Acta Archaeologica vol. 72:1, 2001, pp.

1150 Printed in Denmark All rights reserved

Copyright C 2001

ACTA ARCHAEOLOGICA
ISSN 0065-001X

TABLE OF CONTENTS
INTRODUCTION PART I. GENERAL 1. Notions of theoretical archaeology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PART II. CONCERNS OF ARCHAEOLOGY 2. Subject matter of archaeology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Archaeological sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PART III. NATURE OF ARCHAEOLOGY 4. Methodological nature of archaeology . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Principles of archaeology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PART IV. ARCHAEOLOGICAL THEORY 6. Empiricism in archaeology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. What is archaeological theory? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8. Structure and working of archaeological theory . . . . . . . . . . . . . . . . . . . . 9. Functions of archaeological theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11 19

32 42

57 66 72 87

PART V. ARCHAEOLOGICAL FACT AND RESEARCH DESIGN 10. Archaeological fact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 11. Archaeological research design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 PART VI. CONCLUSION 12. Panorama revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 APPENDIX: The Commandments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

Acta Archaeologica PART I. GENERAL

1. Notions of theoretical archaeology

1. THEORETICIANS AND PRACTICAL ARCHAEOLOGISTS In archaeology there has long been a difference of interests. Some practitioners were devoted to eld archaeology and at the centre of their lives were the expeditions, surveys and excavations. Others were committed to the possibility of substantiating an idea, such as Schliemanns work to prove the reality of Homers Troy, and some were museum curators, keen to sort through collections and systematise their material. Identifying artefacts, managing exhibitions and the publication of catalogues of material all offered them great delight. Happiness was to nd an analogy, to place an artefact into a system. There were also those who were drawn to ancient artefacts by their mystery and enigma, who were interested in ideas that might be obtained from the artefacts. From this type of enthusiast either empty dreamers originated, ideological speculators, or historians with a spade. But sometimes, if they were rapt by the extraction of ideas as a process itself, they developed as theoreticians. Gustaf Kossinna and Gordon Childe, Walter Taylor and Mats Malmer, David Clarke and Lewis Binford all undoubtedly belonged to this latter type, and developed theories. Less known theoreticians brought statements to more exact wordings, added new arguments, and systematised theoretical knowledge. I am attracted to archaeology by its enigmas, but particularly by its theory. It is generally said that theory serves practice and that it is only problems originating in practice which provide stimulus to theoretical considerations. In Archaeological Typology (Klejn 1991a) I have also paid debt to this idea when describing the practical problems that stimulated me to turn to typological theory. I pointed out then that many students are not led to theory by the same difculties. However, the demon tempter was probably always within me. I began student life not in archaeology, but in linguistics and folklore-studies, dealing with problems of ethnogenesis and semiotic searches for original sense, under the teaching of Vladimir Propp. Then I became interested in archaeology and was taught by Mikhail Artamonov, who also was interested in ethnogenesis. My rst published work in archaeology appeared in 1955 in the form of a critical review article on the origins of Slavs. Among my articles in the 1960s were theoretical issues concerning questions of typology and ethnogenesis, and since the 1970s I have continued to write theoretical papers. In general, practice certainly does build a foundation for theory, but only with the help of theory can our knowledge grow in bounds. Scientic revolutions actually occur through the process of theorizing. There would be no science and no scholarly study without theory. In archaeology this truth was ignored for longer than in other disciplines, and it is still not fully realised now. This raises the question: Are specialists in theory needed? Is the demarcation of such specialisation in archaeology justied? I know from my own experience that to deal with theory one must study many non-archaeological matters that are not necessary for practical archaeologists, and in archaeology itself a great deal is also brought to light which is required knowledge. One might well combine theory with practical studies, and to some extent it is even necessary otherwise theory risks becoming airy and without object but few archaeologists are able to succeed in both elds. The post-processualists, or post-modernists, are against isolating the theoretician. The opposition was announced at TAG 1992 by Julian Thomas (1995, 351). He stated ironically that each university department [of archaeology] should have one theory specialist, just as it should have one lithic specialist, one ceramic specialist and one environmental specialist. In his opinion the truth is ignored that all archaeology is theoretical. The theoretician Michael Shanks declares that, I am ... very concerned that

Metaarchaeology
theoretical archaeology is being dened more now as another expert eld. It is argued that just as an excavation report or department must have its specialists in techniques, environment evidence, pottery, dating, leather, bone, quantitative analysis, whatever, so too they need an expert in theory seen as a separate matter of explanation and interpretation (A dialogue 1994, 19). Shanks sets against this a view that everyone must know theories and everyone must be able to apply them. Must know, must apply. Yet is it reasonable to insist that everyone must be able to work them out? Speaking against Thomas and Shanks is their own experience; they have not carried out very much practical work in archaeology as compared with their theoretical activity. It is clear that for expeditions and for museum processing of materials many more people are needed than for interpretation. However, there was also something else being said here. First, in order to deal seriously with theory one needs another branch of knowledge and another mind set than one does for practical studies. A theoretician needs a sharp ability to establish distant associations, a capability of systematisation, a vivid imagination, a certain courage in thinking and good articulation of ideas too thoughts must be put into very tting words. It is desirable that he or she is trained in logic, mathematics, linguistics and philosophy. A possession of the main languages would not go astray either. Since these needs were not realised, such a combination of traits very rarely occurred among archaeologists. Secondly, theory was not regarded with much respect for a long while, it was in fact equal to empty rhetorics. For quite some time all theorising was regarded as speculation, and all speculative thinking (i.e. thinking abstracted from direct consideration of facts) was seen as unscientic and unscholarly. Looking at other disciplines later convinced archaeologists of the necessity and respectability of theories but it did not acquaint them with the complexities of the business and did not impart them with great prudence for theory. So, a person having made a new guess on, for example, the chronological distinction between charnieres of two bulas or something else of the same crucial importance, in a somewhat bombastic manner declares: ... according to my theory .... (Gjessing 1962, 504).

Another contributing factor that raised the status of theoretical work was the connection of archaeological theories with philosophy, methodology, and strategy of studies. However, strange as it is, this did not undermine the notion that practical experience and factographical erudition is everything and the only tools we need for successful theoretical work. This led to the point that producing something theoretical began to be thought of as a matter of prestige and privilege for renowned empirical workers. A similar situation in ethnography gave the Hungarian scholar Hofer occasion to rally for the fact that in Europe, as distinguished from America, theorising ... is generally reserved to the peers of the science (Hofer 1968, 315). Finally, in more recent years, with the reinforcement of scholarly knowledge, a realisation began penetrating archaeology that the theoretical level of studies has its own specicity, that it demands a special, different education, needs its own specially prepared cadre by the inux of youth, and that it be separated into a special branch of archaeology (Zakharuk 1971b, 12). The relation of archaeologists to theory is two-fold. On the one hand, a lot of practising archaeologists eld people and museum curators regard theory as an addle business and relate to it with some irony. As Kluckhohn noted (1940, 46), the greater number of anthropologists (and he counted prehistorians among them) still feel that theorising is what you do when you are too lazy or too impatient or too much of an armchair person to go out and get the facts. On the other hand, many practical workers harbour a secret envy at theoreticians since the latter always have a wider horizon, broadened by their knowledge of philosophy and anthropology. The same envy arises on account of the fact that theory is always evident or latent in interpretation, the real nub of archaeology. And as theory is nearer to philosophy and to outward thinking, i.e. to ideology, theoretical studies are considered a privilege that should be offered for merits either for great achievements and experience in practical archaeology or for a high administrative seat. Therefore, especially in the socialist states, where theory was considered a leading force, old practical workers and heads of research institutions frequently advanced in the shoes of theoreticians.

Acta Archaeologica
What is essential here is that theoretical archaeology is inscribed in a whole system of analogical branches of other disciplines. Monographs systematising the theoretical knowledge in these disciplines are simply entitled Theoretical ... (with the designation of the discipline following). There are examples of such book titles in biology (Bertalanffy 1932), geography (Bunge 1962), sociology (Merton 1967), linguistics (Zvegincev 1967) and so on. Evidently a general notion of the theoretical branch in these disciplines is established, and it remains only to apply the format to archaeology. It seems in this case that it would be simple enough to give a denition of theoretical archaeology. It is clear that theoretical archaeology must embrace theoretical studies in archaeology. In fact the question of this denition has its own difculties that arise from various circumstances. First, it is disputable as to what can be considered theory in general on the one hand and theory specic to archaeology on the other. Correspondingly, which studies can be considered theoretical those only concerning concepts and laws or others too? For instance, there are notions that theory is simply generalisation of facts, and another that it is a set of methods. If one accepts the former, then it would be very difcult to delimit archaeology from numerable surveys of materials. If one accepts the latter, then all the works on methods used in archaeology must be included. How to deal with the methods of other disciplines used in archaeology? Secondly, the question of differentiating between ideas arises. Many people call any interpretation of sources a theoretical activity, and there is a reason for this the interpretation constantly involves theories. Indeed, is not the whole business of interpretation a job for theoretical archaeology? It is without doubt that the analysis of the mental process of an archaeologist, of the development of his or her ideas, is closely connected with theoretical thinking. During the second half of the 20th century it became clearer and clearer that the development of ideas of the entire discipline in its historical course also has a direct relation to the theoretical grasp of archaeology. But how can the history of theoretical thinking be divided from the history of archaeology in general? How can it be separated from the history of discoveries, excavations and biographies of scholars?

The idea was put forward that experience is the best guarantee of a good theory, but they had the background for creating theory no more than they did for choreography. Their constant attempts at theoretical pirouettes merely increased the disregard other archaeologists had for theory. The opposite extreme frequently occurs among theoreticians. Tilley, known for his theoretical works and radical ideas, illustrated this in discussion with me when he recounted with resentment how an older colleague advised him to make an initial study of a remarkable amount of material a monument, site or a period before advancing with theories (Tilley 1991). The advice was not entirely misplaced. Both eld experience and museum practice are necessary for a theoretician if he or she is to realise what is hidden behind idealised objects. Though no more necessary than for a theoretical physicist like Einstein, who needs to be experienced in experimentation. It is desirable to avoid the mire while gaining experience. A solid experience in practice is usually reached after many years, by which time it is too late to set about the theory.

2. THE CONCEPT OF THEORETICAL ARCHAEOLOGY What is presented here as theoretical archaeology was in some circles termed, and still is termed, differently. In Soviet literature there was methodology of archaeological science (Zakharuk 1969); in America theory and method of archaeology (Hawkes 1954; Willey and Phillips 1958) or epistemology of archaeology; and in Britain general or central theory of archaeology (Clarke 1968, 663; 1973, 15f). There is also a suggestion to call it archaeosophy (Szekely 1977). However, for this branch the designation theoretical archaeology is becoming more and more xed. The rst appearance of the term was ascribed to me (Gardin 1980, 126; Barich 1977; 1982, note 10), but this is not quite correct for it emerged spontaneously in different places and by different authors (Sher 1972; Fowler 1977, ch. 5; Gardin 1979; 1980; Klejn 1977a; Barich 1977; 1982; Hodder 1982; Holtorf and Quensel 1992; Guliaev 1993 a.o.). I have even found the term in publication as early as 1971 (Cuadernos).

Metaarchaeology
Lastly, the question must be asked whether theory in archaeology is possible at all or if archaeology is a descriptive, factographic discipline. There are views that even if theories do act in archaeology they are not archaeological theories but anthropological or historical ones. This would be a case of improper theoretical archaeology. Therefore, it follows from all these reasons, since the very beginning of the separation of the new branch and of having realised this, that debates take place on the reasonable effectiveness, the subject matter and the borders of theoretical archaeology. One of the rst denitions of theoretical archaeology was given in my Panorama of theoretical archaeology (Klejn 1977a, 1f). When looking at theoretical monographs in other disciplines I noted frequent extremes; some authors dening the theoretical branch broadly, including both philosophical problems and particular theories, others prefering a narrow denition, embracing merely the general theory of the given discipline. As for myself, I have chosen the middle of the road. I issued from the fact that the existing notions of theory were various and diffuse, and to leave them outside the branch was inconvenient as my task was only a survey. I also derived my view from an understanding that people have to specialise in a whole agglomerate of interconnected problems usually designated as philosophical, methodological, logical, theoretical and in part historical. Yet from these problems closely interconnected and acting under the guise of theoretical archaeology, I considered only those which were specic to archaeology and which embraced the whole of archaeology. Fowler leaned to a broader denition. Although he admitted the practicability of a narrower denition he referred to the involvement of theory in any interpretation, be it unconscious or not, and came to the conclusion that in a certain sense the whole of archaeology is theoretical (Fowler 1977: 131). More radical inferences were made by Michael Shanks from the same idea. In the discussion with Iain Mackensie he declared that ... there is no such thing as theoretical archaeology. Theory should be part of what everyone does critical self-consciousness (A dialogue 1994: 19). On the contrary Gardin regarded my accommodating denition as too broad. He indicated that I

introduced it into Soviet archaeology where it was equal to general archaeology but it is clear from the above references that it was not introduced by me and not only in Soviet archaeology. That general archaeology is broader than the theoretical one, is not difcult to guess from the amount of handbooks, reference-books, dictionaries and encyclopaedias on archaeology, which are all general archaeology but not theoretical. Gardin himself denes theoretical archaeology as the logical structure of scientic constructions in archaeology (Gardin 1980: 126; 183: 196) or as science on considerations produced by the pens of archaeologists, expressed in a symbolical, sign form, or more simply as science on symbolical constructions of archaeologists (Gardin 1983: 245). Gardins view is also close to that of Sher, who views theoretical archaeology as the theory of processing of archaeological data (1973: 55), and as the disciplines section which studies general rules of description, analysis and interpretation of archaeological records (1976: 68). Theory is a concept from the logic of science, but to reduce the whole of theoretical archaeology to logic, to rules and sign constructions seems to me unreasonably limiting. It is hardly fruitful to cut off the philosophical, methodological, cognitive and historiographical aspects. This can lead to simplication and schematization of problems, as can be found in Gardins book. On the question of the validity of theoretical archaeology, even if archaeology is viewed as rather descriptive, it will at the very least need theories of scientic description of archaeological material. Best of all, the question of where theoretical archaeology belongs is solved by the life of archaeology. Despite the frequently diagnosed crisis (Gaffney and Gaffney 1987) and the prognoses of its end (Kolpakov 1992) theoretical archaeology still exists. It passes in and out of channels of the discipline, broadening in places and narrowing in others, it introduces certain problems as fashionable while offering no attention to others, but all in all it continues to grow.

3. POINTS OF DEPARTURE Every theorist, including myself, has his own point of departure and his own approach. To a large extent they are determined by training, background and so-

Acta Archaeologica
often betrayed them. The role of personality and of chance makes history unpredictable on exactly the points that are usually most desirable to foresee. Rather, the very gravitation towards theory was to a certain extent cast by Marxist training: Marxism is in its ideals theoretical. Marxism used dialectics as substantiation for the necessity of revolution. Yet this is a too situationbound use of the method. If dialectics is considered as teaching on contradictions, not only of development but of modern world and conscience, it appears much more interesting. Contradictions, antinomies and oppositions must be considered not as anomalies this is just the way the world is built. Marxism necessarily compelled the removal of the contradictions by revolution. However both sides of a contradiction can be right. And one must learn not to take them away, but to live with them. Turning different sides, one phenomenon and one concept reecting it can manifest themselves as different concepts. Different concepts can appear as different sides of the same one. The spirit of such dialectics can be felt in my work. From Hegel, Marxism borrowed not only dialectics but also the bureaucratic spirit of Prussian state organisation as an ideal. The interest to Neo-Kantian teaching served me as an antidote against such Hegelism and Marxism. I rst became acquainted with Windelbands and Rickerts works through readings of Marxist criticism of them. Yet this criticism was enough to cause the reverse impact and it was the criticised conception that impressed me. The explanation of the specicity of humanist knowledge, the view of the classication of disciplines, which was new as compared with the view of Comte and Engels, and the view of the interrelation of fact with law in them all gave me an anchorage outside Marxism. My background and milieu, Russian archaeology, has its own elaborated traditions, notably distinct from West-European. Of course I received these local traditions, and they used to introduce not only a peculiar colouring into my works but certainly permitted me to take an independent and prospective position. So, in Russia there is no division of archaeology into two disciplines, as is commonplace in the West: classical (often with oriental) and prehistoric (with medieval). In Russia archaeology was and is one. For me

cial positions, which in turn are probably conditioned by temperament. My own training was, like all my compatriot colleagues, a Marxist one. I began with a rejection of Marxism while still in high school, where I hammered out an underground organisation of liberal standing. Then there was a short period of being rapt by Marxism, and then, still on the student bench, it changed to a criticism of Marxism, a criticism that grew stonger and stronger. I identied with Marxism only outwardly and at any rate, I needed to in order to keep the possibility of working at university. So, I tried not to use the term Leninism. I certainly did not accept the false Utopia of the Dictatorship of the Proletariat and I was very sceptical of the bright communist future, nor did I ever enter the Communist Party. In my opinion Marxism has three main drawbacks. Firstly, Marxism ignores the biological basis of mans nature and reduces man to interaction of economic factors and interests. This transforms Marxism into a narrow populist doctrine, and its aims into Utopia. For instance such an instinct as the love for ones own children prevents discarding the tradition of inheritance, but without discarding the tradition, real equality is unachievable. The second drawback of Marxism is the preference it gives to the future over the present and puts the state above the personality what is evidently anti-human. Thirdly, Marxism was always distinguished by great zeal and was ready to apply any means for the sake of its utopian aim. All of this sickened me. Yet the short affair with Marxism did not leave me without trace. Many views of materialism and methods of dialectics were deep-rooted in me and did not meet with inner resistance. I am a denitive atheist and up until now can only look at religious practice with irony. I give attention to the material and socialpolitical conditions of scholarly creation and analyse social roots of theories seriously, although I do not consider these roots as my only ones. In my opinion, the claims of indeterminists of absolute free will of the individual are simply laughable. There are limits to free will. While a person can do really whatever you imagine, although the more extreme the more unlikely, a mass acts with the regularity of nature. Otherwise sociologists would have had nothing to do. On the other hand the rectilinear thinking of Marxists

Metaarchaeology
it was the initial norm, and this determined much of my understanding of archaeology. It was not a part of history, but a source-studying discipline common to all branches. This was also a protest against ideology and politics involved in archaeology. Hence the strive to create an integral theory for it. No doubt, the impact of my teachers shows too. The founder of Russian semiotics and Russian structuralism Vladimir Jakovlevich Propp, under whose direction I had a fortune to study, was often accused of Formalism, both by Marxist ideologists and by Levi-Strauss. Yet how is content to be understood if not through form? I listened closely to lessons on such Formalism and this is why I researched so long and carefully on the cognitive potentials of typology. My second teacher was Mikhail Illarionovich Artamonov, archaeologist and art-specialist, many-years Director of the Hermitage Museum. He was notable for the independence of his judgements that were advanced against several conceptions produced for the situations of the moment. Noteworthy too for his evident orientation to the West. I liked this position and it seems to me very contemporary even now when the alternatives are discussed about which way Russia is destined for, the European civilisation or some special way, essentially Asiatic. Firstly, like my teacher, I evinced an interest for Western literature and common world science, and secondly, I devoted my entire work in theoretical archaeology to working out a system of theories and methods which could grant strictness and objectivity of archaeological cognition, in order that archaeology would be able to resist the intentions to serve the political situation of the moment. Propps Formalism was a support for me in this too. On the other hand, criticism of Neopositivism in general and of Analytical philosophy in particular, both very developed in Soviet methodological literature, kept me back from the theoretical task of creating an analytical machine for archaeology. Nevertheless the task of axiomatisation of archaeology remains attractive for me. Yet in that and in general I tried to escape simplications. It altogether assured me of my own position in theoretical archaeology, which I hope will not be without interest for the reader.

4. THE STRUCTURE OF THEORETICAL ARCHAEOLOGY The coverage of theoretical archaeology is very broad. A number of diverse problems are united here, and for the sake of comprehension one has to order them somehow. One should present theoretical archaeology as a system, rationally differentiate its parts and place them in a logical order. The rst person who presented the structure of theoretical archaeology was David Clarke, a very analytical and conceptual thinker. In his Analytical Archaeology (1968) he presented a general theory of archaeology. It consisted of three models: for archaeological research design, archaeological objects and for archaeological processes. The rst model dictates how to proceed from archaeological records to their interpretation, describing what steps are needed on this route. The second model grasps the nature of archaeological material, the nature of which is conditioned by the material belonging to culture and secondly by the peculiarities of human creativity, which is complex and diffuse, despite its regularity. The third model provides a framework to study the dynamics behind the archaeological material the changes, growth and evolution of ancient cultures. As one can see, the second model is directed to the material and to records, the third one to the past hidden behind it, and the rst one to the process of cognition connecting the second with the third. Some years later Clarke (1972b, 238) described the three-part division of theoretical archaeology otherwise. He divided it into archaeological metaphysic, archaeological epistemology and archaeological logic. Under metaphysic he understood clarication of archaeological concepts as well as their interrelations with reality and their limitation. Under epistemology he had in mind the prospect of revealing the specicity of archaeological information and consequently modes of archaeological cognition. Archaeological logic was understood as the development of the rigor and explicitness of archaeological reasoning. In his article Archaeology: the loss of innocence Clarke complicated the structure. He considered the components of theoretical archaeology from the point of view of archaeological philosophy. These components consisted of: (1) theory of concepts, (2) theory of information, (3) theory of reasoning, and nally, (4)

Acta Archaeologica
chaeology is structured very simply; it has taken account of only one aspect of theorising, the logical one. According to the suggested treatment of archaeological cognition by Gardin everything is divided in two parts: compilation and explication. Roughly speaking, these are the collecting of material and its processing by interpretation. This can be equalled to Clarkes general theory which is in a similar position of a research design with steps, but Gardins scheme is more scarce and simplied. In Soviet scholarship the late Juriy Zakharuk thought much on the structure of theoretical archaeology (1969; 1973). As he worded it, on archaeological theory. Within cognition he distinguished between the object and the subject matter. The rst one to him is a fragment of the reality under study, the second its reection in scholarly knowledge. The object consists of aimed object (particular societies of the past) and immediate object (results of their activity the material culture). The subject matter is also divided into the object of the discipline, which is the archaeological records and sources, and the results, and these are reconstructed societies of the past and their history. On Zakharuks schemes these components are situated in the sequence in which the information proceeds, and the sequence is obtained in the process of cognition. Everything is like in Clarke and Gardin, only the number of stages is different: real societies of the past their material culture archaeological records reconstructed societies. Accordingly he postulates theories building the united structure. At such approach the following row would be expected: (1) theory of society, of social development, (2) theory of culture (or more correctly, of material culture), (3) theory of archaeological sourcestudy, and (4) theory of reconstruction. The whole pathos of Zakharuks building was directed to the idea of showing that archaeology is not to be reduced to naked artefactology, that is to say, archaeological source-study is only a part of archaeology. This was in the watercourse of the general ghting for the afrmation of the Arcikhovskij Rybakov conception dominated in Soviet archaeology that equated archaeology with history (Arcikhovskij: archaeology is history armed with the spade). However at such treatment one should include theory of social development and theory of

general theory of archaeology. The rst is characterised by him as metaphysical, the second as cognitive, the third as logical (1973: 15) and the fourth is not characterised at all. Evidently it is the correct structure of archaeology for him. In the text he connects the rst component with notions on objects of study and on the dynamics of the cultural process, the rest with research design. The general theory is divided up by the steps of the research design, that is, by steps of the archaeological cognition. It seems to me, it is here that the most important contribution of this article is contained. Clarke enlisted these parts: (1) Pre-depositional and depositional theory it has to consider the interrelations between activity and ideas of ancient people and their remains and traces; (2) Post-depositional theory it considers changes that occurred with the remains and traces after their positioning in the earth or on the earth; (3) Recovery theory considers interrelations between the archaeological objects that appeared on the place after all the changes produced by the time and all what is obtained by excavations and transformed into data (it is mainly theory of selection); (4) Analytical theory it is mainly theory of elaboration of the material with various methods; (5) Interpretation theory it establishes interrelations between the obtained results of analyses and the events and processes inaccessible to observation. This line is clear and reasonable but its interrelations with the rst three of Clarkes schemes are not so clear and neither are their interrelations with each other or their comparative important. Why is a theory of concepts favoured, but not theories of laws, of principles, criteria or language of theory? If theory of archaeological information is included into the list, why does theory of archaeological systems, of ordering of archaeological material not deserve this too? If theory of reasoning is here, why is theory of modelling absent? Something here is not well thought over, and perhaps a broad eld of systematisation remains left out of the parts of research design. Only some components of different ranges are selected for study. As could be expected, Gardins Theoretical Ar-

Metaarchaeology
culture into theoretical archaeology but they are theories of other, independent disciplines such as sociology, history and culturology. In my Panorama of theoretical archaeology I positioned literature by periods and directions, I did not need other structure. Yet when I worked on Archaeological Typology the problem arose in front of me to determine the place of typology among other archaeological theories, to establish their logical interrelations, i.e. the structure of theoretical archaeology. Incidentally, I also considered then the general theory of archaeology. Like my predecessors Clarke and Gardin on this question I positioned some kernel of archaeological theories in line along which information proceeds from facts to their interpretation. These theories deal with archaeological material and are intended to give ideas for its processing, for interpretation, for extracting historical information from it. I have called these theories endoarchaeological (inner-archaeological, intra-archaeological, or properly archaeological) as they are in the very kernel of our activity. Theory of classication and typology belongs to them as well as theory of migration recovery. However, many archaeological theories, including those I was dealing with, do not t into this kind. They are not directed toward archaeological material, its peculiarities and possibilities of its interpretation, but toward the archaeological discipline itself, toward its perfection. These are philosophical, cognitive and scientic problems as well as being problems of logic and psychology, so to say self-criticism and self-cognition of archaeology. When determining the subject matter of archaeology, its methodological nature, its meaning, the character of its facts and of theory itself all these tasks belong here. That is a theory of theory ensues. Thus, it is archaeological metatheory. As soon as Renfrew applied the term metaarchaeology I have taken it into my tool kit and called this group of theories metaarchaeological. For many a long day archaeologists saw only nds and what is hidden behind them an Indian behind the artefact, after the catchwords of R. Braidwood (1959a, 79). There was, and is, meant exactly this by archaeology. Renfrew (1969a, 243) called the methodological study of the discipline itself, of its general theory and methods, metaarchaeology. Under this term he seemed to imply ironically a cliche of D.

Clarke. It would be more exact to speak of archaeological metatheory, but Renfrews phrase possesses potentials for a broader use and therefore has come to stay. According to norms of the Greek language one should write this term as metarchaeology (when a prex in Greek ends in a vowel and is followed by another vowel, then the vowel from the prex is dropped). However, in newer languages the borrowed prex meta- only occurs in the full form (probably due to words like metaphysics, metaphor, metamorphosis, metastasis). This does not exhaust theoretical archaeology either. Still more theories remain that do not properly belong to archaeology but have a signicant meaning for it. These theories developing mainly in other disciplines like sociology, history, culturology, anthropology, ethnology, linguistics, geography and others, nevertheless elaborate on concepts and laws which are condition sine qua non of functioning of archaeology. Among these concepts are culture, artefact, ethnos, society, epoch, etc. These theories establish regularities of migrations, diffusions, autochthonous development, inuences, borrowings, independent inventions and so on. I do not mean their hallmarks in the archaeological material, but the laws according to which they were developed in living societies in the past. I have called this group of theories paraarchaeological. Belonging to this is concepts of culture, material culture and historical process. Archaeologists need to study them very carefully, often also to work them out with regard to special needs of archaeology. So the former elaboration of ethnological theory did not take into account the special interest of archaeologists to know how in the remote living peoples their entire culture is connected with their material culture, especially with its part which is able to last during thousands of years. It was necessary to create a special discipline, ethnoarchaeology. Thus, I have outlined the structure of theoretical archaeology as three-part, consisting of endo-, metaand paraarchaeological theories. I consider them only in this grouping. The three-part division later suggested by Lester Embree (1989; 1992) of the structure of American theoretical archaeology coincides to a considerable extent with my structure. There is metaarchaeology in both our systems, his proper theoretical archaeology corresponds to my endoarchae-

10

Acta Archaeologica
though philosophical issues are touched on, as they are in metaarchaeology. I begin with metaarchaeology because it allows me to present a general systematic survey of the discipline and present its initial concepts and principles.

ology and his philosophy of archaeology corresponds to some extent to my paraarchaeology since it concerns culture, historical process etc. However from my understanding paraarchaeology does not coincide with Embrees philosophy of archaeology, al-

Metaarchaeology PART II. CONCERNS OF ARCHAEOLOGY

11

2. Subject matter of archaeology

An object is perceived differently by an eye and by an ear. So, the object of the eye is a different object than that of the ear. Karl Marx

1. REACHING UNDER THE CARPET When in 1987 heads of archaeological departments from all British universities elected a Committee for outlining the future common policy in the difcult years of reduced budgets, the Committee spotted the unexpected truth. Absolutely fundamental is the denition of archaeology as an academic discipline ... Do we regard a degree in archaeology as a collection of topics gathered together for convenience under this heading ...? Or do we see it as dependent on a compulsory inner core, the irreducible heart of the discipline, without which no student can be properly taught? (Austin 1987, 230). The Committee chose the second alternative. Yet some problems were brushed under the carpet. Most of our subject has strong links with other disciplines ... with history, ethnology etc., so the borders of the discipline appeared to be diffuse. However the question is not new. As distinct from some other problems of metaarchaeology, the determined concerns of archaeology, its eld and its subject matter, have for a long time appeared the necessary topic of any solid general textbook in archaeology. What has changed with the appearance of metaarchaeology? First, the role allotted to the analysis of such problems has changed. The exposition of useful results from a study used to be aimed at rather than the elaboration and checking the means of obtaining them. So, having exposed his considerations on aims, means and method of archaeology, Sophus Mller (Mller 1898) did not begin with these considerations to his monumental survey of northern antiquities but placed them at the end. Such a practice was usual even for works of more recent times. Now it is impossible to begin a work without exposing and analysing its premises. ... No archaeological study

can be better than the ideological assumptions which underlie the development of its arguments (Clarke 1968, XV). Secondly, the way matters are considered has changed, and with it the types of implications. In the past they were advanced with some awkwardness as a statements of banal truths. In the same book Sophus Mller saw it necessary to set the reader at rest by promising to make the methodological problems as clear as possible but also as briey as possible. Because truthfully speaking it is hardly tempting to create such an outline or to become acquainted with it. Yet nevertheless ... (Mller 1898, 292). Meanwhile Buschor would scarcely agree to writing a methodological preface to a handbook of archaeology since he supposed that explanations on the concept and method of some discipline can only bear a character of self-cognition (Buschor 1936; 1969, 3). That is they are subjective and divested of all objective grounds. Nowadays the necessity is accepted not only to elaborate strictly on methodological problems, but to rethink well-tried initial principles of archaeology (Chang 1967b). Thirdly, the perception of the structural building of archaeology has changed. Various branches of archaeology were formed within different spheres of knowledge and were traditionally experienced as the separate disciplines of prehistory, classical archaeology and other categories. These branches have seen great difculty in undergoing rapprochement. The progression from sources to inferences was considered in a far more simple way than in todays clear divisions into a step-by-step process. Childe was one of the rst to note that archaeology is one (1956, VI), and despite the conservatism of some classical scholars, archaeology as an explanatory science or as a system of methods is now viewed as gradually more united in the chronological span of its subject matter. It is also becoming more complicated and segmented in its steps of cognition. More than one hundred years

12

Acta Archaeologica
subject matter to the sum of the aims or tasks of cognition (Rakitov 1964, 375). These suggestions seem unlucky. First of all they introduce again an unnecessary synonymous nature, a duplication of terms. Secondly they return again to a view of reality as exclusively separate from the tasks of cognition and the concept thereby disappears in which they could unite. Besides, it is hardly fruitful to take away the laws and regularities from the object. Suggestions also occur to place with the term of subject matter of the discipline the community of knowledge on the object, the reection of the object in the sign system of science and the like, i.e. the result of research activity (Sadovskij 1965; Melnikov 1967, 21; cf. Rakitov 1971, 3763, 75). However, here the necessary terms are also present without such designation (knowledge, facts, evidence, theories). Such understanding can only produce doubts in the necessity of the very distinction of object and subject matter (cf. Trudzik 1971, 5558). Besides, through such understanding the aspect of the consideration is also ruined and the concept fades away that was able to grasp the activity, which is the actual purpose of research cognition. As the tasks of the cognition are an important component of the subject matter of the discipline their presence or absence is told in the whole of its structure. That part of the reality which is involved in both the object and subject matter of the discipline is dismembered within the subject matter. Two poles can be recognised in it (Zak 1968: 456462, 471, cf. Rakitov 1969: 169178): (A) Extrinsic phenomena accessible by direct observation and registration these are immediate objects or initial materials, sources or records; (B) Intrinsic, hidden essentials (laws, causal links etc.) which while characterising the object are just desirable, outlined and attainable aims of studying these are aims of cognition. The task of science, Marx noted, consists in reducing the visible movement only manifested in the phenomenon to the actual inner movement. This division is important for the determination of the distinctness of a particular discipline. The right of a

ago Ivan E. Zabelin lamented that it is very difcult to clear up for yourself even the very subject matter of archaeology. Indeed, what exactly is the business of this discipline, what does it elaborate? (Zabelin 1878, 1).

2. INITIAL PRINCIPLES AND CONCEPTS In the systematic description of any particular discipline its subject matter is the rst component. The concept itself, the subject matter of a particular discipline, was elaborated in detail by Soviet philosophers and scientists. However, even among them divergences exist and there is still not an agreed terminology. This is why it is necessary to stipulate which understanding of the subject matter of archaeology is of present concern. The subject matter of cognition, in particular of the scientic, is interrelated with the object of cognition. The separate piece of reality that is included in research cognitive activity is termed the object of the given discipline, if its objective character is intended, independent of the researchers existence. This is known as the ontological approach. If however the intended meaning is the role of this piece of reality as the source of knowledge and information, thus a gnoseological (the Russian term for epistemological) approach, then this piece is termed the subject matter of this discipline (Zotov 1973, 1829). The role of fragments of reality as sources of information is determined not only by their own properties but also by the tasks and possibilities of the cognising conscience. So on the one hand the subject matter is narrower than the object because not all the facets of the object can enter into the tasks of a given branch. On the other hand the subject matter is broader than the object for it includes the tasks of cognitions in it as well, the aspect of the consideration. It was formerly conventional to identify the object and subject matter. As is still ordinary in the West, these terms were synonyms, and they were never envisaged with the tasks of cognition taken into consideration (or just the reverse, by only considering the impact of the subject of cognition). Now it is sometimes suggested to reduce the concept of the object to immediate objects of study, i.e. to sources (Grushin 1961, 12f, 74; Rakitov 1969, 174), and the concept of

Metaarchaeology
particular discipline for its separation, for the isolation of its own subject matter, is determined rstly by the specicity of initial materials (sources), demanding a special methodical apparatus (Sommerwill 1960, 155190; Kelle and Kovalzon 1966, 52). The distinctiveness of a discipline is also given by the homogeneity and coherence of the entire chain of methods, techniques and procedures leading from these initial materials to the aims of cognition (Frolov 1968, 152 179). Thirdly it is given by the systemic tie or unity of those entities which are the aims of cognition (Sokolov 1972, 7172; Zotov 1973, 29f). The subject matter of a discipline is broader, and by comparison its knowledge is the more complete part. Therefore optimally separated subject matter of a particular discipline is reached when all the three conditions are met above. If some of them cannot be realised at an outlined version of the subject matter separation, it is evident that the subject matter is separated unreasonably, and one must change its borders by dividing up some sources, and leave further advancement of the deeper aims to another discipline. The nal aims of cognition of each large piece of reality, those characterising the object, are not necessarily immediate aims of all disciplines studying this piece of reality, this object. The modern process of the research cognition is deeply divided (Ovchinnikov 1968, 26, 27, 29 on the justication of dismembering of the subject matter).

13

3. THREE TRADITIONS There is no agreement among archaeologists as to the question of the subject matter of archaeology. The only point on which all concur is in the determination of immediate objects of archaeological studies. These are material antiquities regarded as sources for knowledge of life and culture of past peoples and their societies. Perhaps the specicity of these sources is acknowledged as sufciently evident. Obtaining, registering and ordering of these materials, even if only briey, is something as a matter-of-course for any understanding of the tasks of the discipline (Spaulding 1960, 439). Views on this are divergent, not only now but also in the past. From the previous development of archaeology, metaarchaeology also inherited the discussion on the determination of the subject matter

of archaeology, in which certain positions were reached long ago. (1) One line of continuation stretches from Renaissance collectors or antiquaries who gradually became archaeologists. Ancient artefacts, or antiques, were initially valued merely as extremely curious. Later, as unique items were selected they became types of monuments, and nally, since the 19th century they have been regarded as sources. Thus mass material has also acquired archaeological value. It is the preparation of such materials that the tasks of archaeology are limited to according to this tradition. The cognition itself can be allotted to other disciplines, rst of all history, to which archaeology surrenders the prepared material. It appears as a source-studying discipline, subject matter may be reduced to sources and records (Bulle 1913; Niemeyer 1968; Rouse 1972). This conception is expressed by the motto Archaeology is the handmaid of history, or as the Americans would express it, the handmaid of anthropology. Sometimes in archaeological materials, especially in recent decades, a self-depending interest is considered, an interest connected with the cognition of material culture as a part of culture in general (Deetz 1967; Clarke 1968; Dunnell 1971, 116). But this cognition is understood in a cultural-anthropological spirit, as a notion that can reveal general laws. In opposition to this Paul Courbin effusively defends the old position, claiming that the establishment of facts is the archaeologists proper role and mission, the thing that distinguishes him from all the para-archaeologists because he is capable of doing this work, and he is the only one capable of doing it correctly ... (1982; 1988). (2) The other line of continuation begins from historians, philologists and anthropologists who assembled tangible remains of the past and became archaeologists. Classicists were the rst in line, attracted by the possibility of illustrating the subjects of ancient authors with artefacts and monuments, to yield new ethnographic texts and later, the chance to put together the history of visual arts. Then scholars studying illiterate peoples and epochs became involved too. Considering survival in the culture of modern civilised peoples and comparing survival with the culture of contemporary uncivilised peoples, they established the existence of the

14

Acta Archaeologica
even, some archaeologists infer two different professions (Rouse 1968, 12; Moberg 1969; Board 1982). An archaeologist is and must be an all-purpose specialist according to this opinion (whether he can be is another question, not put forward by these archaeologists). As to the direct studies of material antiquities (description, classication etc.), they sometimes separate it into a special branch of archaeology Ja. A. Shers and Ju. N. Zaharuks archaeological source-study, or C.-A. Mobergs archaeography (Sher 1966; Zaharuk 1970c; Moberg 1969, 42). The empiricist notion of history as the simple chronological exposition of facts that speak for themselves (Ranke, Langlois and Segnobos, et al.), naturally reduced the whole scientic procedure to preparing the sources and, consequently, history reduced to source-studies. With such an approach archaeology had the same right to be termed history as the analogous preparation of written sources. To consider it part of history or a separate parallel discipline would fall into complete dependence on the appraisal of the specicity of records and not least on the ambitions of archaeologists. The neo-Kantian expulsion of laws from history and neo-Positivist eviction of causal links lead to the same consequences. Some historians however admitted that the main business of history is the revealing and judgement of hidden connections between events and the search for laws in causal links (Bernheim). Quite different consequences for archaeology issued from this understanding of history. According to such understanding the main business of a historian, after source-studying processing of materials, corporeal or written artefacts, is to begin with the historical synthesis. This is quite a complicated business. Hence the choice. If we acknowledge that the preparation of sources is comparably simple and similar in character to the procedure of synthesis, then it is possible to equate source-studying with synthesis, thus regarding them both as one discipline called history. The processing of written sources was frequently treated in precisely this manner because both at the descriptive and synthesising stages the scholar deals with sentences written on paper. Archaeology was less often viewed this way because the procedures of processing and of synthesis are evidently dissimilar. In the case of archaeology

illiterate, prehistoric period in the past of civilised peoples. Having established this period they placed their hopes in archaeological remains as the main sources of studying this period. For the study of preliterate periods these sources were considered nearly the only ones, at least as important as written sources for the study of the next period, and in addition much more objective than them (Bernheim 1889) and more simple. Hence, for these scholars prehistory practically coincided with primordial (prehistoric) archaeology (Mortillet 1883; Dechelette 1908; Daniel 1950; Narr 1957), and the separation of a special sourcestudying discipline was thought unnecessary. This understanding is well expressed in the title of LeroiGourhans (1961) article on archaeology, History without Texts. Or in the same vein as Arcikhovskiy said, archaeology is history armed with a spade. In contrast is the archaeology of the literate times of the classical, medieval and other periods, which is treated as an auxiliary discipline that provides illustrative material for history and is regarded as a supplement to history. Alternatively it is identied with the ancient history of visual arts, as an individual branch of history (Conze 1869; Furtwngler 1908: 231; Rumpf 1953; Narr 1957; Deetz 1967), or else it is attached to prehistoric archaeology as a branch of history, homogenous with it and only conditionally separated. In this case they all are fused with history due to sharing the same subject matter with it, namely the historical process. The subject matter is reduced to the nal aims of cognition. Correspondingly, Randsborg (1997: 189) distinguishes traditional history based on the written texts and history based on past material reality, or, archaeology. (3) Lastly the third line of subject matter. In the extent of one discipline it is supposed to have all the stages of cognition from preparation to the historical interpretation predetermined by it. The subject matter is considered broadly, and two levels are distinguished within it, source-studying and interpretation (Jacks handbook after Dikshit 1960; Sankalia 1965). An intermediate level is also possible which operates as a history of culture (Sophus Mller 1898; Crawford 1960). Sometimes this intermediate level is declared as nal for archaeology (Brndsted 1938). From this double notion on the division of subject matter in two branches of archaeology, two disciplines

Metaarchaeology
one can of course call this merged discipline archaeology, but in fact a discipline parallel to history is actually intended, i.e. another history, or a history without texts as it were. However, with the development of source-studying, the drive becomes stronger to admit that the processing of sources is quite a complicated business and strikingly different to the procedure of synthesis. If so, the processing of sources, both the written and especially the archaeological sources, must be separated into special disciplines. Here, only the role of the source-studying discipline, the handmaiden of history, remains as a part of archaeology. Certainly, many archaeologists will not be able to resist putting forward historical interpretations and conrming them with the facts they have gathered (or can gather); some of them ... will formulate behavioural laws, cultural laws. But where they were, in a way, irreplaceable during the establishment of facts, by this stage anthropologists and historians will be able advantageously to take their place ... Archaeologists can become historians, epigraphists, or anthropologists: but as happens every time one leaves ones speciality, the result will perhaps or probably not be as good as if the work were done by a historian or an anthropologist. It would be no doubt better to use a specialist. Otherwise specialisation has no value, and absolutely anything can do absolutely anything .... And perhaps some archaeologists will be earnest historians ... The important thing is that he be aware that, in doing this, he is acting not as an archaeologist but as an anthropologist or a historian. He is no longer doing archaeology, but something else (Courbin 1988, 154, 151).

15

4. CRITICISM OF CONCEPTIONS Not all of this is a scholastic debate. The discussion is connected with philosophical and social positions and they have a direct consequence for the practice of research activity and teaching. The acceptance of the rst, narrowly objectoriented, point of view on the subject matter of archaeology can lead and frequently leads to the break away of archaeology from history and sociology. Archaeologists close themselves within studies of all that is antiquarian, lose the aim of the studies and, even if

studying things intensively, fail to obtain from them the information that is necessary for historical and sociological reconstructions. V. I. Ravdonikas (1930) accused the Soviet archaeologists of an older generation of this and W. Taylor (1948, 4594) showed how it occurred with the domestic (in this case American) Taxonomists. P. Martin (1968) demonstrated this with an example of Conjunctivists, the followers of Taylor. Concurrently historians and sociologists either do not use archaeological sources at all or undertake sheer dilettante attacks of little signicance on archaeology. Historians have no special methods of interpretation of archaeological material (Hachmann 1970, 11). The establishment of the second, narrowly aimoriented, point of view which dominated several decades of Soviet archaeology is also associated with undesirable consequences. The specicity of archaeological records and their fundamental incompleteness are forgotten, leading to a tendency to solve broad historical and sociological problems (for instance those of ethnogenesis) exclusively with archaeological material alone, and with methods of archaeology, while ignoring the important, sometimes crucial role of other sources, such as linguistic or ethnographic sources (cf. thereabouts Klejn 1969, 21 35). In an attempt to compensate for their absence in areas of sociology and history, archaeologists make crude criticism of the study of ethnonims, toponims, folklore etc. (thereabouts Propp 1962, 8791; Nikonov 1965, 45; Filin 1972, 25; Klejn 1991b; 1998). It is as if archaeology duplicates history, but this other history appears biased and weak. In lecture courses not an as-if but what manifests itself is the real duplication of history of primordial society by archaeology, partly of ancient and medieval history. In order to avoid duplication, it appears necessary to schematise history of primordial society and to illustrate it with ethnographic examples, while the basic aims of archaeology remain to make a survey of sites and cultures chronologically, as found in standard texts on archaeology. Also signicant is the changing titles given to courses on archaeology. It was initially termed Introduction to Archaeology, which was later to be changed to Basics of Archaeology, and eventually ended being called Basics of Archaeology of the USSR. Although the meaning of the word archaeology in the designation of course titles

16

Acta Archaeologica
this separation does not demand the acceptance of archaeology as the handmaiden of history or necessarily the full break of archaeology from history. Neither does it entail the loss of history and sociology in archaeology. Inclusion of truncated archaeology into a deep staged process of cognition is denitely possible. In a number of contemporary works an idea is held that archaeology creates a new historical source from its materials, a system of antiquities which becomes an important object of palaeohistory (prehistory and early history) and has the form of a written source (Labuda 1957, 22; Grigorev 1973: 41; Bochkarev 1973, 59, abs.11; Lebedev 1973, 57; Gening 1975, 11). Feedback from archaeology to history is stressed too at the selection of attributes and other operations (Sher 1970, g. 1; Bochkarev 1973, 58). Archaeology separates to a certain extent from history, like in the rst conception, but does not break entirely from history. For this a considerable modication of the rst point of view is necessary. Suggestions of some contemporary researchers are directed on such modication. These researchers introduce into the subject matter of archaeology not only things with their simplest order but also the system of deeper connections and relations into which these things are included (Grigorev 1973, 42). Yet these connections and relations have many levels of complexity, and ultimately everything that proponents like Binford of the third point of view want to introduce into the subject matter of archaeology, such as laws of historical process, can be represented as the last and deepest level of complexity. Then we return to the middle and broad treatment of the subject matter of archaeology, if not to the third, narrowly-aimed one. The point is however to determine the reasonable limit on which the penetration of archaeology as a special science into the depth of these connections and relations must reach. Such a shift will x the fourth understanding of the subject matter of archaeology, also a middle road between the rst and the third one, though rather narrow in character.

changed somewhat unnoticed, it actually began as an archaeology that designated a discipline, but which by the end had undergone a shift in meaning to an archaeology that was the sum total of material antiquities, chronologically and regionally ordered and supplied with historical comments. A demand of publications emerges to include not only processed materials of excavations but obligatory and well-founded historical and sociological inferences too (Artamonov 1959: 58; Houbowicz 1961). This greatly hinders the publication of the nds and sometimes leads to publication of inferences instead of materials. The compromising third point of view, where a broad subject matter is conceived, does not win great acceptance. As yet its advocates have not succeeded in showing that the tasks of studying both parts of such broad subject matter as material antiquities and the life of ancient societies are possible by means of one and the same discipline, without including the sources with other specicity. There is no hope that it is possible to unite the properties of both parts of the subject matter, to show their common features, while still achieving coherence and continuity in the chain of methods. In practice this compromise point of view appears reduced to one of the above extremes, more often to the third one, with which it has in common an imposition onto archaeology of the functions of history and sociology and their social and political role, something that many archaeologists long for. Another point relevant to these conceptions of archaeologys subject matter is the separation between prehistory and archaeology. From a narrowly object-oriented viewpoint on the subject-matter of archaeology, prehistory is a separate discipline, while the other viewpoints see a fusion between archaeology and prehistory.

5. DETERMINING THE CORE OF THE PROBLEM With the separation of archaeology from prehistory, history and sociology now nd more and more support in archaeology (Grigorev 1973, 4143; Bochkarev 1973, 5660; Klejn 1977d; 1978b; 1991c; 1993; Tabaczynski & Pleszczynska 1974, 1013; Hensel, Donato and Tabaczynski 1986). Despite current prejudice (Clarke 1968, 1213; Masson 1973, 4649),

6. A PROPOSAL The cognition in the frames of a single discipline must continue until the point where the systemic connec-

Metaarchaeology
tion between its materials is discovered and the totality of facts appear in the system of functional links. Yet neither the sum of antiquities nor the restored material culture constitute an integer system. They are blocks of such a system that are isolated, unclosed, separated on the basis of second-rate attributes of material and safety. Such a system would be culture, which leads archaeologists to take the study of material antiquities to the level on which these antiquities are included into the system of culture recorded in statics and dynamics. Meanwhile culture as a whole cannot become the subject matter of archaeology because an adequate study of it in statics and dynamics demands a huge involvement and application of other methods. Only one aspect of culture enters the subject matter of archaeology as its nal unit, and this aspect includes material antiquities in the system of functional links within culture. The basis of this inclusion consists of three processes: (1) Continuous development of culture with integration, disintegration, convergence and interaction of local cultures (Artanovskij 1967, 2650; 84207; Sokolov 1972, 5891); (2) objectication, i.e. conversion of ideas and events into material things (Deetz 1967, 105134); (3) archaeologisation, i.e. egress out of life, transformation of results of objectication into artefacts and monuments (Eggers 1959, 199297; Bochkarev 1973, 59f). This process contains its own causal mechanisms and regularities, and they of course crop up in the subject matter of archaeology. If these considerations are correct, then there are three levels of depth in the subject matter of archaeology: 1) things and their relationships, 2) phenomena, events of the past, and 3) their inclusion into the system of culture and into the cultural-historical process. This allows a category composition of the subject matter of archaeology to be determined as follows: (1) The material antiquities per se as remains of the defunct material culture of the past. (2) Their links and relationships in the system of culture with each other, with other elements of cul-

17

ture and with particular chains of the culturalhistorical process. (3) Regularities and causal mechanisms in the basis of all these links and relationships. This means in particular that the concerns of archaeology are not limited, as V. S. Bochkarev (1973, 5960) supposes, with the separation of archaeological cultures. The formation of track sequentions is still involved entirely in the competence of archaeology. These sequentions look like genetic lines but are only cultural traditions. It is in these catenae or chains of cultures that the historical process has owed (Klejn 1973, 14; 1974, 4748). This means that the so-called historical reconstructions of phenomena (authochtoneity, migrations, inuences, transformations etc.) are the business of an archaeologist, not of a historian. But the causal explanation of these events and revealing of laws governing them is in the main not the business of an archaeologist. These problems are of particular concern to an archaeologist but only from one side. He returns to their elaboration only as soon as it allows him to establish the systemic link between his materials, to form sequences. In the rest, despite the directives of L. Binford (1967, 267275), these problems retreat to the competence of historians and sociologists. It is here the border lies. Under such an understanding, the subject matter of archaeology does not appear to be the work of history, a part or duplicate of history, or an absolutely independent discipline. It emerges as a methodologically independent and a particular social-historical discipline. Social-historical in the same sense as can be said for ethnography, linguistics, literary criticism etc. It has its own subject matter with specic initial materials, its methods, and its theories. It can advance in the role of a source-studying discipline in relation to other disciplines, as do ethnography, linguistics, and philology. Similar ways out of archaeology lead into sociology, history, culturology, history of culture disciplines synthetic by nature. Ambitions of status and overt, though well meant, pretension of alien roles do not support its authority. Archaeology is a discipline of a narrower but sufciently important purpose and to ensure a fruitful integration of it with other disciplines one must rst delimit archaeology,

18

Acta Archaeologica
cessing, then there is a basis for two disciplines: archaeology and source-studying of texts. To provide synthesis this should naturally be carried out in one discipline. But if synthesis is necessary then it is a question of which discipline to tackle the problem with. Archaeology should be allowed to attain what it reasonably can from the materials accessible to it, carried out to the extent that the methods it has developed will allow. This would be far more benecial both to archaeology and to history. Let us render what is Caesars to the Caesar of archaeology and what is Gods to the God of history.

clearly determining its functions and localising its subject matter. Reluctance to take into account the borders of disciplines and a trend towards archaeology losing its subject matter has a long tradition in the Soviet Union. Besides the methodological substantiation critically considered here, this tradition creates positive ideational stimuli too: a trend towards the integration of disciplines, overcoming the worship of facts, and the rise of the authority of history. However if the specicity of different types of sources demands differences in methods of pro-

Metaarchaeology

19

3. Archaeological sources
1. SOURCES IN THE COGNITION OF THE PAST The term archaeological sources (arkheologicheskie istochniki) is widely used in Soviet/Russian archaeology and in the archaeology of formerly Socialist countries; it appears both in general handbooks and in the titles of certain works (Arkheologija ab 1961; Houbowicz 1961; Kameneckij et al. 1975; Klejn 1978b). Due to the extensive use of the term it conveys the impression that it is natural and commonly accepted. Additionally that it is understandable and always used in this matter-of-course meaning. Yet this is not the case. Even skimming through the main pre-Revolution Russian handbooks on archaeology by Uvarov, Gorodcov or Spicyn, or the Proceedings of Russian Archaeological Congresses, is enough to be convinced that they are concerned with archaeological monuments, sites or materials, but they were not usually called sources. In modern Western usage the terms corresponding to our Russian archaeological sources are in general known. They are present in archaeological vocabulary in German as archaol ogische Quellen and in French as sources archeolo giques. Yet the full correspondence only occurs in continental Europe. In English the similar term material sources is very seldom used. In its place we meet the term archaeological records. The word record carries the meaning of a xed message, registration or entry. While the word source implies the possibility of using information, the word record is abstracted from this aspect and stresses only the imposed nature of the information onto the object. In general, even if these terms are used in foreign literature, they occur not in an archaeological parlance but in general theoretical considerations in philosophy and methodology of history. In the archaeological literature they are hardly ever encountered (some exceptions will be considered later). In contexts where they could be expected, other terms appear: archaeological evidence, archaeological data, archaeological monuments, archaeological nds (German Bodendenkmaler, Bodenfunde), ar chaeological documents (German archaologische Urkunden) etc. In order to understand what archaeological sources are and to learn how to deal with them correctly, one needs rst and foremost to determine their place among other sources of information, to reveal the specicity of archaeological sources and clear up their interrelationship with the concept historical sources. This means complementing the historical approach with the structural approach and considering the concept archaeological sources in the established system of concepts characterising and serving the research cognition. Only in the system, in a net of connections and relations can each concept receive adequate completeness of determination, and only from this can a clear-cut notion on its functions and on the limits of its application emerge. In each theory, the scholar who wants to work strictly and methodically will usually meet with harmonious and rigid systems, systems of main concepts necessary for the given theory. Yet such systems emerging in different theoretical trends are often incompatible. The most general concepts of a discipline are called to serve and connect its different theories and even different disciplines. In real functioning such concepts gradually accommodate themselves to each other and spontaneously constitute one system, although it does not always appear complete and clearcut at once. The task emerges to order and specify this system beginning from its most general parts. Every positive discipline has its own factual base. The facts constitute this base, the initial data, and the concrete evidence on objective reality, which this discipline elaborates on with its methods and enlightens with its theories. Each of these disciplines isolates, carefully seeks and systematises those objects and processes in which data necessary for it are contained. These are its means of obtaining information and they differ in structure. As a rule any given discipline does not limit itself by one kind of means but nevertheless some are more important than others. There are disciplines on functional dependencies, behaviour etc. (biology, sociology, psychology) for which the main such means is natural observation. Other disciplines like physics and chemistry are supported by experiment the observer intervenes in the investiga-

20

Acta Archaeologica
the time of recording) customs and incidents, on wars, donations and so on. This rich information is accumulated by the conscience of bygone informants and expressed in sign systems especially predestined for transmitting the information, mainly in the natural language of human communication, the language of words. In other cases this deep information is associated with regular connection of phenomena from when some phenomena can be judged from others. Then it is recognised only by the conscience of an investigating historian whose conscience equates this regular link to a sign system. Through it, he proceeds from immediate evidences to inferred ones. In such ways this bygone reality becomes accessible to historical conscience. The reection theory accepted in Marxism claims independence and priority of the reected object as viewed against the surroundings but it does not reduce the result of the reection (the image) to the reected object, and it does not consider the image and the object identical to each other. It takes into account the active character of the cognition process, the possibility of removal of the image from the object, of the cognition from the reality. In such formulation this theory accumulates conventional general ideas of Materialistic philosophy and does not contradict sound reasoning. However the possibilities of mismatch of an image with the object remained unrened in the Marxist-Leninist theory of reection; as applied to the practice of research it usually struggled for rapprochement of the image with the object. Instead it preferred simplications and reduced everything to direct correspondence. Historical cognition can be considered as a specic kind of a reection (Ivanov 1962; 1963; 1973a; 1973b; Danilov 1963; Pushkarev 1970, 6484), and it is especially clear how complicated in reality the process of reection can be. This is an indirect reection, which as a rule is made without direct contact with the object of reection by the investigator. It is outstretched in time, with its distribution in phases and with delay (Fig. 1). First objects are reected in the conscience of observers where these images are xed and expressed in a sign system. Later the images are passed to other people while being complemented en route with new results of reection, and are xed again anew. If we isolate one such xation that im-

tive process and changes its conditions in order to trace how it can be explained and thus to grant the cognition of hidden mechanisms. There are certain disciplines, for instance linguistics and geography, which need comparatively less to organise special observation. They use mainly contemporary materials always accessible to description and act as the main object of immediate study. Then there are other disciplines whose objects accessible being analysed have by themselves secondary interest but which are important for the discipline in question primarily as mediators which transfer the information, or in other words as sources. It is from sources that one can extract evidence about the events of the past which this discipline strives to cognise about. Such disciplines deal with processes of development during the past (palaeontology, history, historical geology, historical literature etc.). In Russian the word source generally means any reservoir or opening from which something ows out of or can be derived from. In the direct sense it is a spring, the beginning of a brook. In the gurative sense it is an object from which information, evidence or knowledge can be obtained about another object, and it is only due to this property that the rst object is interesting for us in some context. Only in this context does the object appear a source. It is in this sense that newspapers use this word when they refer to diplomatic sources or to ofcial, governmental sources etc. In describing their means of obtaining facts historians characterise the specicity of their discipline, the peculiarity and difculty of the initial methodological situation in normal historical study. The facts interesting for a scholar are very frequently inaccessible to his immediate observation and effect. At the same time this statement stresses the complexity too, the cognitive depth of those objects that are accessible and liable to immediate investigation of a historian: chronicles, treaties etc. Indeed, being themselves facts of the past products of a certain time, a certain medium and certain actions, like the creative work of a chronicler or the recording of diplomatic negotiations they contain information on some other, more distant facts of the past, facts often especially important for the historian. These are facts about bygone migrations of peoples, about extinct (at

Metaarchaeology

21

Fig. 1. Historical cognition: reection in written sources.

printed not only old images but also the condition of their selection and xation, this is altogether just a source. After a while, usually a long interval, this xation is perceived by an investigator-historian and xed again in his conscience and in a sign system. The reection appears multi-step like in a periscope, but the mirrors of this periscope are displaced in time and they are disconnected. At the moment when the investigator has perceived the images in their nal form, the initial objects and many intermediate forms are already gone. When the ray reaches the eyeglass, many mirrors that it had passed are gone, let alone the very objects. Checking is extremely difcult but this is the nature of historical cognition. Sources are an important chain in the machinery of reection characteristic for historical cognition. It is the sources that ensure the agency between the perceived reality of the past and the subject of cognition, the historian-investigator. They contain the connections between times and the information contained in

them is realised in images (Gulyga 1965). An image contains some general features expressible in abstractions. But it unites them with individual features, peculiar, and inherent only in the given piece of reality. Like in any reection, by and large images could not emerge in sources if there were no objects (prototypes) ready to be reected, or if there were no systems able to provide the reection. In early stages of the development of Positivist historical knowledge the reality of objects was viewed as self-evident and the validity of images did not raise doubts. The reecting system seemed simple, obedient and easily controllable. Later, there was an uprising of scepticism and frustration in historical studies. Real objects may remain hidden behind this array of images but not necessarily behind each one in particular. Historians recognised the activity and complexity of the reecting system and became terried at the size of its contribution to the formation of images. It had lost not only the credulous nature of the

22

Acta Archaeologica
tial historical information) during the process of its consumption. It does not ow by itself from the source it must be extracted, by squeezing the source like a vessel, by turning it inside out. Its quantity and quality depend not only on the objects of reection but also on by whom and how this information is extracted and consumed. In that milk or cream form in which it is often imaged, it exists neither in archaeological sources nor in any historical sources in their initial state. In that sense historical and archaeological sources are similar. Thus if not all sources that inform about the past are historical sources, they nevertheless all deserve careful checking for aptitude in this respect. Any of them can still appear as such in the future and all need an attentive attitude. It has been a point of discussion among Russian theoreticians of source-studying to consider whether it would be worthwhile to include a special category, a potential historical source or pre-source category, into the concept apparatus of source studies. There are different methods by which to group historical sources epochs, territories, themes etc. Most customary is division by way of reection, in texts, corporeal (material), or lingual etc. L. N. Pushkarev (1975) suggested the most exact denition of this criterion, in which classication was made by the method of coding and keeping the information in the source. From the viewpoint of the information approach now widely applied in many disciplines, information contained in any objects can be conventionally considered as messages made up in some sign systems or in certain languages, i.e. enciphered in a certain clue. In order to read these messages and extract information one must know this clue or code. But it appears differently in different kinds of sources as soon as information is extracted from them in various ways. Strictly speaking by this criterion we distinguish four groups among historical sources as well as in general among sources of information on the past: 1) lingual or speech sources (written and oral), 2) behavioural (customs, rituals, games etc.), 3) material, and 4) gurative. If one keeps to conventional use of words it is clear that the path to the denition of archaeological sources goes via the concept material (corporeal) sources.

images, but the credit for the reality of their prototypes. What a historian sees in a source depends on his preparation, education and talent, on the level and activity of his conscience. The more he knows, the more he will see (Collingwood 1956, 237). Indeed, the reecting system of historical cognition deserves some attention. Included in such a system are observers, informants, creators of sources and the sources themselves as well as source-studying specialists and the users of historical information. However, as the mental reection is not mechanical, not mirrorlike, and not direct and not repeatable in history, and as images are not identical to the objects in principle, there is no sense in attempting to look behind each image for an object imprinted exactly in it. Images can be more or less near to real prototypes and can be absolutely distorted and complete fantasy. Of course there is the possibility of checking if and how the images correlate with the objects (prototypes), but this checking in the historical cognition is very difcult due to its specicity. In principle there are two main possibilities of making such a proof. Either to collate the obtained images with what is known from other sources about the same objects, or to monitor the most vulnerable moment of reection, risking injury and frequently leading to the appearance of distorted images, i.e. to control the creation of the source. Data for such a control are partly contained in the source itself and partly in other sources enlightening this source as a fact of the past. Thus, there are at least two kinds of evidence in each source: a) images of certain objects in relation to which this source serves only as the means of reection, and b) images-facts of which a part is the source itself as an event of the life. The point is reached of asking whether there is historical information contained in the sources. The answer is both yes and no. One should not imagine that historical information is contained in a source like milk in a jug. If the jug is tipped the milk will spill and as much as was originally poured into the jug will run out; its state inside will be the same outside the jug. And it retains one and the same calorie content whoever drinks it. It is not so with historical information. It does not exist beforehand in the source, but neither is the information brought in from outside. Rather, it grows from information on the sources on the past (poten-

Metaarchaeology
2. REFLECTION IN MATERIAL SOURCES In the direct sense of the words archaeological sources are those sources which are relevant to the concerns of a special discipline called archaeology. This means that these are sources deserving separation into a special category by considerations of a methodical character. In order to extract information from them a special professional preparation and training is needed, special methods, a special set of concepts, in a word a special discipline. Intuitive understanding of such a necessity emerged long ago, and a discipline was spontaneously formed; archaeology was established. Initially, since classical times until the 18th century the term archaeology in full correspondence with its literal sense (from the Greek a|caioV, ancient and logoV, a word, learning, or knowledge) designated a very broad branch of knowledge, that of any true evidence about the ancient past. Since the 19th century the meaning of the term narrowed to the study of material monuments of antiquity. This was probably necessary as this subject matter demanded specialisation. But because this occurred spontaneously and was not substantiated theoretically, these naturally formed borders and functions of such a formed discipline were often damaged. Moreover, suggestions were frequently advanced to change its borders to shift them or to broaden them, or to annihilate such a discipline completely. This can be illustrated by several examples. In the rst textbook on archaeology published under the Soviet power it was stated that both material and written sources on the ancient belong to the competence of archaeology. What, then, distinguishes it from history? An inclination of interest to the material side of things and chronological depth. The author of the textbook, the prominent historian S. A. Zhebelev (1923, 4, 27, 130), apparently did not acknowledge in material sources any specicity important for the methods of research. He held that any special archaeological method is hardly necessary to invent, similarly there is no reason to speak of some special attitude in studying archaeology. In 1932 in the introduction to a most authoritative collaborative work, Handbuch der Archaologie, the inuential German archaeologist E. Buschor (1969, 3) stated that there is not any special archaeological method, and that methodical tricks are the individ-

23

ual business of each investigator. In his opinion an archaeologist is anyone who attributes a thing formed by a mans hand to the past epoch, while archaeological sources are simply a mirror in which an archaeologist sees historical life (my emphasis L. K.). Objectivity of a researcher depends entirely on the investigators acumen and power of observation. Meanwhile, to clarify the idea of archaeological sources as a means of reection, they can be compared to an optical device, in which they are likened not to a mirror but to a periscope-like dive into the past. This would not be a simple periscope like the Minos eye (a simple periscope lowered by Lerici into ancient burial chambers of the Etruscans), but a much more complicated periscope consisting of a system of mirrors: now straight, now concave, now convex, now bizarrely distorted, and different ltres rotating, opaque, coloured, etc. It would be foolish to ignore this device on the pretext that it is not the device itself that is interesting for us but the object being observed through it. Without meticulous studying of the device we cannot present, even mentally, the object in its real form, and cannot understand which real object is hidden behind that incomplete, split and often senseless image which is presented to our eye in the ocular.

