada of 1588 and the Battle of Britain,
1940. In Shipwreck Anthropology, edited by R.A. Gould, pp. 105-142. University of New Mexico Press,
Albuquerque.
Held, D., 1980, Introduction to Critical Theory. University of California Press, Berkeley.
Leone, M.P., 1988, The Relationship Between Archaeological Data and the Documentary Record: 18th-Century
Gardens in Annapolis, Maryland. Historical Archaeology 22( I ):29-35.
Leone, M.P., and Crosby, C.A., 1987, Epilogue: Middle-Range Theory in Historical Archaeology. In Consumer
Choice in Historical Archaeology, edited by S. Spencer-Wood, pp. 397-410. Plenum Press, New York.
Leone, M.P., and Potter, P.B., Jr., 1988, Introduction: Issues in Historical Archaeology. In The Recovery of
Meaning: Historical Archaeology in the Eastern United States, edited by Smithsonian Institution Press,
Washington, DC.
Leone, M.P., Potter, P.B., Jr., and Shackel, P.A., 1987, Toward a Critical Archaeology. Current Anthropology
28(3): 383-302.
McClelland, L., 1986, Guidelines for Completing National Register of Historic Places Forms. National Register
Bulletin 16. Government Printing Office, Washington, DC.
Potter, P.B., Jr., 1988, Preserving Historic Shipwrecks in New Hampshire Waters. New Hampshire Division of
Historical Resources, Concord.
Potter, P.B., Jr., 1989, Archaeology in Public in Annapolis: An Experiment in the Application ofCritical Theory to
Historical Archaeology. Ph.D. thesis, Department of Anthropology, Brown University, Providence, RI.
Potter, P.B., Jr., and Switzer, D., 1989, Shipwrecks. Historic Context 65. New Hampshire Division of Histori-
cal Resources, Concord.
Schmidt, P.R., and Mrozowski, S.A., 1983, History, Smugglers, Change, and Shipwrecks. In Shipwreck
Anthropology, edited by R.A., Gould, pp. 143-171. University of New Mexico Press, Albuquerque.
Schmidt, P.R., and Mrozowski, S.A., 1988, Documentary Insights into the Archaeology of Smuggling. In
Documentary Archaeology in the New World, edited by M. Beaudry, pp. 32-42. Cambridge University
Press, Cambridge.
Schuyler, Robert, 1970, History and Historic Sites Archaeology as Anthropology: Basic Definitions and Rela-
tionships. Historical Archaeology 4: 83-89.
 Additional Bibliography
for Part VI.A

Anderson, R.K., 1988, Guidelines for Recording Historic Ships. National Park Service. U. S. Department of
the Interior, Washington, DC.
Delgado, James P. et al., 1991, National Register Bulletin No. 20 Nominating Shipwrecks to the National Reg-
ister, National Park Service, Interagency Resources Division, Washington, D.C.
Giesecke, Anne G., 1989, States and Their Shipwrecks. Underwater Archaeology Proceedings from the Society
for Historical Archaeology Conference. edited by 1. Barto Arnold, pp. 35--41. Society for Historical
Archaeology, Tucson, Arizona.
U. S. Department of the Interior, 1992, National Register of Historic Places Information. U. S. Department of
the Interior, National Park Service, Interagency Resources Division, Washington, D.C.

427
 Part VI.B

Specific Criteria

In Chapter 39, Watts discusses the site-by-site approach to assessing the significance of
wrecks, under pressures such as salvage, maintenance, and environmental, which must
be changed to include a firm database. Primary considerations must be addressed for
each wreck: (l) historical assessment ("facts" about the ship), (2) extent of the resource
base for the particular wreck type (gaps in our knowledge), (3) site reconnaissance data
(physical environment), and (4) historical, archaeological, and anthropological value
scale. Due to the popularity of the Civil War period, an assessment scheme developed for
ships from this period in North Carolina can act as a model for other wreck categories.
In Chapter 40, Alford stresses the need for a strong typology in the field assess-
ment of the significance of particular underwater sites. The scheme of classification
and hierarchical status of such a typology is particularly useful concerning the diversity
of small boats in North Carolina. The best of these systems, in Alford's consideration,
is one based on structural morphology. Dangers lie in the possibility of oversimplifica-
tion and in erratic data resulting from designs not included in the typology (and not
built in North Carolina). He includes a typology for North Carolina boat types from
1790-1920.

429
 Chapter 39

Toward Establishing
Research and Significance
Criteria for Civil War
Shipwreck Resources

GORDON P. WATTS Jr.

Determination of significance and assessment of research potential for shipwreck sites

have generally been carried out on a vessel-specific basis as a response to one of a vari-
ety of threats to the resource. Those responsible for the protection and management of
shipwreck archaeological sites are frequently put in the position of having to make
decisions about site value as a reaction to rapid environmental changes, developmental
pressures, channel or beach maintenance activities, salvage interests, looting, and on
occasion, proposals for scientific research. Unfortunately, most of these decisions are
made without the benefit of an examination of the associated historical context or an
assessment of the nature and scope of the data base.
With the apparent perception of "significance" accepted to identify all things that
are "not significant" and vice versa, options have been polarized. The remains of ves-
sels are consequently determined to be either "significant" and worthy of preservation
or mitigation, or "not significant" and unworthy of the effort and expense. Decisions
made working within this dichotomy, under the normal pressure politic and without a
supportive base of historical, archaeological, and anthropological data, frequently do
Source: Underwater Archaeology Proceedings from the Society for Historical Archaeology Conference, edited
by Paul F. Johnston (1985), pp. 133-137. Reprinted by permission of the Society for Historical Archaeology,
Tucson, Arizona.

431
432 G.P. Watts Jr.

not accurately reflect the real values of the resource, the interests of the historical,
archaeological, and anthropological communities, and the general public. Where miti-
gation is required, the research is often inappropriately conceived and ill-suited to the
real needs associated with the resource.
To improve the process for assessing research potential and significance criteria
that support those decisions and provide some assurance that the limited resources
available for protection, investigation, and development of shipwreck sites are not mis-
appropriated for activities that will generate only nominal returns, criteria must be
developed and utilized by responsible managers. Although most states with submerged
cultural resource management programs have developed shipwreck files to assist in site
identification and protection, few agencies have made a concerted effort to properly
research vessels, categorize resources within a historical, archaeological, and anthropo-
logical context, identify criteria for assessing significance and research potential, and
develop site protection and mitigation requirements that are reflective of established
resource values.
In North Carolina, historical research and shipwreck investigations associated
with the remains of vessels lost during the American Civil War have generated suffi-
cient data to identify some initial considerations essential in developing significance
and research criteria. Although much remains to be done, the research to date has iden-
tified a number of primary considerations. A brief synopsis of this research and an
examination of these considerations can perhaps serve to focus attention on the prob-
lems of significance and research potential.
Naval and maritime activity played a decisive role in the conduct of Civil War
activities in North Carolina. A unique coastal geography provided the environment for
virtually every type of water-related military and civilian operation. As a result, the
coastal and offshore bottom lands of North Carolina contain the remains of more than
100 Civil War vessels (see Table 1). A preliminary inventory of Civil War shipwrecks
provides an insight into the nature and scope of the resource base.
This shipwreck population, exclusive of those wrecks recovered or destroyed by
postwar salvage activities, includes 34 of the approximately 85 categories of Union and
Confederate vessels lost during the war (Shomette, 1973). Of the approximately 185
Union naval and military vessels lost, a total of 29 were sunk in North Carolina waters.
More than 246 Confederate naval and military vessels were also lost during the con-
flict. Of those, 27 have been determined to be in North Carolina. Out of approximately
182 ships lost while engaged in Confederate associated maritime commerce, at least 55
are known to have been sunk in North Carolina. In sharp contrast, only 2 vessels out of
the more than 182 lost in carrying on Union commercial activity were lost in the state.
In rough figures, the Civil War shipwreck population of North Carolina contains
approximately 14% of the total Civil War shipwreck data base. Of these wrecks,
approximately 30% have been located and to varying degrees examined.
Although only a portion of that work has been carried out in accordance with ar-
chaeological standards, all of the activity has resulted in the generation of data concern-
ing Civil War shipwreck sites. Initial interest in Civil War shipwrecks was generated
by the discovery and subsequent salvage of the blockade runner Modern Greece.
Through the office of the Governor, the North Carolina Division of Archives and His-
tory obtained assistance in salving the cargo and a variety of material associated with
the ship structure (Bright, 1981).
Establishing Research and Significance Criteria for Shipwreck Resources 433

Interest generated by activities associated with the Modern Greece resulted in the
location and investigation of at least ten other Civil War vessels. Investigations associ-
ated with these wrecks concentrated on identifying their location, conducting a prelimi-
nary assessment of the condition of the wreck, and recovering material from the sites
for displays associated with the Civil War centennial celebration. That celebration and
the recovery of material from the blockade runners stimulated interest in the recovery
and display of the remains of the Confederate ironclad CSS Neuse (Bright, 1981).
The activities of Underwater Archaeological Associates, Inc., a nonprofit group
formed to investigate the remains of North Carolina shipwrecks, produced additional
information concerning Civil War vessels. Operating in the early 1970s, the group
explored the remains of blockade runners lost in the vicinity of Cape Fear, south of
Wilmington. For the first time, an effort was made to recover both material preserved
at a site and the associated archaeological record (Watts and Bright, 1973).
In conjunction with these and other activities, the North Carolina Division of
Archives and History was able to develop an in-house underwater archaeological
research capability. The passage of protective legislation in 1967 confirmed state own-
ership of submerged cultural resources and delegated management authority to the
Division of Archives and History. By 1971, funding was approved for a professional
staff to manage the underwater archaeology program and expand research associated
with shipwreck site location, identification, and investigation. By developing coopera-
tive research projects with other state agencies, educational institutions, and historical
societies, the Underwater Archaeology Branch was able to significantly increase the
number of known Civil War shipwreck sites and refine data about previously identified
vessels (Watts and Bright, 1973; Watts, 1975).
Within the past 2 years, shipwreck site location and assessment surveys associ-
ated with channel maintenance activities of the Wilmington District United States
Army Corps of Engineers have generated additional information concerning Civil War
shipwrecks in North Carolina. In 1983, investigation of the remains of two vessels in
the vicinity of Carolina Beach Inlet led to their identification as Civil War blockade
runners (Watts, 1984). The following year, the remains of three additional vessels were
located and investigated in the vicinity of Lockwoods Folly Inlet. Examination of
exposed hull structure at each site and historical research confirmed that two of the ves-
sels were the remains of blockade runners; the third was identified as a Union gunboat
(Watts, 1985).
Historical research initiated to identify the Carolina Beach Inlet and Lockwoods
Folly Inlet shipwrecks was expanded to generate insight into the significance and
research potential of the wreck sites and assist in determining appropriate measures for
mitigation should the sites be threatened by channel maintenance activities. To date,
the research has confirmed the existence of extensive historical source materials. Many
of these have not been previously examined and most have never been considered in
determinations of shipwreck significance and research potential. The data identified in
these sources shed new light on the vessels involved in blockade running, the cargoes
carried by blockade runners, and the activities of those engaged in various aspects of
the trade.
One example of this research activity and perhaps the most extensive source of
new data concerning blockade running was found in the court records associated with the
adjudication of captured vessels. Extensive collections in regional repositories of the
Table I. Preliminary Civil War shipwreck inventory
Total lost
Total
Vessel type U.S. N.C. located/examined
Union military
First rate
steam frigate I 0 0
ship 2 0 0
Second rate
steam screw 3 0 0
frigate 4 0 0
ship 2 0 0
Third rate
side-wheel steamer (iron) 0 0
side-wheel steamer (wood) I 0 0
side-wheel steamer (wood/double end) 2 2 0
Fourth rate
side-wheel steamer (iron) 2 0 0
side-wheel steamer (wood) 2 2 0
screw steamer (iron) 2 2 I
screw steamer (wood) 7 5 3
screw tug (iron) 7 0 0
screw tug (wood) 3 0 0
Monitors 6 I I
Submarines 0 0
Ironclad steamer 2 0 0
Cotton clad steamer I 0 0
Tinclad steamer 2 0 0
Total 51 12 S
Auxiliary military service vessels
sloops 3 0 0
schooners 32 3 0
barks I 0 0
brigs I I 0
ships 3 0 0
canal boats II 0 0
picket boats I 0
mail boat 0 0
ferry 4 4 0
unidentified steamers 19 I 0
paddle-wheel steamer (wood) 3 3 I
paddle-wheel steamer (iron) 4 I I
paddle-wheel tug (wood) I 0 0
screw tug (wood) 4 0 0
stem-wheel steamer (wood) 4 0 0
screw steamer (wood) 2 2 I
screw steamer (iron) 4 0
Total 98 14 3
Stone fleet vessels
bark 10 0 0
whale bark 5 0 0
ship 14 0 0
whale ship 7 0 0
Total 36 0 0
Union commercial
pilot boat (steam) 2 0 0
mail steamer I 0 0
schooner 43 I 0
bark 27 0 0
Table 1. (Continued.)

Total lost
Total
Vessel type U.S. N.C. located/examined
Union commercial (Continued.)
whale bark 18 0 0
collier 2 0 0
brig 16 0
hermaphrodite brig 0 0
ship 37 0 0
whale ship 23 0 0
clipper 3 0 0
Total 172 2 0
Confederate military
Second rate
frigate raised
Third rate
steam sloop 3 0 0
Ironclad ram 18 4 2
Cotton clad 8 0 0
Screw gunboat (iron) I 0 0
Screw gunboat (wood) 6 I
Screw steamer (iron) 4 2 I
Screw steamer (wood) 5 0 0
Screw tug (wood) 2 0 0
Screw torpedo boat (wood) 0 0
Submarine torpedo boat (wood) 0 0
Submarine 0 0
Total 51 8 4
Auxiliary military service vessels
steamer (unidentified) 46 4 I
paddle-wheel steamer (iron) II 0 0
paddle-wheel steamer (wood) 40 3 0
lightship I 0
floating battery 6 0
barge 4 0 0
sloop 7 0 0
schooner 63 10 0
bark 3 0 0
brigantine I 0 0
ship 0 0
unidentified 12 0 0
Total 195 19 1
Confederate commercial
steamer (unidentified) 10 0 0
paddle-wheel steamer (iron) 31 15 9
paddle-wheel steamer (wood) 12 3 2
screw steamer (iron) II 8 6
sloop 28 0 0
schooner (center board) 0 0
schooner (center board) 0 0
schooner 80 23 3
bark 4 2 0
brig I 0
ship 2 I 0
unidentified 2 2 0
Total 182 55 20
Ship losses 785 110 33
Resource base % 100% 14% 4.2%
436 G.P. Watts Jr.

National Archives at Boston, New York, Philadelphia, Atlanta, New Orleans, and Lea-
venworth, Kansas, preserve information associated with each case. A typical record set
might contain a detailed and certified cargo manifest, frequently prepared to facilitate
auction of the goods. Information about the ship often survives in records associated with
the survey of the vessel carried out to facilitate auction or purchase of the ship by the
United States Navy. To ensure that the crew of each prize was properly interrogated, an
inquiry form containing a standardized series of questions designed to produce intelli-
gence concerning the activities of the captured vessel and others operating out of the
same ports had to be filed for each prisoner. Captured ships' papers frequently included a
vessel's log, enrollment and registry, manifest, accounting disbursements, certificates of
the ship's officers, and roster of the crew. Navigational charts and copies of regulations
that governed shipboard life were occasionally included among newspapers, mail, per-
sonal receipts, invoices, and diaries held as evidence. With 198 cases adjudicated in New
York alone, these court records may well represent the largest collection of blockade
running material available for investigation outside investigation of shipwreck resources
(Robinson, 1945).
Establishing the significance and research potential of shipwreck sites under
investigation at Carolina Beach and Lockwoods Folly has provided an opportunity to
examine both the nature and scope of information generated by 20 years of Civil War
shipwreck investigation in North Carolina and the historical record sources associated
with that period. Although that investigation is still under way, several primary consid-
erations have been identified.
First, assessment of vessel significance must be made within a well-developed
historical context. That political, social, military, economic, and technological context
is essential to understanding the nature of human activity associated with the remains
of any ship. Although sufficient data exist to develop an excellent historical context to
support assessments of shipwreck significance and identification of research priorities,
little effort has been made in that direction.
Investigation of the historical source materials associated with Civil War vessels
confirms that many gaps in our knowledge of blockade running can be filled through
archival research. Detailed cargo-related records survive in judicial records, newspa-
pers, auction advertisements, and manifests. Although architectural and construction
records are minimal where sailing vessels engaged in blockade running are concerned,
the records associated with steamers are quite numerous. Steamships were one of the
most dramatic products of the Age of Engineering that changed ship construction.
Engineers published extensive records of their work and patented their designs. A pro-
liferation of secondary treatises on engineering associated with ship design and con-
struction document practices that became progressively more uniform as construction
materials were standardized and regulations for ship construction became increasingly
complex. A systematic investigation of these and other record sources would identify
data available only through on-site investigation of shipwreck remains.
Second, assessment of shipwreck significance and identification of research pri-
orities should be made with specific consideration for the nature and extent of the asso-
ciated resource base. While it is perhaps possible to identify those shipwrecks that are
of paramount importance without a detailed examination of the associated resource
base, it would be difficult to responsibly assess the significance of other shipwreck
sites where values must be determined in accordance with representative criteria. Until
Establishing Research and Significance Criteria for Shipwreck Resources 437

the shipwreck population of a given period can be quantified and categorized, it is vir-
tually impossible to make judgments concerning significance and/or priorities for
research. This research and an assessment of the resource base are particularly critical
to the development of responsible priorities that realistically reflect resource values.
For Civil War period sites, this can be accomplished with a high degree of reli-
ability. Nineteenth-century shipwreck references became both more frequent and pro-
gressively more detailed as insurance became an essential aspect of maritime
commerce and newspapers broadcast the details of daily life. Naval communications
became more extensive and a higher percentage of these documents have survived to
preserve a record of activity at sea.
Third, shipwreck significance must be established with specific consideration for
the nature and scope of the archaeological record both at the site in question and with
respect to other resources in that respective category. Where wrecks are considered as
representative of a particular category, the condition of the site and integrity of the
archaeological record must be evaluated in determining significance and research
potential. This makes site reconnaissance data particularly important. In some cases,
valuable sites have been virtually destroyed by post-Civil War salvage activity, others
have been damaged or destroyed by the activities of modem looters and development,
and in some cases, high-energy environments have destroyed the very aspects of the
archaeological record that make the site valuable. Conversely, other sites in similar
high-energy environments contain excellent material records. Without on-site recon-
naissance surveys, these considerations cannot be adequately addressed.
Fourth, based on an assessment of the historical, archaeological, and anthropo-
logical interests in Civil War period shipwreck sites, it is apparent that significance and
research potential are too complex for a polarized binary system. As there must be lev-
els of significance designed to reflect a variety of historical, archaeological, and
anthropological values, an ordered scale appears to be more appropriate. One approach
to establishing such a scale might include the following categories:

1. First-order shipwreck sites would include only those vessels determined to

(a) be associated with historical events of national or international impor-
tance, (b) preserve a unique archaeological record that does not survive else-
where in the data base, and (c) contain historical and/or anthropological data
essential to developing an understanding of past human activity.
Sites determined to be of first-order significance would be identified and
preserved for scientific investigation. Protection of the site would be ensured
by designation as a first-order resource. With the exception of approved
research, all on-site diving would be limited to nondestructive activities.
2. Second-order shipwreck sites would include only those vessels determined to
(a) be associated with events of importance to states or regions within the
United States, (b) preserve an archaeological record that represents a major
source of information within the data base, and (c) contain historical and/or
anthropological data that make an essential contribution to developing an
understanding of past human activity.
Sites determined to be of the second order of significance would be iden-
tified and preserved for scientific investigation to recover the scientific infor-
mation contained in the archaeological record and, if possible, preserve the
438 G.P. Watts Jr.

remains of the vessel. With the exception of approved research, all on-site
diving would be limited to nondestructive activities.
3. Third-order shipwreck sites would include only those vessels determined to
(a) be associated with events of local importance, (b) preserve an archae-
ological record that represents one of several sources of information within
the data base, and (c) contain historical and/or anthropological data that make
a contribution to enhancing our reconstruction and understanding of past
human activity.
Sites determined to be of third-order significance would be identified
and available for scientific research, educational instruction, and salvage
designed to recover pertinent data preserved at the site. With the exception of
approved research, educational and recreational projects, and salvage, all on-
site diving would be limited to nondestructive activity.
4. Fourth-order shipwreck sites would include only those vessels determined to
(a) have limited association with the historical events of the period, (b) con-
tain a minimal archaeological record related to circumstances associated with
loss, environment, and/or man's activity, and (c) preserve archaeological and
anthropological data contained and reproduced at other higher-order sites.
Sites determined to be of the fourth order would be identified and avail-
able for student and directed amateur research, educational exercises, and sal-
vage activity restricted by minimal data recovery requirements. With the
exception of approved research, recreational and educational activities, and
salvage, all on-site diving would be limited to activities that do not disturb
the archaeological record preserved within the bottom sediments. Surface
material associated with the site could be recovered if reported.
5. Fifth-order shipwreck sites would include only those sites determined to have
no significant association with the events but represent a resource worthy of
protection for educational and recreational activity.
Sites determined to be of fifth-order significance would be identified and
available for student and amateur research, and educational and recreational
activities that are determined to be nondestructive.
6. Sixth-order shipwreck sites would include only those sites determined to (a)
have no significant association with period events and human activities and
(b) preserve no significant archaeological record.
Sites determined to be of sixth-order significance would be identified
and included in resource inventories but remain unprotected and available for
any public or private use. No restrictions would be imposed on diving activi-
ties at the site.
Although human activity and development have been continuous and uninter-
rupted, they have frequently been categorized for the purpose of investigation. By adopt-
ing an arbitrary but functional chronological framework, the comparative process could
be structured to address levels of significance within a manageable context of human
activity. Isolating periods of development, such as Exploration, Colonization, Revolu-
tionary War, Federal, Antebellum, Civil War, Early Modern, and Modern, would facili-
tate use of both primary and secondary historical source materials in the development of
background chronologies. Resources associated with each period could be quantified and
Establishing Research and Significance Criteria for Shipwreck Resources 439

assessed to establish significance and research priorities. Period-specific significance and

research priorities could then be ranked in accordance with criteria developed to support
assessment of the total resource base.
Given the need for well-developed criteria for assessing the significance and
research potential for shipwreck resources, it is perhaps worthwhile to consider a
method of addressing the problem. Because of the extensive attention that Civil War
period shipwrecks and historical research have received, that period could serve as a
model. Standards for assessing significance and identifying research priorities devel-
oped in conjunction with Civil War shipwreck resources could be generalized to assist
other period and geographical areas. However, until the question receives the serious
attention of historians, archaeologists, and anthropologists with interests in shipwreck
resources, management decisions will continue to be made without the benefit of ade-
quate guidelines.

REFERENCES
Bright, L.S. et aI., 1977, The Blockade Runner Modern Greece and Her Cargo. North Carolina Division of
Archives and History, Raleigh.
Bright, L.S. et al., 1981, C.S.S. Neuse: A Question of Iron and Time. North Carolina Division of Archives and
History, Raleigh.
Robinson, M.R., 1945, An Introduction to the Papers of the New York Prize Court. 1861-1865. Columbia Uni-
versity Press, New York.
Shomette, D.G., 1973, Shipwrecks o{the Civil War. Donic Ltd., Washington, DC.
Watts, G.P., Jr., 1984, Underwater Archaeological Reconnaissance Carolina Beach Inlet, New Hanover County,
North Carolina. Submitted to the U.S. Army Engineer District, Wilmington, NC.
Watts, G.P., Jr., and L.S. Bright, 1973, Progress in Underwater Archaeology in North Carolina 1962-72. Inter-
national Journal of Nautical Archaeology, 2(1 ): 131-136.
Watts, G.P., 1r., 1985, Underwater Archaeological Reconnaissance Lockwood's Folly Inlet, Brunswick County,
North Carolina, Report in preparation.
Watts, G.P., Jr., 1975, Report on the Activities of the 1975 Field School in Underwater Archaeology. Project
report (draft on file with the Program in Maritime History and Underwater Research, East Carolina
University).
Wise, S.R., 1983, Lifeline of the Confederacy: Blockade Running During the Civil War. Ph.D. dissertation,
University of South Carolina.
 Chapter 40

A Small~Craft Typology:
Tool for Archaeological
Research

MICHAEL B. ALFORD

Three common questions facing archaeologists in the field on confronting the remains
of a small craft or larger vessels are: What is it? How old is it? Is it significant? As pre-
sumed experts in the area of marine vessel evolution and history, the maritime staff at
North Carolina Maritime Museum is frequently approached for assistance in obtaining
answers to those questions.
These questions are not necessarily easily answered, identification and evaluation
being complex issues. Most often, the vessels in question are situated in underwater
sites, and the archaeologist has gathered what little data as was possible under the most
trying of conditions. The combination of limited visibility, hazardous currents, physical
hardship, and poorly preserved, usually largely obscured structural components con-
spire to prevent accurate observation of critical features.
Finding answers to these questions is made even more difficult if there happens
to be no common language between the asker and the hopeful responder. Nomenclature
of ship structures is often vague and ambiguous.
In some aspects of vernacular boat types there may be no recognized standard
nomenclature. Boat-building practices can vary widely from region to region, or even
within the same general locale. Often, the questions come over the telephone, with each
Source: Underwater Archaeology Proceedings from the Society for Historical Archaeology Conference, edited
by J. Barto Arnold (1989), pp. 61--63, Reprinted by permission of the Society for Historical Archaeology, Tuc-
son, Arizona.

441
442 M.B.Alford

party groping for descriptive phrases that will somehow transform murky images into
concise identities. Inevitably, the tables are turned and the hunter of information
becomes the hunted, the brunt of questions rather than the recipient of answers.
This is a frustrating situation for everyone involved. The archaeologist in the
field needs immediate answers. Often, a contractor or government agency is complain-
ing of needless delays and escalating construction costs. Analysis of data and reluc-
tance to make decisions based on sketchy observations may be viewed as incompetence
or intolerable regulatory quagmires.
Although archaeologists are on the "front line," and thus take most of the "heat"
in these situations, it is unrealistic to expect them to be specialists in all of the fields
they encounter on a dig, or a dive. That is particularly true in the case of boat and ship
structures, for which we are still gathering the data (for regional vessels) that will
enable the teaching of what is a fragmented and difficult subject. But as the specialist
in this case also may not be a diver, and cannot (or will not) go on site (I freely confess
to being a terrestrial being: I am not and have never been aquatic or even semiaquatic
in the sense that the acronym "scuba" suggests), and because in most cases the 'object"
cannot be delivered to the expert's laboratory, the underwater archaeologist becomes
the "eye" of the specialist. In order to "see," the archaeologist must know what to look
for, and must be able to communicate back to the specialist the observations that will
lead to answers.
This is the function to which the small-craft typology attempts to contribute. It is
designed to reduce a maze of types, and variations within types, to a manageable number
of related groupings, with the key characteristics by which they may be identified. The
idea has sprung from the urging of state Underwater Archaeology Unit personnel, who
needed a guide to North Carolina's diverse historical small craft. After years of collect-
ing data from abandoned boats, from boats stored in barns, from pieces of boats lying in
marshes or behind fish houses, by examining fragments of boats raised from underwater
graves, and from repeated efforts to combine, or reconcile fieldwork with historical
accounts, records, and documents, the unwieldy nature of this mountain of data had
become obvious. Work was begun on a scheme to classify known examples ofvernacu-
lar North Carolina boat-building techniques into a simplified checklist that would pro-
vide assistance to divers who needed to make quick, on-site assessments relating to
identification and historical significance of submerged resources. Patterns and relation-
ships emerged when examples were grouped by similarity of major structural elements,
and when those elements were given a hierarchical status.
The system is based on the premise that structural systems are a more reliable
indicator of origin and related cultural aspects than is shape, and what we usually refer
to as the "lines of a boat." This concept may not be universally accepted as yet, but at
least in the present situation it has removed several blocks that had continually proved
troublesome previously.
Acceptance of a typology based on structural morphology does not eliminate
shape as a consideration. In the first place, boat shape is very much related to the meth-
ods and materials selected for construction. However, shape can and does vary within a
type, within limits, and is also, like fad and fancy, subject to the whims oftaste and the
winds of change. A shape typology would probably have to begin with a classification
separating hulls into the traditional three classes: round-bottom, vee-bottom, and flat-
bottom. Immediately, this invokes artificial criteria which will prejudice many future
A Small.Craft Typology: Tool for Archaeological Research 443

decisions regarding classification. True evolutionary lines, if present, will be camou-

flaged and may go undetected. There are other instances in which inconsequential fac-
tors conspire to confuse the issue when shape is used as a classification device.
The structural system recognizes the role of boat-building technology in the
social and economic development of a societal group. As boat use and economic fac-
tors change in a community, shape, size, and arrangement of the boat and its equipment
may alter, but the basic structural elements will evolve more slowly, because of the
dependency of the building on the technology "status quo" in a vernacular setting.
A structural typology has the added advantage of being based on tactile, defini-
tive objects that are readily recognizable and readable in the field. It must be empha-
sized that the present typology is based on data collected to date and is thus limited. It
is still undergoing testing in the field and also with accumulated file data.
The structural elements incorporated into the typology are planking style, keel
morphology, and the system of transverse frames. It is helpful if the diver or observer
can determine the general structural class, designated as "plank on frame," "skiff," etc.,
but at this point the system is not so cumbersome as to prevent starting in the middle
and then eliminating the obvious disparities.
The designation "flat-bottom," "deadrise," and "round-bilge" are used in the
sense of construction types, and are not descriptions of shape. Many times, any of these
three construction types can be flat or relatively flat in shape over some portion of the
bottom, leading to confusion and misinterpretation. There is an apparent allusion to
shape in the descriptives "square both ends" and "sharp stem," but the reference is
intended for structural features, transverse structures truncating each end in the first
instance, and planking terminating at the stem and stem post in the second.
Boats hollowed from logs or incorporating log shells in their construction occupy
a separate major category. In a system based on "round," "vee," and "flat" bottoms,
these clearly distinct boats usually get tucked away in irrelevant or inappropriate cate-
gories. Log boats derive essential strength from the monolithic nature of their shells.
When logs or log parts are joined, however, and in some lightly constructed sin-
gle log boats, additional stiffening and strength is achieved through the use of trans-
verse frames and sometimes longitudinal devices. In the more sophisticated form of log
boats, we may see techniques applied to "extend" the capacity of these craft. In North
Carolina this consists of either raising the freeboard through the use of a strake or
strakes installed along the upper edges of the log shell, or by separating the port and
starboard portions of the log shell and combining them with a centerline structure
resembling the stem, keel, deadwood, and stempost of plank-built boats. The two tech-
niques are often seen in the same example.
Apparent limited success with the typology suggests that eventually it may lead
to a dichotomous key by which, through a series of simple choices between paired
structural features, identification of vessels at the specific level may be accomplished.
Such a key is a tool ideally suited to the needs of the archaeologist who may lack the
specialization required for identification and evaluation of the wide variety of vessels
of diverse origins potentially residing in underwater sites.
There are some negative aspects to this typology. It has previously been stated that
it is limited and experimental. I suspect also that it is oversimplified, thus resulting in the
risk that field operators may assume a vessel is insignificant if a vessel's remains do not
immediately settle into a niche in the typology outline. The typology does not directly
444 M.B. Alford

address significance, admittedly one of the initial objectives. On the other hand, it does
say that vessels or remains that can be placed on the outline match certain minimum cri-
teria for vernacular craft historically used in North Carolina. Every boat of vernacular
origin is unique; each has significance of greater or lesser degree. Identification is nine-
tenths of the significance questions, and the typology is a step toward identification.
An extremely troubling deficiency in this typology is its lack of application for
boats whose origins lie outside the boat-building techniques historically practiced in
North Carolina. Boats are obviously mobile and, by purpose, move about and may
operate some distance from their points of origin. Some of these will obviously come to
rest in foreign ports. Application to vessels that have come from outside the implied ge-
ography may result in misleading conclusions, or may fail to register at all. Hopefully,
though, observations that are made within the scope of this typology will result in data
that will be helpful in arriving at a valid identification of any vessel type for which
there is available documentation. Its success ultimately lies with the experience of the
user and his or her willingness to pursue matters to fruition
The references in this paper provide a guide to literature relating to small-craft
typologies.

REMARKS TO ACCOMPANY TYPOLOGICAL OUTLINE

The typology suggested in its outline is an evolving system based on limited in-
formation. The structural features on which the categories, as they are presently config-
ured, were established represent concepts based on a combination of field observation
and historical records; none are speculative. They are valid only when applied to the
vernacular boat types native to North Carolina and, specifically, to those generally not
more than 40 ft in length. The typology is further limited, for the most part, to boats
and methods used prior to about 1920.
The author makes no claim that this system is either accurate or complete at this
time, only that it appears to work as a device to classify boat types within its scope and
to indicate possible relationships. Further testing and evaluation in light of recent and
forthcoming data will enable a continued honing process to increase the reliability and
effectiveness of the typology.

TYPOLOGY OF NORTH CAROLINA INDIGENOUS BOAT TYPES

FOR THE PERIOD 1790-1920
Typological Hull Classification (Draft Outline December, 1988)
I. Hollowed log shell (dugout, canoe, canoe-boat, kunner [coiL], piragua)
A. Single log, simple, through-form or,rudimentary
B. Single log, simple, modeled form
C. Single log, complex (inserted frames, modeled form)
D. Split log, complex (inserted frames), extended, modeled form
E. Multiple log, complex, extended, modeled form (unconfirmed verbal report)
II. Plank-on-frame
A. Round bilge, molded cross-section keel
I. Transom stem
2. Fantail stem
A Small-Craft Typology: Tool for Archaeological Research 445
3. Staved round stem
B. Round bilge, flat keel (or hog.)
I. Transom stem
2. Round staved stem
C. Round bilge, lapstrake
I. Sharp stem
2. Transom stem
a. Skiff
b. Deep keel, decked or partially decked
D. Deadrise bottom, longitudinally planked bottom, molded cross-section keel
I. Transom stem
2. Fantail stem
3. Staved round stem
E. Deadrise bottom, bottom planked longitudinally, flat or plank keel, or keel-less
I. Transom stem
2. Fantail stem
3. Staved round stem
4. Log forefoot
5. Staved forefoot
6. Planked forefoot
F. Flat bottom, longitudinally planked, more or less rockered bottom
I. Transom stem
2. Staved round stem
3. Sharp stem
4. Square both ends
5. Side frames molded
6. Side frames straight
7. Side planked
8. Sides molded cross-section
G. Plate bottom and knee, little or not bottom rocker
I. Sharp stem or transom stem
a. Side frames molded
b. Side frames straight
III. Skiff constructed (slab or post)
A. Flat-bottom skiff, bottom planked transversely
I. Transom stem
2. Sharp stem
3. Swim head (square both ends)
B. Deadrise skiff, bottom planked longitudinally small angle of deadrise (modified skiff const.)
1. Log forefoot
2. Staved forefoot
IV. Flats, flatboast, and scows (square both ends)
A. Essentially skiff-built but larger and heavier construction
1. Bottom planked transversely
2. Bottom planked longitudinally
B. Archaic form (molded side construction)

REFERENCES
Chapelle, H. I., 1951, American Sailing Craft. Norton, New York.
Greenville, B., 1976, Archaeology of the Boat. Wesleyan University Press, Middletown, Connecticut.
Johnstone, P., 1980, The Seacraft of Prehistory. Harvard University Press, Cambridge, Massachusetts.
McKee, E., 1983, Working Boats of Britain. Conway Maritime Press, London.
Phillips-Birt, D., 1979, The Building of Boats. Norton, New York.
 '.~~.' Additional Bibliography
~

o i' .
for Part VI.B
q;,O_" .': .
.j'~l

Carrell, Toni, 1989, In All Things Remembered: An Oral History Approach to Understanding Small Craft
Remains, Underwater A rchaeology Proceedings from the Society for Historical Archaeology, edited by
1. Barto Arnold, pp. 76--81. Society for Historical Archaeology, Tucson, Arizona.
Litwin, Jerzy, 1995, Working Boats of Poland, Gdansk. Maritime Museum Proceedings, Vol. X. Gdansk, Poland.
Yoshiahi, Kaoru and Peter Hoagland, 1994, The Value of Historic Shipwrecks: Conflict and Management.
Coastal Management 22(2): 195-214.
Tolson, Hawk, 1991, Vernacular Watercraft of Isle Royale National Park. Underwater Archeology Proceedings
from the Society for Historical Archaeology Conference, edited by John D. Broadwater, pp. 132-33.
Society for Historical Archaeology, Tucson, Arizona.

