Natural Disasters: A Framework for

Research and Teaching

DAVID ALEXANDER

Natural disasters are defined in this paper by relating the impact of extreme
geophysical events to patterns of human vulnerability. Hazard perception is shown to
be a factor that limits the mitigation of risk. The historical development of disaster
studies is traced and five different schools of thought are identified. The current
International Decade for Natural Disaster Reduction (ZDNDR) is evaluated critically
zclifh regard to its potential for unifying the disparate strands of knowledge and its
scope as a vehicle for education.
A pedagogical framework for disaster studies is presented. Time and space provide
valuable unifying factors, while the subject matter can be differentiated according to
the continua and dichotomies that it presents. In disaster studies as in other branches
of higher education, an ecocentric approach is preferable to a technocentric one, as
inany of the poorer nations of the zuorld, which are most afflicted by natural
catastrophe, will have to rely for mitigation on maintaining their ecological
sustainability, instead of depending on sophisticated technology. Valuable insights into
the impact of environmental extremes on mankind are gained from the study of
disasters as human ecology.
The United Nations has designated the framework for teaching it at the level of
years 1990-2000 as the International Decade adult education.
for Natural Disaster Reduction (IDNDR). The key to a successful didactic ap-
Education is a vital component of this global proach to natural disasters is to consider
initiative to reduce the toll of casualties and them as whole phenomena, in which the
destruction caused by natural catastrophe, demands of the problem (such as search
and it is primarily at the post-school level and rescue, the provision of shelter and the
that tomorrow's hazard managers and management of mass casualties) replace
scientists will be instructed and trained. narrow disciplinary specializations. Thus, in
Obviously, it is essential that they be given the emergent field of disaster studies it is
a broad and rigorous grounding in the imperative that the approach be fully inter-
theories, methodologies and examples that disciplinary. Moreover, there is consider-
constitute the "proto-discipline" of natural able scope for considering disasters within
hazards and disaster studies.' In this paper the compass of ecology, as extreme forms
I review the past development and current of ecological - or more properly human
state of this field, examine some of its ecological - relationships.*
theoretical underpinnings and propose a