3. WORDS AND THINGS Things as sources of information differ radically from other sources of information by the way in which information is encoded in them and consequently by the means of extracting it (Fig. 2). Why however is a professional historian of traditional prole unable to extract this information? Any discipline functions only in the sphere of thought and speech. History is no exception. As distinct from some other disciplines history is formulated by means of everyday literary language. This means that it operates with information expressed in concepts, judgements and deductions exposed in natural language with the help of some minimum of special terminology. It is also how information is organised in written sources in chronicles, memoirs, treaties etc. This is why it is difcult to make sharp borders between original sources (annals, memoirs), secondary sources (compiled chronicles, chronicle convolute)

24

Acta Archaeologica
the investigation, or had an ashtray been knocked and the butt fallen out before the event? The list of possible correspondences between the things and events is not limitless it is quite apparent the cigarette was not smoked by a horse belonging to the criminal or by the investigators dog but nevertheless the list is very large, and it is difcult to grant its completeness. Here one cannot rely on the simple expansion of the professional training of a historian into the new eld, one cannot do with the usual notions, life experience and sound reasoning. Absolutely new knowledge is necessary, special methods and another discipline is needed. For processing material evidence used in law a special discipline is created, criminology. Even more necessary is a special discipline for processing ancient material sources because it is even more difcult to elaborate on them.

and academic works of historians. The elaboration of pre-source for its transformation into a source does not differ in principle from the main labour of a historian from establishment of causal connections, from trying to grasp interaction of laws and chances. In both elds in written source-studies and in history a scholar confronts common concepts, throws together and transforms everyday judgements in life according to strict rules of logic. So written source-studying is often worked out completely in a professional way by the same historian who thereafter uses the information obtained for historical research. With material sources matters stand otherwise. It is impossible to operate immediately with material things in the sphere of thought and speech. Without preliminary remaking, information xed in things is not suitable for scholarly use. One must re-code it and do it twice. First one must translate it from what is metaphorically called the language of things to any of the natural languages; to describe things and their relations. At this stage one has to use special terminology much more than in history. Then it is necessary, by comparing acquired evidence with others, to establish what events and processes of the past have been reected in the material sources described. Only then does the information obtained reach the form demanded by the historical discipline. It was once widely supposed that this was a very simple task. Now however it is clear that to realise the demanded re-coding of information of material things is much more difcult than to translate a text from one natural language to another, say from Russian to German. Despite the polysemy of words and pliability of their use, it is practically established that there is plenty of very narrow correspondences between natural languages, and the context will prompt which of them to choose. Material things are even more polisemic, or multiple in meaning. Full lists of all their possible correspondences are not available, and their context is often distorted. The researcher takes the classical position of a detective criminologist who notices a cigarette stub at the scene of a crime. Has the stub been left by the criminal, victim, witness or a chance passer-by who saw nothing? Was the cigarette smoked during the moment and at the spot of the event of interest for

4. ARCHAEOLOGICAL REFLECTION For many a long day scholars observed only one essential distinction in the earliest of ancient material sources. The oldest sources available are sparse, monotonous, and poorly preserved. From the Palaeolithic period comes a plenitude of int tools and fragments of bones, but the rest is badly preserved. In the Neolithic ceramics appear in great abundance, and from later epochs metallic ornaments, coins, and foundations of buildings are found, albeit only very seldom with things like wooden utensils and leather shoes. Yet from the more recent past furs, laces, textiles, paper etc. can still be found in a preserved state. Ancient material sources are fragmentary and incomplete, but these are purely quantitative differences. H. J. Eggers, the German archaeologist, was the rst to give heed in 1950 to the qualitative differences between the living material culture (eplebendes Gut) studied by ethnographers, or observed by ancient informants, and the dead material culture (totes Gut), which falls into the hands of archaeologists (Eggers 1950). In addition to this qualitative difference, archaeological sources are one-sided, caused by the uneven preservation of different assemblages and the gaps within them that cannot be compensated for. Indeed, more often than not, things that found their way onto the scrap-heap and into the garbage sediments of

Metaarchaeology

25

Fig. 2. Reection in archaeological record.

26

Acta Archaeologica

dwellings were items that had broken or were not often needed, especially if they were simple to produce. Metal items would be used over and over again for a long time. Fragments that had broken and been ground off during production were not thrown away but were re-melted for further production. Pots however do occasionally appear in rubbish heaps. Archaeological sources can be intentional just like written sources. The dead were not often laid into graves wearing the clothes and accompanied with the things which they used for daily use (living Christians do not strut about in shrouds). For the next or other world the dead were supplied with equipment especially selected according to conventional ritual, or otherwise especially made for this purpose. In some female burials of the Bronze Age in Central Europe or Caucuses the sets of ornaments weigh so much that it would be impossible to adorn them, it was possible only to lie with them on. Golden ornaments of Scythians and Sarmatians are not always cast, frequently they are only covered by a very thin gold leaf bearing

convex picture imprints. To wear such ornaments was risky for they would crumple at once. They can be placed into a grave however and the appearance is magnicent, even though the total use of precious metal is small. In a burial of a leader of the Scythian time in Altay (Pazyryk barrows), as Professor Grjaznov has established, the wooden details of the bridle, different for each one of the nine horses, were produced by different artisans. Tribesmen probably donated the horses as a burial sacrice, but it is doubtful that the leader used such bridle when living. Thus upon the transformation of the living culture into a dead one its constitution was rebuilt and the rates of its components were changed. In the same work, but especially following it, Eggers added considerations on subsequent changes occurring to culture of the dead after its deposition, that is, after its resting in the earth, on the earth or underwater. In essence he has brought the reader to the inference, without formulating it as such, that however a dead culture differs from a living culture,

Metaarchaeology
the third state of culture differs even more; the remains of a culture dead long ago the state in which it properly speaking passes to the business of archaeologists (Eggers 1959: 262270). The term for the changes leading to this state were later suggested by David Clarke as post-depositional (Clarke 1973). Changes affecting the culture of the dead act selectively upon it. Different materials react differently to long-term effect of chemical substances. Metals are altered by corrosion, wood decays and falls to pieces, bones can be better preserved, and stone and ceramics resist well against time. This exaggerates the disproportion between metals and ceramics in cultural remains. Archaeologists nd ruins of settlements almost devoid of metal but this does not mean that it was absent from daily life. Yet these settlements are over-saturated with ceramics there are literally mountains of ceramics, but this does not necessarily mean that when these settlements existed their houses were full of all of these wares. Not only the composition of the dead cultural material changes but its layout and structure too. A river can undermine a citadel and dislocate its remains, and a coastal dune will be re-scattered by wind causing sediments of different epochs to be sprinkled into the sand and mix together into one deposit. More often still sharp changes are introduced by the activity of people, such as re-digging of old deposits in a hill fort with storage-pits, wells, and ditches for dwellings etc., levelling of ground for fortications, clearing of ruins, secondary use of building materials, robbing of rich graves and so on.

27

5. MATERIAL ANTIQUITIES When comparing antiquities with other material sources Ju. N. Zakharuk noted their differences in terms of diagnostic possibilities. Diagnostibility of antiquities, i.e. the possibility to recognise and understand them, sharply declines as compared to sources of later origin. Younger sources are often more or less known to the investigator or to his contemporaries the things themselves or their apparent analogues, or closely kindred things. Their functions are experienced, their connections are congnised and their place in the system of culture is made known. These may be rather old things too, circulating for a long

time and may even have become defunct long ago. Yet if such information about them is saved in the memory of living people these are not antiquities. The category of antiquities contains only those whose information is lost or severely distorted. Russian terminology here is richer and more accurate than the Western equivalent. In Western languages only one border among monuments of culture is distinguished. It distinguishes modern or in general younger objects from those that came from the depth of a matter of centuries and are called antiquities (in Russian drevnosti). This simplication conceals the distinction between approaches of archaeologists and ethnographers to the material remains of the past. It would be simpler to use terminology with one border, i.e. the division into two, if antiquities were the concern of archaeology and modern things were the concern of ethnography. Yet ethnography studies not only things of modern life! Eggers felt this lack of a concept and introduced an additional term into his terminology. Between living material culture and dead material culture he placed one more term dying material culture (sterbendes Gut), and in this way designated the defunct, obsolete, out-of-date part of living material culture. The border between the completely living material culture and faintly living, or dying, runs in different parts of material culture at different levels. Everyday clothes are completely replaced on a widespread basis in perhaps ve year cycles, but furniture is only discarded with every generation or two, and table silver and jewellery remain in a family for a number of generations. Sometimes, however, things of an age greater than this limit are kept in use, and they are perceived as old-fashioned. While studying ethnographic collections from Pomerania, Eggers has noted that this state does not have an age limit, and it is not equal in different parts of culture (Eggers 1959: 258261). In other words, the dying culture also has an opposite end behind which things are no longer utilised in peoples lives. They are not kept together with more new ones but fall out completely into the dead material culture and are deposited there. Designation of this stratum of dying culture, in some places millennia old, is strangely enough not a commonly accepted term in German, leaving Eggers termino-

28

Acta Archaeologica
science. Both terms in biology and in history of culture are determined not as applied to each single object but to the whole typological totality; to a species in biology, to a type in archaeology. In the churches of Kievs Pechery Lavra divine service still goes on today, but this is nevertheless an antiquity and object of archaeology, since its functioning is a rare exception. All the other monuments of this type and time were long ago left by their inhabitants or users and since then have altered. Material antiquities are things which, judging by their type, fell into disuse long ago and were forgotten, left in the earth, on the earth or underwater by the time of their discovery by an observer. The observer in this case is a scholar or a collector who extracted them from the environment for use not according to the direct original destination, but as a source of information on the past. Although there are archaeological trends in which regularities in history are accepted and which are optimistic towards reconstructions and explanations, something else restrained theoreticians from reconciling the border of antiquities and archaeology with the break of living memory. The point is that as a rule memory about the events of long ago does not vanish at once, it declines gradually. And with it explanatory possibilities must decline gradually. Gradually and smoothly the diagnostibility must weaken! If so, changes on this scale would be quantitative and smooth. This means that one can establish only conventional or sliding borders. Only recent studies in semiotics showed that this is not so. The Russian archaeologist Vera Kovalevskaja stated with a ne mathematical analysis that the sign system of archaeological material she studied, of ornaments of belt sets of Early Medieval steppe nomads, contains redundant information, i.e. more signals than it was necessary for ancient people to comprehend the sense given to a thing. This permitted them to comprehend the sense even in difcult conditions, say partial alteration, and thus strengthen the reliability and vitality of the system. As it happened, the redundant information of the sign system in archaeological material appeared quantitatively the same as earlier measurements discovered in natural language a redundancy of nearly 80%. Consequently, optimal redundancy (such that there will be sufcient

logical innovations to remain only in his personal glossary. In Russian there is another term, starina, which comes from the word staryj meaning old, and it belongs between the two extreme concepts antiquities and modern objects. Starina is everything that emerged long ago and it no longer corresponds to modern conditions, demands and standards. It is everything that belongs to defunct types still circulating in living culture which are applied or can be applied because they are still understandable to common people in function and use. Material starina are things technically or morally outdated, of obsolete or defunct types, but preserved in living circulation or at least in continuous possession of persons who know how to use them until the moment when these things come under observation or were extracted for studying or collecting. These are materials of ethnographic studies although these are not all materials of ethnography. Not everything old and decrepit is called starina. A thing modern by type can simply be old if it was produced long ago and if the type has still not become archaic, or can be neglected if it is very worn out and badly preserved. Some narrower concepts are antiquarian things and survivals. Antiquarian here are antique things having value as a commodity and designated as such from this point of view. The term survivals (perezhitki in Russian) is applied in the same sense as in Western languages, but sometimes it only covers musty ideas and stereotypes of behaviour, but not things. To the latter the term relicts is applied in this meaning. Naturally, monuments of the comparably recent past have more chances of receiving the status of starina, although it is not time that is the main criterion in question. Antiquities are remains of the culture of the past, usually of the distant past, separated from our lives not only by a signicant interval of time but also importantly by a break in tradition. It is such breaks that hinder the understanding of the remains; one should clear up the purpose of objects and search places of fragments in the system of the whole. In a certain sense antiquities take a similar place among monuments of culture as fossil species do in biology. A species is called fossil when it is now non-existent, and ceased to exist before the beginning of biological

Metaarchaeology
spare but not excessively so) appears approximately equal for the main sign systems working in social communication. This suggests that common regularities of social and psychological comprehension act within them (Kovalevskaja 1970). Issuing from language-material psychologists established that with a gradual and smooth decline of information, by gradual distortion of a text for instance, comprehension does not decline gradually, but in a jump. Initially it weakens slowly, followed by a rapid decline and complete lack of comprehension. That is, the cognisability is subjected to the dialectical law of converting a quantitative accumulation into a qualitative shift. If the distortions of the text touch no more than 30% of its words, leaving 70% intact, the text is still comprehensible. But if more is distorted, the text becomes incomprehensible. Apparently this phenomenon is not limited to languages (Moles et Valancien 1963; Frumkina 1965; Frumkina and Dobrovich 1971, 31, 36f). The discovery of the threshold of comprehension allows to archaeologists nally to establish the criterion of the separation of antiquities from starina. Assume that the memory of things declines gradually, and the diagnostibility and comprehension lower not just gradually but fall in a jerk. Here oblivion is difcult to measure and to determine the threshold quantitatively is also difcult but one thing is clear: behind some threshold of oblivion a thing becomes an antiquity. The necessity to decode emerges the need for archaeology.

29

6. SPECIFICITY OF ARCHAEOLOGICAL SOURCES Much has been written on the specicity of archaeological sources. The whole body of study came to the conclusion that there is such specicity. Nevertheless it was contested again and again. In dispute was either its existence, its detectability or its importance. Now we see that it was impossible to construct the best and deepest criteria for determining and isolating the specicity of archaeological sources without and before the development of the semiotic approach. Therefore investigators spontaneously approaching the determination of archaeological sources usually stopped at attributes which would not even lead to a

fruitful division. Either these were attributes with only seeming divisive ability for the given material such as indications about the unintentional and objective character of remains, as opposed to the intentionality and tendentiousness of written sources. Or they were weak attributes of narrow coverage, not obligatory as, e.g., underground position, which would exclude petroglyphs and medieval churches from archaeological sources. Or they were attributes that were diffusely formulated and unclear (like outlook, character, material nature, etc.). Or they were attributes such as incompleteness, indeed peculiar to archaeological sources, but separating them only in the function of historical sources, and only from other main kinds (i.e. from modern). This is a very scematic distinction, because to admit absolute completeness of any sources is impossible. Investigators did not usually attempt to motivate their choice and explain why a particular attribute demands strict isolation of a group of sources. Neither did they attempt to determine if one attribute is enough for the isolation of sources. Indeed, one attribute is not adequate. This can quite obviously be derived from the above facts and considerations. Archaeological sources belong at once to two broader groups, things (or artefacts), and antiquities. There are other material sources that are partly historical sources. For example the architecture of buildings, coins, smashed cars at the scene of an accident or a bloody knife at the scene of a crime. There are other fossil and ancient sources, such as skeletons of pangolins, traces of glaciers, chronicles, letter patents, and Russian Laws, including cultural remains (antiquities). Archaeological sources differ from other material sources by their antiquity, and from other ancient ones by their materiality. This double separation is probably important for the isolation of this group into such a category of sources which demands a special discipline. Double isolation produces extreme difculty in cognition. Turning to material antiquities the investigator nds a double break. It occurs in the traditions between the distant past and our time and in objectication, i.e. in the forms of how information is incorporated (a break between the world of things and the world of ideas able to operate in the research). This double break is the main specicity of archaeological

30

Acta Archaeologica
of text or other source. And he did not have to guess about destiny and connections, he knew them and sometimes saw them as an eyewitness, not only things but also actions and events associated with them. Thus although they hinder cognition, taken separately any one of the breaks is not so hazardous. The loss of acquaintance with the context that interests us may be compensated by the transmission of restorative information through another channel. In ethnography through the channel of live communication with the participation of the investigator, and by direct observation, and in ancient history through the channel of a written tradition. Only both breaks together in combination like in archaeology, lead to a radical isolation of the scholar from the past reality, to a shattering of the whole into pieces, sometimes to the complete loss of the sense of the information. To restore it some external supports are needed in the form of the possibility to attract additional information that would allow a bridge to be built from the known to the unknown. This may be the knowledge of the sense of other things. Among them are archaeological objects already deciphered. These are other, non-archaeological sources pertaining to the same past reality; general evidence on the structure of objects similar to those under study; knowledge of laws according to which the incorporation of ideas into things happened and still happen; transformation of things and events into archaeological traces and remains, etc. And of course strict and reliable methods of using this knowledge for recovery of the lost sense is also necessary. This is why only a double break demands isolation of sources into a special category and creates a special discipline for their study. For this reason ancient material sources those that it is customary to call archaeological constitute this category. From these considerations it follows that it is impossible to establish a strict and common chronological date for the separation of archaeological objects from non-archaeological. It follows too, that excavations are not obligatory for attributing objects and works to archaeology and are not exclusive to archaeology (there are excavations without archaeology and there is archaeology without excavations). It is clear then that some studies are attributed to archaeology due to the broad or confusing understanding of the

sources. To tackle the problem of this double break does not simply mean doubling the research effort. The double break creates a new qualitative difculty. Let us compare archaeology with disciplines dealing with only one break, such as ethnography and history. In ethnography when we take a thing and it is clear to us what type of an object it is. Perhaps a rock, a clasp, a vat, or an oven fork, things that must be noted and described. This involves information which must be translated into another form, like in archaeology. Of course it is not easy to formulate information, to translate it into ideas, into words, i.e. to isolate the essential and to discard the inessential attributes. This can be done in various ways and one can certainly go wrong. Yet in ethnography much is already at hand to aid in overcoming this difculty. The connection of the thing with the actions of people is present, the date is known and there is a context of functioning. There is an entire assemblage, the whole. In archaeology this is absent. There are only fragments in front of us, fragments no longer actively connected to life and very indenitely connected with each other. Contemplating them is fraught with risk and demands special methods of study. An example of this occurs in hill forts of so called Djakovo type in the Moscow region and its neighbouring districts of two thousand years ago, where small earthen disks with a central hole are always found. Some archaeologists believe they are spindle-whorls, while others think they are loom-plummets, burners from oil lamps, weights for measuring grain, or objects of a sun cult. They were rst discovered in the 19th century and are still debated today. In the history of ancient written sources the opposite picture is found. Here like in archaeology the break in tradition is also present. Much is forgotten and is incomprehensible to the modern investigator. Yet this is not as detrimental as it is felt in archaeological material. It is told most of all by conversion of information from material form into word form, an operation unnecessary for a historian to take. In ancient written sources it has already been accomplished by an ancient author. He was most probably biased and as distinct from modern scholars he had no broad interests or deep understanding of general trends. But in return he, as distinct from an archaeologist, had to translate not fragments but a whole body

Metaarchaeology
term archaeology. In fact, strictly speaking, they do not constitute archaeology and have no grounds to do so. In Archaeology of Cognition the French historian of science Michel Foucault (1966) uncovers past layers of the mental culture of scholars. In it he presents how scholars approached their study in the 19th century, in the Renaissance period and earlier still, and which principles were adhered to in the study. It is clear that Foucault designated his theme archaeology in a sheerly metaphorical sense (probably in order to alienate the consideration from descriptive history) this is a simple case. A more complex case is the so called industrial archaeology, which would rightly be involved in archaeology if archaeology was identical to the history of material culture. But if it is not then industrial archaeology also has to be left outside archaeology. This is a branch

31

of history of material culture using some methods of archaeology, ethnography and museum activity. Another example would be if archaeologists together with criminologists were invited as experts to participate in the opening of graves from World War II. An archaeologist would call this excavation activity, while a criminologist would say exhumation work. In this case their work becomes neither archaeology nor criminology. It is an expertise for history. Similarly a radiochemist determining the date of a Neolithic hearth by radiocarbon dating is practising within archaeology, not in radiochemistry. In summarising what characterises the place of archaeological sources among other kinds of sources of information on the past, we can say that they have a specicity that forms the basis for the separation of archaeology as a special discipline.

32

Acta Archaeologica PART III. NATURE OF ARCHAEOLOGY

4. Methodological nature of archaeology

1. METHOD AND METHODS Despite long debates on the subject matter of archaeology, most archaeologists themselves still declare their perplexity as to the nature of it. Sir Mortimer Wheeler wrote: What in fact is archaeology? I do not myself really know (1956: 17). Robert Braidwood added that archaeology is what archaeologists do (1959b: 1), a comment that agrees nicely with the opinion of Friedrich Koepp, There is no archaeology, there are merely archaeologists (Hausmann 1939: 11). The denition and description of the subject matter of archaeology alone does not exhaust its determination. In order to comprehend what archaeology is, it is insufcient to answer the question of what it studies. In addition one has to determine how it studies its objects (Taylor 1967: 23). Formerly this was described by determining its methods (cf. Reinach 1911). In the West even now the expression theory and method of archaeology is used (Hawkes 1954; Willey and Phillips 1958; South 1977; Schiffer ab 1978; Stjernquist 1984), and sometimes in Russia such an expression can be seen (Rogachev 1978). In a recent book by Evzen Neustyp ny (1993) Archaeological method specic methodology in general is described as distinct from a textbook of archaeological methods (such as, say, by Hrouda 1978, Martynov and Sher 1989 or Djindjian 1991). On no account is one particular means of research meant by Neustupny, Hawkes and others when they use the word method (evolutionarytypological method, cartographic method, method of horizontal stratigraphy and so on). Nor is the sum of different means of research implied by the word. The point is to inquire which general path of cognition it is that gives archaeology its methodological distinction, choosing from the few principal possibilities there are for a discipline. What is its specicity in respect to its methodological nature? This requires asking and solving questions of a classicational character. Is archaeology a humanistic discipline or a science akin to physics, chemistry and sociology? Does it operate as a discipline distinct in itself or as part of some other discipline? Or is archaeology just the reverse, an entire complex of disciplines? Is it a self-dependent discipline, or is it an auxiliary, subsidiary thing? Can archaeology be counted among the fundamental disciplines? A practical conclusion from this circle of questions is to ask what tasks, procedures and methods are admissible in archaeology, and which ones are not? Of course, in such an understanding the method of the discipline, i.e. its methodological nature, its epistemological character, and its status and prole are determined largely by its subject matter. The typical properties of this material and the aims of study condition how it approaches its material.

2. HISTORY, SOCIOLOGY, AND ANTHROPOLOGY In Anglophone countries the main division of research disciplines falls into a) natural and exact ones and b) the humanities. There is no separate term for research activity (like the Russian nauka or German Wissenschaft). Two terms exist instead, science used for the natural and exact disciplines and the term humanity for humanistic disciplines. If there is a need for designating the research activity without this specication only approximately similar terms are used: knowledge, scholarship, discipline, research eld etc. Sometimes the term science acquires a more general meaning and it is used in this way, but this can cause confusion. When considering the position of sociology it is mostly located in science, together with physics, chemistry and political economy, while history is placed in the humanities with literary criticism and philosophy as its neighbours. In Russia we are very unaccustomed to this, but we feel that in the methods of sociology there is something that in fact makes it

Metaarchaeology
akin to physics and chemistry, namely its revealing laws by generalisation of facts, its negligence of individual facts, and the proving of hypothesis by independent facts. In sociology facts are interchangeable and not central to analysis as compared with laws. On the contrary, it is difcult to deny the methodological similarity of history with literary criticism and philosophy, although typologically history is rather closer to geography. In USA the anthropological complex of sciences exists separately, consisting of physical anthropology, cultural anthropology with ethnography, prehistoric archaeology, linguistics, and psychology. This entire complex is located in science. To such a grouping a Russian is even less accustomed, but for Americans these disciplines do indeed use similar methods, approximately the same as used in sociology. The exception is archaeology, for which it is better to speak of separately. Other things, and in some respect more important ones, depend on these criteria of division rather than on the object. In particular the methodological nature of the disciplines depends rst and foremost on the aims of study. Meanwhile in the classication of disciplines there are known conceptions based purely on the difference between methods and aims of study. In the natural sciences F. Bacon distinguished between physics of abstracts and physics of concretes (1977). A. Comte built his ve-unit scheme of the increasing complication of objects just for the row of abstract sciences, but beside each abstract science he placed the corresponding concrete one, for example alongside sociology and its generalising nature he placed purely fact-describing history (1900). The most well known among such conceptions is W. Windelbands and H. Rickerts scheme dividing disciplines into nomothetic (aimed at revealing laws in a mass material), and idiographic (concerned with exposing and explaining facts in their individuality). We can liken the world in which we live to a at. When moving into a new home we learn laws of possession, rules of house keeping and using the home. Standard norms about its installations and furnishing are interesting to us too. But besides these things, it is important for us to know how these norms are realised in the given individual home where in its walls the electric wires run, who the neighbours are,

33

where the windows open onto etc. Geography serves to such orienting in space, as history does to time. Issuing from this example, one can determine historys interrelations with sociology. Sociology reveals laws of functioning, the development of society, and laws of historical process, while history analyses events of the past and nds causal links between them. History, one could say, reconstructs the historical process. Both sociology and history are concerned with facts but as Gulyga showed, they manifest different relations between fact and law (Gulyga 1969). Sociology generalises some totality of facts, usually a representative sample, extracts a law from them, and in this function facts are mutually interchangeable for sociology; thereafter they lose any interest for it. As distinct from sociology, history is interested in facts themselves, in each one separately, in their concreteness and individuality. For History, Julius Caesar cannot be replaced by another great general, like Napoleon. Sociology needs facts only for revealing the laws. History needs laws mainly for explanation of facts.

3. ARCHAEOLOGY AS A SCIENCE David Clarke had a vision to make archaeology a rigorous discipline. Lacking an explicit theory, he wrote, ... archaeology has remained an intuitive skill ... (1968: XIII). He welcomed the use of methods borrowed from natural and technical sciences, but he noted, scientic aids no more make archaeology a science than a wooden leg makes a man into a tree .... However, the use of theory, models, mathematics etc. being introduced into the very core of archaeology would transform it, he hoped, into an analytical discipline comparable with sciences and similar to them in status (1968, 635). Nevertheless, according to Clarkes notions, it would still not be a science. The temptation to see this aspect of archaeology as an archaeological science must be resisted. Analytical archaeology is not a science but it is a discipline, its primary machinery is mathematical rather than scientic. The archaeologists pursuit of the scientic mirage has long obscured the point that a study may be based on empirical observation, experiment, induction and formation of hypotheses without necessarily being a science

34

Acta Archaeologica
MacNeish published his book entitled The Science of Archaeology? (1976), a title noteworthy for its question mark. In Poland Zbigniew Kobylinski (1981), stimulated by the example of New Archaeology, called to provide nomothetisation of archaeology, to make it into a science revealing laws and ruled by theories. Really it proceeds from axiomatisation too, on building a system in which everything is logically co-ordinated and built on principles. Urbanczyk responsed to Kobylinski with an extremely sceptical response. Przemysaw Urbanczyk (1981, 58f) admits that it would be ideal to have a scientic theory in the full sense as an ordered system of statements linked by outlined logical relations. Yet it is difcult to expect this to be a deductive system like in mathematics. Such deductive sciences completely abstracted from concrete objects do not expand our knowledge on reality. Only theories as those of empirical sciences can be in question such as the electromagnetic theory of light or classical mechanics. Some of the axioms in them belong to determinative and statistical laws. However in archaeology they have not achieved exact quantitative expression and will still not achieve it in the near future. The empirical theory of archaeology still remains in the sphere of dreams. In general the position insisting that archaeology be turned from humanity or even art into a science may be considered as occupying the middle ground between two extremes (Fig. 3). At one extreme are the scholars who assume archaeology is a part of anthropology as a nomothetic social science or, as the majority of Soviet archaeologists see it, archaeology as a part of sociologically oriented history, Archaeology with a capital S (Flannery 1973). The other extreme is represented by scholars who insist that archaeology is not an exact science (Amalrik and Mongajt 1959, 252). It must be admitted that both extremes seem to be more logical and consistent than the middle position. Indeed, disciplines can hardly change their methodological nature arbitrarily. The methodological nature of a discipline is stipulated by its object, aims of study and place in the system of knowledge. If it is changed, we simply get quite a different discipline, usually with a different name, for instance, astronomy instead of astrology. So archaeology, by its methodological na-

Fig. 3. Archaeology in the system of disciplines conventional look.

(1968, 663). So for Clarke, mathematics is not a science, and archaeology as an analytical discipline is not a science either. Then it appears that analytical operations theory building, models, and use of mathematics do not qualify as scientic studies. Clarke views a science as distinguished by a high degree of certainty, while archaeologists can reveal only more than chance regularity. On this basis sociology and anthropology are not sciences either. What Clarke described as an analytical discipline, others would safely attribute to science, even to a hard, exact science in the strict sense of the word. The differentiation of exact sciences from inexact ones is very poor though, especially if using a degree of certainty in the differentiation. Physics and chemistry were evidently ideal for Clarke, whereas archaeology was his love. Why did he not claim the term science appropriate for analytical archaeology? It may be a psychological matter that could be most desirable to reveal. We can surely say that Clarke had a dream to transform archaeology into something like science. Strangely enough, Christopher Hawkes, from whom nobody would expect, shared this dream with Clarke and was even more radical in his view of it. At the anthropological section of the British Association he gave a presidential address, which he entitled Archaeology as Science, and which took place ten years before Clarkes Analytical Archaeology appeared. My theme, he declared, is not archaeology and science, but archaeology as science (Hawkes 1957: 94). In America, Robert

Metaarchaeology
ture, is either a science or a humanity. Or even art. However the whole of this apportionment is correct only if: (A) The nature of archaeology is dened by its alternative association either with anthropology (correspondingly with sociology or sociologically oriented history) or with idiographic history; (B) These two domains are really two halves of research knowledge: science and humanity. This is not the case however. Generally I share Walter Taylors opinion that archaeology is neither anthropology nor history (Taylor 1948, 44). I fully agree with D. Clarkes insistence that archaeology is archaeology, is archaeology (1968, 13). Clarke explained very simply that archaeology cannot be regarded as history because the nature of the archaeological record is such that there is no simple way of equating our archaeological percepta with ... lost events (Clarke 1968, 13). It is worth adding that archaeology and history have different inspirations of knowledge; history strives to understand unique events and heroes, whereas archaeology is obsessed with generalisation, typication, and its central concept is type. I must add that archaeology cannot be considered the same as anthropology or sociology either, because the system of archaeological data is lifeless. It has no simple and direct correspondence with a society that lived in the past and with the cultures that lived in the past. So regularities we reveal in archaeological data are not the same as the laws that were in force in the living societies and cultures of the past. I would also add that to me archaeology, even prehistoric archaeology, is not equivalent to prehistory, although D. Clarke assumed such equivalence (1968, 12). To him Archaeology and archaeologist contain prehistoric studies and the prehistorian. It follows that the prehistorian is always an archaeologist ... archaeology is often synonymous with prehistory. On that point, David Clarkes teacher Graham Clark seems to me to have been more logical when distinguishing clearly between archaeology and prehistory in the rst chapter of his book Archaeology and Society, entitled Archaeology and prehistory (Clark 1957, 9). I am fond of Childes aphorism Archaeology is

35

one (Childe 1956, VI). This is the idea to which David Clarkes well-known demand for general theory fully conforms. He demanded a central theory which should unite studies within the discipline regardless of area, period and culture (1968, XV). He meant the wider scope of archaeology stretching beyond the perimeter of prehistoric studies and mentioned such aspects as classical, medieval, recent colonial and industrial archaeology (1968, 12). If Childe and Clarke are right, and I think they are, then we must treat prehistoric times within archaeology just as we treat the classical ones. So if we distinguish between classical archaeology and ancient history, why do we need prehistoric archaeology and prehistory to merge together? There are some other arguments in favour of delimiting between these branches of knowledge which are presented in my article To separate Centaur (Klejn 1993). According to D. Clarke, recent development inclined to distort these terms in a dangerous fashion. There is currently a tendency to take the term prehistory as meaning a writer of history covering periods without written records with the implication that the prehistorian is an armchair synthesiser of the analytical work of the archaeologist. Here the term archaeologist is warped to mean the unintelligent excavator or the narrow-minded specialist ... (1968, 12). Indeed, the archaeologist is an excavator and a narrow-minded specialist and cannot be different because of the limited nature, scantiness and incompleteness of his data. Quite the opposite is true for the prehistorian, who should be a synthesiser because it is necessary to put together, compare and combine results of studies covering different kinds of sources. Not only are there archaeological, but also ethnographic ones, linguistic, osteological, folkloristic and so on. Yet to undertake comparative studies and interdisciplinary synthesis a special kind of knowledge is needed. It has its own methods, it requires peculiar erudition, and needs special training. Nobody forbids an archaeologist to do the prehistory too, but then he must realise that it is another profession and he is thereby faced with the problem of mastering a second profession. Essentially, the root of the confusion lies in the conventional dichotomy dividing the whole totality of knowledge in the sciences and humanities. Other

36

Acta Archaeologica
most part as restoration and recovery. Taking into account this undesired coincidence a suggestion was made to determine the essence of the reconstruction procedure by comparing it with prediction (to which it is similar by initial data and by composition of actions) and so it was designated as retrodiction. This is the procedure, and correspondingly the totality of methods applied in it, of indirect acquirement of inferred knowledge on past objects on the basis of present objects or other past objects (Nikitin 1960). Reconstruction is far from the main task of history, indeed, if it was we would reduce history to a chronicle. In history we seek something more: the systemic order, explanation, and causal connection in the past. But it is not only this I refer to. In the spirit of historical science the task of reconstruction as recovery is not manageable, or to be more exact, is manageable in part. It is possible to reveal regularities or laws of the historical process. They are general sociological laws, and in Marxism it was historical materialism that was allocated to establish these laws. It is possible to introduce correctives that are subject to conditions of territory and environment (general sociological learning concerns the principles of an impact of these conditions on the historical process). Yet going along this path one cannot reach individualisation. It would be necessary to enact not only main regularities or laws and main local conditions but an enormous number of small particular circumstances which are not even always known to us. Besides it would be necessary to take into account a personal choice of many people, which is also regulated by some laws, but those are probability laws with a great extent of freedom for each individual case. Now nally to the question of how to take chance into account. P. N. Tretjakov once remarked that, ... There are no contingencies in the development of culture ... (P. N. Tretjakov 1962). Really, this statement itself was not a chance, rather it reected the pathos of the day. None other than T. D. Lysenko (1948, 520f) pronounced that Such sciences as physics and chemistry became free from chance events. Therefore they became exact sciences. Ejecting Mendelism-Morganism-Weissmanism, we automatically banish contingencies from biological science. We must rmly remember that science is the enemy of chances. Marxist theory refused absolutisation of determin-

Fig. 4. The same: Klejns suggestion.

classications of knowledge were proposed (in particular by F. Bacon, Comte and others) which provide division in abstract and concrete disciplines. My own scheme is developed according to this trend (Fig. 4).

4. RECONSTRUCTION IN HISTORY Reconstruction of the past is called the main task of history, where reconstruction means recovery, reinstatement and reproduction. This is a very difcult task since a historian deals only with fragments of the past and the complete picture of the past is not reected in the sources, it remains for ever unknown to the historian (Erofeev 1976). Nevertheless many historians consider the task of fully adequate reconstruction as feasible and even suppose that the reconstructed past has preferences over the past reality because the general picture of the past is better observed from a distance and as the sophistication of historians grows (Dondt 1969). Naturally it is the mental reconstruction that is implied here. In confrontation with the terms restoration, reproduction and the like it becomes clear that the term reconstruction is not unambiguous or simple. Due to its similarity to the concept of constructing it entails notions on the complexity of the procedure and on the inadequacy of the result to the past itself. Nevertheless reconstruction is perceived for the

Metaarchaeology
ism but Marxist practice demanded and favoured it because the belief in rightfulness and destiny of the complete and nal victory of socialism was supported by determinism, and here it was less dangerous to overdo than underdo. In knowledge on prehistory this practice had a theoretical tradition too, which was spontaneously continued by Marxist archaeologists and ethnographers. The same simplication was also characteristic for evolutionists. For them history was moving with the exactness of a railway timetable, and periods, epochs and stages changed like passing stations on a one-way track. Yet we know that trains sometimes experience delays, let alone accidental crashes, and the route is not usually along only one track. As for history, it does not run on rails at all. How is the historical process recovered however, in all its completeness and in its diversibility of individual manifestations? Strictly speaking it cannot be recovered at all in this way. It lies fragmented and is then assembled from saved remains and fragmented reections, and of course never recovered in its full entirety. The remains produce doubts because they can be damaged. The reections are of course subjective and need to be corrected according to other reections if they are present. Without the remains and reections such operations are only connected with facts as interpolation, extrapolation and inference by analogy. In any connections with facts they are nevertheless hypothetical of course. Still more distant are disposed models based on these areas of knowledge and on theories, i.e. on sociological, demographic, psychological and other types of regularities. The general principles of historical reconstruction are formulated more completely by the late geologist S. V. Meyen (1978, 80ff). Only by historical ction is a substantially more concrete and colourful recovery obtained, but this is accomplished with the rich and in many ways free play of imagination, so that any guaranties of adequacy fail. All this activity is fully justied. It deepens and broadens our knowledge, and brings us to a more sophisticated understanding of the past. Apart from history being unable to return the past to us (this in itself is a banal truth), it does not even recover it mentally in the esh in all obviousness and concreteness, although sometimes it bears such an illusion. The past produced by historical studies appears fragmented in

37

its solid core and schematic in its all-embracing attempts. This fragmented past is like tatters of events, lumps of what happened, and they are hung on fragile constructions of theory. History cannot retrieve facts that have disappeared without trace, and can no longer make them concrete or retrieve them in their concreteness and uniqueness. History can only build plausible hypotheses about past existence and on the character of the past as seen through its hypotheses. It sees the past in terms of certain general characteristics, which are expressed by its hypotheses. In a sense, this is actually sociology rather than history. It does not reconstruct, but it synthesises historical past on the basis of its sources and the result of the synthesis is not one of reconstruction. If it were able to reconstruct then it would also be able to predict, which is far from what it does! Indeed Hegel said that people did not draw lessons from history, for each situation was too individual to do so (1935, 7f). In Marxs introduction to his 1857 Economical manuscripts we nd a very interesting statement which is often cited without recognising its full meaning. According to Marx, the categories and structures of the most developed formation give us the possibility to penetrate into the organisation of all the obsolete social forms ... The anatomy of man is a clue to the anatomy of ape. On the contrary, the hints of higher species in lower species of animals can only be understood (sic! L. K.) in the case of the higher already being known. Bourgeois economy gives us thus a clue to classical times etc. (Marx 1958 [1857], 731). This is an unambiguous statement on the impossibility to predict in history! And consequently on the impossibility to reconstruct it What is this principle conditioned by? Evidently by the openness of the prospects of development, by the subjective factor, by the interaction of regularities with those contingencies without which history would bear a very mystical character (Marx 1964 [1868]: 175), it would be doomed in a fatal way. K. Popper proposed that predictions in the form of prophecy are in principle impossible (1957). Avowed to the likening of history to natural sciences, A. I. Rakitov advances serious counter arguments, but it appears that he only defends the possibility of making predictions of a special kind only probabilistic predictions of stable types and successions of socially sig-

38

Acta Archaeologica
chaeology is a source-studying discipline and archaeological sources are ancient artefacts and assemblages, its task is to convert information from the language of things into the language of history, the language of historical phenomena, events and processes (Klejn 1981b: 1617). In other words to look behind preserved remains and traces, and behind their relations to reveal ancient phenomena, events and processes. How do archaeologists go about doing this? The simplifying approach, to which Soviet sociologists and even others were inclined, consisted of ascribing a corresponding social meaning to every kind of element and trace, and to their relations. In a word, a social meaning for every type of archaeological pattern, where the type is considered a sign, as a stable correspondence to a certain category of historical phenomena, events and processes. J. Hill and M. Schiffer later called these kinds of archaeological patterns archaeological correlates to social phenomena. Correspondingly the latter could be called social and cultural correlates to archaeological patterns. They are composed like an archaeological-sociological alphabet for reading the past. One only needs to know this alphabet in order to read history freely from archaeological texts. The problem however is that social phenomena and historical events in different conditions leave different traces in material culture. Both sling and bowand-arrow served the same function. On the other hand archaeological traces of historical events are not unambiguous, behind them different phenomena of the past can be hidden. A dwelling burnt to the ground can be interpreted in several ways: natural disaster, arson, war or ritual could have caused the re. Even the causes of a contemporary re are a headache for criminologists, while in the archaeological situation many centuries have passed and possibly anything but the necessary distinguishing details remain. Step by step archaeologists reconstruct former states of sources, monuments, artefacts and assemblages, retrospectively from the end of their existence to the beginning. Even if the law is not revealed or is unknown to the archaeologist, he always has the handy means of substituting the law or doing without it, although this puts the reliability of the result at

nicant activity in certain time intervals, not on concrete and exactly xed results, i.e. events (1982, 282ff). Time and again, assurance is given that there is no fatal predestination in concrete turns of the historical process. At each step history appears before an open situation, with opportunities of alternative solutions or choices (Gurevich 1969). When events have already happened, it seems that everything could have be foreseen and predicted, and such predictions use to be made behindhand: vaticinia ex eventu. Yet predictions of concrete events are made only as guesses. As inter alia the geologist V. I. Onoprienko noted, retrodiction and prediction are similar in their logical structure and they are connected with each other. These two procedures have time symmetry, opposite in direction to the ow of cognition ... Retrodiction looks like the basis of prediction (1976: 158). Consequently they are similar in their strength of inference. The impossibility of historical prediction implies the impossibility of full retrodiction, an unrealisable in terms of adequate detail in reconstruction of the concrete course of history, and in terms of those details which did not leave clear traces.

5. RECONSTRUCTION IN ARCHAEOLOGY Both in Russian and Western publications one can frequently see the statement that, as a prominent American scholar writes, the aim of archaeology is historical reconstruction (Kroeber 1937, 163). V. D. Victorova noted that Soviet archaeologists applied the term historical reconstruction or reproduction, recovery of historical reality by archaeological materials and did so from the late 20s until early 30s (1988: 225). In practice when real reconstruction of artefacts and assemblages is concerned, one speaks of an archaeological reconstruction, while when mental reconstruction of phenomena, structures and events is concerned, one speaks of a historical reconstruction in archaeology. At rst sight, by its role in the discipline and by its procedure, reconstruction in archaeology is completely like reconstruction in history. But while in history reconstruction is only one of the tasks, everything in archaeology can be reduced to reconstruction. It is the all-embracing task of archaeology. As far as ar-

Metaarchaeology
stake. This method uses analogy in order to compare objects. An investigator seeks objects belonging to the same type or objects of a similar kind, and looks for similar assemblages that are already known or other ones that are better preserved. Thus he provides an interpretation by close comparison. As Chang stated with some exaggeration, archaeology as a whole is analogy .... He explained: ... for to claim any knowledge other than the objects themselves is to assume knowledge of patterns in culture and history and to apply these patterns to the facts (Chang 1967b, 109). J. Hawkes stated that the main aim of archaeology is the reconstruction of individual events in time (1968: 255). If the word individual is understood as separate, then this statement is doubtful, and if understood as unique, taken in their specicity, it is simply erroneous. As is known to all archaeologists since their student days, archaeology is oriented on the typical, on typology. It retrieves only what is general from artefacts, assemblages and cultures. In this way it is fully consistent with history. But while for history it means that reconstruction receeds from its central position and appears largely irrelevant, archaeology does not squirm by such refusal. It is content with a much smaller degree of individualisation, while it supports culturology (anthropology in Anglophone understanding) and history of culture, sociology and sociological history the history of social structures which the Annales school propagated. Archaeology helps protohistory history of early peoples that were already neighbours to literate societies to retain the general canvas of the cultural process. Archaeology connects prehistory with palaeoanthropology and palaeontology since for a signicant part of that period man was little different from animal, and animals do not possess biographies and have history. This is a time for which individualisation was to a large extent objectless. Does this mean that archaeology is not concerned with unique phenomena at all? Not for toffee, of course. Unique objects are not so infrequent among archaeological nds. Furthermore, individuality can be observed in every object as the opposite side to what is typical. Indeed, in the discovery of individuality it is not so much the presence of such properties in the object that is shown (although this is of course

39

important too) as the attitude of the subject of inquiry on the object. Of course the possibility of such an attitude is to a great extent conditioned by the presence of such properties in the object. A considerable amount of unique things appears where an archaeologist nds objects well preserved. Having obtained such an object he must transmit it to a historian untouched, but of course archaeologically processed having been dated by assemblage, having passed through an expert investigation, undergone analyses etc. Archaeology has the means to convert unique objects having no analogies into more amenable objects of typological study. One should only dismember mentally such an object in a number of details so that each of them appears without complicating elements (they have formed neighbouring details), resulting in the objects appearing very simple and no longer retaining their uniqueness. So a complex pattern disarticulates itself in its elemental components and in very commonly used motives. The combination of these elements and motives remains unique. But it does not affect archaeology. The uniqueness exists before archaeological analysis and returns to the thing after this analysis. Archaeology transmits objects to historical investigation having reconstructed them and, as much as it is accessible to archaeology, interpreted them; but they will undergo individualisation after their inclusion into history.

6. ARCHAEOLOGY AS DETECTIVE ACTIVITY So far we have considered the place of history and of sociology in the system of knowledge. What about archaeology? We have seen how it differs from both history and sociology so which group of disciplines does it belong to? Certainly it contributes to history and, as yet in lesser measure, to sociology, and to anthropology too. However this fact says absolutely nothing about the methodological nature of archaeology, it only uncovers its working contacts, connections and partners. The fact it co-operates with history makes archaeology no more a kind of history pace Clarke than contacts with men make horses human, or vice versa. Archaeology is archaeology, is archaeology ...

40

Acta Archaeologica
1997, 295f), but usually proposed as a metaphor. Yet it is not only this, it lies at the very core of things. The very rst of these authors who made comparisons, Clark, was quite serious when he began his foreword to a compendium of scientic methods in archaeology with precisely this analogy. In his mode of work, he wrote, and in his general approach the archaeologist resembles in several signicant respects the detective. Like the disciples of Sherlock Holmes he seeks to recover the activities of men in past time from clues which compensate for their incomplete and often vestigial character by abundance and diversity. Most of this evidence is necessarily circumstantial it can only be made to speak by bringing upon it the resources of natural science; and the more effectively these can be harnessed the more complete the information likely to be won from traces, which in themselves may appear to the layman to be almost as slight as the blood stains and nger-prints used by skilled detectives to reconstruct crimes. Bartel Hrouda (1978, 15) remarked in a textbook that, Basically archaeology is a kind of criminology. That is why archaeology adopted so many working methods from criminology. It enriched itself with them especially during trials with archaeological subjects (e.g. the Gloselle scandal) the custom of protocol investigation, the reckoning of scientists and experts etc. In the study of objects a detachment of the place of origins is demanded from legal experts and similarly from archaeological classicators (Daniels 1978). That is why some prominent archaeologists are known as popular writers in ction as well, both in the detective genre (Casson, Glyn Daniel) and in science ction (Francois Bordes as Francis Carsac). One can object and ask if the task of a historian is not also of the same kind? The historian also tries to restore the events and processes of the past, although he mainly relies on material traces or on considering the stories told by participants, eyewitnesses and other informed persons. Let us introduce itemisation, not really the task of history, but a eld of historical source-studying, the preparation studies of written sources, and there are specialised professionals for this task. A proper historian analyses the results of such an investigation and tries to answer other questions on the causes of the restored events, on their causal connection, on their meaning and signicance, on

Archaeology is unconditionally connected with history, but it is connected by business relations, as a partner. In other areas they are sometimes not only dissimilar but not even akin. They are of different methodological nature. While history is an individualising discipline and, taking into account the individuality of human creatures, history is a humanity, archaeology is not. This is a fundamental difference. Having said this however, they do have something in common. In particular it is apparent that neither archaeology or history reveal laws, they only use them. History uses laws for explaining facts while archaeology does so for reconstruction. This means that archaeology does not belong to the fundamental sciences but to the applied ones. Something similar is remarked on by G. A. Antipov with respect to narrative reconstruction in history: Differences between ordinary research cognitive procedures and narrative reconstruction is fully analogous with, for instance, the difference between the activity of a specialist in the sphere of theoretical mechanics and the building of bridges, houses, machines etc. (1987, 94f). For even greater reason this can be said about archaeological reconstruction about real creation of missing details and constructions (like modern reparation and restoration activity). Due to the similarity and even identity of logical operations this also applies to the whole of historical reconstruction in archaeology. As soon as one tries to search not for its partners but its nearest relatives among disciplines, neither history, sociology or anthropology would be discovered in such roles. Unexpectedly, one would nd similar methods, principles, aims far from pure scholarship, namely in criminological tracing, in professional knowledge and the skills of a detective or police inspector and court investigator. The only difference is that a detective serves society by exploring merely the pathological side of social life and that he arrives at the place of the event without such an enormous delay as the archaeologist does. But the nature of the aim is the same; to restore the events and processes of the past by their material traces and fragmented remains. The analogy with detective activity was sometimes used by archaeologists to describe their trade (Clark 1957, 20f; 1963; Braidwood 1959b, 5f; Klejn 1967; 1978b, 48; Adams 1973; Shanks 1992, 53ff; Olsen

Metaarchaeology
regularities in the course of history and how they manifest themselves in the events. Thus it is more similar to the function of a court and the activity of judges. The archaeologist is nearer to the stage of the criminal case in the related frame of detective activity and the prosecutions investigation. Restoration of antiquities and works of art is another similar occupation. There is no need to prove likeness. The only thing to be stipulated is that restoration is not just a skill or a trade. It has its own basic rules, principles, methods and system of concepts. This makes it is a scholarly discipline too. So

41

what kind of disciplines are these? I suppose they belong to a wide group of applied sciences, and this goes for archaeology too. This means that Walter Taylor was near the truth when he afrmed that archaeology is no more than a method and a set of specialised techniques for the gathering of cultural information. Archaeologist as archaeologist is really nothing but a technician (Taylor 1948, 44;1968, 41). Nevertheless archaeology is not just a method or technique. It has its own system of concepts, its own theory and its own principles.

42

Acta Archaeologica

5. Principles of archaeology
1. FOR THE SAKE OF CLARITY TOWARDS MURKY PRINCIPLES According to the physicist T. Hertzs statement, there were two paths of development in the history of physics, Babylonian and Greek. Along the Babylonian route scientists established a great number of concrete facts, particular regularities and a numerous connections. The Greeks in contrast tried to reveal axioms from which it was possible to deduce everything else (Hertz 1980, 173203). This is called the axiomatisation of science and it was along this route European science developed. One of the most important theories developed in physics is Thermodynamics, which is formed around three main principles. Another important theory is Classical Mechanics, based on three fundamental laws of motion and formulated by Newton in his Mathematical Principles of Natural Philosophy in 1687. Newton also formulated his general idea on certain principles: To derive two or three general principles of motion from phenomena and afterwards to tell us of how the properties and actions of all corporeal things follow those manifest principles, would be a very great step in philosophy, though the causes of those principles were not yet discoverd (1730, 377). Einstein contended: Fundamental ideas and laws which cant be reduced any longer, form an integral part of a theory ... The most important aim of any theory is to make the number of these fundamental elements as little as possible and to make the elements as simple as possible (1933, 183). There are two such principles in his Theory of Relativity. So far only physicists have been mentioned but in biology and psychology, as well as in the humanities, fundamental principles are not usually marked out, though they are implied. Apparently they have simply not yet had the necessary time to take shape. In contrast to fundamental or pure sciences, applied ones are, by their nature, more oriented to serve other sciences and practices and this is the reason they are less theoretical. Their objects are marked out less clearly in terms of structure and so it is more difcult to understand their scope with a common theory. Correspondingly it is more difcult to defend their scientic, or at least scholarly status. It is more difcult to prove that they are something more than a collection of methods, that they have their own strict theory, harmony, stable unity and systemic order. For this reason the problem of initial principles is particularly important for these sciences. Having enumerated some examples of how scientic methods and technical analyses are used in archaeology, C. Hawkes then concluded the following: But, whatever kind of archaeologist or prehistorian you are, your own claim to scientic status here, for yourself, rests only on the care which you take to provide as far as possible properly guarantied, undamaged, and uncontaminated specimens for the appropriate examination. And that it is genuine claim, and not simply a matter of taking common sense precautions against getting things mixed up, broken up, or messed up, must depend on the principles embodied in your part, our part, the archaeologists part, of the interpretation of the nd. If these are scientic, then we to that extent are scientic; if they are not, we are here no more than painstaking and inquisitive individuals who have managed to get some information from other people who are scientists while we are not (Hawkes 1957: 94, My emphasis L.K.). In a recent book by Lewis Binford we nd a similar idea. Speaking about inferences that the archaeologist draws from investigated data, Binford concluded that: All these inferences were made tenable by his linkage of archaeological observations to principles and laws of causation drawn from the sciences of mechanics, physics, and the related elds of applied engineering ... Archaeologists cannot wait for other elds to permit them to make reliable inferences about the past. They must themselves develop a science of archaeology (1983 198: 17). Gibbon, writing in his 1984 book Anthropological Archaeology includes a section on Methodological principles and rules. He notes that more often than not these principles and rules are untestable or at least untested tacit assumptions that lie hidden in our thinking (1984: 93). In a later work, Explanation in Archaeology, he divides these principles in two sets (1989: 3f). The rst set consists of primary presuppo-

Metaarchaeology
sitions without which any archaeology as it is usually understood would be unthinkable. As examples Gibbon provides such banal truths like Archaeological assemblages are material remains of past human activity and We can learn about the history of the human species by studying archaeological remains. Such primary presuppositions can be regarded as the ultimate justications of archaeology, and they lead directly to general methodological directives such as Preserve and examine archaeological sites and materials! . Denial of the primary presuppositions of archaeology ... would entail the collapse of archaeology as we know it. The second set of principles is formed by the secondary presuppositions, i.e. presuppositions which are supplementary or subsidiary to the primary set. For instance, An archaeological assemblage is a mirror of past human activity. Being unnecessary for the very existence of the archaeological discipline such presuppositions determine the strategy of interpretation. Whether the past human activity is reected in archaeological monuments as in a mirror is a debatable question, yet it has nevertheless conditioned their forming. The methods of interpretation depend anyhow on this connection. These principles are not in any way banal, they are worthwhile studying. In any case one can deny neither the necessity of some general principles in archaeology, or their a priori status, nor the fact they must be principles of interpretation. I have examined in brief their necessity and I shall deal with it further at the end of this chapter. For now let us remain with the other two characteristics. The nature of archaeology as an applied science is connected with its aim of serving other disciplines. It serves scholarly research just like other applied sciences serve practical life and it is this indirect goal of archaeology that makes interpretation so crucial to it. Principles of interpretation appear as central in its theoretical system. They control its relations with other disciplines it serves and thus the principles of interpretation make it relevant for them. Interpretation is placed at the very output of archaeology, but the discipline exists purely for the sake of it. The a priori nature of these principles is connected with their axiomatic character. As we saw, this is also how principles of other disciplines are that evidently belonged to sciences. When Newton considered the

43

disciplines he began to assume an uncertainty of their grounds. The axiomatic character of these principles was clear since the times of Aristotle, who stated that the existence of the principles is to be accepted, all the rest to be proved (Aristotle, Anal. Sec., I,10). For the most part, a remark made by Mill on this matter proves to be correct: There is nothing more dark in each science than the fundamental principles of the science itself (as quoted by Regirer 1966: 90). Fundamental concepts and principles of any science or discipline appear to be fundamental for it precisely because they do not nd grounds in this discipline itself on the contrary, the discipline is based on them. The grounds are found either in other disciplines or in practice and in common sense. It is the business of philosophy to deal with them. In short, getting to the core of axioms is an obscure task best left to philosophers. From our side, archaeologists must ensure axioms are reliable, reasonable, stable and arranged so that the other propositions of our discipline can be inferred from them. Inferrence should be explicitly possible according to strict rules of logic. Is the axiomatisation of archaeology possible?

2. THE SEARCH FOR PRINCIPLES How many and what principles build the basis for archaeological interpretation and make it explicitly reliable? When looking for axioms it is difcult to suggest any preliminary order, save perhaps starting from within the discipline, from the supposed derivatives. If something is done already, one can begin with a review of existing versions, with the experience of a science. Then one may correct accumulated propositions and determine what is in shortage. In his book Foundations of Archaeology (1976) Jason Smith, the American archaeologist of Marxist orientation, writes that it is possible to reach scientic level by holding the following four assumptions: 1) that the world and universe are real and independent of human consciousness; 2) that all things are directly or indirectly interconnected through cause and effect; 3) that there are no things which by their very nature are unknowable, that is, supernatural or magical; 4) that change is a universal constant, that no things stay as they are forever. Yet these assumptions are applicable to all the

44

Acta Archaeologica
ments may be advanced exists between the things made by people, the manufacturer of these things and its users (1967, 201). The relationship is not determined, and surely one may advance statements about anything you like. Sangmeisters premise is so diffuse and weak that it is useless to rest upon it. This scholar appeared to be too cautious. In Gibbons book Anthropological Archaeology the principles of interpretation do not build a system but examples are provided and there are some very apt ones among them: The processes of the past were the same as those in operation at the present (the principle of uniformitarianism) (1984: 9397); People did things in the past for the same reasons they do things today (Roes principle of cultural uniformity). The last principle is entitled after Roe and the reference to Roes article (1976) follows after this In Gibbons book. However Roe has only narrowed and made more exact the principle formulated earlier by Hawkes.