447
 Part VII

Data Recovery

The major thrust of archaeological recovery is to learn about the past, both the people
and their interrelation with their environment. Archaeologists, like salvors and treasure
hunters destroy sites. The difference is that the archaeologist mitigates the adverse im-
pact by recording the details of research plans, survey, the site before and during exca-
vation and then reports on what was found and what was learned. Data recovery is the
basic acquisition of information for use in interpreting the past and the accurate record-
ing of data is the single most important aspect of archaeological field work.
Obviously, there will be time and financial constraints on the fieldwork. Thus, a
strategy to learn as much as possible while disturbing the site as little as possible is
very important. Sampling strategies provide options to full excavation and can provide
a wealth of information without totally disassembling the site.

449
 Part VII.A

Pre . . Excavation Recording

Working with the barest necessities of diving and measuring equipment, Gluckman
(Chapter 41) succeeded in reporting not only environmental site conditions but also
provided a rough draft of expected findings, sketch maps, photographs, and clearly
delineated areas of material remains. The results are interpreted as tests of stated
hypotheses. Gltlckman acknowledges at the end that "this preliminary work has pro-
duced more questions than answers."

451
 Chapter 41

Preliminary Investigations
of a Shipwreck, Pumpata
Cahuita National Park,
Costa Rica

STEPHEN J. GLUCKMAN

INTRODUCTION

In late June 1978 I received a letter from Dr. Carlos E. Valerio, a professor in the Biology
Department at the University of Costa Rica. Dr. Valerio was seeking advice about the
protection of shipwrecks located in a national park off Costa Rica's Caribbean coast.
The situation he described was all too familiar. The wrecks are located in shallow
water with good visibility and are easy to reach. Local and visiting divers were removing
material and Dr. Valerio wanted to know what could be done to stop this destruction.
I responded with the usual suggestions about surveillance and the imposition of
civil and criminal penalties for the illegal removal of materials. I also pointed out that
this was a worldwide problem and that no country had been very successful in stopping
such activities. In addition, I suggested that without the support of the local diving com-
munity and a program for conservation and management of these historic resources, it
was unlikely that the situation would get anything but worse.
During the exchange of letters that followed, I was extended an invitation to
come to Costa Rica and assess the situation. Final arrangements for my visit were made
in late August and in the middle of September I went to Costa Rica for 10 days.

Source: Underwater Archaeology: the Proceedings of the Eleventh Co'!ference on Underwater Archaeology,
edited by Calvin R. Cummings (1982), pp. 6-19, Fathom Eight, San Marino, California. Reprinted by permission.

453
454 S.J. Gluckman

Because of the short time frame during which arrangements were finalized and
the shortcomings of long-distance communication, I had only a sketchy idea of where
the work was to be done, who would be involved, and what equipment might be avail-
able. Consequently, it was difficult to do any real preparation.
It was only after I arrived in Costa Rica that information needed to plan the work
became available. Two of Dr. Valerio's students, Jorge Campos and Hector Guzman,
met me at the airport in San Jose, the capital of Costa Rica. During the next 2 days I
examined maps and aerial photographs of the park where the wrecks were located and
discussed the project with archaeologists at the University of Costa Rica and the
National Museum. The project was also discussed with the director of the Servicio de
Parques Nacionales, Alvaro F. Ugalde, and the administrator of Parque Nacional
Cahuita (the park that contains the wrecks), Sigifredo Marin-Zuniza.
These discussions provided my first specific information about the wrecks, the
diving conditions, and how I might proceed with the assessment. It quickly became
apparent that no one knew very much. Hector Guzman had dived on the wrecks but it
was not clear from his description whether it was one or two wrecks. I did learn where
the sites were, that one area contained cannon and a large amount of brick, and that a
second area contained a number of cannon and a couple of large anchors.
The last day in San Jose was spent gathering diving equipment and filling tanks.
The equipment was somewhat deficient from an archaeological perspective. There was
no depth gauge, only one underwater compass was available, and a 2-m wooden folding
rule was the only measuring device. No land surveying equipment could be obtained.
This lack of mapping equipment introduced data collection restraints.
On the evening of September 19 we arrived at the park. We made our first dive
the next morning and briefly examined both wreck areas.
The following day the remaining two members of our diving team, Quin Rodri-
quaz Toledo and Ricardo Soto Soto, joined us. Toledo was affiliated with the National
Museum and Soto with the Biology Department at the University of Costa Rica.
During the 4 days spent at the park we were only able to spend three mornings
diving. Outboard motor problems cost us one morning and onshore winds in the after-
noons reduced visibility to the point that productive work could not be done much
beyond midday.
Even with the equipment and diving problems, enough information was collected
to make a preliminary assessment. Perhaps the most important results of this brief visit
are that we now know what questions to ask and how to proceed toward the answers.

LOCATION AND PHYSICAL ENVIRONMENT

Parque Nacional Cahuita is located on the Caribbean coast of Costa Rica. The
park is midway between Costa Rica's only Caribbean port, Puerto Limon, and the
Panamanian border. Limon is about 20 miles north of the park and Panama about 20
miles to the south.
Costa Rica's eastern coast is essentially featureless. From Panama to the south to
Nicaragua to the north, there are no major embayments or headlands. River mouths
tend to be low lying and swampy and good anchorages of any kind are rare.
The coastline projects into the Caribbean in only two places: Punta Mona, just
north of Panama, and Punta Cahuita, about 20 km north of Punta Mona. Neither of
Preliminary Investigations of a Shipwreck 455

these points provides much in the way of a lee shore. Their maximum elevations are
quite low and the waters on each side are studded with reefs and shallows. In addition,
both points project north and the prevailing and storm winds tend to be northeast to
northerly. Either, or both, of these points could conceivably have been used for naviga-
tional reference by coastal sailing vessels though it is doubtful if either can be seen
from any distance offshore. But, as the 100 fathom line ranges from 3 to 22 miles off
the beach, there is plenty of water close enough to shore for any vessel that wanted to
use the points for a position fix. However, both prevailing winds and currents are
onshore along this part of the coast which would make it dangerous to come too far
inshore for a position fix (Anonymous, 1976).
Parque Nacional Cahuita was established in 1970. The park encompasses 200
hectares of land and 600 hectares of ocean (Anonymous, 1978:23). The northern
boundary of the park is a small stream, the Rio Suarez, just south of the small village of
Cahuita. The park's southern boundary is the Rio Carbon. There are about 6~ miles of
beach between the two rivers.
The park's major land feature is Punta Cahuita. The point runs in a northeast
direction and is about 2.5 km long. The average width of the point is 1 km. Punta
Cahuita is covered in dense, moist tropical forest with a canopy over 30 m tall. The
point is covered in standing water much of the year.
The park's major attraction, and the reason why the park was established, is the
coral reef. The reef follows the configuration of Punta Cahuita on the north, extends
around the point, and forms a good sized shallow lagoon south of the point.
Alfred's Shoal extends about a mile in a northerly direction from the reef off the
tip of Punta Cahuita. The water depth is about 5 m and waves break across the shoal
under certain wind conditions.
The reef depths are 1 m or less. Reef face depths are about 7 m and the 10-fathom
line is about a mile off shore.
The park is composed largely of recent alluvial deposits. Just west of the park is a
small range of mountains, the Fila Carbon. Their maximum elevation is about 300 m.
The Fila Carbon date from Oligocene to Pliocene (lower slopes to peaks). Uplifting and
faulting are apparently the major processes involved in the geological history of the
park area.
The climate of the park is classified as tropical wet. There are two drier seasons,
February through April and August through September. Average rainfall is 137.84
inches per year. The temperature averages 27°C with a variation of only 2°.
The park's marine environment is dominated by the fringing coral reef and asso-
ciated communities. The main communities and zones of the marine environment are
the beach zone, the rocky intertidal zone, the Thalassia community, the Sargassum
community, the sand community, and the coral community. These overlap to some
degree and combine to form the primary elements of the marine ecosystem.
A wide variety of food resources are supported by the marine environment. These
include several varieties· of shellfish, numerous species of fish, lobster, and sea turtle.
None of these, with the possible exception of turtle, seem to be available in any real
abundance. These resources are scattered throughout the various zones or, like the sea
turtles, are only present on a seasonal basis.
Water temperature in the park ranges from the low to high 80s (degrees F).
Depending on location and time of year, underwater visibility can be as low as I m. In
456 S.J. Gluckman

the shal10w lagoon on the southern side of Punta Cahuita, the visibility is over 13 m
during the drier parts of the year.
The park's land transport system tends to be along the shore and from south to
north around the point. None of the rivers within the park carry large sediment loads.
There is little beach erosion as the coconut palms (introduced around the tum of the
century) that line the beach have a root system which acts to stabilize the beach.
The only information on currents in the park is a hypothetical model based on
surface currents. It suggests that in the smal1 cove north of Punta Cahuita the surface
current pattern is circular in a clockwise direction. In the remainder of the park it
assumes that the currents follow the general pattern for this area and set generally north
to northeast with longshore currents northerly on the south side of the point and west-
erly on the north side. Because of the reef, turbulence would be the best way to charac-
terize water movements in most of the park.
In general terms, the climate of the park is hot and moist. Undisturbed land areas
are covered in tropical jungle which is low lying and wet much of the year. The west-
ern margin of the park, and surrounding areas, are cultivated with the principal crop
being cacao. The 6 or so miles of beach are covered with sand and coral rubble and
fringed with 30-m tall coconut palms. The water is warm and shallow with good to
excellent underwater visibility. Currents are mild to nonexistent. Some 100 species of
coral exist in the park. The reef is a fringing type and while quite beautiful, it is not as
spectacular as barrier reefs. As a working environment for underwater archaeology, it
is good to very good. But, because all reef systems are extremely fragile, great care
must be taken in the conduct of any underwater work so as not to upset the equilibrium
of the system.
Unless otherwise referenced, al1 specific data in this section have been taken
from an excel1ent, unpublished study of the park's marine environment done by Peace
Corps volunteer Gerard M. Wellington some time in 1973 or 1974.

RESEARCH DESIGN

Little of the information necessary to research design development was available

prior to the first dive. The general questions that needed to be answered by the prelimi-
nary examination were known. These, of necessity, were particularistic rather than
explanatory questions. That is, until the location, numbers, age, nationality, condition,
etc., of the wrecks were known, it was not possible to formulate certain kinds of test-
able hypotheses. Consequently, the work was planned to col1ect the data required to
answer the basic questions as well as the data needed to develop a research design to
guide future work.
The following are the questions it was hoped could be addressed through the pre-
liminary work:
1. How many wrecks are located within the boundaries of the park?
2. Where are the wrecks located?
3. What is the physical condition of each of the wrecks?
4. What is the age of each wreck?
5. What is the country of origin, or nationality, of each wreck?
6. What are the potentials for research, display, and interpretation of each wreck?
Preliminary Investigations of a Shipwreck 457

7. What conservation and management actions are required, or recommended,

short and long term, for each wreck?
8. Are there sites other than shipwrecks in the park or the potential for such sites?
9. What sort of research and management program should be established so as
to best utilize the park's underwater cultural resources?
These questions were designed to be progressive and cumulative yet independent.
Data were expected to accumulate to allow the addressing of increasingly more general
questions. And, each question was designed to be independent of all others so that none
needed the answer to a previous question in order to be answered.
The questions were formulated prior to the work so that the expected outcomes,
or results, could dictate what kinds of data should be obtained as well as how the data
should be collected to make them most useful. This approach, it was felt, would dictate
the choice of data collection methodologies which would allow application of the data
to the largest number of questions.
It was apparent at the outset that some of these questions were unlikely to be
answered. It was also obvious that some of these questions could only be answered
with the presence of specific artifacts and context. As there was no way to predict the
presence of such information it was necessary to choose a data collection methodology
that would return useful data even under "worst case" circumstances, i.e., the absence
of diagnostic artifacts and meaningful context.
Additional factors that impinged on the methodology to be used were:
1. Lack of mapping equipment which would restrict the collection of context
information.
2. Lack of time which meant that extensive rather than intensive methods would
have to be used.
3. Lack of archaeological training on the part of the crew (all were marine
biologists) which meant that general observations were more likely to be
accurate than specific observations.
A last concern was the choice of a methodology that would allow some control
over negative data. Because of the above-mentioned constraints and the presence of an
unknown amount of previous collecting at the sites, it was felt that only the absence of
certain general kinds of context information would be meaningful. That is, a lack of
portable artifacts could be the result of previous collecting and thus have no archae-
ological meaning. Only rigidly controlled, highly systematic, data collection which was
not possible under the circumstances, would allow one to assess the value of this kind
of negative data. On the other hand, the absence of ballast, at a wreck site, is a kind of
negative data which could be meaningful.
The questions were grouped into three independent categories on the basis of the
kind of data required for meaningful answers. Category one included the questions con-
cerning age and nationality. These questions required specific, largely artifactual, data.
Contextual information, such as the relations among and between the locations of
wrecks, as well as details of ship construction, might also be useful. But, there was less
likelihood for this kind of data return and a greater chance that potentially unavailable
historical information would be needed to make it immediately useful.
Category two included the questions concerning the numbers, locations, condi-
tion, and potential of the sites. These questions could be addressed using observational
458 S.J. Gluckman

and contextual data. Artifact information would also be important, but was not neces-
sary. If sites could be located and inspected, these questions could be answered. Any
data collected relative to the questions in category one would add detail and precision
to the category two answers.
The third category included the management questions. All data would have some
bearing on these questions but much of the required data would be nonarchaeological in
nature. Even though archaeological considerations would shape the answers, environ-
ment, park policy, and the ability to institute and carry out a management program would
be the critical elements. General observational data would be sufficient to deal with the
questions in this category.
Given the questions that had to be answered, the data that had to be found to pro-
vide answers, and what could not be done, there were no real methodological choices
available. Examination of as many sites as possible combined with partly controlled,
nonsystematic collection was the only viable possibility.
The plan was to first visit each known site making general observations. If site
organization or patterning could be determined, this would be used to divide the site
into data units. The data units could be cultural, natural, or a combination of the two,
depending on which were apparent and which were most useful in the context of the
questions asked. Then, a sketch map of the site would be made and each data unit
examined in sufficient detail to decide what information should be collected and how.
Artifact collection was to be the final activity.
Following examination of the known sites, suspected site locations were to be sur-
veyed. It was hoped that park personnel, fishermen, or local divers would be able to sup-
ply this information. If no one knew of possible sites within the park, and if there was
time, areas that might contain sites would be surveyed. Prediction of high-probability
areas would be based on environmental factors and cultural-historical assumptions.
Any sites found would be treated as previously mentioned. All sites would be
located on a map of the park and by cross bearings from identifiable land features. Pho-
tographs, including nonrectified mosaics, would be made where possible and if they
could be expected to aid in recording or interpretation.
As no conservation facilities were available, collection was to be limited to mate-
rials that required no, or minimum, treatment. Should materials have to be removed
from the sites in order to protect them, it was planned to store these in fresh water until
arrangements for conservation could be made.
Plans were made to send a preliminary report on the work to the Park Service
shortly after completion of the project with a more complete report to follow. The pre-
liminary report was. completed 3 months after the fieldwork. It provided a brief descrip-
tion of the sites and contained a number of recommendations for protection of the
wrecks and for future work.

DESCRIPTION OF THE WORK AND FINDINGS

On the afternoon of September 19 Hector Guzman, Sigifredo Marin-Zuniga, and

I arrived in Puerto Limon. We located Eduardo Fernandez, a local diver who had dived
on the known wrecks in the park and others in the vicinity, our purpose being to look at
some of the material he had collected and ask about wreck locations. Mr. Fernandez
showed us artifacts from various wreck sites (Figure 1) and turned over to park person-
Preliminary Investigations of a Shipwreck 459

,_ ...."
~,' \
I \
" I CARIBBEAN SEA
I
I
I
I
/
~ / PUERTO
~~ " VARGAS
,~"
"
I "
I
/
I
I ,
\._ ............-. I \
,
I
\_ .........""'~,
2 \
\""
.......-"

--N

Figure I. Manillas. Upper piece recovered from Brick Area (Area I) site. Complete specimen re-
portedly recovered fTom Brick Area site.

nel an anchor that had reportedly been recovered in the park. These materials will be
discussed later.
No locational information was obtained for any of the wrecks from which Mr.
Fernandez had recovered artifacts. Some of the materials were identified as having
come from the known wrecks in the park. There was only time to make a few photo-
graphs of certain of the artifacts and to make tentative arrangements for a more thor-
ough examination prior to completion of the project. Unfortunately, the return visit was
not possible. Consequently, it is not possible to be certain which artifacts came from
which sites and no complete descriptions or measurements of the material were made.
Mr. Fernandez knew of only the two already known wreck areas within the park.
As he was the most knowledgeable person in the area, the decision was made to con-
centrate our efforts on these two sites. Later conversations with a number of people
about wreck locations turned up only vague, secondhand rumors about additional
460 S.J. Gluckman

,-------------------1
I I
18
I 1I
Figure 2. Map of Punta Cahuita National Marine
park showing wreck areas. Area I is the Brick Area,
L________________
I
I ~
I
I Area 2 is the Cannon Area.

wreck locations. None of these locations was inside the park or even in the immediate
vicinity. We made a trip to Puerto Viejo, a small cacao processing village south of the
park, to ask about sites. It was hoped that "old Port" was old enough to be of interest.
This proved not to be the case and no one we talked to knew of any wrecks.
The two wreck sites are located on the northern side of Punta Cahuita (Figure 2).
Water depth at the wreck sites varies between 1.5 and 5 m. For convenience, the sites
were referred to as the Brick Area (Area 1) and the Cannon Area (Area 2).

Brick Area (Area 1)

The Brick Area consists of a large pile of brick and two cannon. An anchor is
known to be in the area but we were unable to locate it. The brick pile is 19 m long and
9.5 m wide. It is oriented 30° west of north and is about 2 m high at its midpoint. The
brick is nearly stacked on edge and is little displaced by the wreck. At the eastern end
of the pile only one layer of brick remains. Structure, presumably hull planking as it
runs in the long dimension of the wreck, is immediately below the brick. The two can-
non are on this end of the wreck. One of the cannon is lying on sand with wood under-
neath. The other is about 2 m to the east and appears to be off the wreck (Figure 3).
Only a small area of wood was uncovered. This was done by fanning off the 5 to
8 cm of covering sand. The planks are about 13 cm wide and in spongy condition. A
small piece of manilla and a piece of chipped stone were uncovered while fanning. The

t ~ J
N
0

~Or
U ~ #
0

Figure 3. Sketch map of Area 1. Wreck is contained

within the rectangular area {not to scale} and is oriented
30o WofN.
Preliminary Investigations of a Shipwreck 461

manilla was on top of the wood. The chipped stone could have come from on top or
from under the wood.
With the exception of the eastern end, the brick pile is covered with Sargassum.
The southern side is against the reef. The remaining area around the wreck is sandy
with beds of turtle grass (Thalassia testudinum). Visibility averaged 3 to 5 m and water
depth at the base of the brick pile is 4 to 5 m.
About ten person hours were spent underwater examining this site. Overall meas-
urements of the site were made with spare anchor line which was later measured with
the 2-m folding rule. Smaller distances were measured with the folding rule and the
dimensions of the heavily encrusted cannon were taken using pieces of marked string.
Numbered tags, made from pieces of plastic bottle, were tied to the cannon to identify
the measurements.
Slides were made with a Nikonos III underwater camera fitted with a 35 mm lens.
Underwater records were made on plastic slates and later transferred to a field notebook.

Cannon Area (Area 2)

Area 2 is west of the brick area. There are at least 14 cannon and two anchors in
this location. One of the cannon is broken with muzzle and breech portions lying adja-
cent to each other. One of the anchors is also broken with the flukes lying some dis-
tance to the south.
Water depth at the anchors is about 1.5 m. Most of the cannon lie in 4 to 5 m of
water. Underwater visibility in this area was about 2 to 4 m. The bottom here is hard
with numerous coral heads.
The only material, other than the cannon and anchors, seen in this area were a few
bricks like those in Area 1. But, this may relate to the small amount of time that was
spent in this area. Boat and weather problems significantly reduced the time that we had
scheduled to work in this location. For these reasons, less than a complete search was
made here and only a sketch map and measurements on the anchors were made.
The cannon and anchors were heavily encrusted with coral and sea fans. This, plus
the lack of measuring equipment, make it unlikely that the measurements are very accu-
rate. In an attempt to get some idea of the cannon scatter pattern, buoys, made from coco-
nuts, were tied to each cannon and anchor and photographed from the surface.
Unfortunately, a squall that produced heavier than normal swells blew in from offshore
and the coconut buoys moved around too much to make any pattern identifiable.
The following measurements must be considered approximate as the material was
heavily encrusted and proper measuring equipment was not available. For these same
reasons, smaller measurements that were taken, such as trunnion diameter, are not con-
sidered accurate enough to be reported here. Measurements were only made on the two
cannon in the Brick Area and the two anchors in the Cannon Area. Cannon and anchors
are iron. The unmeasured cannon appear to be the same size as those measured. The
two cannon measured were the same size so only one set of figures is given. The same
is true for the anchors. All measurements are in meters.
Cannon
length 2.55
breech to trunnion 1.0
trunnion to muzzle 1.55
462 S.J. Gluckman

muzzle diameter 0.80

Anchor
shank length 3.30
arm tip to arm tip 2.10
diameter of ring
inside 0.35
outside 0.50
shank width (square in cross section) 0.15
palm
width 0.50
length 0.82
Brick
length 0.22
width 0.10
thickness 0.04

The chipped stone artifact is of blue-gray flintlike material. It is not known if the
material is local. It is the rear portion of what could have been a blade. It appears to
have been unifacially flaked and may have been retouched around the edges of the
opposite face. It shows heavy usage along the edges. Its greatest width is 3.6 cm and its
greatest length is 2 cm. It is 1 cm thick.
The piece of manilla that was found is from the midsection of a so-called slave
bracelet. It is 5 cm in length and 0.9 cm in diameter. It appears to be made of copper.
Mr. Fernandez reported finding a clump of 8 or 10 manilla in the Brick Area. The piece
found matches these. These are the standard % round, expanded end bracelet shape.
They are made of solid, cast metal, apparently copper. The greatest inside diameter is
7.2 cm. The opening between the ends is 4.5 cm. The thickness is about I cm with the
thickened, expanded ends being about 1.7 cm (Figure I).
Mr. Fernandez has in his collection one almost complete ceramic vessel and the
base of another, both reportedly found in the Brick Area. Both of these are unglazed,
unslipped earthenware. The paste and finish are much like those of Spanish olive jars
although the vessel walls are not as thick. The smaller, more complete vessel is about
30 cm tall with a flat base and rounded globular shoulders. Two, apparently extruded,
almost luglike handles have been applied above the shoulders. The mouth is about 10
cm in diameter.
The larger vessel, represented only by its basal section, is about twice the size of
the smaller and was probably the same shape. These vessels would appear to belong to
the storage jar category or, at least, are highly reminiscent of ceramics so identified
from a wide variety of land and underwater sites.

INTERPRETATION

The data recovered from the known wreck locations in the park are not very great.
However, the guidance provided by the research design allows these data to be used in a
number of ways. First, a hypothetical model of how the wreck(s) occurred and how the
two locations relate can be proposed. Second, hypothesis concerning the function of the
vessel(s) can be formulated. Third, a time frame for the material can be suggested.
The first question that must be raised about the material examined is the number
of wrecks represented. Given the structural remains and the size of the brick pile, there
can be no doubt that Area 1, the Brick Area, represents at least one vessel (Figure 3).
Preliminary Investigations of a Shipwreck 463

The negative evidence, lack of additional ballast, structural remains, and artifact con-
centrations, argues against this location containing more than a single wreck. A more
complete visual examination of the area combined with a magnetometer survey should
provide confirmation of this. A testable hypothesis can be formulated relative to this
question: If a systematic visual examination and fine-grained magnetometer survey of
Area I and vicinity return no evidence of ballast, structural remains, or artifact concen-
trations, then Area 1 represents the remains of a single vessel. The rejection of this
hypothesis, the finding of any of the stated evidence, would not allow the conclusion
that Area I represents more than a single wreck unless such evidence could be clearly
removed from any relationship to the wreck that is known to be there. Given the
number of variables, and particularly the number of dependent variables possible if the
hypothesis is rejected, testable alternatives cannot be formulated without more data.
Because of human or environmental destruction of evidence, confirmation of the
hypothesis seems unlikely - given the remains that are present. Still, this is a possibil-
ity and should be considered. It also represents a situation of such complexity as to be
presently untestable.
Area 2, the Cannon Area, lacks the remains that allow the interpretation of the
Brick Area as a wreck site. It contains no ballast, no structural remains, and no wreckage
scatter pattern. At least some of these would be expected if this location represented a
wreck site. This is particularly true given the presence of the 14 cannon and 2 anchors,
the hard bottom, coral reef, environment, and shallow water. That is, it is hard to imagine
Ii vessel carrying this number of guns and this size anchors that did not also carry ballast.
If the cannons and anchors were ballast, it is hard to imagine that ship fittings and fasten-
ers plus other artifacts would not be visible on the bottom or trapped around the base of
the coral heads. However, it is also possible to conceive of environmental conditions that
would destroy all evidence of a wreck except cannon and anchors.
Thus, two sets of hypotheses must be formulated and tested with respect to the
Cannon Area. The first set of hypotheses would deal with whether or not Area 2 is a
wreck site. The second set of hypotheses would deal with an interpretation of the mate-
rial if the testing of the first set of hypotheses indicated the location is not a wreck site.
The first hypothesis to be tested can be stated as follows: If a systematic visual
examination and fine-grained magnetometer survey do not return evidence of ballast,
structural remains, or artifact concentrations, then Area 2 is not a wreck site unless it
can be determined that such evidence has been destroyed.
As there are only two possible ways that the evidence of a wreck could have been
destroyed, salvage or environmental conditions, the remaining hypothesis in this set
must deal with these. A salvage effort that left cannon and anchors while removing all
other remains is too unlikely a possibility to require testing. Environmental conditions
that destroyed all evidence except cannon and anchors are possible but represent a situ-
ation with too many variables to be adequately tested.
Because of this, confirmation of the first hypothesis in the set is not sufficient to
reject Area 2 as a wreck location. Rejection of the first hypothesis, the finding of any of
the stated evidence, would allow the determination of Area 2 as a wreck location.
Therefore, confirmation of the hypothesis would require the formulation of an
alternative theory in order to explain the presence of the material in Area 2, that is, an
alternative to the theory that the material represents a wreck site. The alternatives would
seem to be limited to four: the cannon and anchors were lost here while being transported
464 S.J. Gluckman

to or from shore; the material was dumped here for unknown reasons; the cannon and
anchors were jettisoned here to prevent a vessel being driven onto the reef; the materials
are part of the wreck in Area 1. A possible expansion on any of the first three alternatives
must be mentioned. Two people from the village of Cahuita stated that they had heard
from grandparents about a battle between two ships with the result being the cannon and
anchors on the bottom. Given the shallow water and the reef, this seems unlikely. It
seems more reasonable to accept this as part of the local oral tradition and probably
means that local people have known about this material for at least three generations.
It is not possible to formulate testable hypotheses relative to these alternative
theories solely with the available archaeological data. The last one may be an excep-
tion. The last alternative, the two sites are part of the same vessel, can explain the
observed data. Only historical records are likely to make the other three alternatives at
all viable as explanations.
One can postulate a vessel being driven onto the reef at the location of the Brick
Area where it goes aground with its upper deck, or decks, at or above the water line.
One can further postulate that during the processes of breaking up, the portion of the
vessel above water floats freely and carries the cannons and anchors with it to Area 2.
If this floating platform tips to port or starboard on hitting the reef at Area 2, the guns
on that side would be spilled overboard. The weight of the remaining guns would cause
overrecovery in the opposite direction, perhaps even causing this raftlike affair to turn
completely over, spilling the remaining cannons and anchors into the sea. With the
weight removed, the deck could have broken up on the reef or floated ashore or out to
sea. If there were a few bricks on deck at the time it broke free and floated to Area 2,
we have a scenario or model that accounts, or could account, for what has been
observed in the Cannon Area.
Additional corroboration for such a model comes from several sources. The
largely intact configuration of the pile of brick argues for a vessel that went aground
rather than one that was battered against a reef. The surface current model for the area
suggested by Wellington shows a flow from Area I to Area 2. And, it was not uncom-
mon for large sections of a broken-up sailing vessel to be carried some distance away
from the site of the sinking.
The testing of this model will not be an easy task. But the process of testing
should tell us a great deal about how the wreck occurred and could settle the question
about the number of wrecks in this part of the park. Testable hypotheses about the
breakup of the vessel in Area 1 could· be formulated from analogies based on the
breakup of other similar vessels. Hypotheses could also be drawn from knowledge of
construction techniques and methods if these were examined from the perspective of
how such a vessel might break up. Lastly, hypotheses could be developed from repli-
cating the sinking with a scale model of the Brick Area vessel.
Hypotheses concerning what materials might have been on deck when the vessel
began to break up and how this material might aid in testing other elements of the
model; and hypotheses related to what other material should be found in Area 2, if the
model is correct, would also help test part of the model.
Until additional archaeological and historical research is undertaken, the hypothe-
ses necessary to test this model cannot be effectively formulated. But, the outcomes
needed for testing are known. Consequently, the model, and previously stated hypothe-
ses, can become an important part of the research design for future work.
Preliminary Investigations of a Shipwreck 465

The data acquired during the preliminary work allow some additional specula-
tions. These can also provide important guidance for future research.
Several alternative models can be suggested concerning the function of the vessel
in Area 1. The size of the pile of brick at this location makes it seem unlikely that this
material was ballast. It is also generally assumed that economic factors argue against
the shipment of brick over any great distances. This would mean that either the brick
was not moving any great distance or that special economic incentives obtained with
respect to this load of brick. It is also possible, of course, that this vessel represents the
remains of a special and idiosyncratic behavior pattern or event. If the latter is the case,
only comparative research showing this to be a unique situation or historical records
describing the situation will be very helpful as explanatory devices. Either of the other
alternatives could be examined through reference to general behavior patterns related
to the shipment of brick.
The presence of manillas in the Brick Area wreck adds an interesting dimension
to the interpretation of the wreck's possible function. Manillas, which are not well
reported in the archaeological literature, are intimately connected with the slave trade
(Quiggin, 1949:89-91). Quiggin points out that they are apparently indigenous to west
Africa where they were long used as a medium of exchange. By the end of the 15th
century, from the Gold Coast to Lower Niger, they were the dominant currency of the
slave trade. At that time their value was 8 or 10 per slave.
The value of manillas as well as their size, style, and material fluctuated widely
though time. By the middle of the 1800s tons of them were made from copper, tin, and
pewter. These would "ring" properly when struck. Other materials were used including
cast iron. The latter were rejected by the Africans as valueless.
Given the dominant function of manillas and their locus of manufacture in Eng-
land, one would not expect to find many in the New World. Their almost complete
absence from the archaeological literature would seem to support this contention. It
might be expected that a few would show up that had not been traded in Africa or that
had been kept as souvenirs or lost aboard slave ships. It also seems possible that some
could have been traded to Indians in the New World.
Possibly ten manillas may have been recovered from the Brick Area wreck. This
association between manillas and a cargo of bricks could mean that the vessel had
some involvement with the slave trade, but a mixed cargo of slaves and bricks does not
seem to make economic sense. Also, the economics of shipping brick plus the bricks
themselves make it almost certain that the bricks were not being shipped from the Old
to the New World.
The bricks are about the same size and shape as the 18th-century Dutch bricks
found throughout the New World. However, their quality is significantly below that of
Dutch brick. They show evidence of some indifference in manufacture; the firing was
not very well controlled and was done at low temperatures. The bricks are soft and
show numerous firing clouds. The molding is sloppy and the clay is incompletely
ground; unground lumps of clay are clearly visible on all surfaces.
One almost inescapable conclusion is that this is not high-quality brick. Shipping
brick to the New World makes little economic sense, unless the brick is shipped as bal-
last. And, some 280 m3 of brick seems like a lot of ballast. The shipment of low-quality
brick to the New World would seem difficult to justify. On this basis, it seems logical
to suggest that the brick is of New World manufacture.
466 S.J. Gluckman

On the basis of the above reasoning, several possible models can be postulated
for the function of the Brick Area wreck. First, the vessel could have brought slaves to
the New World and picked up a short-haul load of brick which would then have been
replaced with a more economically viable cargo for shipment to the final port of call.
Second, the vessel could have been a coastal tramp involved in short-hauling bulk car-
goes. In this situation, presumably, a cargo of brick would not be uneconomical. A
variation of this would be a privately owned vessel hauling brick for private use.
The single manillas thus far found on the wreck and the small number reportedly
found there, could fit any of the postulated models. The piece of chipped stone tool, if
it came from on, rather than below, the wreck, could mean a relationship between the
vessel and Indians, possibly as part of the crew. This would fit best with a pattern of
local or regional use for the vessel rather than one of an international carrier.
Another possibility is that the stone artifact is part of a flint and steel or strike-a-
light. The edge wear pattern is consistent with this interpretation. If the material should
prove to be nonlocal, particularly if it should prove to be European in origin, support
would be added to this interpretation.
It is also possible that neither the manillas nor the chipped stone artifact have
anything to do with the function of the vessel. And, of course, if the stone artifact came
from under the wreck and is of Indian manufacture, the vessel might be lying on a sub-
merged site of former human habitation.
With the small amount of information available at this point, the postulated mod-
els are mainly useful in providing guidance for future work. They strongly suggest that
questions about the economics of shipping brick must be examined. They also bring up
questions about the differences between long-haul and short-haul vessels. And, they
point to a number of areas where the formulation of testable hypotheses will be valu-
able in setting the direction of future work on the wreck.
The age of the remains in the two areas is an important factor in interpretation in
that the use period of the vessel(s) would make certain assumptions more or less likely.
There are very little data available from which a use or loss period can be established.
The cannon and anchors are potentially datable but this cannot be done with the infor-
mation in hand. The brick can potentially be tied to a period yet the possibility that they
are of local or regional manufacture might also mean that they do not fit the presently
known patterns. It should be possible to date the manillas; two sources, neither particu-
larly reliable, seem to suggest that these date from the middle 1700s. One shown for
sale in the World Treasure Society's December 1978 catalogue is about half the size of
those recovered from the Brick Area wreck. It is identified as having come from the
wreck of the "schooner Duoro that sank near England's southeast coast in 1843." A
photograph in a National Geographic Magazine article (Peterson, 1977:725) shows
manillas that appear to be the same size as those recovered. These came from a 1740s-
period wreck off Bermuda. Given the normal pattern for manufactured mediums of
exchange, it would not be unexpected to find that manillas became smaller through
time. Unfortunately, neither the above date nor the previous assumption can be given
any great credence due to the evidence on which they are based.
The last potentially datable items are the ceramics recovered by the diver from
Puerto Limon which he said came from the Brick Area wreck. These have not yet been
identified. A preliminary check of the literature served only to indicate that they are
somewhat similar to a number of unglazed utility wares in paste. Nothing of similar
Preliminary Investigations of a Shipwreck 467

shape in this vessel size has been located. The slides have been seen by a number of
knowledgeable people. None of them were able to suggest a date or country of manu-
facture. It is possible, that, like the brick, they are of local or regional origin.
Unfortunately, confirmation that the vessels were found on the Punta Cahuita
wreck is not available. Thus, even if the material can be dated, questions will remain.

CONCLUSION

As is often the case, this preliminary work has produced more questions than
answers. These questions clearly indicate that the potential for significant data recovery
is present. These remains also have great potential for use in the park's interpretive pro-
gram. The brief investigation produced enough information to set the needed directions
of future research. It is hoped that more work will be possible in the future.

ACKNOWLEDGMENTS

I would like to thank all of the following people for the help they provided in
organizing and carrying out the work described in this report. Dr. Carlos E. Valerio,
Department of Biology, University of Costa Rica, was mainly responsible for setting up
this project. Funding was provided by the University of Costa Rica. Alvaro F. Ugalde,
director of the Servicio de Parques Nacionales, and Sigifredo Marin Zuniga, adminis-
trator of the Parque Nacional Cahuita, were both very helpful. Jorge Campos, Hector
Guzman, Quin Rodriquaz Toledo, and Ricardo Soto Soto made up the diving team.
They were an excellent crew and deserve special thanks. Mr. George Ballo and Ms.
Beth Horvath, graduate students in the Department of Anthropology at the University
of South Florida, prepared the figures for this report.
Without the help of these people, the work could not have been accomplished.

REFERENCES

Anonymous, 1976, Sailing directions (Enroute) for the Caribbean Sea. Defense Mapping Agency. Hydro-
graphic Center, Publication 144. Washington, DC.
Anonymous, 1978, Los Parques Nacionales de Costa Rica. 1977 Servicio de Parques Nacionales, Ministerio
de Agriculture y Ganaderia, San Jose, Costa Rica.
Boza, M.A., 1978, Los Parques Nacionales de Costa Rica. Servicio de Parques Nacionales, Ministerio de
Agriculture y Ganaderia, San Jose, Costa Rica.
Peterson, M., 1977, Reach for the New World. National Geographic Magazine, pp. 724-767.
Quiggin, A.H., 1949, A Survey of Primitive Money. Methuen, London.
Wellington, G.M., n.d., An Ecological Description of the Marine and Associated Environments at Monumento
Nacional Cahuita. Unpublished manuscript on file with the Servicio de Parques Nacional, San Jose,
Costa Rica.
 Part VII.B

Sampling
".~i).:
~

':.i
-r/: ..