DISASTERS VOLUME 15 NUMBER 3

220
ESSENTIAL DEFINITIONS
According to Burton and Kates (1964,
p. 413) ’‘Natural hazards are those elements
in the physical environment [which are]
harmful to man and caused by forces
extraneous to him”.3 Turner (1976,
pp. 755-756) offered a more detailed defi-
nition of disaster, as
” . . . an event, concentrated in time and
space, which threatens a society or a relatively
self-sufficient subdivision of a society with
major unwanted consequences as a result of
the collapse of precautions which had hitherto
been culturally accepted as adequate”.
From this it is clear that disasters occur
when physical impacts coincide with
human vulnerability, and that the concept
of risk principally involves the likelihood of
loss sustained by elements in the human
landscape which are threatened by future
hazard impacts. These impending losses
can be mitigated by adjustment in the form
of modifications to extreme natural events,
reductions in human vulnerability, or in the
last resort by taking emergency action
(Kates, 1971).
In cases where there is some degree of
choice, risks are assumed because they
entail benefits to the risk taker. Thus, the
impact of disasters can be viewed as a
simple conceptual equation (Burton, Kates
and White, 1978):
net impact total benefits total costs
of disasters = of inhabiting - of disaster -
risk zone impact
costs of
adaptation
to risk
According to an alternative formulation4
the sum total of risk is a product of the
magnitude and extensiveness of the natural
hazard impact, the number and size of
elements at risk, and their vulnerability in
terms-of probable levels of damage and
destruction:
DISASTERS VOLUME 15 NUMBER 3
David Alexander
total - impact of elements
X X
risk hazard at risk
vulnerability of
elements at risk
All the constituents of this equation
may vary. While in certain fields the impact
of natural hazards is fairly immutable, or
subject to definablenatural trends, in others
it can be altered by human activity (as in the
cases of induced seismicity, flooding related
to the “greenhouse effect” and subsidence
resulting from groundwater withdrawal).
Planning laws, development initiatives and
evacuation plans are examples of measures
which can alter the number of elements at
risk. Their vulnerability is a function of
three factors (Alexander, 1991):
total risk risk
vulnerability = amplification - mitigation f
measures measures
risk
perception
factors
Risk amplification occurs as a result of
the continued development of past and
future disaster areas, but it can be reduced
by mitigation efforts. Micklin (1973) lists
four ways in which the latter can be applied:
engineering mechanisms, including
technological innovations and
application;
symbolic mechanisms, including
culture and its constituent norms and
roles;
regulatory mechanisms, defining
public policy and social control;
distributional mechanisms, which
specify the movement of people,
activities and resources.
The extent of risk amdification or
mitigation will depend on the degree to
which it is perceived. Mileti (1980) noted
that perception can be divided into that
Natural Disasters: A Frameuiork for Research and Teaching
pertaining to the likelihood of damage and
that relating to the role of mitigation. The
level of perception will depend on the
ability to estimate risk and perceive its
causes, the level of past experience with
hazards, the propensity to deny that a risk
exists, the level of access to appropriate
information and the size of the unit
analysed. Preston et al. (1983) described the
psychological adjustments which are
necessary to reduce "cognitive dissonance",
which they defined as the psychological
discomfort that arises when two conflicting
beliefs are held simultaneously, as when a
person perceives his environment to be
hazardous, but continues to live in it.
SCHOOLS OF THOUGHT
In the English-speaking world the field of
natural disasters has grown in close associ-
ation with the applied natural and social
sciences, but has suffered from problems of
fragmentation and over-specialization, as
well as the insuIarity that disciplinary
studies tend to foster. To some extent this
has prevented it from acquiring a separate
identity.
During the last 125 years there has been
a gradual shift of emphasis in the natural
and social sciences from the impact of en-
vironment upon mankind to humanity's
impact on environment (Holt-Jensen, 1988).
In one sense, disasters represent an extreme
class of human-environmental or ecological
phenomena.s They cannot truly be con-
sidered natural, in that human vulnerability
seldom results from purely natural states
(rather than locational decisions based on
socio-economic criteria) and human in-
tervention often results in aggravated risk
of geophysical impact.6 Hence, because
they represent an environmental imperative
to society, natural disasters have been
studied using the tenets of human ecology,
which Mileti (1980) defined as " . . . seeking
the determinants of human behaviour in
natural environment, and the processes that
211
facilitate human adjustment to the physical
world through social organization".
Although an all-embracing concept such
as human ecology is capable of unlfying dis-
parate subject matter, it has not been
adopted universally. The increasing div-
ision of knowledge into disciplines, which
have in turn spawned sub-disciplines, has
meant that natural hazards and disaster
studies have had to struggle for identity
against a wide range of groups and sub-
groups, many of which have substantial
overlap and common ground but often very
different viewpoints, approaches or objec-
tives. Regrettably, the field has responded
to increasing disciplinary specialization by
becoming fragmented. Hence, the following
schools of thought and expertise can at
present be identified.
(1) The geographical approach to natural
hazards stems from Harland Barrows'
work in the 1920s on human ecological
adaptation to environment (Barrows,
1923) and Gilbert F. White's seminal
1940s monograph on flood perception
(White, 1945, 1973). Social science
methods are widely used and emphasis
is given to the spatio-temporal distri-
bution of risk, impacts and vulner-
ability (Palm, 1990).7
(2) The anthropological approach, as
evinced by the work of Torry (1979a),
Dudasik (1982) and Oliver-Smith (1990)
has focused on the role of disasters in
guiding the socio-economic evolution
of populations, in dispersing them and
in causing the destruction of civiliz-
ations (although Torry [op. cit.] noted
that the last of these is very debatable).
A strong concern for the Third World
has led anthropologists to search for the
threshold points beyond which local
communities can no longer provide the
basic requirements for survival of their
members. They have also studied the
"marginalization syndrome" arising
from the impoverishment of disad-
DISASTERS VOLUME 15 NUMBER 3
212
vantaged groups in underdeveloped
societies.
(3) The sociological approach stems from
the work of Russell R. Dynes, Enrico
L. Quarantelli and others. Vulnerability
and impacts are considered in terms of
patterns of human behaviour and the
effects of disasters upon community
functions and organization (Quaran-
telli, 1978; Drabek, 1986; Dynes, De
Marchi and Pelanda, 1987).In addition,
psychologists have studied disaster in
relation to factors such as stress (Glass,
1970),bereavement (Church, 1974)and
the "disaster syndrome", a psycho-
logically determined defensive reaction
pattern (Wallace, 1956).
(4) The development studies approach
considers problems of providing aid
and relief to Third World countries,
and addresses questions of refugee
management, health care and the
avoidance of starvation. Logistical as-
pects are emphasized, as are nutritional
studies (Cuny, 1983; D'Souza and
Crisp, 1985; Seaman, Leivesley and
Hogg, 1984). Davis (1978) stated that
over 80 per cent of disaster impacts
occur in developing countries,' and it
is clear that the epiphenomenon of
poverty increases human vulnerability
to natural hazards: locational con-
straints tend to place the poor more
firmly in the path of impacts, while
credit, savings, capital and alternative
options that normally cushion the
impact are lacking (Tory, 1979a). In
addition, Rivers (1982) found that in
less-developed societies the burden of
coping with disasters falls dispro-
portionately on the unemancipated
woman.
(5) A new field of disaster medicine and
epidemiology has recently been
founded. It focuses on the manage-
ment of mass casualties, the treatment
of severe physical trauma and the
epidemiological surveillance of com-
DISASTERS VOLUME 15 NUMBER 3
David Alexander
municable diseases whose incidence
rates may increase during the dis-
ruption of public health measures
following a disaster (Beinin, 1985;
Manni and Magalini, 1985).
(6) The technical approach prevails among
natural and physical scientists (Smith,
1985). Emphasis is given to seismology,
volcanology, geomorphology and other
predominantly geophysical approaches
to disasters (Bolt et al., 1977; Stein-
brugge, 1982).
Clearly, the prevailing theories and
models in hazard and disaster studies have
not succeeded in preventing the emergence
of quite separate schools of thought that
deal with what are essentially the same
phenomena. Dichotomies exist between the
technocentric school, to whom the solution
to the disaster problem lies in the ap-
plication of measuring and monitoring
techniques and sophisticated managerial
strategies, and the development school, who
point out that, as a result of poverty, such
luxuries are denied to the majority of
disaster victims. Concomitantly, most of the
theory used by hazards specialists has been
formulated for the developed world (es-
pecially the United States of America) and
is of doubtful validity elsewhere.
One opportunity to remedy these prob-
lems is presented by the IDNDR, which
therefore merits a brief critical examination.
THE INTERNATIONAL DECADE FOR
NATURAL DISASTER REDUCTION
Dr Frank Press, the President of the U.S.
National Academy of Sciences, first pro-
posed the IDNDR in 1984 at the Eighth
World Congress on Earthquake Engineer-
ing. Eventually, the idea was adopted by
the United Nations, under Resolution 421169
of 1987.9 The United States continued to
take the lead by issuing two prospectuses
on American efforts connected with the
Decade (U.S. National Research Council,
Natural Disasters: A Framework for Research and Teaching
1987, 1989). Several proponents and prac-
titioners have published their ideas on how
the Decade should be operationalized
(Holland, 1989; Housner, 1989; Lechat,
1990; Oaks and Bender, 1990).
It seems that the principal thrust of the
Decade will be to promote the international
sharing of ideas and data (through con-
ferences and other co-operative efforts), and
the development of national and world-
wide networks for monitoring the agents
which produce disasters. Laudable as these
aims are, the initiative has not escaped
criticism, calling into question the American
prediction that it will be possible to halve
the impact of natural disasters by the year
2000. Mitchell (1988) has made the following
points about the IDNDR in general and the
U.S. Decade for Natural Disaster Reduction
in particular.
(1) New research, such as that proposed
under the auspices of the Decade, is
not easily justified when much current
knowledge and expertise remain un-
utilized.
(2) Broad, societal processes are not easy
to change, yet are fundamental to
human vulnerability. They are not con-
sidered explicitly in the prospectus for
the USDNDR.
(3) Expensive and sophisticated monitor-
ing, experimentation and management
initiatives are unlikely to work in poor
countries. But non-structural ap-
proaches and the applications of alter-
native, low-level technology are not
receiving adequate consideration under
the terms of the Decade.
(4) Co-operative science has a mixed
record of success in reducing the toll of
natural hazards. Its efficacy should be
evaluated before any large new
projects are proposed.
(5) As initially specified, the Decade lacks
a strong focus on a few relatively
simple issues. Resources will undoubt-
ably be insufficient to solve, or even
213
tackle, all major problems of disasters
around the world.
Hence, in the eyes of some practitioners,
the IDNDR has assumed the status of a
“technofix”, in which the proponents of
technology and hard science use it as a
justification for generating yet more of the
same. Sociologists (Dynes, 1990), Third
World development specialists and ecolo-
gists (Burin, 1989) have been quick to point
out the lack of reference to their own special-
ities. An extreme interpretation would be
that the Decade represents an attempt by
engineers and physical and natural scien-
tists to concentrate academic power and
funding opportunities into their own hands
in the name of applying their sciences.
Unless the character of the IDNDR
changes as it evolves, it would seem that it
offers only limited scope in the field of
ecological education. It can, however, be
used profitably as a justification for any
initiative to increase public or student
awareness of disasters.
TOWARDS A PEDAGOGICAL
FRAMEWORK FOR NATURAL DISASTERS
Although the discipiinary structure of
disaster studies is still fragmentary, the field
does not lack regularities. Some of these
take the form of natural or social “laws”
(albeit usually of the statistical variety), such
as the following.
(1) Geophysical events of high magnitude
tend to occur with low frequency, while
those of low magnitude tend to be
numerous in time. In physical terms,
the amount of work done by events
of a given size is a product of their
magnitude and their frequency of
occurrence (the ‘magnitude-frequency
principle”).
(2) People tend to overestimate the impact
of disastrous or sensational hazards
and to underestimate that of pervasive
hazards which claim only small
DISASTERS VOLUME 15 NUMBER 3
214 David Alexander