Fig. 5. Principles of interpretation according to Hawkes.

sciences and to cognition as a whole. It is, as Smith himself notes, the basis of materialist philosophy (1976: 391). There is no sense in taking into account these principles, at least not in this form, for they are too general for our topic and they do not convey the specicity of archaeology. Vadim M. Masson built theoretical principles for archaeology from a Marxist positions. He took three basic principles of historical materialism from standard manuals (historicism, determinism and primacy of production) and, carrying them into archaeology, examined the resulting realisations (sociological orientation, preference to mass materials, interest in functions, Stadialist conception etc.) (Masson 1978: 32). The trouble was not so much that these principles were narrow and dogmatist. The problem resided mainly with the way they determined the nal appearance of the cultural-historical process under reconstruction, but not with the way to it from the material or the methods of reconstruction. These principles made the result deliberately known, but the way of reaching it was unimportant and was left to common sense, cunning speculations or tacit consent with an authority. James Grifns 12 principles, axioms of the cultural process, are of a similar kind (Grifn 1956). They set some reference-points for when the reconstruction is nished, what the cultural process must look like, and which conditions it must meet to be recognised by the scholarly community. Still these axioms, irrespective of whether they are correct and felicitously worded say nothing about how to carry out the reconstruction or interpret the archaeological data. In contrast to these principles the scholar Edward Sangmeister considered archaeology to have only one fundamental principle. The only premise which any archaeology works with ..., he wrote, is that a relationship on which some state-

3. THE ANALYSIS Two papers, those of Christopher Hawkes (1957) and of Robert Dunnell (1978), seem to be the most interesting and serious elaborations on the theme. It has proved useful to make the analysis of the problem just by considering these works, by examining their choice and selection of principles. In Hawkes presidential address not only reasoning about the methodological nature of archaeology is of interest, or the insistence of revealing the fundamental principles of interpretation, the selection of the principles themselves is attractive. What are the archaeologists principles, which are embodied in his interpretation of his excavations, measurements, and comparisons?, asks Hawkes. I have often asked myself this question ... (1957, 95). In Hawkes opinion there are only two principles controlling archaeological interpretation (Fig. 5): (1) People in the past did things for purposes which can, in the most cases if not in all, be correctly understood; (2) If different people did the same or closely similar things, ... their doing is not fortuitous, but requires a single ... explanation to cover them together,

Metaarchaeology
unless these cases are too separated in time or space or otherwise. If we look closely (Fig. 6) at Hawkes rst principle (comprehensibility of ancient artefact), one can see the criterion of simplicity is not maintained. At least two principles have become one here. The rst is the principle of cognisability of functions in material culture, meaning that one can comprehend the purposes things are made for. The second is the principle of connection between epochs aims that existed in the past can be understood by contemporary people as it was worded by Roe. When examining the second of Hawkes initial principles, the principle of non-fortuity of similarities in things, it does not appear very simple either. It implies two different propositions. First, that formal similarities in things are important since they are not merely formal some correlation between form and sense is implied. And second, that similarities are important because similar things have something common in their origin; Hawkes meant the real common origin of similar things from one cultural source and then spreading by diffusion. Convergence can also be intended here because there is also something common in it in terms of common causes and common regularities. Thus, where Hawkes saw two initial principles, at least four are present. The rst of them is rather general. It concerns all the disciplines on man and culture. I should even like to say, it is not very axiomatic: it is based on other principles. For instance, the principle of cognisability could not act if every thing were absolutely unique and was made for absolutely new purposes, ad hoc. So, to be able to rely on this principle one should accept the existence of some regularities in culture, some laws of satisfying the basic needs. That means the principle of determinism in culture is necessary and some others alongside. Thus behind, or rather beneath the rst principle implied by Hawkes, no less than three very important initial, and really fundamental principles should be placed: the unity of human nature and correspondingly the community of human culture, the existence of social and cultural laws, and the systemic order in culture. The second implied principle of those two into which it appeared reasonable to divide the rst prin-

45

Fig. 6. Hawkes principles corrected.

ciple of Hawkes is known as the principle of actualism. The third deals with the correlation of material parts of culture with its non-material components (ideas, social functions and so on). The fourth principle is not exposed in detail by Hawkes. He did not explain how and why the fortuity should be determined. Yet it was clear from the context that Hawkes saw an afnity between the separate cases, and correspondingly something behind the formal similarity namely a relationship caused by spreading. Behind this briey uttered reection a discourse is hidden, reminding one of Graebners criterion of form, which was worked out by him for the sake of estimating the likeness while determining the propinquity (homology) of ethnographic objects. Graebners criterion demanded the occurrence of a rare combination of attributes, not conditioned by the nature of the object. Graebner also had another criterion, that of quantity the particular combination should appear repeatedly (Graebner 1911). However, Hawkes had predecessors in these matters in archaeology itself. Sophus Mller wrote: Where there is a likeness, there must be a relation, a connection of some kind (1884, 185). This principle does not look initial or axiomatic. It is inferential, quite evidently deduced from the recognition of laws in culture and society, i.e. from some kind of determinism. Indeed, if law-like regularities exist in the material, then likeness must exist as well. In that case neither differences would be fortuitous. They show either the manifestation of different regularities or the interference of some other factors. In contrast, if there are no regularities, then the differences are quite natural even if they are because of random dispersion. What is more, it is often impossible to distinguish between random and non-random events. True, Hawkes himself did not suppose it was necessary to use regularities for explaining the non-fortu-

46

Acta Archaeologica

Fig. 7. Principles of interpretation according to Dunnell. Fig. 8. Dunnells principles corrected.

itous character of similarities and differences. He was ready to look for explanation exclusively in identity and heterogeneity caused by diffusion. In turn, diffusion has its own regularities. The importance of formal similarities and differences in the archaeological material depends also on the recognition of regular relations between material forms and social meanings. All this leads to other, deeper principles. An article by Robert Dunnell (1978), or more precisely one section of it, is a central work on this topic, although both the article and the section in question are entitled without indicating this topic. It concentrates on functions but questions in fact the initial principles of the functional interpretation of archaeological evidence. The system of interpretation fundamentals (Fig. 7) is constructed by Dunnell in two axioms and in some derivatives inferred from them, which consist of three theorems and one corollary. Everything looks strict and logical like in mathematics. Although mathematics is not a science to David Clarke, archaeology when presented so structurally in its fundamentals looks very scientic. Dunnell made an impressive attempt in structuring the basis of archaeology. According to the rst axiom, the purpose of archaeology is anthropological description. Two theorems are consistently inferred from this, the rst one that archaeological data are incomplete and the second that it is not identical to anthropological data. Hence, enrichment of the data is necessary. The second theorem is that archaeological interpretation may take the form of functional reconstruction. The sec-

ond line of discourse consists of an axiom and two inferred propositions: a theorem and a corollary. It leads to the legitimate use of ethnographic analogies in archaeological interpretation. Although Dunnells construction is serious and interesting, its realisation can hardly satisfy (Fig. 8). Dunnells rst axiom is awkwardly formulated. The aim of archaeology is narrowed and its orientation on history is dropped. Overall, if it was necessary to ground the inherent incompleteness of archaeological data, the grounds might be chosen more correctly. Archaeology aims to be of service to other disciplines on the past like palaeoanthropology, its subject matter does not coincide with the nal object of their common interest and is merely a source of cognition of this nal object. It could be said that Dunnells rst theorem is no theorem at all as it does not need substantiating. It is simply a corollary to the rst axiom. As far as archaeological sources reect the nal object of disciplines studying the past, it is quite clear that they relate only incomplete information. The reection through the mists of time leads to remains and traces, and they give naturally less information than the object itself. The second theorem indeed deserves its status, since it needs to be proved. However, to ground it one needs rst of all to rely on positions which are not contained explicitly in the initial axiom. They are contained neither in Dunnells wording nor in the suggested substitute. For the substantiating, some propositions on the signicance of function in culture are necessary, on its connection with reconstruction of

Metaarchaeology
historic events. Also necessary is a very broad denition of function that includes semantic aspects and the use of the thing as well. Otherwise why would reconstruction be limited to function? On the whole, all this line of discourse does not lead to the revealing of interpretative mechanisms. It leads solely to the proving of its necessity in archaeology and to its goals of functional reconstruction. It does not explain how to carry out the reconstruction. Hence none of these propositions (the axiom, theorems and corollary) are to be included into the principles of archaeological interpretation. Rather they concern the foundations of the structure of archaeology and are similar to the rst series of Gibbons principles. From the recognition of incompleteness of data I would prefer to infer the demand for enrichment of archaeological information by non-archaeological sources, and not only by ethnographic analogies, a demand that I have advanced before (Klejn 1977d; 1978b, 3944, 5456). The second line of Dunnells discourse is of a different kind. It begins with the second axiom, the constancy of laws (uniformitarianism), and derives therefrom the corollary of actualism. Dunnell discriminates between these two principles. As distinct from the Evolutionists and in accordance with the task of his study he formulates the initial principle very narrowly according to the task of his study and disputably: Function, form and relationship between them are constant through time (Dunnell 1978, 45). Hence he deduces the theorem: Modern peoples are functionally representative of past time. After which his corollary follows: Functional interpretation of prehistoric data is by reference to ethnographic occurrence of similar forms. Dunnell sees this as the foundation of why ethnographic parallels are applicable in archaeology. The foundation being rather insufcient for the reason that ethnographic parallels would be better considered in the frames of archaeological signicance of analogies in general. Thus, having in mind the initial, fundamental principles of the archaeological interpretation, Dunnells work broaches them indirectly and incompletely, but his stipulations and the title of his work do not allow one to hurl criticism at him. Yet one of the principles, which has been noted by many practitioners for a long time, is sustained by him successfully. Moreover,

47

the whole construction presenting the base of interpretation, and by this I mean the built up system of principles and their derivatives, may become a model for further studies. While analysing these two important articles, Hawkes and Dunnells, it becomes clear that some fundamental principles were observed by the two scholars, while others were only dimly implied and popped up during the analysis. In my analysis here propositions of other scholars were also used.

4. SET OF PRINCIPLES After analysing these and other works we may proceed to a conclusion that six principles of archaeology have been selected in the long experience of its studies. Some scholars relied upon one part of them, some upon another, but on the whole six interconnected principles appear. (1) Determinism As Michael Schiffer (1976, 5) writes, archaeologists ... found it necessary to draw a wide variety of behavioural laws to facilitate documenting and explaining past events. These laws may appear in some variety as determining the course of evolution, or as keeping society and culture dependent in their development on natural changes, or as enforcing technological progress, and so on. For instance, Montelius Law concerns things of the same category and the same tradition the most similar things in shape, are also the closest to each other in time. Thus it formulated a correlation between formal similarity and chronological proximity. In other words, it described the development of things by their regular character (Clarke 1972a, 45). The principle of determinism, however, is not limited to the recognition of laws. The corollary of this recognition is no less important for science. Bohr declared, Really, I think ... all of us agree with Newton: the most profound basis of science is condence that in nature identical phenomena appear in identical conditions (Bohr, 1961 [1932], 22). The base of our deductions concerning culture is the same. How important is it to have the condence that in every case, cultural community or similarity is based on the identity of conditions either geographical, or racial,

48

Acta Archaeologica
ogy. This principle is built on the idea of bringing together the epochs. That is, as soon as the unity (or community) of basic traits in processes of the past and present are recognised, we can, with knowledge of contemporary processes, infer processes of the past. In other words, through studying contemporary processes we can transfer this knowledge to processes of the past. Thus two principles are operating here, an axiom and a corollary. The same line of discourse is present in Smiths principles of stratigraphy. Smith derives it from the geologist Hutton (1785) and expresses it with the aphorism The present is the key to the past (Smith 1976, 515). However, according to British and American tradition, he covers both principles with the term uniformitarianism. Meanwhile, since long ago the term actualism has been used too (see Guntau 1967), and also for the designation of both principles as one, which in my opinion is not very appropriate. The ethnologists J. Lubbock and P. Ehrenreich applied the rst principle to prehistory. The second was common place in Evolutionists works where it usually took the form of a label like backward peoples are like living fossils (Tylor 1958, 16, 2124; Morgan 1877, 10f). In archaeology Malmer (1997, 7) divides its application into three kinds: ethnoarchaeology, experimental archaeology and reasoning of an archaeologist based on his personal experience. On the basis of the actualistic principle a thing or an artefact is transformed into a source of information not simply on events and ideas, but exactly on events and ideas of the past, a potential historical source or record. (4) Systemic order in culture These general propositions exist in archaeology only as theorems and in order to deduce them they need not only the corollary of parallelism but also the principle of systemic order in culture. In general form this principle was worded as applied to archaeology by Sophus Mller (1884, 185). This scholar saw archaeological thinking as being underlain by our condence that there is order and arrangement in this world. Contemporary authors describe the systemic principle in culture with items that are characterised as interconnected and interdependent. This principle was accepted not only by Binford and his New Ar-

or of social stage. This proposition may be called the corollary of parallelism. It grounds the phenomenon of convergence. From this corollary the applicability of analogies in archaeology should be logically deducible. The signicance of analogies for archaeology, and in particular the signicance of ethnographic analogies for archaeological interpretation is recognised everywhere. Chang Kuang Chih has even found it possible to declare that As to analogy, archaeology as a whole is analogy (Chang 1967b, 109). Objecting to the exaggerated hopes pinned on ethnographic analogies, Binford did not deny the signicance of analogies in general and described the applicability of analogies as a separate principle. If two or a few phenomena have many traits in common they probably hold some additional traits in common, and the probability is higher if you are lucky enough to reveal the inherent tie linking the resemblances (Binford 1967, 1ff). (2) Universalism This is the principle proclaiming the unity of human nature (consequently of psychology too) and in connection with this the community of human culture. Haag mentions the psychic unity of mankind among the main archaeologists presumptions (1969, 47). It is implied that the psychology of all normal people is basically similar. So for any person the behaviour of another person is predictable in a concrete situation. It is also implied that any culture must in any case satisfy some basic needs of a person, and these requirements in food, rest, sleep, posterity etc. are mainly the same. This is the principle of universalism and it was formulated for ethnography by Evolutionists (Tylor 1958, 6). (3) Uniformitarianism This principle is most important for providing a foundation for interpretation. The foundation forms from the same regularities that are being observed now and which acted in the past (i.e. the principle of uniformitarianism, the constancy of laws). Only then can we interpret the results of ancient processes by resting upon contemporary experience the principle of actualism). This second one is a principle well known in all the disciplines studying remains of the past geology, palaeontology, palaeoanthropology, archaeol-

Metaarchaeology
chaeologists, but also by Trigger, a supporter of the contextual approach. The last scholar calls a basic proposition the idea that all elements of a cultural system are interrelated and that alterations in any one component will result in changes of varying degree in all features of the system and in their interrelationships (Trigger 1973, 103). (5) Material/non-material correlation To comprehend the functions of things one needs to rest upon another principle, very specic to archaeology. One may consider it as an aspect or corollary of the principle of systemic order in culture. The link is not evident and yet the specicity to archaeology is extraordinary. The essence of this principle is expressed by Raymond Thompson (1956, 329) as a correlation between a certain set of archaeological material objects and a particular range of sociocultural behaviour. James Deetz (1971, 3) has worded this in the following way: The sine qua non of archaeology is a concern for the relationships between mans visible and measurable modication of his environment and his invisible and less easily measured social and ideological life. Both are regular, patterned, and interrelated. Some archaeologists call these relationships correlates (Hill 1970, 63; Pebbles and Kus 1977; Schiffer 1976, 12f). One can even say more denitely that for every particular case the ideas of people and the corresponding behaviour are recognised as factors determining the form (in broad sense of the word) of things and assemblages directly. This is the principle of objectication of ideas and events in things. Hence the possibility to make judgements on the conditions of things, of their formation and the causes of their formation based on the form of these things. That is, the possibility to apprehend a thing as the source of information on events and ideas. Similar to Hawkes, Malmer expresses this principle very succinctly: physical similarity entails a probability of every other form of similarity, i.e. similarity in respect of time, use, name and environment (1963, 264). (6) Fundamental sufciency of data Many archaeologists have insisted on the idea of the sufciency of data more than any the ideologists of Soviet archaeology. In the acceptance of the incompleteness of the data they saw a view that declared

49

the inability to know about the past. Obviously this would be in contradiction with their inherent epistemological optimism. The idea that we can cognise the real world is one of the main principles of materialist dialectics. In the eyes of most primitive Marxists a disbelief in this looked simply like an attack upon Marxist ideology; Marxism is almighty, and there cannot exist anything that it could not get to know! The other proponent was New Archaeology which had its own epistemological optimism. Everything in culture had to be interdependent with such regularity that after the saved fragments one could completely restore its lost parts. The leading gures among these proponents were Arcikhovskij in the late 20s and Binford in the early 60s. They each advanced a statement on the essential sufciency of the archaeological data for full reconstruction and reliable interpretation. All in all, six fundamental principles are outlined: 1) of determinism in material culture, 2) of universalism, 3) uniformitarianism with its corollary, actualism; 4) the systemic order in culture; 5) of objectication of ideas and events in things (material/non-material correlation), and 6) fundamental completeness of archaeological data. For many archaeologists, whether they are conscious of that or not, the whole system of interpretation in archaeology is based on these principles.

5. PRINCIPLES IN CRITICAL EXAMINATION Enthusiasm for any of these principles has often raised hopes, but warnings immediately followed from the side of sceptical scholars. In an article from the thirties published in French and in English, Tallgren noted a widespread opinion that method adopted by students of the humanities was as certain and as rigid as that of natural sciences. The achievements of Darwinism, he noted, transferred to the humanities, and the course of evolution, made subservient to the laws of history, had to make history as exact as natural sciences. Tallgren inquired whether the formal method is correct when applied to the humanities and he concluded as follows: No, I do not think it is, and probably most archaeologists, who are of the same opinion, are sceptical. Scepticism is a powerful aid to scientic thought ... One must be bold enough

50

Acta Archaeologica
interpretation are called into question and are subjected to criticism. Determinism is avoided, Hodder stated the fact (1991, 9). Hodder opposes determinism as counter to the emphasis on the active role of an individual in culture (1991, 610). (2) Universalism Universalism has become rmly established together with the idea of unity of the human race and culture. This idea typical for Evolutionists and supported by Soviet Marxist archaeologists sounded quite progressive. Evolution was considered as development of the total culture, homogeneous in principle, of the whole of mankind. The subsequent trends Migrationism, Diffusionism, and Ecological conception have undermined this idea. Tallgren stated, No state of culture, no evolutionary stage is or ever has been uniform; there have always been differences. In every culture one discovers rudiments, survivals, archaisms, marginal features, whatever word one may employ ... In every stage of culture there are dialects, if I may use a philological metaphor ... Any reconstruction of a given culture which disregards dialects may lead to serious errors ... It would be a fault not merely in the application of a method but in the method itself (Tallgren 1937, 155f). Walter Taylor made the next step by rejecting comparative manipulations of types and gave paramount importance not to type but the individual artefact and its position in a single particular context. It is this position that allows the functional use of the artefact to be exposed. The broad trend of Contextual Archaeology begins with this appeal. (3) Uniformitarianism and actualism The principle of actualism, which originally gained a foothold in geology, met with opposition from the very beginning. Among ethnoarchaeologists the sceptical opinion of R. Gould has been heard, The less the archaeologist must depend upon uniformitarian assumptions to infer past human behaviour, the more valid his explanations will be (1978, 254). In archaeological interpretation Atkinson considers only conclusions on technology to be reliable because there is no reason to suppose that the laws of physics and chemistry have changed. The inferences are made here within the framework of an invariant

to cast doubt both upon the theories of others and upon ones own, and even upon the foundations of ones own science and its method ... (Tallgren 1937, 153f). None of the principles included in the set presented here have avoided criticism. (1) Determinism Evolutionists, Soviet Marxist sociologists and Neopositivists all took a liking to this. Pjotr N. Tretjakov uttered casually that There are no contingencies in the development of culture (1962, 262). This could not, of course, convince practical archaeologists who run up against strange occurrences, surprises and mysteries at every step in their work. As for Hawkes, he did not see similarities and differences as chance phenomena. Nevertheless he did not interpret nonfortuitous events as a display of social regularity, convergence and the like in the spirit of Evolutionism. He saw in it simply cultural and genetic kinship the evidence of diffusion, migration and common origin. However, the non-fortuitous character of similarities and differences so fundamental for Hawkes was still rejected by R. Lowie in 1912. The comparison of forms can never do more than establish the identity of forms; that such identity is to be explained by genetic relationship is an hypothesis ... (Lowie 1912, 28). Contemporaries of Hawkes developed Lowies harsh scepticism. Some of them, for example Thompson, emphasised that subjectivity of choice in the activity of both ancient masters and todays classiers undermines the effectiveness of laws. Others rejected the truth in the proverb that an exception just conrmed the rule. There are too many exceptions in cultural reality, so many in fact that the concept of law itself caves in (cf. Kbben 1967). Having entered archaeology, the idea of stochastic processes and regularities of probability saved the conception of law, but changed the content of determinism. It has become particularly stronger as regards to mass material, but it has stopped determining every isolated fact and, hence, lost touch with its cause. I do not touch here upon the orientation of archaeology on searching for laws in the framework of anthropology criticism of this enthusiasm is well known, but I do call attention to the point that the availability itself of the laws of cultural process and/ or the possibility to use these laws for archaeological

Metaarchaeology
system. This is manifestly not true, however, of inferences about social organisation or religion, where there is no one-to-one correspondence between intention and action (or between cause and effect), and therefore no possibility of unambiguous inference in the reverse direction. The essential frame of reference here is a system of inference which gives due weight ... to the strength of inference that the individual data will bear, but no more (1975, 176f). Very important was the refusal of ethnographers to consider the currently less developed peoples and social strata as living fossils representing the prehistoric past. The ethnographers have come to understand that these populations had changed greatly by the present time because they exist in new conditions: it is unjust to consider any of the now living groups as our contemporary ancestors (Herskovitz 1949, 71). The proposition of M. Foucault (1968) on the peculiarity of every epoch and even its comparative isolation from other epochs has sharpened the problem even more. A. Leroi-Gourhan (1964) is known for his aspiration to reconstruct primitive religion without adhering to direct ethnographic analogies. He even considered whether they had caused more damage than benet. Guided by the fact that conditions and causes of every cultural phenomenon are multiple and extraordinarily complex, Bruce Trigger contends that the armchair prehistorian, no matter how much general theory he knows, is unable to produce a detailed reconstruction of the course of human prehistory on the basis of what he knows about man at the present time (1973, 104). (4) The Systemic approach Trigger has also stated the principle of systemic order as the interdependency of every element of a system on any other element of it. In such formulation it contradicts the ordinary experience of every person. We often observe evident independence of one or another important component of a cultural system by the behaviour of many others in it. Even where dependence is present, it often proves to be one-sided. The idea of the universal interdependence in culture is very close to Binfords notions. Yet from the very beginning, from his rst theoretical article, Binford came out against recognition of all the elements as

51

equal and comparable. He noted that subsystems, not single elements of a system, interact and inuence each other in a system. He noted also that elements have different signicance depending on their belonging to a subsystem (Binford 1962, 218). In 1971 in a paper presented to the Shefeld Seminar I directed the attention of archaeologists to the structural organisation of a system adopted in the Systemic approach. I emphasised how signicant the position of an element is in this hierarchical structure for determining its role in it, for the inuence it has on the other elements of the system (Klejn 1973, 702f). (5) Objectication of ideas and events The idea of a one-to-one correlation between material elements of culture and non-material ones has lost strength too. Since historical events and essential social divisions of prehistoric peoples dont nd an adequate expression in material remains, Margaret A. Smith (1955, 7) reasons, it cannot be right to try to arrive at a knowledge of them through archaeological interpretation. Michael Schiffer (1976, 11f) declared the idea of material/non-material correlation, the principle recognised by Binford, to be incorrect in archaeology. Binford supposed that social structures had left a direct adequate imprint in material remains. Opposing him Schiffer means that due to the action of natural and cultural forces, material remains have reached us in very much a destroyed condition. The principle I offer, he noted, is that archaeological remains are a distorted reection of a past behavioral system (Schiffer 1976, 12). He did not call into question the availability itself of some imprints in the material culture under archaeological study and he does not deny the correlates with ancient social structures, processes and events. On the contrary, he afrmed their presence, though in distorted condition. However this idea has also come under strong critical re. To Raymond H. Thompson (1956, 330f), Although a correlation between the artefact types and various cultural generalisations is the ultimate goal of an archaeological reconstruction, and consequently considered as possible in principle, it is nevertheless not granted as objective. Resting on analogous relationships in ethnography, it is based on judgements by analogy which are known to be logically

52

Acta Archaeologica
reduced to mutual positioning of components. However, many data are simply absent they have disappeared forever and in archaeological practice we recognise this at every step. Since Winckelmanns time authentic parts of artefacts are to be distinguished from restored, newly created ones. Flinders Petrie left spaces free for early points in his sequence of dates just as there are plenty of blank spots on archaeological maps, as well as hypothetical chains in typological series (historians of language would say forms under asterisk). Childe admitted his beloved short chronology to be just as invalid as the opposite, long one the last appeared however to be better substantiated. Binfords belief in the fundamental reconstructability of cultural components on the strength of their mutual reection makes this principle in turn dependent on another of the systemic order in culture. Thus it is not primary and appears vulnerable from many sides. Despite all the stumbling blocks reconstructions have indeed been carried out. Archaeologists albeit with hesitations and doubts, give credence to them and pin their hope on them overall. The principles I talk about stick implicitly in our heads. Here some discrepancy may be observed between our theory into which we sometimes dive and our everyday practice. In theory we disprove these principles and now and again jeer at them because of their naivety, but in practice we follow them and act as if we justify the use of them by other people, i.e. give credence to the results inferred therefrom. Are there any grounds for this? It seems to me there are. Firstly strict determination of laws, correlations and inferences is, as it turned out, not the only possible determination. Stochastic laws, probabilistic inferences, diffuse or fuzzy sets, incomplete dependences take more and more signicance in our constructions. An inference adopted for a totality does not necessarily cover each member of the totality. A law incorrect with respect to a single part may be true for all the totality. This impels us to decide with circumspection if it is possible to rely on the fundamental principles I have enumerated above. It is important to dene the kind of the phenomena we are going to apply these principles to. Secondly, the principles are none other than the

unnecessary, and is permeated with subjectivity. ... [The] archaeologist injects a subjective element into his inferential reconstruction ..., unavoidably and not only once. Karl Heider called the ethnographic validity of this correlation into question. Such conclusions of archaeologists can only be tested against the facts of ethnographic cultures. Although for the most part they can never be proven universally valid, only a few counterexamples can throw their usefulness into question. Heider provides such counter-examples, one of them in the statement of Thukidides who, long before the birth of archaeology, reasoned as if he foresaw the risky discourses of archaeologists. He remarked that in comparison with magnicent Athens, the military Sparta looked just like a large village, but it would be a mistake in future, looking at its scarce ruins, to draw the conclusion that Athens dominated Sparta politically. While exploring contemporary New Guinea aborigines Heider observed an unexpected picture. He has found that Between the formal and functional typologies of the aborigines, as well as in other aspects of culture, there are a number of discrepancies which would mislead the archaeologist (Heider 1967, 55). Hodder has explained this theoretically: it is ideas, beliefs and meanings which interpose themselves between people and things. How burial reects society clearly depends on attitudes to death (Hodder, 1991, 3). Holding an archaeological conference on an analogous problem, Carl-Axel Moberg (1981, A12f) called both the conference itself and his introducing paper to it Similar nds? Similar interpretations?, all under question marks. He sums up in this work: The thematic question in the title is, in reality, the question whether archaeology is feasible at all. Actually, an answer in the afrmative, even if often entirely inexplicit and unconscious, underlies all archaeology: Similar nds do indicate similar interpretations. But we know so well (or ought to know) that in many concrete cases the reply has to be negative more often than not. In this special research situation, similar nds are poor indicators for similar interpretations. Is archaeology then feasible at all?! (6) Fundamental sufciency of data If the data were sufcient, reconstruction could be

Metaarchaeology
main, most inuential, governing laws. But are they universal? Edward Tylor wrote: The tendency of modern inquiry is more and more towards the conclusion that if law is anywhere it is everywhere (as cited by Kbben 1967, 3). Binford reiterated the teaching of Leslie White with the aphorism: Laws are timeless and spaceless (Binford 1972, 8). However the inclination to formulate universal laws has since greatly decreased and belief has grown that the more universal a law is, the more trivial or unreliable it is (Trigger 1973, 102). Well known is the bitter irony of one of the leaders of New Archaeology who admitted that after lots of trouble many of his associates produced rather trivial and petty Mickey Mouse laws (Flannery 1973, 51). Many students have seen the way to save the conception of law by limiting it. Arguing with Kbben I contended that a law would not be banal and would get out of difculties with its exceptions if the limits of its use were indicated (Klejn 1972a). Thirdly, a principle does not just cease from acting outside the limits of its applicability. We may nd the opposite principle is acting there. In general this is clear: a principle formulates some elementary truth, which has no gradations but expresses polar relations. So if one thing is not true, then the opposite is true.

53

material are comprehensible. Heiders ideas are also contrary to the belief that ethnographic analogies are reasonable. Unfortunately for the archaeological process, writes Heider (1967, 52), cultures are generally quite unreasonable. In his concluding address to the Shefeld Seminar the ethnologist Edmund Leach (1973, 764) warned archaeologists: The proper analogy for human behaviour is not natural law of a physical kind but a game of chess. The eld of play and the rules of the game are laid out in advance but the way the game is played out is unpredictable. Authoritative archaeologists, especially in Britain, were prone to think in this way. Even more so in Germany, where as Ulrich Fischer (1987, 184) has written, cultural laws dont exist. (2) Individualisation (particularism) The principle of universalism has its opposite too; the belief in the unique character of every phenomenon in culture, in a nutshell, the principle of individualisation. In implicit form it was already inherent in the passion with which the Diffusionists opposed particularism (the other designation of the principle) to generalisations of Evolutionists (Buettner-Janusch 1957). Now this principle is exposed and clearly formulated: each archaeological object and situation is unique (Chang 1967a, 230). (3) Historicism What is counter to the principle of uniformitarianism, for which all epochs are equal? It is of course the principle of historicism that demands that all phenomena in development are taken into account and substantial differentiation of epochs is made. This principle arose even before Evolutionism in teachings of the catastrophe and progress in history. It is the fourth initial proposition of materialist philosophy in Jason Smiths book. On the whole Marxism has very much propagated it as documenting the law-dependent character of changes which lead to Communism. Historicism is not alien to New Archaeology oriented to the study of cultural process, of sociodynamics. The acknowledgement of development alone does not require the identication of laws that acted in different epochs. Nevertheless a very narrow principle of historicism existed, i.e. reduced to the demand for perceiving sharp qualitative differences of epochs. It

6. THE OPPOSITE PRINCIPLES As soon as we try to formulate the opposite idea as an independent principle, its corollaries and theorems will immediately spring up from it, the neighbouring principles with their inferred theorems join them, and all of this having turned into a complex system of propositions, into theory, will lose all the former simplicity and banal look. What is more we will most likely nd that we already know these principles and having existed for a long time they are quite respectable. They can be examined as follows. (1) Indeterminism Contrasting determinism is the principle of indeterminism. This principle is suggested both in history and archaeology. Some critics of determinism came to such an opposition of determinism. The conviction of Heider is contrary to Hawkes belief that the aims of prehistoric peoples in their production of cultural

54

Acta Archaeologica
tive than direct access objects in a living culture. Between the two is the destructive activity multiplied by time, a point that was stressed by Dunnell. The recognition of the initial incompleteness of data is very important for archaeology. It leads to the necessity of interpretation and the involvement of non-archaeological information, to the interdisciplinary synthesis in studying the historic and prehistoric past. Sir Mortimer Wheeler (1952, 180f) expressed this very graphically. He noted that the archaeologist will nd the tub and will completely miss Diogenes. He will write profound papers on the typology of tubs; he will classify tubs in categories A, B and C; he will discover a Tub-bearing Folk and plot their wanderings upon geographical and geophysical maps ... Only, he has overlooked, and could not help overlooking, the one signicant thing about that Tub, namely, that it sheltered the eminent Cynic and symbolized his philosophy for all time. Hawkes made the cognisability of ancient purposes one of his two principles, yet three years earlier he had introduced essential limitations. At that time he had built his well known ladder of incomprehensibility. It consisted of different cultural spheres gradually more inaccessible to the archaeological cognition according to the growing incompleteness of the sources (Hawkes 1954). He is one of several who formulated this principle for archaeology.

was present in Marxism and in some other teachings close to archaeology in the teachings of Foucault for instance. (4) Irregularity It is not difcult to see that opposite to the principle of systemic order is the principle of irregularity, of disorder. More difcult is to nd this principle realised in research practice. The question would apparently not be contained in what existed before the principle of systemic order was introduced but in the reaction of its opponents. Very seldom are its uses clear but it is implicitly contained in the initial ideas of archaeologists who do not believe in the wisdom of the Systemic approach and the boundlessness of cognition. Either they do not consider that culture is an ordered enough system to be worth serving as a base for further conclusions, or they recognise living culture as such a system but refuse to do consider archaeological material as such a system for it is too fragmented and incomplete. A system can only be introduced into such material from outside it will be a subjective creation of an archaeologist, just the position of J. O. Brew and J. Ford (Brew 1946; Ford 1954b; 1962 a,b). (5) Polysemism In the rst row of principles the fth one is the principle stating that material components of culture must correlate with its non-material ones and with events of history. This principle has its opposite in the principle of polysemism of archaeological facts. I defended the idea of polysemism in my paper to the Shefeld Seminar of 1971 (Klejn 1973; see also 1978b, 48). In his Theoretical archaeology (1979) Gardin equates interpretation to the logical paraphrase establishing the likeness of monuments dispersed in space and time, and he advises not to forget that this paraphrase is almost always one of many, and the less we take this simple fact into account, the more plausible such paraphrases will seem to be (Gardin 1983, 154). Now Shanks and Tilley (1987, 115) note that material culture is unreducibly polysemic and Stutt and Shennan (1990, 767) proceed from the view. (6) Initial incompleteness of data By denition archaeological objects are less informa-

Thus, this is the other group of initial principles of archaeology, a group that is quite opposite to that one which was investigated above. It must be admitted that these different principles are also realistic and reasonable, by no means without foundations and in any case useful. They also form an integral system with its derivatives corollaries, theorems etc., and all of them rank high in archaeology. So the demand to concentrate the study on the context (the principle of contextuality) is guided by the principles of indeterminism and of individualisation. A further support of this principle of contextuality comes from the functions of things. The context principle was instilled into archaeology by followers of W. Taylor, in particular by Chang and Deetz, and in a somewhat different way by Trigger.

Metaarchaeology
7. DIALECTICS OF PRINCIPLES In logical space the fundamental principles of archaeology are arranged in pairs like the quarrelling Gods before the Battle of Gods in Homers Iliad the principles are in opposition to each other in every pair (Fig. 9). This reminds me of a maxim by Niels Bohr. When visiting Moscow in 1961, Bohr said that the most fundamental truths were so profound that quite the opposite ones were also true! Having discovered such truths, he explained, a science enters its best period (Fainberg 1988, 30). Malmer pointed out something similar in prehistory: To some scholar a vague hint on political conditions may seem more important than a lot of information on daily life but the opposite view is equally reliable (Malmer, 1963, 249). Zbigniew Kobylinski pointed out the opposite views of M. Schiffer and R. A. Gould, both stated in the same year of 1978, with regards to the problem of uniformitarianism. Schiffer calls to formulate universal laws as the same in both prehistoric and industrial societies, while Gould denies the existence of universal laws because the research concerns unique combinations of factors. Instead Gould insists on siteoriented studies, and both of them seem to be right (Kobylinski, 1981, 30f, 46f). In an interesting article en titled Axioms in archaeology A. B. Johansen (1984, 36) writes that having revealed the fundamental axioms and having built a system of knowledge on this basis it would be reasonable to derive thereafter a contrastive alternative, could not the prehistoric people be counting and non-counting at one and the same time? At one and the same time hating and loving? Reacting both constructively and destructively onto outer challenges?. However, as applied to physics this dialectic of principles remains somewhere in the deepest philosophical comprehension of science. Thus inside of every physical discipline physicists manage to construct its basis without intrinsic contradictions. In archaeology the picture is different. Universal laws are inessential, while essential laws not universal. Data are always terribly incomplete, and meaning is hidden which must be deciphered. So it is impossible to avoid contradictions in the basis of archaeology. The Battle of Gods takes place right in the discipline itself. When Gods are ghting, defeats and victories are illusory no corpses are left on the battleeld, wounds

55

Fig. 9. Dialectics of fundamental principles in archaeology.

recover and all as though nothing had happened. Only mortals involved in the conict lose their lives. What can a poor archaeologist do by watching Gods ghting, when far from being God? The decision suggests itself which if suggested for a real ght would be deemed cynical: ignore one side and move on. It would evidently be the most stupid way out and ruin would be inevitable. One has to adjust to the situation and this means estimating which of the sides is stronger in the given situation, each time anew. One must also rely on intuition, on ones own sense of measure. Apparently this matter contains the fundamental methodological difference between archaeology as an applied science, besides studying culture, and physics as a fundamental science studying natural phenomena. David Clarke seemed to feel this irremovable difference when he declared that the conversion of archaeology into a science was not his ideal. He merely called for the formation of an analytical discipline, i.e. an objective discipline acting strictly and explicitly. However, does not the result obtained here undermine the very possibility for archaeology to be analytical? For Clarke as well as for all who derived inspiration from analytical philosophy the ideal was undoubtedly an analytical machine. By the phrase analytical machine is meant the logical research structure, elaborating the information explicitly, unambiguously and equally, always and everywhere. This ideal has been realised in the modern computer, by offering some researchers the possibility of con-

56

Acta Archaeologica
man the multiple devices processing the information (the neurones) co-operate in parallel processing. On a macro level this is expressed in that the brain is split in two hemispheres. These are relatively independent of each other and able to think differently, sometimes even generating polar guide lines. In the most lower portion of the brain, the connection and co-ordination of the two cerebral hemispheres occurs, located in the brain stem. This means that the duality or duplicity of the initial principles which is discovered in the very foundations of archaeology does not border on schizophrenia. It makes this part of archaeological interpretation closer to human thinking, partly intuitive, but also plastic and very rich in ideas, and less like the strict and rigid logic of computer. I consider here archaeological interpretation as a purely logical procedure segregated from the executor and hence I compare it with human thinking as something different. Perhaps a system of two parallel computers (may be with a third one for co-ordination) would be able to imitate the interpretative thinking of an archaeologist with good approximation to reality. When speaking of parallel computers I am simply trying to catch the essence of the matter (professionals in computer science would prefer more exact and cumbersome expressions), but this seems to satisfy the task at present. The idea of parallel computers is active in the science of articial intelligence, but with the aim to reach more speed in calculations (Deering 1985; Feldman 1985; Hinton 1985). Apparently the task to approximate structurally the modelling machine system of human thinking should nd its solution this way too. Hewitt (1985) works with such ideas in the open systems of many computers with a different logic, based on the hypothesis of contradicting axioms. Some qualities a modelling system must have preference over the brain of a researcher, otherwise what use would it have? The preference may be attained by means of strengthening the relative role and capacities of the third, uniting computer. In the human brain the choice between contradictory guiding lines is made to a great extent under inuence of irrational stimuli, but in a way that is individual and not explicit. Computers may help in avoiding these factors. Thus, the axiomatisation of archaeology is possible but the net that is realised appears very complicated and contains inner contradictions.

tructing the articial intelligence of an archaeologist. Thus they hope to make a computer program able to solve the tasks of interpretation. The rst intention of analytical archaeologists is, of course, aimed at rectilinear formal logic issuing from a non-controversial system of principles (Watson et al. 1971; Salmon 1982b). An almost schizophrenic split into two is revealed however in the very logic-foundation of the discipline. This is enough to frighten anyone who hopes for vigorousness and unambiguity, and for a non-contradictory basis of inferences. There are suggestions how to avoid this difculty. One can construct the program of interpretation as an expert system (Ennals and Brough 1982; Gardin et al. 1987; Gallay 1989) like those already existing in medicine or geology. However, such a system would not differ much from simple generalisation of existing subjective estimations. It would only make their search easier. One can construct the articial intelligence on the basis of hypertext (Stutt 1988; Stutt and Shennan 1992) as simulation of the usual arguing of hypotheses which would however be supplied with explicit argumentation. Nevertheless a scholar does not get rid of the feeling of dissatisfaction unless the logic is traced to the very fundamentals. Yet if one is going to model the discourse of an archaeologist at interpretation, then the task is much more complex than it may be imagined. Nevertheless the conclusion is not so pessimistic as it may seem. In one of the rst collective volumes on the application of computers in archaeology M. Doran (1970) formulated the differences between machine thinking from the usual thinking of an researcher archaeologist. He had in mind how one should reconstruct and discipline the archaeologists thinking in order it to make it easier for conversion into formulations accessible to the machines comprehension. Indeed, it manipulates only mathematical expressions and thus needs exact denitions, unambiguous orders, in full formalisation. The brilliant Russian mathematician and linguist V. V. Nalimov at the end of his life wrote a strange book entitled In search of other meanings (1993). To him the thinking of a machine is uni-directional and accurately segmented, whereas human thinking is diffuse, uent, and not without inherent contradictions, but very mobile and plastic. This qualitative difference is conditioned I suppose by the fact that in the brain of a hu-

Metaarchaeology PART IV. ARCHAEOLOGICAL THEORY

57

6. Empiricism in archaeology
1. READING THE PAST, ORIGINAL CAST Ian Hodder is known as a theoretician who advanced against the theoretical enthusiasm of New Archaeology, and against the determinism of laws and belief in objectivity. In his book Reading the past he opposes his simple reading of material culture (with multiple interpretion of material culture from understanding contexts) with Binfords reading off human patterning from material culture patterning (with the help of deterministic correlation) (Hodder 1991, 4). He considered it appropriate to explain that his reactionary views do not return us to simple empirical description and inductive generalisation. He insisted that archaeology can be distinguished from antiquarianism by its concern with the context of material objects. It seems to him that archaeologists can incorporate inductive methods in building up from contextual associations ... (1991, 190f). But if theory is dismissed as depending on biases and if at the same time uniqueness of events and cultural situations is propagated, how can contexts save archaeology from being reduced to inductive generalisations and guesses on facts, let alone in contexts? Reading the past, reading the monuments without universal clue to their symbolic meanings is no innovation, it is something our empiricist grandfathers also tried to do. The use of contexts is not new either Walter Taylor, K. C. Chang and settlement archaeology used this approach. Does the new contextualism (Hodder 1991, XIV) differ considerably from the old one? It initially came as a shock to me that post-processual archaeology was having little impact on data acquisition (Hodder 1992, 171). People continue to collect facts as before. It is a remarkable situation practical archaeologists are left in who are devoid of their recent belief in laws, analogies and objectivity. Nearly three decades ago I showed what the impact was of the ideas of British Hypersceptics, with their belief in uniqueness and their dismissal of theories, on the practice of archaeology. They were brilliant theoreticians (Daniel, Hawkes, Piggott), but I saw them (Klejn 1972c) under the device that he who sows Hyperscepticism, harvests empiricism. The pupils of Piggott appeared to belong to this trend, and Hodder is in the same vein. What will practical archaeologists do who follow his principles? After much criticism has been directed to post-processual works, young archaeologists are already left without any theory and feel happy in such a state; Archaeology does not need another Grand Theory (Campbell 1994, 141). They are ready to collect facts quietly. Really, the two decades of postprocessualism have re-enforced the empiricism that never died in archaeology. Since the time of Francis Bacon the idea of experience entered science as the master: purely philosophical a priori schemes broke away; facts became the only legal source of positive knowledge, and the inductive method became the main principle of science. At least until the late 19th century empiricism was a very respectable way of research. However, later on its authority often rose up in one country or branch of science, then in another, especially where the dominating theory plummeted into deep crisis. In archaeology this way of thinking had especially steadfast practitioners, largely because archaeological material is so demonstrative and large-scale. Sophus Mller worked in just this way and regarded it as universal; gradual generalisation leads an investigator from the material to the inference. The proper method of archaeology is ... the safe induction (Mller 1898, 298f). For a long time classical archaeology has formed the safest accommodation for empiricism, more so than prehistoric archaeology, and both retain this way of thinking more in Germany and Russia than among British or French archaeologists. Having said this, the outstanding British archaeologist Flinders Petrie even wrote a book called Inductive methodology in 1877. In pre-Revolutionary Russia even the most deep

58

Acta Archaeologica
merely shows that empiricism has had its time. Nevertheless it pops out now and again more or less openly, not as an entire movement but in a scattered way. Sir Mortimer Wheeler looked with open eyes and forthright conscience at the humanistic and empiricist qualities of his profession. He concluded his book Archaeology from the earth (Wheeler 1954/1956, 230) in clear diction: As archaeologists, then, we are at the same time collectors and interpreters ... Let it be agreed that the two words archaeologist and antiquary shall in future be exactly synonymous ....

thinking of archaeologists Vasilij Gorodcov was convinced that facts speak for themselves. In the result of methodical cross-section or excavation of the earth, he declared, as exact and coherent a reading of sites must appear as the reading of a usual book. For this process no subjective intervention of an investigator is needed since every phenomenon, every thing, must speak for themselves, like characters and words written on sheets of paper tell the reader everything they were ordered to tell (1908, 11). Soon after the Civil war the historian Sergey Zhebelev, at that time one of the leaders of Soviet archaeology, wrote a textbook for archaeologists. In this textbook he reduced archaeological methodology (for the sake of irony he placed these words in quotation marks) to description, comparison and generalisation of monuments. Therefore, in his opinion, one needs only diligence, erudition and experience. The experienced archaeologist needs only to look at monuments in order to read them, i.e. to understand them. The skill to read monuments in order to interpret them constitutes the subject matter of archaeological hermeneutics. The latter must be based on rules inferred empirically and not by metaphysical ways (1923, 132f). Even the vocabulary is the same as that of the Post-Processualists. Ernst Buschor, the authoritative classical archaeologist whose introduction opened the model handbook in classical archaeology in the editions of 1939 and 1969, could say nothing commonly valid on archaeological method. There was no general method for him: There are so many archaeological methods as monuments and groups of monuments under the consideration, and so many methods as research personalities. This shows good correspondence to Hodders irreducible meanings unique to each context. Nevertheless, Buschor continued, one initial point is unmistakable; looking at the object. The archaeological method, even though its application is executed intensively in an apparatus of thinking, always issues from eye experience, from seeing (Buschor 1969, 5f). So simply reading the monuments and interpreting them by free comparison with the like, and then concluding from an intuitive understanding of contexts was one of the features of empiricism. The fact that it is hidden under the cover-up of theoretical discourse

2. OVERCOMING EMPIRICISM In archaeology the rst opposition to rise against the common belief in inductivist methodology was young Soviet archaeology, and this movement was due to its Marxist claims. Marxism dictates that it stems from theory, namely from Marxist theory of Historical Materialism. So archaeologists should issue from Marxist theoretical dogmas and not from facts, however much in contradiction to facts these dogmas may be (remember the matriarchy thesis). Despite all difculties facts had to be subordinated and could not contradict theory. If they did, woe betide them and the authors. One of the gures who initiated the outt of Soviet archaeology with Marxism was Vladislav Ravdonikas. When he attacked traditional archaeology with sharp criticism he started with the impeachment in empiricism: It was still not long ago that it was considered enough to make archaeological excavations and to print a report on them, i.e. simply to establish a fact, and the scientic work would be fullled and the author can rest on his laurels, especially if there was a chronology in the report, together with a description of usual life, and some notorious ethnic attribution .... Ravdonikas condemned the empirical slavery of thinking, and pointing to the Holy Scriptures of Marxism he concluded: Now we have no right to be empiricists but unfortunately empiricism holds many of us in its captivity (1930f 51f). Paradoxically as it may be, the ght of Soviet dogmatists to support theory corresponded with the general development of sciences. Ten years later Clyde Kluchhohn with similarly clear eagerness advanced against the empiricist outt of American anthro-

Metaarchaeology
pology, and he included archaeologists. In my observation, he remarked, the greater number of anthropologists still feel that theorising is what you do when you are too lazy, or too impatient, or too much of an arm-chair person to go out and get the facts (1940, 46). This especially concerns that part of American anthropologists who dig and are oriented to history, i.e. Americanist archaeologists. If they use the word theory at all, they tend to use it as a pejorative synonym for speculation (1940, 44). In Kluckhohns judgement archaeologists are but slightly reformed antiquarians (p. 40), and he concludes that they remain on the intellectual level of stamp collecting (p. 45). Having in mind their belief in selfobviousness of facts the critic states a methodological and theoretical naivety of his archaeological colleagues (p. 51). He repeats words of the sociologist Parsons when saying: The facts do not speak for themselves; they have to be cross-examined (p. 42). Nearly a decade after Kluckhohn, in the late 40s his pupil Walter Taylor (1948) went for American archaeologists because of their empiricism with more developed, sarcastic and individually addressed criticism. Taylor noted that for American archaeologists history is nothing but hypotheses. As an example he refers to the passing but very meaningful remark of J. Thompson (1944, 23; Taylor 1948, 59) who presented his monograph in the series Theoretical approaches to problems as follows: The fact that the paper is published in the present series is sufcient indication that the case is not considered to have been completely proved. Taylor produced a reproach of his fellow specialists that they considered theorising as something not worthy of proper science, and that they hang on to the untenable position to hold on facts which means to avoid inference, hypothesis and proof (1948, 113). He refuted their outt wait-until-all-thefacts-are-collected (p. 61) and concluded his critical analysis with the words: Americanist archaeology is not in a healthy state. Its metabolism has gone awry. It is wasting and not assimilating its foodstuffs (p. 92). Two decades later Lewis Binford continued this criticism. He turned against methodology reducing the actions of an archaeologist to elementary ones to generalising about the facts he uncovers

59

(Binford 1968, 21). Indeed, an empirical generalisation of data no matter how accurate it is is never an explanation for the data (p. 15). His disciple James Hill a few years later (Hill 1972) analysed in detail differences between the inductive way and the opposite, deductive procedure, he stressed preferences of the latter and made consequent inferences therefrom. So New Archaeology began its triumphant raid through the Western world; everywhere it condemned empiricism and aggrandised theory. In other circles of Western archaeology the understanding started to take roots that empiricism was detrimental. Still during the time of the World War II the French historian Marc Bloch wrote: ... Texts or archaeological nds, seemingly very clear and accessible, speak merely when you know how to ask them. The int tools in the Somme alluviums were abundant both before Boucher de Perthes and after him. Yet there was no man who knew how to ask and there was no prehistoric times (1973, 38). In Germany Rafael Uslar published an article (1955) under the drastic title On the advantage of speculative consideration of prehistoric monuments. In the VII International Congress of Pre- and Protohistoric Sciences in Prague in 1966 Geza RohanCsermak from Paris presented a paper that was later published against empiricism in 1971. The axiom of the empirical approach reads: Primo observare, deinde intelligere, postremo philosophare (rst observe, then understand, nally philosophise). As soon as observation is not possible without previous setting of rules and concepts for sampling, perceiving and description, this author adds to the above axiom an initial phrase: Primo praejudicare (rst foreordain, or predestine). According to Durkheims designation the author calls his own outt a priorism. One can doubt if his choice of the terms is lucky (predestination, prejudices, preconceived ideas, and biases are hardly things archaeology badly needs) but the idea itself is reasonable. With his a priorism Rohan-Csermak partly joined the assault on New Archaeology, which incidentally also likes formulations of ostentatious sharpening. The struggle of New Archaeology against empiricism is weakened by one circumstance. Empiricism remains in neopositivist methodology too, i.e. in the methodology of New Archaeology itself. The same

60

Acta Archaeologica
and not a few of the younger generation. Many of them felt (and many still feel) that if we only wait until all the facts are in they will speak for themselves. Only 10 percent of archaeologists belong to the processual school. The rest aim their re freely at both camps, but at least some of them are more on the traditional side. This was the impression of a participant in the movement on the side of the processual school. It is still a favourable estimation for processualists. Later, when Ezra Zubrow (1980) looked at the quotation statistics of the most creative years of 1968 72 of New Archaeology, it appeared that the most cited authors were empiricists Zeuner, Hole, Braidwood (the lively opponent of Binford) and Helbk, and only then do Binford and Flannery follow, while the whole of New Archaeology took merely 212%! True, in the years 1977 and 1978 the most cited authors were already the theoreticians Taylor, D. Clarke, Hill, Flannery and Binford. Yet this was already the time when the domination began shifting from USA to Britain and from processualists to post-processualists. The place of theory in Great Britains academy for the period of the highest peak of post-processualist success is described by post-processualist Julian Thomas (1995, 349f). As he notes, in the United Kingdom there are more than 200 archaeologists who hold appointments at universities. Over 100 of these scholars have a traditional outlook they are largely concerned with the extraction, description, classication and compilation of archaeological evidence. They believe that it is their task to accumulate as exhaustive and well-documented a record of their chosen material as possible, on the understanding that when complete this record will constitute a selfevident account of past human activity. These atheoretical archaeologists are contrasted to 85 theoretically inclined archaeologists with determinable views. Among them somewhat less than 40 individuals are listed whose approach to archaeology is broadly processual (inuenced by Binford, Clarke and Higgs). A further group of around 30 persons is indicated who are strongly theoretical but belong to several trends (Marxism, feminism etc.) that cannot be included strictly into mainstream post-processualism. Merely fewer than 15 persons are true post-

point was stated by A. Paubicka (1973) with regard to Rohan-Csermak. Nevertheless more than two decades went by under the title of New Archaeology which strongly and inuentially advanced against empiricism. Then post-processual archaeology broke the domination of New Archaeology and, as was shown, built the ground for revival of the empiricist approach, although it managed to do so under the banner of theory.

3. THEORETICAL GHETTO In the 1970s theoretical archaeology as a special branch of archaeology was formed. It does not mean, however, that archaeology in general changed its sentiment towards theoretical studies. The anti-theoretical mood of German archaeology is well known (Hrke 1991; 1995), and is even somewhat exaggerated (on this issue see Klejn 1995b). Yet look at France. Hodder and Tilley one evening in 1986 put a pressing question to some well-known French archaeologists over an excellent beer. French archaeologists are fortunate enough to speak the same language and breathe the same air as Althusser, Bourdieu, Foucault, Levi-Strauss, so why do these archaeologists use so little theory? And French archaeologists (Gleziou et al. 1991, 91) agree: To describe the position of theoretical archaeology in France seems indeed equivalent to posing the question: Why is there no theoretical archaeology in France? . They conclude that: The best current French examples suggest that interpretative work will remain very empirically grounded in an interaction between the data on the one hand and analogy, experimentation and analytical techniques on the other (Gleziou et al. 1991, 119). These archaeologists, known as the most theoretically oriented among French archaeologists, dropped such comments as, The irony is that in actual fact the most conservative position is no doubt the most promising (Gleziou et al. 1991, 105). When New Archaeology was on its climb in the USA, holding up high the banner of theory, Kent Flannery (1972, 103, 106f) estimated that perhaps 60 percent of all currently ambulatory American archaeologists belong to the traditional, normative school. This includes most of the establishment

Metaarchaeology
processualists professing hermeneutic interpretation, contextualism and critical theory. Of course Thomas regards only the last little group as the true theorists, while to me post-procesualism, as I have noted already, is a path to empiricism. He concludes: Academic archaeology in Britain is overwhelmingly empiricist in tenor. Thus some 70 theoretically oriented archaeologists oppose the 100 empiricists and 15 hidden empiricists plus all the archaeologists working outside universities and this is the world stronghold of theoretical archaeology! As Peter Ucko (1995, 2) notes, theoretically inclined archaeologists are in the minority in all countries, and are thus forced to form a ghetto within their discipline. This picture of theoretical archaeologists applies in Russia too. Although Marxism demands theory and negates empiricism, after the violated collapse of Stadialist theory in 1950, any enthusiasm Soviet archaeologists had for theory evaporated and Soviet archaeology saw an atheoretical period for the following decades with the restored domination of empiricism.

61

evident facts, on facts speaking for themselves, he demonstrates his stubbornness. When he invokes to reduce archaeology to facts and in the essence limits the aim of archaeology with establishing of facts, i.e. disposes archaeology of theory, he essentially weakens possibilities to cognise the facts. (2) Belief in pure facts Many veteran archaeologists still crave for pure facts. They are convinced, like K. Godovsky in Poland (1962, 81f), that the researcher sees objectively existent reality in archaeological records, that in archaeology the main approach is induction and that in its basis pure sources should reside, not already interpreted and bearing on them a load of views and conceptions. Deduction is permitted by this author solely in attempts to reveal as much as possible the gaps in our sources or unclear phenomena and facts. (3) Revealing types The simplest method to reveal archaeological types is with statistics and correlation, a method discovered by Binfords teacher Albert Spaulding (1953, 1960) which is widely disseminated. Spauldings outt supposes that our types, properly speaking, are contained ready-made in the material. The Marxist archaeological school, Sher stated, has developed a notion on types of artefacts, objects and monuments as of real classes between which the differences are produced by chronological, ethnic, local, technological and other natural-historical conditions and not by the opinion of a researcher (Sher 1970, 13, note 7). (4) Description instead of thinking This may be manifested in several attitudes, for instance, in preferring ostensive denitions. Cristopher Hawkes (1973, 177) wrote in the journal Antiquity that If cultures, as a viable concept, are indeed to be recognised, they can be dened only by total enumeration of their parts. (5) Induction as the only valid way Domination of empiricism is expressed in that a simple generalisation of facts became in practice to be taken as the only way of study for a reliable guaranty against unrealistic ideas. L. N. Gumilev (1989, 30) refers to the example of natural sciences: instead

4. SYMPTOMS OF EMPIRICISM Thus, there are deep roots to empiricist methodology in the post-Soviet landscape, while both the positivist methodological ideas of New Archaeology and the activity of post-processual archaeology formed conditions for retaining empiricism in the West. What are the symptoms of this now anomalous narrowness of thinking in archaeology? (1) All hopes on fact Gorodcovs belief that things speak for themselves like characters and words and an archaeologists belief in simply reading monuments like normal books still rules archaeology openly or in a guise form. As Hrke (1995, 50) remarks, A. E. van Giffen, the founder of modern Netherlands archaeology between the wars, liked to use the German motto: Die Interpretation schwankt, die Tatsachen bleiben (The interpretation changes, the facts remain). When P. Courbin (1982/1988) fought against simplications of New Archaeology he was mostly right in the exchange. But when he hastily insists on raw facts, on neutral, objective, fundamental,

62

Acta Archaeologica
of purest factography against empty theorising without any connection with facts. However while inductivists eject theory completely from science, their opponents do not think it possible to manage without facts. Even the New Archaeologists who exaggerate their worship of theory, use facts which they often borrow from empiricists, as Courbin (1982, 162ff) showed. Binford himself admits that The reader may well nd some incompatibility in my insistence on inductively formulated research problems and my advocacy of an essentially inductive data collection procedure based on sampling theory. I, too, nd some incompatibility (1972, 133). (7) Generalisation as the road to theory Going beyond theory considered as speculation, empirical generalisation, especially broad coverage of facts, came to advance as theoretical study in contrast to description of facts. In our current archaeological use the words theoretical generalisation fused to a phraseological unit, a common cliche. If there is a generalisation, a summing up, many are inclined to suppose that it is already no longer at the empirical level of study. Even if one adopts theory, it is considered in an inductivist vein as extracting regularities from the generalisation of facts. For instance, Martin JaguttisEmden (1977) builds a symmetrical scheme in which artefacts are xed with protocol sentences, some protocol sentences united form a hypothesis relating to layer or site, and a compound of such hypotheses relates to the whole culture and is called theory. It follows that theory and hypothesis are distinguished from fact-xing sentences merely by their coverage, by a higher stage of generalisation. Evidently for this author one fact is a fact, two similar facts form coincidence, and three are adequate for regularity, and this means law, and law lies in the basis of theory. So there it is, three generalised facts make theory. (8) Voracious hunt for facts It is supposed that the more facts are collected, the more valid are inferences, and in the case of the generalisation held as theoretical, the more facts that are accumulated, the deeper the theory is. Hence the silently approved uncertainty of aims in many project plans. Indeed, in any case something will be exca-

of philosophical postulate the natural scientists apply empirical generalisation having after V. I. Vernadsky validity equal to observed fact. It was meant that in empirical generalisation petty facts are fusing together and instead of an array of small facts one large fact appears. We can leave this without further analysis with the truth expressed by L. von Bertalanffy (1969, 53): a simple junction of empirical data though standing for a certain progress does not properly constitute science . We can also leave aside the idea that empirical generalisation is not a simple sum of observed facts and is not equal to the individual result of observation (recall here the problem of choice and sampling). (6) Data vs. speculation The same empiricist outt is expressed in the strive to confront solid empirical studies of material with all sorts of theory anywhere it is possible as speculation, as empty philosophising. Binford characterises the thinking of his teacher James Grifn as antitheoretical. To Grifn, he writes, theory was to be equated with speculation, and one only did that if there were no data. If data were available, it was clear what one did: one summarised the data and the selfevident units of meaning were historically synthesised. In Grifns mind there was no question about what data meant or what they were telling us about the past (Binford 1972, 3). In the USA many remain on the side of Grifn. The situation in modern Germany is characterised by Heinrich Hrke (1995, 48) as follows: The word theory, to German ears, sounds airy-fairy, it implies speculation without foundation (i.e. without evidence), and it seems to exclude practicality which is considered highly desirable. To call somebody a theoretician, or to call something theory, invariably carries derogatory undertones. In order to avoid these connotations, German scholars outside the natural sciences avoid the term theory as far as possible, and if they have to, they put it under the heading of methodology. Methodik has a solid ring; it sounds practical, systematic, goal-oriented and efcient. Hrke refers to the textbook by Ziegert as an example. For such a position the notion itself is characteristic; these people view the argument between empiricists and practitioners of theory as the confrontation

Metaarchaeology
vated, new facts obtained, and they will be possibly generalised or included into previous generalisations, with broadening and strengthening of the inferences. As the saying goes, one more extra brick will be put into the building of the discipline. The hunters for the facts are ignorant that not all facts allow the inference to broaden, that often the simple multiplying of the facts does not strengthen the grounds, and that the extra brick can appear extra indeed, i.e. a waste. In this case excavations risk appearing to dissipate money and ruining the monument if, of course, it is not a salvage excavation. The simple accumulation of the same evidences brings nothing new. When the borders of the necessary representative sample are overridden, the further accumulation of the same facts is useless. (9) Horror vacui As Kent Flannery (1972, 106f) noted, the traditional archaeologists were often deathly afraid of being wrong ... They spoke in awe of the incompleteness of the archaeological record and of the irresponsibility of speculating on scanty data. Somehow they seem to feel that if they could get together a few more potsherds, a few more projectile points to a few more architectural details, their conclusions would be unshakeable. The exceptional care or abstinence from inferences came to be regarded as prudent. The archaeologist is advised to wait until the sufcient quantity of facts is provided. S. N. Zamjatnin (1951, 92) emphasised that in archaeology the accumulation of facts usually [go] slowly, and one should nd forces in oneself to refrain from temptation of solving and sometimes even from putting the question if the sources did not allow this. (10) Unfeasible desire of completeness But how many facts are necessary in order that the setting of the question and the assumed conclusion be permissible? One cannot nd criteria of sufciency if one admits incompleteness of data in general. Therefore the demand emerges to dig and survey literature until all the necessary facts relating to the theme are collected, the desire to collect all the facts, that is the completeness of the sources. This view penetrates even the considerations of theor-

63

eticians. The narrowing of specialisation, supposed Kolchin, Marshak and Sher (1970, 4), derives not from the loss of interest to theoretical work, as it is sometimes said, but because to deal with theory is impossible if you do not possess the entire completeness of initial data. The entire completeness! Yet Engels (1952, 256) objected against such approaches: If we should want to wait until the material is ready in pure form for the law, it would mean suspending the mental research until then and by this cause alone we could never obtain the law. The American anthropologist Julian Steward (1949, 24f) developed the same idea in the middle of the 20th century. For him it is obvious that the minutiae of cultural history will never be completely known and there is no need to defer formulations until all archaeologists laid down their shovels and all ethnologists have put away their notebooks. If anthropology is not interested in unique and exotic details it is necessary that attempts be made to formulate the law however preliminary they be. This formulation allows us to establish new kinds of problems and directs our attention onto new kinds of data which escaped us earlier. Fact-collecting of itself is insufcient scientic procedure; facts exist only as they are related to theories ... and theories are not destroyed by facts they are replaced by new theories which better explain the facts. As logic specialists showed long ago, induction cannot in principle be complete. (11) Practical experience as a permit to theorising It is common belief that theoretical and methodological judgements are only allowed to be made by those who have a long empirical experience. Joachim Werner worded this most openly and sharply: Publication of archaeological sources is a very laborious business, it demands tenacious and accurate working. Only he who has fullled these imperative obligations and possessed them receives the right to get into the questions of method; this is the border between science and journalistics (1951). Werners highbrow disdain to certain exponents of political journalism in the array of scientists is quite understandable, nevertheless his argumentation is a bit naive. On the basis of this view lies belief in theory as merely inductive generalisation of collected and pro-

64

Acta Archaeologica
completely different from those of objects under study are given to us immediately. A foggy trace of a particle in the ionisation chamber is distinguished by its properties from the particle itself, just as a visible trace of an aeroplane high in the sky is not comparable with the aeroplane itself, and is actually completely different from it ... In the cognition of nature such a situation emerges in which the striving to confront a thought with an object on the basis of experience is actually the confronting of an experience with an experience and thought with a thought. Archaeological materials are interesting for scholars not only in themselves, but as sources of cognition of extinct social and cultural systems and long past events. These sources are merely some fragments of what is necessary to reconstruct, only imprints, traces, and they are also in themselves incomparable with societies and events. One cannot make good progress by means of a simple generalisation of them. As B. M. Kedrov (1966, 36) rightly notes, empirical discoveries by themselves, however important they be, do not produce revolutions in science for this theoretical discoveries are needed. The entire history of science reects this. Among philosophers a saying by Einstein is popular: There is no inductive way which may lead to fundamental concepts of physics (Einstein 1965, 47). This is true for any discipline, including archaeology. Confronting a thought with a thought, theoretical archaeology inevitably appears in front of a necessity to turn to that discipline which specialises in the matching of a thought with a thought to philosophy. The reckoning to philosophy is inevitable for any particular discipline in the critical moment of the change of its theoretical conceptions. In front of the task of reconstruction and cognition of complicated systems, inaccessible by direct observation, archaeology naturally has to turn to the application of models. Among them the most similar and accessible models appear as the main ones for archaeology, namely ethnographic parallels. Therefore the way of overcoming empiricism in archaeology demands its integration with ethnography on the basis of a program which should be worked out by theoretical archaeology. Without this the rise of archaeology on a new, higher level would be impossible.

cessed facts. The most profound and careful among empirically oriented archaeologists avoid indicating the exclusively inductive way and speak of theory being merely the rationalisation of experience (Randsborg 1995, 221; see also 1994). (12) The not-the-time-for-theory argument In view of successes of theory in other disciplines and the glory of great theorists of the past in archaeology, a climb-down is sometimes suggested: there is place for theory in archaeology though only at a later stage or in its earlier stages, but not now. The rst statement was developed as an excuse for the absence of theory during some decades in Soviet archaeology. Ju. N. Zakharuk (1970a, 10) argued that it was necessary to go through an empirical period and rst accumulate enough facts in order to reach an analytical period and begin building theoretical constructions. The second solution has been developed by Ulrich Fischer (1987, 194): All theoretical methods of prehistoric archaeology ... were developed in the 19th century ... One can justiably state that the theoretical section of the methodology of our subject has been completed. Additions are to be expected in the practical section.