Spatial analysis of the Kennemerland site required quantative methodology. Muckel-

roy's analysis, in Chapter 42, is a good starting point for the statistical investigation of
a scattered site and evaluating the scattering factors involved with a ship's sinking.

469
 Chapter 42

'.~~.' The Analysis of

~

o i' .
Sea--Bed Distributions
q;,O_" .': .
.j'~l Discontinuous Sites

KEITH MUCKELROY

In the course of the last section [of Maritime Archaeology], the sophistication of the
methods of analysis rose as the degree of reordering of the material on the seabed in-
creased, and as the significance of the observed distributions became intuitively less
apparent. In the present section, this process moves on a stage further, because with dis-
continuous sites there has been not only a considerable degree of reordering, but also
the loss of any defining structure within which to consider the remains. The distribu-
tions on sites such as those at Yassi Ada could be directly related to the framework of
the structural remains within which they were formerly contained, and even with the
more obscure type of situation such as that represented by the Trinidad Valencera, the
assumption could still be made that the scatter was related to a single nucleus repre-
senting the remains of the vessel. This is no longer true in a situation where the ship
has broken up over a considerable distance, or where the seabed has presented greatly
varying conditions for the preservation of remains within the area of the wreck site. In
this total absence of any defining framework, a discontinuous wreck site is fundamen-
tally different from nearly all other archaeological situations; perhaps the nearest anal-
ogy on land is a midden site.
This presentation of some appropriate modes of analysis is concerned exclusively
with the Kennemerland wreck site, on which such studies have proved particularly
worthwhile .... As has already been described, the overall distribution of these remains
Source: Maritime Archaeology, New York: Cambridge University Press, 1978. Copyright 1978 by Cambridge
University Press. Reprinted with permission of Cambridge University Press.

471
472 K. Muckelroy

reflects clearly the breakup of the vessel and the assimilation of some of its elements
into the seabed, so there are a priori grounds for suspecting that a similar level of sig-
nificance may lie in the details of these distributions. It is important to be satisfied that
such significance is at least conceivable before embarking on any detailed analysis, as
no amount of statistical manipulation can extract information from totally scrambled
data. Determining that significant patterning is possible does not, of course, guarantee
that it will be present, let alone detectable, but it does oblige the archaeologist to search
for that pattern, and to so order his excavation and recording processes that the infor-
mation is available for such a study.
In the case of the site in the South Mouth, discontinuity is imposed not only by
the nature of the remains, but also by the limited extent of the areas so far excavated.
These have been concentrated in only one part of the site, albeit the central area, in
which a good deal of the wreckage was deposited when the foundering ship struck the
surf line (see Figure 7 of Chapter 25). Furthermore, within this area, only certain sec-
tions have been systematically excavated, as shown in Figure I. These seven excava-
tion sites have sampled the main types of seabed in this area, including open areas of
shingle and boulders (Sites A, C, and G) and enclosed gullies (Sites B, D, E, and F).
Similarly, they have sampled a range of deposits, including large expanses of iron con-
cretion (in Sites C, E, and F) and matted organic remains (Sites F and G; see Figure 2)
as well as clean shingle. Any results emerging are accordingly highly provisional. Nev-
ertheless, the analysis of this information has been proceeding since the end of the first
two seasons of excavation (1974), both as a check on the suitability of the excavation
methods and as an indication of the kind of questions to tackle in subsequent seasons; a
preliminary discussion of some of the initial studies and their findings has already been
published (Muckelroy, 1975).

Figure 1. A plan of the specific areas excavated within the main excavation zone on the Kennemer,
land site.
The Analysis of Discontinuous Sea-Bed Distribution Sites 473

o Orgamc depoSit

-7 o Loose splinters

• Tarred deposit

-6 III Iron concretion

l~: ~ I Wood

o 2 3
metres

Figure 2 . Plan of the main features within sites F and G on the Kennemerland site.

Even during the excavation seasons, there were some seabed associations that
were obviously significant. A particularly clear example concerned a collection of per-
sonal possessions (jewelry, thimbles, eating utensils, a small pile of coins, etc.) concen-
trated in one area of Site F (Figure 2: "P"). These, coming in quantities too great for
purely personal use, but too small for Company trading, suggest a spot of (illegal) pri-
vate trading among the contents of a private chest (Price and Muckelroy, 1974:262). In
fact, the whole of Site F, and the western part of Site G, was notable for the concentra-
tion of such items, perhaps reflecting the part of the ship that ended up in this area
close to the western shore of the South Mouth. Within Site G, the almost complete
absence of any finds in the eastern part of the site, despite their profusion in the west-
ern half and in Site D farther east, is equally important for present purposes; because
the preservational characteristics of this area were apparently identical to those else-
474 K. Muckelroy

where, the paucity of finds must reflect some other factor, presumably related to the
process of wrecking. Another marked feature of the distributions that was quickly
appreciated was the concentration of lead shot of all types in Site A (Price and Muckel-
roy, 1974:262). However, a wide range of other items appeared to be irregularly scat-
tered over all of the sites, with no patterns immediately apparent.
As a first stage in the analysis, and in order to quantify the variations between
sites intuititively felt during the fieldwork, the incidence of different classes of finds in
the seven sites was tabulated, in each instance giving the total number of that class in
that site, their density (i.e., items per square meter), and the percentage of the totat
number of that class of find recovered from that particular site (see Table I). In every
case, the unit of measurement was the smallest identifiable entity, on the grounds that
any other approach involved unjustifiable assumptions of relative importance. There
was thus no a priori weighting, a procedure that, in such contexts, is generally regarded
as undesirable (Sneath and Sokal, 1973: 109-113). The classes being considered were
as follows:
1. Stoneware potsherds, all apparently from Bellarmine flagons (unit: a sherd).
2. Green bottle glass, from square-based glass bottles (unit: a piece of glass).
3. Pewter bottle tops, from the bottles of class 2 (unit: a piece of pewter identifi-
able as coming from atop).
4. Clay pipes, subdivided into:
(a) pieces of pipe stem,
(b) pieces of pipe bowl,
(c) all fragments of clay pipes.
5. Lead shot, larger than 0.001 m in diameter, i.e., not scatter shot.
6. Personal possessions (jewelry, thimbles, eating utensils, items of clothing,
etc.).
7. Items associated with the ships armament.
8. Fragments of bronze sheeting.
9. Bronze nails.
10. Fragments of lead.
11. Bones.
12. Pieces of rope.
13. Glazed earthenware potsherds (unit: a sherd).
Together, these classes embraced over 80% of all of the finds made on this site.
While considerable variations between sites within each class are apparent in
Table 1, it is difficult to make comparisons between the various elements in it, and
impossible to bring out any patterns. A simple way of clarifying such a tabulation is to
lump together certain units, according to a common feature, so as to bring out the effect
of that feature on the overall figures. In Table 2 this involved aggregating all of the
open sites (i.e., A, C, and G) and all of the enclosed, gully sites (i.e., the remainder), in
order to inspect any general trends. While confirming that enclosed sites had a consis-
tently higher density of finds, a fact appreciated during the excavation seasons, it also
served to refute another suspicion, namely, that there was a distinction in the relative
survivals of clay pipe stems (class 4a) and the more fragile pipe bowls (class 4b)
between exposed and protected areas. A similar aggregation bringing together finds
from different types of deposit demonstrated no noteworthy trends.
The Analysis of Discontinuous Sea-Bed Distribution Sites 475

Table I. The incident of artifacts in the seven main excavation

sites on the Kennemerland wreck site

Sites A 8 C D E F G Total
Areas (m 2) 66 4 8 4 4 12 19 117
Class I
n* 185 31 16 33 55 319 313 952
d 3 7.8 2 8.3 13.8 24.5 16.5 8.1
P 19 3 2 3 6 34 33 100
Class 2
n 91 134 118 59 72 379 110 963
d 1.4 33.5 14.8 14.7 18 31.5 5.8 8.2
p 9 14 12 6 7 39 II 100
Class 3
n 9 5 11 4 7 41 3 80
d 0.1 1.3 1.4 I 1.8 3.5 0.2 0.7
P II 6 14 5 9 51 4 100
Class 4"
n 21 20 8 18 25 117 167 376
d 0.3 5 I 4.5 6.3 9.8 8.8 3.2
p 6 5 2 5 7 31 44 100
Class 4b
n 2 8 I 5 25 39 81
d 0.1 2 0.1 0.3 1.3 2 2.1 0.7
p 2 10 6 31 48 100
Class 4'
n 23 28 9 19 30 142 206 457
d 0.3 7 1.1 4.8 7.5 11.8 10.8 3.9
p 5 6 2 4 7 31 45 100
Class 5
n 1594 0 o 2 I 46 I 1644
d 24.8 0 o 0.5 0.3 3.8 0.1 14.1
P 97 0 o 0.1 0.1 3 0.1 100
Class 6
n 13 2 3 10 4 157 25 214
d 0.2 0.5 0.4 2.5 I 13 1.3 1.8
P 6 1 1 5 2 73 12 100
Class 7
n 9 o o o o o o 9
d 0.1 o o o o o o 0.1
p 100 o o o o o o 100
Class 8
n 92 0 o 4 o o 1 97
d 1.4 0 o I o o 0.1 0.8
p 95 0 o 4 o 0 100
Class 9
n 389 0 o o 19 410
d 5.9 0 o o 0.3 1.6 0.3 3.5
p 94.8 0 o o 0.3 4.8 0.3 100
Class 10
n 7 o 520 6 I 21
d 0.1 o 0.6 0.5 0 0.5 0.1 0.2
p 33.3 0 23.8 9.5 0 28.6 4.8 100
(Continued.)
476 K. Muckelroy

Table I. (Continued.)

Sites A B C D E F G Total
Class II
n 5 3 0 0 20 20 49
d 0.1 0.8 0.1 0 0 1.7 0.4
P \0 6 2 0 0 41 41 100
Class 12
n 0 0 7 0 0 8 6 21
d 0 0 0.9 0 0 0.7 0.3 0.2
P 0 0 33.3 0 0 38.1 28.6 100
Class 13
n II 0 22 2 I 42 57 135
d 0.2 0 2.8 0.5 0.3 3.5 3 1.2
P 8 0 16 31 42 100
·n. number of items; d. density (items per square meter); p. percentage within
each class.

Table 2. The incidence of anifacts on the Kennemer-

land wreck site. contrasting open and enclosed sites·

Sites, areas (m2) Open, 93 Enclosed, 24

Class I
n 514 438
d 5.5 18.3
P 54 46
Class 2
n ~19 644
d 3.4 26.8
P 33 67
Class 3
n 23 57
d 0.3 2.4
P 29 71
Class 4"
n 196 180
d 2.1 7.5
P 52 48
Class 4b
n 42 39
d 0.5 1.6
P 52 48
Class 4c
n 238 219
d 2.6 9.1
p 52 48
Class 5
n 1595 49
d 17.2 2
p 97 3
Class 6
n 41 173
d 0.4 7.2
P 19 81
The Analysis of Discontinuous Sea,Bed Distribution Sites 477

Table 2. (Continued.)
Sites, areas (m 2) Open, 93 Enclosed, 24
Class 7
n 9 0
d 0.1 0
P 100 0
Class 8
n 93 4
d I 0.2
P 96 4
Class 9
n 390 20
d 4.2 0.8
P 95 5
Class 10
n 13 8
d 0.1 0.3
P 61.9 38.1
Class II
n 26 23
d 0.3
p 53 47
Class 12
n 13 8
d 0.1 0.3
P 61.9 38.1
Class 13
n 90 45
d I 0.9
P 67 33
• Abbreviations as in Table I.

In order to eliminate the effects of the different sizes of the excavation sites, as
well as to allow a more detailed investigation of the distributions, a reasonable proce-
dure would seem to be to impose a grid of meter squares over the sites, as in Figure 1.
This size of square was the smallest justified by the degree of accuracy actually
achieved in the recording underwater. They were arranged so that the total number of
squares on each site equaled the total area (in square meters) of that site, while at the
same time ensuring that every artifact could be placed in one square; the parts of each
site thus excluded from the grid were those that were sterile. Any items falling on the
boundaries between squares were placed in the numerically lower square (i.e., the one
nearest the relevant point of origin) on the grounds that a measurement taken by tape is
more likely to overestimate a distance than underestimate it, as the tape can be slack
and not straight, but it cannot be overstretched. Class·es 7 (items of armament), 10 (lead
fragments), and 12 (pieces of rope) were omitted from these further stages of analysis,
as they were too small to yield significant results. Of the 117 squares, 20 were also
excluded from further consideration because they contained no finds from any of the
classes under consideration; these squares are indicated in Figure 3 by underlined num-
bers. The number of items from each class in the remaining 97 squares is given in
Table 3, following the numeration of Figure 3.
478 K. Muckelroy

Table 3. The number of items from each of 12 classes occurring in each meter square within the
excavation sites on the Kennemerland wreck site*
Class 2 3 48 5 6 8 9 II 13
2 13 70 2 17 2 19 o 3 o I o o
3 13 18 6 7 o o o o o
4 14 18 2 6 I 7 o 2 o o o o
5 14 18 2 6 2 8 I 2 o I o o
6 14 18 2 7 I 8 o o o o o o
7 100 91 10 14 9 23 o 3 o o o 3
8 22 o 5 o o 0 o o o o o o
9 I 10 8 o o 0 I 6 o o 2 4
10 4 2 o o o o 2 o o o o
11 8 14 4 o 4 0 2 o o o
12 50 57 10 20 9 30 21 II o o o 7
13 4 10 6 o 5 0 I o o o 2
14 8 15 4 o 5 0 2 I o o o
15 8 33 5 2 7 0 o o o o
16 8 34 2 5 2 7 0 2 o o o
17 87 46 2 22 5 27 25 100 o 9 4 14
18 3 10 o o o o o o 4
19 8 15 5 I 5 I 3 I o o I
20 7 33 5 2 7 0 o o o o o
21 8 34 5 2 7 0 2 o o o
22 34 67 o 31 o 31 0 29 o 9 14 5
23 4 14 o 6 o 6 0 2 o o o 3
24 9 15 I 5 o 5 I 3 I o o I
25 2 o o o o o 0 o o o o o
26 o 4 o o o o 0 o o o o o
27 o 5 o I o I 0 o o o o o
28 o o o o o 0 o o o o o
29 o 4 4 o o o 0 10 2 o o o
30 2 o o o o o 0 o o o o o
31 4 o o o o o 0 o o o o o
33 6 2 o o o o 0 o o o o o
34 o 7 o o o o 10 I 3 6 o o
37 7 7 o I o I 0 o o o o o
38 o 5 o o o o o o o o o
39 o o o 5 o 5 o 2 6 o o
40 o o I 4 o 4 12 I o o o
42 3 6 o 2 o 2 I o 2 o o o
43 3 5 o o o o 0 I o o o
45 o I o o o o 0 o o o o o
46 o o o o o o 23 o I 3 o o
47 7 5 o 2 o 2 7 o o 3 o 3
49 3 o o o o o 0 o I o o o
50 15 6 3 o 3 71 o o o 3 o
51 17 4 o o o o 41 o 4 95 o
52 6 o o o o o 0 o o o o o
53 o o o o o 0 o o o o o
54 10 I o I o I 54 o 5 4 o o
55 4 o o o o o 142 o 25 o I I
56 I o o o o o 120 o 3 I o o
57 o o o o o o 218 o o o o o
58 6 I o o o o 30 o I o o o
59 7 I o o o o 100 o 2 5 o I
The Analysis of Discontinuous Sea-Bed Distribution Sites 479

Table 3. (Continued.)
Class 2 3 4' 4' 5 6 8 9 II 13
60 5 3 o o o o 67 o o 13 o o
61 o 3 o o o o 28 o 9 II o o
62 7 2 I o o o 0 o o 2 o o
63 6 o o o o o 6 o o 2 o
64 30 o o o o o 83 o I 18 o o
65 2 o o o o o 30 o o o o o
66 6 3 o o 2 2 30 o 10 28 o o
68 I o o o o o 6 o o 8 o o
69 o o o o o o 60 o o o o o
70 12 3 o o o o 6 o 4 o o I
72 3 o o o o o 8 o o 8 o o
74 o 3 I I o I 0 o o o o o
78 2 o o o o o 0 o o o o o
80 3 o I o 225 o 14 35 o o
82 7 o o o o 213 o 2 69 o 3
84 o o o o o 0 o o o o o
85 o I o o o o 0 o o o o o
86 o o o o o 0 o o 71 o o
88 2 o o o o o 0 o o o o o
91 3 4 o 2 o 2 0 I o o I o
92 6 9 o 3 I 4 o o o o o
93 3 3 o 7 3 10 0 I o o o o
94 50 o o 2 3 0 o o o o o
95 89 20 I 18 o 18 0 6 o o 2 5
96 2 8 o 4 3 7 0 4 o o I 8
98 I o o o o o 0 o o o o o
99 5 8 o 25 9 34 0 o o o o 2
100 24 IS o 28 29 0 o o 5 6
101 24 6 o 5 3 8 0 o o o 2
102 93 24 o 38 II 49 0 9 I o 20
\03 6 3 1 14 4 18 0 I o o 6 II
104 o o o o o o 0 o o o I o
105 o o o o o 0 I o o o o
106 I o o o o o 0 o o o o o
107 3 o o 4 2 6 0 o o o o o
108 I 5 o 16 3 19 0 I o o 3 2
109 2 5 o 5 o 5 0 o o o I I
110 2 18 2 5 o 5 0 I o o o 6
III I 5 2 o o o 0 o o o o 5
112 2 28 o o o o 0 o o o o I
113 3 21 3 3 I 4 0 2 o o o 8
114 13 o o o 0 o o o o 2
115 4 13 2 o o o 0 o o o I o
116 3 9 o o o 0 o o o o o
117 o II o o o o 0 o o o o o
"For the location of the squares, see Figure 3.
480 K. Muckelroy

2 11 E I I I I
3 4 5 6
I
7 8 14
D
9 10 19 C 110 111

[ffiB
F t-~--t

12 13 24 112 113 114 115

17 18 116 117 B 20 21
r---

22 23 91

92 93 25 26 27 28 29 30

94 95 96 97 98 99 100 31 32 33 34 35 36
- - - -
101 .102 103 104 105 106 37 38 39 40 41
-
107 108 109 42 43 44 45 46
-
G
47 48 49 50 51 52
-
53 54 55 56 57

58 59 60 61

62 63 64 65

66 67 68 69
-
70 71 72 73
- -
74 75
-
A 76
- -77
78 79
-
80 81
-
82
-83
84 85

86 87 88
-
89
~
90
0.......=.-
Figure 3. Key to the grid square numbers, Kennemerland site (for locations, see Figure I).
The Analysis of Discontinuous Sea,Bed Distribution Sites 481

Table 4. Conventional designation of the four possibilities of

presence/absence relationships between two classes in a matrix

Class X
ClassY Present Absent
Present a b
Absent c d

Subjectively, it is apparent that there is a strong patterning in the distribution of

every class across the 97 squares, an impression confirmed by the extremely high val-
ues resulting in every case from a calculation of an Index of Dispersion (Greig-Smith,
1964:63; Hodder and Orton, 1976:34), lying in the range 373.4 (class 3) to 11224.3
(class 5). Further analysis thus seemed appropriate. The ordering of the squares is obvi-
ously arbitrary, and the distances between them vary, so that the problem of determin-
ing any pattern within these data in fact resolves itself into a problem of correlation
between the 12 classes. For such an analysis, a series of techniques are available within
the discipline of numerical taxonomy (Sneath and Sokal, 1973), some of which will be
explored below.
A common approach to such a problem would be to simplify the data into a table
of presence and absence for each class in each area, recording presence for each occa-
sion in Table 3 when the score is other than zero, and absence when it is zero. A coeffi-
cient of correlation can then be produced between every pair of classes relating the
number of squares in which they achieve the same state to the number in which they
achieve opposite states (Sneath and Sokal, 1973:129-137; Hodder and Orton, 1976:
201-204). Considering classes X and Y in Table 4, this means relating the number of
squares in which they are both present (a) or both absent (d) to those in which one is
present and one absent (b and c). Such an analysis was attempted on the first batch of
data, but the results were not encouraging, as the simplification to presence/absence
states minimized the impact of classes which were highly concentrated in a small
number of squares, and the introduction of an abundance state in addition to presence
and absence did not totally solve the problem (Muckelroy, 1975:178--185). However,
the study showed clearly that with such material, the Simple Matching Coefficient, cal-
culated using the formula (a + d)/(a + b + c + d) (see Table 4), performed badly as it
registered unreasonably high scores between rare classes whose only similarity lay in
the facts that they were mutually absent from most of the site and that the Coefficient
of Jaccard, calculated using the formula a/(a + b + c) and thus totally discounting the
effects of d (mutual absence), was much more satisfactory. Because this characteristic
of some rare and some common classes is likely to be typical of many archaeological
situations, this conclusion will be widely applicable. The similarity matrix calculated
using the Coefficient of Jaccard for this simplification of Table 3 is given in Table 5;
general inspection of it readily shows that there remains an unfortunate tendency for
large classes to achieve high scores with each other.
While such presence/absence methods have the advantage of simplicity and can
be applied (albeit laboriously) without the aid of a computer, their usc inevitably
involves the loss of a good deal of information; if precise figures, such as those given
in Table 3, are available, they should, all other things being equal, be utilized. A sim-
482 K. Muckelroy

Table 5. Similarity matrix calculated on the Coefficient of Jaccard for the presence/absence
simplification of the Kennemerland distribution data in Table J
2 3 4" 4b 4· 5 6 8 9 II 13
I +
2 0.64 +
3 0.37 0.44 +
4" 0.51 0.57 0.44 +
4b 0.33 0.34 0.32 0.54 +
4· 0.51 0.64 0.43 0.98 0.55 +
5 0.30 0.28 0.15 0.19 0.11 0.21 +
6 0.34 0.43 0.48 0.50 0.38 0.49 0.14 +
8 0.21 0.23 0.11 0.17 0.06 0.18 0.48 0.17 +
9 0.24 0.21 0.07 0.15 0.11 0.16 0.48 0.12 0.37 +
11 0.23 0.23 0.20 0.27 0.25 0.27 0.10 0.28 0.02 0.08 +
13 0.40 0.46 0.28 0.39 0.29 0.39 0.20 0.39 0.17 0.14 0.28 +

Table 6. Analysis of Kennemerland data; similarity matrix, using a Euclidean distance measure
2 3 4" 4b 4· 5 6 8 9 11 13
I +
2 2.314 +
3 3.103 2.508 +
4" 2.087 1.442 2.137 +
4b 2.054 1.814 2.290 1.135 +
4· 1.941 1.599 2.092 0.301 0.831 +
5 2.773 2.786 3.187 2.350 2.099 2.370 +
6 2.295 1.912 1.697 1.487 1.132 1.410 1.699 +
8 2.963 2.831 3.050 2.120 2.107 2.230 1.739 1.397 +
9 2.763 2.865 3.294 2.206 2.135 2.338 1.417 1.670 1.523 +
11 2.673 2.570 2.968 1.752 1.382 1.796 1.938 1.334 2.091 2.035 +
13 2.575 2.064 2.629 1.578 1.528 1.877 2.054 1.162 2.014 2.046 1.564 +

Table 7. Analysis of Kennemerland data; similarity matrix, using product-moment correlation

2 3 4" 4b 4· 5 6 8 9 11 13
I +
2 -0.014 +
3 -0.177 -0.034 +
4" -0.119 0.138 -0.068 +
4b -0.019 -0.024 -0.087 0.086 +
4· -0.043 0.029 -0.050 0.750 0.362 +
5 -0.103 -0.236 -0.223 -0.381 -0.261 -0.376 +
6 -0.181 -0.105 0.280 -0.311 -0.080 -0.192 -0.127 +
8 -0.180 -0.256 -0.165 -0.250 -0.300 -0.298 0.180 0.046 +
9 -0.097 -0.274 -0.267 -0.296 -0.294 -0.360 0.341 -0.123 0.275 +
11 -0.123 -0.217 -0.208 -0.124 0.060 -0.127 0.003 -0.053 -0.114 -0.059 +
13 -0.142 -0.012 -0.112 -0.075 -0.098 -0.269 -0.097 0.052 -0.108 -0.103 0.048 +
The Analysis of Discontinuous Sea-Bed Distribution Sites 483
plified data set was explored in the preliminary study, using the Robinson-Brainerd
index of similarity (Robinson and Brainerd, 1951), and this served to show the greater
potential of numerical analysis (Muckelroy, 1975:181-185). However, by harnessing
the greater speed and power of a computer, more complex statistics can be investigated,
and a wide range of possible approaches tested. This has been done, using the data pre-
sented in Table 3, making use of the CLUSTAN package. In order to suppress the
influence of variations in the sizes of the classes, it was found necessary to express all
scores as percentages for each class, and then to standardize these percentages.
The first requirement in such a study is the production of a similarity matrix for
the 12 classes, similar to that given in Table 5. For this, two widely used correlation
coefficients have been tested: a measure of Euclidean distance and the product-
moment measure of similarity. The matrix based on the former measure is given in
Table 6, while that derived from the latter is shown in Table 7. Inspection of these
shows that there are only minor differences between them, and there is no reason to
prefer the one before the other. However, the Euclidean distance measure might be
considered the more suitable generally, as it allows a greater range of subsequent clus-
tering exercises. Eight of these clustering procedures from the CLUSTAN package
were applied in this instance, of which only the first four could be applied to the prod-
uct-moment correlation results. The dendrograms produced by five of the more satis-
factory clusterings are given in Figures 4-8.
The criteria applied in deciding on whether a given procedure had produced a sat-
isfactory pattern were twofold. First, and most fundamental, it had to produce clear and
unambiguous patterning, and it is the fact that several techniques did produce such sat-
isfactory and essentially identical results which gives rise to confidence in the general
trends so indicated. And, second, it had to show a close association of classes 4a, 4b,
and 4c (different aspects of clay pipes), as it could be assumed that these were origi-
nally closely related; again, the fact that some procedures did behave satisfactorily on
this criterion constitutes grounds for confidence in the approach. A third criterion could
be applied, consisting of a search for close relationships between classes 2 and 3 (glass
bottles and their tops), but this appeared in no instance, and it must be presumed that
the different preservational factors affecting these two classes had obscured any origi-
nal association.
A simple and intuitively straightforward clustering procedure is that of linking
together nearest neighbors; in fact, some statisticians consider this technique to be the
only completely valid one (Doran and Hodson, 1975: 176). The dendrogram resulting
from this approach using the Euclidean distance measure is given in Figure 4, while that
using the product-moment coefficient appears in Figure 5. The former suffers from un-
due chaining onto one central cluster initiated by classes 4a, 4b, and 4c, a common fail-
ing of this procedure, but the results from the latter are reasonably satisfactory,
producing the three discrete clusters indicated along the top of Figure 5. A similarly sat-
isfactory couple of dendrograms arose from the use of a related technique, that of group
average clustering; the one produced from the Euclidean distance matrix is given in Fig-
ure 6 and that arising from the product-moment matrix in Figure 7. While the former
produced a couple of clear clusters, but with still some signs of chaining in the central
cluster, the latter displayed a series of very clear groupings, at two distinct levels, as indi-
cated at the top of Figure 7. As has been observed in several other archaeological situ-
ations, this approach proved in many respects to be the most satisfactory; however,
 484 K. Muckelroy

2 4a 4c 4b 6 13 11 8 5 9 3
o

"-r-

0.5

1.0

1.5 L I

2.0 I
Figure 4. Kennemerland site analysis; dendrogram I, using distance correlation and nearest neighbor
clustering.

2 4a 4c 4b 11 3 6 13 5 9 8
10

~r-

05

-- T

o I

I
Figure 5. Kennemerland site analysis; dendrogram 2, using product-moment correlation and nearest
neighbor clustering.
The Analysis of Discontinuous Sea-Bed Distribution Sites 485

I 1-1- - - - I
2 4a 4c 4b 6 13 11 5 9 8 3
o

--
0.5

1.0

..... -

15
T
2.0

2.5

3.0
Figure 6. Kennemerland site analysis; dendrogram 3, using distance correlation and group average
clustering.

serious mathematical objections have been raised concerning its operation, and complete
reliance on it is not recommended (Doran and Hodson, 1975:177). A third related tech-
nique is clustering by considering further neighbors; in the case of the data derived from
the Euclidean distance matrix this produced a result that did not cluster all of the sub-
groups of class 4 together, while the use of the product-moment data produced a dendro-
gram almost identical to Figure 7. The final technique that can be applied to both
matrices involves the use of McQuitty's coefficient; in both instances the results were
practically identical to those produced by using the group average approach.
Four other procedures were tested which could only be applied to data from the
Euclidean distance matrix. Of these, one, the Lance-Williams coefficient, produced a
result that did not satisfy the second of the above criteria, while two others, involving
the related concepts of centroid and median clustering (Lance and Williams, 1967),
exhibited undesirable chaining characteristics parallel to those in Figure 4. However,
the final procedure, known as Ward's Method (Ward, 1963), produced a very satisfac-
486 K. Muckelrciy

I I
I----! I I t----i
2 4a 4c 4b 3 6 5' 9 8 11 13
1.0

...... -

0.5

T
o

Figure 7. Kennemerland site analysis; dendrogram 4. using product-moment correlation and group
average clustering.

tory pattern, as shown in Figure 8, although only after a relocation procedure had been
applied at several levels, as a result of which class 4b migrated clusters. This was the
result that came nearest to achieving the association of classes 2 and 3, and it is based
on a procedure that is widely considered to be particularly satisfactory (Wishart, 1969).
If one considers only the first group of four procedures, then there can be no
doubt that the product-moment correlation data consistently produced more useful
results, being much less susceptible to undesirable chaining. Similar considerations
lead to particular faith in the results of using the group average procedure, especially
because they are duplicated by those achieved using the McQuitty coefficient, which is
not open to the same mathematical objections. However, the satisfactory nature of Fig-
ure 8 suggests that the distance measure is not to be totally discarded, especially when
associated with a clustering procedure to which it is particularly well suited. For a brief
discussion of the archaeological consequences of these results given below, attention is
focused on the dendrograms in Figure 7 (product-rnoment correlation plus group aver-
age clustering) and Figure 8 (Euclidean distance measure plus Ward's Method).
In both of these figures, there is a clear association of classes 5, 8, and 9 (lead
shot, bronze sheeting, and bronze nails), which might together indicate an origin in the
forward part of the ship, where the bosun's stores and the armory would be situated
(see the parallel of the Dartmouth discussed above). A similar close association of
The Analysis of Discontinuous Sea,Bed Distribution Sites 487

2 4a 4c 3 4b 6 13 11 5 9 8
o

0.5

1.0

T
1.5

+@ @~ T
2.e

2.5

3.0

3.5

4.0

4.5
I
Figure 8. Kennemerland site analysis; dendrogram 5, using distance correlation and Ward's Method
clustering (with relocation).

classes 11, 13, and, in Figure 8, 6 (bones, earthenware pottery, and personal posses-
sions) indicates items for use by those on board during the voyage, in the form of salt
meat, crockery, and personal valuables and utensils; an interesting point here is that the
earthenware might equally well have been considered as cargo if it were not for these
results. And finally, there is a consistent, if rather more loose cluster including classes
4 (clay pipes), 1 (stoneware vessels), 2 (glass bottles), and, in Figure 8, 3 (bottle tops),
which may be taken as indicating items from the ship's cargo. In the case of clay pipes
this is a particularly interesting conclusion, for they might otherwise have been
regarded as predominantly items for use during the voyage. The stoneware vessels and
glass bottles were not themselves the cargo, but rather the containers in which brandy,
wine, and other liquids were being carried. The higher level of clustering indicated in
Figure 7 also suggests a primary division between items in use during the voyage in
one context or another, and those permanently consigned to the hold, although this dis-
tinction is not so clear in Figure 8 or the other dendrograms. This brief summary of
some of the possible archaeological implications of this study has obviously not been
exhaustive, but it has indicated some of the potential of such an approach.
488 K. Muckelroy

Despite these relatively satisfactory results, note must be taken of some of the
disadvantages inherent in any quadrat technique, arising from the heavy bias imposed
on the results by the location, size, and shape of the squares (Hodder and Orton,
1976:36-38,204). In this instance, a few tests comparing the results from different grid
sizes, following the procedure for comparing the "mean square between blocks"
(Greig-Smith, 1964:85-93; Hodder and Orton, 1976:34), failed to show that consider-
ing larger blocks would improve the discrimination, and suggested that the most satis-
factory approach was to consider the smallest possible squares. Furthermore, as the
nature of a discontinuous site rules out the use of any measure for distributional analy-
sis depending on the actual distance between items, there are few alternatives to the
quadrat method. One other possibility is some form of Runs Test (Hodder and Orton,
1976:208-210), considering the relative ordering of items from two assemblages, and
not precise distances, and seeking to reject the null hypothesis that the two assemblages
could be derived from the same population. However, it would probably not be valid to
use the scores achieved as a detailed measure of association between such groups, and
so the approach would seem to be less useful than the quadrat method.
The statistical procedures discussed above have been operating at essentially a
descriptive level, seeking to summarize and clarify large bodies of data so that the main
patterns within them may be brought out for archaeological interpretation. There are,
however, more powerful techniques available, in which the use of the techniques them-
selves possesses an interpretive potential, in suggesting how the observed pattern came
to be generated. For example, useful insights into the processes by which new land may
have been colonized by Neolithic farmers have been gained by applying procedures
developed for the study of epidemics. In the absence of any helpful analogous proce-
dures, a useful technique would seem to be the computer simulation of the shipwreck
process in any particular set of circumstances, to be tested for "goodness of fit" against
the observed patterns. In the early stages of such an exercise, the results would
undoubtedly be confused and unsatisfactory, but the hope would be that in time the sig-
nificant factors would become clear, along with any important gaps in current knowl-
edge or understanding. A site such as the Kennemer/and wreck site would seem to be
particularly suitable for developing such an approach, in the light of the favorable char-
acteristics outlined above, but more general interpretative models should appear in due
course, based on those features that are common to all shipwrecks, to be used on sites
that are less well evidenced, and especially those from earlier and less well-understood
periods. This approach would represent the ultimate consolidation of everything dis-
cussed in this chapter, the interpretation of the wrecking system presented in Figure 1
of Chapter 25, and the maximum exploitation of the information contained in the sea-
bed distributions; only time will tell whether its promise can be fully realized.

REFERENCES
Doran, J.E., and Hodson, F.R., 1975, Mathematics and Computers in Archaeology. Edinburgh University
Press, Edinburgh.
Greig-Smith, P., 1964, Quantitative Plant Ecology. Methuen, London.
Hodder, 1., and Orton, C., 1976, Spatial Analysis in Archaeology. Cambridge University Press, London.
Lance, G.N., and Williams, W.T., 1967, A General Theory of Classificatory Sorting Strategies; I. Hierarchical
Systems. Computer Journal 9:373-380.
Muckelroy, K. W., 1975, A Systematic Approach to the Investigation of Scattered Wreck Sites. International
Journal of Nautical Archaeology and Underwater Exploration 4: 173-190.
The Analysis of Discontinuous Sea-Bed Distribution Sites 489

Price, R., and Muckelroy, K.W., 1974, The Second Season of Work on the Kennemerland Site, 1973: An Interim
Report. International Journal o/Nautical Archaeology 3:257-268.
Robinson, WS., and Brainerd, G.W., 1951, A Method for Chronologically Ordering Archaeological Deposits, and
the Place of Chronologically Ordering in Archaeological Analysis. American Antiquity 16:293-313.
Sneath, P.H.A., and Sokal, R.R., 1973, Numerical Taxonomy: The Principles and Practice 0/ Numerical Clas-
sifications. Freeman, San Francisco.
Ward, J.H., 1963, Hierarchical Grouping to Optimise an Objective Function. Journal o/the American Statistical
Association 58:236-244.
Wishart, D., 1969, An Algorithm for Hierarchical Classifications. Biometrics 25: 165--170.
 Part VII.C

Extreme Data
Recovery Conditions

Zero visibility and a strong current seriously hamper diving on the CSS Georgia. Baker
et al. (Chapter 43) combined remote sensing techniques and computer graphics to come
up with a rough sketch of the site's major features. Side-scan sonar, magnetic anomaly
maps, and detailed bathymetric data enabled investigators to determine approximate
length of ship, as well as the dynamic nature of the river bottom. These techniques do
not replace divers on the bottom, but they help in planning further work.
In Chapter 44, Sellers deals with the importance of recognizing and overcoming
psychological barriers inherent i'n zero visibility training and dive conditions. The
author approaches the larger problem of stress on all those who work underwater.
Van Tilburg (Chapter 45) presents information on the East Carolina University
training for field work in extreme environmental conditions, namely strong currents
and zero visibility. This article also includes some recording techniques in zero visibil-
ity environments. Few student programs have anything to do with work in these condi-
tions, and documentation of any of their activities is extremely rare.