TABLE 1
Continua, dichotomies and polychotomies in disaster studies

Confinua in hazards and disasfers:

events:
anthropogenic disaster t-) natural disaster
sudden impact disaster ++ creeping disaster
short-term aftermath ++ long-term aftermath
(restoration) (reconstruction)
scientific organization:
technocentric approach t-) ecocentric approach
natural hazards t-) environmental geology
(social science) (natural science)

risk amplification
-
attitudes and approaches:
symbiosis with environment
t-)
parasitism (exploitation)
risk reduction
optimizer t-) satisficer
mitigation .-, laissez faire
fatalism
environmental determinism
t-)

t-)
activism
probabilism - possibilism

Dichotomies and polychotomies:

recurrence interval time-scale of

for most disasters geological events
[IO-’-IO* years] [103-109 years]

simple impact composite disaster secondary disaster

[earthquake] [earthquake & tsunami] [post-earthquake fire]

structural mitigation non-structural mitigation

[retrofitting of buildings] [insurance]

numbers of victims each time they
strike (Preston et al., 1983).
(3) The more chronic and well-known the
threat, the more integrated it will be
with the local culture, the more
uniform will be the reaction to it and
the less shifting will be the focus of
concern (Anderson, 1967, p. 304).
DISASTERS VOLUME 15 NUMBER 3
(4) As physical or emotional distance from
the disaster increases, so the level of
psychological impact falls, unless
death, destruction or loss increase pro-
portionately. This is Turner’s modifi-
cation of Kastenbaum’s so-called ”Law
of Inverse Magnitude” (Turner, 1976).
(5) Rather than stimulate a massive out-
Natural Disasters: A Framework for Research and Teaching 215