5. THE STANDPOINT OF MODERN PHILOSOPHY The absolutisation of induction and the blind veneration of facts is the Positivist principle. Inuential in the philosophy of the 19th and 20th century, it is rejected by many modern philosophical and methodological schools, including Marxism, right from its beginning! Empirical observation of itself never can prove sufciently the necessity, Engels wrote (1961, 244). Science, writes the Soviet philosopher Bazhenov (1968, 316), develops not by way of simple generalisation of particulars but by advancing hypotheses and including them into deductive systems with the aim of their subsequent checking. Another philosopher, N. F. Ovchinnikov, notes (1968, 12f) that nobody observed in immediate sensual experience Democritus atoms and elementary particles of modern physics. Only through a long chain of theoretical deductions is a thought on these particles mediated by experiment in which properties

Metaarchaeology
Of course, this is not to simplify. A number of attainments of archaeology are connected with the domination of the empirical method in the 1950s and 1960s. It is always the case where the empirical method works in a previously empty place, where materials have still not been collected there were such sections in archaeology or where there are no methods at all, which has also occurred, or where it confronts schematism and dogmatism, preconceived ideas, and a priori schemes. Here it gains victories. The absence of new and proper archaeological methods in Soviet and post-Soviet archaeology can be conceded as an essential hallmark of the general weakness of the empirical method that dominated Soviet archaeology for many years. All methodical

65

novelties of the last decades in archaeology have been borrowed in almost ready form from natural sciences or worked out for the rst time abroad (Semenovs functional-technological method of traceology was created long ago). The need for archaeology in its own archaeological theory is realised more and more strongly, and the trend towards theoretical studies is indeed observable. In order to advance further in our approach towards a general theory of archaeology, we need to realise clearly what theories are, as applied to archaeology, what their distinctions are from empirical generalisations of archaeological material, and what we expect from archaeological theory of the future.

66

Acta Archaeologica

7. What is archaeological theory? A systematic analysis of the issue

It is not the question Is my theory true? that now appears crucial for a theoretician, but rather Is what I am creating a theory? Dyson (1967, 118f).

1. OUTLINING THE PROBLEM What is theory?, asked the archaeologist Pete Rush recently (1994, 6676). His answer did not make for great clarity: ... no general and complete answer to this can be given, he writes, for even where theory has been explicitly formulated in archaeological discourse, the range of conception of theory, of its limits, capabilities and purpose is vast. He comes to the conclusion that it is impossible to nd anything common in archaeological theories, and they form something like Wittgensteins family resemblance that is some of them are similar to some others, but altogether they are connected only by their common participation in a net of links. Eventually they are linked with practice. But then any reasoning of archaeologists could be called theory. Julian Thomas (1995, 350) distinguishes a widespread misconception of what theory is and what it does. As Peter Ucko (1995, 1) notes, there is little agreement about the relationships of archaeological theory to practice, nor even, perhaps, about what constitutes archaeological theory . It appeared clear that until now archaeologists had no united and clear notion on the essence of theoretical work. With only limited experience and their archaic, most often unrecognised, inadvertently borrowed ideas of philosophers (mostly of empiricist ones), archaeologists approached research work with extremely simplied notions on its tasks and demands. Meanwhile, going now through the stage of scientically-technical revolution, archaeology needs a check-up and substantiation of its means of cognition. This stimulates it to venture more intensively into philosophy for help. Whereas in various philosophical schools as well as in various disciplines the inclination towards cognitive problems and the interest in metatheory is already manifest.

2. CURRENT NOTIONS (1) Ordered totality of facts According to such an understanding any generalising

inquiry is theoretical (induction is excluded from empirical studies; solely description is assigned to them), and the wider the circle is of generalised facts, the higher the level of theory must be. Only theory, Zakharuk claims (1971a, 48; 1971b, 8), ... is the single form of scholarly generalisation (Note: the peculiar double emphasis of this author has been preserved in this translation). It is in compliance with this that Brjusov (1954) entitled his article where the factual basis for chronology of the Neolithic of the forest region was generalised, entitled Some theoretical fundamentals of the Neolithic chronology. Such treatment of course reects the broad authority of the empirical method (see Klejn 1977c) the summary of facts is raised to the level of theory! Even to an archaeologist who ghts for theoretisation of archaeology (Gening 1982, 7), early theoretisation appears to be a matter of sorting and simple functional identication of things a pot, a knife, a grave though it is recognised on the spot that this is purely empirical study, and additionally that this is empiricism of the lowest descriptive level. Further theoretisation, transitional from an empirical level to a theoretical one, must be secured by generalisation, classication, revealing of regularities all this is already present in empirical study. Gening writes that ... A certain degree of theoretisation of the knowledge is reached but this does not move it [archaeology] out of the limits of an empirical science. And he elucidates this with a quotation from a philosophical work: one should not reduce it [the empirical discipline] to an accumulating of facts ..., on its empirical stage, too, certain conceptual positions rest in its basis, ... on the logical activity (Gening 1982, 10). This means that as far as scholarship is not reduced to the accumulation of facts, but is concerned with generalisations, manipulations with sets of concepts, acts of logic, then theoretical activity has already started. So what is the distinction of theoretical activity from the empirical one?! Goethe said everything factual is already theory. But, everything factual, and not everything empirical! Leaning to some inapt innovations of certain philosophers Gening attaches the so called descriptive theories to real the-

Metaarchaeology
ories, which must take an intervening place between the empirical and theoretical levels. According to these philosophers descriptive theory immediately describes a group of objects ..., solves rst of all the task of ordering of ... facts. Its laws are generalisation of the empirical stuff (Popovich and Sadovskij 1970, 206). But then, is there in general a ground for naming it theory? Of course, not least to take into consideration the desire of some empirical scholars to be called theoreticians. For those who deny the priority of the empirical method, there will be a resulting lack of understanding of theory too. Our philosophers suppose scholarship as a theoretical base of knowledge is not and cannot be the totality of facts ... (Ovchinnikov 1968, 23). (2) Ordered totality of concepts Chang writes, Archaeological theory ... is a systematic set of concepts which we believe our archaeological facts ... can be consistently dealt with and explained (Chang 1967b, 128). However, concepts are also to be considered not as a priori ones, but as simple generalisation and registration of facts. In this case the whole conception appears as the perfection of the preceding one. Therefore it is not infrequent that the adherents of the rst one are also ready to accept the second. Zakharuk (1970a, 10ff) awards rst place to the working out of concepts, the systematisation of them and the co-ordination of terminology in the theoretical development of Soviet archaeology. He presents this work as the main means for transition from empirical stage in the history of our discipline to its synthetic stage, while identifying synthesis and generalisation with theory. Of course, concepts are necessary for theoretical work, and theory puts them into a certain system. But the concepts themselves, even if ordered, do not constitute theory. The kernel of theory consists of judgements in which concepts are connected with each other so that the dynamic interdependence of variables expressed by the concepts is established. This alone allows the prediction of changes in some variables based on the changes in others. From the point of view of contemporary philosophy, if a logically non-contradictory system of concepts is unable to predict some observable consequences, then there

67

are no grounds in general to believe it is theory ... (Kopnin and Popovich 1968, 88). With such correction this idea was recently exposed by Chapman and Dolukhanov (1993, 1): The simplest denition of theory is the selection of the relevant variables or factors in a dened problem and the establishment of the inter-relationships between those key variables. So working at theoretical archaeology means working at three connected stages: (1) denition of problem focus, (2) selection of variables, and (3) characterisation and explanation of interactions between and transformations of variables. The issue, however, consists in what kind of interactions between these variables are implied and how they could be established. (3) System of laws In a programmatic book of New Archaeology the equation is given: Scientic theory, formally speaking, is a body of related laws (Watson et al. 1971, 163). Laws are to be formed into the logical form In C, if A, then B and are understood as expression of a regularity (Fritz and Plog 1970, 411). In Soviet literature it is difcult to nd such clear wording of the equation theorylaws, but the idea itself is very broadly accepted and directs many studies. A fully Marxist directive to reveal laws or regularities, underpins it here, albeit reduced by the inuence of empiricism to the level of a simple generalising of iteration (lawsimilarities). However, as A. L. Mongajt (Amalrik and Mongajt 1966, 173) illustrated with simple examples, law-type regularity is not reducible to an exposition of iteration, to an enumeration of common outer attributes, but presupposes a common cause of analogous phenomena. The leader of New Archaeology himself, Lewis Binford recounted a remark by his teacher Leslie White: Julian Steward doesnt know the difference between a universal fact and a law. At the time, admits Binford, I really didnt understand what White was saying. Were not law statements universally true? The difference of course is the role of theory. An empirical law and a covering law are very different. One implies lots of looking; the other implies lots of thinking (Binford 1972, 18). Thus, it is not that laws grow from empirical basis and form theory, but quite the reverse, they are

68

Acta Archaeologica
from strict interpretation of them, from the mental processing of them. The methodological apparatus should be inferred from theory, and theory should not be substituted with exclusively empirically-based methods. In the opposite case, a situation described by Hole (1973, 25) arises: Rather than an explicit set of theory we have a set of procedures: typology, numerical taxonomy, attribute analysis and the like. And even worse, since we must work with what we have, we tend to grasp at straws, hoping that ... artefacts ... will somehow inform us on the behaviour that we question. Neither established procedures nor the artefacts at hand necessarily result in information that is meaningful in terms of the cultural categories we wish to understand. (5) Projection of philosophical theory There was a time when in Soviet social learning only dialectical-and-historical-materialism was ranked with the status of theory. No dialectical issue was intended, albeit only one realised in a narrowly dogmatic sense and scholastically strict. Even the Marxist dialectic was opposed to theory simply because somebody from the classics of Marxism somewhere labelled it so. It was thought that in any social science dialectical-and-historical-materialism is the only general theory for this discipline. So the dialectical etc. not only appeared in the function of philosophical and sociological learning but also was allotted with functions of special theories of various disciplines also having the tasks to reveal the specicity of each concern. This dogmatic directive of Naturphilosophie (which would be better paraphrased as Sozialphilosophie), having reduced theoretical work to an arrangement of the materials according to general schemes set beforehand, was abolished in the late 50s during the years of the Thaw. The point of view refuted by Engels that arises from the treatment of philosophy as the science of sciences has lost recognition by many Marxist philosophers (Kopnin and Popovich 1968, 94ff). The importance of philosophy for the development of positive knowledge, of sciences and humanities, is doubtless. Marxist philosophy did contribute to them with interesting ideas. However, penetration of philosophical ideas into research practice of positive disciplines is a complex and mediated

formed by theory and are applied in the explanation of regularities in the material. It is not the other way around, There is simply no logical path which could lead from experimental material to building of theory (Bazhenov 1973, 403). This is actually an old idea and was expressed by Albert Einstein (1965, 10) with certain brevity: ... [a] theoretical system is practically unequivocally determined by the world of observations, though no ways lead from observations to fundamental principles of theory. (4) A set of interconnected methods What does it mean to proceed from empirical tasks to theoretical ones? Some Soviet scholars understood it as the following: For example, it is necessary to pass from an empirical procedure of comparing a given vessel with another vessel ... to general rules of comparing any two or more things or objects, and to build general rules of classication, dating etc. (Kameneckij et al. 1975, 5f). Sher (1973, 55) considers theoretical archaeology as the theory of elaboration of archaeological data. This theory is a system of general models, and the latter are exposed as general rules of comparison ..., various methods of statistical elaboration ... and so on. Among the determinations of central theory of archaeology by David Clarke there occur procedures common to archaeology everywhere (Clarke 1968, XV). For many New Archaeologists theory means simply a mathematical model a formula for calculating the relations between observable variables. Patty Jo Watson with her co-authors suppose, however, that one should not call methods and techniques of excavation and analysis archaeological theory, for these methods and technologies are mostly borrowed from other disciplines. Although it is not logically objectionable to group all these general rules together into a body to be called archaeological theory, ... it surely would best be referred to simply as the set of methods or procedures used in archaeology (Watson et al. 1971, 165). The essence is obviously not in the fact that methods are often borrowed from other disciplines, but that archaeology also needs its own methods for its specic tasks which are suitable for its subject matter. These methods do exist and it is optimal if their apparatus is not built from the facts themselves but

Metaarchaeology
process. They enter a positive discipline not in place of its special theories and not disguised as these theories supercially but are refracted, transformed and assimilated by them (Zakharuk 1970a, 14f).

69

has many meanings but it usually designates any of the following: (a) Every mental activity of a scholar, as opposed to practice (e. i. Strong 1973; Hensel et al. 1986; Shanks and Tilley 1987). (b) The solving of practical tasks on the basis of generalisation of facts, as opposed to speculation. (c) Simply a hypothesis as opposed to fact (see Strong 1936; Johansen 1974; 1982); (d) A proven conception as opposed to hypothesis. (e) Any explanative idea as opposed to other such ideas (e. i. Artamonov 1947). This is not merely homonymic use of terms but an exaggeration of one and the same concept which has some narrower meanings, which happen not to coincide with each other. In general, if despite some inconveniences people prefer to do with a single term by assigning to it various meanings in different conditions, it is not without reection. Concepts designated with the same term can coincide in certain conditions, they can transform into others, or they can transit from one to another it is exactly this partial transformation and transition that are xed by the transmission of the term. Thus, for archaeologists, it goes without saying that facts play an important role in forming theory, as its source according to one opinion, and according to another, as the touchstone for its establishment as a theory, or at least as its aim or object of application. It is a little more difcult to understand the inclusion of theory into the structure of facts, but if we recall concepts by which we are only able to apprehend mentally and describe artefacts for a message ... The British archaeologists M. Shanks and C. Tilley (1989, 2) phrase this idea extremely sharply and paradoxically: Theory and data are not in outer relations to each other ... Any set of data must have a theoretical orientation ... Our dialectical approach means that theoretical structure becomes a part of denition of data and vice versa. In fact, this idea has been around for some time. Even with the opposition of theory (as hypothesis) to fact, Goethe still managed to grasp the impossibility of a pure factography and stated: The main thing is to understand that everything factual is already theory (Maximen und Reexionen, Nr. 886).

3. THE CONCEPT IN OPPOSITION The obligatory study of the general methodology of research is not realised in Russian university programs. The rich and rened conception of theoretical knowledge on science and more broadly speaking on scholarship elaborated by philosophers and specialists in the sciences, including Marxist ones, is simply unknown to archaeologists, and only comparatively weak gleams of it are reected in the theoretical work of archaeologists. It is highly necessary to introduce archaeology to the contemporary notions on what scholarly theory is and from that point of view to analyse the simplied notions, evident or latent in archaeology and as realised in daily research work. With some philosophical snobbishness L. B. Bazhenov once declared the following: In general, in our opinion, a strict determination of theory in the frames of content-philosophical study is impossible and not necessary, (Bazhenov, 1973, 392). Nevertheless, it seems to me that he outlined, and partly realised in the same magnicent article, the possibility of four attitudes for such a determination of theory in archaeology: (1) The logical-deductive approach by indication of a covering concept and specifying distinction (this is the proper denition). (2) The ostensive approach by enumeration of examples. (3) The philosophical-content approach by the totality of the main characteristics of theory. (4) The systemic-structural approach by observing how the place of the concept in the system is changed at its various turns, if we consider the participation of this concept in different oppositions. The last method of determination listed here seems to be the most tting one to initiate a general acquaintance with the complex and many-sided object of theory. It helps to orient oneself and to select the visual angle suitable for the task. The word theory

70

Acta Archaeologica
lected and xed, but on the level where it is creatively processed. This is the level of manipulation with ideal objects, which the real objects, individual and (mainly) group ones, are replaced with. Under certain conditions results of these operations on ideal objects are translated onto real objects, which in the end permits explanation and prediction of phenomena. All this presupposes strict rules that form a program aimed at the information on the data. Under this meaning, theory is a program based on some fundamental idea for processing information. Here theory is the opposite of empirical. To me this is the main denition, and the most modern, well specied and embracing of denitions. By transforming the mechanism of processing into a stereotype, theory turns into method (Ovchinnikov 1968f). On the other hand, to react exibly upon impulses of empirical workers and for the sake of a free evolution of theoretical thought, theory is not enfettered with application tasks in each of its links, and it does not include a stereotyped methodical apparatus. It leads to it only in the end, separating it, and in this sense, it opposes the method. It is not by chance that more than half a dozen very well-known archaeological publications contain the opposition theory and method in their title (see Hawkes 1954; Willey and Phillips 1958; South 1977; Schiffer ab 1978; Gening 1980; Stjernquist 1984). Further, if prediction is only possible once theory correctly grasps and reects the regularity inherent in its object, then it opposes the object in its reection. The reection is inevitably incomplete. Finally, theory itself serves as an object of metatheory and in this capacity it is in opposition to metatheory. However, at the same time it opposes metatheory in terms of identity, because metatheory is also a theory.

Fig. 10. The concept of theory in oppositions.

(g)

Or let us take another aspect. Speculation is the opposite of theory if theory is considered a generalisation of facts (see determination (b) above). However, for a theoretician, speculation is his practice, closely connected with other kinds of scholarly activity in archaeology. Exaggerating this truth, Shanks and Tilley declare in a previously cited passage: We make stress on theory as practice ... Such position puts an end to theory considered as something sharply divided from practice of real activity in archaeology and standing behind of this practice. What we were working for is theory from practice and inside practice, the notion that archaeology is theoretical practice. Sic! On the other hand, speculation is pure thinking separated from practice from immediate dealing with facts. And such thinking is part and parcel of operating with concepts, that is, of a scholars mental activity in other words, an integral part of the creative activity of a scholar. Surely it is present and it cannot be absent in the work of an archaeologist too (Uslar 1955). From this point of view, theory (see determination (a) above) opposing practice naturally implies and includes speculation. In the methodology of research inquiry, in all the sciences, it is considered correct to designate the term theory with: (f) Strict logical systems which arise not on the level where the information on concrete objects is col-

(h)

(i)

(k)

Thus, in each of the indicated oppositions theory can also occur as its own opposition, but does so with the transition into another opposition. So the oppositions themselves are linked in pairs. Each pair contains three concepts of which the middle one is theory (Fig. 10).

Metaarchaeology
The philosopher B. S. Dynin (1972, 69), while considering the categories theory and law also discovered the necessity of opposite (by their meaning) explications of one and the same categories of methodology. Such an abundance of dialectically contradictory outer relations, with possibilities of distribution, shifts and reconstructions of a concept, presupposes a complex inner structure. This inner structure appears to be too complex for archaeologists. Too often they substitute some single components of theory for theory as a whole. The contemporary simplied notions on theory in archaeology have more to do with this substitution than with an inability to explicate, to narrow and to specify the concept of theory.

71

4. THE SYSTEM OF THEORY It is time to move away from simplications reducing theory to one of its components. Such impoverishing is especially threatening in archaeology because in the materials of this discipline, the driving forces of the process being studied (and of its most important regularities) are hidden deeper and are more distant from empirically observable phenomena than in other spheres. This means the components of theory are more separated in archaeology. We would like to navigate through the history of peoples and the laws of their development, but we see merely traces and remains of things. To summarise facts, to specify concepts and their interrelationships, to establish regularities, to elaborate methods, formulas and algorithms, and to apply philosophical principles separately or even together, is not sufcient to explain the similarities and differences in the material, to translate the facts of archaeology into the language of history and sociology, and

to predict reliably what the missing material was in short, to reconstruct the past. And without such explanation and prediction there is no theory. Something else is needed that will tie the components of theory together, transform it into an efcient working mechanism, into an instrument of cognition. Concrete explanative ideas are needed and one needs to know where they should be taken from, how they should be collected and how they should be proved. This is what forms the system of theory. If, for instance, observed similarities are to be explained, and the idea of diffusion is drawn in from ethnography, then one needs to assess the principle of actualism, and specify the concepts of borrowing, assimilation, area and others, and confront chronological data. One must also modify the typological method the genetic linking. After this many other considerations would still remain. The simplied notions on the nature of theory do not continue to linger in the environment of archaeologists because it is short of clear-cut denitions they are far from rare in the methodological works of philosophers, and the addition of yet another denition would not solve anything in archaeology. Much more important is to comprehend how particular philosophical ideas and empirical information enter into theory, and which of its components they inuence. For this, one needs to know which places the various components occupy in theory, how its cognitive mechanism is built and how it works. Thus, the critical analysis of the current, customary notions on archaeological theory turns us to the problem of the cognitive structure of theory. So far, at least in Soviet philosophical literature, there are very few works on this topic, and in theoretical archaeology there are none at all. Understanding this theme is a vital task.

72

Acta Archaeologica

8. Structure and working of archaeological theory

1. AN EMPTY BOX? The sociologist Myrdal (1958, 233) once made the following remark of theory in sociology: In our present age the task is not, as it was some time ago, in ... lling the empty box of theory with empirical data. Our theoretical boxes are empty, because rst of all they are not constructed so as to perceive realities. True, but one must still understand that the construction does not only mean the form of the box, and that it is not every form that will make theory workable. The point is not a box that one should ll, but the mechanism within the box that one should bring into play to perceive and interpret reality. This idea is now entering archaeology, so it is important to examine this mechanism in a theoretical framework. As far as this matter is concerned, references to denitions in philosophical literature can be very differently expressed and offer no solution. One has to reveal how philosophical ideas and empirical information enter theory, and which of its components they affect. It is necessary to nd the components that archaeological theory consists of, their roles within it, how its cognitive mechanism is built and how it functions. It is a matter of the epistemological structure (gnoseological in Russian terminology) of theory and it is desirable to differentiate this structure from that of a logical one. The logical structure of theory is a system of its statements and of the deductive connections of these statements, i.e. of the possibility to make a statement that follows from another. The logical structure of theory is also a system of corresponding hierarchy. To this can be added strong means induction and deduction for instance which must provide the substantiation for and the use of these statements. By epistemological structure something else is to be understood. This is the set of means of cognition (laws, concepts etc.) and the interrelations between them. It is this set that forms theory and offers it the heuristic and creative force, and provides it with the ability to transform information and to create new knowledge (Shtoff 1972). Both themes are clearly connected with each other but the archaeological elaboration of the logical structure does not really clear up very much in the epistemological structure. In archaeological literature there is still no work on this second theme, although some of its aspects were touched upon by American archaeologists. It was done when they considered the neighbouring theme of the logical structure of theory, in particular such sections of it as explanation and verication (Fritz and Plog 1970; Watson et al. 1971; Gibbon 1989; a.o.). Outside archaeology, however, in works of Western philosophy of science the epistemological structure of theory was very intensively elaborated, though mainly for natural sciences (bibliography see in Harvey 1969). Culture has its own specicity, which is exemplied by several traits: polisemy, a high degree of chance, the interest of researchers in individuality, the signicance of estimating etc. The specicity concerns peculiarities and differences in the way of proof and in the role of particular components as well as in the methods of formalisation, in a word, in many things but in the epistemological structure in general, in its general contours and plan. Let us separate the main components of the epistemological structure of theory in general and consider them in the light of the understanding of archaeology developed in the preceding chapters. It is clear from the previous chapters that archaeology is to be understood as a discipline which draws and processes material antiquities as sources of information on the past. The task of archaeology embraces historical interpretation of archaeological materials, i.e. recognition by traces and remains, as well as the reconstruction of things, historical events, processes and cultural phenomena of the distant past, while the causal links between them are established by history. As soon as these recognitions and reconstructions are not just an art but a research activity (science in Russian, French and German), it is implied that they contain some regularities in their basis, even laws, and among them there are laws specic for archaeology.

2. THE SYSTEM OF LAWS In order to explain and to predict the behaviour of objects, which is a primary demand of theory, one has to know the laws of this behaviour. In propositions

Metaarchaeology
asserting laws (nomological propositions), the dependability between variables is stated as well as conditions and limits in which this dependability is valuable, that is, the character of this dependability. Very often when archaeologists speak about laws they mean empirical generalisations, i.e. a general summary of facts, and they try to build theory purely on this basis. Thus A. E. Matjukhin (1975, 20) hopes to achieve theoretical explanation by means of general regularities such as the chronological correspondence of different types of tools with the character of production and the economy of ancient people as well as with their intelligence, or the increase of the coefcient of the use of implements and of entire assemblages from epoch to epoch, etc.. V. F. Gening (1975, 7) justly considers such regularities as empirical generalisations. One cannot build a well-grounded explanation and a strong theory on such a framework. Theoretical work in archaeology has no prospects without recognising this. Theoretical laws differ rst of all because they embrace dependencies between phenomena, causal or functional. There are works especially concerned with substantiating the presence of laws in archaeology (Plog 1973), and to establish that there are the same kinds of laws in archaeology as in other disciplines: causal, functional, dynamic, laws of probability etc. (Stickel and Chartkoff 1973). It would be no less essential in archaeology to reveal the specic type of these laws and establish their grouping. This specicity is in particular connected with a peculiarity of archaeology; in archaeology the objects directly observable (archaeological materials, material sources of information, or the material record) are separated from those objects which properly speaking are only of interest to the investigator, namely from the sociocultural systems of the past. An ethnographer observes such objects directly and a historian receives data on them from other people. The archaeologist meanwhile has access to neither of these and instead must view the behaviour of objects that once lived on the basis of the behaviour (the variability and diversity) of other objects the dead ones. However the behaviour of these others, as soon as they are dead, is not independent. Therefore in archaeology the laws of behaviour of the directly observable objects, the archaeological material, can be only empiri-

73

cal. To explain this behaviour the archaeologist needs two series of theoretical laws. A statement along these lines can be found in several works, especially in Watson et al. 1971, 24; and Fritz 1972, 140. The rst series embraces the laws of behaviour of active objects whose behaviour only interests the archaeologist in the nal analysis. This analysis consists of the functioning and development of sociocultural systems. So these are processual laws of the cultural process (in Russian terminology specied as culturalhistorical, for cultural embraces also contemporary processes, not only development). In other words these are statements on the connections between variables of the cultural (cultural-historical) process, and in particular the stages it traversed long ago. These are the laws of evolution. The second series embraces the laws of the surrendering of the cultural-historical process to the archaeological record, of the reection of this process in archaeological materials. These are statements on the interrelations between the sociocultural phenomena of the distant past and the patterns of the preserved archaeological material. These last laws express the dependence of the saved archaeological materials on sociocultural phenomena of the distant past. The matching dependencies or correspondences, according to Binford (1972, 249), form the basis of the bridging argument, so one can call these bridging laws. More to the point, these are the laws under the ruling of which certain processes subsequently proceed: (1) transformation of ideas into things (objectication of ideas), (2) materialisation of events, (3) mortication and archaeologisation of material culture (wear and tear deterioration, runout, selective accumulation, dilapidation etc.), (4) deposition of results, and nally traces and remains entering into research processing. A lack of understanding in how practice differs between these two groups of laws often leads to unclearness in important inferences. For instance, D. Clarke (1968) explains the behaviour of archaeological cultures directly from laws of their adaptation to the natural environment as if they were dynamic systems like animals or living cultures. This would mean that each change could be explained as the result of some impact of an environmental event. Let us look at the struggle between the two strat-

74

Acta Archaeologica
archaeology too. Thus both of the contesting strategies ignore the diversity of laws and appear to make incorrect simplications. Until recently it was not custom in archaeology to formulate laws clearly after having been discovered, or to give them names, as it is customary to do so in physics. But this does not mean that things did not come off satisfactorily when establishing empirical and theoretical laws. Rather, the weight of the formulation was produced by the diffuse probability character of these laws and by a lack of hope in developing them into a strict system. Even in traditionally more strict sciences, when the decisive role was occupied by laws-trends and archaeology began moving towards the strict sciences, the basis for purity became weaker. This saw the beginning of the explicit formulation of the old, long established laws and even authors being awarded with laws named after them on the basis of priority. For stratigraphic relations Steno laws were adopted from geology, in particular the Law of Superposition. They were named after a 17th century Danish physician (Hole and Heizer 1969, 16f). This law states that providing there has been no redeposition of material, then the upper layers of material are younger than the ones beneath. For work on burial assemblages the Worsaae Law was designated after the Danish archaeologist of the 19th century. According to this law things that appear in the same burial (and in general in the same assemblage) were used at the same time (Rowe 1962). Since then more new laws are being introduced and discussed. For instance in the collection of laws presented on prehistory by the American R. Carneiro is a law stating that in the relative sequence (in a typological series), the extent of similarity of any two attributes is directly proportional to the evolutionary distance between them (Carneiro 1969, 492ff). That is, the more distant these elements are from each other by their level of development, the fewer the cases of their reverse chronology. There is some very obvious criticism to be made concerning the laws of archaeology, for surely these truths are too banal and self-evident to be declared theoretical laws. Why are they claimed to be laws of archaeology when each of them embraces a sphere much broader than the subject of archaeology? One of the initiators of this raid on archaeological

egies of research. The contextualists (the traditional archaeology of the USA) believe that it is rst necessary to reconstruct the historical events of migrations, inuences, wars, building activities etc. and then based on this establish the laws of history (Sabloff and Willey 1967). The Processualists of New Archaeology demand quite the contrary. First to reveal the laws of the cultural process and only then, with their help, to reconstruct events (Binford 1968). There is clearly a muddle here. To recognise and prove new regularities in past stages of the cultural-historical process one must really know its course and direction, and this means a need for a preliminary reconstruction of events. On the other hand, in order to reconstruct them correctly and reliably one has to be in possession of the laws. But which laws are these exactly? Here is the key to this circulus vitiosus. With regards to the processual laws, they are of course necessary to the archaeologist in the selection of the structural models used in reconstruction, they are his only hope, as G. Clark (1957, 170) describes them. If it were the case, then the cognitive scepticism of C. Hawkes, G. Daniel, S. Piggott would receive metaarchaeological substantiation, for the role of the processual laws in archaeology appears twofold. They are one of the main aims of cognition because they are necessary for prognostication. At the same time they serve as the means for cognition because they are needed for the explanation of archaeological material and to reconstruct historical events. However, this contradiction is solved in the dynamics of research. The rst role (of the processual laws) is appropriate for the laws that are under investigation and the second role for the laws already discovered (partly discovered by other disciplines). It remains without doubt that the laws, once discovered and accepted, would lead the researcher to quite denite reconstruction models that would strive to afrm and support these laws, and this would not only hinder but delay the whole progress. Where is the way out of this circle? Most important is that the bridging laws play a crucial role in reconstruction, and that these are inferred not from archaeological materials but from ethnographic observations and from experiments. The processual laws are used in the passage from archaeology into history rather than in archaeology itself, although they are also used in

Metaarchaeology
laws, K. Flannery, says that it seems to him that in order to discover a natural law in the allotted six weeks of his eld season, the investigator was forced to tackle a problem of the utmost trivia; this has produced series of low-level generalisations that some critics have called Micky Mouse laws. These laws have even emerged from the lips of colleagues whom I regard as sane, serious, and competent people. For instance, at a genuinely exciting seminar on the Bushmen I learned that the size of a Bushman site is directly proportional to the number of houses on it. From another colleague I recently learned that as the population of a site increases, the number of storage pits will go up . And Flannery adduces the cry of his co-worker R. Whallon: If this is the new archaeology, show me how to get back to the Renaissance. (Flannery 1973, 51). This is challenging criticism, though one might seek consolation in the fact that theoretical laws do not exist in such a simple form each one of them is inscribed within an entire system. Theory is extensively useful and is able to separate and embrace the subject matter of a discipline. It can be used for systematisation and it is able to self-develop. These capabilities are provided by its structure; the net of laws that make up the kernel of theory is constructed deductively and shows a branch structure out of one or a few postulates (main laws) quite a number of derived laws are drawn according to strict rules of deductive logic, in the way a syllogism is drawn from premisses. What usually escapes the attention of methodologists is that the previously stated connections between phenomena serve as the second premisses of the syllogisms (and because ..., and as soon as ...). Or quite the contrary, the old law forms the rst premiss, while the new law forms the second one. Carneiros Law is in actual fact derived from postulates of Evolutionism, while notions on the dispersion and normal distribution of single artefacts around the selected positions serve as second premisses. The main postulates of archaeological theory are always evolving from some fundamental explanative idea which potentially holds the clue to the whole theory in compact form. For Evolutionism this is the idea of evolution, for Migrationism the idea of migration, for Environmentalism the idea of the impact of the natural environment on culture, for Stadialism the idea of qualitative leap, and so on. All

75

these ideas are able to suggest an explanation for the main enigma of archaeology, namely the phenomenon of the abrupt interchange of archaeological cultures, and in the same stroke to explain similarities and differences of culture in various territories (Klejn 1975a). Therefore when people speak of a crisis in the currently dominant archaeological theory, this means that it is time for a new fundamental explanative idea. When explaining the breaks between archaeological cultures one must always choose between explanative ideas based primarily on some kind of law either on a processual law or a bridging law. In the basis of an archaeological theory an idea of one of these kinds may be laid down. When Migrationists explained breaks by the occurrence of invasions, and the Stadialists by leap-form transformation, they appealed to processual laws. When Evolutionists explained breaks with gaps in our knowledge, and when E. Wahle offered the obscurity of the rst shoots of each new culture as a reason, they addressed laws of reection, i.e. bridging ones. However each leading theory explains not only this enigma, for in its collateral, accessory explanations it can support other laws. As a result Migrationist theory constantly explained the smallest similarities by means of migrations and made every effort to make the criteria of substantiation of migrations less strict, while Stadialists, on the contrary, made these criteria more rigid. Operating in this way, both theories appealed to bridging laws. On the other hand, Evolutionism was unthinkable without the notion of regular and gradual progress and its uneven development (with which they explain differences and similarities). Wahles theory implies an elitist production of culture, and both Wahles theory and Evolutionism employ processual laws. Thus, systems of laws in archaeological theories have a complex structure (Fig. 11).

3. THE LANGUAGE OF THEORY Provided that a law can directly manipulate the same objects as represented in empirical studies, then in its wordings, only the interconnection of two particular phenomena or (by generalisation) of the interconnection of two groups of homogenous phenomena would be xed. Yet the investigator would always remain on

76

Acta Archaeologica

Fig. 11. Structure and working of theory.

Metaarchaeology
the level of the observable surface of the world. The theory would be doomed to remain a hypothesis ad hoc; neither explanation, nor verication with independent data would be possible. However, the power of theory consists in the way it operates with special ideal objects abstract concepts (type, inuence, evolution, and qualitative leaps etc.), and in its ability to establish relations and dependencies between them. The properties and relations of objects in the real world, such as pots and arrowheads, are reected on by generalisation, transformation and by nding meaning. How is the transitiveness of archaeological knowledge implemented? How is the transfer to sociocultural phenomena of the past possible? The Swedish archaeologist M. Malmer places this task with archaeological typology. Among archaeologists the word typology has long carried a special meaning: the term designated certain means and results of classication and ordering (Klejn 1979a; 1982; 1991a). Under typology Malmer understands the set of classes randomly established by an archaeologist. Malmers types are established by means of denitions, until there is denition there is no type (Malmer 1962). This is, of course, idealistic exaggeration, even absolutisation of the relativity of our concepts, and it received critical appraisal in archaeology (Spaulding 1953; 1954; Klejn 1973c). It is essential that Malmer correctly grasped the necessity for every generalisation and for every interpretation to break away from particular artefacts. Archaeologists who believe they avoid typology and rely exclusively on stratigraphy or on combinations of artefacts in assemblages remind Malmer of Molieres Le Bourgeois Gentilhomme who did not realise that he was speaking prose. In fact these archaeologists use typology permanently, but only unconsciously and therefore they often do so incorrectly. The typological action helps the archaeologist to formulate empirical laws, but this is not sufcient for theoretical laws that reveal hidden interdependencies which have explanative character. Among ideal objects in this case are inevitably ones that have no correspondences in the real world at all (theoretical constructs). Either they still have no correspondences (they are designed to receive them in the course of practical creation, or at least can receive them) or

77

they have already lost them due to disappearance, destruction, both quite common in archaeology, or they are unable to carry the correspondences (the error of the conscience). Or these objects must in general have no such correspondences by denition, because they are relative assumptions, auxiliary concepts, or secondary abstractions. From the projection of the laws and concepts of theory onto reality some general aggregate theoretical object is accumulated, an ideal and abstract one, and theory regards this object as the essence of the real, empirically observable fragment of reality, a fragment which it purports to explain. If these concepts remain ideal, i.e. only as mental objects, they could not be grasped in the research and one could not manipulate them. Meanwhile clear dependencies must be established between them to save this information and to transfer it. In order that theory can manipulate these ideal mental objects successfully, they have to be partly materialised without losing their ideal properties and while still remaining abstract. This means that they should be conceived in sign form. Consequently, in order to be unambiguous and sufcient for the aims of theory, its terms should be placed in a system a glossary must be compiled, rst and foremost in natural language. Natural language is however very elastic and exible, its words and expressions not always clear. This has its merits as well as its aws. Drawbacks can be caused by the separation of dependencies, leading to complications and possible confusion. Many words are needed for detailed descriptions and this results in particularisation, which contradicts the actual essence of theory. Therefore it is necessary to work out some special terms. Terms are special words with a single meaning that are precisely delimited and provided with denitions. They are limited and required by the aims of theory and can be granted by systematisation as if creating a glossary. Connected with the concepts, they form the vocabulary of theory, and together with laws as a kind of grammar, the basis of the language of theory is built. It is only when one translates information from the concrete language of observations into this abstract language that one is then able to build theory. For a long time in archaeology, concepts and terms were selected and accumulated rather spontaneously.

78

Acta Archaeologica
sidered on equal ranking and interchangeable. In cultural material individual combinations are also important and the choice of them is not irrelevant. Therefore the task of unifying the language of archaeology is laminated, and the possibility of nding a solution appears differently in different layers. The ordering of elements of analytical classication necessary for the primary description proceeds most successfully. Of course such ordering, descriptive archaeology, implies some theoretical basis for the description (Gardin 1963; 1967), but this basis is not connected with the cultural content of the material, merely with its physical substance. Only topological parameters, physical-chemical properties and the like (K. Pyke termed these etic) have been taken into account. For their xation a joint conventional net of standards, templates and procedures is actually quite possible and such a net is minimally dependent on archaeological theories. However, as soon as it comes to unication of the nomenclature, i.e. names of types having cultural signicance, known as emic after Pyke, things become more difcult. In such ordering, in taxonomic classication, and in cultural typology no unity of criteria is possible. The criteria differ by tasks which are in front of the student, and by cultures which he describes, and these circumstances condition the different nets of types for the same material (Hill and Evans 1972; Klejn 1982; 1991a). The dependence on theories is quite apparent, as indicated when revealing evolution or migrations, inuences or imports, an archaeologist dismembers the material differently. Archaeologists long for systematics, as in biology, but end up only making an inventory. The overall works appear cumbersome and an indisputable result is reachable only when the author introduces order into the conventional practice of naming and builds the whole scheme in the fashion of identifying directories with convenient clues. Unfortunately, as a rule, archaeologists avoid this and do not recognise the qualitative difference of either task systematics and inventory (Klejn 1979a). Among the disciplines dealing with mass materials archaeology is the only one which did not create even a single identifying directory (cf. directories of bones, leafs, owers, languages and even classic mythological heroes). This is the payment for theoretical illiteracy.

Additions of terms were scarcely made, only a handful appearing with each new theory. Evolutionists have introduced development, connection by origins, typological series, typological rudiment, epoch, period and others. Migrationists added type area, origins of culture, Urheimat (original homeland), invasion, and Autochthonists added continuity. Diffusionists introduced inuence and borrowing. From the Theory of Stadiality archaeology received sinstadiality. New Archaeology has brought model, cluster, pattern (conguration), structural pose, and tool-kit among others. In the beginning it is usual practice for each theory to manage with old concepts and terms (Akchurin et al. 1968, 56f). However, because of the terminological conservatism and thrifty ways typical of the Old World archaeologists, this state is consolidated and as a result akin but different concepts are designated by one term: the archaeologist constantly tries to strike a chord on the piano with a single nger. But this means that terms lose their status and become simply words of a professional jargon. Today archaeologists think more accurately about their sets of concepts and terminology, and they introduce more radical and new terms more often, both for new concepts and for explication of the old ones. It became accepted to supply theoretical works with explanatory guides consisting of lists of denitions. We meet expansive dictionaries of this sort in the works of M. Malmer (1962), D. Clarke (1968), R. Dunnell (1971), I. Rouse (1972), Klejn (1991a), and in Malinas textbook we nd a summary glossary (1975; 1980). Another example actually arose as a separate book (Klassikacija 1990). The danger appears however which previously struck American archaeologists who boldly introduced new terms. The danger threatens that soon every archaeologist will speak his or her own language and others will not understand properly without the dictionary of the author in question. The elaboration of a unied language then enters the agenda and there are in fact works especially devoted to tasks of this kind. It is inevitable that one stumbles at this difculty, one of the important peculiarities of archaeology, and one of its distinctions from natural sciences. In the natural sciences not only structural elements but also their combinations are stereotyped and are con-

Metaarchaeology
More complex is the unication of variously designated theoretical constructs. Strictly speaking, on this level the joint glossary implies joint theory. Each theory needs a certain set of concepts with certain interrelations between them, although there are invariant concepts descending from theory to theory. Yet in archaeology, naive hopes are very popular, rst and foremost to establish a unied and commonly accepted glossary, to agree on the necessary set of theoretical concepts, on their delimitation and designation and then, on this basis, to build theories and to dispute them (Zakharuk 1970; Dunnell 1971, 4; Rouse 1972, XXV). Only the contrary approach can be realistic: rstly to create and rene our general theoretical object, to ascertain its needs in terms of concepts, then to work them out and confront them with the concepts of other theories. The systematisation of all the theoretical concepts of archaeology implies systematisation of theories in archaeology, and this is the task of metaarchaeology. The requirement of terminology is of an auxiliary, technical character and can proceed in another more simple way, but then it should be reduced purely to glossary tasks: registration and conventional formulation of meanings, without establishing the normative net and hierarchy.

79

4. MATHEMATICAL APPARATUS As a rule, modern physical theory proceeds in mathematical form. Propositions are formalised, variables and their relations receive quantitative characteristics and postulates become initial equations. Admissible mathematical transformations are made in full abstraction from the content of the variables. As a result new derivative equations emerge, with additionally discovered relations between the variables or with new variables. It remains to take these new relations and variables back to the content and to nd their correspondences in the language of the given theory, i.e. to supply mathematical symbols again with the content of theoretical object. Usually it is called content interpretation of formalised propositions, but Harvey calls it transliteration, leaving the term interpretation for the switching from theoretical concepts to realia. This entire procedure allows the shortening and

standardisation of form of operations that infer laws from postulates, to mechanise operations, and, using the sophistication of mathematical technique of transformation, to discover additional laws that exist in very complex connection with the initial ones. In such a complex connection it would hardly be possible to reach them by means of direct study of the subject content. The principle of using mathematical apparatus in archaeology is the same as in other disciplines (Sher 1970; Kameneckiy a.o. 1975), but here the conditions are different; there is a distinct specicity to the material and there are different tasks to tackle. Thus the functions and the role of the mathematical apparatus are rather different. First of all by studying the complex cultural material, more attention is given to the task of searching for oppositions and evaluating discrete relations. This means evaluations for which gradual changes are inessential, only the question of whether the relation is present or not is important. Therefore many laws of archaeology do not hold quantitative character, for example the Stenos and Worsaaes Laws already mentioned. It is true that modern mathematics is not reduced to quantitative operations, and it is possible to nd a number of quantitative laws in ancient cultural material (Ford 1962a,b; Carneiro 1970), and there are ways of quantitative expression of qualitative dependencies (i.e. Hill 1968). Nevertheless the role of mathematical apparatus in archaeological theory is more modest. To narrow down and establish the functions of mathematical apparatus in archaeological theory, density and cascade character of links in cultural material are especially important, as well as frequent inclusions of random non-controlled factors, which comes from the difculty of isolating the connections under study from noise. But most important is the uncertainty of cultural characteristics of archaeological material. The quantitative analysis demands evening out, which stems from the assumption of units of counting, for different variables can have equal value in different conditions. In ancient cultures these units had a different meaning and different range, but exactly what these were remains unknown to the archaeologist. This is why mathematical apparatus in archaeological cultures has not been developed as a rule since the very beginning of working on theory. It was

80

Acta Archaeologica
oretical conception or can be obtained from natural sciences with their methods (pollen analysis etc.). For the followers of Empiricism, including the early Positivists, the empirical basis was the only permissible source of theory, and its core was reduced in essence to the execution of empirical laws. One could explain an observation by means of covering it by empirical law, and one could explain the empirical law by means of covering it by a broader empirical law, and all while remaining on the surface of phenomena, without diving into hidden causes. In these frames, solving the tasks of archaeology would appear absolutely hopeless: to reconstruct and cognise unobservable processes on the basis of the principle of actualism this cannot be empirically defended! The opposite extreme is a refusal to rely on empirical basis. It is performed by reducing the function of factual material to a purely passive role to become the object of application of explanatory ideas and to provide an illustration for them (various scholastic and dogmatic conceptions). A looser variety of such an attitude is connected with the attempt to reduce the role of the empirical basis to the testing of theory. This occurred in Post-Positivist Deductivism and in archaeology, which has accepted this as its initial principle in New Archaeology of the USA and Britain. In fact, only in the latter role can the empirical basis serve as the starting-point for a strict deductive inference able to connect it with theory. Is its meaning exhausted here? Can the other basis, the theoretical one, serve as the only source of theory? The theoretical basis consists of two parts: (1) the scientic view of the world and, (2) the collection of previous theories of the given discipline. A scientic view of the world is specied as applied to each discipline and consists of principles (the most general laws of this discipline) and universalia (its most fundamental concepts). The principles build borders for what is possible in terms of the laws introduced by the new theory, and through the universalia culture or artefact we dene concepts introduced by the theory. For the determination of the universalia themselves, as they are the most fundamental concepts of the whole given discipline, one must go outside its borders. Universalia can maintain a number of kin-

added considerably later and does not embrace the whole corpus of its system of laws, it only renes single branches of this system. So the Theory of Diffusions has taken a remarkable place in archaeology since the beginning of the 20th century and was strengthened since the middle of the twenties by Childe, but only in the 1960s did quantitative studies based on it appear. These were estimations of the diffusion rate (M. Edmonson 1961), connection of percentage distribution of types with territorial propagation (M. Malmer 1962), and they acceded quantitative diffusion models elaborated in geography (Hgerstrand 1952; Brown 1968; Hudson 1972).

5. PREMISSES OF THEORY Theory does not emerge from an empty space. In order to explain one must have something that pertains to explanation, some body of empirical data, and sources are needed from which explanative ideas can be taken when they are not contained in the empirical data. Thus, for creating a theory two groups of premisses are needed: the empirical basis and the theoretical basis. The comparable evaluation of their roles in forming archaeological theory brought about a most polemical attitude. The empirical basis consists of the collection of, 1) single data of experience (observation and experiment) and, 2) empirical laws derived by means of their generalisation. For instance, the discovery of a burial with a bronze awl found in the barrow is an empirical observation, while the statement that bronze awls occur (or can occur) in barrow graves, but int hand axes never do, is an empirical law. The empirical basis will be reliable only in the case of this knowledge being obtained by means of strict methods. But methodical proof and improvement is one of the functions of theory (Klejn 1978a; 1995c), and theory cannot come before its premiss. So what methods can be applied to the building of these premisses? Only methods derived from the previous theories or borrowed from other disciplines. So for checking Migrationist constructions one needs relative dating of similar sites on the assumed original homeland and in the new area populated. This dating can be provided by the Evolutionist-Typological method elaborated on the basis of the earlier Evolutionist the-

Metaarchaeology
dred disciplines simultaneously and archaeology uses this. Age old principles of archaeological interpretation remained tacitly implied and in recent decades attempts have been made to formulate them. In these attempts the conviction of archaeologists has been discovered that there are very few such principles if not only a single one. As earlier mentioned, Hawkes believes that there are merely two principles which offer scientic status to archaeology (1957, 94; see above, ch. 5.3) However, in other attempts more principles are mentioned those of determinism, actualism, historicism etc. When the full list is compiled they are many in number. This becomes clear as soon as the question arises on the possibility of handing interpretation over to the machine ... There are a number of recent works devoted to universalia to the concepts culture and archaeological culture, type, attribute, artefact etc. (Dunnel 1971; Bochkarev 1975 a.o.). The scientic view of the world is closely connected with intertheory. Intertheory is the environment that is made of theories of other disciplines and, with its ideas, inuences the theoretical development of archaeology. These interconnections are most apparent in the case of Evolutionism which is represented in biology (Darwinism), geology (T. Calvin, Ch. Lyell), linguistics (A. Schleicher), ethnography (E. Tylor and J. Lubbock), archaeology (G. de Mortillet and O. Montelius). The idea of complementarity formulated by N. Bohr for natural sciences did much to achieve popularity for the model conception in a number of disciplines. G. Gjessing made an attempt to apply this idea directly to archaeology and connected this with models in archaeology. Intertheory and the scientic view of the world developed under the command of philosophy, although not always under its leadership. Changing abruptly, they form subsequent styles of the scientic thinking, after M. Borns expression (1953, 501; 1955, 102; cf. Klejn 1973b). Metatheory is connected with the laws and concepts of previous theories. Relying on the methodological function of philosophy, metatheory studies, generalises and gives them sense, and through them it describes the main characteristics of the given discipline in order to enable the production of new the-

81

ories. Some contemporary investigators consider just the theoretical basis as the main source of new theories and the empirical basis only as a touchstone for verication of new theories. Of course, many concepts and terms are delivered from the stores of theoretical foundations in ready form for the needs of a new theory while universalia allow the formation of new concepts. Metatheory helps to manipulate them, to establish the relations they have with the material and to previous theories. Many laws of the previous theories will pass to the new theory too and serve as secondary premisses of syllogisms. The principles will also be useful by helping to formulate postulates. In a word, much more of the building material comes from the theoretical basis to the building site of the new theory than from the empirical one. However, even with this there is no fundamental idea, strictly speaking delivery is not guaranteed. A fundamental explanatory idea may be borrowed from the theoretical store of an adjacent discipline as ideas of evolution and migration were borrowed, and may be transformed from the inherited theoretical luggage by means of broadening a narrow idea, by means of release of limitations this is how Contextualism was formed. Methodological expansion of a branch that developed successfully is considered a normal case (Ovchinnikov 1968: 22). In archaeology organised searches for a prospective source of borrowings are discussed (e.g. Renfrew 1969b). For those searching after ideas a directory of explanatory ideas was compiled which contains a thousand such ideas able to explain human (Berelson and Steiner 1964). An idea can also come from non-scientic knowledge or can emerge anew as did Wahles conception of the idea of a latent development of a culture, and there are no strict rules for its creation. In retrospect, one can estimate the role of various inuences: (1) of the situation in the empirical basis, (2) of theoretical resources, (3) of the state of social conscience, e.g. the role of the ideas of equality, progress, competition and reforms in forming of Evolutionism, and (4) of an individual practice of an investigator. However, any stimuli act indirectly through the creative conscience of a researcher, through his or her imagination. It is here where the notions that were earlier unconnected now touch each other, where new unexpected associations emerge and surprising ideas are

82

Acta Archaeologica
problem emerges of the interrelation of two worlds: the world of theory and the world of facts. Empirical generalisations are partial whereas a theoretical law expresses necessity (essentiality) and in its own sphere it is complete and universal. If and when facts are generalised and rise to the level of empirical laws, then to bring the world of facts closer to theory is impossible while they are still distant from the form of necessity and universality of research thinking. We are left with the only solution: to transfer the world of facts over to theory. There is only one way to do this, namely to transform the mental objects and propositions of theory into real ones, i.e. to render concrete the fragment of reality under study by means of this theory. This means constructing a third world, ideal but real, and formulating a particular statement in the language of laws and concepts of this theory. For this sake one must rst and foremost describe the factual material in concepts and terms of the given theory. In general there is no such thing as pure description as well as pure observation free from the demands of theory (or of fuzzy mixing of old theories) and free from theoretical notions. When describing something, we inevitably use concepts, and concepts are inevitably components of a system of notions. This system is of a theoretical character, and let this theoretical setting be dim and dull, implicit and non-strict. However, concepts and terms with which the material was described before forming the new theory, belonged to the glossaries of old theories. Now one has to describe this material, in the glossary system of the new theory and using the old glossary only to the extent in which its parts are assimilated by the new system. Using this description, one can already infer particular empirical regularities from abstract theoretical laws, i.e. to build models that realise theory as applied to every particular fragment of reality. From each law one can infer a number of such models, because the conditions of realisation are different. These models dictate the expectation of observations. When confronted, the expected empirical laws and observations must coincide with real ones if the theory is true (they can however also coincide if the theory is not true). In order to describe strictly the material by means of the glossary of the new theory, one must possess a

born those from which it is necessary thereafter to select intuitively a suitable one. Thus, some creative act which is equated with abduction would be positioned before induction and deduction, more exactly before their inclusion into theory (Hanson 1965, 86f; Hempel 1966, 15). This thought has already found its way into archaeology and introduced a confusing re-estimation of values. Good abduction does not depend upon the quantity of data a given investigator has examined ... It is a function of his creative ability as a scientist, and no more ... 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 ... 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 (Plog 1974, 19). There is, of course, some exaggeration in such strong distancing from the real material. To create new ideas material is necessary too but the creative process does not extract general, repetitive characteristics from the material in the way induction does and it does not cover facts by imperatives derived from more general regularities like deduction. It is a mixing of information that leads to new connections and associations which forms part of the creative process. By itself this act is alien to the discipline and does not belong to strict science, but without it there is no theory. When completely embraced by a strict discipline, without its divisions, knowledge is faultless and lifeless. In order to shape a theory it is insufcient to know facts and to think strictly. Creative boldness is necessary too and although it can sound impolite, sometimes talent does not go astray.