491
 Chapter 43

'.~~.' Mapping and Site

~

o i' .
Characterization in
q;,O_" .': .
.j'~l Zero Visibility
The C.S.S. Georgia

JAMES G. BAKER,
RICHARD J. ANUSKIEWICZ,
and ERVAN G. GARRISON

INTRODUCTION

The purpose of this presentation is to share the ways in which computer-generated

graphics were an aid to wreck site visualization for the C.S.S. Georgia. The Georgia
lies in the totally black water of the Savannah River where a combination of strong cur-
rents and a 6-ft tide fall limits the work time for divers to brief periods at slack and
high tide. During the spring there is only an hour each day of acceptable diving condi-
tions. To further complicate the picture, indications were that dredging had scattered
the wreckage over a broad area of the channel and a salvage attempt in 1868 involving
the use of dynamite had destroyed an undetermined portion of the superstructure. An
extensive search of archival sources failed to produce a set of ship's plans, but six
sources gave her length and beam - each different, varying from 150 to 260 ft in
length and 40 to 60 ft of beam (Garrison and Lowery, 1980).
Because of the impossibility of mapping the site with conventional techniques, a
variety of remote sensing methods were applied. These included: side-scan sonar, mag-
Source: Underwater Archaeology: The Challenge before Us; The Proceedings of the Twelfth Conference on
Underwater Archaeology, edited by Gordon P. Watts, Jr. (1981), pp. 10-20. Fathom Eight, San Marino, Califor-
nia. Reprinted by permission.

493
494 J.G. Baker et at.

netometric, tight beam bathymetric, and video. The video proved unable to see through
the thick sediment load of the river and was abandoned. The side-scan sonar produced
informative images; one will be discussed later. The magnetometer and bathymeter
produced large amounts of data; 10,000 bits of data per run was typical. To image and
process such quantities the Amdahl and Prime computers at Texas A & M University
(T AMU) were employed and the final images were plotted on the Versatec and Prin-
tronix plotters. We will look at three ways of displaying the data base. The magne-
tometric data were plotted as a curved line contour map only. The bathymetric data
were plotted as:
I. Curved line contour maps
2. Gray tone contour maps
3. Three-dimensional grid element contour displays, which could also be used
to produce stereoscopic views.
The combinations of these techniques with diver verification of salient features
allowed a definition of the nature and extent of the wreck site which will serve as an
essential part of the planning process for the future disposition of the vessel. Plans by
the Corps to enlarge the ship channel necessitate the removal of the historic monument.
A cofferdam of driven steel piles is under consideration as a method to still the current
and tide, clarify the water, and allow excavation. It was, therefore, essential that the
structural limits be located to prevent the expense of enclosing vacant areas or driving
piles through the wreck.
The U.S. Corps of Engineers Savannah District, under the direction of Colonel
Tilford C. Creel, provided funding and much of the raw data base. Dr. Ervan G. Garri-
son of TAMU was the principal investigator in charge of data collection and project
operations. Richard J. Anuskiewicz was the archaeologist for the Corps. T AMU pro-
duced and interpreted the computer graphics. The data base was collected by the
Corp's vessel Carlson and its onboard systems.

FIGURE 1

The ironclad ram Georgia was built in the city of Savannah to serve as a harbor
defense vessel but was used as a floating battery because of the lack of effective
engines. She was clad with railroad rails with the casemate extending below her water-
line. No plans survive to record her exact dimensions or features of her construction.

Figure I. Artist's conception ofeSS Georgia.

Mapping and Site Characterization in Zero Visibility 495

As Sherman's troops closed in on the hopes of the Confederacy and attacked Savannah
on December 20, 1864, the Georgia was scuttled to prevent her falling into Union
hands. Despite the aforementioned salvage attempt and the occasional hits by dredges,
the vessel and her contents have remained largely intact.
Contemporary reports indicate that the wreck should contain a full complement
of ship's stores and personal belongings of the period, as the sailors took with them
only their personal sidearms. It is a concern for saving these objects, the vessel contain-
ing them, and the associations of these that motivates the project. She lies beneath 30 to
40 ft of water in the area where she spent most of her short life, opposite Fort Jackson,
in the middle of the Savannah River.

FIGURE 2

To note a number of features in this aerial view of the River and the city of
Savannah for the purpose of orientation on successive images will be useful. The river
flows from left to right, the main channel being the south or lower one on the left and
the north or upper one on the right. The northern shore is South Carolina and the south-
ern shore is the state of Georgia. The wreck is located off Fort Jackson in the middle of
the river at the limits of the juncture of the main dredged channel and the channel
dredged for the Back River.

FIGURE 3

This enlargement will help orient us to the wreck site. To the south we see Fort
Jackson sitting on the Georgia shore. Across the river is the South Carolina shore. The

Figure 2. Vicinity and locality maps of CSS Georgia wreck site.

496 J.G. Baker et at.

- -------_..-/'
South Carolina /"
,/"

--- C.S.S. GEORGIA

WRECK
./
MI.n Oredoed Channel
/'
Old Dredged - - - -
Beck Rive' Chlnnel /--
/
-/~~-= . / GeorOl1

-- - FEET

Fig. 3. Wreck site of the

Pi

f.i.i. Georgia.

Figure 3. Wreck site of the CSS Georgia.

19.000-
IIAQNETOOIIn A CONTOUII MAP
C' I .OOGAM ,I.,;

,\ .
•r. o

o 20 MUElIS 00

Figure 4. Contour map of magnetic intensity at the CSS Georgia wreck site. Grid spacing = 100 ft.
Contour interval = 100 gammas.
Mapping and Site Characterization in Zero Visibility 497

main dredged channel of the river is shown clearly, as is the Back River's dredged
channel. The wreck site was located at the northern limit of the intersection of the two
channels. Dredging activity for the back channel was impacting the upstream end of the
wreck in the area where it entered the main channel. This back channel was later
moved south to avoid the vessel and now enters the main channel about where the mid-
channel is indicated on this figure. Keep in mind the angle of intersection of the lines
of the back and main channels as they will be useful to orient the other graphics.

FIGURE 4

The first computer graphic is to be viewed as a planimetric presentation of mag-

netic anomalies at the wreck site. The contour interval is 100 gamas; each square of the
superimposed grid is equal to 100 ft. The anomaly on the lower right is the result of a
modern anchor which was later recovered; the anomaly on the lower left is the chain
and anchor to the wreck buoy marking the site. The other contours record an intense
magnetic concentration to the north of the main channel at the point of intersection
with the back channel and to the east of that intersection.
This information does serve to approximate the location of the wreck; but be-
cause the Georgia was an ironclad and she is an object of intense magnetism, it is diffi-
cult to obtain detail using this technique. Data for this image were obtained by passing
over the site taking magnetometer readings in coordination with onshore transit shoot-
ings to locate spots where the readings were made. These data were then fed into the
Amdahl through a program that performed trigonometry and positioning of readings.
Contour maps were produced using a "Conrec" program subroutine (Reid, 1980). This
technique has been used for some years and more complex graphics of magnetometric
data have been produced (Arnold and Clausen, 1975; Breiner, 1975; Clausen and Ar-
nold, 1976; Upham, 1979).

FIGURE 5

This sonar graph was produced onboard the Carlson and shows the fine resolu-
tion possible if the right pass is made over a wreck site. Such images are produced by
sound waves and should not be regarded as a photograph. What it does show is the two
large pieces of casemate which remain intact. This side view is from the South Carolina
side looking toward Fort Jackson. Here the Back River channel enters from the west
(right). The area between the two pieces of structure may well reflect the dynamiting
and salvage operations attempted in the late 1860s.

FIGURE 6

One way to present the bathymetric data is to represent depth differences by dif-
ferent shades of gray. On this gray-tone contour map each shade of gray represents 1.5
ft of depth; it is based on data gathered by the Carlson in December 1979. This is an
aerial view with the darkest shades representing the deepest parts of the dredged chan-
nel. These lines are the intersection of the main and back river and its impaction on the
wreck shows up on the left of the map; the scatter of debris into the channel is recorded
498 J.G. Baker et at.

Figure 5. Side-scan sonar record of wreck area. View from the South Carolina side; dredged chan-
nel of Back River enters from right; downstream is to the left. View is from north looking south.

as the area of light grayout in the channel. The numbers record diver-located points on
the wreck which were buoyed and shot with onshore transits.
This next series of three figures will illustrate the development of our knowledge
as the computer-processed data base was expanded. (Thanks are due to Mr. Sid Theis
of TAMU Remote Sensing Department for his assistance in the production of these
maps.) For these maps the same bathymeter-generated data base, used in Figure 6, was
imaged as a regular smooth line contour map, with a contour interval of I ft. Three dif-
ferent maps are shown for February, March, and August of 1980. North is to the top of
these maps.

FIGURE 7

The February run shows the upstream limits of the wreck as a projection of high
relief extending into the intersection of the old Back River channel and the main chan-
nel. The damage to the wreck at this point appears significant. A scatter of debris
extends into the main channel. We lacked sufficient information on the downstream
end and requested that the next run cover that area more fully.

FIGURE 8

Figure 8 is based on data gathered during an intense thunderstorm in March 1980.

The same features are visible on the upstream end, but the definition we sought of the
downstream end has begun to appear. An area of sharp relief can be seen in the north-
easternmost square of the map, defining the area where the dredge rammed the down-
stream end of the wreck. This feature is defined more clearly in the final contour map
fr()m August.
Mapping and Site Characterization in Zero Visibility 499

Figure 6. Gray-tone contour of Georgia wreck site. Shade interval = 1.5 ft. Letters indicate buoys;
numbers are diver-located points on the wreck. Later bathymetric data indicated additional wreckage
to the east of points 6 and 7.
 500 J.G. Baker et at.

./

Figure 7. Computer-generated contour map of the CSS Georgia wreck site. Contour interval = I ft;
grid spacing = 50 ft. Based on February 1980 bathymetric data.

Figure 8. Computer-generated map of the CSS Georgia wreck site. Contour interval = 1 ft; grid
spacing = 50 ft. Based on March 1980 bathymetric data.
Mapping and Site Characterization in Zero Visibility 501

/
/,'''-.

,\\
.~./ --- \,,---/

Figure 9. Computer-generated map of the CSS Georgia wreck site, Contour interval = 1 ft; grid
spacing = 50 ft. Based on August 1980 bathymetric data.

FIGURE 9

Except for the movement of sand bars, relief and major features shown in Figure
9 are similar to the previous contour maps. A problem arises, however, when the maps
for August and March are compared, for there is a difference of 50 ft between the area
of dredge impaction downstream and upstream on the two successive maps. Only addi-
tional data will be able to resolve this problem. The downstream end of the wreck
(northeastern part of map) shows clearly the 40 ft relief line and just upstream the oval-
shaped 28 ft contour, which is the smaller piece of superstructure seen earlier in the
sonargraph. The 40 ft depression is thought to have been caused by the dredge making
a solid hit at the downstream end of the wreck and digging in. This series allowed us to
state with a degree of surety that the 250 ft estimates of the length were the correct
ones. It also allowed an accurate construction of the site as it is today. However, to the
uninitiated, contour maps are often difficult to read; even to the experienced, subtle.
features may escape notice. We also felt an even more dynamic presentation was
needed to communicate area features for divers and to use as an orientation model for
divers and Corps personnel working on or near the site.

FIGURE 10 TO IJ

Production of three-dimensional grid element contour displays provided the most

readable and communicative graphics. (These graphics were produced by Dr. Doug
Green and his student, Steve Hyde, ofTAMU.)
502

-~~
GfClllU:::]OlIlE.
,
N ~~
J.G. Baker et a1.

~H CAlQU ... Sill!

1
~

~~
v

VIEW FROM NORTH lOO111ll' SOUTH

Figure 10. 3-D isometric plot of upstream area of wreck side. Back River channel enters ITom right.
Grid spacing = 5 ft. Vertical axis is exaggerated to more clearly show relief.

Figures 10 to 13 are representations of the February 1980 data base displayed as

3-D graphics. The last few years have seen dramatic advances in this type of graphics
(Nagy, 1971; Hartig, 1973; Weymouth, 1976; Upham, 1979; Frankel, 1980). The verti-
cal has been exaggerated to show more clearly the relief. A 5-ft grid was used. The
image is similar to laying a net over the wreck site and drawing a picture of the net
allowing it to define the object underneath.

FIGURE 10

Figure 10 represents a view similar to that in the sonar graph. It shows the large
piece of casemate seen in the sonar graph; the smaller piece lies outside the area in
these graphics as does the deep hole caused by dredge impaction on the downstream
end. The view is to the south looking toward Fort Jackson from the South Carolina
side. Dredging from the Back River channel enters from the right and impacts the
upstream end of the Georgia; the debris resulting from this area is seen in the channel
beyond.

FIGURE 11

Another advantage of this graphic technique is the ability to image it on a TV

monitor and to rotate the image as though one were swimming around the wreck. When
an informative view is found, it can then be printed in hard copy. Figure 11 is gener-
ated from the same data base as 10, but viewed from a different angle. Here we are
looking downstream along the main channel. The Back River channel enters from the
left; the debris in the foreground represents parts of the Georgia and results from the
old channel dredging.

FIGURE 12

Proceeding around the wreck, Figure 12 is the view from the south looking
toward the South Carolina side. The features are familiar by now, namely, the Back
River and Savannah channels, the relief of the intact areas of structure, the debris, and
the repose of the South Carolina side.
Mapping and Site Characterization in Zero Visibility 503

'1IEW FROM WEST I.OOklNr. t.AST

Figure II. 3-D isometric plot of upstream area of wreck site. Back River channel enters from left
foreground. Grid spacing = 5 ft. Vertical axis is exaggerated to more clearly show relief

FIGURE 13

Completing our tour of the site is Figure 13, looking upstream along the axis of
the wreck with South Carolina on the right and the dredged channel of the Savannah on
the left. The wreck forms a long, low mound with sharp relief along her southern side
caused by the dredging; divers verify that wreckage is exposed along much of the far
end (west end) of this face. It is also practical to produce pairs of images which may be

Figure 12. 3-D isometric plot of wreck area. Back River enters from left back, northwest side. Grid
spacing = 5 ft. Vertical axis is exaggerated to more clearly show relief
 504 J.G. Baker et at.

Figure 13. 3,D isometric plot of upstream area of wreck. Back River channel enters from right back
side. Grid spacing = 5 ft. Vertical axis is exaggerated to more clearly show relief

viewed with a stereoscope. This provides a level of information above that offered by
the 3-D image. Stereo images can be produced by changing the point of view 6°. When
viewed with a stereoscope these pairs give a greater sense of three-dimensionality than
a flat 3-D image. This allows a fuller mental picture of the site to form in the mind of
the viewer and aids in the process of conceptualization.

CONCLUSION

Thanks to computer graphics, the major features of the wreck site have been
characterized and structural elements have been defined. The length of the ship has
been determined to be about 250 ft; the beam must await excavation for a determina-
tion. A complete and working image of the site has been created which will allow and
assit planning for future activities. The dynamic nature of the bottom in the area
became evident as sand bars appeared and disappeared on successive months' maps.
These techniques have allowed us to see through the sediments of the Savannah and to
characterize a major wreck site in zero visibility.

REFERENCES

Arnold, J., III, and Clausen, C.J., 1975, A Magnetometer Survey with Electronic Positioning Control and Cal-
culator-PIotter System. International Journal of Nautical Archaeology 4(2):353-366.
Breiner, S., 1975, Marine Magnetics Search. Geometries Technical Report No.7. Palo Alto.
Clausen, C.J., and Arnold, J.B., III, 1976, The Magnetometer and Underwater Archaeology: Magnetic Deline-
ation of Individual Wreck Sites, a New Control Technique. International Journal of Nautical Archaeology
5(2): 159-169.
Mapping and Site Characterization in Zero Visibility 505
Frankel, D., 1980, Contour Plans and Surface Plotting: Aids for the Field Archaeologist. Journal of Field
Archaeology 7:367-372.
Garrison, E.G., and Lowery, L., Jr., 1980, Archaeological and Engineering Study ofthe C.S.S. Georgia, Parts I
and 2. Texas A & M University, College Station.
Hartig, G.W., 1973, Contur-A Fortran IV Routinefor Plotting Contour Lines. National Technicallnfonnation
Services, U.S. Department of Commerce, AD-760437.
Nagy. N.J., III, 1971, The Graphic Representation of Two-Variable Data. Los Alamos Scientific Report No.
LA-4796, November.
Reid, T., 1980, Computer Graphics Softwarefor the Amdahl. Data Processing Center, Texas A & M University,
College Station.
Upham, S. (ed.>, 1979, Computer Graphics in Archaeology Statistical Cartographic Applications to Spatial
Analysis in Archaeology Contexts. Arizona State University Anthropological Research Papers No. IS.
Tempe.
Versatec, A. Xerox Company, 1978, Versaplot Software Programming Manual. Santa Clara.
Weymouth J. W., 1976, A Magnetic Survey of the Walth Bay Site (39WW203). Midwest Archaeological Center,
National Park Service, Lincoln.
 Chapter 44

Bridging the Experience Gap

Techniques for Reducing the Stress
of Zero Visibility Training

STEVEN H. SELLERS

Zero visibility diving may be the most stressful activity the vast majority of divers ever
experience. When zero visibility conditions are encountered, even seasoned clear water
divers may find themselves stressed to the point that they must terminate the dive.
Some training techniques useful in preparing divers for the zero visibility environment
emphasize composure, others have more emphasis on problem solving abilities while
still employing stress. Techniques are available which will aid in the training process
while reducing the stress element. Designed to be conducted with supervision in a shal-
low pool, the Zero Visibility Maze and associated techniques offer divers an introduc-
tion to the sightless diving environment with a minimum of stress. Unlike an obstacle
course, there are no entrapments, entanglements, restrictions or time limits (other than
air availability). The divers must solve only minor problems of direction and orienta-
tion, while completing a series of objectives. This approach allows divers new to the
sightless or extremely limited visibility diving environment time to become acclima-
tized and aids in developing confidence, tactile and orientation skills.

INTRODUCTION

At some point on the experience curve, divers usually become comfortable with
their abilities under a variety of diving conditions. As this happens, the diver may lose
contact with the fact that certain skills or conditions were difficult and stressful when
first experienced.
Source: American Academy o(UndelWater Sciellces: Proceedillgs, edited by John Heine (1993), pp. 127-131.
American Academy of Underwater Sciences, Monterey, California. Reprinted by permission.

507
508 S.H. Sellers

As divers become more and more comfortable with specific tasks and conditions,
the gulf of understanding widens until they can no longer comprehend another person's
apprehension of a given situation. This experience gap produces a variety of reactions.
From the view of the inexperienced diver, someone with the ability to function, or in
some cases thrive, under conditions they find barely tolerable may be looked upon with
awe. The diver with a mastery of certain situations may look on someone less comfort-
able and wonder: "Will this person ever be able to handle this, or am I going to have to
go in and recover the body?"
Both views are flawed, blinded by the experience gap that neither can perceive
because of their relative positions. New divers doubt they will ever possess "what it
takes"; the experienced diver sees only the gulf between abilities and can not under-
stand why.
This is one of the challenges of instruction; to bridge the experience gap by tak-
ing a person with "minimal diving abilities" (a relative term) and tum him or her into a
diver capable of handling almost any situation smoothly, efficiently, and safely. This
challenge can be formidable under the best circumstances, but the topic here concerns
less than optimal conditions.
Zero visibility diving may be the most stressful activity that the vast majority of
divers ever experience. It removes the sense most people use to perceive the world and
replaces it with a wet blackness where imaginary monsters lurk, waiting, and hungry.
All those fears of the things that go bump in the night have an avenue into the con-
scious mind. Without orientation aids, direction becomes only up or down, all other
bearing is lost in the murk. Reason can be forced out and the unprepared may find
themselves rapidly approaching panic.
Working in zero visibility conditions requires divers to rely on senses other than
sight for primary information. They must learn to use tactile information and develop
the ability to picture in their minds what they feel with their hands and bodies. In many
cases the physical structure of the dive site can provide divers with the cues necessary
to keep track of their position. On sites where physical structure alone is inadequate,
the use of a guide line or other introduced navigational aid may be necessary. However,
under field conditions, experience has demonstrated that the availability of structure or
guide lines will not insure that divers remain oriented. This is especially true for less
experienced divers. To aid in the development of these skills there are a variety of
approaches which may be utilized.
Some training techniques emphasize composure by placing divers in stressful
situations. An example of such a training exercise is the "Regulator Stress Course"
employed by Julius Wiggins of the National Academy Of Police Diving, Miami, FL.
Other techniques such as the "Zero Visibility Obstacle Course" (Sellers and Scharf,
1990) have more emphasis on problem solving abilities while still employing stress.
Both courses can be effective tools in training divers for the zero visibility environ-
ment. However, these techniques may require divers to demonstrate composure levels
which many do not posses due to limited experience. To reduce the experience gap,
other less stressful methods may be employed, thereby allowing divers to ease into zero
visibility diving.
The concept of the Zero Visibility Maze and the exercises associated with it were
developed for use with the East Carolina University (ECU) Program in Maritime History
and Nautical Archeology. The Maritime Program attracts participants covering the entire
Bridging the Experience Gap 509

spectrum of diver certification, but most have never experienced zero visibility diving.
As a result, the majority of the divers coming into the program do not have the compo-
sure or orientation skills required to perform tasks that would otherwise be routine.
Brad Rodgers, Underwater Archeologist with ECU stresses the similarities of the
proper movement pattern for zero visibility diving to the slow, controlled movements
of Tai Chi. This concept appears to be one of the most difficult for divers new to this
environment to grasp. Almost without fail, divers inexperienced in zero visibility con-
ditions will move too rapidly and with such heavy handedness that they can be a hazard
to themselves and anyone or anything in the water with them.
Slow, deliberate movements must be stressed. Paying greater attention to the tac-
tile signals being received from the entire body is crucial. Divers experienced with zero
visibility conditions, in tune with their bodies, can identify certain objects as readily
with other body parts as they can with their hands. It is as if the entire body becomes an
antenna. This too is a concept that does not come easily to those new to the sightless
environment. Couple this inattention with movements too rapid for the conditions and
it is no wonder that inexperienced divers become disoriented.
Working with the ECU Maritime Program in the field has reinforced the ideas
that: To accomplish a specific task under zero visibility conditions, it is essential that
the divers know where they are on the dive site and: That the divers require the ability
to identify specific items by touch. The use of the Zero Visibility Maze during training
can go a long way toward providing divers the background needed to later maximize
the time spent working under blackout conditions.

THE TRAINING PROCESS

Begin the process of acclimatization/desensitization by blacking out the divers'
mask and having it donned in a staging area. While limiting vision may be accom-
plished by tying something over the mask, a more realistic effect is achieved if a mask
with an opaque skirt is modified. Spray painting the faceplate inside and out with sev-
eral coats of paint will reduce the light coming in considerably. By removing the face-
plate and replacing it with a material such as rigid gray PVC, an effect of total
blackness may be achieved. Painting and replacing the faceplate require dedication of
equipment for zero visibility training. An alternate method is to blackout the divers per-
sonal mask by covering the faceplate with duct tape. This method will produce the
desired effect, but it tends to leave a sticky residue which may not be appreciated.
After vision has been blacked out, have participants assemble their diving equip-
ment unassisted. This process should be observed to assure it is done correctly. The
objective is to insure equipment familiarity and enhance diver confidence. If a mistake is
made, inform the person of the mistake and have it corrected. A variation here is to tell
the individual that there is a mistake in the way they have assembled their equipment, but
not to identify the mistake. The diver must then locate and correct the mistake.
When divers have correctly assembled their equipment, they are ready for the
next task. Give the divers a tactile puzzle to solve. this can be as simple as a series of
nuts and washers on a bolt, to be removed and replaced in the same order; or as com-
plex as a childs' shape puzzle.
The Playschool shape puzzle is a wonderful challenge under blackout conditions.
It consists of 18 different shapes to be placed into a box with 18 matching holes. This
510 S.H. Sellers

puzzle may be completed in less than five minutes (usually by parents of small chil-
dren) or may be perceived as totally baffling. A hint about the Playschool puzzle; the
shapes have a tactile design present on each piece that corresponds to the side of the
puzzle on which it belongs.
In addition to the shape puzzle, divers should be put through a series of item
identification exercises both in and out of the water. By handling a variety of common
every day articles and items that are project specific, divers can develop a mental pic-
ture of objects they may encounter in the field. The divers may be asked to identify
every day articles based on life experience. With more unusual project specific objects,
the identifications should be made prior to the divers handling of the items.
Items should be as varied as possible and should include objects which could pro-
duce minor physical injury if mishandled. The objective here is not to injure the divers,
but to make them aware of the potential for cuts, scrapes, or punctures. The concept of
grasping with a light touch is reinforced by the manipulation of items such as hooks,
frayed wire, metal with rough edges, etcetera. The divers should be cautioned before
they are given these items to handle. After the divers have completed the item identifi-
cation orientation they are ready to progress to the Zero Visibility Maze.
Unlike an obstacle course or other higher stress exercise, the Zero Visibility
Maze does not require one on one supervision. Spot checks of the divers involved in
the drill are usually all that are necessary. For most circumstances, four to six divers
are easily supervised by one safety diver . The reduction in supervision requirements
translates into larger numbers of participants in a given time period.
A primary purpose of the maze exercise is to allow divers time to develop and
improve personal movement and orientation skills. Therefore, direct contact with the
divers by the supervisors should be kept to a minimum. The supervising divers should
be instructed to limit their contact with participants to situations that would contain a
threat of injury, or when individuals are being so heavy handed that they are destroying
the maze structure.
As with any underwater activity, the divers and supervisors should be briefed on
predetermined emergency and recall signals. When the exercise is conducted in a shal-
low pool, personnel overseeing the activity are outfitted in snorkeling gear to facilitate
ease of movement.
The design of the maze need not be overly complicated, but it is important that
the structure be stable. If it is applicable, design the Maze to be similar in structure to
what the divers will experience in the field. For example, if a grid will be used on an
upcoming dive site, the pattern of the grid should be duplicated as closely as possible.
This preconditioning allows the divers to become accustomed with "the feel" of the
structure they will encounter.
Setting up a Zero Visibility Maze does not entail a major investment, though the
heavy handedness of many divers require the use of taut lines and rigid or semi-rigid
materials. Items common to many pools (for example; aluminum benches) may be
assembled to provide a recognizable structure. With this established, a series of guide
lines overlaid to provide an orientation aid is all that is necessary for the basic maze
structure.
For inexperienced divers to receive the maximum benefit from this exercise, the
maze should contain easily identifiable landmarks. Features such as distinctive comers,
knots, direction indicators, and the use of a variety of construction materials will pro-
Bridging the Experience Gap 511

vide attentive divers the cues necessary to quickly become and stay oriented. Objects
similar to those the divers have handled on the surface should be placed at various
points throughout the maze. These objects are used for item identification and location
exercises as well as orientation aids.
The divers enter the maze by way of descent/ascent lines. The number of
entrances, while exercise specific, should reflect what the divers will encounter during
field work. They are required to use these lines for ingress or egress.
The divers may be assigned any number of tasks as part of the maze exercise, but
the first assignment should always be to explore the maze and become oriented. Exam-
ples of diver tasks may include:
• Have the divers draw a map of the Maze before visually inspecting it.
• Have the divers list on a slate the objects they encounter.
• Have the divers retrieve specific objects from specific locations.
• The Grope. Have the divers identify other divers they encounter during the
exercise.
• Show the divers a drawing of the maze and have them follow a specific path
from entry to exit.
• Place the divers in contact with the Maze at an unknown point and have them
locate a specific object or exit point.
• Have the divers assemble or disassemble portions of the maze.
Assignments are limited only by imagination, but it is suggested that the progres-
sion be from the simple to the more complex. The tasks should reflect what the divers
will be expected to accomplish in the field.

CONCLUSION

The Zero Visibility Maze and related techniques were conceived to address spe-
cific problems encountered while training inexperienced divers for blackwater and
brownwater conditions. The techniques discussed here, while designed for controlled
zero visibility sites such as those associated Underwater Archeology, have application
to any diver training where zero visibility is a concern. As with any other stressful
activity, experience and training under zero visibility conditions are the keys to mas-
tery. No training program is perfect and all training programs should constantly evolve
if the needs of the individual diver are to be met. It should be emphasized that the train-
ing techniques discussed here, while stressing the development of individual skills, do
not profess the elimination of safe diving practices, but are a useful training aid for pre-
paring divers for the zero visibility environment.

LITERATURE CITED

Sellers, S.H., and R.J. Scharf, 1990. Training scientific divers for zero visibility diving, in W.Jaap (ed.), Diving
for Science, pp. 313-322. American Academy of Underwater Sciences, AAUS, Costa Mesa, CA.
 Chapter 45

Zero~Visibility Diving
on the Maple Leaf
The Tricks of the Trade

HANS K. VAN TILBURG

INTRODUCTION

The initial dive on the Maple Leaf (Cantelas and Rodger, 1994) was in total darkness,
producing an almost unimaginable feeling of disorientation. Even halogen lights pro-
duced no more than a dull glow in the silt-laden current. An upper layer of viscous mud
thickened into a jellylike substance below. That was all. The ascent line was reached by
memory.
Four weeks later hundreds of cubic yards of mud had been removed and a meas-
ured drawing of the first 50 ft of deck aft from the bow had been produced. This paper
describes some training techniques for zero-visibility underwater work and also offers
some solutions developed by the East Carolina University (ECU) field teams faced
with this extreme working environment.

TRAINING BACKGROUND

At ECU training for each season's field session begins a few weeks before the
actual site work. During this time both newly certified and experienced divers are intro-
duced to survey and recording techniques. Divers practice on a half-scale fiberglass
Source: InternationaiJournal o/Nautical Archaeology (1994),23(4):315-318. Reprinted by permission of Har-
court, Brace and Company, Limited.

513
514 H.K. Van Tilburg

model known as the Sinkentine (Watts, 1989). Increasingly, ECU training has empha-
sized low-and zero-visibility conditions, reflecting the acknowledged potential of river-
ine environments for nautical archaeology. This has changed the training schedule. In
addition to basic recording skills, preparation now includes working with more exotic
types of safety and dive gear, and increased practice in simulated and real zero-visibil-
ity conditions.

ADDITIONAL EQUIPMENT

Although the safety of surface-supplied lightweight diving gear with its hardwire
communications and infinite supply of air is ideal for limited situations, the lack of
mobility resulting from the umbilicals and the high number of divers needed in the
water at one time make it unsuited for this site. Modified scuba units fulfill the crew's
needs. Full-face masks enable the use of wireless underwater communication sets.l
Ocean Technology's Aquacoms, models SSB-2000 and SSB-3000, supply each diver
with communications. A surface set allows diving safety personnel to monitor under-
water operations and assist in sometimes garbled transmissions. Voice communications
are almost essential to operations in zero-visibility conditions.
The difficulty in reading air-pressure gauges presents another problem to safe
diving procedures. On the project, each 80 fe aluminum tank while in use carries an
attached 13 ft 3 safety bottle. This provides ample reserve for work at the relatively shal-
low depth of 30 ft. The air whip from this source leads to a manifold block attached to
the buoyancy compensator, and can be opened by the diver if low on air.
Finally, a battery pack on the tank is connected to a powerful halogen light
mounted above the mask, for those rare occasions when the site offers a few inches of
visibility. Usually these lights are switched off. They are, however, important to have
on hand because maximum advantage must be taken of any sort of visibility. The entire
bulky unit (BC, safety bottle/manifold, com set, face mask, and light), stored and trans-
ported in a separate carryall box, provides necessary communications and a sufficient
supply of reserve air. Every effort is made to streamline and condense the equipment's
array of hoses and wires.

SIMULATION AND TESTS

To ensure the proper simulation of the actual dive-site conditions, masks used for
all training drills are completely blacked out with a nontranslucent material. Preparation
for the truly dark conditions thus begins as early as possible, and in a safe, controlled
environment. The psychological barrier presented by the loss of the visual sense, much
more than any of the physical factors, is the toughest to overcome (Sellers, 1993 a,b).
Before the students leave the pool they face two zero-visibility (or black mask)
tests. The first consists of a simulated wreck site. "Descent" lines from the side of the
pool lead blind divers to the "silt barrier" (aluminum benches arranged in a perimeter).
Divers navigate to and from the barrier, locate certain items by touch, and transport and

1 Such as MD-II AGA, Diving System International's EXO-26, and Ocean Technology's Tec-I mask. Several
different styles of mask also provide a better fit for the variety of face shapes in the class.
Zero-Visibility Diving on the Maple Leaf 515

secure dredge heads. Other tasks include setting up polypropylene lines as guides and
learning the fine art of encountering and getting around another blind diver on the same
guideline. The difficulty this presents is always surprising at first, but soon the students
become familiar with blind encounters.
The second test, the black mask underwater obstacle course, is traditionally the
highlight of the training period. Usually the entire day is set aside as instructors and
diving safety personnel construct, in secret, the convoluted maze of entanglements.
Students, in scuba gear and blacked-out masks, are led by hand into the water and
placed in contact with the guideline; they are told to follow the line through whatever it
may encounter. If they become helplessly stuck, they may signal for assistance. Safety
divers provide one-to-one supervision, monitoring air consumption and composure.
Obstacles include hanging drift nets, tangled polypro, barrels, chicken wire, and mon-
ofilament line, among other nasty surprises. The test ensures that, while blind, the
divers will become entangled and practice freeing themselves from a variety of situ-
ations. While students are never allowed to dive under such threatening conditions on
an actual site, the course graphically demonstrates the importance of always moving
slowly and cautiously in zero-visibility environments. Divers learn to rely fully on the
tactile rather than visual sense (Sellers and Scharf, 1990).
The final portion of the training takes place in blackwater rivers around the eastern
North Carolina area. The students are expected to repeat the basic skills of tactile naviga-
tion and communication etiquette while confronting new challenges such as strong cur-
rents and the ever-present possibility of sharp objects. By this point the divers have been
exposed to difficult conditions above and beyond what they will encounter at the actual
site; they have had multiple opportunities to grow accustomed to the physical and psy-
chological stresses of zero-visibility diving, as well as the additional diving gear. Most
importantly, they are well trained in moving cautiously while on the bottom, taking time
to "think" their way around obstacles and remain free of entanglements.

The Unseen Site: Some Solutions to the Problems

Until the first drawings of the ship emerged as a coherent plan, the site existed, in
a sense, only as a series of mental constructs in the mind of each diver. Each had a per-
sonal image of the system of lines, moorings, barriers, and, as the dredges removed the
mud, the deck of the ship itself. As a response to working in this strange environment,
several solutions evolved to deal with various difficulties. These may be of some use to
other nautical archaeologists.
1. Individual tag lines: Divers often carried short, 8- to lO-ft lengths of 14-inch
polypro line with a snap shackle firmly secured to one end. These lines
allowed an increased radius of movement away from fixed guides or struc-
tures, while still maintaining contact to a known point on the site.
2. Object feelers: Similar to the curb feelers on large automobiles, plastic wire
ties (also known as electricians' snap ties) were sometimes positioned to pro-
trude from parts of the divers' bodies. The best place to attach these feelers is
the mask strap, so that they project above the diver's head, warning of any
imminent obstacle.
3. Gear attachment points: As it is not possible to put something down and then
find it in a zero-visibility environment, all tools must be secured somewhere
516 H.K. Van Tilburg

about the body. A simple scuba cam strap, rigged with D-rings and placed
around the lower portion of the tank, allows objects to be clipped securely
behind the diver, out of the way of frontal entanglements. It is also easily
transferable from tank to tank.
4. Visibility bags: In zero-visibility environments where even bright halogen
lamps will not penetrate the water, divers can read surveying tapes and slates
by means of a chemical light source {such as a cylume stick), activated and
dropped into a clear plastic bag filled with clean water. The bag is then
securely sealed and carried with the diver. Placing this directly between the
object and the mask permits relatively clear observations.
5. Cutting clippers: A pair of garden pruning shears operate much more effi-
ciently than divers's knives for cutting through monofilament or polypro line.
The ability to cut polypro not only relates to work, but must be considered
from a safety standpoint as well.
6. Game bags: Mesh bags are a practical solution for dealing with large
amounts of polypro underwater. Once the line is fed into the divers'mesh
game bag, it can easily be pulled or stuffed back into it while on the bottom.
7. Survey tape floats: Though not a perfect solution, small rounded Styrofoam
floats with holes through the center (like those found on many fishnets) can
be easily threaded onto fiberglass surveying tapes. Invariably, when pulling
the tape from one point to another, these will slide to the center point and
take up the slack, floating the tape over unseen obstructions.
8. Coveralls: In water too warm for wet suits, some protection from the threat
of sharp debris such as iron fasteners or twisted sheathing is still necessary.
Loose-fitting coveralls, procured from a local thrift shop, not only provide
this protection, but allow the crew to attach badly needed tool pockets, some
with Velcro enclosures and grommets for drain holes.
9. Search lines: Recovery of lost items such as survey tapes, slates, or wrist
watches helps to keep the overall costs of the project at a minimum. Small
spools of nylon cord (called "jump reels" by cave divers), when attached to
appropriate points underwater, allow searchers to swim arcs around the last
known locations of missing gear.