migration of displaced populations, TABLE 2

most disasters give rise to a ”conver-
gence reaction”, in which many groups
and organizations come to the disaster
area (Fritz, 1957).
Hence, there is ample scope for basing
the teaching of disaster studies on verifiable
Structural and non-structural methods of
disaster mitigation
Structural methods:
Retrofitting of existing structures
Reinforcement of new structures
- design features

generalizations. However, there are also - overdesign

considerable uncertainties. For example, the Safety features
process of turning the prediction of an - structural safeguards

impending impact into a public warning, - failsafe design

and of ensuring that appropriate action is Engineering phenomenology

taken (such as evacuation), is a very impre- Probabilistic prediction of impact strength
cise one in which the warning message is Nun-structural methods:
as likely to be ignored or misinterpreted as (a) short-term:
it is to be heeded (Foster, 1980).In consider- Emergency plans
ing the role of information in inducing - civil
people to take immediate action to prevent - co-ordinator(s)
the worst effects of disaster, Sims and - police and firemen
Baumann (1983) were only able to conclude - Red Cross and charities
with the following highly tentative state- - volunteer groups
ment: ”Information may lead to behaviour - medical services
change . . . under highly specific conditions - military forces
. . . if properly executed [i.e., delivered to Evacuation plans
recipients] . . . with specific targets”. - routes and reception centres
Another focus for study is the presence - for the general public
of various highly significant continua, - for vulnerable groups: the very
dichotomies and polychotomies in the field young, elderly, sick, or handicapped
(Table 1).For instance, a basic dichotomy Prediction of impact
exists between structural and non-structural - monitoring equipment
methods of hazard mitigation (Table 2). The - forecasting methods and models
former involve engineering methods and Warning processes
architectural design, while the latter com- - general message
prise most other methods, including oper- - specialized warning (e.g. ethnic)
ations research, land-use planning, risk
(b) long-term:
analysis and economics.
Building codes and construction norms
One of the most significant of the
Hazard microzonation
continua is that which exists between dis-
- selected risks
asters that strlke abruptly (e.g., earthquakes
- all risks
and tornadoes), and those that have long
Land-use control
drawn-out impact and are hence known as
- regulations, prohibitions, moratoria
“creeping disasters” (e.g ., desertification
- compulsory purchase
and accelerated soil erosion). Tables 3 and
Probabilistic risk analysis
4 show that the geophysical agents which
Insurance
produce disasters can be classified, not only
Taxation
according to speed of onset, but also with
Education and training
respect to potential duration of forewarning
DISASTERS VOLUME 15 NUMBER 3
216 David Alexander

TABLE 3
Classification of disasters by duration of impact and length of forewarning

Duration of impact Length of forewarning (if any)

Type of disaster 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 1 0

Lightning
Avalanche
I
Earthquake
Landslide
Tornado
Intense rainstorm
Hail
Tsunami
Subsidence
Windstorm
Frost or ice-storm
Hurricane
Snowstorm
Environmental
fire
Volcanic eruption
Insect infestation
Fog
Flood
Coastal erosion
Drought
Crop blight D D

Expansive soil
Accelerated
erosion
Desertification
and precursors, as well as by frequency and
pattern of occurrence.1°
Being a linear measure, time is essen-
tially the "backbone" of disaster studies. For
example, Figure 1 illustrates the temporal
sequence of solutions to the post-disaster
housing problem, with reference to a
sudden-impact disaster that strikes a devel-
DISASTERS VOLUME 15 NUMBER 3
..
oped country and generates an abrupt and
massive demand for shelter. Table 5 lists the
time-phases into which sudden or relatively
abrupt disasters are usually classified, and
gives three hypothetical examples. A more
simple approach would be to use only four
time periods: impact and emergency, repair
of essential services, replacement recon-
Natural Disasters: A Framework for Research and Teaching 217

TABLE 4
Classification of disasters by frequency or type of occurrence

Type of disaster Frequency or type of occurrence*

Lightning Random
Avalanche Seasonalldiurnal; random
Earthquake Log-normal
Landslide Seasonal-irregular
Tornado Seasonal; negative binomial
Intense rainstorm Seasonalldiurnal; Poisson
Hail Seasonalldiurnal; Poisson, gamma, negative binomial
Tsunami Random
Subsidence Sudden or progressive
Windstorm Seasonallexponential
Frost or ice-storm Seasonalldiurnal; Markovian, binomial
Hurricane Seasonallirregular
Snowstorm Seasonal; modified Poisson**
Environmental fire Seasonal; random
Volcanic eruption Irregular
Insect infestation Seasonal; random
Fog Seasonall diurnal
Flood Seasonal; Markovian, gamma, log-normal
Coastal erosion Seasonallirregular; exponential, gamma
Drought Seasonallirregular; binomial, gamma
Crop blight Seasonallirregular
Expansive soil Seasonal or irregular
Accelerated erosion Progressive (threshold may be crossed)
Desertification Progressive (threshold may be crossed)

* Frequency distributions adapted from Hewitt (1970, pp. 333-4)