6. THE OPERATIONAL APPARATUS AND TEXT OF THEORY Theory is always exposed in a special language and establishes laws in a special ideal world by postulating links between ideal objects. If so, in order to provide proofs from the facts and to apply it to the real world of particular things and events, it is necessary to translate concepts and inferences of theory into the language of facts and particular operations. The complex

Metaarchaeology
set of rules that regulate the covering of particular phenomena with concepts of theory. These rules of correlation, connecting the ideal objects with real ones, control the selection of the chosen things, properties and relations of reality, norms of measuring the material, permitted operations with terms, i.e. the functions of the present terms. In the same vein, in order to infer particular models of real phenomena from theoretical laws, one must possess a set of rules that regulate the transformation of dependencies between the variables into particular ones. This transformation includes: a) simplication imposed by the unevenness of the material: by encumbrances, noise, gaps; b) delimitation imposed by the specicity of the material, by the impeding conditions. The rules must predict reliable operations with concepts and formulas of the theoretical laws and determine in this way their real sense, the semantics of the matter. The operational apparatus of theory is built by these two codices that regulate operations with the laws and glossary of the theory in the theoretical rendering of reality. If by concepts and terms of theory we mean the glossary and by the laws we mean the grammar of a certain language, then the operational apparatus of the theory forms the semantics of this language. Most specialists of scientic methodology make no distinction between the text of theory and its operational apparatus. In descriptions of the cognitive functioning of a theory one can nd a text (Brown 1963) or the operational rules (Nagel 1961) and one can nd direct statements on their make-up (Harvey 1974, 85). Meanwhile their functions are very different as they are in every language the functions of semantics and the statements or text of a theory are different from the functions of the rules and the results of their use. It is therefore reasonable to distinguish between these two members of the cognitive machinery of theory because they are distinct by their connections with other parts of this machinery. In archaeology, perhaps in history too, the task of theoretical rendering of reality appears to be more complex than in other disciplines. The point is that theoretical laws of archaeology, while providing historical reconstruction, also establish dependencies of rendering. These are dependencies between variables of either the rendering procedure (bridging laws) or

83

of the object to be rendered (processual laws). Therefore, when formulating the statements of theory on the basis of theoretical laws and rendering initial reality as if it were empirical material, it will be already a secondary rendering that is arrived at, and the result will have been rendered twice. In the beginning in fact, on the basis of the archaeological record we must render the real world of the past, and then we must render the record on the basis of this rendered world. However, reconstruction of the real world on the basis of the archaeological record has many possible outcomes since cultural objects are polisemic and the record is fragmentary in several ways. Going back to the record from our hypothetical reconstruction we must face not only the manifestation of known laws in the formation of records but also the contribution of chance. So in both ways to the past real world and back to the record distortions and losses of information can occur. This brings us back to the question of the specicity and importance of archaeological laws (Klejn 1979b). Really, both the main categories of archaeological laws processual and bridging often seem quite ordinary. They have been revealed with methods and materials of other disciplines, such as sociology, ethnology and anthropology, or simply through some experience by means of sound reasoning. Processual laws also fully realise their explanative and organisational potentials in other disciplines like the history of material culture, and historical sociology. However, even the laws of this group are particular to archaeology and are used differently than in other disciplines. They demand special rules in operational apparatus, for intsance the need to separate the diffusion of ideas from the diffusion of things in archaeology. Furthermore, in this group there are laws where archaeological materials are necessary and irreplaceable in order to reveal those laws. In particular are the specic laws of early stages of the cultural-historical process (Watson et al. 1971, 168f). As concerns bridging laws they are simply unnecessary for any other discipline, except other areas of source-studying. Not by accident the means by which these laws are obtained and veried, through scientic and sociological experiments as well as ethnographic observations, formed several branches not of sociology or natural sciences or ethnography, but

84

Acta Archaeologica
sometimes still are stubborn in seeing only ancient ethnicity in archaeological cultures, despite all the cautions of ethnographers (for criticism of these views see Mongajt 1967; Klejn 1970). For a long period the source criticism in archaeology was limited to the socalled extrinsic criticism, that is, checking the authenticity and preservation state of the material form of the records; whether they are false or not. Only recently the demand of the inner criticism of sources, i.e. evaluation of their cognitive potentials, expanded into archaeology from history (Eggers 1950; Klejn 1978b; Kristiansen 1978). This led to the situation that even now the possibilities contained in a number of old theories have not been exhausted, despite their one-sidedness. This is because in their time these theories did not receive a satisfactory operational apparatus. The elaboration of such an apparatus presents large heuristic opportunities. We rightly criticise Migrationsim and Diffusionism for converting the ideas of migration and diffusion into a skeleton key for the interpretation of any changes in a culture. The migration theory in Kossinnas doctrine was built on extremely primitive rules of correspondence and particularisation (see Klejn 1974). But as it is, we have no need to recoil from these phenomena migrations and inuences per se. These phenomena acted signicantly in the cultural-historical process, and archaeology needs the theory of migrations and theory of inuences and borrowings. The development of operational apparatus for the archaeological theory of migrations means searching for archaeological hallmarks of migrations, for qualitative criteria of the validity of migration hypothesis in each particular case (Klejn 1973a; 1999a; Titov 1982). Such apparatus await development for other old archaeological theories, provided they have revealed essential characteristics of some real phenomena. These old theories will still work further in the discipline and will gradually become more dened cases of a more general theory. Of course, one cannot take for granted that this will not produce new exaggerations, and it is possible that archaeology still awaits Neoevolutionism, Neomigrationism, Neodiffusionism etc. However, archaeology has not only the experience of plummeting into extremes but also the experience of overcoming them.

just of archaeology: experimental archaeology (Semenov 1957; Ascher 1961; Coles 1966; 1973), urgent archaeology, living prehistory, and action archaeology. From this series of new disciplines only living prehistory belongs more to ethnography or ethnology, although it is oriented to serving archaeology (Kleindienst and Watson 1956; Asher 1962; McKern and McKern 1974; a.o.). In essence, the rules of correspondence in operational apparatus must receive substantiation from these laws. It is customary to make a clear distinction between rules and laws in methodology of science (Popov 1972, 121, 151), but in archaeology the distinction, to a large extent, is lost. Laws of archaeology, as was mentioned above, are involved in the operational apparatus of archaeological theory by receiving status and meaning from this system of laws. Thus laws in archaeology display a far from trivial and ordinary character. Let us take the laws of Steno and Worsaae. Surely, what is simpler than a sequence of sediments or the contemporaneity of things in one and the same grave? It can however prompt us to think of conditions in which these laws are not valid, situations where superposition does not hold. Many natural phenomena can present contradictions to the law of superposition volcano eruptions, levelling of a surface, the activity of burrowing animals, geological faults etc. The law of Worsaae can also be opposed as a consequence of certain human actions crypts, secondary burials, reburials, the robbing of graves etc. More complicated still was the task of expanding this law, performed by O. Montelius, in order to embrace non-burial assemblages like stores and dwellings. In other words, it is the incomplete wordings so far typical of archaeological laws that make them so banal. This can happen when one forgets that, as Engels wrote in the Dialectic of nature, relations that are extended to a law are present only in certain conditions, i.e. the realisation of laws always occurs wherever these conditions are present (Engels, 1961, 549). Archaeologists have felt the necessity of an operational apparatus for their theories for a long time but until recently it was compared with the operational apparatus of the physical sciences and endeavoured to nd close correspondences of its theoretical objects in the real world. This meant archaeologists were and

Metaarchaeology
7. THE STATEMENTS AND VERIFICATION OF A THEORY Formulating the statements of a theory is made possible by verication against reality its test. For New Archaeologists this is a comparably simple procedure of empirical verication, or the slightly more complicated procedure of empirical falsication. In the case of empirical verication the statements are conrmed by looking for correspondences in the real world, whereas empirical falsication checks for unexpected phenomena that do not correspond with the expected or predicted phenomena of the theory, i.e. to disprove the theory. However, verifying a theory is not limited to a meeting with reality on an empirical basis (let us call this procedure the E-test). The point is that this type of testing has a number of weaknesses. It is only a strict procedure when applied to material in which facts are always unambiguous, always occurring in one-to-one relations: when the cause of or the reason for a type of material is able to produce only this one type; while each such consequence can only have a cause of one kind. Yet even in the study of the physical world this condition is not always granted. In culture, at least similar causes in similar conditions can lead to similar results, and by means of altering conditions one can recognise regular connections. Nevertheless the procedure of deduction of expectations is there unreective, it cannot be turned back. Unforeseen observations disprove a theory of course, and predicted observations substantiate it, although not completely for they only make it more probable (and without quantitative determination of the degree of probability) because the expected phenomenon could be due to other causes. Many ways have been suggested to compensate for the weaknesses of empirical verication: to enlarge the number of the deduced predictions, to provide the diversity and mutual independence of the predictions, or even to substitute falsication for verication, i.e. to check the competing hypotheses for the possibility of disproving them. Yet none of these means give assurance. Basically this is conditioned by the imperfection and incompleteness of induction that is the foundation of the empirical basis. In the cultural material the role of the conditional choice of forms, of conditional offering of meanings

85

to them, is considerable. So quite often similar factors even in similar conditions can lead to different consequences, and behind similar phenomena in similar conditions different causes can be hidden plural causality dominates in cultural material (Kbben 1967). Archaeological fact is polisemic (Klejn 1973b), and regularities are therefore discredited. In order to avoid having these regularities discredited, it was suggested such disturbances of laws could be considered excusable exclusions as statistical or explainable errors (Kbben 1967), or as borders of the sphere of the law (Klejn 1972a). Although this disposes discrediting of laws, it does not remove the difculty of their testing. There is more asymmetry here than in the physical world: even an unexpected observation does not disprove a theory and does not demand its revision and correction it might be simply the result of the inclusion of an untting fact. In this situation there are additional difculties in rendering reality. In order to understand them one has to return to the contents of the text or statements of a theory. Beside the enlisted links in rendering reality, two more are involved: explanation and prediction. Both use the means of projecting the inferential expectations onto empirical matters and onto practice. However, in the case of explanation it is achieved through the comparison with reality (the E-test). So, prediction is a simple projection of expectation (and its repeating as applied to practice), while explanation uses connections and meanings suggested by theory and carries them onto reality and embodies in itself the power of theory, its authority. For the activity of theory in a puried environment, for instance in physical science, symmetry of both projections is implied: if predictions are realised and they are identical to expectations of a theory, then the theory is true. Even in physics symmetry falls under doubt, but only from one side: not every explanative inference provides explanation (Pechenkin 1973). In archaeology prediction about the past is extremely asymmetric to explanation. Even when a cultural-historical law and some conditions of its realisation are known, we are nevertheless unable to render the result of its impact unambiguously if we do not know all the conditions of realisation. The historian and the archaeologist must always take into account interaction of

86

Acta Archaeologica
beautiful theories. When a theory is veried on the basis of some multiple facts in one sphere and turns out to be a asco in adjacent arrays of material, it does not necessarily mean that the theory is wrong at all. A new, competing theory may explain the facts which had troubled a previous explanation but may not necessarily include all the other facts explained by the old theory it simply leaves them aside. But if the new theory is to be stronger then it is able to include the old theory as well. In natural sciences this is considered the grand road of the development of science. In social sciences the development is more frequently directed another way. Here the confronting social forces, while taking opposite and extreme positions, often defend one-sided views, strive to maintain the universality of their own theory and try to refute and crush all the others. However, the grand road mentioned is not contradicted for these disciplines either. In the search of that grand road for theoretical archaeology there is perhaps a tendency to think of a theory that could include invention, migration, diffusion and other causes for the interchange of cultures as particular cases (see Klejn 1972b; 1981c).

many factors and reckon with the role of contingency as sometimes very productive. Such are the weak points of the E-test in general and in archaeology in particular. In many disciplines a T-test is included in verication that checks concurrence with previous theories on the basis of Bohrs principle of correspondence. The new theory should not contradict old theories already proved for their sphere of competence and must include them as limiting cases into its own body. This is the demand behind which belief in the unity of laws of the world is hidden. It becomes a criterion of reliability of a new theory (Kuznecov 1948). Such a criterion can only strengthen the respectability and prospects of a theory, it raises its creditability so to say, but cannot provide complete condence needed in every case. The less reliable the checking by the E-test, the greater the signicance of the T-test for checking by means of theory. Archaeology is accustomed to discarding fashionable theories rather easily and this ckleness is used by some authors both in the substantiation of theory and in the advocacy of freedom from theory. Some archaeologists describe the failures of theoreticians and enthusiastically collect facts that have destroyed

Metaarchaeology

87

9. Functions of archaeological theory

In his article The philosophy of archaeology, anticipating the posthumous collection of David Clarkes papers, Glynn Isaac (1979, 15) addressed both those scholars who are interested in theory and those who ask: Why do we need to bother with theory?. Until now did we really live badly without theory?, asked Jakov Sher in his preface to the Russian edition of J.C. Gardins Theoretical archaeology. He asks this question from the position of an imagined empirical archaeologist, although there are many who ask this question quite seriously. Well, the last question could immediately be answered with reference to Clarke. In his Analytical archaeology Clarke (1968, XIII) demonstrated that it has long been the case that Archaeologists do not agree upon central theory, nevertheless regardless of place, period, and culture, they employ similar tacit models and procedures based upon similar and distinctive entities the attributes, artefacts, types, assemblages, cultures and culture groups. Later Clarke (1970, 29) stated: the central theory uniting archaeology is implicit in what archaeologists do and constitutes a real central theory however weak and inadequate any written account of it may prove to be. But this latent theoretical conception inevitably appears dim, incomplete, insufciently elaborated, and weakly grounded. An explicit presentation of theory allows us to use its merits more effectively and to detect its shortcomings at an earlier stage. However the question set out by Isaac and implied by Sher still remains. Both authors stress that they do not share the scepticism of their imagined opponents. The use of theory in archaeology seems, to them, beyond doubt, and they see the ready answer in the works of Clarke and Gardin. Meanwhile, these scholars themselves regretted the fact that it was still impossible to answer this question positively and popularly. Having divided archaeologists into those who accept theory as necessary for archaeology and those who think theory is a hash, Clarke (1968, 22) declared that his Analytical archaeology was written for the rst group whereas the latter will doubtless continue to ... blinker themselves to narrow aspects of narrow problems. Theoretical archaeology has still not proved its necessity, noted Gardin (1983, 32). There is, in his words, an understanding that arguments are to be expected from theoretical archaeology. If one approaches this task from the position of a theoretician, it is evident that the utility of theory must be demonstrated not by examples of a useful application of several theories (because however multiple these examples may be, there will always be counter-examples which can be adduced), but rather by strict logic, that is, in the framework of methodology. The question of the use of theory in archaeology, as translated into the language of methodological analysis, turns into the question of the functions of archaeological theory. Works devoted to this issue are nearly absent in archaeological literature. Non-Russian works of this kind are simply not known to me. Among works in Russian two may be mentioned: the short article of V. V. Radililovskij (1985) and a chapter in V. F. Genings book The structure of archaeological cognition (1989b, 156193). Both authors have transferred schemes from Soviet philosophical literature mainly for the exact and natural sciences onto archaeology (Pechenkin 1972; Bazhenov 1973; Ruzavin 1978). These elaborations, in turn, rest upon works of well-known non-Russain specialists in the philosophy and logic of science. In Radililovskijs list, which he has borrowed from Bazhenov, there are four functions of archaeological theory: descriptive, explanatory, predictive, and synthesising. Gening retains only two of them, descriptive and explanatory, but adds another one, systematising, which he has taken from Ruzavin. Both archaeologists try to nd arguments for likening archaeological theory to physical, chemical, and sociological theory: archaeologists must have all the traits of these subjects laws, and systematisation, and explanations, and predictions. Proceeding from an understanding of the concerns and nature of archaeology and from different notions on theory, it seems to me sensible to consider (consecutively) various stages of archaeological inquiry in order to detect whether, in each of them, archaeological theory plays any role, and if it does, what role. It is also worth evaluating

88

Acta Archaeologica
cal. Here, also, a theoretical analysis of the problem is needed. Speaking about studying historical objects in general, G. A. Antipov (1987, 49f) enumerates possible errors of identication and accompanies his list with archaeological examples. It appears, from his selection, that two possible errors may occur in each of the two characteristics of the archaeologicality (antiquity and the belonging to the material culture) over-estimation (when non-archaeological objects are considered as archaeological ones) and under-estimation (when some archaeological objects are omitted). In sum, there are four categories of mistakes: mistakes of modernisation, archaisation, sociologisation (when natural phenomena are taken for cultural ones), and naturalisation (when cultural objects are not recognised, they are taken for natural ones). Examples are, correspondingly: the pictured pebbles from Mas-dAsil, int parts of modern threshers, and eoliths. Antipov did not nd an example for the fourth kind of mistake, but one can easily adduce such an example: the ancient pots which, in Poland, were initially taken for naturally produced objects. Thus, the determination of archaeologicality is founded on theory and at the same time this is an established initial task for archaeological studies. According to differently formulated theories this task will be solved differently! When archaeological theory, diffuse and unclear though it was, was considered by archaeologists as a part of geography (and there was such a time), then sites were considered as archaeological objects they included barrows, hillforts, citadels, and temples. When archaeological theory developed in art history, it was mainly monuments of ancient art that were added to the range of archaeological objects. The orientation of the discipline towards history gave birth to the notion of the archaeological source or archaeological record and prompted archaeologists to study all antiquities, not only aesthetically expressive ones. Despite this, merely the unique, spectacular, or especially eloquent monuments were taken into account seriously. For mass material (pottery, int akes, etc.) to be included in the archaeologists scope, it appeared necessary to realise the importance of the history of culture and, as the French historians use to say, the history of structures. Then not only proper

traditional general scientic notions of the functions of theory how signicant are they in archaeology, and to what extent are they incorporated in it?

1. THE IDENTIFICATIVE FUNCTION In such an approach, one function which is absent in the general reasoning of philosophers, the identicative function, must be included in the list. The allocation of the eld for archaeological study is itself established in the theoretical sphere. Consequently, so too is the general delimitation of appropriate materials for the study, as well as the attitude to be taken to them in brief, the determination of the sphere of archaeology. What we consider as the object of archaeology, what constitutes (pace purists of the Queens English) archaeologicality is determined by theory. If there is no clear theory, then the direct consequence is a confusion in attribution of some objects to the competence of archaeologists. Hence, debates on applicability of its methods to those objects occur. Archaeology has long been established as a study of material antiquities, of ancient things. Under things, objects of material culture are understood, the material products of humanity. Thus, archaeologicality (it sounds better in Russian!) is dened by two characteristics: belonging to material culture and antiquity. Neither concept is as simple as it may seem. Both demand empirical substantiation and theoretical elaboration. The characteristic of belonging to material culture is ascertained by the presence of attributes of human treatment, as understood on the empirical basis of generalised results of experience and observation. But culture presupposes, if not a purposeful, then at least an expedient treatment in accordance with certain cultural norms and traditions. This means that theoretical analysis of the concept of culture and its subdivision, material culture, is needed, as well as of its components norms and traditions. Antiquity is ascertained by empirical analysis too, by the selection of attributes derived from the long experience of the discipline. However, such empirical generalisations do not dictate what date an object must be for it to be recognised as an archaeological object, how ancient must a thing be to be archaeologi-

Metaarchaeology
things to study became the object of archaeological study, but all material trace of human activity. Debates about the extreme chronological limits of archaeology how far its competence stretches and at what point it begins usually display an abstract nature. Too great an exactness was not usually demanded. In such discussions another matter of debate is the material origins of archaeological data to what extent and in which cases should natural objects not produced by people (i.e. wood, bones, pollen etc.) be considered archaeological data. Such debates are more important, for they concern issues of the delimitation of the discipline and of the integration of these materials into it. These are issues of professional specialisation and of the approximate scope for each discipline. All these problems are connected to the key question of how archaeology recognises its own material, that is, with the ascertaining of the archaeologicality of objects the establishment of the concerns and limits of archaeology.

89

2. THE INTEGRATIVE FUNCTION This function follows on from the preceding aspect, in the sense that it relates to the same material, is based on the same statements, and is also connected with the processes by which the concerns of archaeology are formed. The unity of archaeology is the question here. Archaeology originated in many countries and not as one discipline. Its branches appeared to be independent disciplines, even belonging to different groups of disciplines. This is shown in the fact that special institutions, journals, university chairs, and societies were created. So, classical archaeology was, and is, developed in nearly the whole of Europe separately from primordial (prehistoric) archaeology. The rst was shaped in the wake of art-studies and philology, the second, however, established itself as a part of biology. In Germany it was guided during some decades by the palaeo-pathologist Virchow with his collaborators, who were also physicians. In Russia prehistoric antiquities were collected initially by Baer in the course of his study of human anatomy, whereas museums were allotted to classical archaeology. In the USA, prehistoric archaeology is part of the anthropo-

logical complex of sciences (parallel with physical and cultural anthropology, ethnography and linguistics), while classical archaeology is not; it is, in general, related to the humanities, not to the sciences. In Germany and USA these branches of archaeology are even named differently: the term archaeology is used only for classical archaeology, while primordial archaeology is named simply prehistory (Vorgeschichte, Urgeschichte). Oriental archaeology was developed in some places together with classical archaeology, in others separately from it. Medieval archaeology was always linked to history. Sometimes outside of the Mediterranean area mediaeval archaeology was combined with prehistoric archaeology to form a local archaeology which was seen as opposite of the classical sort. In Germany, about a century ago, the division between these branches was intensied by the competition and mutual hostility of these two branches to such a level that it produced a schism in the discipline. Some scholars are trying to substantiate this sharp division Kubler (1961) among the classicists and Chang (1967b) among prehistorians. So deep-rooted and entrenched in history is the division of archaeology, he writes (1967b, 137f), and the tradition has already carried so much momentum, that the status quo is inviolable, its rationalisation futile, and any grand design of confederating these rival approaches into a single master discipline of archaeology is foolhardy. Chang is aware of the possibility to search for common ground: The question is real, however, about whether a common methodology underlies all archaeological schools. Or, less ambitiously, are some of their concepts, methods, and terminological structures essentially interchangeable lenses for the same master camera?. He answers this question negatively. Hawkes (1954), in England, and Rogachev (1975; 1978), in Russia, suggested that the absence of written sources was the principal distinction of prehistoric archaeology from historical archaeology, as Rogachev called all the other branches. Hawkes not only divided archaeology into text-aided and text-free, but also introduced subdivisions into the former: protohistory, parahistory, and telehistory. These subdivisions and terms suggested by Hawkes did not catch on (except the rst one which already existed). But

90

Acta Archaeologica
fundamentally differed from the rest of the archaeological world: the comprehension of the MarxistLeninist scientic outlook and the development in the conditions of a socialist state created, as a matter of fact, a completely new discipline which has taken an honourable place in the ranks of other sections of historical knowledge ... (Rybakov 1967, 583). All this was instilled into the conscience of Marxist orthodox archaeologists to such an extent that the leading theoreticians of the establishment considered it possible to publish works on theoretical archaeology without using any foreign literature. This was not because using foreign works was forbidden (to some extent it was always allowed) and not only because they did not speak foreign languages (although admittedly most did not), but because they believed that, abroad, archaeology was another discipline, which had not reached the level of a proper science. There was no point in concerning oneself with it, may be simply to unmask and expose its folly. Western archaeologists often responded in the same way. Americans, in their archaeological works, ignored not only Russian but also German archaeological literature, though on essentially different grounds. Their next turn in direction aspired to be considered as the New Archaeology. These were not Changs rival archaeologies, these were hostile archaeologies, ghting with each other. But they still argued on the same basis, and often in the same scholarly language. As a rule, the more a school upholds its uniqueness, its monopoly on the right to be held as science (or scholarship), the more it is isolated and the more it falls into decay. No doubt archaeologists have plenty of philosophical and political orientations there always were plenty of these. But when some people, observing discord, speak of many archaeologies, it is useful to consider how deep a level of divergence of interests can condition preferences in the choice of facts and methods, as well as certain probabilities of error. All have to select out of one and the same set of methods, and use the same facts. Dissimilar in their convictions, archaeologists try to ll their concepts with different contents, but these are concepts derived from one system and contents from one world, and this difference in meanings is not so signicant as to prevent archaeologists from understanding what the thing is all about.

can presence or absence of written sources inuence the content and specicity of archaeological studies and their opportunities?, asks Zakharuk (1978, 25). The experience of Russia is very interesting in this respect: here archaeology is traditionally one discipline. It is remarkable that in a number of countries the term archaeology is kept in the names of both halves of archaeology. All workers in these disciplines are archaeologists, and they all use principally the same methods and main concepts. All these branches have, of course, one and the same theoretical basis. Childe (1956, VI) meant just this, when he declared archaeology is one. More than that, Childe observed that prehistoric archaeology outstripped classical and mediaeval archaeology, up to his own time, in terms of the development of concepts and methods. So he remarked: If Romanists and medievalists could be persuaded to adopt the techniques and the categories elaborated for older periods, many problems in history might be resolved. This problem also has another aspect. As soon as theoretical thinking is commonly connected with ideology, it is held that the ideological differences of scholarly schools in archaeology, of national schools, and so on, must lead to a sharp divergence in general archaeological theories, and hence of archaeologies in general. Pointing out the disintergration of New Archaeology, which pretended to be a Kuhnian paradigm for archaeology, B. Myhre (1961, 161) stated: There is not one archaeology or one prehistory any more, but many, depending on the philosophical or political theory of the archaeologist. The ideologies of Soviet archaeology were especially efcient in this schismatic activity. In the Soviet Union the belief was imposed in every way possible that Marxist archaeology was an absolutely new, quite different science as compared to all other archaeology, which in Soviet parlance was called bourgeois. It was said that it had different theory, different concepts, and different methods. The question was discussed whether such a science should retain the old term with its new content (Ravdonikas 1930, 20). A replacement of archaeology by the term history of material culture was considered, and for a short time this became practise. The old term survived, but the Soviet archaeological establishment long continued to assure everybody, and itself, that Soviet archaeology

Metaarchaeology
The theory of archaeology must allow archaeologists to see this situation clearly and must help them, despite all the differences, to keep their discipline intact, to keep the connections within its framework. It is not by chance that there are no special theoretical works applicable only to prehistoric archaeology. When a theoretical book by an archaeologist is published, whether on specicity of sources, or on classicatory concepts, or on ways of interpretation, it is sought with equal interest by archaeologists from all branches of the discipline.

91

and tasks. We need to supply these with materials, and information, as if for a questionnaire. It is sensible to gather information according to a broad set of questions. In recent decades the methodology of collecting evidence was intensively elaborated as a theoretical problem (sampling theory) in American and British archaeology (e.g. Binford 1964; Mueller 1975; Cherry et al. 1978). There is no need to prove that the selection of evidence depends on the special problem being studied. This problem either results directly from the dominant theory, or is due to contradictory information which, itself, produced the problem.

3. THE SELECTIVE FUNCTION This function appears to follow directly from the identicative one: after identifying the archaeological materials the choice and selection of which to use in each study is made according to the theoretical line to be adopted. If a zero theory (the principle of empiricism) is accepted, in other words, if there is no theory, then any material can be tted in indiscriminately, but in essence unwittingly on the part of the researcher, for it is selected according to his or her taste and other biases. However, it is usually held desirable to gather more data in order to provide a wider base for generalisations. Underlying this view is the idea that any material must provide something new. In general empiricists suppose that one has to strive for maximal recovery, ideally to collect all of the material. If a theory exists and the principles of statistics are included in it, then one can make do with representative samples. If theory is guided by hypotheticodeductive logic and accordingly the inquiry is aimed at the checking of a hypothesis, then the collection of material appears narrowly directed: only that is gathered which can conrm or refute the hypothesis. Of course, such an attitude is dangerous, especially in excavations. Since excavations destroy part of, or all of, the monument, some or all of its other material which is unnecessary for the inquirer at the moment will simply perish with all the potential information it contains. Finally, if one assumes, like Clarke, that there is a general theory, some set of principles, concepts, methods and procedures, common to all the subdivisions of archaeology, then it is reasonable to conclude that there is also a standard set of problems

4. THE DESCRIPTIVE FUNCTION This function is mentioned both by archaeologists writing on the functions of theory and by some philosophers. Archaeologists, however, in the main merely explain the place of description in archaeological studies and how it is to be conducted. The task here is different: to reveal whether it or some aspects of it are a function of archaeological theory, and if they are, to show this connection. The connection does exist. Description is nothing but the translation of information about material into the sign system for xating, storing, processing and transmitting it. Language serves as such a sign system for description. But the initial point at which description is supported by theory is before the moment scholarly description begins, because the language for it must already exist. It existed beforehand, and existed with all of its concepts with which it is able to mirror the material, its components, properties, measurements, connections, changes, movements, etc. That is to say, a sufciently developed categorisation must be established beforehand. Description and even primary systematisation are dependent on language, on pre-exising categories. As Caws (1965, 33) remarked, mastering a certain language can result in characteristics for its division and segmentation of what is immediately perceived. Indeed, to English-speaking archaeologists camp-sites are distinguished from settlements in that the rst were populated only for a short time, while the latter were places long inhabitated. Russian archaeologists traditionally distinguish stojanki (camp-sites or

92

Acta Archaeologica
layer. The partisans of Diffusionism (including Migrationism) could not do without the concepts localisation, site, and area. Ecologists began speaking of the environment and niche. Taxonomists elaborated the concepts of attribute. Understanding of the basic objects of archaeology also changed initially it was antiquity, then monuments and relics, now artefacts and the archaeological record (or archaeological source). First and foremost, natural language appears in the role of the language of science. Phenomena become described with everyday words and expressions, usually without any indication of quantitative characteristics for the aims of comparative analysis of styles (the direction was evidently theoretical!). Winckelmann set out some rules for scholarly description of ancient arts by means of natural language, and these rules completeness of description, attention to small points, personal survey (to describe de visu) etc. are still in use, and not only applied to monuments of the arts. However, for all its force and exibility, natural language displays a number of shortcomings when playing this role it is too unspecic, diffuse, redundant, and emotionally coloured. Special branches of language, therefore, come into existence, a sort of scientic or scholarly slang. One of their peculiarities is the presence of special terms for scholarly concepts, including concepts intended for scholarly description. Ideally, these terms must be inter-coordinated and compose a united system, at least within the framework of a given study, but preferably within the frame of the entire range of sciences or humanities. Terms become conventional and if theory is well considered they are specic and clearcut, supplied by full denitions. The other peculiarity of a special scholarly language is the dry, strict, laconic, and neutral style, which avoids every ambiguity, or latent metaphoric meanings, and substitution of emotions for argumentation. The third peculiarity is the use of abbreviations instead of often listing words, terms and expressions. The fourth peculiarity is the application of formulae for showing relationships and ties, especially in the cases where quantitative measurements and calculations are possible. In archaeology such style of description was elaborated gradually, beginning with the records of collectors and travellers, but was fully established only in

stands) as Stone Age settlements from selishcha (remains of villages) as settlements of the early Metal Ages even if the rst were inhabited for longer periods than the latter. Only the date of the site is taken into account, not the duration of its life. Besides, both categories are included in the Russian terminology into poselenija the general category settlement, including hillforts and towns. Categorisation exists in articulated language from the very beginning. It grows and evolves spontaneously. However, archaeological material has its own claim to categorisation. Firstly, many archaeological objects are things not existing in daily life and therefore have no correspondence in the actual system of categories. Secondly, nearly all archaeological objects have suffered from the effects of time and natural processes. Hence they have a peculiar appearance not met with in usual practice. Thirdly, archaeology is interested in properties and connections which usually do not attract attention outside of archaeology (for example, presence of patina, striking bulb, stratication, and so on). In the development of a discipline (in particular of archaeology) the language of scholarship is, therefore, being perfected all the time. This perfection consists mainly of the enrichment and renement of categories, in the ordering of their totality, and in replenishment of it with abstractions, setting out the interrelationship of categories more precisely, and organising them into a more and more strict, balanced and rich system. This is, of course, the business of theory. Our usual language is lled with theories, Popper (1965, 59) has said; observation is always observation in the light of some theory. There are philosophers who reduce theory to the language of science. Although this is an exaggeration, an extreme view, the elaboration of the scholarly language, particularly the language of description is undoubtedly one of the tasks of theory, and such language is one of its components. Change in the leading theoretical concepts (the -isms of archaeology) introduced into everyday practice newer and newer concepts and terms of descriptive language. So, since the time of the earliest adherents of the concept of the Three Ages, other concepts have entered the language of archaeological description such as assemblage, in situ, and

Metaarchaeology
the second half of the nineteenth century in museum catalogues and eld journals of expeditions. It began to be applied in published expedition reports, too. In Russia Gorodcov in the early twentieth century applied the tabular form of data presentation. As a unied language of description in the discipline it is still absent in the second half of the twentieth century, large archaeological works are often supplied with terminological glossaries containing denitions of the main special terms (e.g. see Klejn 1980, 104; 1991a, 343). Finally, in the recent times, under the pressure of computerisation, specic languages have been formed especially for the aim of scholarly description descriptive (document, informational) languages which are distinguished by extreme formalisation, explicitness and structured organisation. A reduction of description to its simplest operations forms their basis. Description is here conducted with special symbols according to an expanded code.

93

5. THE SYSTEMATISING FUNCTION To some extent this is the logical continuation of the preceding function because, in the process of categorisation, every description not generalises (this is an empirical operation) but also covers the material with categories created within a certain ordering system, no matter how it orders material, and even if only primitively. Strict intercoordination and subordination of these categories in accordance with the specicity of the material and with the tasks of the study builds the scholarly scheme of ordering. If on the basis of similarities and differences the material is distributed into strictly delineated cells, covered by standard categories, a scheme of classication results. If the material is clustered around concepts on which the centres of such categories are xed, a scheme of typology is built. How to organise the totality of classicatory or typological concepts in due accordance with the specicity of the material, and with the tasks of the study, are of course decided by that system of propositions, that program of actions, which properly is theory. To the extent to which this theory regulates these actions, it is called theory of archaeological classication (cf. Kolpakov 1992) or theory of archaeological typology (Klejn 1991).

The following must be stipulated: the distribution of empirical material into cells of classication or its grouping around ideals being called types and styles are empirical operations ascertaining similarities and differences, identication, and generalisation. But the organisation of the system of concepts itself, allotting them to hierarchical levels, delimitation and intercoordination, as well as their subordination to the general tasks of the study all of these are theoretical actions. There are a number of difcult questions here. Is it possible to combine the tasks of classication and those of typology in a single scheme? To what extent are the cells, the grouping, or the borderlines which we ascertain in the archaeological material inherent to that material itself, and to what extent do they depend on the arbitrariness of the inquirer? Do they mirror connections and delimitations that really existed in the past, in the onceliving culture? One might cite further questions. Quite a number of these questions have been extracted by Chang from Kluckhohns article on typology (Kluckhohn 1960; Chang 1967b, 83), and some more can be seen in other works (cf. Hill and Evans 1973, 231, 268; Klejn 1991a, 13, 31f). All of these are theoretical questions. Classications were made by collectors of antiquities in the Renaissance and even in classical antiquity. Even then, many of their classications sorted things by their functional purpose, according to analogies with things that existed in daily life weapons, tools, and ornaments as seemed appropriate. To this, Winckelmann added classication on stylistic principles. In the time of Evolutionism, distribution by supposed functions became the main kind of classication. At the same time typologies began to appear, where the material was grouped by stylistic and, in general, formal similarities and differences only within categories separated by the supposed functional purpose of things (and only for such categories was the term category used in Russian archaeology). That there is a hypothetical basis of this categorisation indicates the involvement of theory. The point is not the future of every hypothesis. Indeed, after being conrmed, some go on to become fact. Hypotheses about function are of this sort. The process of ascertaining the functional purpose of things is based on notions of the functional organisation of an ancient

94

Acta Archaeologica
6. THE PROBLEM OF EXPLANATORY FUNCTION This function is present in the lists compiled by both archaeologists and, even more commonly, philosophers. It receives the most attention from all of these scholars, whether archaeologists or philosophers, and is regarded as the main function. The connection of explanation with theory is difcult to deny if one holds that explanation directly appeals to laws, then laws are the kernel of theory. If, however, one understands explanation more broadly and questions the equation of history with sociology by their shared interest with laws, then the problem becomes more complex. The essence of explanation as exposed by the most rigourous, most neo-positivist wing of the American archaeology (cf. Spaulding 1968; Fritz and Plog 1970; Watson et al. 1971) followed the philosophers Hempel and Oppenheim: there must be at least one scientic law by which the facts to be explained (the explanadum) can be covered. Explanation through law is called nomological or deductive-nomological. The position of Deductivists has been subjected to a devastating critique (e.g. by Schuyler 1973; Morgan 1973; Levin 1973). Three main types of explanation are mentioned in the literature causal (in terms of cause and effect), structural-functional (nding the place in the structure and ascertaining the functions), and genetic (ascertaining the origin). Refusing absolute determinism in culture, one group of researchers inclines toward the idea that in archaeology structuralfunctional explanation is applicable (M. Salmon 1982a; Renfrew 1982), the other part persists in causal explanation as a special kind of probabilistic explanation most appropriate to the historical disciplines, including archaeology (W. Salmon 1982; Smith 1982; Eggert 1982; Mellor 1982; Trigger 1982). But what is liable to explanation in archaeology? Binford, who was among the rst to introduce the deductive-nomological principle into archaeology, thought initially that it was events and processes in socio-cultural systems of the past that were to be explained. Later he thought that, before one begins to study the cultural process through the sources, one has to explain the sources themselves. But as he still considered that one had to explain them through the events and processes of the past, then the problem

culture and the means of its cognition. This is theory. In culture, the connection of function with form is not determined rigidly. Usually there is a certain freedom of choice in selecting a form for a given function. We archaeologists are forced to ascertain forgotten functions from the present excavated forms. The task in every case is extremely difcult. We have to search for the key in the system of forms of the given culture itself and in the supposed system of functions corresponding logically to this. We must already have some general notion on the given culture, some expectations about it in respect to functions which were performed in it. This means that the discovery of ancient cultural structures of our archaeological materials by means of purely empirical procedures is impossible. This was noted rst by Taylor (1948) and this is the central idea of my work on typology (Klejn 1977c; 1982; 1991a). To this observation concerning the dependence of the earlier stages on the further stages in an inquiry (a sort of a paradox of outstripping), one can add that, by becoming standard, the classicational cells soon move into the professional language of description. Then there is no need in description to enumerate the attributes of a thing a whole permanently repeated complex can be replaced by the simple attribution of a type that was formerly determined. Formerly determined here is the clue to the paradox of outstripping the initial steps of an inquiry, resting on a later stage of the study, draw information from the results of earlier steps of general inquiry more advanced than the earlier stages of the special study being undertaken. Systematisation is inherent in theory by its nature, by denition so to speak. For in some essential aspects theory is just a system of laws, concepts and conclusions. Theory systematises not merely concepts, but also empirical regularities or laws, trying to establish relations of intercoordination and inference between them. The most organisational perfection in theory is reached when it is able to derive all its laws from one single general principle. To do this they must all be connected with this principle by logical subordination, that is, they must be explained by it. In other words, this level of systematisation is unattainable without explanation (again the phenomenon of outstripping!) and, properly speaking, it appears to be an aspect of the explanative function.

Metaarchaeology
appears to be a vicious circle. Binford came to the conclusion that a special discipline for the study of archaeological sources is needed, and a corresponding theory of archaeological record (Binford 1982). For him both of these steps of cognition the cognition of sources and the cognition of the past itself are steps within archaeology, but to me, as to Rouse (Rouse 1972; Klejn 1976; 1978b; 1991c, 1993) they must be separated. Archaeology is a discipline studying material antiquities as sources of information about the distant past, whereas the study of the past itself, of its events and processes, on the basis of processed and prepared sources is palaeohistory (prehistory and most ancient history). In establishing itself as a mature discipline, archaeology learnt to study not only each individual artefact, but the entirety of material culture. Archaeology studies the sources for the understanding of the cultural-historical process. It is not difcult to observe that for the time being debates in the archaeological literature are about the nature of explanation in palaeohistory, not in archaeology. Explanation has been concentrated into only a part of the range of understanding of the past on the peripheries of cultural-historical process. As soon as archaeology appears a special source-studying discipline, the question of the nature of explanation must be considered anew. At this rst stage in understanding the past we are interested to know what an object that we have discovered was, what its functional purpose in an ancient culture was, and of which ancient events and processes does it give evidence. These questions are applicable to any amount of material whether a single artefact, or many similar artefacts, whether an assemblage or a group of uniform assemblages. In the research design of archaeology this is the task of culturalhistorical interpretation, a task which is akin to the tasks of translation from one language into another, and also to the tasks of a detective. But it is not at all the task of a historian! Thus, in application to archaeology proper, the explanative function appears the interpretative one, and this puts archaeology outside of the historical disciplines in respect to its immediate tasks and methods, and its methodological nature, although not of its ultimate aim. What, then, is the nature of explanation in archaeology? That is to say, what is the mechanism of interpreta-

95

tion? Since events and processes that we were able to explain the appearance and content of archaeological sources occurred in the distant past and are inaccessible to direct observation, the main type of explanation in archaeology must be explanation by analogy that is, by using a model. Analogies in archaeology (including ethnographic ones) have received less attention than they deserve (cf. Ascher 1961; Smolla 1964; Morwood 1975; among others). With some exaggeration, but not without grounds, Chang (1967b, 107) declared: archaeology as a whole is analogy. Despite the enormous number of archaeological books and articles with the word model in the title (Klejn 1973, 7577), still less has been written on models proper. From the works of philosophers, and of methodologists of science, it is known that explanation by analogy belongs to the group of inductive, probabilistic methods (Nikitin 1970; Popov 1972, 181183). Besides explanation by a model, other approaches are also applied in archaeology. As soon as one has to ascertain the functional purpose of artefacts and features, one cannot do without functional-structural explanation. It is also produced, in the main, inductively. As soon as archaeological objects come to rest in a condition or in a place that was not typical of them in the past (when they were working in the culture) one has to do some explaining. In order to comprehend their functions one must mentally move from their present state and place to their past state and place; one must reveal what has led them to their present place and state, why they occur where they do, and so on. This requires causal explanation. The causal form belongs to the deductive group, although it does not immediately require a universal law, or necessarily presuppose absolute determinism. Now, it is possible to ascertain whether such explanation is the function of theory, that is, to decide what the role of theory in archaeological explanation is, and of which theory. As it is known, theory presupposes the presence of a strong law or a block of laws. Let us begin with causal explanation. Behind every cause some regularity, even if probabilistic, is hidden (Kon 1969), and every scholar seeks to strengthen the probability of their explanation being correct. So, in the end, behind the cause looms an absolute law (Mellor 1982) as an unreachable ideal. The causal

96

Acta Archaeologica
predict new archaeological discoveries, the places of new nds, and their appearance. But, rstly, such specic predictions (they are called heuristic) rest more upon empirical generalisations than upon theory and general laws. These are simply inductions, broadening a generalised fact. Secondly, it is clear that while these new nds belong to the future, it is not their past appearance which is to be predicted (it is already known), but new discoveries. This means that not the past of the nds is predicted, but the future of the study. Other scholars (for example Nikitin 1966; 1970; Binford 1972, 333f; Trigger 1973, 105; Onoprienko 1976) have chosen an opposite approach in order to retain the past as the point of the function, they sacriced prediction as a form of action appealing to the future and speak instead on retrodiction, or postdiction, inferences appealing to the past. Binford (1972, 334) explains: The only way ... contemporary observations are converted into statements about the past is through our ability to retrodict to establish on the basis of present evidence what were the conditions in the past which produced the contemporary observations .... But how do these inferences differ from usual explanations of the past? Many rely on the propinquity of the prediction, or retrodiction, and explanation (Dray 1957, 2; Hempel 1965, 176; Nikitin 1970, 222238; Trigger 1973, 105). Their logical structure is identical; both are inferencesa and both hypothetical ones. Explanative hypothesis is to be checked through confronting the expectations inferred from it with independent facts. Further, what is expectation? This is just prediction (or retrodiction) with a stressed uncertainty (Zubrow 1973, 246). On the other hand, prediction or retrodiction, made merely on the basis of analogy or correlation (that is, an empirical way) remains vague. It becomes strong only if the basis receives explanation from law, from theory. Retrodiction and explanation do not coincide. The distinction consists of the fact that explanation presupposes the linking an object with many other objects, whereas retrodiction, undertakes to describe, even if only approximately, supposed objects of the past (or of the future). But what does it mean to describe supposed objects of the past on the basis of its modied remains, frag-

explanation presupposes laws of dynamics, whereas the structural-functional one presupposes laws of the static aspect of structure. Finally, explanation by analogy as it appeals to particular occurrence, seems to avoid law. Behind the object and its model, some set of similarities besides those which underlie the approximation are presupposed. Consequently, analogy presupposes that both objects are encompassed by a regularity that determined their resemblance. Of course, one seldom appeals to this regularity directly: it is merely implied. In an obvious way other regularities are acting here, more general ones. They are probabilistic, guiding the choice of the model for the specic object or site and, in general, relations between every object, site, and model. It is already evident that the basis for all these explanations forms a theory: a theory of archaeological record (of archaeological sources) and, more broadly, a theory of interpretation. This leans upon empirical generalisations from experimental archaeology, ethnoarchaeology and ethnography, as well as upon that theory of synthesis which lies behind historical reconstruction and unites all the source-studying disciplines as a foundation of history.

7. THE PROBLEM OF THE PREDICTIVE FUNCTION Philosophers dene the predictive, or prognostic, function as the most important function of every really scientic theory (Ruzavin 1978, 23). Only the presence of the predictive function, elucidates Bazhenov (1973, 416), prevents theory from being selfcontaining, closed in itself, and lets it become practically useful. Apparently, there is some difculty for the realisation of this function in archaeology. This difculty probably results from the fact that prediction is directed to the future whereas, according to Triggers (1970) apt expression, the future of archaeology is the past. How then can predictions be made, when prediction belongs to the past? Some scholars (for example, Rakitov 1982, 289), striving to retain the function of prediction intact for historical and archaeological theory (since each decent theory must have a predictive function!), modied the past. They retained from it only some hints, and substituted the future for it. They say one can

Metaarchaeology
ments and traces in the present to describe it as such as it looked in the past? This means carrying out its mental reconstruction. However in many cases, when fragments are small and there are few of them, a great freedom of choice arises. In piecing them together one can compose them differently, and the traces can be differently interpreted. This, in turn, means that besides facts, imagination and ideas participate in reconstruction. If they are strictly organised and substantiated, then this is theory. Thus, in archaeology the predictive function appears to be a reconstructive one. The specicity of history, to which archaeology relates, means that it cannot abstract itself too far from reality, it cannot refuse detail altogether. It is directed not to revealing the general laws, but to establishing specic facts of the past in their causal connection (Dray 1957; Trigger 1973). Consequently, historical reconstruction proper is possible only in rare cases when it concerns the recent past, very specic situations, and those well provided with sources. In the majority of cases, in history, not properly historical but sociological or anthropological (culturological) reconstructions occur. As for archaeological reconstructions, these do not usually pursue the task of exactly and minutely restoring individual aspects. Even as applied to single artefacts they content themselves with restoring type-characteristics. So, in archaeology reconctructions are quite real, but the degree of their completeness, and adequacy, is inversely proportional to their connection with purely historical tasks.

97

forms with epoch and came to the stratigraphic method. The theory of stadiality encouraged interest in productive forces and, having fought against the formal things-study (a translation of an ironic Russian term for artefactology or potology), produced Semenovs functional-traceological method. Processual archaeology, with its interest in a systemic approach and models, introduced multivariate methods and computer simulation. Behavioural archaeology gave rise to branches (including ethnoarchaeology and experimental archaeology) specially aimed at the elaboration of interpretative methods.

8. THE INSTRUMENTAL FUNCTION Under this term I mean the transformation of theory into method. The history of science, or of scholarship, teaches that theory is akin to method: driving its operational apparatus to the level where it becomes stereotyped, theory turns into method (Ovchinnikov 1968, 20f). In not one list have I found this function. Yet, it serves to bring theory into practice. So, proceeding from the theory (or, pre-theory) which correlated style with epoch, Winckelmann and Gerhardt elaborated fundamentals of the comparative-stylistic method. The Danish progressionists Thomsen and Worsaae, turned their scheme of the Three Ages into the method of correlating artefact

9. THE HEURISTIC FUNCTION Kaplan (1964, 268, 302), an American philosopher and a specialist in the logic of science, writes: Every theory serves, in part, as a research directive; theory guides the collection of data and their subsequent analysis, by showing us beforehand where the data are to be tted, and what we are to make of them when we get them ... Theory functions throughout inquiry ... it guides the search for data and the search for the laws encompassing them. Archaeologists also note something of the sort. In their statements some heuristic directions are mentioned. The rst one is, so to speak, intrinsic, in the frame of the chosen theory. The main purpose of a theoretical model, writes Redman (1973, 20), is to help the researcher to select relevant variables and signicant hypotheses from a large number of possibilities. Mainly this is work in the existing paradigm, following the terminology of Kuhn. The other direction is the external one. It is a search for new ideas, for new theories. In his entertaining article on the interaction of typical gures in archaeology during the formation of every next paradigm, Schwartz (1978) has separated a gure, Searcher, who initiates the whole affair by nding new, unexpected data, not predicted by the current theories. Thereafter the results come to the main gure, Genius, who creates new ideas, new explanations, new theory, and who thereby sets into motion the entire team-Systematiser, Advocate, Contributor, Teacher, and Eclectic. The scheme is very nice, but the beginning is not quite aptly pictured. Nobody can deny the value of new data, new facts, especially if they are unexpected, but experience

98

Acta Archaeologica
stress that all of his new ideas grew from processing old ideas and old theories. If we turn to Walter Taylor, we nd behind him the gure of his teacher Clyde Kluckhohn. Kluckhohn supervised the writing of his well-known book when it was still a dissertation. Many scholars borrowed ideas from Taylor, not only among Processual archaeologists but also their opponents, Contextualists like Chang and Trigger, or the new opponents, the PostProcessualists, such as Hodder. David Clarke had his own intellectual antecedents in archaeology and outside of it Childe, the philosopher Wittgenstein, new geographers, numerical taxonomy, and systems theory. Thus, to a very great extent theory gives birth to theory. A new theory results from the processing of old theories. As Steward (1949, 25) has put it, theories are not destroyed by facts they are replaced by new theories which better explain the facts. This is so partly due to the addition of new ideas from other disciplines and from life in society, and partly because the growth of the discipline, the development of its concepts has its own logic. Their succession and change are not fully fortuitous. Evolutionary typological series themselves drew on the theory of catastrophe as applied to archaeology. So each existing theory has got much potential to condition new theories. One must only know the theories and be able to work with them. On this heuristic aspect Randsborg (1982, 425) draws a general conclusion: he proposes to proceed inductively using various theories as indicators of possible further directions of research and to form approaches rather than dogmatic, static frameworks.

teaches us that their signicance, at least in archaeology, is rather in checking the old theory in order to correct it, to modify it, to limit it, or even to disprove it. Its opponents are not necessarily innovators, they can be the partisans of one of the old theories. This is quite understandable: the unpredicted fact exerts a negative inuence it denies and refutes but, by itself, it is not a positive contribution to theory it does not create new ideas, it does not provide new explanations. A new theory usually originates in another way (for theoretical analysis of this problem see the preceding chapter and Klejn 1980). The principal new ideas for a new theory must come from outside the discipline, or be produced inside by a creative individual able to combine and process a lot of information, often alien to the discipline, but sometimes produced inside this time arising from the confrontation of existing theories. Let us look closely at the examples with which Schwartz has illustrated his Genius gure Taylor and Binford. Lewis Binford is the obvious example of an apparent paradigm-maker in recent American archaeology. His work has attached many young scholars, to the extent that they think of themselves as the school of new archaeologists. His role has stimulated new work by others ... (Schwartz 1978, 158). However, I have not found in the article an example of Searcher who supplied Binford with unpredicted facts from the very beginning, nor have I found there indications of such facts themselves. Moreover, I have not found them elsewhere. Apparently, they did not exist. More than that, it appears that in the initial creative phase, Genius was oriented, in general, not towards new facts but towards old ideas. Yet, how new are his ideas?, asks Schwartz (1978: 158) about Binford. They certainly were foreshadowed at least by Steward and Setzler (1938), Kluckhohn (1948), Steward (1949) and Caldwell (1958). Binford (1972, 219) himself points to Walter Taylor, Albert Spaulding, and Leslie White as the sources of inspiration for his ideas and he describes very picturesquely how he used these inuences in creating his innovative programme. One may add to his antecedents Neo-positivistic philosophers, notably Hempel and Nagel. I do not want to say that Binford was not mentally independent or that he created nothing new. Far from it. I wish only to

10. THE CONTROL FUNCTION To advance dogmatic, static frameworks is also a task of every new theory and one cannot escape this. The point is not in creation of a paradigm for normal science, following Kuhn, this being a consequence of other tasks. The thing is that every scientic (scholarly) law is more absolute or, if it concerns probabilistic laws, the stronger (i.e. the probability of it being correct is the higher), the more limited is its sphere of application. Laws not only dictate or, if probabilistic ones, permit, but also forbid.

Metaarchaeology
The limiting role of a law inuences essentially its predictive functions and is most manifest itself in theory, writes the philosopher Vinogradov (Popov 1972, 189). And he continues: besides classes of assumed phenomena theory sets a class of forbidden phenomena ... A theory in which nothing is forbidden can neither describe nor predict anything. Therefore the process multiplying the principles of interdiction in theory is connected with broadening the class of phenomena being described, explained, and predicted. If, nevertheless, a fact is reliably conrmed, then it forms the basis of revision of the theory which forbade it a basis for a limitation or abolition of the theory. So the conrmation of the Palaeolithic age of cave art essentially undermined the authority of Evolutionism, which believed in the gradual progress of the human mind and so forbade the existence of too early art. Evolutionists therefore declared it a forgery. However, unless condence in the theory is not undermined, and while its basis remains broad and stable, it serves as the criterion for verication of new facts and new laws. Thus, theory appears a criterion of the scholarly range of new discoveries and directions in archaeology. More than that, however, theories, although they contradict each other, nevertheless form some theoretical fund consisting of contributions from various theories into the theoretical picture of the world, by correspondence with which we evaluate new discoveries. Only very stable (and the most reliable) contributions from various theories go into the theoretical picture of the world principles, concepts, and methods of verication. This is why we reject with clear and calm conscience all claims for the abominable Snowman, Atlantis, ancient astronauts, and so on. What rests behind this calm condence is our knowledge of the theoretical (or scholarly) picture of the world, the knowledge that we have to correct and replenish, and this also goes into the control function of archaeological theory.

99

11. THE SYNTHESISING FUNCTION Any theory synthesises the material it condenses the empirical information and simplies its mosaic. The trend to principal simplicity is expressed in

that theory seeks to explain as wide as possible a range of appearances on the basis of as few as possible premises. Therefore, a trend to expansion is inherent in any theory: theory begins penetrating into the area of competence of other theories and disciplines. It is especially noticeable when this trend is realised by the rise of new theories: each new one has to show its validity by being able to explain also those phenomena that the previous one could not. However, the factor of retaining results acts here too. As Regirer (1966, 95) wrote, it is only in preliminary stages that one simply throws out the existing theories in order to free a place for the new ones. When theory has been conrmed by experience of facts, usually it is modied, so that it is included into the new, bigger theoretical conception as a particular case, with all its results conrmed by the experience. When a success is reached in the explanation of facts, passing on to another theory appears possible only if the number of the facts explained does not decrease, but quite the reverse, if it increases: the attained explanations do not get so easily rejected by science (or scholarship). I have always been amused by the vehemence with which adherents and enemies of migrations or independent development advocated their own explanations as a principle and rejected fully the competing explanation. Whereas all that was actually needed was to determine the conditions under which one or another explanation appears correct. Inuences and borrowings on the one hand, and migrations on the other, long competed with each other as explanations of the sudden multiple appearance of new elements in culture. Albeit from the very beginning some researchers used one or the other of these explanations at choice, as if equal both phenomena are, as a matter of fact, alternative kinds of diffusion. However, diffusion too, is merely one among many possible explanations of culture change evolution, cultural and social revolutions, environmental effects, and others. They might be considered as potential parts of a more general theory of culture change (or theory of the change of cultures). This theory, in its turn, may be viewed as a part of a still more general theory of archaeological interpretation, and so on.

100

Acta Archaeologica
Russian we say by the method bustling about). Theory evolves before their very eyes as a scheme of interrelations in their material and an ordering scheme for their set of instruments, it determines their aims and tasks. Of course, both schemes are incomplete, in some respects distorting the picture. But from eld experience we know that it is better to use imperfect maps than to have nothing and to travel without maps at all. The more so as that nothing often means the blind condence in a chance guide.

12. THE ENLIGHTENING FUNCTION The last, but not least, function of archaeological theory is to enlighten. None of the scholars cited before mentioned this function, probably because it is the most evident. As one scientist (Regirer 1966, 101) said, theory is like a light-tower on which oodlights are mounted to illuminate a building site. Indeed, without theory, however experienced archaeologists may be, have either to subordinate completely to tradition, or to search by trial and error (in

Metaarchaeology PART V. ARCHAEOLOGICAL FACT AND RESEARCH DESIGN

101

10. Archaeological fact

1. FROM FACT TO DATA What constitutes an archaeological fact?, three modern archaeologists, including Binford, ask and reply: We maintain that in reality the record contains no facts, if we take this term to mean given empirical truths (Sabloff et al. 1987, 203). They imply the known motto of new philosophers that fact is laden with theory. Even such avowed empiricists as Sir Mortimer Wheeler (1960, 199) understood that there must be an element of theory there, if our facts should mean something. This shows that the problem of archaeological fact is not as simple as it was once thought. In the early 19th century the British excavator Sir Richard Hoare (1812) was proud of having not resorted to theory. We speak from facts, not from theory, he wrote in the introduction to his book. Fact had long been contrasted with theory as true and solid reality had with speculation. Later, as theories received weight in science, fact became considered as a necessary base of any theory, then as their touchstone, as a criterion of proof. However there is little in the way of theoretical works on this problem in archaeology. The rst articles on archaeological fact appeared in the West in the 30s of this century (Strong 1936; Steward and Setzler 1938), and in the USSR in the middle of the 70s (Viktorova 1975a; Zacharuk 1977; Sher 1985). Historians realised much earlier than archaeologists the complexity of the problem of scientic fact. Their understanding of fact undoubtedly inuenced the notions of archaeologists. For a long time archaeological fact (AF) was seen by archaeologists as something simple, hard and self-evident. Archaeology, like history, began with nave empiricism but lingered at this stage longer than history. The Curator of the Ashmolean Museum, J. H. Parker, said, archaeology is ... history taught with eyes, by showing a series of tangible objects (Parker 1870, 9). In this, V. A. Gorodcov wrote on stratigraphic data, one needs no subjective intervention of the investigator because each phenomenon, each thing should speak for itself ... (Gorodcov 1908, 11). The accumulation of errors and fakes showed archaeologists that the obviousness of an AF could be deceptive (Munro 1905; Arnau 1959 or 1961; Paul 1962; a.o.). There was nothing for it but to establish methods of checking the authenticity of archaeological objects. Although these methods were in unquestionable relation to the external criticism of written sources, archaeologists called it simply source criticism (for instance Jacob-Friesen 1928, 98; Kruglov i Podgaeckij 1935, 1431). Any other criticism of the sources was not thought of as necessary. According to the classication of historical sources, written ones fell into the bracket of tradition and were held as intentional and tendentious, while archaeological records were included in the category of remains and were considered objective and incidental (for a survey and critic of these views see Pronshtejn 1971, 17, 2325). H.-J. Eggers was one of the rst who noted that information contained even in genuine material antiquities is not completely adequate to reect the events of the past because not all data are maintained. Eggers noted the intentional nature attached to specially selected articles laid into graves with the dead (Eggers 195, 52). Eggers found three phases in the process of converting information: living material culture dying material culture long ago defunct, museum material culture (Eggers 1959, 262270). In general Eggers and other archaeologists of the BRD (E. Wahle, R. Hachmann) while revealing the limits of the cognitive possibilities of archaeology (Wahle 1941; Hachmann, Kossack and Kuhn 1962, 1628) refer to the distortions of the information that take place during the course of the dead culture. American and British archaeologists also noted the ltering of information that occurs in the rst and last stages of the process. What are the rst effects of ltering in the process? From W. Taylors viewpoint culture consists of

102

Acta Archaeologica
ivist philosophy of New Archaeology caused it to advance against empiricism and produced some ignorance of facts to the advantage of theory. Facts do not speak for themselves, taught Binford. From the very beginning facts are laden with theory, they do not exist by themselves outside the theoretical approach, sampling and processing. The main basis for the New Archaeologists custom to neglect a single fact is their anthropological and sociological training, their hunt for universal laws of the socio-cultural process. This is why facts are frequently interchangeable for them and important only so far as on their basis it is possible to infer some law. They are important only until such a law is inferred. Thereafter facts lose their meaning. They are selected with the special aim of solving some problem there is simply no need in other facts. With their interest in theory and aiming for laws the New Archaeologists were distinguished in this respect from Taylors direct followers, the Contextualists. The latter also admit the cognitive inertia of facts, their dependence on interpretation and their theory-ladenness. Yet they respect every single fact; for them the array of facts forms not the basis for universal law but a unique context. P. Courbin, an antagonist of New Archaeology, sees with irony that as a counter to the old, traditional archaeology, New Archaeology declares there are no raw, neutral, objective facts, there are no basic data, no unstructured samples, no data which could speak for themselves. Facts emerge only in the system of connection, in the approach system explicitly given previously ... everyone has and had always a system of connection ... (Courbin 1988, 119). Well, this is somewhat exaggerated. As we have seen, there were other approaches too. Even Courbin himself after disclosing New Archaeologists secret condence in facts obtained by traditional archaeology, concludes: In essence, although it is wrong in reality, is it not simpler to admit (from practical considerations only) that facts exist independently of the exact attitude to problems? (Courbin 1988, 119). As to the secret condence of New Archaeologists to facts, he denitely caught sight of it. It was conditioned by the following circumstance: New Archaeologys relation to facts depends rst and foremost on its theoretical orientation on the systemic approach. The

ideas, an ethnographer observes objectication of behaviour while an archaeologist is left with only the tangible remains of this behaviour (Taylor 1948, 97 115). Thus after Taylor AF is the third step from the essence of culture and in each step shifts of meaning and loss of information are possible. Childe (1954, 739) noted that many deeds (gestures, words etc.) do not pass into fossil state and do not leave durable results after them. Neither language, social relations nor ideology are deposited in archaeological materials, and this, G. Daniel writes, is the fundamental limitation of prehistory (Daniel 1962, 127f). J. Grifn used to say he never saw anybody who was lucky to excavate the kinship system (Binford 1972, 8). Ideas are objectied differently: much depends on the subjective factor the free will of personalities, which is alien to regular rules. How is the ow of information discussed in British and American studies? Here modern critics exaggerate the subjective contribution of an investigator. Yet not before long Childe believed: Cultures are observed facts (Childe 1936, 3) until Daniel challenged: Cultures of the modern archaeologist ... are only instrumental concepts (Daniel 1950, 319). Once again Krieger and Spaulding maintained the idea that, types are discovered in the material and characterise the state of facts (Krieger 1944; Spaulding 1953). J. Brew and J. Ford challenge this: types are constructed by the investigator and are imposed onto the material (Brew 1946; Ford 1954b; 1962a,b). David Clarke summarises these sceptical views as follows: In archaeology the only facts are artefacts (Clarke 1968, 41). However one cannot describe artefacts without covering them with the set of concepts held in the mind of the investigator. Clarke notes: These facts appear observations in which the nature of the observer and his intentions play a great role ... (Clarke 1968, 21). In such case there are no facts at all in archaeology, or one has to admit all its facts are to a large extent constructions of the investigator and, in this new sense, artefacts. These will not merely be ancient artefacts but modern ones too, our artefacts, not in the sense of being hand-made but in the sense of being articial. Clarkes sympathy for these sceptical views was not accidental. They reected a climate that reigned in New Archaeology. The neo-positivist and post-posit-

Metaarchaeology
New Archaeology sees in culture a system where all components are closely interconnected and interdependent. So for an archaeologist armed with the system analysis method it is no grief that facts of archaeology are laden with theory and contain elements of interpretation. Due to the correlation of facts in the system, this interpretation is hard and potentially unambiguous. An early stage Binford sorted artefacts into technomic, sociotechnic and ideotechnic, according to self-evident treatment of their content (Binford 1962). Later he admitted himself that this division was somewhat silly (Binford 1972, 17). Other New Archaeologists started searching for narrower determined correlates (Hill 1970, 63; Schiffer 1976, 12f). The depth of facts here is theoretically admitted but in practice the face of the fact appears at. Facts as before do speak for themselves though under their breath, in the investigators ear. Archaeological systems as they were outlined from the direct generalisation of facts were, in the New Archaeology, equal to cultural systems of the past as they had lived. Tracing these archaeological systems (by extrapolation and interpolation) through subsequent stratigraphical planes through time proles was considered as the true development. Any system changes along the way were assumed to be changes of cultures of the past, changes that demanded causes to be found in the past such as impact of the natural environment, of neighbouring cultures, or the activity of intrinsic factors. Meanwhile some of these changes in archaeological systems were not at all reections of phenomena of past cultural systems. Instead they were merely distortions introduced in these systems, post-mortem so to say, when they were purely archaeological systems: some whole fractions of material are lost because of unequal erosion of different materials. Unequal studies also lead to differences in results, etc. However a new approach was forming in the rst half of the 1970s, and more decisively in the second half, on the basis of the New Archaeology. This approach was called Behavioral Archaeology. It moved the centre of gravity from the study of cultural and historical process to the study of archaeological source formation. The essence was in realising the depth of archaeological fact. In the very rst scheme M. Schiffer pictures the passage of materials from living contexts

103

(he calls it systemic) to archaeological context: obtaining the raw material production use discarding deposal in rubbish (Schiffer 1972). Later the scheme became more detailed and elaborated on particular examples (Fig. 12 Schiffer 1976, g. 4.3). S. G. H. Daniels from Nigeria suggested the most multistage scheme of AF (Fig. 13 Daniels 1972, g. 5.1). His scheme follows the idea stated by Montelius long ago about the losses of information in archaeological cognition. Montelius (1888, 5) wrote: Only a small part of what once existed [1] was buried in the ground [2]; only a part of what was buried has escaped the destroying hand of time [3]; of this part all has not yet come to light again [4]; and we know only too well how little of what has come to light has been of service for our science [5]. Almost all of the nds of past centuries have disappeared without a trace, and even much of what has been discovered in the present century has been destroyed. Thus Montelius envisaged 5 stages of losses of factual material, indicated in the above passage with the addition of brackets. Danielss model of origins (or of conditioning) of archaeological information was formulated as applied to a single site and shows 7 levels: (1) Potential population of artefacts suggested by the cultural matrix; (2) Actually deposited population; (3) Its preserved part; (4) Part in the excavated unit; (5) Discovered artefacts; (6) Data taken into consideration; (7) Published data. Perhaps not without inuence from this work, which was given to Clarke for publication, he elaborated on his scheme of information movement in the levels of AF from the predepositional situation through to positional one and postdepositional to the situation of discovery and then to analysis and interpretation (Clarke 1973, 15f). In this setting Binford presents a scheme of relations more complicated than he did in the late 1980s, where this time he places fact, phenomenon, event, data, although not completely itemising every one of these concepts: In the science the term fact relates to aspects of actual appearing of the phenomenon. More import-

104

Acta Archaeologica

Fig. 12. Flow of information in archaeology (after Schiffer 1976).

ant is that scientists give the status of the fact to recognisable single event that was realised in a certain time. Fact exists in the event ..., which occurred once and then is gone forever, while data are presentation of the fact by means of relatively stable established documentation. If we admit the equalisation of facts with events then we must conclude that archaeologists can never work with the facts of the past. However archaeologists produce much data as the result of their study and of observation on the archaeological record... What events do archaeologists describe when they produce data? They x events of the observation in which they participate ... These observation records reported as data of archaeology belong to modern facts to modern events of observation. No historical facts (past events) are accessible to archae-

ological observation. Archaeologists produce data from the facts of contemporary observations on artefacts. Observations on one sherd or arrowpoint are treated as one event and consequently as the source of the fact. When the archaeological event is discovered in archaeological context ..., it seems probable that artefacts must belong to past events ... Nevertheless we cannot muddle implications with facts, There are no historical facts that remained for us in order that we have seen them and recorded (Binford 1989, 5557). Events proceeded in the past and became the property of history, or historical facts. Archaeological remains, artefacts, sources (or records) are preserved from them. Observations on these sources are events,

Metaarchaeology
but they are modern ones, or modern facts. Recordings of these events are data produced by archaeologists. On the basis of data we judge about past events, but data in archaeology are for us not facts but merely implications or inferences. In Binfords scheme, fact of archaeology has been laminated in the historical, an event of the past, and in the modern, as an act of archaeological observation in the present. In this situation there are two more stages in the movement of information, two more steps of fact, both reecting an event in material form. One is artefact (or source), the other is data (record of observation of the source). These last two steps are not called facts because only event deserves the status of fact after Binford (although in Latin, factum, literally means made or done). Binford notes four steps in the ow of information. Taking into account that events of the past do not exist, likening facts to events allows Binford to allot the status of fact with observation and to place observation over real objects. This is a neo-positivist line. In theoretical considerations by the New Archaeologists it is the last stage of the ow that features most strongly, that is, the data, as opposed to facts. The data is the only reality that exists for us while facts are merely implications of data. Binford carried the term data into the title of the article quoted above. He also used this term much earlier (a chapter from his book in 1972 was entitled Method of data collection), but it did not receive a place in the general scheme until the 1980s. The term is borrowed from mathematics and was initially coloured in a statistical hue. The statistician and archaeologist Daniels (1978, 31) even remarked about his archaeological work that: I use the term information rather than data, since data tends to suggest only tables of gures, and I am intentionally including ... apparently nonquantitative information. Gardin (1974, 18) considered the denition of the term data (donnee) without such overtones. To him data implies an indication of individual object in a community and a distinctive trait of this object, the trait being intrinsic (morphologic, technical etc.) or extrinsic (origine, context, date etc.). So, Binford had a justication in retaining the term data for the last step of the transformation of fact. This stress did not pass unobserved. R. Gould, whom Binford accused for his empiricism, noted an-

105

Fig. 13. Research design model (after Daniels 1972: 5.1).

other exciting idea by Binford that has failed because it lacks a reliable empirical basis (Gould 1985, 641). Binford rejoined: Gould is forced to infer that statements on the past are impossible so far as no empirical data remained for us which we could see. He notes: The past is gone, Richard. There are only modern statistical data. It is our task to understand so these modern facts that reliably reconstruct (to infer, if you like) the real world of the past. To insist that the real world of the past is the same real world that is present in archaeological records is a sheer nonsense (Binford 1989, 116). Nevertheless due to the belief in strict systemic organisation of culture and in the existence of laws (laws of cultural process and

106

Acta Archaeologica
modern western archaeologists who recognised the cognitive multistage nature of AF saw the absoluteness of its subjective side. Comparing schemes of Daniels and Clarke, A. Sullivan elects Clarke as his preference. Daniels proceeds by counting artefacts, which is treated simultaneously both as a channel of information communicating to an archaeologist and as the communication (or message) itself which has to be processed. Clarke analyses traces (in the wide sense), and these are the message, while artefacts are merely the channel through which the message is communicated to an archaeologist (Sullivan 1978, 193f). Sullivan does not use the terms clearly here and does not clearly differentiate the concepts. In the same survey he differentiates evidence or data from material remains: archaeologists obtain evidences (or data) from material remains (Sullivan 1978, 189). Evidence has been frequently translated as information and fact, while the term material remains is used here because, in English, the term source, which is more suitable here is not used. The term record is usually applied instead which does not t the designation a channel or reservoir: it means written information, and this is imprinted and xed. Fact and sources are more exact designations of those sides of the relation which are presented here by Sullivan as traces and artefacts.

those governing the archaeologisation of materials), Binford is convinced that it is quite possible to infer reliably the real world of the past from archaeological records. He considers methods worked out by him and by other New Archaeologists as adequate and reliable means for such inference. In the 1980s however, the ideology of the next generation began setting the tone in archaeology, at least in British archaeology. They had lost faith in systemic order and with this faith the Neo-Positivist optimism as well. The main theoretician of this new direction, Ian Hodder, summarised in his book Reading the past all the inferences of the New Archaeologists and their predecessors on the facts laden with theory and the dependence of facts on it, on the non-existence of pure facts and so on. The bare bones that are left, he summarised, are the facts in the real world which we can never observe (Hodder 1986, here 1991, 16). He rejected Binfords assurance of the reliability of his methods as independent instrument able to measure relations between ancient society and modern archaeological material. This is not true, declared Hodder, because many measurements depend on our perception and our categorisation, and they are very weakly connected with the material. Over and above there is no independent measuring instrument at ones disposal: the methods themselves are dependent on theory. There can be neither checking of theory with data nor independent measuring installation nor valid knowledge on the past. Hodder stresses the destroying effect of his inferences: ... the whole fabric of archaeology as a scientic discipline ... is threatened (Hodder 1991, 18). If the basis of facts of archaeology is so unreliable and inaccessible, from where does an archaeologist obtains his knowledge on the past? As a result, the theories one espouses about the past depend very much on ones own social and cultural context ... In other words, the data-theory relationship is conceived and manipulated within cultural and historical contexts (Hodder 1991, 17f). Thus, if in the beginning AF was perceived as something even more simple and evident than fact of history, in the end AF appeared in front of investigators as something far more complicated and deep. Having regarded AF as something simple archaeologists once saw its objective side as absolute, while

2. CONVERSION AND RECONVERSION OF INFORMATION In his article Archaeology: the loss of innocence (1973) David Clarke considers archaeological research as information processing (or elaboration), where information enters through archaeological channels. The information may be extracted from complex, integrated relationships encapsulated within ... sites ... from data (Clarke 1973: 13). He holds this elaboration of information as part of the general theory of archaeology it is just these ... steps which underlie the critical leaps in archaeological reasoning. Without such a body of theory these critical leaps do indeed take-off and become a free-ight of creative fancy an irresponsible art form (Clarke 1973: 16). Clarke was indeed impressed by Daniels article in Clarkes collective volume Models in archaeology, which was printed the year before The loss of inno-

Metaarchaeology
cence. This can be seen in Clarkes choice of particular levels being separated in any archaeological interpretation, namely: 1) Hominid activity patterns as well as social and environmental processes that once existed; 2) The sample and traces of these patterns, deposited contemporaneously with (1); 3) The part of that sample which survived and is recovered; 4) The part of the sample in (3) which was recovered by excavation or collection. Clarke suggested theoretical conceptions concerning the transition from each level to the next one: (a) (b) (c) (d) Predepositional and depositional theory (12), Postdepositional theory (23), Retrieval theory (34), Analytical theory analysis of the circumstances discovered which became data.