CONCLUSIONS

These few examples represent some of the nonarchaeological lessons from

ECU's work in zero visibility. Most are inexpensive and simple. All are designed to be
effective under low- or zero-visibility conditions. Communication and safety gear offer
the technological edge, while the training process at ECU produces divers who are able
to operate safely in an unfamiliar and potentially stressful environment and to over-
come the psychological barrier of working in total darkness.
These suggestions do not amount to the "perfect" solution, but they do allow divers
to accomplish work in, what still remains as, a hazardous environment.
Easily accessible clear-water sites are, quite naturally, the first to be investigated.
As time goes on, however, underwater archaeologists will find themselves faced with
more challenging types of environments. What was once considered an impossible task
Zero-Visibility Diving on the Maple Leaf 517

because of poor visibility becomes, with the proper training and technology, a site of
great potential archaeological value.

REFERENCES
= =
Cantelas, F.J., and Rodgers, B.A., 1994, The Maple Leaf A Case Study in Cost Effective Zero Visibility
Riverine Archaeology. International Journal of Nautical Archaeology 23(4):271-282.
Sellers, S. 1993a, The V-cam: A Variation on a Load Bearing System. Sources (3):53-54.
Sellers, S. 1993b, Bridging the Experience Gap: Techniques for Reducing the Stress of Zero Visibility Train-
ing. In American Academy of Underwater Sciences: Proceedings, pp. 127-131, edited by J. Heine.
Monterey, CA.
Sellers, S., and Scharf, R., 1990, Training Scientific Divers for Zero Visibility Diving. In Diving for Science,
edited by W. Jaap. American Academy of Underwater Sciences, Costa Mesa, CA.
Watts, G., 1989, The 'Sinkentine': A Fiberglass Shipwreck Model to Assist in Teaching Three-Dimensional
Mapping. International Journal ofNautical Archaeology 18: 151-156.
 '.~~.' Additional Bibliography
~

o i' .
for Part VII
q;,O_" .': .
.j'~l

Ballard, Robert D., 1990, The Discovery of the Bismarck. Warner/Madison Press, Toronto, Ontario.
Haggett, Peter, 1966, Locational Analysis in Human Geography. St. Martin's Press, New York.
Hodder, lan, and Clive Orton, 1976, Spatial Analysis in Archaeology. Cambridge University Press, New York.
Lipke, Paul, Peter Spectre, Benjamin A.G. Fuller (eds.), 1993, Boats A Manual for Their Documentation.
American Association for State and Local History, Nashville, Tennessee.
Mendenhall, William and Lyman Ott, 1976, Understanding Statistics. Duxbury Press, North Scituate, Massa-
chusetts.
Mueller, Janice W. (ed.), 1975, Sampling in Archaeology. University of Arizona Press, Tucson.
Thomas, David H., 1986, Refiguring Anthropology - First Principles of Probability & Statistics. Waveland
Press, Prospect Heights, Illinois.

519
 Part VIII

Conservation
".~i).:
~

-r/: ..
':.i

It is not enough to go into the water and simply recover artifacts. A key part of the plan-
ning process and an essential part of the postexcavation analysis and curation is the
development of methods to keep recovered materials available for additional research.
Conservation is keeping artifacts in the same condition, or in better condition, as they
were found. This means the application of resources and time to bulk up wood, arrest
rust, and clean glass, ceramics, and cloth.
Conservation is an essential part of the archaeological process because retention
of the artifacts is crucial to reevaluating early interpretations when techniques improve.
If the artifacts are allowed to disintegrate, great amounts of information are lost with
the vanishing material. Conservation must start immediately, at times, even before arti-
facts are brought to the surface.
In this part, one article is presented. However, the bibliography that follows pro-
vides a variety of essential readings that archaeologists will find useful. They range
from museum handbooks to more practical "cookbooks" that provide formulae for con-
serving different types of artifact classes.
In Chapter 46, Oxley notes that the environment affects preservation of underwa-
ter materials and specifies the techniques used for conservation in situ. That no under-
water archaeological references are yet available from the National Clearing House for
Archaeological Site Stabilization indicates the reluctance of underwater archaeologists
to recognize the chemical environment of sites. Sites such as the USS Arizona and
Yorktown need to be understood in these terms. Oxley suggests that modeling environ-
mental factors be based on the United Kingdom's Digital Marine Atlas Project, a recent
development in information technology.

521
 Chapter 46

The Investigation of
the Factors That Affect the
Preservation of Underwater
Archaeological Sites

IAN OXLEY

INTRODUCTION

In recent years site environmental assessments have been recognized as a fundamental

part of effective cultural resource management. However, the inclusion of an assess-
ment of the natural and burial environment in underwater archaeological investigation
has not been adopted universally. This paper reviews the various benefits of environ-
mental impact assessments and geographic information systems (GIS), to develop
methodologies to enable such evaluations to be carried out routinely on underwater ar-
chaeological sites in the future .

THE BURIAL ENVIRONMENT

The nature of the burial environment of a site plays a fundamental role in deter-
mining what evidence survives, in what form, and in what position. Certain specific
conditions will promote the survival of particular material types. Studying the nature
Source: Underwater Archaeology Proceedings from the Cotiference on Underwater and Historical Archaeol-
ogy. edited by Donald H. Keith and Toni L. Carrell (1992), pp. 105-110. Reprinted by permission of the Society
for Historical Archaeology, Tucson, Arizona.

523
 524 I. Oxley

and impact of the environment of a site is vital to understanding the quality of the evi-
dence that is eventually recovered. The environment of a site also dictates the tech-
niques and methods that will be most effective throughout the archaeological
investigation, from the initial survey stage to the postexcavation analysis. Objective
data collected at the predisturbance survey stage will therefore indicate the most effec-
tive techniques that should be used on the site.

SITE FORMATION AND SITE CLASSIFICATION MODELS

Site formation and site classification models are becoming an increasingly com-
mon feature of shipwreck studies as tools to help workers understand the formation of
sites and the effect the formation processes have had on the archaeological evidence
contained in the site. The concept of extractive filters has been used to attempt to dis-
tinguish between the effects of natural processes and cultural ones. The formation of a
site can be extremely complicated as, for example, where a shipwreck lies on top of
prehistoric remains (Murphy, 1990).
Classification models have been based on perceived levels of preservation (Ced-
erlund, 1980) or the zonation of the seabed into areas of specific environmental condi-
tions thought to have affected the preservation of particular artifact types (a strategy
used to analyze the distributions of finds in Orange Bay, St. Eustatius, as described by
Nagelkerken, 1985).
Muckelroy concluded that there are several types of "intermediate" sites on
which the remains are neither "perfectly preserved nor smashed to pieces" and he dem-
onstrated the importance for the survival characteristics of a site of the variety of forces
acting on it. This was reflected in the relatively high correlation between the survival of
material and factors relating to the site's position, for example, the extent of the sea
horizon open to the site (Muckelroy, 1977).
It is clear that site assessment studies provide a fundamental role in improving
the viability of any site classification system or site formation study (Gibbins, 1990).

CONSERVATION SCIENCE AND FINDS HANDLING

Conservation science will obviously benefit from an increased knowledge of the

burial environment of objects and materials and more effective conservation treatments
will be developed if the properties of the environment that surrounded an object (and
which contributed to its deterioration) have been considered (Pearson, 1987). The
analysis of concretions and associated corrosion products from nonferrous artifacts
from Australian wrecks, together with a characterization of the marine environment for
each site, contributed toward establishing the previous history of the artifacts for the
period between the wreck and the excavation of the vessel (Macleod, 1991).
In finds handling, it is important to minimize the damage caused by moving the
object from one environment to another, completely different one, during the recovery
process. This can be attempted by reproducing the characteristics of the objects' burial
environment or by applying holding treatments to help the object survive until it can be
safely transported to a conservation laboratory. In order to construct and maintain cor-
rect storage environments, the conservator must have objective information about the
conditions of burial (Carpenter, 1987)
Factors That Affect the Preservation of Underwater Archaeological Sites 525

CULTURAL RESOURCE MANAGEMENT (CRM)

Site environmental assessments should form a fundamental part of any CRM
scheme as such evaluations provide the basis for achieving a better understanding of
the site and its formation and indicate ways in which it can be better managed and pre-
served for the benefit of future generations. Strategies such as the taking of baseline in-
formation followed by periodic monitoring form a fundamental part of the management
plans for underwater heritage parks in many parts of the world.
The investigation of the USS Arizona represents a major landmark in site assess-
ment studies as no one had previously confronted the problem of developing a long-
term preservation program for a whole ship in situ. The program included collecting a
baseline inventory of biological communities on the structure of the 600-ft battleship
that would help determine the biochemical processes impacting the vessel fabric. Sta-
tions were established to enable quantified measurements of the state of deterioration
of structural elements to be collected at periodic intervals (Lenihan, 1989).
An example of the complicated nature of resource management in a archaeology
underwater is the study conducted at Yorktown. Its purpose was to assess the environ-
mental impact of measures taken in order to carry out the archaeological investigations,
particularly the effect of the introduction of the large ferrous metal structure of the cof-
ferdam, and the use of visibility enhancing chemicals (Rodgers, 1989).
In the Legare Anchorage Shipwreck Project, the bottom topography and vegeta-
tion patterns were mapped as part of the initial survey to determine placement of the
excavation units. Site stabilization strategies to encourage the deposition of protective
sediment through the use of artificial and natural seagrasses were also investigated
(Fischer et al., 1984).

PRESERVATION IN SITU AND/OR SITE STABILIZATION

It is well known that artifact deterioration on site can be lessened by action based
on an understanding of the chemical and physical processes of the marine environment
(McCarthy, 1982). The further development of the quantified monitoring of underwater
sites will help conserve archaeological material in situ using techniques such as sacrifi-
cial anode systems that continually "treat" metal objects, for example, cannon on sites
in the John Pennekamp State Park (Bump and Muncher, 1987).
Significant work has been carried out in Australia on the measurement of the fac-
tors on which the corrosion of iron objects is dependent. On-site measurement of corro-
sion potential together with the use of sacrificial anodes have demonstrated the link
between environmental assessments and the benefits of in situ conservation methods in
CRM and development of new conservation treatments: on SS Xantho the predistur-
bance biological, chemical, and electrochemical survey followed by the recovery of the
engine, the treatment of the anchor from HMS Sirius, and the cryogenic deconcreting
of the Trial cannon (McCarthy, 1982, 1988; MacLeod, 1987).
An important technique for the recovery of information from archaeological sites
without incurring the substantial costs of conservation and storage is the reburial of
archaeological material after suitable recording has been carried out. It is likely that
variations on this method of resource management will become increasingly popular in
the future, but its effectiveness depends ultimately on how well the reburial environ-
526 I. Oxley

ment mimics the original burial conditions. This, in turn, depends upon how precisely
the preservation conditions of the site are.known.
In addition, the development of quantified measurements and monitoring tech-
niques is crucial to the success of any predictive survey aimed at identifying areas that
might have a high potential for containing archaeological sites. Various physical,
chemical, and biological parameters have been determined to be important for the
wreck site location as well as for predicting expected states of preservation.
An important initiative in furthering site stabilization studies is the National
Clearing House for Archaeological Site Stabilization (see below for contact address),
which maintains a bibliography intended to support the conceptualization, design, and
development of site stabilization and preservation projects. The bibliography is divided
into four sections: Philosophy, Technical Support, Management Recommendations, and
Practical Applications. It is salutary to note that virtually no references are available for
underwater archaeology at this time.

PRINCIPLES OF ENVIRONMENTAL ASSESSMENT

The perceived reluctance by archaeologists to carry out site environmental
assessments may be related to factors such as a lack of an accepted methodology (for
example, what physical, chemical, or biological factors should be measured and how
often?) and an inbuilt wariness when relating data and terminologies that are derived
from the natural sCiences (such as pH or the measurements of obscure chemical spe-
cies) to archaeological problems. There have been few objective data published that
might be useful for comparing one site against another, with regard to factors such as
the differential preservation of the materials present; therefore, the opportunity of
building up a useful corpus of evidence on the nature of archaeological sites underwa-
ter as a whole has been lost (Wildesen, 1982).
With a view to suggesting an achievable methodology for site environmental
assessments, it is instructive to examine the environmental assessments process that has
emerged from the recent implementation of environmental impact regulations in many
parts of the world. Those organizations that have been forced to carry out environ-
mental assessments because of statutory obligations have had cause to examine the fun-
damental basis of assessments methodology (National Research Council, 1990).
Environmental assessments are the collection of data and information on the basis
of which decisions are made. Good data and information help to make good decisions.
Environmental systems are recognized to be made up of: individual environmental
parameters (any single characteristic of the total environment that can be measured by
an objective methodology, e.g., pH or temperature) and interactions between parame-
ters or components (e.g., degradation processes).
Environmental criteria are selected parameters based on a current understanding
of importance, such as those factors thought to be important for preservation that might
include the criteria that describe sedimentary environments (Robinson, 1981). Termi-
nology must be used carefully as there is a danger that the decision-maker who has to
use the information will be faced with an overload of highly specialized information,
none of which they know how to relate effectively to their problem.
Modeling an environmental resource means essentially constructing a map of
how various components are interrelated, how a change in one component can instigate
Factors That Affect the Preservation of Underwater Archaeological Sites 527

a change in another, the conditions that must be met for the change to occur, and the
rate at which it occurs (Erickson, 1979).

IDENTIFYING PROCESSES

Once the nature of the environment has been assessed, then any potential altera-
tions must be considered. These provide the key to understanding the development of
the environment and can be categorized into five basic ways: introduction, transforma-
tion, translocation, sequestration, and dissipation. All site formation processes, whether
cultural or natural, can be categorized in terms of these alteration types.

INTERDISCIPLINARY COOPERATION

No one individual has sufficient knowledge to identify precisely all of the inter-
connections represented in the complex archaeological site environments so specialists
from other disciplines must be enlisted. Initiatives taken to involve other disciplines in
archaeology underwater are to be applauded, for, although there is a recognized need, it
is not always easily achieved effectively (Smith et aI., 1981; Wildesen, 1982).

UKDMAP

A recent development in the application of information technology in archaeol-

ogy is the availability of geographic information systems (GIS) (Allen et al., 1990).
The United Kingdom Digital Marine Atlas Project (UKDMAP) has been developed by
the British Oceanographic Data Centre with funding provided by various U.K. govern-
ment organizations. UKDMAP is intended to be a reference work on the marine envi-
ronment of the United Kingdom that will be of use to the scientific, educational,
government, and commercial sectors, and it enables widely environmental parameters
to be directly compared. In this case it is possible to compare these parameters with at
least one category of underwater archaeological site as the locations of the UK Pro-
tected Wreck Sites have been included in the data base.
In addition to the presentation of spatially referenced information in the form of
maps, which may be zoomed and/or overlaid on each other, and individual points que-
ried to obtain detailed information, the Atlas presents accompanying textual informa-
tion that enables the user to contact the source of data and technical expertise in the
relevant subject. This feature has proved useful in bringing the underwater cultural
heritage to the attention of other major sea and seabed operators in the United Kingdom
(e.g., commercial fishing, aggregate extraction, and dredging operations).

CONCLUSION

Site environments are complex systems, and although the benefits of environ-
mental assessments have been recognized widely in contributing toward understanding
these complexities, they have not been universally adopted in archaeology underwater.
A full characterization of the physical, chemical, and biological environment of a site,
as it was when it was occupied or active as well as through the later stages that brought
528 I. Oxley

it to the condition in which it was found, should be considered an integral part of any
investigation. What is required is a broader discussion of what is required to be meas-
ured and how the data are to be collected, on all sites, but in ways in which the data can
be directly comparable. Such assessments could then be routinely carried out as part of
preliminary site surveys. It is hoped that established environmental assessments impact
methodologies and the availability of geographic information systems will enable some
of these problems to be addressed.
Information about the National Clearing House for Archaeological Site Stabiliza-
tion and the Archaeological Site Stabilization Bibliography can be obtained from:
Robert M. Thorne, Center for Archaeological Research, University of Mississippi, Uni-
versity, MS 38677.
Further information about UKDMAP can be obtained from: UKDMAP Project
Manager, British Oceanographic Data Centre, Proudman Oceanographic Laboratory,
Bidston Observatory, Birkenhead, Merseyside, UK L43 7RA.

REFERENCES
Allen, K.M., Green, S.W., and Zubrow, E.B.W., 1990,lnterpreting Space: GIS and Archaeology. Taylor &
Francis, London.
Bump, H.D., and Muncher, D.A., 1987, Conservation of Historic Artifacts from Marine and Coastal Environ-
ments. In Underwater Archaeology Proceedings from the Society for Historical Archaeology Conference,
edited by A.B. Albright, pp. 52-53. Savannah, Georgia.
Carpenter, J., 1987, The Uses of Soil Stabilizing Gel Media in the Conservation of Large and Small Shipwreck
Artifacts. International Journal ofNautical A.rchaeology 16(2): 95-107.
Cederlund, C.O., 1980, Systematic Registration of Older Sinkings and Wrecks in Swedish Waters. Interna-
tional Journal q{ Nautical Archaeology 9(2): 95-104.
Erickson, P.A., 1979, Environmental Impact Assessment: Principles and Applications. Academic Press, New
York.
Fischer, G.R., Donnellan, R.A., Skowronek, R.K., Wild, K.S., Jr., Vernon, R.H., and Johnson, R.E., 1984,
Investigation of the Legare Anchorage Shipwreck, Biscayne National Park, Florida. In Search of Our
Maritime Past: Proceedings of the Fifteenth Coriference on Underwater Archaeology, edited by J.W.
Bream, R. Folse-Elliott, C.V. Jackson, III, and G.P. Watts, pp. 145-167. East Carolina University, Wil-
liamsburg, VA.
Gibbins, D., 1990, Analytical Approaches in Maritime Archaeology: A Mediterranean Perspective. Antiquity
64: 376-389.
Lenihan, 0.1 (ed.), 1989, USS Arizona Memorial and Pearl Harbor National Historic Landmark. Submerged
Cultural Resources Study, Southwest Cultural Resources Center Professional Papers No. 23, Santa Fe,
NM.
McCarthy, M., 1982, A Wreck Inspection Program as an Aid to the Coordinated Management of a Large
Number of Wreck Sites. International Journal of Nautical Archaeology II (I): 47-52.
McCarthy, M., 1988, S.S. Xantho: The Pre-Disturbance Assessment, Excavation and Management of an Iron
Steam Shipwreck off the Coast of Western Australia. International Journal of Nautical Archaeology
17(4): 339-347.
MacLeod, 1.0., 1987, Conservation of Corroded Iron Artifacts - New Methods for On-Site Preservation and
Cryogenic Deconcreting. International Journal of Nautical Archaeology 16( I): 49-56.
MacLeod, 1.0., 1991, Identification of Corrosion Products on Non-ferrous Metal Artifacts Recovered from
Shipwrecks. Studies in Conservation 36: 222-234.
MacLeod, 1.0., et al.. 1986, The Excavation, Analysis and Conservation of Shipwreck Sites. In Preventive
Measures during Excavation and Site Protection. ICROM, International Center for the Study of the
Prevention and Restoration of Cultural Property, Rome.
Muckelroy, K., 1977, Historic Wreck Sites in Britain and Their Environments. International Journal ofNautical
Archaeology 6(1): 47-57.
Factors That Affect the Preservation of UndelWater Archaeological Sites 529

Murphy, L.E., 1990, 8SLl7: Natural Site Formation Processes of a Multiple-Component Underwater Site in
Florida. Submerged Cultural Resources Special Report. Southwest Cultural Resources Center, Santa Fe,
NM.
Nagelkerken, W., 19185, Preliminary Report on the Determination of the Location of the Historical Anchorage
at Orange Bay, St. Eustatius, Netherlands Antilles. In Proceedings of the Sixteenth Conference on
Underwater Archaeology, edited by P.F. Johnston, pp. 60-76. Boston.
National Research Council, 1990, Managing Troubled Waters: The Role ofMarine Environmental Monitoring.
National Academy Press, Washington, DC.
Pearson, C. (ed.), 1987, Conservation of Marine Archaeological Objects. Butterworths, London.
Robinson, W.S., 1981, Observation on the Preservation of Archaeological Wrecks and Metals in Marine Envi-
ronments.lnternational Journal ofNautical Archaeology 10(1): >-'14.
Rodgers, B.A., 1989, The Case for Biologically Induced Corrosion at the Yorktown Shipwreck Archaeological
Site. International Journal ofNautical Archaeology 18(4): 335-340.
Wildesen, L.E., 1982, The Study of Impacts on Archaeological Sites. In Advances in Archaeological Method
and Theory Vol. 5, edited by Michael B. Schiffer, pp. 5-96.
 Additional Bibliography
".~i).:
~

for Part VIII

-r/: ..
':.i

Clarke, R.W. and S.M. Blakeshaw (eds.), 1982, Conservation of Iron. National Maritime Museum Monographs
and Reports Number 53, Greenwich.
Coles, John M., 1990, Waterlogged Wood. English Heritage, London.
Cronyn, J.M., 1990, The Elements C?f Archaeological Conservation. Routledge, London.
Doroman, E.A., 1970, Conservation in Field Archaeology. Metheun, London.
Grosso, Gerald H. (ed)., 1976, Pacific Northwest Wet Site Wood Conservation Conference: Proceedings. Dis-
cussions ... vol. 1-2, Neah Bay, Washington.
Hamilton, D.L., 1976, Conservation of Metal Objects from Underwater Sites: A Study in Methods. Texas
Antiquities Commission, Austin.
Lewis, Ralph H .. 1976, Manual for Museums. National Park Service, Government Printing Office, Washington,
DC.
Ofen stein, Sharon K., and Thomas H. Spiers, Jr. (comp.), 1984, Index to the Bulletin of the Association for
Preservation Technology . Association for Preservation Technology, Washington, DC.
Pearson, Colin, 1987, Conservation of Marine Archaeological Objects, Butterworths, London.
Rodgers, Bradley A. 1992. Conservation of Water Soaked Materials Bibliography, Program in Maritime History
and Nautical Archaeology, East Carolina University, Greenville, North Carolina.
Rodgers, Bradley A., 1992, The ECU Conservator's Cookbook. Program in Maritime History and Nautical
Archaeology, East Carolina University, Greenville, North Carolina.
Thomson, Garry. 1978, The Museum Environment. Butterworths, Boston.

531
 Part IX

Interpretation
".~i).:
~

and Exhibition
-r/: ..
':.i

If the research, location, and recovery of cultural materials is seen as science, the art of
archaeology involves interpretation; what do the remains really mean and why? Inter-
pretation is almost the end product of an archaeological investigation because it repre-
sents the story behind the material that was found and how it relates to both the past (its
context) and the present.
Archaeological excavation is a destructive process because the site itself is
destroyed; interpretation is how the archaeologist explains the site that was destroyed
to provide the information. A detailed presentation of what was learned is crucial and
creates a basic distinction between archaeologists and those who recover materials sim-
ply for their financial or curiosity value.
Unfortunately, interpretive statements about the results of archaeological research
do not seem to reach the public very often. The public often receives only media com-
mentary and this is usually preliminary and overstated. After all the research and writ-
ing have been done, the public only has media-level information because reports are
usually filed away or presented as professional papers. Rarely is an archaeological
report a good seller, even if it does reach the publication stage.
Consequently, to reach the public and educate them about archaeology and what
was found is often the task of museums and their exhibits. A close cooperation between
archaeologist and exhibit designer can produce a very worthwhile and informative dis-
play which conveys information about a site. All too often this is not done, but the gen-
eral public is the constituency for whom archaeologists work and prepare their
interpretations. Without popular support, archaeology will wither on the vine because it
will not be sustained by funding, protective legislation, and enforcement of existing laws.

533
534 IX. Interpretation and Exhibition

It is essential that archaeologists get their results before the public in as many ways as
possible. In this section, suggestions are made about how to reach the public.
One journal does provide a great deal of information, both about curation, resto-
ration and interpretation. This is the APT Bulletin, the journal of the Association for
Preservation Technology. A particularly good example of the variety of articles relating
to maritime interpretation and restoration can be found in Volume IX, Number 1
(1987).
 Part IX. A

Interpretation

Ross's examination, in Chapter 47, of the extensive remains of the San Juan serve as a
good example of material-based interpretation. Combining archival research with
remains of the stowage barrels (barrica), he was able to discuss issues such as l6th-
century staved container repair and coopering technology. Stowage pattern of the casks
aboard the San Juan is also apparent from their remains.

535
 Chapter 47

16th~Century Spanish Basque

Coopering Technology

LESTER A. ROSS

INTRODUCTION

In the early 16th century, Spanish Basques began exploring the North Atlantic fisheries
off the coast of present-day Labrador, then known as Terranova. By the 1540s they had
established annual fishing expeditions and had also begun exploiting the vast
untouched whale populations in the area. Leaving from the Basque country in early
summer, whalers arrived in Terranova in midsummer, established shore facilities, and
proceeded to hunt Atlantic right and possibly bowhead whales from small boats. Kills
were towed to shore to be butchered, primarily for their blubber which was minced and
boiled for its oil. After clarifying, the oil was drawn into wooden casks, loaded aboard
ship, and sent back to the Basque country, England, or Holland in early winter when the
whalers left Terranova (Figure 1).
Archival research into this North Atlantic whale fishery by Selma Barkham has
shown that one of the largest whaling stations annually established in Terranova was at
Buttes, now known as Red Bay, Labrador (Figure 2). Buttes, being well protected by a
large offshore island known as Saddle Island, had two good harbors in which as many
as ten galleons could be found at anchor during the whaling season. Toward the end of
one season, a vessel known as the San Juan was blown ashore with its return cargo of
1000 casks of whale oil. During her archival research Ms. Barkham noted the existence
of this wreck and, later in 1978, Parks Canada marine archaeologists under the direc-
Source: Underwater Archaeology: the Proceedings of the Eleventh Conference on Underwater Archaeology,
edited by Calvin R. Cummings (1982), pp. 130-140, Fathom Eight, San Marino, California. Reprinted by per-
mission.

537
~
""
...~" '~~
lc~~~ ' ,'~~ " :-~
\" • {<- . l'..,! :'·' .I
':"'~ 'I
, . ~J." .~SJt,~' ~4I . y, ~ '. " ..
of ~ :'~ 'rl' ;'..,f,' ' .. ,,~ ".~ "., :. ;I. . . .. . ~ ,
~
00
-/.Jc.'j.t>~.i.: ~~ ";'..-.,. r, : :-:. J~
.'': ~
,~ . " , \!,: .' •