** Eggenberger and Polya modification: see Hewitt (1970)

struction, and developmental reconstruc-
tion (Kates and Pijawka, 1977, and other
sources).
Geography is also relevant here, and
Figure 2 gives a simple model of spatial
relations in disaster. Unfortunately, few
geographers have attempted to apply their
skills to spatial modelling of natural catas-
trophe (Alexander, 1989; Montz, 1982), and
hence there is a dearth of models for the
longer term spatial effects of disaster.
Nevertheless, time and space both provide
essential foci for teaching about the evolu-
tion of particular disaster situations.
Another framework for study is pro-
vided by the distributive effects of disaster
(Table 6; Burton, Kates and White, 1978).
As the magnitude of individual impacts
decreases, so the size of the population
affected increases (Burton, Kates and White,
1978).Using this framework involves start-
ing with the medical, public health and
DISASTERS VOLUME 15 NUMBER 3
218 David Alexander

OF HOUSING
1

CAUTIONARY PERMANENT
pending pending pending
survey repair resettlement

SOLUTIONS

BUSES AND PUBLIC

AUTOMOBILES BUILDINGS

TENTS HOTELS SPONSORED MOBILE

OUTMIGRATION TRAILERS

-1

PREFABRICATED

HOUSING UNITS

RECONSTRUCTION
permanent
reurbanization
of the site

FIGURE 1 Shelter, rehousing and reconstruction after a disaster (the example of a developed country)

DISASTERS VOLUME 15 NUMBER 3

Natural Disasters: A Framework for Research and Teaching 219

MOOEL OF WALLACE (i956) -modified

5000 iW0 500 100 50 5 0 km

NON - ISOTROPIC

FIGURE 2 Simple rriodels of spatial relafzons in d i s a s t u

TABLE 5
Summary of time periods in disaster (with examples)