107

The theory consists of joining the data by means of models preserved in archaeology. To that Clarke added: (e) Interpretative theory which, from all these analysed data, concludes that the structures and processes are unobservable directly (Clarke 1973, 16f). Although Clarke has employed Daniels treatment, he remained with his four-level structure of archaeological information (he does not apply the term fact here). However, it traces clearly the ow of information from processes of the past through archaeological monuments and to the written report of an archaeologist. The full process of information conversion is the cognitive basis and logical structure of the archaeological procedure, as Clarke calls the archaeological research design. This conversion process leads from factual material (objects of archaeology) to its interpretation in terms of history. This is what united the sequence that lies at the basis of certain research design models: inductivist, deductivist and problem design. The selection of these schemes is determined by the philosophical and methodological position of the investigator. However, what conditions the struc-

ture of the subject itself the overall quantity of stages, their content and sequence? How is this structure connected with the philosophy of science? It can be shown that these characteristics are conditioned by the understanding of AF. If the fact is simple and may be reduced to what is seen on the surface, then merely collection, description and generalisation are necessary. If the fact is complex and contains not only information which we search for, and beside it there is much other information and that which we search for is distorted, then the elaboration must be more sophisticated and extensive. Critique, clarication and replenishing are needed. Archaeology studies material antiquities (both artefact and non-artefact ones) and relations between them. The necessary postulate of archaeology is to admit that its objects are traces and remains of the past, i.e. results of historical events and sociocultural phenomena. These traces and remains are considered to be information on and produced by the events and phenomena that once occurred. Information went through several transformations before it was deposited as facts of archaeology. Thus, it is not initial information but a transformed type. It now reects not only events, processes and phenomena of the past but changes in the reection too. Not only a history of a society is deposited in this information, but also its own history. AF is not simple or at, it has depth. This second history is not merely extra complication. To cognise it is necessary in order to rebuild the rst history. In order to establish what is initial information, one must recognise all the transformations in the information. In the essence the analyses made by us that the material is exposed to is not direct transformation of the information but the reverse one. It is not the conversion, but the reconversion, not the divergence from the past reality but an attempt to return to return to it. Clarke indicated that his interpretative theory connects [levels] 41 (Clarke 1973, 17), that is, it traces information in the reverse direction into the archaeological material.

3. THE DEEPEST LEVELS OF AF Optimally, the researcher would strive for accurate reconversion and the separation of subjective and objective elements in AF as completely as is possible. It

108

Acta Archaeologica
duction. In the early 1930s A. V. Kiparisov (Kiparisov 1933, 79) and K. R. Megrelidze (Megrelidze 1935, 7181) suggested holding the principle of object-practical activity as a leading idea in Marxist archaeology. In the 1970s this was caught up by V. D. Viktorova (Viktorova 1975b; 1989, 16f, 6578). V. F. Gening considers AF as the result of objectication and puts objectication theory on the foreground in archaeology. Objectication of what? Of particular ideas? Not at all, objectication of some very abstract forces of human essence, of social live activity in general (Gening 1989, 64f). From the object-technological reconstructions (OTR) he leads the investigator (Gening 1989, 224270) to socio-technological theoretical model (STTM) and so on, and from there directly to the end target, to socio-economic formations, which as it is known are exactly ve in number (the phraseology of Bulgakovs Voland ts well here). In my opinion, to be a Materialist does not mean refusing the role of ideas in the conditioning of behaviour, and does not mean sticking to the tangible, visible behaviour and to boggle over revealing ideal motives of behaviour. It is just the reverse, to be a rational and consecutive materialist means, after passing through all the levels of cognition, to see behind the ideas and recognise those stimuli by which these ideas are produced in the conscience of people. Not to shrink back one step but to advance a step further. Ideas are conditioned by the being, by the entire social and personal practice of people by productive, economical, political, ideological, everyday life activity etc. Within this practice, the stimuli of creation of social norms, standards, and mental templates emerge and are rooted to ideas. The problem of conscience, the Soviet philosopher E. S. Markarjan noted (Markarjan 1969, 43, also 3742, 44f), is rst of all the problem of the stimuli of its formation. It is these stimuli that were contained in the material-production practice. Behind the matrix of ideas one has to see the matrix of stimuli. These stimuli have to be considered as the last, the deepest level of AF. They provide the passage to the causal machinery and sociocultural or historical regularities, i.e. output off the limits of the AF, into theory. The approach to these last levels of the AF and to their separation was difcult. These difculties resided in the double relationships of practical activity with

is necessary for this sake that the scheme of the cognitive multistage structure of the AF appreciates the levels of the AF (i.e. stages of the conversion and its ltering) as well as it can. Most schemes manage this only for a few levels, but as there were many such proposed schemes, even a simple generalisation of them is able to aid a more complete examination of AF. The authors of these schemes spent their entire investigation efforts mainly on attempting to mentally picture the material culture by issuing from archaeological remains. Rouse and Taylor aimed to establish what is hidden behind things and their relationships. Behind things and relations these archaeologists saw customs, stereotypes, and idioms of behaviour, and still deeper, behind those, they saw social norms and individual motives, that is, ideas (Rouse 1939; Taylor 1948, 97124). Theoreticians in the West concerned with the concept of AF and the problem of archaeological research design did not venture further than this. It is understandable by itself: sharp criticism can be made of the objective idealism in the basis of the methodology because it is reducible to the notion of ideas as prime causes. Not only is idealism presented in the methodological fundamentals of modern archaeology in the West but materialism is too. Soviet Marxists used to consider this materialism at, non-consecutive, and supercial, since it is full of neopositivist ideas and thus is akin to subjective idealism. As applied to the cognition of AF this is clearly expressed by the line of the New Archaeology to limit the culture concept within behaviour and its material results, and to eliminate ideas from this concept since they are not observable directly (Binford 1965; Watson, LeBlanc and Redman 1971, 6365). Yet the concept of culture (in such limits) appears fundamental for the conception of cultural process, and the cognition of cultural process and its laws are considered as the main aim of archaeology (Binford 1968; Watson et al. 1971, 22f). In other words, in reconversion the New Archaeology theoreticians envisage advancing only to the level of behaviour actions, not deeper. In their notion AF appears atter than in the notion of their idealistically disposed predecessors. The orthodox Marxism affected archaeology in a similar way to refuse seeing ideas in the basis. It concentrated its attention on tools, objects, and pro-

Metaarchaeology
ideas. Must we place this activity on a level higher or a level lower than the ideas? It seemed that there were grounds for both cases. In theoretical works of Soviet authors concerning this problem, a more general approach was taken: the activity was related not to ideas but directly to the material results indeed, these are undoubtedly its derivatives (Zacharuk 1970, 13f; 1973, 44f). The root of the difculty was lurking in the double essence of the very concept of practical activity. One thing is the particular operations, dynamic actions of people, and components of behaviour. They can be considered as realisation of the ideas. Quite another thing is the total aggregate practice of the society, situations which the activity brings people into, and the characters that result.

109

(7) (8)

(9)

(10)

4. THE COMPLETE COURSE OF THE CONVERSION Issuing from these considerations we can present the full scheme of the multi-step cognitive structure of AF as follows (the levels are enumerated according to the conversion of information): (1) Stimuli social and personal practice of people (production, economical and political relationships, ideology or mentality, everyday life etc.). (2) Ideas social norms, customs, standards, individual motives (already rooted here are plans of behaviour and the potential population of artefacts, i.e. the cultural matrix). (3) Actions operations, deeds, events, i.e. that which realises the customs, stereotypes and idioms of behaviour. (4) Incorporations (embodiments) things, artefacts, marks and traces on them, material objects as well as their interrelationships in the living culture. Their total population is present. This is the objectication of behaviour, or, indirectly, objectied ideas. (5) Deposits things, fragments of them, traces on them and left from them, as well as interrelationships of all of them in the dead culture. Not the total population of things but only its deposited fraction. (6) Remains the same in the archaeological material in situ in or on the earth by the time of

(11)

(12)

(13)

(14)

excavation (or of sampling, surveying or the like). This is the preserved fraction. Destructions displacement of a part of the material by non-professional collectors. Observable that which is in the archaeological material which was in the space under research (in the dig, the section, the prole, the collection etc.). This is what an archaeologist is able to look at directly. Discoveries things, traces and their interrelationships discovered, noted by researchers in the eld or in collections. This is the fraction that is seen by an archaeologist. Samples things, traces and their interrelationships recorded and liable to registration (visual, written, and other descriptions) or taken from the eld. This is the fraction selected in the eld or in collections. Filtrates the same objects less the discarded ones, i.e. objects (and their relations) accounted erroneously: pseudo-artefacts, non-genuine traces, fakes, counterfeits, forgeries. This is the puried fraction. Descriptions reection of the information from the previous level with the means of a special elementary alphabet and by comparison with concepts of analytical classication. Compacts (generalisations) the result of information having been compressed by separating out its essentials by means of taxonomic classication, type and systematisation. Evidences information from one of the previous two levels (or from both of them) reported in a publication or in a lodged work (eld report, dissertation etc.), in les or in computer memory as well as in personal archives and at the worst, in the researchers own memory.

Of these 14 levels of AF, the rst three deal with potentials, the next eight (from 4 to 11) with material, and the last three with reections.

5. FILTERING IN THE FLOW OF INFORMATION Between these levels there are lters or oodgates in which information is transformed: some passes

110

Acta Archaeologica
Realisation of stimuli (12) This process is also determined and therefore customs of the peoples having no contacts often appear similar (the phenomenon of convergence) but the segment of probability determination here is much bigger. Even Marxists admit, at least in theory, the relative independence of ideology (here meaning the world of ideas). In the nal appearance of ideas, the individual peculiarities of people are represented. Thus customs of peoples are very diverse. Incorporation of ideas in behaviour (23) Here the mark of various obstacles are met: natural disasters, conict of ideas, conicts between people and between groups of people, and between personality and society. Social norms are less rigid than the biological program of behaviour. Personalities and human populations dispose freedom of will, freedom of choice, and this freedom is not limitless but not ctional either. So the realisation of ideas in behaviour and with it the behaviour itself are predictable, but not strictly predictable. Objectication of behavioural acts (34) The New Archaeology as well as orthodox Marxists established here clear relations. Schiffer retained this notion when he placed correlates before his distorting C- and N-transforms (Schiffer 1976: 1214). This is correlation between objects (artefacts) and ideas. In fact there are no unambiguous correlations, there are only ideals weakly realisable in behaviour. It is clear that behaviour intervenes between ideas and objects (artefacts), behaviour that leads from ideas to objects, and it usually does not realise the ideas exactly. Besides, a gap always exists between the behaviour and its objectied results (artefacts) (Krause and Thorne 1971). A simple difference of materials will condition different results even in equal series of actions directed on these materials. Things of different appearance will emerge. However, it is not only the materials that are different, but the conditions of actions too. H. Obermeier, with a group of French archaeologists, arranged experiments of throwing int into stone crushers in order to show that int is split differently in warm conditions and in frost and that it can lead to different sets of int industry (Obermajer 1913: 468). Therefore ideals are ideal, manufacture

Fig. 14. C- and N-transforms in archaeological information ow (after Schiffer 1976: 2.1).

through but some is lost, distorted, or a kind of polluting enters (Klejn 1975b; Sullivan 1978, 193). What factors affect the information at these oodgates? (Daniels 1972, 204209; Collins 1975) M. Schiffer divides them into cultural (C-transforms) and natural (N-transforms) (Fig. 14 Schiffer 1976, 14f). In order to investigate the sequence the direction outlined by Daniels and Clarke is more convenient along the steps of information ow. Let us consider them in the same order of information conversion. For the sake of convenience in comparing, ordinal designations of oodgates are given. Crystallisation of stimuli (01) Led to by sociocultural-historical regularities and causal machinery. The situation is complicated by the fact that many of these regularities act only as trends and provide merely probabilistic (not rigid dynamic) determination, while some of the causal mechanisms do not produce unambiguous causal connections: under the impact of different conditions, frequently random ones, similar causes can lead to different effects, and behind similar effects different causes can be hidden (the phenomenon of plural causality).

Metaarchaeology
does not coincide with template, or realisation with model, or imitation with original. The rst act of archaeologisation (45) This is predepositional: death, becoming defunct, departing, deposition on, and later, in the earth. In this way the selection proceeds, one may say, depositional selection) the bearers of culture themselves accomplish it having in mind peculiarities of situations and things (Eggers 1959, 264268, see also 276294). People themselves selected what to discard, and what to keep. They determined in which situations the worn-out, obsolete, damaged, or broken things go, in a natural way, to deposition, and in which situations this way is open to fresh, intact, sometimes specially made things, necessarily in an articial way. In comparison with living culture, here proportions are changed as well as appearance. The make-up of the living is not repeated in the summary distribution of the bone fragments of animal species in the occupation layer. Age composition of the once living community is not reected in a cemetery as in a mirror: during the life of one mature generation several generations of children could have lived. Some components of the living material culture are simply lost (usually organic parts) while components absent in the living culture have entered into the dead culture, for instance, specially manufactured ornaments with gold foil and ceramic models of things. Many sub-systems in the material culture are open they do end up as compact deposits and are frequently not covered by other distinct layers. This is why in several cases elements that were separated by enormous time intervals in the living culture coincide and are mixed together (compression, Ju. N. Zakharuk 1975). Second act of the archaeologisation (56) This is postdepositional. The rst group of factors is contained in fossilisation. The changes that take place make the orginal use of an object no longer possible. The other aspect of fossilisation is the belonging to extinct species. The nal appearance of the objects is a consequence of natural forces and of time decomposition of organic materials, corrosion of metals, and so on. The position of remains is changed by

111

redepositing under the action of natural forces. All this increases the compression effect. The second group of factors is ancient articial destruction. During the post-depositional period the subsequent generation of people also had an impact on deposits through activities such as piling up the habitation layer, robbing the graves, secondary digging and destruction of monuments. They disturb the structure of assemblages and change the composition of material. This also increases the compression effect. Both groups of factors acted in parallel ways, without strict alternation in time. Modern articial destruction (67) The activity of non-professionals, obtaining things without recording nd conditions or falsifying them, which displaces a part of the material from assemblages. It is at this stage that the majority of fakes arise in the material. The choice of unit for investigation (78) Archaeologists are forced to limit the area of sampling and mapping. It is clear that this will inuence the amount and content of the information received. This will depend on many circumstances the aims of research, nancing, equipment available, accessibility etc. The detection of information bearers (89) The detection of artefacts, traces, and relations by researchers in the space outlined. It is insufcient to look at the objects one has to see them. What the investigator will be lucky enough to see depends on a number of factors: the competence of archaeologists, the methods applied, the technical outt of the project, the level of the development of science at the time, conditions of work etc. Selection of the necessary information (910) Everything worth recording and preserving is necessary to select. There is usually too much observed information. To record it in totality is practically impossible, not to speak of the total amount of ceramics, all the bones, the content of the graves and the entire content of the occupation layers. Transfer to a museum of all the relationships of artefacts with the environment is in general impossible in principle, so

112

Acta Archaeologica
Generalisation or minimisation (1213) This is the comprehension of information for the sake of convenient record keeping and use. In the basis of this process is separation of the essential by means of taxonomic classication, type and systematisation. Several kinds of connections are revealed: type, assemblage, culture etc. Taxonomic classication implies hierarchy of attributes, estimation of comparative importance of various object parameters. Such estimation has objective footholds in the material. However, culture is a complex and multi-sided system, and the interests of observers are broad and do not coincide on many points. So the choice of the essential and the estimation of parameter importance are not rigid. Some divergence of solutions is determined by difference of approaches by systems of views, and by interests. Narrative (1314) This is guration and objectication of the elaborated information by special means of storage and distribution in publications, papers and les, in computer memory as well as in natural memory, notes of researches themselves, in museum exhibitions etc. Publications, electronic and printed, occupy the main place among these reservoirs and distributors of information. In preparing for publication the last distortions of information are possible editing, changes to drawings and photos, editorial additions and changes in accentuation, and uncorrected mistakes. Having traced the course of archaeological information transformations from beginning to end, we can hardly wonder there are errors and divergences in archaeology. How strong are the regularities and sense connections between events and their traces, if indeed after this entire cascade of transformations they do still show through the mosaic of descriptions and interpretations? And how sophisticated must the researcher be to catch them, to reveal and to reproduce them in a state similar to their initial one?

selection is unavoidable. Technical constraints of the project will affect the selection. Notions of the researchers on the comparative importance of different parts of the information discovered forms another group of factors conditioning the selection. These notions rest on aims set by the given project and on general evidence on an estimated set of data for the project. Again, the competence of the researchers and the general level of science development will be relevant. Subjective intentions and views of the researchers also play an important role. Selection of genuine information (1011) One must recognise young, often even modern artefacts, traces and relations mistaken for ancient ones, and natural objects mistaken for human made objects. In this oodgate distortions are possible too: the loss of some information and relations or the loss of unwanted, corrupting information. Recording of the discovered and selected (1112) Recording of artefacts, traces and relations is a rule of every professional expedition. The monument is lost to a certain extent and to excavate it a second time would be impossible. Recording should rescue the maximum possible quantity of information and grant adequate access of the collected information to any user distant in space and time. Written description is the reection of real things by means of not only natural but also a special conventional language with an elementary alphabet, vocabulary and grammar (scientic terminology, abbreviations, formulas, tables, indexes, codes etc.). Such desription implies comparison with the concepts of analytical classication. All these means must provide maximum and exact transmission of information. The character of the transmission depends on the set of concepts and terms in which the material is described, the richness of language, its exibility and adequacy for the material, and its terminological clarity. Graphic xation also implies a certain conventionality, the transmission of three-dimensional reality with plane projections, reduction of the light and shade relations with linear contours, and often detachment from colour. This is also a language and its rules imply quantity and choice of projections, the degree of generalisation, sign conventions, and scale.

6. RECONVERSION, REFLECTION AND CRITIQUE To consider in such details the conversion of information, the process offering multistageness and depth

Metaarchaeology
to AF, was necessary in order to present clearly the full course of reconversion. Since it is reconversion that lays in the basis of archaeological research design, the planned succession of steps in archaeological inquiry is intended. The point is not to repeat the list of stages of conversion in reverse order. Reconversion does not simply mean the mirroring of the conversion. One has to trace its full course anew. It is necessary to consider all the stages of reconversion completely and consecutively, because at every stage the task of reconversion is to restore and reconstruct the state of information on the preceding level, while each time taking into account what occurred in the ltering oodgate what is lost, added and distorted. Conversion was the movement of information in the formation of an archaeological record, the forming of the archaeological facts, where objective processes of culture acted. Reconversion is the revealing of information, the tracing of the ow of information, the advancement of archaeological thought into the depth of AF. The research of this process is reection: self-checking, self-proving, self-control and self-critique by the researcher as an abstract person, as a researcher in general. With respect to particular people it can not be a self-criticism at all, but simply a critique. Indeed, different features of this abstract person can be realised in various scholars. This is also the analysis of the portions of information being received about the traced sections of reconversion. This procedure is called source criticism and is usually limited to a particular step or a few steps of the entire ow. Here it is understood in the broadest sense. Practically the whole inquiry can be understood as source criticism. No wonder then, that archaeology is considered as a source-studying discipline. By moving the reverse way through the preceding survey, towards the information, let us designate the stages of reconversion or criticism with a different numeration, this time with Roman numerals and correlated with the previous list in brackets. I. Textological criticism (1413) This is proof of the adequacy of narrative exposition. It relies on the inner logic of the narrative. Other aids are the comparison of various realisations and versions of exposition, and verication of the exposition with the preserved parts of information from

113

the preceding levels. The outer characteristics of the narrative (accuracy, detail etc.) are taken into account, too, as well as general ndings of the authors personality and on the process of the exposition. II. Critique of concepts (1312) This is the reliability proof of generalisation and it can be executed in two ways. The rst consists of repeating (even if only partially by sampling) the work done and including an analysis of the methodological foundation of classication, and analysing the attempts of its application to the same material. This route is convenient since it does not demand any output beyond the borders of the material studied, but it is not attractive because it entails twice the amount of work. The second way is to ascertain from outside whether the given system of generalised concepts is realistic. This means comparing it with other analogous systems, to apply it to a broader circle of materials and to clear up whether it works whether the new materials t it and whether its concepts correlate with other concepts. This way is more interesting, but it leaves the researcher unsure whether all the particular errors were revealed. III. Critique of description (1211) Critique means foremost proof of the completeness and accuracy of description. This proof includes estimation of possibilities and limits of every method of description applied and testing of the accuracy of their application. This allows us, if not to restore unrecorded details, then at least to establish where and what kinds of omissions one can assume. IV. Critique of authenticity (1110) The proof of the original information selection, including estimation of the applied methods of selection. The rest is reduced to repeating the selection because the checking of methods application coincides with the selection due to the critical character of the very operation of selection. V. Setting critique (109) This is the proof of the selection of information needed. This proof concerns criteria and conditions of this selection, demands to characterise the personality of the researcher and the level of scientic devel-

114

Acta Archaeologica
is necessary: in order to reconstruct the rst destruction. In many cases, remains that survived after destruction are unambiguous or have little meaning they allow only one or a few certain varieties of reconstruction. Secondly, in several cases, the causes of destruction are known, so one can establish what deposits would have been distorted (and thus possibly had existed) and what would have survived. Thirdly, analogous objects that are better preserved can serve as models for such a reconstruction. Thus critical work on reconstruction at this stage is reduced to three series of operations: 1) counting of degrees of freedom in reconstruction from remains of various kinds, b) estimation of destructive factors that once acted and estimation of their impact, c) considering objects under study against a background of better preserved objects of the same kind. X. Critique of deposits (54) An estimation of the changes that occurred when objects passed from the living culture to the dead one. This estimation has to embrace depositional selection of the bearers of ancient culture, necro transformations (changes of proportions etc. in the dead culture in comparison with the living), and compression effects. It is here, in these operations, that the centre of gravity lies in the estimation of cognitive opportunities of various kinds of archaeological sources. XI. Critique of things (32) This is an estimation of the difculties met in the objectication of behavioural acts. Problems of objectication embrace connection of work and artefact, process and result, actions and traces. What difculties can be expected here? Establish regular correspondences, trace regularities, calculate correlations and its all over but the shouting. The network of universal correlates will appear and the only task remaining is to formulate the rules of correspondence and apply them. This is not quite the case however (Klejn 1981a). Culture is polysemic. There are no stable correspondences in it, there are only more or less noticeable preferences, usually local, temporal and each specially conditioned. Each case has to be considered individually. It is necessary to uncover in which cases equal

opment at the time of research. It is not always possible to make the selection again. What was discarded in the eld is mostly lost beyond hope. One can only establish which cases do not exclude objects of a certain kind in the space studied. VI. Critique of observation (98) This is proof of the completeness of objects discovered in the area under study. This proof must assess the methods of survey, equipment of the project, competence of researchers, take into consideration the level of scientic development and the conditions of work, and the intensity of study if carried out over a large area. As a control, it is of course possible to make a second study but only if the excavation of the object or of its given part is still proceeding. Afterwards it only remains to infer undiscovered details from the above estimations and by comparison with analogous assemblages. VII. Critique of limits (87) The analysis of the choice and limitation of the space in which the research is done. For this it is sufcient to estimate the size of the area and its content of archaeological materials in comparison with other areas. Factors which determined this are considered by historiography. For critical proof they are interesting only in the case data on the choice and on the limitation of the area. VIII. Museum critique (76) This is an estimation of the scope and consequences of the modern destruction and separation of the information. It is essential to assume what information may have been contained in the lost parts. The main foothold in this selection is museum and eld documentation. The comparison is suggested by itself. If the eld documentation is absent, the only hope is in analogies and various evidence from elsewhere. IX. Critique of remains (65) This is an estimation of the scope and consequences of ancient destruction natural (fossilisation) and articial. However, strictly speaking, one can only estimate the consequences of destructive processes when it is known what existed before the destruction. But this is not known, and it is the reason the estimation

Metaarchaeology
actions led to different results. This is important in order not to overestimate the differences and not to overestimate their meaning. It is also necessary to unravel situations, fortunately not so frequently, where different actions have led to equal results (Adams 1968; 1973). Critical estimation of complexity orients the investigator in search of decisive details, for instance, stratigraphic traces of secondary burial. It orients the researcher too, in considering the object under study in a broader system for example, a comparison with analogous tombs where the second half of the tomb space prepared for the second corpse remains unlled (Itina 1954). Modern behavioural archaeology does not want to go further than this level of cognition. It is afraid of shaky ground on the route ahead, but are further areas really impracticable? XII. Critique of behaviour (32) This is an estimation of the divergence between ideas and their realisation in behaviour. In essence this is a psychological task to conjecture and grasp ideas from behaviour. This psychological task is very difcult even with living people, whose feelings, thoughts and intentions we understand better than the psychology of people more distant by culture and time. Nevertheless, in these cases, when it depends on familiarity, we can rely on plenty of minute details, imagine ourselves in the same situation and make stipulations based on discerning Other from Us. All of these fail almost completely when we advance ideas from archaeological remains. In order to restore an individual idea in such a situation one has to nd a more adequate realisation of the same idea the imitation of an original artefact for instance. The methods of determining the direction in the typological series (after typological rudiments) and the direction of the inuence (after separate modes) are the keys to such a riddle. Shared ideas, or social norms, can be restored more reliably: if there is much deviation from the ideal but they show a normal distribution, then revealing the ideal is not difcult. A question emerges however of whether the deviations from ideal always conform to the law of normal distribution. Can systemic shifts not emerge under the impact of equal vector forces? This should establish the critical estimation of the whole situation.

115

XIII. Critique of ideas (21) The task of this stage is to judge the extent to which life phenomena, those that stimulated certain behaviour, were reected in particular ideas of ancient people. The critique of ideas lies at the core of the reconstruction of life of the past, reconstruction after ideas deposited in things, after archaeological facts in general. How did people react to phenomena in their lives, and how did they perceive them? Is it possible to restore their ideas through their reactions and then to make judgements based on these phenomena? This estimation has to appreciate not only realisations of individual chance divergences, not only local differences connected with race, ethnicity, gender, age, class, estate, caste, etc., but also the specicity of the prehistoric and ancient consciences. Since things like mythology played a large role in the thinking of people, it does not coincide with the thinking of modern man. Even the actions of medieval people will remain completely incomprehensible to us if we imagine them as sheer rational agents, and we reason strictly logically about them, soundly calculating their interests and the historic situation. We often are inclined to judge situations in the prehistoric and ancient world after ideas and actions of ancient people and forget these traps, seeking naive rationalistic explanations for ancient ideas and actions with the help of modern common sense and patchy data on the past. These explanations participate actively in our system of evidences on past life, and so condition the formation of a false historical picture. A critique of ideas attempts to prevent this deviation. How prehistoric people thought and felt, and how they cognised about their life and themselves, we learn from ancient pictorial and symbolic monuments, from ethnographic and psychological studies, and from theoretical reconstructions. It is signicant that in recent decades researchers of prehistoric economics venture more and more into these themes. XIV. Critique of stimuli (10) We have to estimate to what degree particular stimuli (historical phenomena, events, and circumstances) were necessary and unavoidable in the conditioning of the behaviour of ancient people. This is connected with revealing the limits and possibilities of the probabilistic determination in each case where such deter-

116

Acta Archaeologica
The complexity of AF is for me not a reason for pessimism but rather the stimulation to work to enlarge the archaeological research design in scope, and to the elaboration of its new stages and operations. It is not to philosophical speculations that this study leads us, but to methods of work. Modern researchers arrived at inferences of AF with four-levels, which would correspond to a fourstage procedure, or four-stage archaeological research design. Only Daniels from Nigeria suggested a sevenstage procedure (without any connection with AF). David Clarke wrote in 1973 on four levels of fact and 5 stages of the research design when he divided the general theory of archaeology into consecutive steps in his well-known article Archaeology: the loss of innocence. In 1975 I suggested a 14-stage research design (Klejn 1975b); substantiation of this procedure was my understanding of the multi-stage nature of AF. The parcelling of procedure and scrupulous itemisation of its stages acquires special attention as connected with the prospective computer elaboration of archaeological facts on the basis of articial intelligence. Computers demand maximum parcelling of operations. Today it is insufcient to know that AF is complex, one has to know how many levels there are. The innocence of archaeology was lost long ago. Now the issue is not in stating contriteness, doubt or satisfaction, but for this event to lead to a strong and healthy paradigm.

mination took place. This is also connected with cases of pluricausality. From this denition it is evident that if the critique of ideas means reconstruction of stimuli, the critique of stimuli leads to reconstruction of historical process, of causal relations, and needs analysis of very broad historical prospects. This analysis is hardly feasible on the basis of archaeological records alone. Consequently, critique of stimuli goes beyond the limits of archaeology proper. Archaeological research in the proper sense is limited to the preceding step.

7. CONCLUSION The analysis undertaken of AF is distinguishable from preceding attempts by one circumstance. Other analyses have concentrated on the philosophical aspect of the theme. They revealed the interrelationship between objectivity and subjectivity in AF, and their aim was to show how signicant the subjective factor was, and to show this against over-condent optimism. They focus very much on terminological hairsplitting subtleties in order that philosophical arguments are stated clearly enough. European archaeologists, mostly British ones, gave more attention to the practical operations of cognition. It seems to me that in continuing the European line I was lucky in revealing the connection between structure of AF and archaeological research design.

Metaarchaeology

117

11. Archaeological research design

1. BIAS AND IGNORANCE S. G. H. Daniels, an archaeologist from Ahmadu Bello University, Nigeria, recognised a stimulating paradox connected with the processing of archaeological information: From the earliest stages of our aquaintance with the academic world, he stated, we are taught that knowledge is desirable, and that ignorance is a condition to be remedied as soon as possible and even at great cost. [Archaeological research] design on the other hand states unequivocally that in matters of classication ignorance is a paramount virtue and one which, as innocence used to be, is irreplaceable if lost (Daniels 1978). He understood classication here in a very broad sense, embracing both selection and measurement. It is not common opinion that the loss of innocence in archaeology is worth the sorrow. David Clarke did not think so when he wrote his article Archaeology: the loss of innocence, and Christopher Hawkes supposed that it is even possible to return to this state he wrote a responce to Clarke with Innocence retrieval in archaeology, as he did not want to know of the new, analytical, archaeology. Due to his mathematical orientation, Daniels, who works at a statistical centre, was one of the rst scholars to escape the sometimes idle philosophical talks on the depth of archaeological fact, and began speaking instead on the cascade ow of information from the past events to the analysis of archaeological records. So he linked the two previously unconnected problems: the epistemological structure of archaeological fact and the normative research design the normative succession of the stages of archaeological inquiry. The course of information is his main interest, and to understand its movement in the most complete way possible and its results as objectively as possible are his main concerns. By the 1970s, two decades had already passed since the beginning of the talks on the subjectivity of inferences in archaeology (since Smith 1955; Thompson 1956), although the critical theory, just emerging in western Marxism, did not enter archaeology in full until the 1980s. Daniels stepped into this line of interest when he made the challenge of subjectivity the main concern of his studies. He was dealing with the ow of information from the past and, correspondingly, with the reverse proceeding of cognition from the present. He had in mind a modern researchers cognition of the past the scheme, the plan, and the design of the inquiry process. The struggle with biases troubles him most: At the heart of the research design problem, he wrote (Daniels 1978, 29), lies bias, and the way in which it enters, or can be prevented from entering, into the data during the research activities of selection, measurement and classication. Since preconceived ideas are usually oriented against some views and facts, the implementation of bias by the researcher presupposes preliminary knowledge of the material under study if the material contains odious things against which bias can act, suppressing or refusing them. So the means to prevent bias from doing this is to grant the researcher the benet of not knowing too early the material on which research operations are to be directed. Such measures are usually applied in the activity of experts the material must be unknown to him before the results of examination are declared. The strange insistence of the researcher on this ignorance is explained by the belief in this measure. However, despite his belief bias is not the only signicant problem in achieving objectivity, perhaps not even the main one, and ignorance is not the only means to reach a validity of results, and perhaps not the best means either. In addition to the subjectivity of a researcher, there are objective factors in play such as the fragmentary state of the artefacts, large gaps in the materials, the polisemic nature of cultural things, and often an absence of relations between parts found and parts lost. As for resorting to ignorance it is not a logic tool, but part of the psychology of research strategy. Overcoming ones own bias does not necessarily demand such measures, much can be done with the help of randomisation as well as strict rules and limitations elaborated for methods. This was more broadly considered in the earlier work of Daniels. In his 1972 work he placed a scheme of a 7-step information ow from the past events to the present remains (see Fig. 13). In the oodgates between the information states in his scheme, factors

118

Acta Archaeologica
procedures and randomisation: the last one suppresses the noise rather than bias.). Post-depositional factors are shown on the right, and it is said hereof: The most that can be done is to estimate their effects and to try somehow to allot them a place (Daniels 1972, 202). It is difcult to say why Daniels later narrowed his view on the means of obtaining objectivity. May be by this time, approaching the 1980s, the advent of critical theory with its preoccupation with biases was already felt. Both articles by Daniels contain the term research design in their titles, a term that implies the normative scheme of archaeological inquiry.

Fig. 15. Model of archaeological procedure (after Clarke 1968: 2).

which make an impact on information are included. They are inserted from two sides: from the left of the scheme factors which can be controlled by the investigator are shown, for instance, the selection of the excavation location, and from the right, uncontrollable factors like redisposition. According to the position on the way the information ows, these factors are divided into three groups: historical, postdepositional (from deposition in the earth until the excavation) and methodical. In the article, main kinds of errors on these levels are considered in detail: noise, non-systematic mistakes, and prejudgement. Correspondingly measures aimed to dispose them are indicated: formalisation of the procedures, introduction of redundancy, and randomisation. (In Daniels scheme 5.3 one should change the places of normalisation of

2. NORMATIVITY OF THE RESEARCH DESIGN Archaeology is a scholarly discipline, in continental Europe it is even called science, since linguistically humanities are there not distinguished from sciences. Since it is a discipline or a science (in the general meaning of this word), the main process of its functioning should be a disciplined activity subjected to certain substantiated rules. It is divided into stages or levels, and their organisation, that is their set and succession or the procedure of archaeological research, is not arbitrary. Clarke calls this normative scheme a model for archaeological procedure and presents it as one of the three parts of the general theory of archaeology (Fig. 15). Since Clarke, the term procedure has come to stay in Russian archaeological literature as the designation of this succession of steps. Rouse calls such a scheme strategy of archaeological and prehistoric researches, while he tolerates exceptions from it as tactics (Fig. 16). B. C. Swartz exposes this scheme as the logical succession of archaeological aims, L. Binford, J. Fritz and F. Plog as archaeological research design, and V. S. Bochkarev as structure of archaeological research (Clarke 1968, 34f, g. 2; Rouse 1972, VII-X, 27f, 62ff, g. 8; Swartz 1967, 487497; Binford 1964; Fritz and Plog 1970, 409411; Bochkarev 1973, 59). The term archaeological research design is now usually applied in Anglophone archaeology. It is possible to formalise some parts of this scheme and transform them into an unambiguous succession of commands, leading to a rigidly determined chain

Metaarchaeology

119

Fig. 16. Strategy and tactics in the logical procedure of archaeology (after Rouse 1972: 6).

of operations. Such succession becomes an algorithm (Sher 1970, 923, g. 1). It is often impossible to reach such a grade of strictness within so many stages, and every one of these stages can be freely dismembered into elementary actions. Yet interrelationships between the stages perceived as whole blocks are rather denite. Bochkarev (1973, 59) insists that the procedure has a hard structure that does not permit omissions of levels or their displacement. As we have seen in the previous chapter, the archaeological fact is practically the long course of information from the past events and phenomena. Along this course the information squeezes through several oodgates where parts of information are distorted and lost. This conversion of information is xed as a series of states, which are levels of the archaeological fact (AF). With the cognition of those remote events and phenomena as the target, an archaeologist by the very nature of things executes reconversion of information. The temptation to make this reconversion a one-act performance is great to pass in one stroke from AF to historical-sociological meaning, presuming it lies

somewhere under the very surface of AF. This would be so simple, so convenient! Even if we know that there are many levels to the fact, with many ltering oodgates, why not consider the changes of information all together or grouped morphologically (e.g. after Schiffers classication: C-transforms, N-transforms), and not according to their natural succession? There are factors that push archaeologists to attempts of such a simplied, or reduced, reconversion. These are now daft attempts, dogmatic adherence to a particular rigid scheme of unambiguous interpretations with a ready set of labels. V. S. Bochkarev (Bochkarev 1973, 59) is right: one should not skip over stages of the research design. In each rendering the ow of information runs through a cascade of qualitative and quantitative transformations (objectication, mortication, idealisation etc.). It seems as if it passes through lters and at each one incurs a loss and suffers distortions. It is desirable to control, correct and compensate for these losses. So, the conversion of information consists of phases separated by oodgates that have a ltering effect.

120

Acta Archaeologica
they will be ascribed to earlier causes. The initial state of the information will be reconstructed wrongly: in the original picture factors will be supposed which in fact were not existent. This is why all the oodgates should be revealed, the full course of the conversion traced, and the strict succession of the reconversion back through the oodgates respected. Discredited long ago were naive notions that did not suspect a multi-step nature to the archaeological fact, and did not conceive how dangerous it is to jump from archaeological facts directly to events and social phenomena of the historical past. Archaeological cognition is digging, not jumping.

In each oodgate every transformation changes the appearance of information and serves as a springboard for the next transformation. Only adjacent phases are connected with each other directly as initial and derived. And quite often some remains, after which one could judge the initial appearance, are kept only in the derived phase. So we make judgements on the initial appearance by issuing from the previous one. If while doing this we skip over some phases, perhaps some essential changes in information will be lost, and distortions of the past reality will be unobserved. As a result we come to false reconstruction. Hence the importance to account for all the phases of the conversion, all the oodgates. This must become the basis for working out the full scheme of the archaeological research design. There is an additional complication to this problem. The action of these oodgates is two-way: The same lter that stipulates materialisation in one direction, stipulates idealisation in the contrary direction. The mischief of it is however that the losses in these transformations are asymmetrical and often irretrievable. If while objectifying an idea (say, the plan of the dwelling), a person could not realise some part of the idea, and in a hurry did not install an oven required by the usual plan, then an archaeologist, while idealising his observations on this (and only this) excavated dwelling, could not reconstruct this part of the plan and will come to new and completely different losses. The archaeologist will lose data on an entrance as soon as it was cut by a later pit-dwelling and will receive false data on the content of storage pits if together with it some materials have been found of the pit-dwelling cut in. So the losses do not let the saved information compensate for them, but they are pooled and not only add to each other but become more complicated too. This means that the rules of interrelation in archaeological theories cannot do with simple correspondences but should envisage a procedure, much more complex than in other disciplines. One should envisage an organisation and permanent attraction of external information not contained in the regulated ow. Why it is necessary to provide such conversion subsequently, without omissions? If we omit a oodgate those changes that occurred to the information at this oodgate may not be identied, and consequently

3. COMPLICATION In my scheme of information ow 14 stages were presented. This list of conversion phases (and correspondingly reconversion) is divided in two parts. As concerns the conversion, the rst 7 acts present building (14) and death (57) of the material culture. The result of this part of conversion is an archaeological record, or in Russian terminology, an archaeological source. The next 7 steps present its initial, presumably empirical cognition (survey, excavation, recording etc.). The research procedure executing this cognition mainly follows the reconversion but with some stipulations. This connection is realised in two varieties. If the investigator builds a study on original sources, his or her procedure coincides with conversion embracing the stages 814 and includes the outer criticism of archaeological records (stage 11). The result of this investigation is the transfer of an archaeological record from material form into generalised concept form, applicable for operating in scholarly knowledge (or in science in the broad sense of the word). The continuation of the study is usually an interpretation study, and its procedure follows reconversion through the stages 71. The core of this study is the inner criticism of the archaeological record (estimation of information change at each oodgate), and the result is the creation of a new record, or new source, a historical one. When an investigator works not with original sources but with literary data, the entire procedure follows the full course of reconversion through all the stages 141, and the outer criticism is widened cover-

Metaarchaeology
ing all the rst 7 stages (148), since it is necessary to check transformations in each of these oodgates too. Only when reconversion (141) having terminated does the historical, sociological or culture investigation begin. In fact the realisation of this procedure is even more complicated because this straight logical scheme of information processing is only a basis on which different approaches to the material are built. These different approaches are connected with different understandings of the interrelationships between fact and inference, with different placing of generalisation, hypothesis and problem. In the research practice, the normative scheme of the archaeological research design (hereafter ARD) is realised only as a trend, and deviations from the strategy frequently go far beyond the borders of any rational tactics. In implementing the ARD, confusion dominates. Two circumstances support this situation in the discipline. Firstly, the schemes of ARD usually appear more directive than normative: they are advanced by theoreticians without serious substantiation or merely with general philosophical substantiation, and not interrelated with the specicity of the archaeological material. Secondly there is disaccord in the notions of theoreticians themselves with respect to this issue: instead of receiving a single normative scheme, practical archaeologists are supplied with an amount of schemes differing both in quantity and in the set of the stages, as well as in their succession. Nevertheless these schemes can be grouped according to the New Archaeology in two main varieties (their own and the opposite), or possibly in three varieties in all. Every one of these varieties changed with time, new blocks of operations were introduced in them, and the connections between the blocks became more complicated. Blocks were grouped differently now closed inside one discipline (archaeology, or prehistory, or palaeohistory), and now the entire succession was more or less radically divided in two parts archaeological and historical (or sociological, or anthropological). Yet behind all these variations one can distinguish three main patterns.

121

many scholars, both scientists and humanists, without musing upon its explication. The rst to clearly formulate it among archaeologists was Sophus Mller (1896, 292307). He described the order of archaeological research as follows: (1) (2) (3) (4) Collection of materials; Observation or inspection of materials; Inferences on details; Generalisation inferences on the rules (establishing of types, combinations, styles, cultural groups); (5) Their conrmation by covering with common human rules (in particular connected to modern ones); (6) Turning to the causes of discovered phenomena elucidation of them by comparison and by conclusion with analogies, setting up and proving the hypotheses. The nal objective is the cognition of causal dependence; through gradual generalisation, induction leads an investigator from the material to this nal objective. The proper method of archaeology is ... the safe induction (Mller 1896, 298f). With various modications, this basically inductive scheme is exposed by K.-H. Jacob-Friesen (where the aim is genealogy), G. Clark (where the aim is reconstruction of the past), I. MacWhite, D. Clarke (feedback connections are introduced), J A. Sher (weighting of the attributes is separated into a special step), and E. Chouraqui (methods are joined with the technique of the research). For Rouse, archaeology in a three-step procedure cognises the nature of the remains and then hands the results over to history, which in a four-step procedure terminates in revealing the causes (Jacob-Friesen 1928, 98152; Clark 1957, 9f, 1820, 169f, 174; MacWhite 1956, 37, pl. I; Clarke 1968; Sher 1970; Chouraqui 1972, 212 228; Rouse 1972, 28; a.o.). This scheme was logically worked out and epistemologically substantiated by inductivists, philosophers of early positivism, and especially by Mill J. S. (1914). It was accepted by archaeologists of various philosophical orientations, perhaps because it reected and logically executed some real aspect of empirical research studies. This aspect remains necessary even in those systems where it is not exaggerated and is not the entire essence of the study: induction is inherent to empirical studies and the

4. INDUCTIVE PROCEDURE One of these patterns was formed in the nineteenth century. This is the scheme which was followed by

122

Acta Archaeologica
In Soviet archaeology the course of methods for historical-archaeological studies at Leningrad State University was built into the teaching program according to a similar but more complex scheme. The path of the researcher was traced three times: from observations to hypotheses (logical heuristics), from data to unknown (psychological heuristics) and especially at large from archaeological materials to historical and sociological conclusions (archaeological methods). With establishing the problem, the second of these advancements began, with characteristics of observations as the rst one. Therefore neither step was considered at the third advancement, and it began immediately with criticism of sources: (1) external criticism, (2) inner criticism, (3) description of materials, (4) classication, (5) revealing of connections and placing them in time and space, (6) historical reconstruction, and (7) sociological interpretation. Preferences in this development were the linking of ARD to general scientic aspects of research process (to logical and psychological ones) and the introduction of H. J. Eggers demand for inner criticism of sources in Soviet archaeology. The imperfection of this design was that from the immediate archaeological line of consideration, the step of selection and sampling of materials was lost. The importance of this step was well shown by Binford (1964, 425441), while the general scientic characteristic of observations does not serve as an adequate substitute for such analysis because it does not solve specic problems of an archaeologist. A similar scheme of Bochkarev repeats this aw (and even aggravates it by the absence of parallel lines of considerations) and terminates the archaeological part of the procedure with separating of fractions and cultures. It is questionable if one can commit the interpretation of similarities between cultures to a historian, and the revealing of migrations, inuences, autochthoneity etc. Taylors problem-oriented scheme nds its philosophical substantiation with pragmatist J. Dewey, in a work on problem situation (Dewey 1955, 104f). The general pragmatic orientation of Taylors contextual approach makes this scheme and even its reinforcement natural for him and his followers. As for archaeologists with other philosophical beliefs, for them this scheme seemed acceptable, because it reects and formalises a real and important

widening of factual basis. The introduction of new facts into research circulation is able to produce changeovers of old conceptions, to serve as a stimulus.

5. PROBLEM-ORIENTED PROCEDURE Another scheme of procedure, established long ago as a norm in sciences, was also used in archaeology although rather sporadically and without strict rules. As applied to archaeology it is rst exposed in a clearcut form in the work of Taylor (1948, 152202, table 4). Taylors order of operations looks like this: (A) Determination of the problem in the frames of a conception; (B) Working with materials: (1) Collection of data; (2) Criticism of their usefulness; (3) Analysis; (4) Interpretation (ascertaining the technique, function etc.); (5) Description; (6) Exposition (in publications and the like). (C) Building of local chronologies; (D) Synthesis by connections in the context of discovery; (E) Comparative study of culture in statics and dynamics; (F) Studying the nature of culture, its constants, laws, and functioning; After Taylor the stages (B) and possibly (C) are the concern of archaeology. Further than this it passes the baton to historiography and ethnography (D), ethnology (E) and anthropology (F). In setting out his steps Swartz (1967) follows Taylor: (1) Preparation surveying the preceding studies and anticipation of eld problems. (2) Acquirement of materials in the eld. (3) Analysis placing the data in the network of time and space. (4) Interpretation ascertaining the technology and functioning of artefacts. (56) Integration (a) Reconstruction of the life of a population, and synthesis of large cultural units, and (b) Abstraction, the formulation of laws or principles.

Metaarchaeology
aspect of research process in archaeology: its organic order, emergence and presence of problems in it, and in connection with this, a purpose to the researches, at least to many of them.

123

6. DEDUCTIVE PROCEDURE Not long ago a scheme of procedure widely practised in physical-chemical sciences was suggested to archaeologists. According to the scheme a study begins with advancing an hypothesis from which expectations are deduced and confronted with the material. It is in this confrontation that the proof of the hypothesis consists. This deductive scheme of ARD was rst exposed in L. Binfords works of 19671972 (Binford 1972, 47f, 92f, 114121, 245260) and especially well by J. Fritz and F. Plog (1970). In their article, the core of the procedure is the explanative hypothesis covering the fact of archaeology with the law of anthropology; thus the explanation is derived from theory. The list of stages of the ARD arranged by them is as follows: (1) hypothesis procurement, (2) expectations deriving from it, (3) design of data collection, (4) acquirement of data, (5) data analysis transforming them so as to make them useful for verifying predictions, (6) proving the predictions (establishment of connections between variables), and (7) hypothesis estimation by the results of the proof, the estimation of the explanation. Other American New Archaeologists follow this model (Watson et al. 1971, 114152). The orientation accepted by them on the proof of a certain hypothesis runs through the entire line of research and determines not only the succession but the gure of all its steps as well, from the collection of materials (Only data considered to be relevant will be collected) to the report publication (with only enough data from the site excavated to support the argument). This procedure is not only legitimate but also it is the life of science (Watson et al. 1971, 14, 114, 157). Of course, this is utter exaggeration the deductive scheme becomes a deductivist one. This time the archaeologists themselves indicate the philosophical source of their methodological ideas. This is the hypothetico-deductive explanation scheme of K. Popper K. Hempel T. Nagel (Popper 1935, 26f; Hempel 1942, 3548: Nagel 1961). According to this scheme the researcher explains particular facts by covering

them with a law. This scheme was born on the basis of logical positivism but became well developed only in post-positivism. It was much criticised and rightfully so, in both philosophy and archaeology for its narrowness, limitation, simplication and exorbitant pretensions (Dray 1957; Tuggle et al. 1972; Morgan 1973; a.o.). However, it is shown that explanation through law remains a very central and determining kind of explanation in historical disciplines (Kon 1969), over and above hypothesis is the necessary stage of any explanation, and explanation is the core of a full study. So to in this scheme of ARD, separating itself more and more from its narrow law-explanative basis, it moulds a completely real aspect of archaeological studies. This is why in Russia as well, many studies devoted to proving broad hypotheses follow practically the same scheme, if not in exposition, then at least in execution.