"''.t~~~:~"
, , ." i#l'4t'I'"i,:
. ~
~~~.,
. i~· ."·,·

~~ ".

,\
b~ ~
.w..: ~

r
- ~ a ZZi
- =
_ ~... :-:--~.....::.. _~ _ ;n~· 7 .:- ' . ~;.~ _ . ...:-=- --.. : :. -=----
»
6'
<II
Figure 1. Basque shaling in the North Atlantic toward the end of the 17th century (Sanez Reguart 1791-95: PI. LXII), <II
16th.Century Spanish Basque Coopering Technology 539

TERRANOVA

•I

Figure 2. Location of Buttes, Terranova, now known as Red Bay, Labrador. Drawing by S. Epps,
Parks Canada.

tion of Robert Grenier, located the probable wreck of the San Juan along the northern
shore of Saddle Island, just offshore of a number of Basque whaling sites being exca-
vated by Dr. James Tuck of the Memorial University of Newfoundland (Figure 3). The
onshore sites consisted of a number of structures including at least one major trying or
boiling oven complex marked by numerous fragments of ceramic roofing tiles imported
by the Basques to cover their shore-built structures.
The northern shore of the island was rocky and relatively steep, thus allowing the
San Juan to be blown within a few yards of the shoreline. It rests in some 30 ft of water
on its starboard side; during low tide its masts, yards, and portions of the upper decking
may have been above water (Robert Grenier, personal communications). Historical
540 L.A. Ross

•
® J._
Figure 3. Location of the wreck of the San Juan just offshore of Saddle Island, Red Bay, Labrador.
Drawing by S. Epps, Parks Canada.

insurance documents from the 1560s and 1570s indicated that the ship was practically
salvaged immediately after sinking and again in the following year of 1566 (Barkham
and Grenier, 1978-79:61-62). Thus, much of its onboard cargo, equipment, and per-
sonal supplies were removed. However, most of the lower hull remained untouched,
and after 400 years underwater, the below-waterline portion of the ship and its cargo
were extremely well preserved.
Excavations on the wreck revealed that the lower hold of the San Juan was
loaded with at least four layers of casks. On clearing away the fine layer of silt cover-
ing the site, crushed cask parts appeared to litter the entire area. The upper portions of
this staved container layer were very disturbed with staves and head pieces scattered in
a random fashion. Just below this disturbed layer, casks were relatively undisturbed,
although they had all been crushed with' their upper parts having shifted forward and
starboard following the offshore bottom slope and the settling angle of the ship.
Excavations during the 1979 season resulted in the recovery of seven types of
staved containers, including four sizes of casks and three possible types of tubs (Figure
16th Century Spanish Basque Coopering Technology 541

• . .~
• ~
~

Figure 4. Graphic reconstructions of the staved containers recovered ITom the 1979 excavation of
the San Juan (drawings by D. Kappler, Parks Canada). a, bever cask; b, one-third barrica; c, half
barrica; d, barrica; e, mess tub; t: large mess kit; and g. small mess kit.

4a-g). The largest casks were barricas, used to ship whale oil from Terranova to
Europe, with a capacity averaging 211 liters or 56 U.S. gallons. Next in size was the
"half barrica," with a capacity roughly half that of the larger barrica. The third size of
casks were the "one-third barricas," and the smallest cask was a "bever cask" having a
capacity equal to approximately 1 U.S. gallon (Table 1). Hypothesized tubs consisted
of a possible "mess tub" made by cutting a barrica in half, a "large mess kit" made
from a one-third barrica cut in half, and finally a "small mess kit."

Table 1. Hypothesized capacities for 16-century units of measure possible used in association with
whale oil commercea

Liters
0.5275 I Libra
13.1875 25 I Arroba
52.75 100 4 I Quintal
70.33 133.33 5.33 1.33 I "One-third Barriea"
105.5 200 8 2 1.5 I "Half-Barnea"
140.67 266.67 10.67 2.67 2 1.33 I Tercerol
211 400 16 4 3 2 1.5 I Barriea of Whale Oil
422 800 32 8 6 4 3 2 I Bota
844 1600 64 16 12 8 6 4 2 1 Tonelada
'Historic terms (underlined) were provided by Selma Barkham from 16th·century lawsuits in the Real Chancilleria at Val-
ladolid and from notarial documents in the Protocolos de Guipuzcoa at Onate. Other sizes based the hypothesized capacites
of casks from the San Juan, using the average capacity of the barrica as 211 liters.
542 L.A. Ross

Table 2. Wood species identified for the staved

containers from the 1979 excavation of the San Juan
Species Major function
White oak (Quercus sp.) Staves and head pieces
Beech (Fagus sp.) Staves and head pieces
Alder (Alnus sp.) Hoops
Willow (Salix sp.) Hoop bindings
Cork Bungs
Elm (Ulmus sp.) Sample hole pegs

All of these containers were made entirely of wood without the use of metal
hoops or nails. This wood appeared to be of European origin, consisting primarily of
white oak, beech, alder, willow, and cork; with limited use of elm (Table 2).

Barricas (Figure 4d)

Parts from approximately 60 to perhaps 80 of these casks were excavated during
the 1979 season, and the following descriptions and interpretations have greatly
expanded initial interpretations published earlier and based solely on remains recovered
from the 1979 survey (Ross, 1980). Barricas were generally constructed of 20 white
oak or beech staves with two five-piece white oak or beech heads, each held in place
with one white oak or beech head reinforcement, all bound with 20 to 24 single-piece
alder hoops, each bound with two to three willow bindings.
During excavation, complete barricas were often uncovered, some with hoops still in
place. When feasible, these in situ assemblages were raised to the surface and brought to
the field laboratory for more detailed study and recording. Because of underwater working
conditions and budgetary considerations, every barrica could not be totally recorded nor
studied in detail. Only the best preserved and assembled examples were raised intact. Of
the half-dozen or so complete barricas recovered from the 1979 season, most had all of
their staves, both heads, and usually one or two of their head reinforcements. Alder hoops
associated with these barricas were extremely fragile and, during excavation, they continu-
ally fragmented into sections averaging about 6 to 10 inches in length. Hoops brought to
the field laboratory were so fragmentary that most were virtually impossible to reconstruct.
Although staves, head pieces, and hoops comprise the major components of any
cask, there are a number of secondary parts that must also be considered. For barricas,
a single cask could be composed of as many as 170 separate pieces comprised of
staves, head pieces, head reinforcements, hoops, hoop bindings, bungs, vent hole plugs,
various reinforcing pegs, and repair wedges. Attempting to reconstruct a complete cask
with so many separate parts would be an extremely time-consuming proposition, but it
could be done. Efforts for now, however, have been directed toward the creation of
graphic reconstructions where all parts recovered in association are illustrated as such
(Figure 5). Even at this level of interpretation, however, a totally complete barrica has
yet to be reconstructed.
Looking at some of the individual parts of barricas, there are the staves, which
average about 3 ft in length and vary from 2 to 6 inches in width. Analysis of fatty resi-
dues found on the interior surfaces of these staves indicates that they once contained
16th-Century Spanish Basque Coopering Technology 543
10 •

"
..

..
... II II II JO • • J•

,I 1
!
t--
1-:..: ~ .- -- ---
\ ~
- -- j
'-
I F- , 1 I-
It - •
lACK

+ , --IIiJ ~ t_
---
, 1

\\ ,"' :

.i'.
.;

'41 I I .t

II II 10

..
II
IOTTOII

'.
"
Figure 5. Graphic reconstruction of a complete barrica (24M 14M9-1-33) sho\Ning the top and
bottom heads \Nith their head reinforcements (note the bottom head reinforcement has been made
from a reused head middle piece), the 19 cask staves with their assembly marks beginning on bung
stave #1, and shippers' marks on staves # 17-19. Drawing by D. Kappler, Parks Canada.

products (presumably oil) from baleen whales. Each barrica had one bung stave which
allowed the cask to be filled with oil, and a cork or occasionally white oak or beech
bung used to stopper the bung hole. From historic illustrations of casks (e.g., "Tom
Truelove's Kneel," published by Haines and Son, 1795), pieces of cloth are occasion-
ally shown between the bung and its stave, serving as a gasket to help seal the bung
hole. From bung staves excavated from the San Juan, fragments of such cloth gaskets
have been recovered; the cloth appears to be a linen twill, possibly from sailcloth.
Looking at barrica heads, the common style was a five-piece head with a single
head reinforcement. Occasionally, spare or discarded head pieces were also used as
head reinforcements (Figure 5). These head reinforcements were held in place by the
use of multiple pegs set into the ends of adjacent stabes. Very few of these pegs sur-
544 L.A. Ross

SINIL[ LIN[ MAliK

• X· MARK

Figure 6. Graphic illustrations of the four types of cooper's

HE MARK marks observed on staves, and occasionally head pieces, re-
covered from the 1979 excavation of the San Juan.

vived intact, but of the ones that were recovered, it appears that most were made from
alder hoop fragments which were trimmed to form a peg.
Hoops from barricas were made from a single piece of split alder, joined by over-
lapping each end and wrapped with two to three pieces of willow binding. To eliminate
slippage of the overlapped hoop ends, the tips were notched - inner ends having two
to three slanted "V" notches with outer ends having two to three rectangular notches.
Individual willow bindings always began at a rectangular outer notch, terminating near
the end of a slanted "V" inner notch.
Barrica staves and heads were commonly marked with a variety of incised lines,
some of which have been identified as assembly and shippers' marks, and others which
may be coopers' inspection, grading, or markers' marks. During original cask construc-
tion in the Basque country, staves were assembled into cases. Lines were scored around
these cases thus allowing each case to be disassembled so the staves could be bundled
together for shipment to Terranova where they could be reassembled in their original
order (Figure 5).
Most of the beech staves, and a few of the white oak staves, also had very fine lines
incised on their exteriors. There were at least four types of these lines, and they are
hypothesized as being inspection, grading, or makers marks made by master coopers in
charge of stave manufacture (Figure 6). Finally, a few examples of shippers' marks were
also noted (Figure 5). These marks appear to be composed of letters and symbols, and
according to historic Spanish shipping manifests, such marks were used by merchants to
denote ownership of cask contents (following Arnold and Weddle, 1978:360-361). Thus
far, none of the owners of these marks from the San Juan have been identified.

Half Barricas (Figure 4c)

The second size of cask reconstructed from the San Juan was the half barrica, and
only a single stave was recovered. Other than gross size, no other attributes of this type
of cask could be reconstructed.
16th~Century Spanish Basque Coopering Technology 545

One:rhird Barricas (Figure 4b)

The third size of casks recovered were one-third barricas, presumably used for
the transportation of provisions to and from the Basque country. They had approxi-
mately 18 beech staves, two four-piece heads, and 10 to 12 hoops arranged in a variety
of patterns. Only 7 staves from at least three such casks were recovered from the San
Juan, and all were about half as long as the barrica staves. Likewise, head pieces were
also few in number, only 30 being recovered, and the majority came from four-piece
heads which normally appear to have been used without head reinforcements.

Bever Cask (Figure 4a)

The smallest cask recovered was a bever cask (following Kilby, 1971:174 and Pl.
98) with approximately 16 beech staves, two two-piece white oak heads, and eight
hoops. Such casks had a rope sling attached to their bung spouts, and were probably
used by whalers and land crews to carry water and cider. Only four staves and two
complete heads were recovered from a single bever cask.

Mess Tub (Figure 4e)

Other staved container remains from the San Juan were extremely rare. A possi-
ble mess tub for preparing food was made by cutting a barrie a in half, and only two
staves from such a tub have been recovered.

Large Mess Kit (Figure 4f)

Another tub size is a possible large mess kit for serving food, made by cutting a
one-third barrica in half. Only three staves from one such tub have been recovered.

Small Mess Kit (Figure 4g)

Finally, a possible small mess kit for serving individual portions of food was
hypothesized, but only a single stave was recovered, and it could also have belonged to
a funnel.

SIXTEENTH~CENTURYSPANISH
BASQUE COOPERING TECHNOLOGY

One of the more significant historical aspects of these staved containers from the
San Juan is the knowledge that can be obtained regarding 16th-century coopering and
woodworking technology. From detailed studies of the tool marks found on barrie a
parts, most of the woodworking techniques and tools used in their manufacture have
been reconstructed or hypothesized.
From archival evidence it is known that the whale oil barricas were originally
manufactured in the Basque country (Quinn, 1979:Vol. IV, 93-94), and that staves,
head pieces, hoops, and bindings were bundled separately and shipped to Terranova
(Selma Barkham, personal communications). Wood for these casks had to be cut, split,
and dried. Staves appear to have been completely manufactured in the Basque country
546 L.A. Ross

Table 3. Hypothesized Spanish Basque cooper's toolkit

reconstructed from an analysis of the tool marks found on
barrica parts from the 1979 excavation of the San Juan

Hollowing adze Gimlet

Straight adze Knife
Tapered auger Hooping lever
Axe Mallet
Dividers Croze groove plane
Curved drawknife Jointer plane
Straight drawknife Saw
Driver Scribe
Frow or splitting wedge Rope winch

while hoops were only split, and hoop bindings were probably shipped unsplit. Recon-
struction of the actual woodworking techniques demonstrates that construction
sequences were extremely similar to those reported by 18th-century French encyclo-
pedists (Anonymous, 1751-65; Anonymous, 1751-80; Fougeroux de Bondaroy, 1763).
Reconstructing the tool types that must have been used, it is also apparent that the 16th-
century Spanish Basque cooper's toolkit was similar to that of an 18th-century French
cooper's toolkit. A hypothetical toolkit is listed in Table 3.
Exact,styles of tools remain to be determined from historical and archaeological
evidence, but close examinations of the tool marks tell much. For example, the chiv or
hollow on barrica staves from the San Juan must have been cut by a hollowing adze,
while the croze grooves had to be cut with a side-cutting plane. Historical illustrations
show something of the techniques and tools used to cut croze grooves, and from other
historical accounts, a more complete example of a croze groove plane can be recon-
structed. Now all that remains is to find parts to such a tool in the archaeological
record. From land site excavations being conducted on the 16th-century Basque whal-
ing stations on Saddle Island by Dr. Tuck, woodworking tools which might represent
articles from coopers' toolkits have already been recovered, but positive identifications
of the tool types have yet to be undertaken.
Once barrica parts arrived in Terranova they had to be reassembled and the final
casks finished. Damaged staves had to be repaired or replaced, heads had to be
matched to cases of the same size, hoops had to be sized, cut, notched, and bound, and
entire casks had to be assembled and made watertight. Again, the techniques recon-
structed for barricas from the San Juan appear to be extremely similar to those of 18th-
century French coopers, and unless new evidence comes to light, there is little to
demons.trate a major technological change between the 16th and 18th centuries.
One aspect of coopering technology rarely discussed historically has been cask
repair. With barricas from the San Juan it can be demonstrated that worm holes had to
be plugged, damaged stave ends reinforced or repaired, and head pieces occasionally
required special pegs to properly secure each piece within its croze groove. The major
cause for damage to whole casks appears to have been caused by the lifting of casks
with canthooks. By such means, stave ends were cracked outwards along the croze
grooves, and to be repaired, a wedge had to be inserted in place of the broken end,
between the stave and the hoops. When entire ends were replaced, a replacement
wedge had to be used, but when the stave was not cracked severely, then a flat rein-
16th-Century Spanish Basque Coopering Technology 547

forcement wedge could be slipped in between the damaged stave and hoops. When
stave ends were severely worn down by friction, the croze groove had to be moved in
toward the center of the cask. This required the cutting of a second croze groove, thus
double croze grooves occasionally appear in the ends of cask staves, obviously result-
ing in casks with reduced capacities.

STOWAGE OF BARRICAS ABOARD THE SAN JUAN

Loading casks aboard the San Juan does not appear to have been greatly dissimi-
lar to techniques reported for the 18th century (e.g., Chapman, 1968-75:PI. 32). Casks
found adjacent to the hull of the ship were arranged in rows both parallel and perpen-
dicular to the keelson. This first layer was covered by at least three additional layers
with each cask offset from those below and above. In an 18th-century cutaway of a
Spanish whaling ship (Figure 7), the hold is likewise shown with four layers of casks.

+
,
\

Figure 7. Cutaway view of a Spanish whaling ship of the late 17th or 18th century showing the
stowage of whale oil casks in the hold [this illustration depicts the processing of whale oil aboard
ship, a practice apparently developed after the abandonment of the Labrador fishery, perhaps in
the late 17th century (Sanez Reguart I 791-95:PI. LXXII)].
 548 L.A. Ross

These casks, however, are incorrectly depicted as laying at right angles to the keelson
(perhaps an error on the part of the artist). A proper illustration should show the casks
with their heads across the hull. Also missing in this illustration is any evidence of bal-
last and dunnage which were generally packed around cargo.
Aboard the San Juan casks were surrounded with angular limestone ballast, short
wooden logs known as dunnage, and at least four types of wedges or supports, includ-
ing rock wedges, long wooden logs with cutouts to support casks, double-ended log
wedges, and single-ended split log wedges. The most popular type of packing material
appears to have been the ballast, followed by the short wooden wedges.

SUMMARY

The major findings from the 1979 research into the staved containers from the
San Juan are that:
I. Staved containers consisted of at least four sizes of casks and three sizes of tubs.
2. The vast majority of the containers recovered were barricas with an average
capacity of211 liters holding 430 pounds (avoirdupois) of whale oil.
3. Sixteenth century Spanish Basque coopering technology appears to have
been very similar to 18th-century French coopering technology.
This brief article is not intended to serve as the definitive report on the staved
containers from the 1979 excavation of the San Juan. For those interested in further
descriptions and interpretations, a more comprehensive report is currently in prepara-
tion, to be completed in Spring 1980. Additional information may be obtained by writ-
ing the author.

ACKNOWLEDGMENTS
I wish to thank Parks Canada for providing financial support for this project, and I
gratefully acknowledge the professional assistance provided by Selma Barkham (Archi-
vist, Canadian National Archives), Robert Grenier (Marine Archaeologist, Parks Canada),
Dr. Edward Mitchell (Whale Biologist, Canadian Fisheries and Marine Service), and Dr.
James Tuck (Archaeologist, Memorial University of Newfoundland).

REFERENCES
Anonymous, 1751--65, Encyc/opedie, ou dictonnaire raisonne des sciences, des arts et des metiers, par une
societe de gens de lettres, 17 vols. Briasson, Paris.
Anonymous, 1751-80, Recueil de Planches sur les Sciences. les Arts Liberaux. et les Arts Mechaniques. avex leur
Explication and Suite de Recueil de Planches, 12 vols. Paris. Arnold, 1.B.,III, and Weddle, R., 1978, The
Nautical Archaeology of Padre Island: The Spanish Shipwreckv of 1554. Academic Press, New York.
Barkham, S., 1978, The Basques: Filling a Gap in Our History Between Jacques Cartier and Champlain. Cana-
dian Geographic, 96(1 ):8-19.
Barkham, S., and Grenier, R., 1978-79, Divers Find Sunken Basque Galleon in Labrador. Canadian Geo-
graphic, 97(3):60--63.
Chapman, F.H., 1968-75, Architectura Navalis Mercatoria and Tractat om Skepps-Byggeriet. Translated from
the Swedish in 1971 by Adlard Coles, London.
Fougeroux de Bondaroy, A.-D., 1763, Art du Tonnelier. H. L. Guerin and L. F. Delatour, Paris.
16th.Century Spanish Basque Coopering Technology 549

Kilby, K., 1971, The Cooper and His Trade. John Baker, London.
Quinn, D.B. (ed.), 1979, New American World: A Documentary History of North America to 1612, 5 vols.
Amo Press, New York.
Ross, L.A., 1980, Basque Whale Oil Casks from a 16th Century Ship Sunk in Red Bay, Labrador. Research
Bulletin No. 123. Parks Canada, Ottawa.
Sanez Reguart, A., 1791-95, Diccionario historico de los artes de la pesca nacional. por el comisario real de
guerra de marina, 5 vols. Impr. de la Viuda de Joaquin de lberra, Madrid.
 Part XI.B

Exhibition
".~i).:
~

':.i
-r/: ..

In Chapter 48, Arnold and Alsup report on an exhibit about the Spanish jlota sunk in
Texan waters. It represents an impressive job of interpretation for the public with an
emphasis on the "nontreasure" aspects of nautical history.

551
 Chapter 48

Shipwreck! -
The 1554 Flota Exhibit
J. BARTO ARNOLD III

and

A Children's
Museum Exhibit on
the 1554 Flota Wrecks
BECKY ALSUP

The Corpus Christi Museum of Science and History opened a brand-new exhibit in the
spring of 1990. Entitled Shipwreck!, the exhibit tells the story of the three Spanish ships
lost on Padre Island in 1554. The Texas Antiquities Committee (TAC) is the state
agency that conducted the archaeological and historical work on the 1554flota and has
reported on it in a series of publications (e.g., Arnold and Weddle, 1976). The artifacts
remained in the public domain. Therefore, detailed studies of the collection and new
contributions to knowledge of the past continue (e.g., Skowronek, 1987). Public televi-
sion produced a documentary film, Graveyard of the Gulf, that the TAC continues to
circulate on videotape to schools and other interested groups. The public enjoyed a
major traveling exhibition in the middle and late 1970s. Several exhibit formats are
described in another JJNA article focusing on the Austin Children's Museum exhibit on
Source: International Journal of Nautical Archaeology (1992),21(4):343-355. Copyright 1992 by Harcourt
Brace and Company, Limited. Reprinted by permission.

553
554 J.B. Arnold III

the wrecks (Arnold and Alsup, 1992). The present article describes the definitive per-
manent exhibition on the 1554 wrecks. Heeding the proverb about pictures, here the
text is brief, and the plans and photographs carry the load.
Corpus Christi has a population of 274,000 and is about 130 miles north of the
United States-Mexico border. The city sits on a bay of the same name inshore of the
northern end of Padre Island, which is one of a series of barrier islands. The 1554
wreck sites are about 75 miles south of Corpus Christi. The city fathers realize that
tourism is a major factor for their town and, accordingly, have created a first-rate
museum as part of their programme to encourage and entertain visitors.
In 1983, the TAC selected the Corpus Christi Museum as the curatorial repository
for the 1554 collections. Rick Stryker, the museum director, soon began planning for a
big new exhibition (Smith, 1989). The city council allocated $2,000,000 to expand the
museum building, and $500,000 to design and build the exhibition came from fund-
raising efforts by the Friends of the Museum organization. The museum retained Ger-
ard Hilferty and Associates of Athens, Ohio, to design the 4800 ft2 exhibit. Southwest
Museum Services of Houston, Texas, assisted in parts of the design and installation
relating to the new video dramatization on the wrecks and the fate of the survivors. The
video is shown in a viewing area inside. a simulated cargo stack (Figure 1) with a
viewer-selectable soundtrack in English or Spanish.
Two-and-a-half million dollars is a lot of money to spend on an archaeology
exhibit for a medium-sized city. The mayor and city council deserve the highest praise
for their commitment and vision. Perhaps they realize that the payback is direct and
dramatic from the positive economic impact on the city coffers. Throckmorton (1990)
tells us, "A study by the Association of American Museums has shown that each tourist
who visits a museum leaves about 10 dollars in the immediate community's tax struc-
ture." In the first 12 months after Shipwreck! opened in May 1990, the exhibit had
227,000 visitors. This represents $2,270,000 in local tax dollars alone, leaving aside the
money spent in the private sector for food, lodging, and so on. The city government
will recoup its costs for the exhibit very quickly indeed.
The exhibit consists first of artifact displays backed up by graphics and murals to
tell the story of the shipwrecks in their historical context (Figures 1-3). The second
major component is an impressionistic reproduction of part of one of the ships at full
scale. From the deck level upwards the hull is present in ghostly shades. A strictly real-
istic facsimile was not attempted because construction details of mid-16th-century naos
are little known. The intention is for the visitor to stand on the deck and get a vivid im-
pression of the incredibly small size of the ship that carried about 100 passengers and
crew on a several-month-Iong voyage across the Atlantic. Stuffed rats in the corners
and plastic cockroaches here and there speak silently of the conditions on board.
Topics covered by the exhibit include the shipwreck story, the 16th-century econ-
omy with regard to Spanish colonialism, ship technology and construction, armament,
"how we know" (i.e., archaeological and historical research and analysis), artifact con-
servation, navigation, cargo, and life at sea.
In summary, the Padre Island shipwreck story began by illicit treasure salvage, as
is so often the case. The State of Texas reacted in the short term to clean up the mess
but, as is often not the case in other places, took positive long-term steps. The Texas
Antiquities Code stopped destructive commercial activities on historic wrecks and set
up a scientific research and management programme. The TAC excavated a second
 ~
"~
&
I
;1
(\)

V1
V1
,J::o.
\
(\ -n
AREAS: , \\\ """"L
~~ ~
[N'1"JI"t Loe,., I, \ m
KleMa. JIIIG,
~
" ~-'<~ .. _
[.
!lUUI"
f!n"-.ooysnOfl' _,'
,.,. l$JIL\.J M* : ~=;;':..
.' ,f '

L
~~~I:~~I
t .. . ~) f1.!t..~~
=
I '\... ff", E"A"J~
,';, AR.
'I 'q~
~~ ',
,/
/f1"
Y ,~ .....
L!
!Lt~,,~/ I A"~ [
&-II!I"~ -.....

-h-
Figure 1. Floor plan of the Corpus Christi Museum's Shipwreck! exhibition on the 1554 Aota wrecks of Padre Island. (Scale I cm = O.5m.) (Plan drawn V1
V1
by Gerard Hilferty and Associates.) V1
 ,•
....
,- - ... _-- ·, ~~", '1..
I VI
I
I
... ,
-""""- VI
0\

fc\. ~C.T'IO't.I/1iI..EVN"10 ....

~~II 'V I"'-o-

. ".... -tt .1..... ::d~

....., ..
(-
__
..............,
fo\
~~-I \ot· • • -O·
I"E\ ""c:no-J/' ...."""""
~ ..c:.a.L.&. IVI-· · ";oa

c:...
~
~
o
Figure 2. Cross-section drawings for the partial reconstruction of one of the Padre Island naos (Area I in Figure I). Visitors walk on the deck, view exhibits, 0:::
and get a good idea of how small and crowded the ships must have been. (Scale I cm = 0.25 m.) (Plan drawn by Gerard Hilferty and Associates.)
 til
~ ::r
~.~

tiI"""', .... c. __ oe.

~ ....
.... '1 .

1
I
"".-
;i
(\I
~~c:,~"""
V1
V1
~
"""1~~"_"""""-O
"T1
fi;::\ E.\.Ev...,.", ...
\!:!J" ~ ,,,, .,,, ,'~.

oDO&.~~""''''-'''''''''''' ___ ~
~
m
~
[.
~,- - ;.;~
... ~ ' TO
~ ,-
== ----.1 ...,
:: .=--~ , ...
=:'~C~

f'I..J>oN - So-I,p tEe,", !\IOk(x.... PlA~O!SM

~c. ..... · ,~ 6

( i\ 'Lt."""'''''
~-""""~~-';" " -O'

Figure 3. Plan and elevation views of one of the exhibit areas containing the San Esteban's keel remains (Rosloff and Arnold, 1984), an anchor and iron fas- V1
V1
teners and fittings (Area 4 in Figure I). (Scale I cm = 0 .25 m.) (Plan drawn by Gerard Hilferty and Associates.) --.)
558 B. Alsup

ship and carried on the project to several socially and culturally valuable conclusions.
The Shipwreck! exhibit is the most recent and large-scale positive result. Not only does
the public gain by the exhibit in many ways, but also the investment of tax dollars has
been repaid and the exhibit will continue to generate actual monetary dividends for the
public coffers in the future.

ACKNOWLEDGMENTS

All figures in this article are courtesy of Rick Stryker, director, Corpus Christi
Museum. Plans are by Gerard Hilferty and Associates.

REFERENCES
Arnold, 1.B., III, and Alsup, B., 1992, A Children's Museum Exhibit on the 1554 Flota Wrecks, /JNA
21 :350-353.
Arnold, 1.B., III, and Weddle, R.S., 1976, The Nautical Archaeology of Padre Island: The Spanish Shipwrecks
of 1554. Academic Press, New York.
Rosloff, J., and Arnold, 1.B., III, 1984, The keel of the San Esteban (1554): Continued Analysis. /JNA
13:287-296.
Skowronek, R.K., 1987, Ceramics and Commerce: The 1554 Flota Revisited. Historical Archaeology
212:101-111.
Smith, H.A., 1989, The Building of the Spanish Shipwreck Exhibit at the Corpus Christi Museum. In Under-
water Archaeology Proceedings from the Society for Historical Archaeology Conference, edited by
1.B. Arnold, III. Baltimore.
Throckmorton, P., 1990, The World's Worst Investment: The Economics of Treasure Hunting with Real Life
Comparisons. Underwater Archaeology Proceedings from the Society for Historical Archaeology Con-
ference, edited by T. Carrell. Tucson, Arizona.

* * * * *

The fruits of the archaeologist's labour, if one is fortunate, go beyond the academic
realm of journal articles and monographs. A museum exhibit, for example, can reach
far more people at many different levels of interest and awareness of archaeology.
There have been four formal exhibits on the 1554 jlota wrecks off Padre Island, Texas
(Arnold and Weddle, 1976). In this article we will briefly mention all four and then
describe the Austin Children's Museum exhibit.
From 1977 to 1981, the Texas Antiquities Committee (T AC) sponsored the first
one, a large traveling show of the 1554 wreck materials entitled Treasure, People,
Ships and Dreams. The exhibit included an anchor, cannons, the San Esteban's keel
and stem-post remnants, and 23 large exhibit cases with many artifacts, extensive text
and graphics. There were 20 showings at museums in all parts of Texas. After the tour,
this exhibit went up permanently at the Corpus Christi Museum where it stayed until a
purpose-built addition to that institution's building made room for the new, more exten-
sive Shipwreck! exhibit. Shipwreck! had 227,000 visitors in the 12 months from its
opening in May 1990. The exhibit located at the Corpus Christi Museum has been
described by Smith (1989).
The Treasure, People, Ships and Dreams exhibit was designed and built for the
TAC by the Institute of Texan Cultures (lTC), a unit of the University of Texas at San
A Children's Museum Exhibit on the 1554 Flota Wrecks 559

Table I. Comparison of four 1554 flota exhibits

Treasure, People, Ships and Dreams (Texas Antiquities Committee)
Contents: 23 cases plus major free standing artifacts
Size: 2400 sq. ft. minimum
Weight: 24,000 Ibs.
Production cost: $75,000
Small Version - Treasure, People, Ships and Dreams (Institute of Texan Cultures)
Contents: 12 panels
Size: 400 sq. ft.
Weight: 900 lbs.
Production cost: unknown
Shipwreck! (Corpus Christi Museum)
Contents: four major areas including 60 ft. deck level replica of a nao
Size: 4,800 sq. ft.
Weight: N/A, not a traveling exhibit
Production cost: $500,000 plus $2,000,000 building addition
Treasure, People, Ships and Dreams (Austin Children's Museum)
Contents: 12 narrative panels and five activity areas
Size: 1,500-2,000 sq. ft.
Weight: 4,900 1bs.
Production cost: $10,000

Antonio. They used the same graphics as the large traveling exhibit and a few replica
artifacts to produce a small, 12-panel version suitable for exhibition in schools or shop-
ping malls. This small version of Treasure, People, Ships and Dreams was produced in
five copies, two of which are still circulated by the ITC. It formed the nucleus for the
Austin Children's Museum (ACM) traveling exhibit for children that is the subject of
this paper. For a comparison of the several exhibits see Table 1.
Importantly, museum exhibits are a means of providing the public with access to
historic shipwreck sites. Not everyone is a diver, after all. Appropriate nondestructive
public access is a requirement of the new U.S. shipwreck legislation.
Children's museums are a phenomenon that is rapidly growing in popularity with
family audiences and as adjuncts to more structured learning situations such as school
classrooms (Pitman-Gelles, 1981; Secor et al., 1987; Craig, 1988). Their focus is on
providing "hands-on" experiences with three-dimensional forms rather than the collec-
tion and display of objects. Activities are frequently presented in contextual environ-
ments, where learning can be related to real experiences. For young visitors and their
families, children's museums encourage curiosity and creativity, promote cross-cultural
awareness, and help demystify science. For archaeologists, they offer an unparalleled
opportunity of introducing to the next generation the value of scientific preservation of
irreplaceable cultural resources, such as historic shipwreck sites.
In the ACM version of Treasure, People, Ships and Dreams, the lTC's panels
provide both the informational backbone as well as the visual graphic background. The
panels tell the story of the 16th-century voyage, shipwrecks, and Spanish salvage ef-
forts, and of the modern recovery process. "Hands-on" exhibit components developed
by ACM allow children to become participants in the story and to investigate the
answers to several questions: What was life like aboard one of these ships? How did
ships of this time navigate on the sea? What caused these ships to wreck? How are his-
560 B. Alsup

toric shipwreck sites explored and mapped using modem underwater equipment? How
do scientists recover artifacts from concretions brought up from the shipwreck? What is
the value of items that are found, be they rusty nails or gold bars? How do we know
what we know about ships and people who sailed 400 years ago?
Visitors on scheduled tours are oriented to the exhibit with a short video version
of a narrated slide show produce by the ITC with the original exhibit. The video intro-
duces the story line of the exhibit and allows museum docents to engage young chil-
dren in the mystery of thinking about the past. To consider life aboard, children board a
scale facsimile of the stem of one of the ships, complete with working rudder and
whip staff. Donning period costumes, they may choose to assume the duties of various
crew members as described in supplementary text panels. At a nearby activity station,
simple materials are available to make representations of some of the navigational tools
of the time. At another station with a background of the sounds of a storm, Velcro-
backed symbols can be arranged on a carpet-covered board to follow a "recipe" for a
storm. In the underwater recovery area, "divers" gear up in scuba wet-suit tops and
masks to search the seabed (sand box) for artifacts (metal washers, etc.), depositing
their finds in a box to be taken to the conservation lab and mapping the location on a
log sheet. In the lab, "scientists" wearing lab coats and safety goggles use blunt hand
tools to recover metal objects from plaster-of-Paris mock concretions.
Archaeologists often decry the easy glory and public acclaim that accrue to the
treasure hunter in spite of the destructive nature of his game. The ACM exhibit gave us
a chance to engage children in the excitement of discovery while impressing on them
the value of the scientific approach to recovery and preservation. The distinction
between the two is implicit throughout the exhibit and was highlighted in the original
ACM installation by a wall graphic depicting a grid map of an underwater recovery site
bearing the title "Treasure Hunters or History Keepers?" This participatory exhibit was
premiered at the ACM from October to December 1987 where it was enjoyed by about
14,000 visitors. Since then almost 75,000 people have experienced it on tour in venues
from Florida to Ohio. It remains popular with family audiences because it tells a fasci-
nating story and involves children in multidisciplinary learning activities as they
become engaged in the archaeologists' detective work of piecing together history.
The value of presenting to children the lessons of archaeology is manifest. The
ACM exhibit on the 1554 flota wrecks is an excellent example of this kind of effort.
We urge our colleagues to keep in mind this avenue for reaching the public.

REFERENCES
Arnold, 1.8.. III, and Weddle, R.S., 1976, The Nautical Archeology of Padre Island: The Spanish Shipwrecks of
1554. Academic Press, New York.
Craig, T.L., 1988, Afterschool Adventures. Museum News July/ August:51-53.
Pitman-Gelles, 8., 1981, Museums Magic & Children. The Association of Science-Technology Centers, Wash-
ington, DC.
Secor, 1.A., Skolnik, L., and Bennett-Mendez, K., 1987, Philosophy and Fun at the Staten Island Children's
Museum. Museum News June:36--41.
Smith, H.A., 1989, The Building of the Spanish Shipwreck Exhibit at the Corpus Christi Museum. In Under-
water Archaeology Proceedings from the Society for Historical Archaeology Conference, edited by
J.B. Arnold, III. Baltimore.
 Bibliography for Part IX

Aston, M., 1985, Interpreting the Landscape: Landscape Archaeology in Local Studies. London.
Casson, Lionel, 1987, Hulls and barrels: underwater archaeology's vital contributions to the history of naval
technology. Tropis II Second International Symposium on Ship Construction in Antiquity, edited by
Harry Tzalas, pp. 85-86. Greek Ministry of Culture, Delphi.
Coles, John, 1979, Experimental Archaeology. Cambridge University Press, New York.
Guilmartin, John F., Jr., 1988, Early Naval Ordnance and European Penetration of the Caribbean; the operational
dimension. International Journal ofNautical Archaeology 17 (I): 35-53.
Gillmer, Thomas, 1987, Authentic replica ships: Theory and practice. Tropis II Second International Symposium
on Ship Construction in Antiquity, edited by Harry Tzalas, pp. 207-212. Greek Ministry of Culture, Del-
phi.
Hodder, Ian (ed.), 1987, The Archaeology of Contextual Meanings, Cambridge University Press, New York.
Katzev, Michael, 1987, An analysis of the experimental voyages of Kyrenia II. Tropis II Second International
Symposium on Ship Construction in Antiquity, edited by Harry Tzalas, pp. 245-56. Greek Ministry of
Culture, Delphi.
Leone, Mark P. and Parker B. Potter, Jr., 1988, The Recovery ofMeaning, Smithsonian Institution Press, Wash-
ington D.C.
Lewis, Ralph H., 1976, Manualfor Museums. National Park Service, Government Printing Office, Washington,
DC.
Sarasan, Lenore and A.M. Neuner, 1983, Museum Collections and Computers. Association of Systematics Col-
lections, Lawrence, Kansas.
Switzer, David c., 1981 , Interpretation of the stem area of the Privateer Defence. Underwater Archaeology:
The Challenge Before Us - Proceedings of the Twelfth Conference on Underwater Archaeology, edited
by Gordon P. Watts, pp. 144-50. Fathom Eight, San Marino, California.
Steffy, J. Richard, 1994, Wooden Ship Building and the Interpretation o.fShipwrecks. Texas A & M University
Press, College Station, TX.
Thomson, Garry, 1978, The Museum Environment. Butterworths, Boston.

561
 Conclusion

Like all written materials, this work is part of a continuum that must be updated. The
text and readings were designed for a course in the history and theory of maritime
archaeology. Initial readings were associated with faculty lectures and student topical
presentations. The readings served as a practical introduction to the field of maritime
archaeology as well as sources for topical information on different types of sites, equip-
ment and approaches. The course has changed and will continue to change as more
information becomes available.
Ideally, this text will be used in conjunction with Robert Schuyler's Historical
Archaeology which contains more practical and theoretical reading relevant to mari-
time work. Other useful readings will continue to appear in the International Journal of
Nautical Archaeology and the Proceedings of the underwater sessions at the meetings
of the Society for Historical Archaeology.
Students who use this book will have a model for conducting documentary and
field research. Aspects of the tools available for their field research are also presented in
such a way as to allow a determination of their utility in a real world where time, fund-
ing, and specialized equipment are limited or unavailable. Guidance for determining a
universe with regard to suggesting importance or significance is also included because so
much research is dependent on funding related to cultural resource inventories.
We wish that our readers will regard this collection of readings as the starting
point for organizing a professional library and for a way of thinking about examining
and interpreting the resource base associated with maritime lifestyles. It is not the final
word but it does include materials we have found useful in our own work and in the
guidance of our students.

563
 Appendix A

Archaeological Organizations
and Journals

The following list is not meant to be all inclusive nor advocate a particular emphasis.
Organizations and journals listed below have proven useful for maritime archaeologists
over the years and will likely continue to be so. They provide resources as well an out-
lets for publishing information recovered through documentary and archeological
means. Details about subscriptions and article submission can be obtained using the
addresses which are included.
Organization Journal
American Institute for Archaeology Archaeology and Journal oj Field Archaeology
656 Beacon Street
Boston, Massachusetts 02215-20 I 0
Association for Preservation Technology Journal of the APT
1100 17th Street NW
Suite 1000
Washington, DC 20036
Australian Institute for Maritime Archaeology The Bulletin of the AlMA
clo Western Australian Maritime Museum
Cliff Street
Fremantle, Western Australia 6160
Australian Society for Historical Archaeology The Bulletin of/he ASHA
Department of History
University of Sydney
Sydney, New South Wales
Australia

565
566 Archaeological Organizations and Journals

Society for Nautical Research The Mariner s Mirror

Department of History
University of Exeter
Exeter, Devon EX4 4RJ
England
International Maritime Economic History Association International Journal of Maritime History
Maritime Studies Research Unit
Memorial University of Newfoundland
St. John's, Nfld. AIC 5S7
Nautical Archaeology Society International Journal of Nautical Archaeology
206 Moorview Way
Skipton, North Yorkshire BD23 2TN
National Maritime Historical Society Sea History
5 John Walsh Blvd
P.O. Box 68
Peekskill, New York 10566
Peabody Institute of Salem and Essex Institute The American Neptune
East India Square
Salem, Massachusetts 01970-0783
Society for Historical Archaeology Historical Archaeology
P.O. Box 30446
Tucson, Arizona 85751-0446
SPAFA (Thailand) The SPA FA Digest
Director: Pisit Chaerowaongsa
Regional Office, SPAFA Building
81/8 Sri Ayuttahaya
Sam sen Theves, Bangkok 10300
 Appendix B

'.f" .,:
'"~e_":
Ii' ..
~ Supplemental Bibliographical
Sources for 17th through
fi"~l 19th Centuries

GENERAL MARITIME ARCHAEOLOGICAL TEXTS

Bass, George, 1972, A History ofSeafaring Based on Underwater Archaeology. Walker and Company, New York.
Bass, George, 1975, A rchaeology Beneath the Sea. Walker and Co., New York.
Bass, George, 1988, Ships and Shipwrecks of the Americas. Thames and Hudson, New York.
Dean, Martin and Ben Ferrari, 1992, Archaeology Underwater: The NAS Guide to Principles and Practice.
Nautical Archaeology Society, London.
Filipowiak, Wladyslaw, 1996, Protection of Maritime and River Cultural Heritage. Proceedings of the Third
Conference of the Polish Maritime and River Museology, Szczecin, Poland.
Green, Jeremy, 1990, Maritime Archaeology A Technical Handbook. Academic Press, New York.
Greenhill, Basil, 1988, Archaeology ()fthe Boat: A New Introductory Study. London.
Muckelroy, Keith. 1978, Maritime Archaeology, Cambridge University Press, New York.
Throckmorton, Peter, 1970, Shipwrecks and Archaeology, Atlantic Monthly Press, Boston, Massachusetts.
Throckmorton, Peter, 1987, The Sea Remembers. Mitchell Beazley Publishers, London.
UNESCO, 1972, Underwater Archaeology - a Nascent Discipline. UNESCO, Paris.

PREHISTORIC ASIAN, EUROPEAN, AND AMERICAN

Adney, E. T. and H. I. Chapelle, 1964, The Bark Canoes and Skin Boats of North America, Smithsonian Insti-
tution, Washington D.C.
Edwards, C. R., 1965, Aboriginal Sail in the New World. Southwest Journal of Archaeology 21: 351-358.
Hornell, J., 1946, Water Transport: Origins and Early Evolution. Cambridge University Press, Cambridge.

567
 568 General Bibliographical Sources for 17th through 19th Centuries

Johnstone, Paul, 1980, The Sea-Craft of Prehistory, Routledge and Kegan Paul, London.
Lao Bomin, 1984, A 7000 Year-old Oar. Ancient Discoveries in Chinese Archaeology. Foreign Language Press,
Beijing.
Meggers, B. J., 1975, TransPacific Origin of MesoAmerican Civilization: a Preliminary Review of the Evidence
and its Theoretical Implications. American Anthropologist 77 (I): 1-27.
Thompson, J. E. S., 1949, Canoes and Navigation of the Maya and their Neighbors. Journal of the Royal
Anthropological Institute 79:69-78.
Waterman, T. T. and G. Coffin, 1920, Types of Canoes on Puget Sound. Indian Notes and Monographs Heye
Foundation, New York.

DROWNED TERRESTRIAL SITES

Blawatsky, Vladimir D., 1972, Submerged sectors of towns on the Black Sea coast. Underwater Archaeology
-A Nascent Discipline. pp. 115-122, UNESCO, Paris.
Carrell, Toni, Sandra Rayl and Daniel Lenihan, 1976, The Effects of Freshwater Inundation ofArchaeological
Sites Through Reservoir Construction: A Literature Search. Cultural Resources Management Division
(Archaeology), National Park Service, Washington, D.C.
Marx, Robert, 1972, The submerged remains of Port Royal, Jamaica. Underwater Archaeology - A Nascent
Discipline. pp. 139-146, UNESCO, Paris.
Romero, Pablo Bush, 1972, The sacred well ofChichen-Itza and other freshwater sites in Mexico. Underwater
Archaeology-A Nascent Discipline. pp. 147-154, UNESCO, Paris.
Ruoff, Ulrich, 1972, Palafittes and underwater archaeology. Underwater Archaeology - A Nascent Discipline.
pp. 123-138, UNESCO, Paris.

MEDIEVAL

Barkham, M. M., 1981, Report on 16th Century Spanish Basque Shipbuilding c. 1550 to c. 1600. Parks Canada,
Ottawa, Ontario.
Brogger, A. W. and H. Shetelig, 1976, The Viking Ships. Arthur Vanous Company, Riveredge, NJ.
Bruce-Mitford, R. L. S., 1978, The Sutton Hoo Ship-Burial. British Museum, London.
Crumlin-Pedersen, Ole, 1970, The Viking Ships of Roskilde. National Museum Monographs and Reports I:
7-23. Copenhagen.
Evans, A. C., 1986, The Sutton Hoo Ship-Burial. British Museum, London.
Hornell, J., 1938, British Coracles and Irish Curraghs. Society for Nautical Research, London.
Olsen, 0 and O. Crumlin-Pedersen, 1958, The Skuldelev Ships (I) Acta Archaeologica 29.
Olsen, 0 and O. Crumlin-Pedersen, 1967, The Skuldelev Ships (II) Acta Archaeologica 48.
Olsen, O. and O. Crumlin-Pedersen, 1978, Five Viking Ships from Roskilde Fjord. The National Museum,
Copenhagen, Denmark.
Shetelig, H. and F. Johannessen, 1930, Das Nydamschiff. Acta Archaeologica I.

SEVENTEENTH~CENTURY EUROPEAN AND AMERICAN

Albright, Alan B. and J. R. Steffy, 1979, The Brown's Ferry Vessel, South Carolina. International Journal of
Nautical Archaeology 8: 121-142.
Bass, George, 1988, Ships and Shipwrecks of the Americas. Thames and Hudson, New York.
Clifford, Sheila A., 1991, A Preliminary Report on a Possible 17th Century Shipwreck at Port Royal, Jamaica.
Underwater Archaeology Proceedings of the Conference on Underwater and Historical Archaeology.
edited by John D. Broadwater, pp. 80-83. Society for Historical Archaeology, Tucson, Arizona.
Curtis, Julia B., 1985, Sherds and Shipwrecks: The Dutch and the China Trade, 1600-1650. Proceedings of the
Sixteenth Conference on Undenvater Archaeology. edited by Paul F. Johnston, pp. 26--29. Society for
Historical Archaeology, Tucson, Arizona.
Einarsson, Lars, 1987, HMS Kronan: Underwater Archaeological Investigations of a 17th Century Man of
War. Underwater Proceedings from the Society for Historical Archaeology. edited by Alan B. Albright,
pp. 99-102. Society for Historical Archaeology, Tucson, Arizona.
AppendixB 569

Fleetwood, Rusty, 1995, Tidecra/t. The Boats of Lower South Carolina. Georgia and Florida. WBG Marine
Press, Tybee Island, GA.
Lundeberg, P. K., 1966, The Continental Gunboat "Philadelphia" and the Northern Campaign of 1776. Smith-
sonian Institution, Washington, D.C.
Solecki, Ralph S., 1974, The Tiger, An Early Dutch 17th Century Ship, and an Abortive Salvage Attempt.
Journal ofField Archaeology I: 109--116.
Thompson, Bruce F., 1985, Rigging Elements from a 17th century Portuguese Frigate. Proceedings of the Six-
teenth Conference on Underwater Archaeology. edited by Paul F. Johnston, pp. 100-\02. Society for
Historical Archaeology, Tucson, Arizona.