Purely hypothetical examples

Time period Earthquake Tornado Riverine
Flood
~~~ ~

”Incubation” or return period 150 years 5 years 100 years

Immediate precursor period none 20 mins 15 hours
Impact 100 secs 5 mins 36 hours
Aftermath or crisis period:
Isolation 8-48 hrs* 2 hours 2 hours
Search and rescue 2-7 days 12 hours 3 days
Repair of basic services 4 weeks 3 weeks 5 weeks
The long term:
Restoration-reconstruction 12 years 2 years 4 years
Developmental reconstruction 25 years 3 years 12 years

* Time lapse represents centre-periphery dichotomy

DISASTERS VOLUME 15 NUMBER 3

220 David Alexander

sociological side of disaster, and moving on TABLE 6

progressively to the engineering, architec- Distributive effects of natural hazards
tural, sociological and economic aspects. and disasters*
The fundamental principle underlying
these approaches is that the key to the S mat 1 popu 1ation affected,
adequate development and teaching of dis- concentrated effects
aster studies lies in making them interdis-
ciplinary; in particular by concentrating on Death (mortality)
the unexplored ground at the point of con- Injury (morbidity)
tact between scientific specializations (e.g., - trauma
between geomorphology and architecture, - severe (hospitalization)
human perception and volcanology, engin- - slight (out-patient treatment)
eering and human behavioural patterns). - disease
This w d require a considerable accession of - starvation and malnutrition
ingenuity and inventiveness on the part of - psychological injury
scientists and teachers who are unaccus-
tomed to break ground outside their own Bereavement
specializations, or to study unfamiliar Homelessness
subjects. - permanent (reconstruction or
It is thus necessary to refocus natural migration)
hazards and disaster studies from the - temporary (repair)
current fragmentary approach to one that
is much more unified and hence much more Unemployment
capable of giving birth to coherent theories Damage and destruction of assets and
of wide general applicability.The key to this possessions
is to treat hazards and disasters as complete
phenomena. A general examination of the Economic loss
nature and evolution of the event, in other - loss of income or investment,
words of its phenomenology or problematic, indebtedness, bankruptcy
will determine which are the most pressing Disruption of activities
questions that it poses, and hence which is
the combination of disciplines and methods Voluntary donation
that can be employed to solve them. This Mandatory taxation
will require re-education of practitioners at
both the theoretical and the applied levels, Large population affected,
so that they have a much wider and more well-distributed effects
integrated knowledge of natural events and
* After Burton, Kates and White (1978)
their human consequences.
The unified strategy outlined here will
only be successful in research and pedagogy bute to each aspect.
if the problem is conceptualized in terms of Political aspects of disaster also deserve
its constituents, independently of any to be examined. With regard to public
disciplinary constraints. This has been administration and natural catastrophe,
attempted in Table 7, which lists some of Olson and Nilson (1982) classified politics
the main geophysical and socio-economic into participatory, specialist, pluralist and
aspects of disasters .I1 Table 8 reconceptu- elitist forms, and public policy into distribu-
alizes a selection of these in terms of the tive, constituent, regulative and redistribu-
disciplines that have something to contri- tive. They noted that political culture,
DISASTERS VOLUME 15 NUMBER 3
Natural Disasters: A Framework for Research and Teaching
TABLE 7
Aspects of disasters pertinent to their
management
Physical occurrence:
Probability
Frequency
Transience (duration)
Physical magnitude
Energy expenditure
Physical effects: direct, indirect and
secondary
Area affected: directly and indirectly
Degree of spatial concentration or
ubiquity
Volume of products (e.g., lava,
floodwater)
Predictability:
Short-term (for avoiding action)
Long-term (for structural and non-
structural adjustment)
Con trollability :
Can physical processes be modified?
Can physical energy expenditure be
reduced?
Can effects be mitigated?
Can effects be modified?
Socio-culfu ral factors:
Belief systems inherent in societies
Degree of knowledge of risk
Complexity of social system and its
constituent groups
expressed as power relations, may inhibit
decision-making by preventing a latent
issue, such as a disaster mitigation strategy,
from becoming a question for decision.”
Furthermore, Mitchell et al. (1989) argued
that political questions may provide a
context which on occasion actually over-
shadows the impact of natural disaster.
Although it is not difficult to map out
the fundamental elements in natural disas-
ter and relate them to each other (Tables
1-8), the exercise is not without contro-
221
versy. Hewitt (1983) argued that the direc-
tion of causality has been mistaken by the
majority of scholars in the field. In catas-
trophe nature is seen to decide which social
conditions or responses will be significant.
Hewitt has argued that vulnerability and
human social organization are, instead, the
critical determinants of both risk and
impact. This view is reinforced by Kreps
(1989)and Drabek (1989),who both redefine
disasters as ”non-routine social problems”
(my emphasis).
CONCLUSION: HUMAN ECOLOGY AND
THE ECOCENTRIC APPROACH
Natural hazards should be considered in an
integrated way in terms of their geophysical
impacts, their human repercussions, and
the opportunities for monitoring and miti-
gating them. The last of these questions
poses the dilemma, which is well-known to
environmentalists, of to what extent techno-
centric, rather than ecocentric approaches
are capable of mitigating hazards. The view
that technology will eventually triumph
over catastrophe is still fashionable, but is
increasingly subject to criticism and doubt.
Economic power, the force behind tech-
nocentrism, is related to the seriousness of
disaster impacts in a non-linear way that is
tempered by political factors and the
tendency to accumulate physical capital and
thus increase vulnerability (Figure 3).
Adaptation, in varying degrees, is the
key to the human ecology of natural dis-
asters. It can take any of four forms
(Alexander, 1991).
(1) Persistent occupation of the hazard
zone despite the risk involved: (a) with
comprehensive measures for risk
mitigation and hazard abatement; (b)
with only warning and evacuation