7. CONTRA-DISTINCTIVE DISCUSSION In some respect the problem-orientation scheme of ARD can be viewed as an interim between the two others, as a transitional scheme: a determination of the problem is still not advancement of a hypothesis, but it does limit data collection and processing from the very beginning, and stimulates internal criticism of sources the estimation of their cognitive possibilities as applied to the problem. Therefore the main argument had run high between the partisans of the two other eccentric conceptions of ARD, while the presence of the middle one was simply not given attention. The followers it did have were seen as belonging to one of the extreme conceptions. J. Hill (1972), who advanced in favour of the deductivists, considered the pleas from both of these outside camps. Proponents of the deductivist scheme accuse their opponents that the inductivist scheme is built on erroneous premises: (1) As if one can work in the eld without a priori notions and collect all information contained in sites (vacuum-cleaner approach, Hill 1972, 67); (2) As if data collected this way are universal and any investigator can later use them for solving any tasks; (3) As if every archaeological fact is unambiguous,

124

Acta Archaeologica
stead this must be some overall hypothesis compiled of all possible hypotheses with respect to the essence of the problem, hypotheses practically necessary and realistic. A problem implicitly contains in itself such a set of potential hypotheses something of the kind of those multiple hypotheses of Lloyd Chamberlain or K. Poppers theories in themselves, that were never produced or understood by people but potentially exist (Chamberlain 1944; Popper 1963, vol. 2: 237). This is like a matrix of hypotheses. Naturally, the problem-oriented scheme of ARD, though not free from limitations either, nevertheless possesses the very broad applicability to studies of the three schemes. Of course, an experienced archaeologist is able to direct the expedition so that, among evidences found, more of those appear which are necessary for proving the hypothesis specied, or, for solving the problem outlined. Yet the material will surely be richer and will present surprises. It would be unforgivable dogmatism to ignore this with appeal to a boundless number of possible observation aspects and to the impossibility of describing everything. Plans of research studies are like plans of battles: they are normally thoroughly worked out only to be shoved aside as soon as the battle has begun. However the modern battle, though far from regular, is not a rough-andtumble affair. In real scholarly practice, with respect to every potential source, a limited and not even very large set of actual problems always exists. A similar set of programs for the collection of evidence corresponds to it. For many sources these problems and programs conditioned by a leading theory are common. A pool of problems and programs comes to hand, if not universal then at least rather widely applicable. It gravitates towards forming a broad but standard program that would describe methods and could serve as a basis for particular modications. One such pool, or a few of them, are present in the mind of every well prepared archaeologist, and it is these that provide the possibility of working with alien material drawn from outside.

and facts speak for themselves, and their simple generalisation is sufcient for comprehending the past. Treatment of the facts depends on the aspect considered, and in collecting the information one program often technically excludes another (for instance, you can remove a layer either horizontally or vertically, but not both ways simultaneously). The inductivist procedure leads one to adjust aims to the character of the data collected, and this character is predestined by the unobserved, unrealised readiness of the investigator. By collecting material blindly an archaeologist will catch a lot of details needless to anybody. What criticisms do advocates of the inductivist scheme have of the deductivists? That their opponents force a single aspect on the material, while ignoring the richness and diversity of information contained in the sources, and that this will lead to faults in the collection of data. An orientation of proving the only hypothesis will be psychologically bafing it will stimulate a search for conrmation of this hypothesis and not to search for refutations: (1) That the deductive procedure is not relevant namely for archaeologists because monuments usually present surprises; (2) That there are some projects (salvage excavations, contract work) which are stimulated not by requests of investigators but by demands of life; (3) That archaeologists usually have no possibility of proving hypotheses by experiment. In fact at least a part of the mutual accusations is correct as applied to absolutisation of these schemes of ARD and it shows limitations of both. Situations are possible in which the inductive scheme of ARD is justied (for example, a study of processed materials of salvage projects). There are studies demanding a deductive procedure (subject studies realising an idea, attribution of nds, polemical works). It is essential that these accusations are mostly non-applicable to the intermediate, problem-oriented scheme of ARD. The problem-oriented directive can be likened to advancing a hypothesis to the outset of the research process, but not a narrowly specied hypothesis. In-

8. STRUCTURAL COMPARISON All three kinds of ARD can be applied, each kind in its corresponding conditions. The validity of this

Metaarchaeology
usage and the existence of the standard program of data collection encourage the assumption that there is some common invariant scheme that reposes in the basis of all three schemes of ARD and contains an objective criterion of completeness of research process. It is like a generalised stratigraphic sequence that is more complete than any of the initial proles. It is not difcult to observe that the root structural difference between the generalisations considered by the schemes of ARD lies in placing the train of operations (one step or several steps) connected with the advancing and developing of the hypotheses. In the inductive procedure, this train follows the long succession of operations dealing with the wide collection and processing of the material. So the results of the processing are considered as the premise of the hypothesis and an object for its application. Independent facts for its proof are to be drawn from outside. In the problem-oriented procedure this train of operations is dismembered and only the matrix of hypotheses is replaced, to the forward position. In the deductive procedure this train is moved in full to a forward position, so that acquiring and processing of facts is narrowly oriented on the proof of the specied hypothesis. However, what serves as the factual premise of the hypothesis, what is the object of its application? Apparently, according to the last analysis, it is information on facts already known before. Thus in all cases one and the same cycle is present. It consists of 4 trains of operations: (a) preparation of initial facts, (b) procuring a hypothesis, (c) engaging independent facts, and (d) proving the hypothesis. In the case of the inductive procedure, fresh materials are regarded as initial facts, while the results of previous studies are taken as the engaged facts. In other cases everything is to the contrary: results of previous studies are the initial facts, while fresh materials are the engaged facts. The difference tells of the character of free material processing, of the processing of new data. In the rst case (the inductive procedure) there is an endeavour to collect as applied to any possible hypotheses, and to collect very widely. The problem-oriented procedure is for a certain matrix of hypothses, and in the third case, the deductive procedure, is applied to a single specied hypothesis (i.e. collection is narrow and selective). It is natural that the nearer the study to the eld work, the more

125

dangerous is such selectivity and the more preferable the shift to the opposite procedure. With regard to a common basis and limitations, in each of the three cases, two mutually independent groups of facts are present (LeBlanc 1973). Without fail one of them consists of fresh facts. It undergoes elaboration in the study being proceeded, and the elaboration is aimed at verifying the hypothesis, be it a particular hypothesis or some total indeterminate hypothesis. Anyhow these facts should be obtained and processed so that it would become possible their comparison with other systems of knowledge, and so it would become a possible translation of the information (through hypotheses and their proof) into other systems of concepts: sociohistorical, anthropological or ethnological. It is clear (the rst limitation) that all three types of procedure are anticipated only to a particular, mainly empirical study, but also to a theoretical interpretation study. A procedure of an archaeologists general theoretical and methodological study is as yet not discussed. Unfortunately none of the authors mentioned stipulated this limitation. The second limitation is that the ideal nature of the scheme of ARD means the overall processing of the material, the outside transformation of the evidences extracted from it, running the information through all the steps of the archaeological and historical (or sociological, or culture-anthropological) study. Yet in research practice particular studies are completely permissible and reasonable too, which are conventionally closed in the frames of one of these disciplines or of even one or a few levels of some of them. Accordingly, on the scale of ARD, such study will take only a narrow cut off: one or a few steps. The third limitation of all the schemes of ARD considered imply an archaeologists work with fresh original sources, with material objects collected by the archaeologist, and not elaborated previously. There are rather many studies however that are carried out on the basis of the so called engaged material, i.e. alien, selected long ago and processed on the basis of the literary data. The procedure of such investigation must envisage a regressive proof of the part of operations made by other research workers. Finally, it is necessary to stipulate a circumstamce: ARD regulates the run of the research, the execution

126

Acta Archaeologica
from sources of sociocultural phenomena through their deposition as archaeological materials to their reections in scholarly literature. Of course an archaeologist receives the information in its last stages and converts it by going the other way round. A single plot lies in the basis of all the three sequences: inductive, problem-oriented and deductive. It is clear enough that the choice of any one of these sequences is determined by the character of the study, while exaggeration of some of them is conditioned by the philosophical position of the archaeologist. Yet what is it that provides the structure of the plot itself, the general number of the steps, their qualitative set and succession, and how is all this connected with the philosophy of science? We must not forget that this is conditioned by our understanding of the nature of the archaeological fact, of its epistemological complicity, depth and multi-stage structure (Klejn 1975b, and the preceding chapter here). Beneath of all these varieties of procedure the common base is reposed the process of conversion and reconversion of information through these 14 ltering oodgates. Since this structure is as straight as a ladder, the invariant scheme of ARD is on no account arbitrary.

of the research, but not necessarily its exposition. Recently however, a trend is observable of approaching the scheme of ARD with the plan of the exposition, or at least to explain to the reader the applied scheme of ARD as well as the other structural components of the research. In Russia, the trend was expressed by autoreferates (authors abstracts) of dissertations for candidate degrees, where the free plan of exposition (according to chronology or following the survey of culture components) was changed and came nearer to the conventions used in natural sciences. It became roughly as follows: (1) The task of the investigator, (2) The signicance of the study, (3) Methods of the study, (4) Factual basis of the study, (5) The execution of the study, (6) Results of the study, (7) Discussion of the results (estimation of the results and their inclusion into more general systems), (8) Structure of the exposition in the dissertation paper, and (9) Publication of the results. Thus, what is the logical core and epistemological basis of the full archaeological research study? It is the process transguring the information and leading from the factual material (the immediate objects of archaeology) to its interpretation in terms of history, sociology, cultural anthropology etc. This is the way

Metaarchaeology PART VI. CONCLUSION

127

12. Panorama revisited

1. SUMMARY In the rst part of this chapter I wish to sum up in brief what has been exposed in this volume. I evaluate theory as an important component of archaeology, and the locus where real scientic revolutions take place. I argue that theoreticians are a distinct specialisation in archaeology and that they need to receive the relevant training from an early stage. My denition of theoretical archaeology is not particularly broad for it does not include all methods and all ideas, but neither is it very narrow as if it were a special logical scheme. Within this notion I embrace all philosophical, methodological, logical, theoretical and in part historical problems that concern the whole of archaeology. I also make a distinction between theoretical archaeology and archaeological theory. As to the structure of theoretical archaeology I divide it into three parts: metaarchaeological, endoarchaeogical and paraarchaeological. In dening what the subject matter of archaeology is, I had to choose from three main points: (1) Antiquities exclusively as the subject matter. If so then archaeology is merely an auxiliary discipline of history. (2) Past historical events and processes as the subject matter, which would make archaeology a kind of or a part of history. (3) Both of the elds, but raising the question whether it is one integral discipline? To me the archaeological subject matter includes material antiquities per se, their links and relationships in the system of culture (presumably material culture) and the regulations and causal mechanisms in the basis of all these links and relationships. Yet not causal explanations of the historical events and processes they are the business of history, while archaeology is a source-studying discipline. The important task of a theorist in my opinion is to dene the specicity of archaeological sources since the validity of separating archaeology into a special discipline reposes on it. I hold that the specicity of material antiquities as archaeological sources consists of a double break between them and the past historical reality, breaks which must be restored by the archaeologist. The two breaks are 1) the break in the coding of information (between the language of things and ordinary language) and, 2) the break in tradition (antiquities are things which have no links with or in living culture). All other kinds of sources have one of these breaks but only archaeological ones have both. As concerns the methodological nature of the discipline, archaeology as a source-studying discipline is neither pure science nor a humanity, but an applied science, although it is in working contact with the humanities. Like history it has a task to restore historical process and also like history it cannot reconstruct it in particular details. Yet for history this means that the realisation of the task is impossible without the help of the imagination, and so every historian makes his own narration, paint his own picture. Archaeology reconstructs only typical features and structures, but reconstructs them in full, reconstructs the cultural process with exact methods, yet of course not with such completeness as history, not to such level of particular realisation. Like Clarke, one of my intentions was to axiomatise archaeology, if not to make it into an analytical machine. Yet having this in mind I soon discovered that the set of principles at the basis of the whole discipline was split in two parts, each consisting of six principles, with both sub-sets opposing each other, and both valid in archaeology! Each principle has its counter-principle, which is valid too. So contradictions are inherent in the discipline, hidden in its very basis. On turning my attention to archaeological theory I rst tried to dene what empiricism in archaeology is, what its main hallmarks are. Then I listed the existing denitions of archaeological theory (the systematisation of facts, the ordered totality of concepts, a set of methods, an imposition of philosophical

128

Acta Archaeologica
research procedure, algorithm of the investigation depends crucially on the understanding of this problem. It concerns the realisation of the multistage reconversion of information and the depth of archaeological fact. However, this is the nature of the general research design. There are three competing alternative models of research design: inductive, deductive and problem-oriented, all reliable in archaeology. All three can be reduced to the general research design.

theory onto archaeological material etc.) and suggested my own denition; a program for the processing of archaeological information, a program which is based on some fundamental explanative idea. By transforming the mechanism of the processing into a stereotype, theory evolves to become a method. Components of theory build a set of oppositions where theory is an intermediate chain. For instance, object theory metatheory, or, empirical data theory idea, and so on. The structure of archaeological theory in its dynamics, the way it functions, is rather complex. It is based both on the theoretical and the empirical basis, yet they serve only in the testing of theory, as hypothesis. The main path of information processing does not stem from these two premises but from the fundamental explanatory idea that is created in the conscience of a researcher, and created not by deduction or induction, but by abduction. Then a system of laws is derived from it, and using the language of theory, they feed into an operational apparatus and into a mathematical apparatus of theory. The result is then confronted with the facts and old theories etc., and nally is transformed into conrmed knowledge and a new method. The functions of theory are a rare subject of theoretical consideration. Yet it deserves some attention since many functions are usually ignored or wrongly understood. So the explanative function in archaeology, the one commonly considered most important, is more connected with interpretation than with explanation. The predictive function is closer to the reconstructive one, for what is prediction when speaking of the past? The problem of fact has been tackled to a greater extent in history than in archaeology. Philosophers and historians discovered the deep structure of scientic fact. In archaeology fact also has a deep structure it is structured according to the levels that information passes through going from the past through archaeological records to the researchers mind. I have listed 14 levels with lters between them, information being changed in each, distorted and losing some elements, with new additions occurring too. The cognition that operates means reconversion of this information occurs. The general research design the research model,

2. VIEWS IN A WIDER CONTEXT My theoretical views have formed over a long period, the main points chiey in the 1960s and 1970s, followed by some expansion and restructuring later. I did not aim for sensational innovations, my intention was to obtain sound and systemic theory suitable for practical needs and based on solid grounds. I gradually nd less to add and to change, not implying that I believe it is perfect, but simply that I have tried to do what I can. When I worked out the main points of my theoretical views in the 1960s New Archaeology was in the weather. American New Archaeology searched for laws and archaeological methods to discover real systems of the past, while the British branch hoped to build an analytical machine able to process archaeological material that would lead to the past reality as the output of the system. Something of these goals and convictions entered into my system, and something grew in opposition to the ambitions of the New Archaeology. In the 1970s a new trend evolved in the New Archaeology, marked by the Behavioral Archaeology of Schiffer, Middle Range Theory of Binford and by Clarke searching for the path of information from the past (and to the past) in his innocence trial. In fact this was Post-Processual Archaeology since the attention to laws of the cultural and historical process was replaced by the attention to the formation of archaeological sources. Independently of these events but not independently of German comparison of archaeological sources with written sources I developed along the same direction. In the 1980s another new trend was once again the focus of attention, in which three different traditions

Metaarchaeology
were oddly linked together, three traditions which earlier had seemed incompatible: (a) Neo-Kantian ideas and an attractive indeterminism (in the spirit of R. Collingwood, G. Daniel and W. Taylor) as well as contextualism (of the type of Chang) grown up on this basis; (b) structuralism from C. Levy-Strauss; and (c) Western university-reared Marxism issuing from G. Lucacs, G. Markuse a.o. In this new trend, pieced together from older ones, a reliable academic wing appeared, exemplied by the creation of Ian Hodder (1982; 1986; 1987 a.o.) and a more journalistic wing, if not purely declamative, that of Shanks and Tilley (1987; 1989). Practically the whole of this trend was not post-processual but post-postprocessual, yet let us not to be too pedantic. This trend also arose mostly in the midst of the New Archaeology, this time in its British branch, as a reaction against its extremities such as the belief in the possibility of full and absolutely objective reconstructions, the hopes upon strength and selfdependence of archaeology and its theory. The main positive contribution of this trend, I suppose, is the undermining of the exorbitant enthusiasm of the New Archaeologists for the regular linking of material-culture elements with social and spiritual appearances of once living societies. Hodder and his followers show that ideational systems and social systems had and still have considerable freedom of choice among forms for their expression in material culture and of course not only in it and this introduces signicant alterations into the current notions about regularities of the cultural world. Yet Hodders post-processual archaeology has some problems, which have already been highlighted in several known publications. What seems to me insufciently discussed is the dismissal of archaeology as a unied subject with a unied theory, and the loss of criteria of validation. Childe said archaeology is one. But for Ian Hodder this has turned back to front: The idea of a unied science of archaeology, still held to in North America and briey glimpsed in Scandinavia and Britain in the mid 1970s, is now in total disarray in Europe. The notion that archaeology should have unied theory, method and aims is widely rejected (Hodder 1991a, 19). Hodder explains this diversication with socialpolitical enmity in the contemporary world. For postprocessualists archaeology has little means of objec-

129

tive cognition of the past and the investigation is doomed to dependence on worldviews and political views of the investigator, on its class position. This view is the post-processual extraction from Marxism. The New Archaeology, especially the American family, dismissed connections of archaeology with history. In Britain however, as well as Throughout Europe, archaeologys closest intellectual ties are with history (Hodder 1991a, 10). Archaeology here is fundamentally historical in emphasis, is strongly Marxsist in orientation, and is undeniably social in construction (Hodder 1991b, VIII). Yet Hodders approach is eclectic; it manages to join together various, hardly conformable, traditions. Hodder writes on the self-dependence of archaeology too, ... Over recent decades ... (archaeology) has increasingly been able to dene itself as a discipline independent of history and Classical studies (Hodder 1991a, 7). And While it is argued that archaeology should reassert its European ties with history, it is also important to see the differences between archaeology and history. He admits that archaeology is part of history, but since written sources are created from some material substance like paper and ink, it can be added that history is part of archaeology (Hodder 1991c, 12). Hodder recounts with sympathy Taylors expression that archaeology is neither history, nor anthropology and David Clarkes statement that archaeology is archaeology is archaeology. Yet to Hodder there is insufcient contextualism in Taylor, while in Clarke it is absent altogether. As to Hodder himself archaeology is distinguished from antiquarianism by the stress on the context of every object (1991c, 190f). Thus, the subject matter of archaeology according to post-processualists is things and their context. Archaeology is a historically oriented discipline on material culture, while the latter is understood foremost as a system of symbols or meanings, which one can read like a written text. Reading the past, Hodders wellknown book, was rst published in 1986 (1991c). The problem of the reading of material remains is more complicated than it seems to post-processualists. Besides this, their determination of the subject matter of archaeology is realistic, although it has not actually moved very far from its position of one hundred years ago, the position of Sophus Mller.

130

Acta Archaeologica
theories) is replaced by politics and political theories. This is a dangerous directive for archaeology. Here the followers of western Marxist intellectuals become united with the zealous and die-hard Mohicans of Marxist orthodoxy who still remain somewhere in post-Soviet archaeology. As the saying goes in Russia, we have studied this already ... The philosophical and political preferences of an archaeologist are reected of course in his scholarly production. Large trends in archaeology are undoubtedly conditioned not only by accumulation of facts and by the logic of scholarly discoveries, but also by social shifts in the surroundings, which inuence archaeologists. However, scholarly cognition is distinguished from other spheres of production by the presence of its disciplining rules of proof and selfproof, of control and self-control, by its strict methods, by means to reveal and eliminate the subjective component and biases, be they individual or collective ones. Theory is, of course, present in facts, and biases stick fast in our heads, but that we are aware of this is exactly the point. We are scholars exactly because we are aware of this and can cope therewith. And we are scholars to the extent that we do cope therewith. The only necessary stipulation is that for coping therewith we should not need to know each bias by sight and to be able to see its roots, which would be interesting in other respects, but we do need to have a regular ltering mechanism which screens any bias in general.

Is archaeology not one? This is refuted by the very result of the present survey. All theories and methods, the whole of metaarchaeology, are relevant to all branches of archaeology. And in viewing the other aspect of the split, Triggers once astonishing statement (1978, 196f) of the striking likeness between Soviet and Western archaeological discussions, inferred from reading my Panorama (Klejn 1977a), today can surprise nobody. The world is no longer split into two camps, for the ideological opposition, articial to science including archaeology, ended in natural asco. And archaeology as a discipline, now as well as then, is one. By the whole span of ideological divergence, by the entire variability of its schools and trends, it has the same problems and a shared set of possible solutions. As to the issue of validation criteria my position must now be clear to the reader. Both archaeological facts and conrmed theories are criteria of validation. The idea is ascertained both from Binfords hypothetico-deductive scheme of validation and from Hodders position. Theory cannot be proved by means of data says Hodder (1986, 16). He rejects the concept of Middle Range Theory proposed by Binford as it is based on presumably very regular correspondences between material culture and social behaviour. The mentality of the past people and their freedom of choice, he states, inuenced the specimens created by craftsmen, specimens to be restored in the present, and this introduced uncertainty in their appearance. Besides, it also depends on how we restore them, on our theory and practice of research. In turn, they are determined and conditioned by our own ideology, and our ideas depend on our social interests and political orientations. So far Hodder and his followers. Theory in this understanding is interweaved with practice, not only with an archaeological but also with an ideological and political one. No talks on mutual inuence of facts and theory will save us from the statement that in the writings of Hodder and his adherents, derived knowledge of archaeology loses its clear dependence on facts, theory loses its dependence on proof by facts and begins to reect simply the ideological position of the archaeologist and of his or her social milieu. Theory turns into a simple reection of such a position. Facts as validation criteria are dropped, while conrmed theoretical knowledge (old

3. SELF-IDENTIFICATION During a series of lectures that I gave in Turku (bo), I wanted to confront my attitude with that of Hodders. Having in mind his Reading the past, understood as reading the monuments, I decided to call my philological lecture on the Homeric epic Digging the Text, for I applied typically archaeological methods (typication, stratication, correlation) to the philological analysis of the Iliad. I must admit this was not only in opposition to Hodder but at the same time in support of him in some respect, since a matching between the two elds was nevertheless present. Although my attitude in this case was precisely the contrary to Hodders. It may seem from this discussion that my main op-

Metaarchaeology
ponent is Ian Hodder. The illusion is produced only by the time aberration; his works are the closest in time to my course of lectures that led to the publication of this book, so I should stress the differences between us (and especially my criticism of the writings of Shanks and Tilley). But Hodder, as well as Binford, or Taylor, or Montelius, are in a broader sense just as much my collaborators as they are my opponents. This is all the more, that in my own system of theoretical views there are both agreements with ideas of New archaeology, Behavioral Archaeology, Post-Processual Archaeology, etc., as well as there are oppositions to them. I reiterate that I am not only in opposition to Hodder, but simultaneously in support of him, as some ideas of his are equally important in my own work. During a session one day I sharply criticised a young Moldovan archaeologist for a typically postprocessual dissertation he had written, entitled Archaeology of freedom, and I suspect he had not read Hodder before. He objected, claiming that his main ideas were derived from me. After raking back through my own work, I observed that there were indeed some points in it that strongly resembled postprocessual archaeology. As early as in my Archaeological Typology of 1981, which means I cannot say that I was inuenced by Hodder at that stage, the main idea was that it was impossible to crush archaeological material into small particles and then, uniting them with the help of correlation, to obtain reliable cultural types. The ultimate step would be to arrive at cultures. I insisted that the route must be just the reverse: to grasp the sense of cultures rst, then to reveal types in them, and only then to divide the types into attributes for checking the whole picture. It was of course a post-processual idea. Then I came to the idea of dialectics of principles, located at the basis of all of archaeology. Most surely a post-processual idea. First I advanced with this idea in my First Clarke Memorial Lecture, Cambridge

131

1993, which I read by invitation from Ian Hodder. He was very kind in general and looked very happy with the lecture. Now I realise that the content of my lecture at the time might well have been to his liking. Yet the post-processual ideas do not determine the general outlook of my work. And neither do Structuralist or Post-Structuralist ideas, though they also are present. I have already indicated that I borrowed some ideas from Marxism but even early on I became critical in my attitude to its cardinal failings. In this sense my own position was Post-Marxist. However, this negative term, like all post- types, indicates nothing except the time and the departure. It does not reect my own positive ideas, and neither does the term Post-Processualism. For Post-Processualism means that in its armoury one cannot suggest some remarkable idea that is new and determining. Contextualism was no candidate for it had been around earlier. Post-Processualism searched for a new and determining idea but in vain. When I wrote Attainments and Problems of Soviet archaeology (1982, later revised as Phenomenon of Soviet Archaeology, 1993), I called my position Echeloned Archaeology, with the implication that the path of archaeological investigation must be methodically divided into subsequent steps, none of which can be dropped. I had in mind precocious historical conclusions typical then of some Soviet archaeologists. This path of archaeological investigation however is too general a feature. Now after working on my Principles of Archaeology I am inclined to call my system of views Dialectical Archaeology, for inherent contradictions and paradoxes were the features I always tried to reveal, beginning with Archaeological Sources, 1978. I know that there are already some Marxist archaeologists that call their works Dialectical Archaeology, but to me Marxism and Dialectic are different things. Yet the label is usually coined by adversaries or historiographers. If indeed there were something to hang the label on!

132

Acta Archaeologica APPENDIX

The Commandments
COINED DURING THE YEARS 19641995 TO THE MEMBERS OF L. S. KLEJNS SEMINARS The aphorisms listed here, named Commandments, were hanging on the walls as humorist devices for many years during Klejns seminars in LeningradPetersburg, making long discussions a more joyous occasion. Participators would copy them down from the wall in their notebooks and eventually they were published in 1999 by pupils of Klejns former pupils in Chisineu in the journal Stratum, without the permission or knowledge of the author. The conict was settled however, as the publication was made with good intentions. As the subject was already published, the author recognised the fact and agreed to its translation and republication here. (1) Archaeology is not history armed with a spade, but a detective story in which the investigator has arrived at the scene a thousand years late. History is pronounced later by judges. So you must decide: to go in for one or for the other. (2) Do not be similar to the historian, for whom work is already settled in two steps: collection of materials and the writing of a text. Between these two you must take the third the research. (3) Where there is a law, there is no problem. In every set of facts do not search for laws, but for contradiction to law. Behind contradiction a problem is hidden, behind the problem a discovery. (4) State the question as a question. With nominative sentences a theme is set but not a problem. A problem is set only when it is formulated by a question. The real question begins with who, what, where, when, whence, whither, how and why. (5) The scholarly world is not a team of friends. What is your discovery is a loss for someone else. And this someone is usually a prominent and powerful person. Therefore having made a discovery do not expect universal delight. Be ready for tough resistance, sudden attacks and a gruelling and lingering war. A scholar needs talent secondly and courage rst. Research is a threefold struggle with the material, with adversaries and with oneself. The last part is the hardest. Every scholar has a right to mistakes if he makes mistakes correctly. If an experiment fails once, the experiment is guilty, if it fails twice, the experimenter is guilty, if three times, the theory. Do not check facts with your tongue, but with your teeth; do not search for something tasty, search for something true. Indeed what you need to recognise is not raisins but gold. Argue skilfully and vigorously, but remember that one does not believe your skill or your rage but your facts. Beware of assumptions. Probability is a ladder with rolling steps, an escalator. Before you know it you nd yourself on the next oor. Apparent means probably, probably means possibly, possibly means may be and may be not. But whether it was present or absent, issue from the point that it was absent rather than present. Forget the phrase for instance. Examples can substantiate whatever you want. There is always a counter-example for every example. An example is permissible only when it represents a generalisation. Classication is like a piano, do not try to strike a chord with one nger. You need a sufcient set of concepts and terms. Weigh pros and cons on the same scales. If the complex truth does not consist of simple truths, it is not a truth. The scholarly position is not a chair, but redoubt. It is only a position when it is attacked

(6)

(7) (8)

(9)

(10)

(11)

(12)

(13)

(14) (15) (16)

Metaarchaeology
and defended. Thereafter it is no longer a position but a pose. Do not confuse a position with a pose. Do not hunt for a fashionable position. In the discipline, not every word said last is the last word in the discipline. Contemporaneity is not dened by the moment of a work but by the productivity of methods, completeness of materials, and cleverness of ideas. Do not hope for chance and luck. The law of gravitation was created in Newtons head and not in the apple. Do not suppose anything is apparent. Collect proofs as much as possible, then people will perhaps understand that your idea did not need proving. Be brief. However, rstly every one of your terms should be dened, every concept reasoned, every sentence grounded, every conclusion limited, every fact accounted for, proved and measured.

133

(17)

(18)

(19)

(20)

(21) When you substantiate, it is important what, still more important with what, but most importantly how. (22) The crowning proof is the one which the author has ditched and allowed the reader himself to nd. (23) The golden middle between two extremes is only the third extreme. It must be proved especially well. (24) Do not argue until you get a frog in the throat. You cannot out-argue your adversary, no matter how right you may be. The task of every scholarly argument is not to convince your opponent but to check yourself, to believe in yourself and to gain supporters. (25) Even if a gold coin rings on a copper coin, the ring nevertheless is golden. Inequality is not a hindrance to fruitful communication.

134

Acta Archaeologica

BIBLIOGRAPHY
A dialogue 1994: A dialogue between Iain Mackensie and Michael Shanks. Archaeology: theories, themes and experience. In: Mackensie, I. M. (ed.): Archaeological theories: progress or posture? Avebury (Addershot et al.). 1940. Adams, W. Y. 1968: Invasion, diffusion, evolution? Antiquity 42. 194215. 1973: The archaeologist as detective. In: D. W. Lathrap and J. Douglas (eds.): Variation in anthropology: Essays in honor of John C. McGregor. Ann Arbor, Mich., Braun (Bruneld). 1729. 1988: Archaeological classication: theory versus practice. Antiquity 62. 4056. Akchurin, I. A., Vedenov, M. F., Sachkov, Ju. V. 1968: Dialekticheskaja protivorechivost razvitija sovremennogo estestvoznanija. In: Omeljanovskij, E.M. (ed.): Materialisticheskaja dialektika i metody estestvennyh nauk. Moscow (Nauka). 4369. Amalrik, A. S., Mongajt, A. L. 1957: Chto takoe arkheologija. Moscow (Uchpedgiz). 1959: V poiskach ischeznuvshikh civilizacij. Moskva (AN SSSR). 2nd ed. 1966 (Nauka). Anikovich, M. V. 1975: Pervobytnaja arkheologija konkretnaja istoricheskaja nauka (k postanovke problemy). Predmet i obekt arkheologii i voprosy metodiki istoricheskikh issledovanij. Leningrad (Nauka). 1617. 1988a: Mesto arkheologii v sisteme obshchestvennykh nauk. Aktuelle philosophische und methodologische Fragen der Archologie. Studijne Zvesti Archeologickeho ustavu Slovenskej Akademie vied 25. Nitra. 7378. 1988b: Tri urovnja arkheologicheskogo issledovanija ili tri stupeni istoricheskogo poznanija. Sovetskaja Arkheologija 1988:4. 218224. 1989a: O meste arkheologii v sisteme obshchestvennykh nauk. Kategorii istoricheskikh nauk. Leningrad (Nauka). 7398. 1989b: Problema opredelenija arkheologii kak nauki v russkoj dorevoljucionnoj arkheologii. Istoriki ob istorii. Omsk (Omskij Universitet). 426. Antipov, G. A. 1987: Istoricheskoe proshloe i puti ego poznanija. Novosibirsk (Nauka). Arcikhovskij, A. V. 1929: Novye metody arkheologii. Istorik-Marksist 14. 136155. 1940: Vvedenie v arkheologiju. Moscow (Moskovskogo universiteta). 1955: Osnovy arkheologii. Moscow (Gospolitizdat). Arcikhovskij, A. V., Kiselev, S. V., Smirnov, A. P. 1932: Vozniknovenie, razvitie i ischeznovenie marksistskoj arkheologii. SGAIMK, 12. 4648. Arkh. otkr. 1974: Arkheologicheskie otkrytija 1973 g. Moskva (Nauka). Arkheologicheskie ekspedicii ... 1962. Arkheologicheskie ekspedicii ... 19191956 gg. Ukazatel. Moscow (AN SSSR). Arkheologija ab 1961: Arkheologija SSSR. Svod arkheologicheskich istochnikov. Moscow (AN SSSR) [multivolume edition, issues since 1961]. Arnau, F. 1960: Kunst der Flscher, Flscher der Kunst. 3 000 Jahren Betrug mit Antiquitten. Dsseldorf (Econ. Verl.). [Engl. transl.: Arnau F. 1961. Three thousand years of deception in art and antiquities. London (Jonathan Cape)]. Artamonov, M. I. 1947: Arkheologicheskie teorii proiskhozhdenija indoevropejcev v svete uchenija N. Ja. Marra. Vestnik Leningradskogo universiteta 2. 79106. 1959: Predislovie k Trudam Volgo-Donskoj arheologicheskoj ekspedicii, vol. II. Materialy i issledovanija po arkheologii SSSR 75. Moscow (Akademija nauk). 58. Artanovskij, A. S. 1967: Istoricheskoe edinstvo chelovechestva i vzaimnoe vlijanie kultur. Leningrad (Prosveshchenie). Ascher, R. 1961a: Analogy in archaeological interpretation. Southwestern Journal of Anthropology 17. 317325 [reprinted in: Deetz, J. (ed.). Mans imprint from the past. Boston (Little, Brown and Co.). 262271]. 1961b: Experimental archaeology. American Anthropologist 63 (4): 793816. 1962: Ethnography for archaeology: a case from the Seri Indians. Ethnology 1. 360369. Atkinson, R. F. 1978: Knowledge and explanation in history. An introduction to the philosophy of history. London et al. (MacMillan Press). Atkinson, R. J. 1965: [Review of K. Pyddoke: What is archeology]. Antiquity 156. 320. 1975: British prehistory and the radiocarbon revolution. Antiquity 195. 173177. Austin, D. 1987: The future of archaeology in British universities. Antiquity 232. 227238. Avdeev, N. 1925: O nauchnoj obrabotke istochnikov po istorii RKP(b) i Oktjabrskoj revolucii. Proletarskaja revoljucija 1. 150 159. Barg, M. A. 1984: Kategorii i metody istoricheskoj nauki. Moscow (Nauka). Barich, B. E. 19771982: Archeologia Theoretica. Origini XI. (Roma). 744. Bashilov, V. A. 1978: Konferencija po probleme istorizma arkheologii. Kratkie sobshchenija IIMK AN SSSR (Moscow). 152. 109111. Bashilov, V. A., Loone, E. N. 1986: Ob urovnjakh issledovanija i poznavatelnukh zadachakh arkheologii. Sovetskaja Arkheologija 3. 192208. Bauer, E. R. 1935: Teoreticheskaja biologija. Moscow (VIEM). Bayard, D. 1978: 15 Jahre New archaeology. Eine kritische bersicht. Saeculum XXIX:I. 69109. Bazhenov, L. B. 1968: Sovremennaja nauchnaja gipoteza. In: M. E. Omeljanovskij (ed.): Materialisticheskaja dialektika i metody estestvennykh nauk. Moscow (Nauka). 294321. 1973: Stroenie i funkcii estestvenno-nauchnoj teorii. Sintez sovremennogo nauchnogo znanija. Moscow (Nauka). 390420. 1979: Stroenie estestvenno-nauchnoj teorii. Moscow. Becker, C. L. 1958: What are historical facts? In: Becker, C. L.: Detachment and the writing of history. Essays and letters. Ithaca, N. Y. (Cornell University Press). 4164. Bekon, F. [Bacon, F.]. 1977: O dostoinstve i priumnozhenii nauk. In: Bekon, F. Sochinenija v dvukh tomakh. Izd. 2-e, t. 1. Moscow (Mysl).

Metaarchaeology
Bem, Ju. O. 1960: Nauchno-teoreticheskaja konferencija, posvjashchennaja 1100-letiju Novgoroda. Voprosy istorii 1. 199205. Berelson, B., Steiner, G.A. 1964: Human behavior: An inventory of scientic ndings. New York (Harcourt, Brace & World). Bernbeck, R. 1997: Theorien in der Archologie. Tbingen, Basel (A. Francke). Berngejm, E. [Bernheim, E.] 1908: Vvedenie v istoricheskuju nauku. Moscow (M. N. Prokopovich). Bertalanffy, L. von 1932: Theoretische Biologie bd. 1. Berlin. 1962: General system theory a critical review. General systems VII. 120. [Russ. transl.: Bertalan, L. 1969: Obshchaja teorija sistem kriticheskij obzor. In: Issledovanija po obshchej teorii sistem. Moscow (Nauka). 2382.] Beurlen, K. 1935: Bedeutung und Aufgabe geologischer Forschung. I. Zur Kritik des Aktualismus. Zeitschrift fr gesamte Naturwissenschaft 12. 2336. Bibler, V. S. 1969: Istoricheskij fakt kak fragment dejstvitelnosti (Logicheskie zametki). In: Istochnikovedenie. Teoreticheskie i metodicheskie problemy. Moscow (Nauka). 98101. Bidney, D. 1953: Theoretical Anthropology. New York (Columbia University Press). Binford, L. R. 1962: Archaeology as anthropology. American Antiquity 28:2. 217225. 1964: A consideration of research design. American Antiquity 29. 425441. 1965: Archaeological systematics and the study of culture process. American Antiquity 31. 203210. 1967a: Smudge pits and hide smoking: The use of analogy in archaeological reasoning. American Antiquity 32:1. 112. 1967b: Some comments on Historical vs. Processual archeology. Southwestern Journal of Anthropology 24. 267275. 1968: Archaeological perspectives. In: Binford, S. R. and L. R. (eds.): New perspectives in archaeology. Chicago (Aldine). 532. 1968/1972: [Review of R. F. Heizer and J. A. Graham: A guide to eld methods in archeology.] American Anthropologist 70. 806 808. [Reprinted in: L. R. Binford: An archaeological perspective. Seminar Press, New York and London, 1972: 182184). 1972: An archaeological perspective. New York London (Seminar Press). 1977: For theory building in archaeology. New York et al. (Academic Press). 1978: Nunamiut ethnoarchaeology. New York et al. (Academic Press). 1982: Objectivity explanation archaeology 1981. In: Renfrew, C. et al. (eds.): Theory and explanation in archaeology. New York (John Wiley & Sons). 125138. 1983/1988: In pursuit of the past. Decoding the archaeological record. London (Thames and Hudson). 1989a: Data, relativism, and archaeological science (1987). In: Binford, L. R.: Debating archaeology. San Diego et al. (Academic Press). 5557. 1989b: Debating archaeology. San Diego et al. (Academic Press). Binford, S. R. and L. R. (eds.) 1968: New perspectives in archaeology. Chicago (Aldine). Blok, M. [Bloch, M.] 1973: Apologija istorii. Moscow (Nauka). Bochkarev, V. S. 1973: K voprosu o strukture arkheologicheskogo issledovanija. In: Tezisy dokladov sessii, posvjashchennoj itogam pole-

135

vykh arkheologicheskikh issledovanij 1972 goda v SSSR. Tashkent (FAN). 5660. 1975: K voprosu o sisteme osnovnykh arkheologicheskikh ponjatij. In: Predmet i objekt arkheologii i voprosy metodiki arkheologicheskikh issledovanij. Leningrad (Nauka). Bohr, N. 1933: Light and life. Nature 131. 421ff. [Russ. transl.: Bor, N. 1961. Atomnaja zika i chelovecheskoe poznanie. Moscow: 1526]. Borjaz, V. N. 1975: Priroda arkeologicheskogo istochnika i obekt arkheologii kak nauki. In: Predmet i obekt arkheologii i voprosy metodiki arkheologicheskikh issledovanij. Leningrad (Nauka). 611. 1976. Metodologicheskie predposylki i principy vydelenija obekta arkheologicheskoj nauki. In: Materialisticheskaja dialektika i chastnye nauki. Leningrad (Nauka). 185216. 1988: Teorija poznanija dialekticheskogo materializma metodologicheskaja osnova. Razrabotki gnoseologicheskikh problem arkheologii. In: Aktualnye losofskie i metodologicheskie problemy arkheologii. Studijne zvesti Archelogickeho ustavu Slovenskej Akade mie vied 25. Nitra. 101105. Born, M. 1955: The state of ideas in physics and the prospects of their further development. Proceedings of Physical Society 66 (402A) 1953. 501520). [Russ. transl.: Sostojanie idej v zike i perspektivy ich dalnejshego razvitija. In: Voprosy prichinnosti v kvantovoj mechanike. Moscow (Inostrannaja Literatura). 102121.] Boucher de Perthes, J. 1863: Sous dix rois, souvenirs de 1791 a 1860 ` vol. 12. Paris (Jung-Treuttel). Board, M. de. 1982: Archeologie et archeometrie. Quelques re exions. Archeologie medievale XII. 715. Braidwood, R. J. 1952: The Near East and foundations for civilization. Eugene (Oregon State System of Higher Education). 1959a: Archaeologists and what they do. New York (Franklin Watts Inc.) 1959b: Archaeology and the evolutionary theory. In: Meggers, B.J. (ed.): Evolution and anthropology: a centennial appraisal. Washington D.C. (Anthropological Association of Washington). 76 89. Braidwood, R. J., Willey, G. R. (eds.). 1962: Courses toward urban life. Viking Fund Publications in anthropology 32. Chicago (Aldine). Brew, J. O. 1946: The use and abuse of taxonomy. The archaeology of Alkali Ridge, Southern Utah. Papers of the Peabody Museum 21. Cambridge, Mass. 4466. Brezillon, M. N. 1968: La denomination des objects de pierre taille. Paris (CNRS). Brjusov, A. Ja. 1954: Nekotorye teoreticheskie osnovy khronologii neolita. Sovetskaja Akheologija XVIII. 1348. Brjusov, A. Ja., Zimina, M. P. 1966: Kamennye sverlenye boevye topory na territorii evropejskoj chasti SSSR. Svod arkheologicheskikh istochnikov E 44. Moscow (Nauka). Brown, L. A. 1968: Diffusion processes and location. Philadelphia (Regional Science Research Institute). Brown, R. R. 1963: Explanation in social science. Chicago (Aldine). Brndsted, J. 1938: Danmarks oldtid. Bd. I. Kbenhavn (Gyldendal Nordisk forlag) [Germ. transl.: Brndsted, J. 1960: Nordische Vorzeit. Bd. I. Neumnster (Wachholtz)]. Buettner-Janusch, J. 1957: Boas and Mason: particularism versus generalization. American Anthropologist 7. 318324.

136

Acta Archaeologica
1972b. Models and Paradigms in archaeology. In: Clarke, D. L. (ed.): Models in archaeology. London (Methuen). 160. 1973: Archaeology: the loss of innocence. Antiquity 47. 618. Coles, J. M. 1966: Experimental archaeology. Proceedings of the Society of Antiquaries of Scotland 99. 120. 1973: Archaeology by experiment. London (Hutchinson University Library). 1979: Experimental archaeology. London et al. (Academic Press). Collingwood, R. 1956: The idea of history. New York (Oxford University Press) [ed. of 1966, Oxford]. Collins, M. B. 1975: Sources of bias in processual data: An appraisal. In: J. W. Mueller (ed.): Sampling in archaeology. Tucson (University of Arizona Press). 2632. Conze, A. 1869: ber die Bedeutung der classischen Archologie. Antrittsvorlesung. Wien (Universittsverlag). Courbin, P. 1982: Quest-ce que larcheologie? Essai sur la nature de la recherche archeologique. Paris (Payot). [Engl. transl.: Courbin, P. 1988: What is archaeology? An essay on the nature of archaeological research. Chicago and London (University of Chicago Press)]. Crawford, O. G. S. 1960: Archaeology in the eld. London (Phoenix House) [1st ed. 1954]. Dajson, F. Dzh. (Dyson Ph. J.) 1967: Matematika v zicheskikh naukakh. In: Matematika v sovremennom mire. Moscow (Mir). 111 128. Daniel, G. E. 1950: A hundred years of archaeology. London (Duckworth). 1962: The idea of prehistory. London (C. A. Watts). 1967a: Editorials. Antiquity 162. 169173. 1967b: The origins and growth of archaeology. Baltimore Maryland, Harmonsworth (Penguin Books). 1969: Editorials. Antiquity 170. 8590. Daniel, G. E., Renfrew, C. 1975: A hundred and fty years of archaeology. London (Duckworth). Daniels, S. G. H. 1972: Research design models. In: D. L. Clarke (ed.): Models in archaeology. London (Methuen). 201229. 1978: Implications of error: research design and the structure of archaeology. World Archaeology 10:1. 2935. Danilenko, V. N. 1969: Neolit Ukrainy. Kiev (Naukova Dumka). Danilov, A. I. 1963: Marksistsko-leninskaja teorija otrazhenija i istoricheskaja nauka. Srednie veka 24. Moscow. 323. Dechelette, J. 1908: Manuel darcheologie prehistorique, celtique et gallo romaine. I. Archeologie prehistorique. Paris (Auguste Picard). Deering, M. F. [Diring, M. F.] 1987. Arkhitektura mashin dlya iskusstvennogo intellekta. In: Realnost i prognozy iskusstvennogo intellekta. Moscow (Mir). 209230. Deetz, J. 1967: Invitation to archaeology. Garden City, N. Y. (The Natural History Press). 1971: Must archaeologists dig? In: Deetz J. (ed.): Mans imprint from the past. Boston (Little, Brown & Co.). 29. 1971 (ed.): Mans imprint from the past. Readings in the methods of archaeology. Boston (Little, Brown & Co). Donna, W. 1922: Larcheologie, son domain, ses buts. Paris (Flamma rion). Desayes, J., Gardin, J.-C. 1956: Le chier mechanographique de loutillage: outils du Bronze de Balkans a lIndus. Beirut (Institut francais dar ` cheologie).

Bulle, H. 1913: Einleitung. In: Bulle, H. (ed.): Handbuch der Archologie. Bd. I., Lieferung 1. Mnchen (H. Beck). Bunge, W. 1962: Theoretical geography. Lund Studies in Geography, series C,1. (2d ed. 1966). Buschor, E. 1939/1969: Begriff und Methode der Archologie. In: Hausmann, U. (ed.): Handbuch der Archologie. Bd. 1. Allgemeine Grundlagen der Archologie. Mnchen (C. N. Beck). 310. Bykovskij, S. N. 1931: Metodika istoricheskogo issledovanija. Obrazovatelnaja biblioteka GAIMK, vyp. 2. Leningrad. 1932a: K itogam Vserossijskogo arkheologo-etnogracheskogo soveshchanija. Soobshchenija GAIMK 1112. 8. 1932b: O predmete istorii materialnoj kultury. Soobshchenija GAIMK 12. 35. Campbell, L. 1994: Traces of archaeology. In: Mackensie, I. M.: Archaeological theory: process or posture? Aldershorst et al. (Avebury). 140151. Carneiro, R. L. 1969: A quantitative law in anthropology. American Antiquity 35 (4). 492494. 1970: A theory of the origin of the state. Science 169. 733738. Carr, E. H. 1964: What is history? Harmondsworth (Penguin Books). Caws, P. 1965: The philosophy of science. A systematic account. Princeton (D. Van Nostrand). Chamberlain, T. D. 1944: The method of multiple working hypotheses. Scientic Monthly 59. 357362. Champion, T. 1991a: Theoretical archaeology in Britain. In: Hodder, I. (ed.): Archaeological theory in Europe: the last 3 decades. London and New York (Routledge).129160. Chang, K. C. (Chang Kuang-Chih) 1967a: Major aspects of the interrelationship of archeology and ethnology. Current Anthropology 8:3. 227234. 1967b: Rethinking archaeology. New York (Random House). Chapman, J., Dolukhanov, P. 1993: Cultural transformations and interactions in Eastern Europe: theory and terminology. In: Chapman, J. and Dolukhanov, P. (eds.): Cultural transformations and interactions in Eastern Europe. Aldershot et al. (Avebury). 136. Cherry, J. F., Gamble, C., Shennan, S. 1978 (eds.): Sampling in contemporary British archaeology. BAR (Brit. Ser.) 50. Childe, V. G. 1936: Changing methods and aims in prehistory. Proceedings of the Prehistoric Society 1935:1. 316. 1954: Archaeological documents for the prehistory of science. Cahiers dhistoire mondiale 1:4. 739759. 1956: Piecing together the past: The interpretation of archaeological data. London (Routledge and Kegan Paul). Chouraqui, R. C. 1972: Languages formels et reconnaissance des formes. Les Methodes mathematiques de larcheologie. Marseille. 212 228. Clark, J. G. D. 1939: Archaeology and society. London (Methuen). [3d ed.: 1960, New York (Barnes & Noble University Press)]. 1957: Archaeology and society: Reconstructing the prehistoric past. 2nd ed. Cambridge, Mass. (Harvard University Press). 1963: Foreword. In: Brothwell, D. and Higgs, E.: Science in archaeology. London (Thames and Hudson). 1719. Clarke, D. L. 1968: Analytical archaeology. London (Methuen). 1970: Analytical archaeology: epilogue. Norwegian Archaeological Review 3. 2534. (ed.) 1972a: Models in archaeology. London (Methuen).

Metaarchaeology
Dewey, J. 1955 [1st ed. 1938]: Logic: the theory of inquire. New York et al. (Holt, Reinehart and Winston]. Dikshit, S. K. 1960: Vvedenije v arkheologiju. Moscow (Inostrannaja Literatura). [transl. from fr. orig. 1952]. Djindjian, F. 1991: Methodes pour larcheologie. Paris (Colin). Donnan, Chr. B., Clewlow Jr., C. W. (eds.): 1974: Ethnoarchaeology. University of California at Los Angeles Institute of archaeology Monograph 4. Doran, M. J. 1970: Archaeological reasoning and machine reasoning. In: Gardin, J.-C. (ed.): Archeologie et calculateurs. Paris (edi tions du Centre National de la Recherche Scientique). 5766. Doroshenko, N. M. 1968: K voprosu o kharaktere istoricheskogo znanija. Problemy losoi i sociologii. Leningrad (Leningradskij universitet). 2429. 1971: K probleme fakta v istoricheskom poznanii. Metodologicheskie voprosy obshchestvennykh nauk 2. Leningrad (Leningradskogo universiteta). 3441. Dray, W. H. 1957: Laws and explanation in history. London (Oxford University Press). Droysen, J. G. 1937: Historik. Vorlesungen ber Enzyklopdie und Methodologie der Geschichte. Mnchen und Berlin (R. Oldenburg). Dunnell, R. C. 1971: Systematics in prehistory. New York and London (The Free Press and Colliers-Macmillan). 1978: Archaeological potential of anthropological and scientic models of function. In: Dunnell, R. C. and Hall, E. S. (eds.): Archaeological essays in honor of Irving Rouse. The Hague, Paris and New York (Mouton Publs.). 4173. Dunnell, R. C., Shimek, J. 1984: What does the archaeological record mean? Reviews in Anthropology 11. 111. Dymond, D. 1974: Archaeology and history. A plea for reconcilation. London (Thames and Hudson). Dynin, B. S. 1972: K voprosu o kharaktere problem metodologii. In: Lektorskij, V. A. (ed.): Filosoja. Metodologija. Nauka. Moscow (Nauka). 4569. Ebbesen, K. 1983: Indledning til en arkologisk kildekritik. Aarbger for Nordisk Oldkyndighed og Historie 1981. 160181. Eckermann, W., Mohr, N. 1966: Einfhrung in das Studium der Geschichte. Berlin (Deutscher Verlag der Wissenschaften). Edmonson, M. S. 1961: Neolithic diffusion-rates. Current Anthropology 2:1. 71102. Eggers, H.-J. 1950: Das Problem der ethnischen Deutung in der Frhgeschichte. In: Dauer, A. und Kirchner, H. (eds.): Ur- und Frhgeschichte als historische Wissenschaft (Wahle-Festschrift). Heidelberg (Carl Winter Universittsverlag). 4959. 1959: Einfhrung in die Vorgeschichte. Mnchen (Piper). Eggert, M. K. H. 1982: Comment I. On form and content. In: Renfrew, C. et al. (eds.): Theory and explanation in archaeology. New York (Academic Press). 139146. Einstein, A. 1933: On the method of theoretical physic. [Quoted after the collection: Ejnshtejn, A. 1967: Sobranie nauchnykh trudov v 4-kh tomakh. vol. 4. Moscow (Nauka). 181186. 1936/1965: Physics and reality. Franklin Institut Journal 221. 313347 [Russ. transl.: Ejnshtejn, A. 1965: Fizika i realnost. Sbornik statej. Moscow (Nauka). 3860. Embree, L. 1989: The structure of American archaeology: a preliminary report. In: Pinsky, V. and Wylie, A. (eds.): Critical tra-

137

ditions in contemporary archaeology. Cambridge (Cambridge University Press). 2837. (ed.). 1992: Metaarchaeology: reections by archaeologists and philosophers. Boston (Kluwer). Engels, F. 1952: Dialektik der Natur. Bcherei des Marxismus-Leninismus 18. Berlin (Dietz). [Russ. transl.: Engels, F. 187383/ 1961: Dialektika prirody. In: Marks, K and Engels, F: Sochinenija. Moscow (Gospolitizdat). Vol. 20, 339756.] Ennals, J. R., Brough, D. R. 1982: Representing the knowledge of the expert archaeologist. Computer applications in archaeology. 132 144. Erofeev, N. A. 1976: Chto takoe istoriya. Moscow (Nauka). Fainberg, E. L. 1988: Nils Bor, Moskva 1961. In: Nils Bor i nauka XX veka. Kiev (Naukova dumka). 2737. Farsobin, V. V. 1983: Istochnikovedenie i ego metod. Opyt analiza ponjatij i technologii. Moscow (Nauka). FE: Filosofskaja Enciklopedija, vol. 2, p. 369. Fedoseev, P. N., Francev, Ju. P. 1964: O razrabotke metodologicheskikh voprosov istorii. Voprosy istorii 3. 422. Feldman, J. A. [Feldman, Dzh. A.] 1987: Setevye modeli. In: Real nost i prognozy iskusstvennogo intellekta. Moscow (Mir). 137147. Fetten, F. G. 1993: Kulturgeschichtlicher Aktualismus und Ethnoarchologie. Thesen zu einer aktualistischen Archologie. Ethnographisch-Archologische Zeitschrift 34. 317339. Filin, F. P. 1972: Proiskhozhdenie russkogo, belorusskogo i ukrainskogo jazykov. Leningrad (Nauka). Fischer, U. 1987: Zur Ratio der prhistorischen Archologie. Germania 65:1. 175195. Fjustel de Kulanzh, N. D. 1907: Istorija obshchestvennogo stroja drevnej Francii. t.3. Frankskaja monarkhija. Saint Petersburg (I. N. Kushnerev). Flannery, K. V. 1972: Culture history vs. culture process: A debate in American archaeology. In: Contemporary archaeology: A guide to theory and contributions. Carbondale and Edwardsville (Southern illinois University Press) London and Amsterdam (Feffer & Simonds). 102107. 1973: Archeology with a capital S. In: Redman, Ch. (ed.). 1973: Research and theory in current archaeology. New York et al. (John Wiley). 4753. Fleming, A., Jonson, M. 1990: The Theoretical Archaeology Group (TAG): origins, retrospect, prospect. Antiquity 243. 303 306. Ford, J. A. 1954a: Comment on A. C. Spauldings Statistical technique for the discovery of artifact types. American Antiquity 19. 390391. 1954b: The type concept revisited. American Anthropologist 56. 4254. 1962a: A quantitative method for deriving cultural chronology. Pan-American Union Technical Manual 1. Washington, D.C. [Ford, J. A.] 1962b: Kolichestvennyj metod ustanovlenija arkheologicheskoj khronologii. Sovetskaja etnograja 1. 3243. Formozov, A. A. 1961: Ocherki po istorii russkoj arkheologii. Moscow (Akademii nauk SSSR). 1977: O kritike istochnikov v archeologii. Sovetskaja archeologija 1977:1. 514. Foucault, M. 1968: Larcheologie du savoir. Paris (Gallimard).

138

Acta Archaeologica
1982b: Struktura sistemy arkheologicheskogo znanija (K voprosu o metodologicheskom analize urovnej znanija v arkheologii). Metodologijcheskie i metodicheskie voprosy arkheologii. Kiev (Naukova dumka). 547. 1983: Predmet i objekt v arkheologii. Kiev (Naukova dumka). 1985: Teoreticheskoe znanie v arkheologii. Metodologicheskie problemy obshchestvennykh nauk. Kiev (Naukova dumka). 622. 1989a: Arkheologija celostnaja nauchnaja sistema ili dieltantskie vylazki i polufabrikat znanija? Sovetskaja Arkheologija 1989:3. 215228. 1989b: Struktura archeologicheskogo poznanija. Problemy socialno-istoricheskogo issledovanija. Kiev (Naukova dumka). Gerasimov, L. G. 1972: Nauchnoe issledovanie. Moscow (Politizdat). Ghoneim-Graf, I. 1978: Mglichkeiten und Grenzen der archologischen Interpretation. Eine aktual-archologische Untersuchung an afrikanischen Gruppen. Arbeiten zur Urgeschichte des Menschen. Bd.3. Frankfurt a.M. et al. (Lang). Gibbon, G. E. 1984: Anthropological archaeology. New York (Columbia University Press). 1989: Explanation in archaeology. London, Oxford (Basil Blackwell). Giro, P. [Guiraud, P.] 1898: Fjustel de Kulanzh. Moscow (I. N. Kushnarev). Gjessing, G. 1962: Comment on epistemology and archaeological theory. Current Anthropology 3. 504505. Gleziou, S., Coudart, A., Demoule, J.-P., Schnapp, A. 1991: The use of theory in French archaeology. In: Hodder, I. (ed.): Archaeological theory in Europe: The last three decades. London and New York (Routledge). 91128. Godowski, K. 1962: Uwagi o niektorych zagadnieniach interpretacji zrode archeologicznych. Prace i materialy Muzeum archeologicznego i etnogracznego w odzi, ser. archeologiii, 8, odz. 79102. Goethe, W. 1949: Maximen und Reexionen. Neu geordnet, eingeleitet und erlutert von G. Mller. Stuttgart (Krmer). Goodman, N. 1951: The structure of appearance. Cambridge, Mass. (Harvard University Press) [3d ed. 1977 Dordrecht, Boston (Reidel Publ.)]. 1965: Fact, ction and forecast. 2nd ed. New York (Bobbs Merill). Gorodcov, V. A. 1908: Pervobytnaja arkheologija. Moscow (Pechatnja Snegirevskoj). Gorsky, D. P. 1981: Denition (Logico-methodological problems). Moscow (Progress). Gould, R. A. 1968: Living archaeology: The Ngatatjara of western Australia. Southwestern Journal of Anthropology 24. 101121. 1971: The archaologist as ethnographer: a case from the Western Desert of Australia. World Archaeology 3:2. 143177. 1977: Ethno-archaeology: or, where do models come from? In: Wright, R. V. S. (ed.): Stone tools as cultural markers. Canberra (Australian Institute of Aboriginal Studies). 162168. 1978: Beyond analogy in ethnoarchaeology. In: Gould, R. A. (ed.): Explorations in ethnoarchaeology. Albuquerque (University of New Mexico Press). 249294. (ed.) 1978: Explorations in ethnoarchaeology. Albuquerque (University of New Mexico Press). 1979: Living archaeology. Cambridge (Cambridge University Press).

[Russ. transl.: Fuko, M. 1977: Slova i veshchi. Archeologija gumanitarnych nauk. Moscow (Nauka)]. Fowler, P. J. 1977: Approaches to archaeology. London (Book Club Associates). Francuzova, N. P. 1972: Istoricheskij metod v nauchnom poznanii. Moscow (Mysl). Fritz, J. M. 1972: Archaeological systems for indirect observation of the past. In: Leone, M. P. (ed.) 1972: Contemporary archaeology. A guide to theory and contributions. Carbondale and Edwardsville (Southern Illinoice University Press). 135157. Fritz, J. M., Plog, F. T. 1970: The nature of archaeological explanation. American Antiquity 35. 405412. Frolov, I. T. 1968: Sistema metodov biologicheskogo issledovanija. In: Materialisticheskaja dialektika i metody estestvennykh nauk. Moscow (Nauka). 132170. Frumkina, R. M. 1965: Ponimanie teksta v uslovijakh ogranichennogo znanija slovarja. Nauchno-tekhnicheskaja informacija 1:4. 4448. Frumkina, R. M., Dobrovich, P. 1971: Kak luzirovat Simenona? Znanie sila 1:1. 3137. Furtwngler, A. 1908: Einleitung zu einer geplanten griechischen Kunstgeschichte. Aus dem Nachlas. Deutsche Rundschau 231. Fustel de Coulanges, N. D. [Fustel de Coulanges, N. D.] 1922: Histoire des institutions politiques de lAncienne France (4th ed.) vol. 4. Paris (Hachette). Gaffney, C. F., Gaffney, V. L. (eds.). 1987: Pragmatic archaeology: Theory in crisis? British Archaeological Reports (BAR) Brit. ser. 167. Gallay, A. 1989: Logicism: a French view of archaeological theory founded in computational perspective. Antiquity 63. 2739. Gardin, J.-C. 1963: Problemes danalyse descriptive en archeologie. ` In: Courbin, P. (ed.): Etudes archeologiques. Paris (Service dEdi tion et de Vente des Publications de lEducation Nationale). 133150. 1967: Methods for the descriptive analysis of archaeological material. American Antiquity 32. 1330. 1974: Les projets de banques de donnees archeologiques. Ban ques de donnees archeologiques. Colloques Nationaux du CNRS, Marseille, 1972. Paris (CNRS). 1526. 1979/1983: Une archeologie theorique. Paris (Hachette). [Russ. transl.: Garden, Zh.-K. 1983: Teoreticheskaja arkheologija. Moscow (Progress)]. 1980: Archaeological constructs. An aspect of theoretical archaeology. Cambridge (Cambridge University Press). Gardin, J.-C. et al. 1987: Systemes experts et sciences humains: les cas de ` larcheologie. Paris (Eyroplles). Gegel, G. V. F. [Hegel, G. W. F.] 1935: Sochinenija. vol. 8. Moscow, Leningrad. Gening, V. F. 1975a: Obekt i predmet arkheologii. Novejshie otkrytija sovetskikh arkheologov III. Kiev (Inst. Arkheol. AN USSR). 1417. 1975b: Specicheskij predmet i nekotorye aktualnye zadachi sovremennoj arkheologii. Voprosy arkheologii Urala 13. 516. (ed.) 1980: Teorija i metody arkheologicheskikh issledovanij. Kiev (Naukova dumka). 1982a: Problema teoretizacii nauchnogo znanija v arkheologii. In: Gening, V. F. (ed.): Novye metody arkheologicheskich issledovanij. Kiev (Naukova dumka). 622.

Metaarchaeology
1985: The empiricist strikes back: reply to Binford. American Antiquity 50:3. 638644. Graebner, F. 1911: Methoden der Ethnologie. Heidelberg (Universittsbuchhandlung. Gotta J. 1839). Grifn, J. B. 1943: The Fort Ancient Aspect, its cultural and chronological position in Mississipi Valley archaeology. Ann Arbor. (Univ. Michigan Press). 1956: The study of early cultures. In: Shapiro, H. L. (ed.): Man, culture, and society. New York (Oxford University Press). 2248. Grigorev, G. P. 1973: O predmete arkheologii. In: Tezisy dokladov sessii, posvjashchennoj itogam polevykh arkheologicheskikh issledovanij 1972 g. v SSSR. Tashkent (FAN). 4143. 1981: O predmete arkheologii. In: Opisanie i analiz arkheologicheskikh istochnikov. Irkutsk (Irkutskij universitet). 35. Grushin, B. A. 1961: Ocherki logiki istoricheskogo razvitija. Moscow (Vysshaja Shkola). Guliaev, V. I. 1993: The theoretical archaeology in the U.S.S.R. In: Theory and practice of prehistory: views from edges of Europe. Santander Madrid (Universidad de Cabtabria C.S.I.C.). 327 345. Gulyga, A. V. 1969: Istorija kak nauka. In: Filosofskie problemy istoricheskoj nauki. Moscow (Nauka). 750. Gumilev, L. N. 1989: Etnogenez i biosfera Zemli. Leningrad (Izd. Leningradskogo universiteta). Guntau, M. 1967: Der Aktualismus in den geologischen Wissenschaften. Leipzig (Deutsche Verlag fr Grundstofndustrie). Gurevich, A. Ja. 1969a: Chto takoe istoricheskij fakt? In. Istochnikovedenie. Teoreticheskie i metodicheskie problemy. Moscow (Nauka). 59 88. 1969b: Ob istoricheskoj zakonomernosti. In: Filosofskie problemy istoricheskoj nauki. Moscow (Nauka) 5179. Haag, W. G. 1969: The archaeologists presumptions. In: Tyler, St. A. (ed.): Concepts and assumptions in contemporary anthropology. Southern Anthropological Society, Proceedings, no. 3. Athens. 4550. Hachmann, R. 1970: Die Goten und Skandinavien. Berlin (Walter de Gruyter & Co). 1971: Die Germanen. Archaeologia Mundi. Mnchen et al. (Nagel). Hachmann, R., Kossack, G., Kuhn H. 1962: Vlker zwischen Germanen und Kelten. Neumnster (Wachholtz). Hgerstrand, T. 1952: The propagation of innovation waves. Lund studies in geography, ser. B, 4. Lund. Handbuch 1939: Handbuch der Archologie, Bd. I. Mnchen (C. H. Beck). Hanson, N. R. 1965: Patterns of discovery. An inquiry into the conceptual foundations of science. Cambridge (Cambridge University Press). Harke, H. 1989: The Unkel Symposia: The beginnings of a debate in West German archaeology? Current Anthropology, 30:3. 406 410. 1995: The Hun is a methodical chap. In: Ucko, P. J. (ed.): Theory in archaeology: A world perspective. London and New York (Routledge). 4660. Harris, E. C. 1979b: The laws of archaeological stratigraphy. World Archaeology 11. 111117. Harvey, D. 1969: Explanation in geography. London, Cambridge

139

(Cambridge University Press). [Russian transl.: Kharvej, D. 1974: Nauchnoje objasnenije v geograi. Moscow (Progress)]. Hausmann, U. (ed.). 1969: Allgemeine Grundlagen der Archologie. Handbuch der Archologie. Bd. I. Mnchen (Beck). Hawkes, C. F. C. 1954: Archaeological theory and method: Some suggestions from the Old World. American Anthropologist 56. 155 168. 1957: Archaeology as science: purposes and pitfalls. Archaeological News Letter (London), vol. 6, no. 4. 93100. 1973: Innocence retrieval in archaeology. Antiquity 187. 176178. Hawkes, J. 1968: The proper study of mankind. Antiquity 168. 255 262. Heider, K. H. 1967: Archaeological assumptions and ethnographical facts: A cautionary tale from New Guinea. Southwestern Journal of Anthropology 23. 5264. Hempel, C. G. 1942: The function of General Laws in history. Journal of Philosophy 39. 3548. [repr. in Hempel, C. G. 1965: Aspects of scientic explanations and other essays in the philosophy of science. New York (Harper and Row). 231243)]. 1965: Aspects of scientic explanations and other essays in the philosophy of science. New York (Harper and Row). 1966: Philosophy of natural science. Englewood, Cliffs, N.J. (Prentice Hall). Hensel, W. 1971: Archeologia a prahistoria. Wrocaw (Ossolineum). Hensel, W., Donato, G., Tabaczynski, S. (eds.) 1986: Teoria i praktyka badannarcheologicznych. I. Wrocaw et al. (wydawnictwo PAN). Herskovitz, M. J. 1949: Man and his works. New York (Alfred A. Knopf). Hertz, G. 1980: [Russ. transl. Gerc, G. 1980: K voprosu o edinstve ziki i edinykh zicheskikh teorijakh. In: Akchurin, I. A. (ed.): Fizicheskaja teorija. Moscow (Nauka). 179209. Hewitt, K. [Khyuitt, K.] 1987. Otkrytye sistemy. In: Real nost i prognozy iskusstvennogo intellekta. Moscow (Mir). 85102. Hill, J. N. 1968: Broken K Pueblo: patterns of form and function. In: Binford, S. R. and L. R. (eds.) 1968: New perspectives in archaeology. Chicago (Aldine). 103143. 1970: Broken K Pueblo: Prehistoric social organization in the American Southwest. Anthropological Papers no. 18. Tucson (University of Arizona Press). 1971: Seminar on the explanation of prehistoric organizational change. Current Anthropology 12:3. 406408. 1972: The methodological debate in contemporary archaeology: a model. In: Clarke, D. L. (ed.). Models in archaeology. London (Methuen). 61107. Hill, J. N., Evans, R. K. 1972: A model for classication and typology. In: Clarke, D.L. (ed.): Models in archaeology. London (Methuen). 231273. Hinton, J. E. [Khinton, Dzh. E.] 1987: Obuchenie v parallel nykh setyakh. In: Real nost i prognozy iskusstvennogo intellekta. Moscow (Mir). 124136. Houbowicz, W. 1961: Jak publikowZroda archeologiczne. Warszawa, Wrocaw (Panstw. wyd-wo naukowe). Hoare, R. C. 1812/1975: The Ancient history of South Wiltshire. Wakeeld (EP Publ. in collab. with Wiltshire County Library). Hockett, H. C. 1955: The critical method in historical research and writing. New York (MacMillan).