EIGHTEENTH,CENTURY EUROPEAN AND AMERICAN

Baker, William A., 1966, Sloops and Shallops. Little, Brown& Company, New York. Reprinted 1988. University
of South Carolina Press, Columbia.
Bellico, Russell P., 1992, Sails and Steam in the Mountains-A Maritime and Military History q( Lake
George and Lake Champlain. Purple Mountain Press, Fleischmanns, New York.
Cain, E., 1984, Ghost Ships Hamilton and Scourge: Historical Treasures from the War of 1812. Toronto, Ontario.
Chapelle, Howard I., 1935, The History of American Sailing Ships. Bonanza Books, New York.
Chapelle, Howard I., 1967, The Search for Speed Under Sail. Norton, New York.
Chapelle, Howard I., 1975, The History of the Sailing Ship. Arco Publishing Company, New York.
Crisman, Kevin, 1986, q( Sailing Ships and Sidewheelers: the History and Nautical Archaeology of Lake
Champlain, edited by Giovanna Peebles. Division for Historic Preservation, Agency of Development
and Community Affairs, Burlington, Vermont.
Dean, N. and R. Snow, 1985, Manhattan's Mystery Merchant Ship. Wooden Boat 63:96-104.
Dedszus, Alfred and Ernest Henriot, 1986, Dictionary of Ship Types. Conway Maritime Press, London.
Fincham, John, 1851, A History of Naval Architecture. Whittaker and Company, London.
Goodwin, Peter, 1987, The Construction and Fitting of the English Man of War. 1650-1850. Conway Maritime
Press, London.
Hahn, Harold M., 1981, The Colonial Schooner 1763-1775, Naval Institute Press, Washington, D.C.
Henderson, James, 1972, Sloops and Brigs. Adlard Coles, London.
Howard, Frank, 1979, Sailing Ships of War. 1400-1860. Conway Maritime Press, London.
MacGregor, David R., 1988, Fast Sailing Ships: Their Design and Construction. 1775-1875. Naval Institute
Press, Annapolis, Maryland.
MacGregor, David R., 1980, Merchant Sailing Ships 1775-1875: Their Design and Construction. Model and
Allied Publications, Watford, England.
Morris, John William, III, 1991, Site 44Y088: the Archaeological Assessment of the Hull Remains at Yorktown,
Virginia. MA Thesis, Department of History, East Carolina University, Greenville North Carolina.
s
Mountaine, William, n.d., The Seaman Vade-Mecum and Defensive War by Sea. New England and Virginia
Company, Salem, Massachusetts. (Reprint of 1781 edition published in London.)
Ollivier, Blaise, 1992, 18th Century Shipbuilding: Remarks on the Navies of the English and Dutch from
Observations made at their Dockyards in 1737. Jean Boudriot Publications, Rotherhite, England.
Steffy, J. Richard, 1978, Construction Details of the Brown's Ferry Ship. Beneath the Waters of Time: The Pro-
ceedings of the Ninth Conference on Underwater Archaeology, edited by J. Barto Arnold, pp. 53-54.
Texas Antiquities Commission, Austin, Texas.

NINETEENTH,CENTURY EUROPEAN AND AMERICAN

Sailing Vessels
Chapelle, H., 1988, The Baltimore Clipper. Dover, New York.
Lubbock, B., 1946, The China Clippers, Brown and Ferguson, Glasgow, Scotland.
MacGregor, David R., 1984, Merchant Sailing Ships 1850-1875: Heyday of Sail. Naval Institute Press,
Annapolis, Maryland.
Paasch, Capt., 1908, Dictionary q(Naval Terms - English-French-<lerman-Spanish-1talian. George Philip &
Son, Ltd., Liverpool.
570 General Bibliographical Sources for 17th through 19th Centuries

Steam Vessels
Bright, Leslie S., 1989, Recovery and Restoration of Steam Machinery from the Snagboat General H.G. Wright.
Underwater Archaeology Proceedings first Joint Conference etc, edited by 1. Barto Arnold, pp. 141-43,
Society for Historical archaeology, Tucson, Arizona.
Canney, Donald L., 1993, The Old Steam Navy. Naval Institute Press, Annapolis, Maryland.
Foster, Kevin J., 1988, Marine Steam Engines, 1807-1875: A Typology. Proceedings of the Seventeenth Annual
Conference on Underwater Archaeology. Edited by John W. Foster and Sheli O. Smith, Coyote Press,
Salinas, California.
Gardiner, Robert, 1993, The Advent ofSteam. Naval Institute Press, Annapolis, Maryland.
Hawkins, N., 1987, New Catechism ofthe Steam Engine. Lindsay Publications, Manteno, Illinois (originally pub-
lished 1904, Theodore Audel, New York).
Johnston, Paul F., 1983, Steam and the Sea. Peabody Museum of Salem, Salem, Massachusetts.
Lardner, Dr., 1990, The Steam Engine. Lindsey Publications, Manteno, Illinois (Originally pub\. 1848, John
Weale, London).
Morrison, John H., 1967, History ofAmerican Steam Navigation. Argosy-Antiquarian Ltd, New York.

ARTIFACTS

Baumgarten, Linda, 1986, Eighteenth-Century Clothing at Williamsburg. Colonial Williamsburg Foundation,

Williamsburg, VA.
Caruana, Adrian B., 1992, Tudor Artillery. Museum Restoration Service, Bloomfield, Ontario.
Cotterell, Howard H., 1963, Old Pewter: Its Makers and Marks in England and Ireland. Olympic Marketing,
New York.
Godden, Geoffrey A., 1964, Encyclopedia of British Pottery and Porcelain Marks. Bonanza Books, New York.
Noel Hume, Ivor, 1970, A Guide to Artifacts o.fColonial America. Knopf, New York.
Johnson, Mary M., Judy Forbes, Kathy Delaney, 1982, Historic Colonial French Dress. Ouabache Press, West
Lafayette, IN.
McConnell, David, 1988, British Smooth-Bore Artillery: A Technological Study. National Historic Parks and
Sites Environment Canada, Parks, Ottawa, Ontario.
Neumann, George C, 1967, The History of Weapons of the American Revolution. Bonanza Books, New York.
Neumann, George C., 1973. Swords and Blades of the American Revolution. Stackpole Books, Harrisburg, PA.
South, Stanley, 1972, Evolution and Horizon as Revealed in Ceramic Analysis in Historical Archaeology. The
Conference on Historic Site Archaeology Papers 6:71-116.
Warwick, Edward, Henry C. Pitz, Alexander Wyckoff, 1965. Early American Dress. Bonanza Books, New York.
 '.~~.' About the Contributors
~

o i' .
q;,O_' .': .
.j'~l

l. RENE BEAUCAIRE. At the time of preparing this paper, Dr. Rene Beaucaire was a terrestrial archae-
ologist working on terrestrial prehistoric sites. The Society of the Friends ofVieil Istres began to exam-
ine the shoreline and then, by a logical process, moved into exploring the water. (This paper was
translated by Thomas Babits, currently Head, Quality Assurance and Methods, Collective Localization
Satellite, Toulouse, France.
2. WILLIAM A. BAKER. Baker earned the S.B. degree from the Massachusetts Institute of Technology.
After a career in shipbuilding, this well-known naval architect worked as curator at the Francis Russell
Hart Nautical Museum at M.I.T. Baker recognized the important contribution wrecks could provide for
understanding naval architectural evolution and recreating historical ships. He compiled plans for the
Mayflower II, the Adventure, and the Maryland Dove.
3. KEITH MUCKELROY. A major figure in maritime archaeology, Muckelroy was associated with many
early site explorations and did not limit his effort to Britain while he was at St. Andrews. He was one of
the first to recognize the importance of a theoretical background for the field and worked to define it.
His numerous publications brought the field to the public eye as he helped lead the way to a more pro-
fessional approach. He died in 1980.
4. 1. BAS KIST. As with historical archaeology, the conjunction of documentary and artifactual evidence
provides a unique way of evaluating the past. Kist presents the documentary and archaeological back-
ground behind the Hollandia and the Amsterdam as well as placing them in dual historical contexts
between the past when they were wrecked and the almost present when they were recovered. Kist was
curator of the historical department of the Rijksmuseum in Amsterdam. Holland.
5. PAUL FONTENOY. Fontenoy was a graduate student at East Carolina University when he wrote this
assessment of the differing approaches to Maritime Archaeology. The author of a work on the Hudson
River sloops, he is now a curator at the North Carolina Maritime Museum.
6. JOEL L. SHINER. This "paper" was presented as introductory remarks followed by a question and an-
swer session at the 9th Conference on Underwater Archaeology, San Antonio, Texas.
7. CHRISTOPHER F. AMER and CARL STEEN. The South Carolina Hobby Diver Program was an innova-
tive approach to integrating information about sites with sport divers. Amer, a graduate of Texas A&M
University, is Deputy State Archaeologist for Underwater Research. Steen is now employed by Dia-
chronic Research Foundation in Columbia, South Carolina. At the time this article was prepared, Amer
was employed by the South Carolina Institute of Archaeology and Anthropology, University of South
Carolina, Columbia, South Carolina, and Steen was living in Williamsburg, Virginia.

571
572 About the Contributors

8. PETER THROCKMORTON. Throckmorton first became involved with undersea exploration in the 1950s.
Since that time, he has worked all over the world and become an advocate for scientific wreck exami-
nation. He has written numerous books and served as an international advisor on maritime archaeology
to several governments. His last teaching position before his death was with Nova University, Florida.
9. WILBURN A. COCKRELL. "Sonny" Cockrell served as underwater archaeologist for the State of Florida.
Cockrell has numerous publications and served several terms on the Advisory Council for Underwater
Archaeology. At the time of this presentation, he was director of the Warm Mineral SpringsArchaeologi-
cal Research Project, Florida State University, Tallahassee.
10. R. DUNCAN MATHEWSON III. Mathewson has a wide experience in underwater archaeology from
both sides of the debate between professional archaeologist and salvors. He currently teaches diving
and underwater archaeology at Union Institute, Isla Morada, Florida.
11. WILLIAM S. DUDLEY. Dr. Dudley is a Navy man and a professional historian. He is currently Chief of
the Naval History Division, Department of the Navy, Washington, D.C.
12. A.H. TAHA. At the time this paper was presented, Taha was Acting Director of Antiquities, Museums
Department, National Museum, Kuala Lampur, Malaysia.
13. w.P. RONQUILLO. At the time this paper was written, Ronquillo was with the National Museum, Rizal
Park, Manila, The Philippines.
14. T.C. CHAMBERLIN. This paper was originally presented in 1890 when Chamberlin was president of
the University of Wisconsin. It was reprinted by Science circa 1932 and again in 1965 after John R.
Platt called attention to it in his 1964 article on Strong Inference (See Chapter 15). Platt noted "This
charming paper deserves to be reprinted in some more accessible journal today, where it could be
required reading for every graduate student - and for every professor." At the time Chamberlain pre-
sented this paper, he was President of the University of Wisconsin.
15. JOHN R. PLATT. When this paper was published in 1964, John R. Platt was Professor of Biophysics
and Physics at the University of Chicago.
16. M.A. SMITH. Miss Smith read this paper at the Conference of the Prehistoric Society, held at the Uni-
versity of London Institute of Archaeology iri April, 1955.
17. FRED T. PLOG. At the time the book from which this chapter was excerpted was written, Plog was a
professor at the State University of New York, Binghamton. He is now retired.
18. LAWRENCE E. BABITS. This report was written for students in an introductory archaeology course at
Armstrong State College, Savannah, Georgia, where Babits was teaching in 1973. Babits is now on the
faculty of East Carolina University, Greenville, North Carolina.
19. MARK L. RAAB and ALBERT C. GOODYEAR. Raab is in the Department of Anthropology, California
State University at Northridge, Northridge, California. Goodyear is in the Institute of Archaeology,
University of South Carolina, Columbia, South Carolina.
20. RICHARD J. ANUSKIEWICZ. Anuskiewicz was working on his dissertation when he prepared this paper.
At that time, he was an archaeologist with Savannah District, US Army Corps of Engineers. He is pres-
ently a marine archaeologist with the Minerals Management Service office in New Orleans. Louisiana.
21. DANIEL J. LENIHAN and LARRY MURPHY. Both Lenihan and Murphy are National Park Service archae-
ologists. They have been assigned to the Submerged Cultural Resources Unit in Sante Fe, New Mexico,
for many years.
22. REYNOLD J. RUPPE. Ruppe's archaeological interests were wide ranging. After a short stint at the
University ofIowa, he moved to Arizona State University in Tempe. He died in 1993.
23. JAMES F. MUCHE. Muche was a contract archaeologist and publisher of Fathom Eight when he pre-
pared this report. He is currently editor of Aviation. Now and Then, in El Monte, California.
24. RICHARD STEPHENSON. Stephenson is a geomorphologist, Department of Planning, East Carolina
University. He has worked on numerous archaeological sites to generate information about site forma-
tion processes.
25. KEITH MUCKELROY. (See Chapter 3.)
26. ROBERT L. GEARHART. Gearhart is a contract archaeologist who was working for Espey, Huston and
Associates, Austin, Texas, when he prepared the report ITom which this chapter is excerpted.
27. ERVAN G. GARRISON. Garrison is now on the faculty of the University of Georgia, Athens. At the
time this presentation was made, he was in the Department of Civil Engineering, Texas A&M Univer-
sity, College Station, Texas.
28. KIMBERLY WATSON. The flow chart was prepared while Watson was a graduate student at East Caro-
lina University. She is now in the doctoral program at Saint Andrews University.
About the Contributors 57.3

29. DONALD G. SHOMETTE and RALPH E. ESHELMAN. Shomette is an archaeologist, researcher, and
writer currently working on the War of 1812 Patuxent River Flotilla commanded by Joshua Barney.
Eshelman was with the Calvert Maritime Museum, Solomons, Maryland. They have been involved in
the long range study of the Chesapeake Bay and the Patuxent River.
30. U.S. NAVY. No author- US Government Compiled Training Manual.
31. 1. BARTO ARNOLD III. Arnold is Senior Research Associate with the Nautical Archaeology Program,
Texas A&M University and Director of Texas Operations, Institute of Nautical Archaeology, both in
College Station, Texas.
32. lACK HUDSON, KAY G. HUDSON, and HARRY W. RHODES III. The Hudsons are now freelance writ-
ers in Seabroook, Texas. Rhodes is now a practicing attorney in the Houston, Texas, area. At the time
this paper was presented to the Society for Historical Archaeology, the three authors worked for Cul-
tural Resource Services, Inc., Seabrook, Texas.
33. MARTIN KLEIN. Klein is one of the most innovative people in the field of remote sensing. A designer
with widespread practical field experience in remote sensing, he was in the forefront of advocating the
linkage of remote sensing with maritime archaeolgoy during the 70's, 80's and 90's.
34. LAURA lEAN PENVENNE and JOHN PENVENNE. At the time this paper was written, the Penvennes
were remote sensing and computer analyses advocates, seeking better linkages between various tech-
nologies. They were employed then at Triton Technology, Inc., Watsonville, California.
35. RODERICK M. FARB. Farb is a PAD! divemaster and an avocational archaeologist. He has numerous
popular publications relating to underwater archaeology and diving on sunken vessels, including Ship-
wrecks: Diving the Graveyard a/the Atlantic.
36. W. KENNETH STEWART. At the time this article was prepared, Stewart worked at Woods Hole Ocean-
ographic Institution and had many publications relating to computer modeling and imagery in the
underwater environment.
37. PETER J.A. WADDELL. Waddell was an archaeologist with the Marine Archaeology Section, Canadian
Parks Service, Ottawa, Ontario, at the time he prepared this study on electronic mapping.
38. PARKER B. POTTER lr. Potter is currently in law school. At the time he presented this paper he was
working in the New Hampshire State Historic Preservation Office.
39. GORDON P. WATTS Jr. Watts has extensive underwater archaeological experience and numerous publi-
cations. His wide-ranging expertise is a boon for his students at East Carolina University. Among his
many research endeavors are participation in locating the USS Monitor and the CSS Alabama.
40. MICHAEL B. ALFORD. Alford is a recognized authority on small watercraft in the southeastern United
States. He worked up his typology based on practical experience in North and South Carolina while he
was a curator at the North Carolina Maritime Museum, Beaufort, North Carolina.
41. STEPHEN J. GLUCKMAN. Gluckman currently lives in Tampa, Florida. At the time this paper was pre-
sented, he was a doctoral student at Catholic University. After earning his doctorate, Gluckman served
as North Carolina's State Archaeologist and as professor at the University of South Florida.
42. KEITH MUCKELROY. (See Chapter 3.)
43. JAMES G. BAKER, RICHARD J. ANUSKIEWICZ. and ERVAN G. GARRISON. Baker is a mathematician
with Texas A&M University. Anuskiewicz is a marine archaeologist with the Minerals Management
Service office in New Orleans. Garrison is a professor at the University of Georgia. This paper was pre-
pared while Garrison and Baker were at Texas A&M University and Anuskiewicz was an underwater
archaeologist with Savannah District, U.S. Army Corps of Engineers.
44. STEVE SELLERS. Sellers, a former police officer, is a member of the American Academy ofUnderwa-
ter Sciences and the Dive Safety Officer for East Carolina University.
45. HANS VAN TILBURG. Van Tilburg was a professional diver before entering East Carolina University.
He combined his diving experience with archaeological techniques to deal with the poor visibility often
found on underwater sites. These techniques were further refined while working as a crew chief on the
Maple Leaf Van Tilburg is now a doctoral candidate at the University of Hawaii at Manoa and teaches
the university's field course in nautical archaeology.
46. IAN OXLEY. Oxley is an archaeologist with the Archaeological Diving Unit, Scottish Institute of Mari-
time Studies, University of St. Andrews, Scotland.
47. LESTER A. ROSS. Ross worked with Parks Canada when he presented this paper. He has since served
the Society for Historical Archaeology as President and Board Member. He is presently a Historical Ar-
chaeology Consultant in Eugene, Oregon.
48. J. BARTO ARNOLD III. (See Chapter 31.)
 '.~~.' Index
~

o i' .
q;,O_' .': .
.j'~l

Abandoned Shipwreck Act, 88, 93, 95, 419 Amsterdam, 7-8,27,274

Abbot Ealdred of St. Albans, 29 full-scale replica of, 42
Abduction, 180 historical records of, 44
Aboriginal sites and material, 245, 253- 255, 332 Amsterdam Foundation, 44
Accuro, 81 Analogy, 176
Adaptation theory, 216 Anchors
Adelaar, 274, 287-288 as ballast, 463
Admiralty (U.K.), 24, 304 nationality of, 18
Admiralty law, precedence over state law, 102 from Parque Nacional Cahuita sites, Costa Rica,
Advisory Council on Underwater Archaeology, 94 461, 462, 463--464
Aiken, Edward, 377 preservation of, 525
Aimable, 363--364, 372 Spanish, 18
Airborne surveys: see Helicopters, use in magnetic Andrea Doria, 87
surveys Angara, Edgardo, 132
Aircraft, underwater wreck sites of, 109 Ankara, Turkey, National Museum, 76
Alabama, Mississippian sites, 217-218 Annapolis, Maryland, 423
Alabama, 73,105,107-108 Anomaly (ship), 366, 377
Alabama Historical Commission, 108 Anomie, 208
Alberti, Leon Battista, 29 Antigua, 76, 77, 79
Alcoholic beverages, as cargo, 277 Antikythera, Greece, 30
Alderice, Robert, 376 Anuskiewicz, Richard J., 494
Alexander the Great, 343 APT Bulletin, 534
All-terrain vehicles, 355, 376 Aqua-lung, 30, 31, 354; see also SCUBA
Alva Bradley, 399 Areal studies, 113; see also specific geographic areas
Amateur underwater archaeology movement, 55; see Archaeologists
also Divers, sports/amateur; Sports diving involvement in salvage activities, 85--86, 88-89,
associations 90,92
"American Rule," of admiralty law, 102 professionalism of, 89
Amphora, 11-l2, 13-14,31,32 Archaeology, see also Historical archaeology; Mari-
Amphora mounds, 278-279, 288 time archeology; Medieval maritime ar-
Amsterdam, Netherlands, Dutch East India Com- chaeology; New Archaeology
pany shipyards, 41--42 behavioral, 211-212, 213, 214--216

575
576 Index
Archaeology (cont.) Baltimore, 81
historical development of, 167-168 Baltimore Sun Magazine, 327
mechanical, 214 Barka, Nonnan, 249
as science, 24-25 Barkham, Selma, 537, 539
Archaeology from the Earth (Wheeler), 182 Barricas, 537, 541-548
Archaeology Underwater (Nautical Archaeology So- coopering technology of, 545-547
ciety),2 stowage patterns for, 537-538, 547-548
Archival studies, 319, 321; see also Historical records Basques, sixteenth-century whaling activities of,
Archives ofArchaeology, 182 537-549
Arizona, USS, 87, 521, 525 onshore sites of, 537, 539
Annada, 236, 423 San Juan shipwreck site of, 537, 539-548
Annstrong State College, 188 Bass, George, 1,2,32,34-35,49,62- 63, 76, 89
Arnold, 1. Barto III, 377 History of Seafaring, 62, 234
Artifacts, see also under specific shipwreck sites Today Show interview, 86
use in identification of shipwrecks, 18 Bastian, Tyler, 328
naval, 108 Batavia, 34, 42
preservation of Bathymetric data, display of, 494
historical background of, 29-30 Bathymetry
from salvaged vessels, 92-93 tight-beam, 493-494
salvaged, 89-91 use in zero-visibility contour mapping, 493-494,
preservation of, 92-93 497-501
public versus private ownership of, 89-90, 93 Battle of Britain, 423
stone, 460-461, 462, 466 Beach areas, magnetic search and survey techniques
trade in, 92 for, 375-377
valueless, 90 Beaucaire, Rene, 48
Association for Preservation Technology, 534 Behavioral archaeology, 211-212, 213, 214-216
Association of American Museums, 76, 557 Belcher, George, 106
Association of Canadian Underwater Councils, 58 Belle, 364, 372
Assyria, 345 "Bends" (decompression sickness), 349, 350-351
Athens, ancient, 23-24 Benoit, Ferdnand, 9
Atlantic Alliance for Maritime Heritage Conserva- Berkeley, George (bishop), 173-174
tion, 102 Bert, Paul, 351
Atocha, 78,97-103 Berth 52 vessel (09CH691), 187-202
disposal of artifacts from, 90 artifacts from, 196-197
economic evaluation of, 81, 82 heel and keelsons of, 188, 194-196
salvage of, 73 hull planking of, 193
legality of, 89, 93 interior supporting structures of, 193-194
professional archaeologists' disapproval of, 87 interpretation of, 197-200
stock sales financing of, 90 methodology of excavation of, 188-192
Austin's Children's Museum, 559-560, 561 recommendations for presentation of, 200-
Australia 202
iron object corrosion research in, 525 Bertrand, 88
maritime archaeology in, 34 Besancon, H.C., 124-125
Bethell's apparatus, 349
Bacon, Francis, 156, 160, 161, 162, 165 Binford, Lewis, I, 2, 210
Bacon, Roger, 345 Blackfriar's Roman ship, 27, 33
Bahamas, 78, 79 Blockade runner shipwreck sites, 432-433, 436
Baker, William, 62 salvage law regarding, 65
Ballard, Robert, 87 Blundell, Odo, 30
Ballast Boat-burials, 27, 33
anchors as, 463 Bodrum, Turkey, crusader castle ruin, 76, 77
bricks as, 280, 465 Bohr theory, 164
cannons as, 463 Bound House, Seabrook, New Hampshire, 424
Baltic region, 32 Boyne, 20
medieval maritime archaeological research in, 50 Bremen, Gennany, cog shipwreck sites, 20, 21
postmedieval maritime archaeological research in, 51 Brenner, Sidney, 157
Index 577
Bricks Caribbean, see also specific Caribbean islands and
as ballast, 280, 465 shipwreck sites
as cargo, 21, 465-466 tourism in, 76-78
Brimstone Hill castle, St. Kitts, 77 Carlson, 494
British Association, 30 Carolina Beach Inlet, North Carolina, Civil War ship-
British Columbia, Underwater Archaeological Soci- wreck sites, 433, 436
ety, 68-69 Carracks, 20
British Museum, 33 Castles, as tourist attractions, 76, 77
British Royal Engineers, 349 Cayman Islands, 76, 78
British Sub-Aqua Club, 55, 59 Central America, maritime archaeology development
British Virgin Islands, salvage activities in, 79 in, 33--34
Bronze Age shipwrecks, 32 Central America, 87, 88, 91
Brooklyn Bridge, 350 Ceramics
Brown's Ferry, South Carolina, 21 Ballarmine stoneware, 124
Bruijn, J.R., 125 from Parque Nacional Cahuita sites, 462, 466-467
Buoyant hull fracture shipwreck sites, 228, 229, 298, from Philippine shipwrecks, 129, 130
299--300 from the Spanish Armada, 236
Buoyant hull shipwreck sites, 228-229, 298-299 Chamberlin, T.C., 2,161-162
Buoyant hull structure shipwreck sites, 228, 229, Charles P. Minch, 406-409
298,299--300 Chattahoochee, CSS, 261-266
Buoys, marine sensor use on, 375, 377 Chattahoochee River, geomorphology of, 263--264
Burials: see Ship-burials Chemistry
Bursledon ship, 27 mathematical formulation use in, 164
Butuan boat, 131 organic, 163
Byzantinos, George, 30 Chiefdoms/chieftainships, 169, 217-218
Children's museum exhibits, 559--561
Caicos, salvage activities in, 79, 80 China
Caissons, 350, 351 maritime history of, 121
Caledonian Canal Company, 30 Ming dynasty, 121, 129
California, salvage activities in nautical archaeology programs in, 121
disposal of salvaged artifacts, 91 China Sea, salvage activities in, 79
public ownership of salvaged artifacts, 90 Circe, 107
state supervision of, 89, 93 Civil War shipwrecks
Caligula, 14 Alabama, 73,105, 107-108
Calvert Marine Museum, 327-328, 329 blockade runners, 65, 432-433, 436
Campos, Jorge, 454 Central America, 87, 88, 91
Canada, shipwreck and portage sites, 34 Chattahoochee, 261-266
Canadian Parks Service, Marine Archaeology Unit, Cumberland, 106--107
SHARPS (electronic mapping) system use Florida, 106-107
by,395,403-410 Georgia, 491, 493-504
Canal boat: see Berth 52 vessel (09CH691) historical records of, 433, 436-437
Cannon Kearsarge, 107
as ballast, 463 Maple Leaf, 87, 89, 91, 513--517
preservation of, 525 Naval Historical Center's cataloging project for,
Cape Artemision, 30--31 108
Cape Dramont, 278 in North Carolina, 432-439
Cape Fear, 433 research and significance criteria for, 431-439
Cape Gelidonya, 32, 49 salvage-related destruction of, 437
economic evaluation of, 81 Clark, Grahame, 23
Capitalism, archaeology of, 423-425 Classical archaeology, 30
Caravels, 20 Classical shipwreck sites
Cargo Blackfriar's boat, 27, 33
alcoholic beverages as, 277 early excavations of, 30, 31
brick as, 465-466 Lake Nemi, Italy, 14,29,31
of Civil War ships, 432 Cliffs, submerged, 270, 278
coal as, 18-19 Clinker planking technique, 19,20
578 Index

Coal, as cargo, 18-19 "Cultural Resource Evaluation ofthe Northern Gulf of

Coastal Environments, Inc., 303-304 Mexico Continental Shelf' (Gagliano), 236
Cofferdams, 250, 525 Cultural resource management, 432
Cogs, 20, 21 Hobby Diver Program for, 65-69
Collingwood, R.G., 169 research design and, 203, 236
Colonial American archaeological sites, see also site environmental assessment in, 525
Revolutionary War shipwreck sites Cultural resource management reports, 236, 237, 411
effect of sea-bed changes on, 247-252 Cumberland. USS, 105, 106-107, 109
Colonialism, Dutch East India Company and, 40
Colonna, Cardinal, 29 Dalian Maritime College, archaeology program of,
Colorado River, Texas, beach survey of, 376-377 121
Colossus, 274 Dalton, John, 157
Committee on Nautical Archaeology (U.K.), 59 Dartmouth, 274
Computer graphics, use in contour mapping, 494-504 effect of seabed movement on, 284-285
sonar graphs, 497, 498, 503 wrecking process of, 279-280, 281
three-dimensional grid elements, 494, 501-504 Data
Computer modeling, 50 acquisition of, 182
Computer video image digitization, of the USS Moni- analysis of, 182-183
tor, 397-402 raw, 181
Conception, 77 testing of, 183
Conception I, 82 Data base, shipwreck sites as, 92, 94-95
Conception II, 81 Deane, Charles, 29, 348-349, 352
Concepts, archaeological, 224 Deane,John,29,348-349,352
Confederate Naval Museum, 261 De Braak. HMS, 109
Confederate States of America, vessels of Deduction, 176, 184,208,210
Chattahoochee, 261-266 comparison with induction, 156
Florida, 106-107 Defence, 21, 34
sovereignty over, 107 Delbruck, Max, 159
Conference on Underwater Archaeology, 93-94 De Liefde, 274
Conservation: see Preservation Demarchi, Franchesco, 29
Contour mapping, 405-410 Denmark
of magnetometer survey data, 293-296, 360 medieval maritime archaeological research in, 50
systems operations and methodology of. 403-404 Viking-age shipwrecks in, 19-20
in zero visibility, 493-505 Derry Sub-Aqua Club, 60
sonargraphs,497,498,502 Dictator; HMS, 327
three-dimensional grid elements, 494,501-504 Discovery of the Sea. The (Parry), 121
Control of Certain Salvage Operations law, 65 Divers
Cooper River, South Carolina, "Two Cannon manuals for
Wreck," 68 first, 349
Corpus Christi Museum of Science and History, U.S. Navy, History of Diving chapter, 343-354
I 554.fTota exhibit, 553-558, 559 sports/amateur
Corrosion, underwater, 275-276, 525 Hobby Diver Program for, 65-69
Costa Rica, see also Parque Nacional Cahuita, Costa involvement in archaeological preservation, 102
Rica, pre-excavation investigations professional archaeolgists' cooperation with, 92
destruction of shipwreck sites in, 453-454 training in archaeological research, 58-63
Court records, associated with Civil War shipwrecks, of U.S. Navy, 108
433,436 zero-visibility training of, 507-511
Cousteau, Jacques-Yves, 31, 34, 48, 49, 354 Diving
Cowan, Rex, 42-43 historical development, 29, 343-354
Cowan, Zelide, 42-43 aqua-lung, 30, 31, 354
Crannog sites, 30 armored diving suit, 352
Creel, Tilford C., 494 caisson, 350
Crick, Francis, 157 deep-sea diving, 347-350
Critical theory, 417, 422-425 early developments (1500--1800),344-350
Crow, William, 366 first diver's manual, 349
C transforms, 234-235 physiological aspects of, 350--352
Index 579
Diving (cant.) Ethics, of underwater archaeology, 73
historical development (cant.) Ethnography, 51
pre-1500, 343-344 Ethnology, 24, 26
SCUBA, 352-354 Europe, amateur underwater archaeology in, 57-
military use of, 343, 344, 354 63
Diving bell, 29, 345-347, 350 Eustasy, 248
Diving suits, 347, 348-349, 352 Evidence, rules of, 175
DNA, helical structure of, 157 Excavations
Dominican Republic, 76, 79 data acquisition function of, 182
Dramont D site, 32 objectives of, 319, 320
Drydocks, 326 terrestrial, 27-28
Dry Tortugas, shipwreck sites, 306 whole versus partial, 237-238
Dumas, Frederic, 31, 32, 270, 278-279 Exhibits, of shipwrecks
Duol"O,466 by children's museums, 559-561
Dutch East India Company, 7-8,49, 51 economic aspects of, 76--77, 80, 81
organization of, 39-40 importance of, 533-534
shipyards of, 41-42 Expentiana, 10
Dutch East Indiamen shipwreck sites, 40 Experimental approach/methods
Australian sites, 34 in maritime archaeology, 50
De Lie/de, 274 in medieval maritime archaeology, 51
Griffin, 130 Explanation
historical records related to, 41-44 Hempel-Oppenheim model of, 176--177
Risdam, 123, 124--125 nature of, 176--181
Royal Captain, 129-130 operational component of, 178-184
Dutch East Indian Wreck Research Foundation, Extraction filters, 234--235, 259
124--125
Faraday, Michael, 162
Ealdred of St. Albans (abbot), 29 Farb Monitor Expedition, 398
Earthenware, from Parque Nacional Cahuita sites, Fashion, 262
462,466-467 Faunal analysis, 211-212
East Carolina University Fermi, Enrico, 162, 165
extreme environmental conditions field work train- Fernandez, Eduardo, 458-460, 462
ing program, 491 Ferriby boats, 27
maritime archaeology thesis topics, 49 Finds handling, 524
Maritime History and Nautical Archaeology pro- Fisher, Mel, 75, 78, 79, 82, 87, 90, 97, 100, 101,
gram, diver training program, 508-511 103
Program in Maritime History and Underwater Re- 501(C)(3) projects, 77, 81
search,261 Flatbottom boats, 442-453, 445
zero-visibility diver training program, 513-517 Fleuss, H.A., 353, 354
Economics Fior de la Mar, 123, 125
of salvage, 78-81 Florida
of shipwreck exhibits, 76--78, 80, 81, 557 lack of shipwreck excavation sponsorship by,
Egyptian tombs, salvaged material from, 79-80 101
Elgin, Earl of, 79 1715 plate fleet site, 78, 79
Empiricism, 207 salvage activities in, 78, 79
Engelhardt, Conrad, 30 disposal of salvaged artifacts, 89-91
Engine sites, 321 salvaged artifacts ownership, 89,90
English Channel, Alabama shipwreck site, 73,105, state archaeologists' involvement in, 90
107-108 state funding of, 87
English East India Company, 40 state supervision of, 88-89, 93
English Harbor castle, Antigua, 77 Florida. CSS, 106--107
Environmental assessments, of shipwreck sites, 525, Florida Keys, shipwreck sites, 306, 310
526--527 Florida Museum of Natural History, 90
Environmental context, of shipwreck sites, 268, Florida State Museum, Alocha artifacts, 90
270--274 Florida State Underwater Archaeologist, 85, 94
implication for preservation, 523-529 Fiola exhibits, 553-561
580 Index
Flow charts, 317 Guide to Archaeological Field Methods. A (Heizer
Flowerdew Plantation, 249 and Graham), 182
Fort Augustus, Scotland, 30 Gulf of Mexico
Fort Caroline, Florida, 101 British shipwrecks in, 304, 310
For Theory BUilding in Archaeology (Binford), coastline stability of, 248
211-212 factors affecting shipwreck patterns in, 259,
Fort Moultrie, South Carolina, 250 303-316
Fort SI. Louis, Texas, 365 areal factors analysis of, 310, 314, 315
Fort SI. Marks, Florida, 250 chronological-trends analysis of, 310-311,
Fos-sur-Mer, underwater excavations, 7, 9-15 314-315
artifacts of, 11-15 hurricane paths, 309-310, 315, 316
Beaucaire's excavations at, 48 ocean currents and winds, 307-309
dwellings of, II ports, 306, 315
kitchen dumping grounds of, 11-12, 13 sailing route, 305-306
Fowles, Austin, 100 shoals, reefs, sand bars, and barrier islands,
French Huguenots, 101 306-307,315
French Navy, Diving School of, 31 French shipwrecks in, 304, 310, 311
Spanish shipwrecks in, 304, 309 310, 311
Gagnan, Emile, 31, 354 Gumbel, Bryant, 86, 95
Galleons, 129 Guns, see also Cannon
Galleys, Athenian, 23-24 date and place of manufacture of, 18
Gamma, 377 Guzman, Hector, 454, 458
Gardening, in Trobriand society, 170-171
Garrison, Ervan G., 494 Haiti
Garwin, Richard, 159 Sans Souci castle, 77
Geldermarsten, 80, 82 Santa Maria site, 88
Gell-Mann, Murray, 159 Haldane, J.S., 351-352
General Services Administration, 107, 108 Halley, Edmund, 346-347
General theory, differentiated from middle-range Hamilton, 106
theory, 212 Harbor Branch Foundation, 397-398
Geographic information systems (GIS), 527 Harrington, J.C., I
Geomorphology Hatcher, Michael, 124-125
of the Chattahoochee River, 259, 263--264 Hatteras. Inc. vs. the USS Hatteras, 106
coastal, 270 Hatteras. USS, 106
marine, 288-289 Heflin, Howell, 107
of sea-level changes, 247-248 Helicopters, use in magnetometer surveys, 355, 365,
stream-formation, 259, 263-264 366-368,372-373
Georgia, 491 Henrietta Bach, 330
contour mapping of, 494-504 Henry VIII, 20
historical background of, 494-495 Heraclea, 279
Gerard Hilferty and Associates, 557 Heme Bay, Kent, 30
Germany Herodotus, 343
cog vessel shipwreck sites, 20, 21 Hierarchical cultures, 217-218
medieval maritime archaeology research in, Historical archaeology, I
50 as archaeology of capitalism, 423-425
Girona, 274 Historical Archaeology: A Guide to Substantive and
Gould, Richard, 422, 423 Theoretical Contributions (Schuyler), 2, 565
Grand Congloue, Marseilles, 31, 48 Historical records
Grand Ribaud A site, 32 of Civil War shipwrecks, 433, 436-437
Graveney boat, 27, 33 computer-based, 304
Graveyard o/the Gu(((fiIm), 553 of Dutch East Indiamen shipwrecks, 41-44
Green, Jeremy, 34, 125 of Gulf of Mexico shipwrecks, 304
Grenier, Robert, 537, 539 international sources of, 304
Griffin, 130 use in shipwreck significance and research poten-
Griffin, James B., I tial assessment, 433, 436-347
Grounded theory, 208 Historic context, 422
Index
Historic preservation, of shipwreck sites, 417,
419--426
historic context concept of, 422
National Register of Historic Places and, 190,
200,420--422
relationship to critical theory, 422- 425
HistOlY of American Archaeology. A (Willey and
Sabloff), 205
History of Seafaring. The (Bass), 62, 234
Hobby Diver Program, 65-69
Hollandia, 42--44, 283
Hollywood Adventure Films, 107
Homan, Brian, 129
Horan, Paul, 75
Huguenots, 101
Hulks, 20
Human Behavior: An Inventory of Scientific Finding
(Berelson and Steiner), 177, 180
Humanistic approach, in archaeology, 168
Hume, Ivor Noel, I
Hunter-gatherers, settlement behavior of, 216-217
Hurricanes, 309-310, 315, 316
Hutchinson Island, Florida, Valentine site, 88
Hypotheses, 178--179
alternative, 156-157, 160
Berth 52 vessel (09CH691), 190-192, 197-200
disproof of, 160-161, 165
formulation of, 179-180
implicit versus explicit, 179
multiple working, 2, 141, 145--154, 161-162
differentiated from ruling theory, 148
disadvantages of, 148, 150, 152
use in field investigations, 143
method of, 148--150
practical applications of, 150-152
operational ization of, 180-182
lie de Planier, 278
Induction, 184, 208, 210, 516
aids to, 165--166
alternative hypotheses use in, 156-157, 160
definition of, 156
use in high-energy physics, 159-160, 161-162,
165
logical tree process of, 156-158, 159
use in molecular biology, 155, 157- 160, 161-162,
165,166
multiple hypotheses use with, 161-162
problem-orientation of, 164
process of, 156
relationship to scientific effectiveness, 163--164
systematic application of, 162-163
Inference
inductive: see Induction
limitations of, 167-174
Institute of Nautical Archaeology, 89
581
Institute of Texas Culture, 559, 560, 561
International Journal of Nautical Archaeology and
Undelwater Exploration, 203, 234,
235--236,565
International waters. salvage in. 88
Interpretation, 533; see also Museums
evidence use in. 143
as maritime archaeology focus. 5
of prehistoric cultures. 168--174
of primitive cultures. 169-171
Interpretative theory, 211-212
Inventory. baseline, 358
Iran, 218
Ironclad vessels: see Georgia; Monitor; Tecumseh;
Virginia
Iron objects. corrosion of, 275--276, 525
Isis modular data acquisition system, for side-scan
sonar, 387-393
Isostasy. 248
Jacob, Francois, 158
Jakarta, Dutch East India Company shipyards, 41--42
Jamaica, 76
James Bay, 81
Jamestown, Virginia, 248. 249, 250
Japan, 34
John F Leavitt, 17-18
John Pennekamp State Park, 525
Jones, Walter, 107
Journal of Molecular Biology, 158
Junks, 121, 126
Katsev, Michael, 32, 76
Kearsarge. USS, 105, 107
Keels, morphology of, 443
Kennemerland, 274, 469
artifacts from, 276-277
historical records of, 269, 276
salvage of, 276-277
effect of seabed movement on, 285--287
statistical site analysis of
clustering procedures, 483--485. 486
dendrograms, 483, 484, 485, 486, 487
Euclidean distance measures, 483--487
meter square grid analysis, 477--488
open versus closed sites, 474, 476--477
product-moment measures, 483, 485, 486-
487
wrecking process of, 280-283
King Philip, magnetometer survey of, 292-301
Klein, Martin, 251
Kuhn, Thomas, I
Kyrenia Castle, 32, 76, 77
Kyrenia shipwreck site, 32, 48
economic evaluation of, 81
technical importance of, 19, 21
582 Index