measures; or (c) without any protection
measures (the state of maximum
vulnerability).
(2) Cohabitation with the damage caused
DISASTERS VOLUME 15 NUMBER 3
222 David Alexander

TABLE 8
Principal disciplines involved in disaster studies

Aspects of the Example type of disaster

disasters problem Earthquakes Floods Hurricanes

Physical impact
- Magnitude and geology, physical meteorology,
frequency seismology, hydrology, oceanography
statistics statistics
- Geographical geology, geographical meteorology,
location and extent physical hydrology remote sensing
~eograPhY /

seismology
- Physical processes geology, hydraulics, atmospheric
at work seismology physical physics
hydrology
Effects
- Risk to individuals medicine, public health public health
(mortality and public health
morbidity)
- Damage and architecture, architecture, architecture,
destruction structur a1 structural structural
engineering engineering engineering
- Other effects: economics, perceptual economics,
e.g. economic, engineering geography, geomorphology
psychological geology, geomorphology
psychology,
sociology, etc
Predictability of impacts
- Prediction seismology hydrology climatology,
techniques meteorology,
oceanography,
remote sensing
- Scope of prediction geology, statistics meteorology
. and likelihood seismology
of success
Mitigation
- Structural and architecture, civil structural
semi-structural structural engineering engineering
engineering
- Non-structural economics, actuarial planning, public
planning science, administration
planning

DISASTERS VOLUME 15 NUMBER 3

Natural Disasters: A Framework for Research and Teaching
Cumulative
impact of
disasters
FIGURE 3 Level of economic development and severity of natural disaster (modified from O’Keefe, in
T o r y , 1979, p . 537)
by past disasters (the state of rnaxirnurn
geographical inertia).
(3) Abandoning damaged or destroyed
structures, but relocating within the
risk zone (the case of secondary geo-
graphical inerfia).
(4) Migration to safer zones: (a) planned;
or (b) unplanned.
As Parker and Harding (1979) have
noted, by emphasizing adjustment, adap-
tation and perception, the study of natural
disasters draws attention to the dynamic
relationship between humanity and en-
vironment and can be an excellent means
of enhancing environmental awareness.
The field can also be broadened to embrace
ecological destruction (Ball, 1979). In fact,
223
on
momy
technology and
fixed capital
accumulation
cause damage to
increase in scope
and complexity-
Bunin (1989) argued that the IDNDR should
take place explicitly within the compass of
efforts to maintain ecological sustainability,
as the main scope for disaster mitigation in
the less developed nations lies in protecting
their natural resources. Disasters must,
however, be presented as holistic, inter-
disciplinary phenomena, as the boundaries
between disciplines tend to impede under-
standing and restrict the creation of theory.
Notes
This paper was read at the International Con-
ference on Environmental Education, “Strategy
for Lifelong Environmental Education in the
USSR”, held at Kazan’ University in the Soviet
Union from 28 October to 3 November 1990
DISASTERS VOLUME 15 NUMBER 3
224
under the auspices of the State Committee for
Higher Education. I thank my Tatar and Russian
hosts for their interest in and support for this
work, especially Dr Marat Khabibullov, Inter-
national Programmes Co-ordinator, and Dr Yuri
Kotov, USSR People’s Deputy.
1. In this account the terms ”hazard” and
“disaster” will be used synonymously. 1 will
not consider the so-called ”technological”
hazards and disasters (such as nuclear
accidents, toxic spills and urban fires) and
the effects of armed conflict.
2. However, the anthropologist William I.
Torry observed (1979a, p. 377) that . . . an
I’
ecological perspective is curiously absent
from the hazard research literature”.
3. According to Torry (1979b), Burton, Kates
and White (1978) confer three distinct
meanings on the term “hazard”: the agent
which produces the impact, risk of exposure
to the agent, and the losses resulting from
impact.
4. Given by UNESCO and the Office of the
United Nations Disaster Relief Co-ordinator
(see Alexander, 1990, p. 4).
5. See, for example, Chandler, Cooke and
Douglas (1976).
6. For example, when a slope is destabilized
by construction activities (Leighton, 1976).
7. The approach of geographers has been criti-
cised by Torry (197913) and Waddell (1977)
on the grounds of lack of rigour and excess
of technocentrism, manifest as insufficient
concern for the viability and appropriateness
of technology created in industrialized
nations and applied to developing countries.
8. For example, Turner (1979) pointed out that
in the first two thirds of the present century
there were no earthquakes of magnitude
2 8.0 in Western Europe and there was only
one in the United States, but there were 16
in Asia and 23 in Latin America.
9. The text of this Resolution was reprinted in
Housner (1989).
10. Mitchell (1974) offered a parallel approach
by classifying disasters according to the form
of causal agent: climatic and meteorological,
geological and geomorphic, floral and
faunal, etc. Foster (1976), on the other hand,
designed a twelve-point ”magnitude” scale
to classify disaster impacts, using meth-
DISASTERS VOLUME 15 NUMBER 3
David Alexander
odology which is closely related to that
employed in earthquake intensity scales.
Finally, social aspects of taxonomy are dealt
with by Kreps (1989) and his collaborators
in the same volume (e.g., Drabek, 1989).
11. An earlier and more discursive approach is
to be found in Stoddard (1968).
12. Boyce (1990) has chronicled a powerful
example of how he believes this is being
done with regard to the flood problem in
Bangladesh.
References
Alexander, D.E. (1989) Spatial aspects of
earthquake epidemiology. Proceedings of the
lnternational Workshop on Earthquake Injury
Epidemiology for Mitigation and Response, Johns
Hopkins University, Baltimore, July 1989,
82-94.
Alexander, D.E. (1990) Calamitu Naturuli:
Lineamenti di Geologia Ambientale e Studio dei
Disastri. Pitagora Editrice, Bologna, Italy,
284 pp.
Alexander, D.E. (1991) Applied geomorphology
and the impact of natural hazards on the built
environment. Natural Hazards 4(1), 57-80.
Anderson, J.W. (1967) Cultural adaptation to
threatened disaster. Human Organization 27(4),
298-307.
Ball, N. (1979) Some notes on defining disaster:
suggestions for a disaster continuum. Disasters
3(1), 3-7.
Barrows, H.H. (1923) Geography as human
ecology. Annals of the Association of American