140

Acta Archaeologica
dologicheskie i istoriogracheskie voprosy istoricheskoj nauki. Tomsk. 12 17. 1973a: Gnoseologicheskie osnovy istochnikovedenija. Filosofskie nauki 3. 2533. 1973b: Istoricheskij istochnik i istoricheskoe poznanie (metodologicheskie aspekty). Tomsk. Jacob-Friesen, K.-H.1928: Grundlagen der Urgeschichtsforschung. Stand und Kritik der Forschung ber Rassen, Vlker und Kulturen in urgeschichtlicher Zeit. Hannover (Helwing). Jaguttis-Emden, M. 1977: Zur Przision archologischer Datierungen. Archaeologia Venatoria 4. Tbingen. Janshin, A. L. 1963: Princip aktualizma i problema evoljucii geologicheskikh processov. Puti i metody poznanija zakonomernostej razvitija zemli. Tezisy dokladov na obedinennoj teoreticheskoj konferencii losofskikh (metodologicheskikh) seminarov nauchnykh uchrezhdenij AN SSSR. Moscow. 12. Johansen, A. B. 1974: Forholdet mellom teori og data i arkeologi og andre erfaringsvitenskaper. Er teoristabilitet et problem i empirisk forskning. Arkeologiske skrifter fra Historisk museum Universitet i Bergen. Bergen. 1982: Arkeologiens teori og data. Fornvnnen 77. 212225. 1984: Aksiomer i arkeologi. Det Kgl. Norske videnskabens Selskab, Museet. Rapport. Arkeologisk serie, no. 1. Trondheim: 2137. Kaiser, E. 1931: Der Grundsatz des Aktualismus in der Geologie. Zeitschrift der Deutschen Geologischen Gesellschaft 83:6. 389407. Kameneckij, I. S., Marshak, B. I., Sher, Ja. A. 1975: Analiz arkheologicheskikh istochnikov. (Vozmozhnosti formalizovannogo podkhoda). Moscow (Nauka). Kaplan, A. 1964: The conduct of inquiry. San Francisco (Chandler). Kaplan, D., Manners, R. A. 1972: Culture theory. Englewood Cliffs, New Jersey (Prentice Hall). Kedrov, B. M. 1966: O dialektike nauchnykh otkrytij. Voprosy losoi 12. 2637. Kelle, Zh., Kovalzon, M. Ja. 1966: K voprosu o klassikacii obshchestvennykh nauk. Metodologicheskie voprosy obshchestvennykh nauk. Moscow (Izd. Moskovskogo universiteta). 4057. Kelley, J., Hanen, M. 1988: Archaeology and the methodology of science. Albuquerque (University of New Mexico Press). Kiparisov, F. V. 1933: Veshch istoricheskij istochnik. Izvestija GAIMK 100. 322. Kiselev, S. V. 1955: O nedostatkakh i novykh zadachakh v izuchenii bronzovogo veka. Protiv vulgarizacii marksizma v arkheologii. Moscow (Izd. Akademii nauk SSSR). 129140. Klassikacija 1990: Klassikacija v arkheologii. Terminologicheskij slovarspravochnik. Moscow (Institut arkheologii AN SSSR). Kleindienst, M. R., Watson, P. J. 1956: Action archeology: The Archeological inventory of a living community. Anthropology tomorrow 5. 7578. Klejn, L. S. 1955: Voprosy proiskhozhdenija slavjan v sbornike dokladov VI nauchnoj konferencii Instituta arkheologii AN USSR. Sovetskaja Arkheologija XXII. 257272. 1967: Glazami Sherloka Kholmsa. Znanie sila 12. 1517. 1968: Proiskhozhdenie Doneckoj katakombnoj kultury. Avtoreferat diss. Leningrad (Leningradskoe otdelenie Instituta arkheologii AN SSSR). 1969: K postanovke voprosa o proiskhozhdenii slavjan. Problemy

Hodder, I. 1982: Theoretical archaeology: a reactionary view. In: Hodder, I. (ed.): Symbolic and structural archaeology. Cambridge (Cambridge University Press). 116. 1986/1991: Reading the past. Current approaches to interpretation in archaeology. Cambridge (Cambridge University Press) [2nd ed. 1991]. (ed.). 1987: The archaeology of contextual meanings. Cambridge (Cambridge University Press). 1991a: Archaeological theory in contemporary European societies: the emergemce of competing traditions. In: Hodder, I. (ed.): Archaeological theory in Europe: the last 3 decades. London and New York (Routledge). 124. 1991b: Preface. In: Hodder, I. (ed.): Archaeological theory in Europe: the last 3 decades. London and New York (Routledge). VII-XI. (ed.) 1991: Archaeological theory in Europe: the last three decades. London and New York (Routledge). 1992: Towards a coherent archaeology. In: Hodder, I. (ed.): Theory and practice in archaeology. London (Routledge). 169180. Hofer, T. 1968: Comparative notes on the professional personality of two disciplines. Current Anthropology 9. 311315. Hole, F. 1973: Question of theory in the explanation of culture change in prehistory. In: Renfrew, C. (ed.): The explanation of culture change. Models in prehistory. London (Duckworth). 1934. Hole, F., Heizer, R. F. 1969: An introduction to prehistoric archeology. 2nd ed. New York (Holt, Reinhart and Winston). Holtorf, C., Quensel, L. 1992: ber archologische Theorie und theoretische Archologie. Bericht von der Theoretical Archaeology Group 92, Leicester, Gro-Britanien. Archologische Informationen 15:12. 7079. Hrouda, B. 1978: Einleitung und Beweggrunde. In: Hrouda, B. (ed.): Methoden der Archologie. Mnchen (Beck). 1317. Hudson, J. C. 1972: Geographical diffusion theory. Evanston, Ill. (North Western University Dept. of Geography). Huizinga, J. 1954: Geschichte und Kultur. Gesammelte Aufstze. Stuttgart (Krner). Hymes, D. 1970: Linguistic models in archaeology. In: Gardin, J. C. (ed.): Archeologie et calculateurs. Paris (Editions du Centre Na tional de la Recherche Scientique). 91120. Ingersoll, Jr. D. W., Jellen, J. E., MacDonald, W. (eds.) 1977: Experimental archaeology. New York (Columbia University Press). Iribadzhakov, N. 1972: Klio pered sudom burzhuaznoj losoi. Moscow (Progress). Istochnikovedenie 1969: Teoreticheskie i metodicheskie problemy. Moscow (Nauka). Istorija 1964: Istorija i sociologija. Moscow (Nauka). Istorizm 1976: Istorizm arkheologii: metodologicheskie problemy. Konferencija. Tezisy dokladov (Otdelenie istorii AN SSSR). Moscow (Nauka). Isaac, G. L. 1979: The philosophy of archaeology. In: Clarke, D.L: Analytical archaeologist. Collected papers. London New York San Francisko (Academic Press). 1520. Itina, M. A. 1954: K voprosu ob otrazhenii obshchestvennogo stroja v pogrebalnom obrjade pervobytnych narodov. Sovetskaja Etnograja 1954:3. 6368. Ivanov, G. M. 1962: Svoeobrazie processa otrazhenija dejstvitelnosti v istoricheskoj nauke. Voprosy istorii 12. 7387. 1963: K voprosu o svoeobrazii istoricheskogo poznanija. Meto-

Metaarchaeology
otechestvennoj i vseobshchej istorii. Leningrad (Izd. Leningradskogo universiteta). 2135. 1970: Problema opredelenija arkheologicheskoj kultury. Sovetskaja Arkheologija 1970:2. 3751. 1972a: On the logic of cross-cultural analysis: comment on Kbben. Current Anthropology 13:1. 131139. 1972b: O prilozhimosti idej kibernetiki k postroeniju obshchej teorii arkheologii. In: Tezisy dokladov na Sekcijach, posv. itogam polevykh issledovanij 1971 g. Moscow. 1416. 1972c: [Rev. of Coles, J. N. and Simpson, D. D. A. (ed.) 1968: Studies in Ancient Europe. Essays presented to Stuart Piggott. Leicester.] Sovetskaja Arkheologija 1972:3. 385389. 1973a: Arkheologicheskie priznaki migracij. IX Mezhdunarodnyj kongress antropologicheskikh i etnogracheskikh nauk v Chikago, 1973. Doklady sovetskoj delegacii. Moscow (Nauka). 1973b: Marxism, the Systemic approach and archaeology. In: Renfrew, C. (ed.): The explanation of culture change: models in prehistory. London (Duckworth). 691710. 1973c: Metaarkheologija na Zapade. Tezisy dokladov na sessii, posvjashchennoj itogam polevykh arkheologicheskikh issledovanij 1972 g. v SSSR. Tashkent (Fan). 4952. 1974a: Kossinna im Abstand von vierzig Jahren. Jahresschrift fur mitteldeutsche Vorgeschichte (Halle) 58. 755. 1974b: [Rev. of: Hachmann, R. 1970: Goten und Skandinavien. Berlin.] Sovetskaja Arkheologija 19743. 278284. 1975a: K razrabotke procedury arkheologicheskogo issledovanija. Predmet i obekt arkheologii i voprosy metodiki arkheologicheskich issledovanij. Leningrad. 4244. 1975b: Problema smeny kultur v sovremennych archeologicheskich teorijach. Vestnik Leningradskogo Universiteta 8. 95103. 1976: Arkheologija i preistorija v sisteme I. Rauza. Sovetskaja Arkheologija 1976:1. 306315. 1977a: A panorama of theoretical archaeology. Current Anthropology 18:1. 142; 18:2. 371373. 1977b: K ocenke empirizma v sovremennoj arkheologii. Problemy arkheologii i etnograi. Leningrad (Izd. Leningradskogo universiteta). 1320. 1977c: Predmet arkheologii. Arkheologija Juzhnoj Sibiri (Izvestija kafedry arkheologii Kemerovskogo universiteta 9. Kemerovo (Izd. Kemerovskogo universiteta). 314. 1978a: Arkheologicheskaja teorija (problema statusa i denicii). In: Problemy arkheologii 2. Leningrad (Izd. Leningradskogo Universiteta). 817. 1978b: Arkheologicheskie istochniki. Leningrad (Izd. Leningradskogo universiteta) [2nd enlarged ed.: Saint Peterburg (Farn) 1995]. 1978c: Tri procedury arkheologicheskogo issledovanija. Voprosy arkheologii Urala 14. 1524. 1979a: Ponjatie tipa v sovremennoj arkheologii. In: Klejn, L. S. (ed.): Tipy v kulture. Leningrad (Izd. Leningradskogo Universiteta). 5074. 1979b: Teorii v arkheologii. In: Novoe v archeologii Sibiri i Dalnego Vostoka. Novosibirsk (Nauka). 3039. 19801981. Panorama teoretske archeologije. Archeo 1. 116; 2. 142. 1980a: Panorama de larcheologie theorique. In: Schnapp, A. (ed.): Larcheologie aujourdhui. Paris (Hachette). 262303.

141

1980b: Struktura arkheologicheskoj teorii. Voprosy losoi 1980:2. 99115. 1981a: Ethnogenese als Kulturgeschichte archaologisch be trachtet. Neue Einstellung. Beitrage zur Ur- und Frhgeschichte (Coblenz-Festschrift) Teil 1. Berlin (Deutscher Verlag der Wissenschaften). 1325. 1981b: O jazyke veshchej. Metodologicheskie aspekty archeologicheskikh i etnogracheskikh issledovanij v Zapadnoj Sibiri. Tomsk (Izd. Tomskogo universiteta). 1618. 1981c: Problema smeny kultur i teorija kommunikacii. Kolichestvennye metody v gumanitarnykh naukakh. Moscow (Izd. Moskovskogo universiteta). 1823. 1982a: Archaeological Typology. British Archaeological Reports, International Series 153. Oxford. 1982b: Vvedenie v specialnost. Programma speckursov po arkheologii i etnograi. Sverdlovsk. (Izd. Uralskogo universiteta). 37. 1986: O predmete arkheologii. Sovetskaja Arkheologija 1986:3. 209219. 1988: Strategija sinteza v issledovanijakh po etnogenezu. Sovetskaja Etnograja 1988:4. 1323. 1990: Theoretical archaeology in the making. A survey of books published in the West in 19741979. Fennoscandia Archaeologica VIII (Helsinki). 6771. 1991a: Arkheologicheskaja tipologija. Leningrad (Akademija nauk SSSR). 1991b: A Russian lesson for theoretical archaeology. Fennoscandia Archaeologica VIII (Helsinki). 6771. 1991c/1993c: Rassech kentavra. O sootnoshenii arkheologii s istoriej v sovetskoj tradicii. Voprosy istorii estestvoznanija i techniki (Moscow) 1991:4. 312. [Engl. transl.: To separate a centaur: On the relationship of archaeology and history in Soviet tradition. Antiquity 255. 339348.] 1991d: Jazycheskij podkhod k lingvistike. Sovetskoe Slavjanovedenie 4 (Moscow). 8892. 1992a: Intervju. Arheo (Ljubljana), 15: 5256 (recorded by prof. Biba Terzan). 1992b: Metodologicheskaja priroda arkheologii. Rossijskaja arkheologija 1992:4. 8696. 1993a: Fenomen sovetskoj arkheologii. Sankt-Peterburg (Farn). 1993b: La arqueologia sovetica. Historia y teoria de una escuela desconoci da. Barcelona (Critica). 1994: Prehistory and archaeology. In: Kuna, M. and Venclova, N. (eds.): Whither archaeology? Papers in honour of Evzen Neustupny. Praha (Institute of archaeology). 3642. 1995a: Arkheologicheskie istochniki. (2nd rev. ed.). Sankt-Peterburg (Farn). 1995b: Is German archaeology atheoretical? Norwegian Archaeological Review 26:1. 4954. 1995c: Paradigmy i periody v istorii otechestvennoj arkheologii (rec. na knigu G. S. Lebedeva). Sankt-Peterburg i otechestvennaja arkheologija. Istoriogracheskie ocherki. Saint Petersburg (Izd. SanktPeterburgskogo universiteta). 173183. 1995d: The functions of archaeological theory. Journal of Theoretical archaeology 3/4 (19931994). 139. 1997a: Das Phnomen der Sowjetischen Archologie. Berlin (Peter Lang).

142

Acta Archaeologica
Kossack, G. 1992: Prehistoric archaeology in Germany: its history and current situation. Norwegian Archaeological Review 25:2. 73 109. Kovalevskaja, V. B. 1970: K izucheniju ornamentiki nabornykh pojasov VIIX vv. kak znakovoj sistemy. In: Statistiko-kombinatornye metody v arkheologii. Moscow (Nauka). 144145. Kramer, C. (ed.). 1979: Ethnoarchaeology: implications of ethnography for archaeology. New York (Columbia University Press). Krause, R. A., Thorne, R. M. 1971: Toward a theory of archaeological things. Plains Anthropologist 16. 245257. Krieger, A. D. 1944: The typological concept. American Antiquity 9:3. 271288. Kristiansen, K. 1974: En kildekritisk analyse af depotfund fra Danmarks bronzealder (periode IV-V). Et bidrag til den arkologiske kildekritik. Aarbger for Nordisk Oldkyndighed og Historie. 119 160. 1978: The application of source-criticism to archaeology. Norwegian Archaeological Review 11:1. 15. (ed.) 1985: Archaeological formation processes: the representativity of archaeological remains of Danish prehistory. Copenhagen (Nationalmuseets Forlag). Kroeber, A. L. 1937: Archaeology. In: Encyclopedia of the social sciences, vol. 2. Kruglov, A. P., Podgaeckij, G. V. 1935: Rodovoe obshchestvo stepej Vostochnoj Evropy. Izvestija GAIMK 119. Moscow Leningrad. Kubler, G. 1961: Rival approaches to American antiquity. Three regions of primitive arts. New York (The Museum of Primitive art). Kudilova, E. A. 1972: Filosofskie problemy v sovremennoy psikhologii. Moscow (Nauka). Kuznecov, I. V. 1948: Princip sootvetstvija v sovremennoj zike i ego losofskoe znachenie. Moscow Leningrad (Gostekhizdat). Labuda, G. 1957: Proba nowej systematyki i nowej interpretacji zrode his torycznych. Studia zrodoznawcze I (Warszawa, Panstwowe Wyd awnictwo Naukowe). 352. Langlua, Sh. V., Senobos, Sh. [Langlois, Ch. et Segnobos, Ch.] 1899: Vvedenie v izuchenie istorii. Saint Petersburg (O. N. Popova). Lappo-Danilevskij, A. S. 19101913: Metodologija istorii vol. 111. Saint Petersburg. (studench. izd. komitet pri Istor.-Filolog. fakultete SPb. universiteta). Leach, E. 1972: Concluding address. In: Renfrew, C. (ed.): The explanation of culture change: Models in prehistory. London (Duckworth). 761771. Lebedev, G. S. 1973: Otrazhenie socialnoj struktury v arkheologicheskikh materialakh. In: Filosoja, istorija, sovremennost. Leningrad (Izd. Leningradskogo universiteta). 5563. 1992: Istorija otechestvennoj arkheologii. Saint Petersburg (Izd. Sankt-Peterburgskogo universiteta). LeBlanc, S. A. 1973: Two points of logic concerning data, hypotheses, general laws and systems. In: Redman, Ch. L. (ed.): Research and theory in current archeology. New York (Wiley Interscience). 199214. Ledieu, A. 1885: Boucher de Perthes, sa vie, ses ouvres, sa correspondence. Abbeville (Caudron). Lenin, V. I. 1908/1961: Materializm i empiriokriticizm. Poln. sobr. soch., izd. 5, t. 18. Moscow (Gospolitizdat).

1997b: Kulturno-istoricheskij process i teorija kommunikacii. Grani kultury. Tezisy dokladov i vystuplenij. Sankt-Peterburg. 107 110. 1998: Zmej s khobotom (Folklor i etnograja v rabote arkheologa). Pamjatniki stariny: otkrytija, koncepcii, versii. St. Peterburg Pskov (Pskovskij muzej IIMK RAN) vol. I. 289296. 1999a: Migracija: arkheologicheskie priznaki. Stratum plus 1 (Sankt-Peterburg Kishinev Odessa). 5271. 1999b: Principy arkheologii. St. Petersburg. (Izd. St. Peterburgskogo Universiteta). Klejn, L. S., Minjaev, S. S., Piotrovskij, Ju. Ju., Khejfec, O. I. 1970: Diskussija ob arkheologicheskoj kulture v Problemnom seminare Leningradskogo universiteta. Sovetskaja Arkheologija 1970:2. 298302. Kljuchevskij, V. O. 1968: Pisma. Dnevniki. Aforizmy i mysli ob istorii. Moscow (Nauka). Kluckhohn, C. K. M. 1940: The conceptual structure in Middle American studies. In: Hay, C. L., Tozzer A. M. et al. (eds.). The Maya and their neighbors. New York and London (D. Appleton Century Co.). 4151. 1960: The use of typology in anthropological theory. In: Wallace, A. F. C. (ed.): Selected papers of the Fifth International Congress of Anthropological and Ethnological Sciences. Philadelphia, 1956. Philadelphia (University of Pennsylvania Press). 141151. Kosciakowski, S. 1954: Historyka: W step do studiow historycznych. Londyn (Polski Uniwersytet na Obezyznie). Kobylinski, Z. 1981: Badania etnoarcheologiczne a nomotetyzacja archeologii. Archeologia Polski 26:1. 747. Kbben, A. J. F. 1967: Why exceptions? The logic of cross-cultural analysis. Current Anthropology 8:12. 334. Kolchin, B. A., Marshak, B. I., Sher, Ja. A. 1970: Arkheologija i matematika. In: Kolchin, B.A. and Sher, Ja. A. (eds.): Statistikokombinatornye metody v arkheologii. Moscow (Nauka). 37. Kolpakov, E. M. 1992: Teorija arkheologicheskoj klassikacii. SanktPeterburg. 1993: The end of theoretical archaeology? A glance from the East. Norwegian Archaeological Review 26:2. 107115. Kolpakov, E. M., Vishnjatsky, L. B. 1990: Current theoretical discussion in Soviet archaeology: An essay. Fennoscandia Archaeologica 7. 1725. 1993: Sovremennaja sovetskaja teoreticheskaja arkheologija. Arkheologicheskie vesti 1 (Saratov). 320. Kon, I. S. 1969: K sporam o logike istoricheskogo objasnenija (skhema Poppera Gempelja i ee kritiki). Filosofskie problemy istoricheskoj nauki. Moscow (Nauka). 263295. Kont, O. [Comte, A.] 1900: Kurs polozhitelnoj losoi vol. II. Saint Petersburg. Kopnin, P. V. 1966: Vvedenie v marksistskuju gnoseologiju. Kiev (Naukova Dumka). Kopnin, P. V., Popovich, M. V. 1968: Problemy dialekticheskoj logiki v ikh svjazi s estestvoznaniem. In: Omeljanovskij, M. E. (ed.): Materialisticheskaja dialektika i metody estestvennykh nauk. Moscow (Nauka). 70104. Kosolapov, V. V. 1964: Gnoseologichna priroda naukovogo faktu. Kiev (Naukova Dumka). 1965: Fakt kak osnovanie nauchnogo znanija. In: Logika nauchnogo issledovanija. Moscow (Nauka). 4567.

Metaarchaeology
Leone, M. P. (ed.).1972: Contemporary archaeology. A guide to theory and contributions. Carbondale and Edwardsville (Southern Illinois University Press). Leroi-Gourhan, A. 1961: Lhistoire sans textes: ethnologie et prehi stoire. In: Samaran, Ch. (ed.): Lhistoire et ses methodes. Paris (Gal limard). 217249. 1964: Les religions de la prehistoire. Paris (Presses Universitaires de France). Levin, M. 1973: On explanation in archaeology: A rebuttal to Fritz and Plog. American Antiquity 38. 387395. Likhachev, D. S. 1962: Tekstologija. Moscow Leningrad (Izd. AN SSSR). Ljapushkin, I. I. 1968: Slavjane Vostochnoj Evropy nakanune obrazovanija Drevnerusskogo gosudarstva. Materialy i issledovanija po arkheologii SSSR 52. Leningrad, Nauka. Loone, E. N. 1973: Ob urovnjakh istoricheskogo znanija i poznanija. Uchenue zapiski Tartusskogo universiteta 301 (Trudy po losoi 16) 1980: Sovremennaja losoja istorii. Tallin (Eesti raamat). Lowie, R. H. 1912: The principle of convergence in ethnology. Journal of American Folk-Lore 25 (45). 2442. Lowther, G. R. 1962: Epistemology and archaeological theory. Current Anthropology 3. 495503, discussion 503509. Lysenko, T. D. 1948: O polozhenii v biologicheskoj nauke. Stenogracheskij otchet sessii VASKhNIL. Moscow (Archive of the VASChNIL). MacNeish, R. S. 1976: The science of archaeology? Hamilton, Ontario, D. G. (Seldon Printing Limited). 1978: The science of archaeology? North Scituate, Mass. (Duxbury Press). MacWhite, E. 1956: On the interpretation of archaeological evidence in historical and sociological terms. American Anthropologist 58. 325. Malina, J. 1975: Archeologie: jak a proc? Mikulovo Breclav (Re gionalni Muzeum OV Socialisticke akademie). 1980: Arcgeologie vcera a dnes, aneb Maji archeologovove ede hmoty vice s za nehty nez za usima? 12. Ceske Budejovice, Jihoceske Mu zeum. Malina, J., Vasicek, Z. 1990: Archaeology yesterday and today. The devel opment of archaeology in the sciences and humanities. Cambridge (Cambridge University Press). Malmer, M. P. 1962: Jungneolithische Studien. Acta Archaeologia Lundensia, series in 8o, 2. Bonn Habelt Lund (Gleerup). 1963: Metodproblem inom jrnalderns konsthistoria. Acta Archaeologica Lundensia, Series in 8, 3. Bonn Lund (Gleerup). 1997: On objectivism and actualism in archaeology. Current Swedish Archaeology 5. 78. Mann, J. 1981: Dons and detection. In: Evans, J., Cunliffe, B. and Renfrew, C. (eds.): Antiquity and memory. London (Thames and Hudson). 203207. Markarjan, E. S. 1969: Ocherki teorii kultury. Erevan (Izd. AN Arm. SSR). Markovin, V. I. 1960: Kultura plemen Severnogo Kavkaza v epokhu bronzy. Materialy i issledovanija po arkheologii SSSR 93. Moscow (Izd. AN SSSR). Marks, K. [Marx, K.] 185758/1958: Iz rukopisnogo nasledstva K. Marksa. Vvednie. (Iz ekonomicheskikh rukopisej 18571858

143

godov). In: Marks, K. and Engels, F.: Sochinenija. Moscow (Gospolitizdat), vol. 12, 709738. (1968 vol. 46:1. 1748). 1974: Ekonomicheski-losofskie rukopisi 1844 goda. Kommunizm. In: Marks, K. and Engels, F.: Sochinenija. [2nd ed.], Moscow (Gospolitizdat), vol. 42. 113127. 1868/1964: Pismo Ljudvigu Kegelmanu 11 ijulja 1868. In: Marks, K., Engels, F.: Sochinenija. Moscow (Gospolitizdat), vol. 33. 175. Marks, K., Engels, F. [Marx, K. and Engels, F.] 1955: Nemeckaja ideologija. In: Marks, K. and Engels, F.: Sochinenija. Moscow (Gospolitizdat), vol. 3. 7544. Martin, P. S. 1968: Revolution in archaeology. American Antiquity, 36:1. 18. Martynov, A. I., Sher, A. Ja. 1989: Metody arkheologicheskogo issledovanija. Moscow (Vysshaja shkola). Masson, V. M. 1964: Srednjaja Azija i Drevnij Vostok. Moscow - Leningrad (Nauka). 1966: Ot vozniknovenija zemledelija do slozhenija ranneklassovogo obshchestva. In: Doklady i soobshchenija arkheologov SSSR. VII Mezhdunarodnyj kongress preistorikov i protoistorikov. Moscow. 150 166. 1973: Arkheologija za bortom gumanitarii. In: Tezisy dokladov sessii, posvjashchennoj itogam polevykh arkheologicheskikh issledovanij v SSSR. Tashkent (FAN). 4649. 1976: Ekonomika i socialnyj stroj drevnikh obshchestv (v svete dannykh arkheologii). Leningrad (Nauka). 1978: Sistemnyj podkhod i issledovanie paleoekonomicheskikh struktur. Kratkie soobshchenija Instituta arkheologii AN SSSR (Moscow) 152. 3036. Matjukhin, A. E. 1976: Ob istochnikakh, roli i tipakh objasnenija v pervobytnoj arkheologii. In: Predmet i objekt arkheologii i voprosy metodiki arkheologicheskikh issledovanij. Leningrad (AN SSSR, LOIA). 1823. McKern, S. S., McKern, Th. W. 1974: Living prehistory: an introduction to physical anthropology and archaeology. Menlo Park, California (Cummings). Megrelidze, K. R. 1935: N. Ja. Marr i losoja marksizma. Problemy istorii doklassovylh obshchestv 1935:34. 7089. Mellor, D. H. 1982: Probabilities for explanation. In: Renfrew, C. et al. (eds.): Theory and explanation in archaeology. New York et al. (Academic Press). 5763. Melnikov, G. P. 1967: Azbuka matematicheskoj logiki. Moscow (Znanie). Merton, R. K. 1967: On theoretical sociology. New York (Free Press) London (Collier-Macmillan). Merzon, L. S. 1968: K voprosu o gnoseologicheskoj prirode faktov nauki. Uchenye zapiski Leningradskogo Pedagogicheskogo instituta im. Gercena, vol. 365. 325. Meyen, S. V. 1978: O naibolee obshchikh principakh rekonstrukcii v geologii. Izvestija AN SSSR, ser. geol. nauk 11. 7991. Meyer, E. 1902: Zur Theorie und Methodik der Geschichte: Geschichtsphilosophische Untersuchungen. Halle a. S. [Russ. transl.: Mejer, E. 1904: Teoreticheskie i metodologicheskie voprosy istorii. Filosofsko-istoricheskie issledovanija. Moscow (tip. t-va I. D. Sytina)]. Mill, Dzh. St. [Mill, J. St.] 1914: Sistema logiki sillogicheskoj i induktivnoj. [2nd ed.]. Moscow (G. A. Leman).

144

Acta Archaeologica
Neustypny, E. 1993: Archaeological method. Cambridge (Cambridge University Press). Niemeyer, H. G. 1968: Einfhrung in die Archologie. Darmstadt (Wissenschaftliche Buchgesellschaft). Nikitin, E. P. 1966: Metod poznanija proshlogo. Voprosy losoi 8. 3444. 1970: Objasnenie, funkcija nauki. Moscow (Nauka). Nikonov, V. A. 1965: Vvedenie v toponimiku. Moscow (Nauka). Njuton, I. 1954 [Newton, I. 1730]: Optika. [2nd ed.]. Moscow (Gostekhizdat). Obermajer, G. [Obermeier, H.] 1913: Doistoricheskij chelovek. S.Peterburg (Brokgaus-Efron). Olivier, L., Coudart, A. 1995: French tradition and the central place of history in the human sciences. In: Ucko, P. J. (ed.): Theory in archaeology: A world perspective. London and New York (Routledge). 363381. Olsen, B. 1997: Fra ting til tekst. Oslo (Universitet). Onoprienko, V. I. 1976: Logika procedury retroskazanija v geologii. In: Metodologija geologicheskikh issledovanij. Vladivostok (Dalnevostochnyj nauchnyj centr AN SSSR). 154167. Orton, C. 1980: Mathematics in archaeology. London (Collins). Ovchinnikov, N. F. 1968: Metodicheskaja funkcija losoi v estestvoznanii. In: Omeljanovskij M. E. (ed.): Materialisticheskaja dialektika i metody estestvennykh nauk. Moscow (Nauka). 942. Paubicka, A. 1973: W sprawie apriorzymu i empirzymu w archeologii i etnologii. Fontes Archaeologici Poznaniensis 22. 252 255. Parker, J. S. 1870: The Ashmolean Museum: Its history, present state and prospects. Oxford (Oxford University Press). Patrik, L. E. 1985: Is there an archaeological record? Advances in archaeological method and theory 8. 2762. Paul, E. 1962: Die falsche Gttin. Geschichte der Antikenflschung. Leipzig (Koeler & Amelang) Heidelberg, (L. Schreider). Pebbles, C. S., Kus, S. M. 1977: Some archaeological correlates of ranked societies. American Antiquity 42:3. 421448. Pechenkin, A. A. 1972: Funkcii nauchnoj teorii. Filosoja. Metodologija. Nauka. Moscow (Nauka). 202218. 1973: Problema simmetrii predskazanija i objasnenija. Voprosy Filosoi 1973:2. 167174. Petrie, W. M. F. 1877: Inductive methodology. London (Hargrave Saunders). 1904: Methods and aims in archaeology. London (Macmillan). Piggott, S. 1958: The Dawn: and an Epilogue. Antiquity 126. 75 79. 1959: Approach to archaeology. London (Adam and Charles Black) [new ed. 1965: New York (MacGraw Hill)]. Plekhanov, G. V. 1957: O knige Rikkerta. In: Plekhanov, G. V: Izbrannye losofskie proizvedenija. Moscow (Gospolitizdat), vol. 3. 508515. Plog, F. T. 1973: Laws, systems of law, and the explanation of observed variation. In: Renfrew, C. (ed.): The explanation of culture change: models in prehistory. London (Duckworth). 649661. 1974: The study of prehistoric change. New York and London (Academic Press). Popov, P. V. (ed.) 1972: Metodologicheskie osnovy nauchnogo poznanija. Moscow (Vysshaja shkola).

Mironov, B. N., Stepanov, Z. V. 1975: Istorik i matematika. Leningrad (Nauka). Moberg, C.-A. 1969: Introduktion til arkeologi. Stockholm (Natur och Kultur). 1976: Introduction a larcheologie. Paris (Francois Maspero). ` 1981: Similar nds? Similar interpretations? A spectrum of approaches. In: Moberg C.-A. (ed.): Similar nds? Similar interpretations? Gteborg (University of Gothenburg). A1A18. Moles, A., Valancien, B. (eds.) 1963: Communications et language. Paris (Gauthier Villars). Mongajt, A. S. 1949: Amerikanskaja arkheologija na sluzhbe imperializma. Voprosy istorii 5. 97104. 1955: Arkheologija v SSSR. Moscow (AN SSSR). 1967: Arkheologicheskie kultury i etnicheskie obshchnosti. Narody Azii i Afriki 1. 5365. Montelius, O. 1884. Den forhistoriske fornforskarens metod och material. Antiqvarisk Tidskrift for Sverige 8:3. 128. 1888: The civilization of Sweden in heathen times. London (Macmillan). 1900: Typologien eller utvecklingslaran tillampad p det mnsksliga arbetet. Svenska Fornminnesfreningens Tidskrift (Stockholm) 10:3. 237268. 1903: Die Alteren Kulturperioden im Orient und in Europa, vol. I. Stockholm. Morgan, C. G. 1973: Archaeology and explanation. World Archaeology 4. 259276. Morgan, L. H. 1877/1934: Ancient Society. New York (Henry Holt). [Russ. trans.: Morgan, L. G. 1934: Drevnee obshchestvo. Leningrad (Institu narodov severa).] Mortillet, G. de. 1883: Le prehistorique, Antiquite de lhomme. Paris. (4th ed. 1910: La Prehistoire). Morwood, M. J. 1975: Analogy and the acceptance of theory in archaeology. American Antiquity 40. 111116. Mueller, J. W. (ed.) 1975: Sampling in archaeology. Tucson (University of Arizona Press). Munro, R. 1905: Archaeology and false antiquities. London (Methuen, Antiquary Books). Murray, M. 1961: First steps in archaeology. Antiquity 137. 813. Myhre, B. 1991: Theory in Scandinavian archaeology since 1961: a view from Norway. In: Hodder, I. (ed.): Archaeological theory in Europe. The last three decades. London and New York (Routledge). 161186. Mller, S. 1884: Mindre bidrag til den forhistoriske Archaeologis methode. Aarbger for nordisk Oldkyndighed og Historie. Kbenhavn. 18971898: Nordische Altertumskunde. Bd. III. Straburg (Karl Trbner). Mller-Karpe, H. 19661980: Handbuch der Vorgeschichte. Bd. IIV. Mnchen (Beck). 1975. Einfhrung in die Vorgeschichte. Mnchen (Beck). Myrdal, K. G. 1958: Value in social theory. A selection of essays on methodology. London (Routledge and Kegan Paul). Nagel, E. 1961: The structure of science. New York (Harcourt, Brace and World). Nalimov, N. N. 1993: V poiskakh inykh smyslov. Moscow (Progress). Narr, K. J. 1957: Abriss der Vorgeschichte. Mnchen (Oldenburg). Narskij, I. S. 1961: Sovremennyj pozitivizm. Moscow (AN SSSR).

Metaarchaeology
Popovich, M., Sadovskij, V. 1970: Teorija. Filosofskaja enciklopedija vol. 5. Moscow (Sovetskaja Enciklopedija). 205207. Popper, K. 1935: Logik der Forschung. Wien (Springer). 1957: The poverty of historicism. London (Routledge and Kegan Paul). 1963: Conjectures and refutations. 2. New York (Harper and Row). 1965: The logic of scientic discovery. New York (Harper torchbooks). Pronshtejn, A. P. 1971: Metodika istoricheskogo issledovanija. Rostov (Izd. Rostovskogo universiteta). Propp, V. Ja. 1962: Ob istorizme russkogo eposa. Russkaja literatura 2. 8791. Pushkarev, L. N. 1970: Istoricheskij istochnik v svete leninskoj teorii otrazhenija. Aktualnye problemy istorii Rossii epochi feodalizma. Moscow (Nauka). 6484. 1975: Klassikacija russkikh pismennykh istochnikov po otechestvennoj istorii. Moscow (Nauka). Pyddoke, K. 1964: What is archeology? New York (Roy Publ.) London (J. Baker). Radililovskij, V. V. 1985: K voprosu o postroenii nauchnoj teorii v arkheologii. Problemy rekonstrukcii v arkheologii. Novosibirsk (Nauka). 1623. Rakitov, A. I. 1964: Statisticheskaja interpretacija facta i rol statisticheskogo metoda v postroenii empiricheskogo znanija. Problemy logiki nauchnogo poznanija. Moscow (Nauka). 375407. 1969: K voprosu o strukture istoricheskogo issledovanija. Filosofskie problemy istoricheskoj nauki. Moscow (Nauka). 169178. 1971: Kurs lekcij po logike nauki. Moscow (Vysshaja Shkola). 1982: Istoricheskoe poznanie. Sistemno-gnoseologicheskij podchod. Moscow (Politizdat). Randsborg, K. 1982: Theoretical approaches to social change: an archaeological viewpoint. In: Renfrew, C. et al. (eds.): Theory and explanation in archaeology. New York (Academic Press). 423430. 1994: Ole Worm. An essay on the modernization of antiquity. Acta Archaeologica 65. 135169. 1995: Hjortspring: Warfare and sacrice in Early Europe. Aarhus (Aarhus University Press). 1997: On archaeology and history. Acta Archaeologica 68. 189 194. Ranke, L. von 1874: Smtliche Werke. Bd. 3334. Leipzig (Dunkur und Humbolt). Rathje, W. L. 1974: The Garbage project: A new way of looking at the problems of archaeology. Archaeology 27:4. 236241. Ravdonikas, V. I. 1930: Za marksistskuju istoriju materialnoj kultury. Izvestija GAIMK (Leningrad), vol. 7:34. 1939: Istorija pervobytnogo obshchestva. Ch. 1. Leningrad (Izd. Leningradskogo universiteta). Redman, C. L. 1973: Research and theory in current archaeology: an introduction. In: Redman, C. L. (ed.): Research and theory in current archaeology. New York (John Wiley & Sons). 520. Regirer, E. I. 1966: O professii issledovatelja v tochnykh naukakh. Moscow (Nauka). Reinach, S. 1911: La methode archeologoque. De la methode dans les sciences, 2me serie. Paris (Falcan). Renfrew, C. 1969a: [Review of Clarkes and Binfords books]. Antiquity 43. 241244.

145

1969b: Trade and culture process in European prehistory. Current Anthropology 10:23. 151169. (ed.) 1973: The explanation in culture change: models in prehistory. London (Duckworth). 1982: Explanation revisited. In: Renfrew, C. et al. (eds.): Theory and explanation in archaeology. New York (Academic Press). 523. Renfrew, C., Bahn, P. 1991: Archaeology: Theories, methods and practice. London (Thames and Hudson). Renfrew, C. et al. (eds.) 1982: Theory and explanation in archaeology. New York (Academic Press). Rickert, H. 1902: Die Grenzen der naturwissenschaftlichen Begriffsbildung: Eine logische Einleitung in die historischen Wissenschaften. Tbingen. [Russ. transl.: Rikkert, G. 1903: Granicy estestvennonauchnogo obrazovanija ponjatij. Logicheskoe vvedenie v istoricheskie nauki. Sankt Peterburg (tip. E. D. Kuskova)]. Roe, D. 1976: Typology and the trouble with hand-axes. In: Sieveking, G. et al. (eds.): Problems in economic and social archaeology. London (Duckworth). 6170. Rogachev, A. N. 1975: O predmete i metode pervobytnoj arkheologii. In: Predmet i obekt arkheologii i voprosy metodoki arkheologicheskikh issledovanij. Leningrad (Nauka). 1114. 1978: O predmete i metode pervobytnoj arkheologii. Kratkie soobshchenija Instituta arkheologii AN SSSR (Moscow) 52. 1723. Rohan-Csermak, G. de. 1971: Apriopisme et empirisme dans larcheologie et ethnologie. Actes du VIIe Congres International des ` Sciences Prehistoriques et Protohistoriques. Prague 1966, vol. 1. Prague. 97100. Rouse, I. B. 1939: Prehistory in Haiti: a study in method. Yale University Publications in Anthropology 21. New Haven, Con. 1968: Prehistory, typology, and the study of society. In: Chang, K. C. (ed.): Settlement archaeology. Palo Alto, California (National Press Books). 1030. 1972: Introduction to prehistory: a systematic approach. New York et al. (McGraw-Hill). Rowe, J. H. 1962: Worsaaes law and the use of grave lots for archaeological dating. American Antiquity 28. 129137. Rumpf, A. 1953: Archologie. Mnchen (Beck). Rush, P. 1994: The use of theory in archaeology. In: Mackenzie, I. M. (ed.): Archaeological theory: progress or posture? Adershot et al. (Avebury). 6676. Ruzavin, G. I. 1978: Nauchnaja teorija. Logiko-metodologicheskij analiz. Moscow (Mysl). Rybakov, B. A. 1946: Arkheologicheskie istochniki russkoj istorii. Kratkie soobshchenija IIMK (Moscow) 13. 181182. 1957: O korpuse arkheologicheskich istochnikov (tezisy doklada na plenume IIMK AN SSSR v marte 1957 g.). Moscow (AN SSSR). 1967: Arkheologija. In: Oktjabr i nauchnyj progress, vol. 2 Moscow (izd. APN). 575595. 1968: Sovetskaja arheologija za 50 let. Voprosy istorii 1968:1. 28 37. 1978: Istorizm arkheologii. Kratkie sobshchenija IIMK AN SSSR (Moscow) 152. 56. Sabloff, J. A., Binford, L. R., McAnany, P. A. 1987: Understanding the archaeological record. Antiquity 232. 203209. 1987. Sabloff, J. A., Willey, G.R. 1967: The collapse of Maya civilization in the Southern Lowlands: a consideration of history and process. Southwestern Journal of Anthropology 23. 311336.

146

Acta Archaeologica
1989: Archaeology into 1990s. Norwegian Archaeological Review22. 154. Shatskij, N. S. et al. 1951: K voprosu o periodizacii osadkoobrazovanija i o metode aktualizma v geologii. Moscow (Nauka) Shatskij, N. S. 1965: Izbrannye trudy, vol. 4. Moscow (Nauka). 85 101. Shennan, S. 1988: Quantifying archaeology. Edinburgh (Edinburgh University Press). Sher, Ja. A. 1966: Tipologicheskij metod v arkheologii i statistika. In: 7-j Mezhdunarodnyj knogress doistorikov i protoistorikov. Doklady i soobshchenija arkheologov SSSR. Moscow. 253266. 1970: Intuicija i logika v arkheologicheskom issledovanii (k formalizacii tipologicheskogo metoda arkheologii). In: Kolchin, B. A. and Sher, Ja. A. (eds.) Statistiko-kombinatornye metody v arkheologii. Moscow (Nauka). 824. 1972: Nekotoryje voprosy teoreticheskoj arkheologii. Tezisy dokladov na sekcijakh, posvjashchennykh itogam polevykh issledovanij 1971 g. Moscow. 401405. 1973: K kharakteristike struktury obshchej arkheologii. Tezisy dokladov, posvjashchennykh itogam polevykh arkheologicheskikh issledovanij 1972 goda v SSSR. Tashkent (FAN). 5356. 1976: Metodologicheskie voprosy arkheologii. Voprosy losoi 10. 6779. 1983: Vstupitelnaja statja. In: Garden, Zh.-K. [Gardin J.-C.] (ed.): Teoreticheskaja arkheologija. Moscow (Progress). 528. 1985. K kharakteristike ponjatija arkheologicheskij fakt. In: Vasilevskij, R. S. and Kholjushkin, Ju. P. (eds.): Problemy rekonstrukcij v arkheologii. Novosibirsk (Nauka). 516. Shtoff, V. A. 1960: Vvedenije v metodologiju nauchnogo poznanija. Leningrad (izd. Leningradskogo universiteta). Simon, R. 1678: La histoire critique du Testament Vieux. Paris. Simpson, G. G. 1980: Splendid isolation: The curious history of South American mammals. New Haven and London (Yale University Press). Smith, B. D. 1982: Explanation in archaeology. In: Renfrew, C. et al. (eds.): Theory and explanation in archaeology. New York (Academic Pres). 7382. Smith, J. W. 1976: Foundations of archaeology. Beverly Hills (Glencoe Press). Smith, M. A. 1955: The limitations of inference in archaeology. The Archaeologucal Newsletter 6. 17. Smolla, G. 1964: Analogien und Polaritten. In: Uslar, R. v. and Narr, K. J. (eds.): Studien aus Alteuropa, Bd.I. Beiheft d. Bonner Jahrbcher 10:1. Kln Graz. 3035. Sokolov, E. V. 1972: Kultura i lichnost. Leningrad (Nauka). Sommerwill, Dzh. 1960: Chto otlichaet nauku ot drugikh form znanija. In: Sommerwill, Dzh.: Izbrannoe. Moscow (Progress). 155190. South, S. 1977: Method and theory in historical archaeology. New York San Francisco London (Academic Press). Sov. arkh. lit. 19591989: Sovetskaja archeologicheskaja literatura. Bibliograja. Moscow Leningrad (since 1969 Leningrad) (Nauka). Spaulding, A. C. 1953: Statistical techniques for the discovery of artifact types. American Anthropologist 18:4. 305313. 1954: Reply to Ford. American Anthropologist 56. 112114. 1960: The dimentions of archaeology. In: Dole, G. E. and Car-

Sadovskij, V. N. 1965: Metodologicheskie problemy issledovanija obektov, predstavljajushchikh soboj sistemy. Sociologija v SSSR, vol. I. Moscow (Mysl). 164192. Safronov, V. A. 1970: Chronologija pamjatnikov II tys. do n. e. Juga Vostocnoj Evropy. Avtoreferat diss. Moscow (Institut arkheologii AN SSSR). Salmon, M. H. 1982a: Models of explanation: two views. In: Renfrew, C. et al. (eds.): Theory and explanation in archaeology. New York (Academic Press). 3544. 1982b: Philosophy and archaeology. New York (Academic Press). Salmon, P. 1969: Histoire et critique. Bruxelles (Ed. de lUniversite de Bruxelles). 2 ed. Salmon, W. C. 1982: Causality in archaeological explanation. In: Renfrew, C. et al. (eds.): Theory and explanation in archaeology. New York (Academic Press). 4555. Sangmeister, E. 1967: Methoden der Urgeschichtswissenschaft. Saeculum 18:3. 199244. Sankalia, H. D. 1965: An introduction to archaeology. Poona (Deccan College). Scheidigger, A. E. 1961: Theoretische Geomorphologie. Heidelberg. Schiffer, M. B. 1972: Archaeological context and Systemic context. American Antiquity 2. 156165. 1976: Behavioral archaeology. New York et al. (Academic Press). (ed.): Ab 1978. Advances in archaeological method and theory Vol. 1 ... New York et al. (Academic Press). Schlzer, A. L. von 1768: Probe russischer Annalen. Bremen und Got tingen (Georg Foster). [Shlecer, A. L.] 1809: Russkie rukopisi na drevleslavenskom jazyke. Slichennye, perevedennye i objasnennye A. L. Shlecerom. Ch. 1. Sankt Peterburg. Schuyler, R. S. 1973: Review of Watson et al. American Antiquity 38. 372374. Schwartz, D. W. 1978: A conceptual framework for the sociology of archaeology. In: Dunnell, R. C. and Hall, E. S. (eds.). Archaeological essays in honor of Irving B. Rouse. The Hague (Mouton). 149176. Selinge, K.-G. 1987: On the prerequisits and needs of archaeological source-kriticism. British Archaeological Reports, Int. ser. 366. 27 42. Semenov, S.A. 1957: Pervobytnaja technika. Materialy i issledovanija po arkheologii SSSR 54. Moscow Leningrad (Izd. Akademii Nauk SSSR). [Eng. transl.: Prehistoric technology. Bath (Adams and Dart), London (Gory, Adams and MacKay), New York (Barnes and Noble).] Senobos, Sh. [Segnobos Ch.] 1902: Istoricheskij metod v prilozhenii k socialnym naukam. Moscow (tov-vo tip. A. I. Mamontova). Serzhantov, V. F. 1972: Vvedenie v metodologiju sovremennoy biologii. Leningrad (Nauka). Shakhmatov, A. A. 1916: Povest vremennykh let. Ch. 1. Vvodnaja chast. Tekst. Primechanija. Petrograd (Izd. Imper. arkheologicheskoj Komissii). Shanks, M. 1992: Experiencing the past: On the character of archaeology. London and New York (Routledge). and Tilley, C. 1987a: Re-constructing archaeology: Theory and practice. Cambridge (Cambridge University Press). 1987b: Social theory and archaeology. Cambridge (Polity Press).

Metaarchaeology
neiro, R. L. (eds.): Essays in the science of culture in honour of Leslie A. White. New York (Thomas Y. Crowell). 437456. 1968: Explanation in archaeology. In: Binford, S. R. and L. R. (eds.): New perspectives in archaeology. Chicago (Aldine Publishing Company). 3340. 1971: Some elements of quantifying archaeology. In: Hodson, F. R., Kendall, D. G., Tautu, P. (eds.): Mathematics in the archcaeol ogical and historical sciences. Edinburgh (Edinburgh University Press). 316. 1973: Archaeology in the active voice: the New Archaeology. In: Redman, C. L. (ed.): Research and theory in current archaeology. New York (John Wylie & Sons). 95111. Stefanov, N. 1961: Po voprosa za predmeta. Metoda i granicite na arheologicheskata i istoricheskata nauka. Izvestija na Instituta po losoja. Soja. 167198. Steward, J. H. 1949: Cultural causality and law: A trial formulation of the development of early civilizations. American Anthropologist 51. 127. and Setzler, F. M. 1938: Function and conguration in archaeology. American Antiquity 4:1. 410. Stickel, E. G., Chartkoff, J. L. 1973: The nature of scientic laws and their relation to the law-building in archaeology. In: Renfrew, C. (ed.): The explanation of culture change: models in prehistory. London (Duckworth). 663671. Stjernquist, B. 1984: Perspective on archaeological theory and method. University of Lund Institute of archaeology, Report series, 20. Lund (Lund Universitets historiska museum). Strong, D. E. (ed.). 1973: Archaeological theory and practice. London New York (Seminar Press). Strong, W. D. 1936: Anthropological theory and archaeological fact. In: Lowie, R. H.(ed.): Essays in anthropology in honor of Alfred Louis Kroeber. Berkeley (University of California Press). 359370. Stutt, A. 1988: Second generation expert systems, explanations, arguments and archaeologists. In: Rahtz, S. P. O. (ed.): Computer and quantitative methods in archaeology. British Archaeological Reports, int. ser. 446. 356368. Stutt, A., Shennan, S. 1990: The nature of archaeological arguments. Antiquity 245. 766777. 1992: Designing a workbench for archaeological arguments. In: Archaeology in the Information Age: A global perspective. London (Routledge). Sullivan, A. 1978: Inference and evidence in archaeology: A discussion of the conceptual problems. Advances in archaeological method and theory 1. New York et al. (Academic Press). 193194. Swartz, B. K. 1967: A logical sequence of archaeological objectives. American Antiquity 32:4. 487497. Szekely, E. B.1977: Archaeosophy, a New Science. New York (Academic Press). Saar, G. P. 1930: Istochniki i metody istoricheskogo issledovanija. Baku (Izd. AzGNII). Tabaczynski, St., Pleszczynska, E. 1974: O teoretycznych podsta wach archeologii. Archeologia Poski 19:1. 1013. Tallgren, A. M. 1937: The method of prehistoric archaeology. Antiquity 42. 152161. Taylor, W. W. 1948/1967: A study of archaeology. Menasha (American Anthropological Association) [(2d ed. 1967. Carbondale

147

(Arcturus Books, Southern Illinois University Press) London and Amsterdam (Feffer & Simons)]. The use 1940: The use of historical objects as sources of history. Detroit (American Association of Museums). Thomas, D. A. 1974: Predicting the past: An introduction to anthropological archaeology. Basic anthopology units. New York Holt (Rinehart & Winston). Thomas, J. 1995: Where are we now? Archaeological theory in the 90s. In: Ucko, P. J. (ed.): Theory in archaeology: A world perspective. London and New York (Routledge). 343362. Thomas, N. 1960: A guide to prehistoric England. London (Batsford). Thompson, J. E. 1944: The sh as a Maya symbol for counting and further discussion of directional glyphs. Theoretical approaches to problems 21. Washington (Carnegie Institution). Thompson, R. H. 1956a: The subjective element in human culture. Proceedings of the Prehistoric Society 5. 209221. 1956b: The subjective element in archaeological reasoning. Southwestern Journal of Anthropology 12. 327332. [reprint in: Deetz, J. (ed.) 1971: Mans imprint from the past. Little (Brown & Co.). 148154.] Tikhomirov, M. N. 1962: Istochnikovedenie istorii SSSR. Vol. 1. Moscow (Izd. soc.-ek. literatury). Tilley, C. 1991: Comments on Klejn: Theoretical archaeology in the making. Fennoscandia archaeologica VIII. 5961. Titov, V. S. 1982: K izucheniju migracij bronzovogo veka. In: Arkheologija Starogo i Novogo Sveta. Moscow (Nauka). 89145. Topolski, J. 1968: Metodologia historii. Warszawa (Panstwowe wydawnictwo naukowe). Toynbee, A. J. 1964: A study of history. Vol. 12. Reconsiderations. London et al. (Oxford University Press). Tretjakov, P. N. 1962: Rev. on E. I. Gorjunovas book. Sovetskaja Arkheologija 1962:4. 258269. 1966: Osnovnye itogi rabot Verkhnedneprovskoj arkheologicheskoj ekspedicii. In: Isaenko, V. F., Mitrofanov, A. G. and Shtykhov, G. V. (eds.): Drevnosti Belorussii. Minsk (Institut istorii AN BSSR). 114126. Trigger, B. G. 1968a: Beyond history: The methods of prehistory. New York (Holt, Rinehart and Winston). 1968b: Major concepts of archaeology in historical perspective. Man 3. 527541. 1970: Aims in prehistoric archaeology. Antiquity 173. 2637. 1973: The future of archaeology is the past. In: Redman, C. L. (ed.): Research and theory in current archaeology. New York (John Wylie & Sons). 95111. 1978a: No longer from another planet. Antiquity 206. 193198. 1978b: Time and traditions. Essays in archaeological interpretation. Edinburgh (Edinburgh University Press). 1982: Archaeological analysis and concepts of causality. Culture 2:2. 3142. 1989: A history of archaeological thought. Cambridge (Cambridge University Press). and Glover, I. (eds.) 19811982: Regional traditions in archaeology. World Archaeology 13:23. Trudzik, Z. 1971: O przedmiocie archeologii i jej procesie badawczym. Acta universitatis Wratislaviensis, Studia Archeologiczne 4 (Wroclaw). 556.

148

Acta Archaeologica
1960: Moderne Archologie: Methoden und Technik der Ausgrabungen. Reinbek/Hamburg (Rohwolt). Willey, G. R., Phillips, Ph. 1958: Method and theory in American archaeology. Chicago (University of Chicago Press). [2d ed. 1965]. Willey, G. R., Sabloff, J. A. 1973: A history of American archaeology. San Francisco (Freeman & Co.) [2d ed. 1980.] Yellen, J. E. 1976: Settlement patterns of the !Kung: An archaeological perspective. In: Lee, R. B. and De Vore, I. (eds.): Kalahari hunter-gatherers. Cambridge Mass. (Harvard University Press). 4772. 1977: Archaeological approaches to the present: Models for reconstructing the past. New York London (Academic Press). Zabelin, I. E. 1878: V chem zakljuchajutsja osnovnye zadachi arkheologii kak samostojatelnoj nauki? Trudy III Arheologicheskogo sezda, vol. I. 117. Zak, J. 1968: Archeologia w systemie nauk historycznych. Archeologia Polski 13:2. 455473. Zakharuk, Ju. N. 1969: O metodologii arkheologicheskoj nauki i ee problemakh. Sovetskaja Arkheologija 1969:3. 1120. 1970a: Leninskoe teoreticheskoe nasledie i arkheologicheskaja nauka. In: Leninskie idei v izuchenii istorii pervobytnogo obshchestva, rabovladenija i feodalizma. Moscow (Nauka). 716. 1970b: Leninskoe teoreticheskoe nasledie i nekotorye voprosy razvitija arkheologicheskoj nauki. Sovetskaja Arkheologija 1970:2. 817. 1970c: Dejaki metodologichni pytannja arkheologichnoi nauky. Arkheologija (Kyiv) 24. 311. 1971a: Do rozrobky teoretychnykh osnov arkheologii. Arkheologija, 2: 817. 1971b: K voprosu o teoreticheskikh osnovakh arkheologii. Tezisy dokladov, posvjashchennykh itogam polevykh arkheologicheskikh issledovanij v 1970 g. v SSSR Arkheologicheskie sekcii (dopoln. vyp.). Tbilisi (AN SSSR, AN GSSR). 4548. 1973: K voprosu o soderzhanii i strukture arkheologicheskoj teorii. Tezisy dokladov sessii, posvjashchennoj itogam polevykh arkheologicheskikh issledovanij 1972 goda v SSSR. Tashkent (FAN). 4445. 1975: K voprosu o predmete i procedure arkheologicheskogo issledovanija? In: Predmet i obekt arkheologii i voprosy metodiki arkheologicheskich issledovanij. Leningrad (Nauka). 46. 1977: O ponjatii fakt v arkheologicheskoj nauke. Sovetskaja Arkheologija 1977:4. 3040. 1978a: Spornye voprosy obekta i predmeta arkheologii. Kratkie soobshchenija Inst. arkheologii AN SSSR 152. 717. 1978b: Arkheologija ili pervobytmaja arkheologija? (po povodu koncepcii pervobytnoj arkheologii A. N. Rogacheva). Kratkie soobshchenija Inst. arkheologii AN SSSR 152. 2333. 1983: Ob odnoj koncepcii arkheologicheskogo istochnikovedenija. Sovetskaja arkheologija 1977:3: 7281. 1989: Arkheologija: nauka istoricheskaja ili istochnikovedcheskaja? Sovetskaja arkheologija 1989:3. 207214. 1992: K diskussii o statuse arkheologii. Rossijskaja Arkheologija 1992:4. 97101. Zamjatnin, S. N. 1951: O lokalnych razlichijakh v kulture paleoliticheskogo perioda. Trudy Instituta etnograi (nov. serija) XVI. Moscow. 89152. Zhebelev, S. A. 1923: Vvedenie v arkheologiju. Nauka i zhizn. Petrograd.

Tuggle, H. D., Townsend, A. H., Riley, Th. J. 1972: Laws, systems and research design: a discussion of explanation in archaeology. American Antiquity 37. 312. Tylor, E. 1958: Primitive culture. New York (Harper and Row). Ucko, P. J. 1995: Archaeological interpretation in a world context. In: Ucko, P. J. (ed.): Theory in archaeology: A world perspective. London and New York (Routledge). 127. Urbanczyk, P. 1981: Archeologia wspoczesnosci. Archeologia Polski 26:1. 4963. Uslar, R. von. 1955: ber den Nutzen spekulativer Betrachtung vorgeschichtlicher Funde. Jahrbuch des Rmisch-Germanischen Zentralmuseums Mainz 2. 120. Viktorova, V. D. 1975a: Arkheologicheskij fakt. Voprosy arkheologii Urala 13. 1726. 1975b: Princip predmetno-prakticheskoj dejatelnosti kak osnovanie dlja sistematizacii arkheologicheskikh faktov. In: Novejshie otkrytija sovetskikh arkheologov. Kiev (Naukova dumka). 1819. 1988: [Comments on Bashilovs and Loones article]. Sovetskaja Arkheologija 1988:4. 224226. 1989: Nauchnyj poisk v arkheologii. Sverdlovsk (Izd. Uralskogo universiteta). 1989. Windelband, W. 1924: Prludien: Aufstze und Reden zur Philosophie und ihre Geschichte (Neue Au.). Tbingen. Vipper, R. 1921: Krizis istoricheskoj nauki. Kazan (Gos. Izd.). Vishnyatsky, L. B. and Kolpakov, E. M. 1990: Current theoretical discussion in Soviet archaeology: An essay. Fennoscandia Archaeologica 7. 1725. Vysockij, B. P. 1961: Problema aktualizma i uniformizma i sistema metodov v geologii. Voprosy Filosoi 3. 134145. Wahle, E. 1924/1964: Vorgeschichte des deutschen Volkes. Vorwort. [reprinted in Wahle 1964. 316333.] 1930/1964: Besprechung A. W. Brgger. Deutsche Literaturzeitung [reprinted in Wahle 1964. 307315.] 1941: Zur ethnischen Deutung frhgeschichtlichen Kulturprovinzen. Grenzen frhgeschichtlichen Erkenntnis. I. Sitzungsberichte der Hedelberger Akademie der Wissenschaften, Phil.-hist. Klasse, 1940/41, Abh. 2. Heidelberg. 19501951: Geschichte der prhistorischen Forschung. Anthropos 45. 497538; 46. 49112. 1954: Ur- und Frhgeschichte im mitteleuropaischen Raum [reprinted in Wahle 1964. 334449.] 1964: Tradition und Auftrag prhistprischer Forschung. Berlin (Duncker & Humblot). 1970: Ur- und Frhgeschichte im mitteleuropaischen Raum. Handbuch der deutschen Geschichte, Bd. I (9. Au.). Stuttgart (Union). 614. Walther, J. 1894: Einleitung in die Geologie als historische Wissenschaft. Bd. 3. Jena (G. Fischer). Watson, P. J., LeBlanc, S. A., Redman, Ch. L. 1971: Explanation in archaeology. An explicitly scientic approach. New York and London (Columbia University Press). Werner, J. 1954: Neue Wege vorgeschichtlicher Methodik? Forschungen und Fortschritte 28:8. 246248. Wheeler, M. 1952: Archaeology and the transmission of ideas. Antiquity 26:104. 180192. 1954: Archaeology from the earth. Harmondsworth (Penguin books).

Metaarchaeology
Zotov, A. F. 1973: Struktura nauchnogo myshlenija. Moscow (Politizdat). Zubrow, E. 1973: Adequacy criteria and prediction in archaeological models. In: Redman, C. L. (ed.): Research and theory in current archaeology. New York (John Wylie & Sons). 239255.

149

1980: International trends in theoretical archaeology. Norwegean Archaeological review 13:1. 1423. Zvegincev, V. A. 1967: Teoreticheskaja i prikladnaja lingvistika. Moscow (Prosveshchenie).

Authors address: Ul. Zhelenovodskaja D. 27, Kv. 27 199155 St. Petersburg Russia klejn/lk1256.spb.edu