Labrador, sixteenth-century Basque whaler-related Maehault, 34

sites, 537-549 Magnetometer
onshore sites, 537, 539 European versus American use of, 57
San Juan shipwreck site, 537, 539- 548 shallow-water use of, 251
Lake Nemi, Italy, Roman shipwreck sites, 14,29, tow system for, 375
31 use on all-terrain vehicles, 376
Lambert, Alexander, 353 Magnetometer surveys
L 'Aimable, 36~364, 372 anomaly source depth estimation in, 300-30 I
La Roche Fouras site, 32 as cultural resource management survey compo-
Larson, Louis, 190 nent, 357-358
La Salle, Rene 63, 365, 377 helicopter use in, 355, 365, 366--368, 372-373
Latimer, Truett, 365 lane spacing in, 355, 358-360
La Tradeliere site, 32 limitations of, 357-361
Laws of Matagorda Bay, Texas, 311, 365-374
anthropological, 177 test excavations, 370-371,372
hypotheses versus, 183 of Ocean Beach, California, 291-301
for salvage, 34, 88-89,91,102,126 anomaly pattern recognition, 298
Abandoned Shipwreck Act, 88, 93, 95, 419 anomaly source depth estimation, 300-30 I
admiralty law, 102 buoyant hull anomaly pattern, 298-299
Control of Certain Salvage Operations law, 65 buoyant hull fracture pattern, 298, 299-300
in Malaysia, 126 buoyant hull structure, 298, 299-300
in South Carolina, 65--{j6 data processing and contour mapping, 29~
in Texas, 61~2, 66 297
of U.S. government vessels, 105-109 interpretative methods, 297
L'Be/le,364,372,377 methodology, 291-293
Lederberg, Joshua, 158 visible anomaly sources, 297-298
Lee, T.D., 159 target mass prediction in, 359, 360
Legare Anchorage Shipwreck Project, 525 use in zero visibility, 493-494, 496, 497
Lehman, John, 106 Mahdia, Tunisia, 30
LePrieur, Commander, 354 Maine, Department of Economic and Community
Leshikar, Margaret, 377 Development, 78
Lethbridge, John, 347, 352 Maine State Museum, Augusta, Maine, 77-78
Levinthal, Cy,. 158 Malaysia
Liberty Bell VII (space capsule), 88 nautical archaeology programs in, 121-122
Licenses, for salvage underwater archaeology of, 12~126
in Malaysia, 125 legal aspects of, 124, 126
in the Philippines, 132 personnel and training for, 124
in South Carolina, 66 Malinowski, Bronislaw, 170-171
Linzer, Claudia, 100 Mallory, Stephen R., 262
"Living Systems: Cross-Level Hypotheses" (Miller), Manillas, 459, 460-461, 462, 465, 466
177 Maple LeaJ
L1oyds,304 salvage of
Loeational Analysis in Human Geography (Haggett), legality of, 91
181 professional archaeologists' involvement in,
Loch Ness, Scotland, 30 89
Lockwoods Folly Inlet, 433, 436 state funding of, 87
Log boats, typology of, 443, 444 zero-visibility diving on, 51 ~517
London Board of Trade, 304 Mapping, three-dimensional: see Contour mapping
Londontown, Maryland, 250 Margarita, 97, 101, 103
Looting: see Salvage Marinas, salvage activities in, 79
Louisiana, salvage in, state supervision of, 93 Marine Archaeology Unlimited Incorporated, 129
Lovely, 283 Mariners Museum, Newport News, Virginia, 304
Lower-level theory, 213-214, 215 Marin-Zuniga, Sigifredo, 458
Lunulae, 168-169, 171 Maritime archaeology, I
Lyell, Charles, 29 academic status of, 49-50
Lyon, Eugene, 103 definitions of, 2~29, 224
Index 583

Maritime archaeology (cant.) Method and Theol)' in Historical Archaeology

development of, 1,29-35,47-50,48 (South), 1-2
19th century, 29-30 Methodology, of maritime archaeology, 143; see
early 20th century, 30 also Magnetometer surveys; Sonar
1920s-1940s, 30-31 European compared with American, 57-63
1950s-1960s, 31-32 logic of, 175-185
1970s,32-34 Miami Herald, 82
differentiated from other archaeology subdiscipli- Middelburg, Netherlands, Dutch East India Com-
nes, 24, 26-27, 28--29 pany shipyards of, 41
multidisciplinary approach of, 5, 7, 8,49, 50 Middle Ages, see also Medieval maritime archaeol-
overview of, 23--29 ogy
standards for, I development of diving during, 344-345
as subdiscipline of archaeology, 35 Middle-range theory, 203, 205-221
Maritime Archaeology (Green), 49 archaeological theory as precursor to, 210-211
Maritime Archaeology (Muckelroy), 2, 7,49, correlation with magnetometer data, 223--231
234-235 cultural systems applications of, 216- 218
Maritime history, relationship to maritime archeol- definition of, 224
ogy, 26 differentiated from general theory, 212
Marquesas, shipwreck sites, 306 potential problems of, 214-216
Marsala, Sicily, Punic War shipwreck site, 19 remote sensing and, 223-231
Marsden, Peter, 42-43, 44 in sociology, 205, 206-209
Maryland Derelict Boat Removal Project, 327-328, Ming dynasty, 121, 129
333 Mission Santa Catalina de Guale, St. Catherines Is-
Maryland Historical Trust, 329 land, Georgia, 224, 225
Mary Rose, 20, 274 Mississippian sites, Alabama, 217-218
effect of seabed movement on, 284, 289 Models, 207
wrecking process of, 279 Modern Greece,432-433
Massachusetts Molecular biology, induction use in, 155, 157-160,
relations with New Hampshire, 424-425 161-162, 165, 166
salvage in activities in, 89, 93 Mombasa, Kenya, 34
Matagorda Bay, Texas Monitor. USS, 108, 395
airborne magnetometer survey of, 311, 365-373 computer video image digitization of, 397-402
test excavations, 370-371, 372 federally-sponsored survey of, 101
shipweck sites, 311 as first National Marine Sanctuary, 397
MAVL,81 National Oceanic and Atmospheric Administration
Maxwell, James, 163 management of, 397, 398
McKinney, Charles M. III, 102 remote sensing of, 397-398
Mechanical archaeology, 214 Monod, Jacques, 158
Medieval maritime archaeology, 49,50-51 Montelius, 0., 29
boat-burials, 27, 33 Montgomery. USS, 262
cog, 20 Moore, George, 365
hulk,20 Moundville, Alabama, 217-218
multidisciplinary approach in, 51 Muckelroy, Keith, Maritime Archaeology, 2, 7, 49,
Mediterranean shipwreck sites 234-235
Bronze Age sites, 32 Musee du Vieil-Istres, 10
environmental context of, 270, 278--279 Museum of Florida History, 90
Israeli research regarding, 34 Museum of the University of Pennsylvania, 32
medieval maritime archaeological research in, 50 Museums
salvage of, 30, 343--344 British Museum, 33
Megalithic tombs, 169 Calvert Marine Museum, 327-328, 329
Merced, 88, 89, 90 children's, 559-561
Merton, R.K., 207, 208--209,214,217 Corpus Christi Museum of Science and History,
Meselson, Matthew, 157 553--558
Mesopotamia, 218 economic importance of, 76-78, 80, 81, 557
Metal(s), electrolyte protection of, 276 Maine State Museum, 77-78
Metal detectors, 57 Mariners Museum, 304
584
Museums (cant.)
Museum of Florida History, 90
Museum of the University of Pennsylvania, 32
National Maritime (U.K.), 33, 57
National Museum (Turkey), 76
National Museum of Mexico, 77
National Museum of the Philippines, 129-132
North Carolina Maritime Museum, 441
Penn, 82
Quazhou Museum of Overseas Communications, 121
Rijksmuseum, Amsterdam, 43
as tourist attractions, 557
Mussolini, Benito, 31
Mustang Island, Texas, 299
Myerscough, John, 76
Nanking, China, Imperial shipyards, 121
National Academy of Police Diving, 508
National Archives of Mexico, 304
National Clearinghouse for Archaeological Site Sta-
bilization, 526
National Geographic, 100
National Library of France, 304
National Marine Sanctuary, 397
National Maritime Museum (U.K.), 33, 57
National Museum (Turkey), 76
National Museum of Mexico, 77
National Oceanic and Atmospheric Administration
(NOAA)
Marine Sanctuary Program, 108
protection of shipwreck sites by, 88- 89, 93
USS Monitor management by, 397, 398
National Park Service
fortification sites of, I
historic context concept of, 422
protection of shipwreck sites by, 88--89, 93, 419
Shipwreck Anthropology Advanced Seminar, 94
National Register of Historic Places, 190,200,420--422
National Register of Historic Places Bulletin 20, 417
Natural philosophy, 172-174
Nautical Archaeological Associates, Inc., 327, 328,
329
Nautical archaeology, differentiated from maritime
archeology, 24, 28
Nautical Archaeology Society, 2
Naval Facilities Engineering Command, 108
Naval Historical Center, 109
Volunteer Training Unit, 108
Navigational charts, use by Confederate blockade
runners, 436
NCR,81
Ne'eman, Yuval, 159
Nelson, Horatio, 20
Netherlands, see also Dutch East India Company;
Dutch East Indiamen shipwreck sites
polder excavations in, 32-33
Index
Nevis, 76
New archaeology, I
middle-range theory and, 205, 206, 210-211
New Hampshire
critical underwater archaeology program, 423-425
relations with Massachusetts, 424-425
New Hampshire State Historic Preservation Office,
419
New Organon, The (Bacon), 160
Newton, Isaac, 163
Nitrogen narcosis, 352
North America, see also specific shipwreck sites and
states
maritime archaeology development in, 33-34
North Carolina Division of Archives and History, in-
house underwater archaeological research
by, 433
North Carolina Maritime Museum, 441
North Carolina shipwreck sites
Civil War, significance criteria and research poten-
tial assessment of, 432--439
indigenous small-craft typology (1790- 1920) of,
429,441--445
North Sea shipwrecks, 33
N transforms, 234-235
Nuclear sites, 321
Nydam boat, 27, 30
Objectives, of underwater archaeology, 319, 320
Ocean Beach, California, magnetometer surveys of,
291-301
anomaly pattern recognition, 298
anomaly source depth estimation, 300-301
buoyant hull anomaly pattern, 298-299
bouyant hull fracture shipwreck patterns, 298,
299-300
buoyant hull structure, 298, 299-300
data processing and contour mapping, 293-297
interpretative methods, 297
methodology, 291-293
visible anomaly sources, 297-298
Octopus, amphora mound building by, 288
Ontario, Save Our Shipwrecks, 68-69
Opticks (Newton), 163
Orange Bay, St. Eustatius, 524
Organizational variability, 216-217
P.K. Younge Library of Florida History, 304
Pacific Rim, marine archaeology of, 121-136
bibliography of, 135--136
Malaysia, 123-126
Philippines, 127-133
Padre Islands, Texas, 1554 flotas exhibits of,
553-561
Palos Verdes Penn insula, aboriginal site, 253-255
Panol, Francisco, 131, 132
Index 585
Paradigm Plat Taylor, Joan du, 32
definition of, I Plymouth State College, New Hampshire Studies
for scientific research, 141 program, 419
of sea-level changes, 248 Poidebard, Andre, 31
Parque Nacional Cahuita, Costa Rica, pre-excava- Pollard, Hugh, 30
tion investigations, 454--467 Popper, Karel, 160-161
Brick Area, 459, 460-461, 462-463, 464, 465-467 Portage sites, 34
Cannon Area, 460, 461-462,463-464,466 Pottery, see also Ceramics
interpretation, 462-467 Gallo-Roman, II
location and physical environment, 454-456 Roman, 30
research design, 456-458 slaves' manufacture and use of, 68
Parsons, T., 206-207, 209 Pre-excavation investigations, 451-467
Parthenon, 79 of Parque Nacional Cahuita, Costa Rica, 454-467
Pasley, William, 349, 351 Brick Area, 459, 460-461, 462-463, 464,
Pasteur, Louis, 163 465-467
Patuxent Indian tribes, 332, 333 Cannon Area, 460,461-462,463-464,466
Patuxent Submerged Cultural Resource Survey, interpretation, 462-467
323-355 location and physical environment, 454-456
historical background, 323-236 research design, 456-458
phase I, 329-331 Prehistoric cultures, 168
phase II, 331 interpretation of, 168--174
preliminary survey observations of, 332-333 South Pacific sailing traditions, 121
purpose of, 326-327 Prehistoric sites
Pearl Harbor, USS Arizona site, 87 firepits of, I
Pelagos site, economic evaluation of, 81 magnetometer surveys of, 358
Pell, Howard, 42-43 Preindustrial societies, "ship as machine" in, 23-24
Penlee site, 274 Preservation, 524
Penn Museum, 82 of anchors, 525
Penobscot River, Defence site, 21, 34 of artifacts, 521
Peter Cooper, 330 historical background of, 29-30
Philippines contlict with salvage, 75
nautical archaeology programs in, 121-122 environmental context of, 523-529
salvage activities in, 79 as public issue, 75-76
underwater archaeology in, 127-133 three-part heuristic model for, 92, 94-95
chronology, 129-130 of underwater archaeological sites, 93, 525-529
legal aspects, 132-133 Primitive cultures, interpretation of, 169-171
Philo Scoville, 409-410 Principles of Geology (Lyell), 29
Philosophy of Social Science (Rudner), 175 Problem-solving, 178-179
Phipps, William, 346 Professional archaeology societies, attitudes towards
Phoenix, 81, 82 salvage activities, 93-94
Photography Project planning, 317
aerial, 31 Protection, of shipwreck sites, see also Laws, for sal-
underwater, 3 I vage; Preservation
remote-controlled robot use in, 106 archaeologists' concerns about, 91-94
Physics Provenience, of shipwrecks, 88
high-energy, use of induction in, 159-160, Public attitudes, towards shipwreck excavations, 93
161-162, 165 Public Law 828, 79
mathematical formulation use in, 164 Punic War shipwreck, 19
Pitt-Rivers, Augustus, 29, 167 Putziger, Barney, 190
Planking
caravel-type, 20 Quadrants, Davis, 21
of pre-I 605 vessels, 19 Quadrat analysis, of discontinuous sea- bed distribu-
Plantations, 68 tion sites, 477-488
Flowerdew, 249 Quantitative analysis, 2
on the Patuxent River, 324 Quanzhou Museum of Overseas Communications,
Platt, John R., 2 121
586 Index

Quarry Wreck, electronic mapping of, 405 Roman villa, submerged, 7

Quinqueremes, 19 Round-bottom boats, 442--443
Rouquayrol, Benoist, 353
"Rapture of the deep," 352 Royal Captain, 129--130
Reburial, of archaeological material, 525-526 Royal George, HMS, 20, 278, 349
Recompression chambers, 351 Royal Malaysian Navy, 124, 125
Red River, Louisiana, survey of, 376 Royal Navy, 351
Red Sea, shipwreck sites, 34 submarine escape equipment of, 353
Reinach, Salomon, 30 Rules of evidence, 175
Remote sensing, see also Magnetometer surveys; Sonar Rushton, W.A.H., 159
use in cultural resources management, 236 Rydberg constant, 164
Replicas, full-scale, 50 Rye, Sussex, 30
of Dutch East Indianmen shipwrecks, 42
Reporter, 292-301 Sabloff, J.A., 213-214, 215
Reproductions, 50 Sail sites, 321
Research design, 17S-184 St. Catherines Island, Georgia, middle-range theory
anthropological issues of, 236-237, 238 analysis of, 223-231
conceptual approaches in, 233-239 archaeological indices, 203, 227, 230
Muckelroy's contributions to, 234-235 instrumental indices, 203, 227-228, 230
recommendations for, 238 maritime model of, 223, 225, 226, 229, 230
of cultural resource management reports, 203, 236 site formation processes of, 225, 227, 230
data acquisition and analysis in, 182-183 Saint-Gervais, drowned Roman villa site, 7
deductive, 233, 238 St. Juan de Ulloa castle, Veracruz, 77
evaluation of research component of, 183-184 St. Kitts, 76, 77
explicitness of, 203, 233-234, 238 St. Lucia, 76
hypotheses formulation in, 17S-180 St. Martin, 76
hypotheses operationalization in, 180-182 Salvage
hypotheses testing in, 183 of artifacts, 89--91
of International Journal of Nautical Archaeology "boom" in, 7s-80
and Underwater Exploration, 235-236 in the Caribbean, 76, 79
problem formulation in, 179 conflict with historic preservation, 75
problem orientation of, 25 criteria for acceptibility of
Research potential, assessment of, 431--439 age of vessel, 87-S8
Research publications, 234, 235-236 disposition of artifacts, 89--91
Research Strategies in Historical Archaeology legality, 8s-89
(South), 1-2 practicality, 91
Reservoirs, 265 professionalism of salvors, 89
Revolutionary War shipwreck sites, 68 profit motive, 90
Defence, 21, 34 provenience of vessel, 88
on the Patuxent River, 324 transfer of legitimacy, 90
Rex 11,81 type of vessel, 88
Ribault, Jean, 101 destructive nature of, 80, 86-S7, 437
Rijsmuseum, Amsterdam, 43 economic aspects of, 7s-82
Risdam, 123, 124-125 historical background of, 343-344
Risk theory, 216-217 in Florida, 78, 79
Rivard, Paul, 77 historical background of, 343-344, 347-348
Riverboats, Dutch, 33 impact on shipwreck sites, 275
Robots, use in underwater photography, 106 as industry, 79
Rodgers, Brad, 509 investors in, 78, 79-S0, 82
Roentgen, Wilhelm, 162, 163 justification of, 97-103
Roman Empire, shipbuilding during, 23 legal aspects of, 61-62, 65-66, 8s-89, 91, 95,
Romano-British shipbuilding, 33 102,105-109,126,419
Roman shipwreck sites, 19 in Malaysia, 123, 124-125, 126
Blackfriar's boat, 27, 33 in the Mediterranean, 30
Lake Nemi, 14,29,31 as motive for excavations, 237-238
the Mediterranean, 30 in the Philippines, 127, 129
Index 587
Salvage (cont.) Seahawk, Inc., 85, 88, 89, 90
professional archaeologists' involvement in, Sea-level changes, effect on Colonial America ar-
85-86,88-89,90,97 cheological sites, 247-252
professional archaeology societies' attitudes to- Searches, differentiated from surveys, 357
wards, 93-94 Sea Search, 81
survey use in, 321 Seaview, 81
of terrestrial sites, 88 Sensors, use in magnetic searches and surveys,
Salvage masters, involvement in training of archae- 275--277
ological divers, 61 Serce Liman (Sparrow Harbor) vessel, 19
Salvors Settlement behavior, 216-217
contracts of, 79 Shallow-water sites, 270
cooperation with archaeologists, 85- 86, 92, deterioration of, 101
97-103 magnetic search and survey techniques for,
cooperation with federal government, 88 375--377
educational background of, 79 seabed movement of, 284
Sampling, 181-182,469 Sharmal-Sheikh site, 34
San Esteban, 556, 559 SHARPS (Sonic High Accuracy Ranging and Posi-
San Jose, 130 tioning System), 395, 403-410
San Juan, 537, 539-548 Sheytan Deresi, Turkey, 288
barricas from, 535, 541-548 Ship-burials, 27, 33
coopering technology of, 545--547 Ship construction
stowage patterns of, 535, 537-538 anthropological aspects of, 236-237, 238
Sans Souci castle, Haiti, 77 in the Philippines, 131
Santa Maria, 87 pre-I 650, examples of, 19-21
Santa Maria de la Rosa, 283 Roman, 19, 23
Santa Monica Bay region, aboriginal site location in, Romano-British, 33
253-255 surveys related to, 321
Santo Domingo, Conception exhibit, 77 of Viking-age vessels, 19-20
Sapphire, 34 Shipwreck Anthropology, 94
Saugus, Massachusetts, submerged colonial iron Shipwreck Anthropology Advanced Seminar, 94
works site, 249, 250 Shipwreck sites, see also names of specific ships
Scale models, 50 age of, 87-88
Scandinavia, maritime archaeology development in, buoyant hull, 228-229, 298-299
30 buoyant hull fracture, 228, 229, 298, 299-300
Scattering factors, 469 buoyant hull structure, 228, 229, 298, 299-300
Schiffer, M.B., 212-213 classification models/systems for, 524, 437-438
School of American Research, 94 common features of, 267-270
Scrambling devices, 243, 259, 268, 278- 289 computer-based data files of, 304
seabed movement, 268, 283-289 continuous, 228
wrecking process, 268, 278-283 definition of, 17
Schrodinger theory, 164 discontinuous, 228
Scientific practice, in archaeological research, 25, disintegration of, 275--276
167-168 environmental context of, 268, 270--274
Scientific research, paradigm for, 141 extracting filters of, 268, 275--278
Scientific Survery and Location Limited, 129, 130 federal jurisdication over, 102
Scourge, 106 identification of specific ships, 18
SCUBA (self-contained underwater breathing appa- identification of type of vessel, 18
ratus), 29, 47, 352- 354 intact, 17
Scuttled vessels, 18 intermediate, 272-273, 524
Chattahoochee, 263 provenience of, 88
of the Revolutionary War, 21 scrambling devices for, 243, 259, 268, 278-279
of War of 1812, 326 process of wrecking, 268, 278-283
Seabed movement, effect on shipwreck sites, 268, seabed movement, 268, 283-289
283-289 types of, 321
Seabrook Nuclear Power Plant, New Hampshire, 424 unidentified, 18-19
Seafaring communities, 25 Shomette, Donald, 328
588 Index
Side-scan sonar, 339, 493-494 Society of Antiquaries, 30
data acquisition and image processing with, 395 Sociology, middle-range theory of, 205, 206-209
Isis modular data acquisition system for, 387-393 Solomon, lsaaci, 326
new capabilities of, 381-386 Somers. USS, 105, 106
alphanumeric record annotation, 384 Sonar, see also Side-scan sonar
auto-ranging, 384 use in zero-visibility contour mapping, 493-494,
combined side-scan sonarlsubbottom profiling, 497,498
385 Sonic High Accuracy Ranging and Positioning Sys-
CRT display, 385 tem (SHARPS), 395, 403-410
digital processor, 383 Soto, Ricardo Soto, 454
full ocean depth operation, 385 South, Stanley, 1-2
graphic recorders, 381-382 South Carolina
K-MAPS, 383-384 Hobby Diver Program, 65--69
magnetic tape recording, 382 salvage activities in
magnetic tape storage, 382 legal regulation of, 65--66
side-scan distortion factors, 383 state archaeologists' involvement in, 90
very high resolution scanner, 385 state supervision of, 88--89, 93
Siebe, Augustus, 348, 349, 352 Underwater Antiquities Management Program, 67
Significance, of shipwreck sites Underwater Archaeology Division, Hobby Diver
criteria for, 431-439 Program involvement, 67
for Civil War shipwrecks, 432-439 South Carolina Institute of Archaeology and Anthro-
of National Register of Historic Places, 420-422 pology, 66, 69
of sea-bed distributions, 471-472 South Carolina Underwater Antiquities Act, 66
Simulation models, 178 Southeast Asia, maritime history of, 121
Sinclair, James, 100 Southwest Museum Services, 557
Sirius. HMS, 525 Space capsule, 88
Site analysis; see also Magnetometer surveys Spanish Armada, 236, 423
of discontinuous sea-bed distribution: see Statisti- Spanish galleons, 129
cal analysis, of discontinuous sea-bed dis- Spanish plate fleets, 33-34
tribution sites Spanish shipwrecks, see also names of specific ships
of medieval sites, 51 exhibits of, 553-558
Site formation processes in Gulf of Mexico, 304, 309, 310, 311
models of, 524 Spargi site, 270
relationship to middle-range theory, 209-210, Spaulding, Albert, I
212-213 Sport Diver Archaeology Management Program,
ofSt. Catherines Island, Georgia, 225, 227, 230 65--69
stream formation geomorphology, 259, 261-266 Sports diving associations, 55, 57
Site location involvement in archaeological diver training,
of aboriginal material, 253-255 58-63
factors affecting, 245 Stahl, Franklin, 157
Site visualization, in zero-visibility conditions, Star Carr, Yorkshire, 23
493-505 State Historic Preservation Offices, 420, 421, 422
Slaves, pottery manufacture and use by, 68 Statistical analysis
Slave trade, 465 of discontinuous sea-bed distribution sites
in the Patuxent River Basin, 324 clustering procedures, 483-485, 486
Small craft, typology of, 429, 441-445 dendrograms, 483,484,485,486,487
Smith, John, 324, 333, 334 Euclidean distance measures, 483-487
"Smudge Pits and Hide Smoking: The Use of Anal- meter square grid analysis, 477-488
ogy in Archaeological Reasoning" (Bin- open versus closed sites, 474, 476-477
ford), I product-moment measures, 483, 485, 486-487
Social systems, 206-207, 208 similarity matrices, 481, 482, 483
Societies, of terrestrial archaeology, 57-58 in hypotheses testing, 183
Society, definition of, 172 of observed variation, 178
Society for American Archaeology, I, 95 of sea-bed distribution sites, 469
Society for Historical Archaeology, 93 Steamboating/steamboats, 263, 326, 434, 435
Proceedings of, 565 engineering records of, 436
Index 589
Steed,81 Texas A&M University, maritime archaeology thesis
Steffy, Richard, 62 topics at, 49
Stemm, Gerg, 85, 86 Texas Antiquities Code, 557
Steward, Julian, I Texas Antiquities Committee, 377, 553, 557
Stockholm harbor, Vasa shipwreck site, 20--21, 32, Underwater Archaeological Research Section.
78,278 Matagorda Bay field investigations by,
Stockton, Maine, Defence shipwreck site, 21 363--373
Stone artifacts, from Parque Nacional Cahuita sites, Thailand
460-461, 462, 466 nautical archaeology programs, 121-122
Straits of Florida, shipwreck sites, 306-307, 308, Sattahip site, 34
309 Thames River estuary, South Edinburgh Channel
Stress, experienced by divers, 491, 507- 511 site, 289
Stryker, Rick, 553, 557 Theory, 2
Sub-Aqua Society (U.K.), 68--69 archaeological, 206, 210--211
Subbottom profilers, 251 comprehensiveness of, 215
Submarine escape equipment, 353 critical, 417, 422-425
Submerged cultural resources management: see Cul- definition of, I
tural resources management generality of, 215
Submerged sites, archaeology technology for excava- lack of, in archaelogical research, 141
tion of, 47 premature formation of, 146-147
Sulla,30 relationship to empirical studies, 207
Surveys, see also Magnetometer surveys ruling, 146, 147-148
definition of, 357 Thompson, James, 78
differentiated from searches, 357 Thomson, C.J., 29
methodology of, 317-321 Throckmorton, Peter, 2, 32, 48, 49
as pre-development requirement, 411 Tirpitz, 354
of specific ships, 319, 321 Titanic, HMS, 87, 88
Sutton Hoo boat burial, 27, 33 Today Show (television program), 86, 95
Sweden Toldeo, Quin Rodriquaz, 454
medieval maritime archeaology research in, 50 Tombs
tourism in, 78 Egyptian, 79--80
Vasa shipwreck, 20--21, 32, 78, 278 Megalithic, 169
Switzer, David, 419, 4230 Torre Sgarratta, economic evaluation of, 81
Szilard, Leo, 158, 159 Tourism
in Caribbean, 76-78
TACMAR (salvage company), 80, 81, 82 in Maine, 77-78
Taillez, Philippe, 31 museum-related, 557
Taylor, W.W., I in New Hampshire, 423-424
Technical importance, of shipwreck archaeology, in Sweden, 78
17-21 Treasure Diver s Guide (Potter), 103
Technology, 339 Treasure hunting: see Salvage
implications for maritime archeology, 49 Treasure Salvors, 97-103
Tectonic shifts, 245, 248 Trial,525
Tecumseh, USS, 105, 108, 109 Trinidad Valencia, 60, 274, 284
Terminology, nautical, 18 Triremes, 19
Terrestrial archaeology societies, 57-58 Trobriand society, 169-171
Terrestrial sites, 27-28 Tuck, James, 537, 539
salvage of, 88 Tulane University, shipweck-related archives of, 304
Territorial waters Turkey, 32
federal jurisdiction over, 102 Turks (islands), salvage activities in, 79, 80
French, U.S. government vessel shipwrecks in, "Two Cannon Wreck," South Carolina, 68
107 Tyre, 31, 343
Texas
Institute of Culture, 559, 560, 561 Ugalde, Alvaro F., 454
salvage law, 61--62 Ulu Burun, economic evaluation of, 81
shipwreck site preservation in, 93 Undersea Research Group, 31
590 Index

Underwater Archaeological Associates, Inc., 433 Warfare, see also Revolutionary War shipwrecks sites;
Underwater archaeology, stages of, 62 War of 1812 shipwreck sites; World War II
Underwater sites, see also names of specific sites archaeology of, 423
archaeological methodology for, 47 War of 1812 shipwreck sites, 106, 324, 326
diving techniques for, 47 in Patuxent River, 327, 328, 330-331, 333
Union Leader (newspaper), 424-425 HVasa,20-21,32,78
United Kingdom wrecking process of, 278
Admiralty, 24 Watson, James, 157
amateur underwater archaeology in, 57 Weinrich, Mareel, 159
economic impact of the arts in, 76 WendorLFred,365
maritime archaeology development in, 29--30, 33 Wentworth, John, 423-424
Romano-British ship excavations, 33 Whaleboats, 19,434,435; see also San Juan
shipwreck sites, environmental context of, Wheeler, Mortimer, 24, 168
271-274 White, Leslie, I
Thames River estuary site, 289 HVhite Squall, 81
United Kingdom Digital Marine Atlas Project (UK- HVhydah, 89, 90
MAP),527 Wiggins, Julius, 508
U. S. Army Corps of Engineers Wiley, William, 366
as areal studies information source, 411 Willey, G.R., 213-214, 215
Berth 52 vessel (09CH691) involvement and, Williamsburg, Virginia, I
187-188 Windmill Hill culture, 168, 172
Mobile District, 376 Wooden shipwreck sites, see also names of specific
New Orleans District, 376 ships
Savannah District, 494 disintegration of, 18
Wilmington District, 433 identification of, 18
U. S. Marines, 326 World Explor, 81
U. S. Naval Academy, 106 World War II, scuba use during, 354
U. S. Navy World Wide First, Inc., 129--130
Director of Ocean Engineering, 108 Wrecking, process of, 259, 278-283
Diver's Manual, history of diving chapter,
343-354 Xantho. SS, 525
Naval Historical Center, 73 Xerxes, 343
Patuxent River facilities, 326 Xi amen University, nautical archeology program,
shipwrecks, 105-109 121

Valentine, 88 Yang, C.N., 159

Valerio, Carlos E., 453, 454 Yassi Ada site, 32, 278, 471
Van der Heide, G., 32-33 economic evaluation of, 81
Variability, organizational, 216-217 Yorktown, 34, 521
Variables, statistical analysis of, 178 Yorktown, 525
Variation, explanation of, 177-178 Yugoslavia, 32
Vee-bottom boats, 442-443
Veracruz, St. Juan de Ulloa castle, 77 Zeewijk,34
Vergulde Draeck, 34 Zero visibility, 491
Verification, standards of, 175 contour mapping in, 493-494, 497-501
Vero Beach, Florida, 1715 plate fleet site, 78, 79 SHARPS use in, 395, 403-410
Verulamium, 29 site visualization in, 493-505
Victory, 20 Zero-visibility diving
Video, use in zero visibility, 493-494 on the Maple Leaf, 513-517
Video image digitization, of the USS Monitor, stress experienced by divers in, 491, 507-511
397-402 Zero Visibility Maze, 507, 508, 510, 511
Viking-age ships, 19--20, 27 Zuider Zee, vessel remains on reclaimed lands of, 27
Virginia. CSS, 106 Zwamrnerdam, 33