Geographers 13, 1-14.
Beinin, L. (1985) Medical Consequences of Natural
Disasters. Springer-Verlag,New York, 271 pp.
Bolt, B.A., W.L. Horn, G.A. MacDonald and
R.F. Scott (1977) Geological Hazards: Earth-
quakes, Tsunamis, Volcanoes, Avalanches, Land-
slides, Floods (2nd Edition). Springer-Verlag,
New York, 328 pp.
Boyce, J.K. (1990)Birth of a megaproject: political
economy of flood control in Bangladesh.
Environmental Management 14(4), 419-428.
Bunin, J. (1989) Incorporating ecological concerns
into the IDNDR. Natural Hazards Observer
14(2), 4-5.
Burton, I. and R.W. Kates (1964) The perception
of natural hazards in resource management.
Natural Resources lournal 3(3), 412-441.
Natural Disasters: A Framework for Research and Teaching
Burton, I., R.W. Kates and G.F. White (1978) The
Environment as Hazard. Oxford University
Press, New York, 240 pp.
Chandler, T.J., R.U. Cooke and I. Douglas (1976)
Physical problems of the urban environment.
Geographical journal 142(1), 57-80.
Church, J.S. (1974) The Buffalo Creek disaster:
the extent and range of emotional and
behavioural problems. Omega 5, 61-63.
Cuny, F.C. (1983) Disasters and Development.
Oxford University Press, New York, 278 pp.
Davis, I. (1978) Shelter After Disaster. Oxford
Polytechnic Press, Oxford, 127 pp.
Drabek, T.E. (1986) Human System Response to
Disaster: An Inventory of Sociological Findings.
Springer-Verlag, New York, 509 pp.
Drabek, T.E. (1989) Disasters as non-routine
social problems. International journal of Mass
Emergencies and Disasters 7(3), 253-264.
D’Souza, F. and J. Crisp (1985) The Refugee
Dilemma. Report no. 43, Minority Rights
Group, London, 20 pp.
Dudasik, S. (1982) Unanticipated repercussions
of international disaster relief. Disasters 6(1),
31-37.
Dynes, R.R. (1990) Social concerns and the
IDNDR. Natural Hazards Observer 14(3), 7.
Dynes, R.R., B. De Marchi and C. Pelanda, eds
(1987) Sociology of Disasters: Contribution of
Sociology to Disaster Research. Franco Angeli,
Milan.
Foster, H.D. (1976) Assessing disaster magni-
tude: a social science approach. The Pro-
fessional Geographer 28(3), 241-247.
Foster, H.D. (1980) Disaster Planning: The Preser-
vation of Life and Property. Springer-Verlag,
New York, 275 pp.
Fritz, C.E. (1957)Convergencebehaviour in disasters:
a problem in social control. Special Report,
National Academy of Science-National
Research Council, Washington, D.C., 102 pp.
Glass, A.J. (1970) The psychological aspects of
emergency situations. In Abram, H.S., ed.,
Psychological Aspects of Stress. Charles C.
Thomas Co., Springfield, Illinois, 62-69.
Hewitt, K. (1970) Probablistic approaches to
discrete natural events: a review and theor-
etical discussion. Economic Geography Supple-
ment 46(2), 332-349.
Hewitt, K. (1983) The idea of calamity in a
technocratic age. In Hewitt, K., ed., Interpre-
tations of Calamity. Unwin-Hyman, London,
225
3-32.
Holland, G.L. (1989) Observations on the Inter-
national Decade for Natural Disaster Reduc-
tion, Natural Hazards 2(1), 77-82.
Holt-Jensen, A. (1988) Geography: History and
Concepts (second edition). Paul Chapman
Publishing, London, 186 pp.
Housner, G.W. (1989) An International Decade
for Natural Disaster Reduction: 1990-2000.
Natural Hazards 2(1), 45-75.
Kates, R.W. (1971) Natural hazards in human
ecological perspective: hypothesis and model.
Economic Geography 47(3), 438-451.
Kates, R.W. and D. Pijawka (1977) From rubble
to monument: the pace of reconstruction. In
Haas, J.E., R.W. Kates and M.J. Bowden,
eds., Disaster and Reconstrttction, M.I.T. Press,
Cambridge, Massachusetts, 1-23.
Kreps, G.A. (1989) Future directions in disaster
research: the role of taxonomy. International
Journal of Mass Emergencies and Disasters 7(3),
215-241,
Lechat, M.F. (1990) The International Decade for
Natural Disaster Reduction: background and
objectives. Disasters 14(1), 1-6.
Leighton, F.B. (1976) Urban landslides: targets
for land-use planning in California. In Coates,
D.R., ed., Urban Geomorphology. Special Paper
no. 174, Geological Society of America,
Boulder, Colorado, 37-60.
Manni, C. and S.I. Magalini, eds (1985) Emer-
gency and Disaster Medicine. Springer-Verlag,
New York, 511 pp.
Micklin, M. (1973) Population, Enuironment and
Social Organization. Dryden Press, Hinsdale,
Illinois.
Mileti, D.S. (1980) Human adjustment to the risk
of environmental extremes. Sociology and Social
Research 64(3), 327-347.
Mitchell, J.K. (1974) Natural hazards research.
In Manners, I. and M. Mikesell, eds, Perspec-
tives on Environment, Publication no. 13,
Association of American Geographers,
Washington, D.C., 311-341.
Mitchell, J.K. (1988) Confronting natural disas-
ters: an International Decade for Natural
Hazard Reduction. Environment 30(2),25-29.
Mitchell, J.K., N. Devine and K. Jagger (1989) A
contextual model of natural hazard. Geo-
graphical Review 79(4), 391-409.
Montz, B.E. (1982) The effect of location in the
adoption of hazard mitigation measures.
DISASTERS VOLUME 15 NUMBER 3
226
Professional Geographer 34(4), 416-423.
Oaks, S.D. and S.O. Bender (1990) Everyday life:
opportunities for integration during the
Decade for Natural Disaster Reduction.
Natural Hazards 3(1), 87-89.
Oliver-Smith, A. (1990) Post-disaster housing
reconstruction and social inequality: a chal-
lenge to policy and practice. Disasters 14(1),
7-19.
Olson, R.S. and D.C. Nilson (1982) Public policy
analysis and hazards research: natural com-
plements. Social Science Journal 19(1), 89-103.
Palm, R. (1990) Natural Hazards: A n lntegrative
Framework for Research and Planning. Johns
Hopkins University Press, Baltimore,
Maryland, 184 pp.
Parker, D.J. and D.M. Harding (1979) Natural
hazard evaluation, perception and adjust-
ment. Geography 64(4), 307-316.
Preston, V., S.M. Taylor and D.C. Hodge (1983)
Adjustment to natural and technological
hazards: a study of an urban residential com-
munity. Enzlironment and Behaviour 15(2),
143-164.
Quarantelli, E.L., ed. (1978) Disasters: Theory and
Research. Sage, Beverly Hills, California,
282 pp.
Rivers, J.P.W. (1982) Women and children last:
a n essay on sex discrimination in disasters.
Disasters 6(4), 256-267.
Seaman, J., S. Leivesley and C. Hogg (1984)
Epidemiology of Natural Disasters (Contributions
to Epidemiology and Biostatistics, vol. 5). S .
Karger, Basel, 177 pp.
Sims, J.H. and D.D. Baumann (1983)Educational
programmes and human response to natural
hazards. Environment and Behaviour 15(2),
165-189.
Smith, J.V. (1985) Protecting the human race
against natural hazards (asteroids, comets,
volcanoes, earthquakes). Geology 13, 675-678.
Steinbrugge, K.V. (1982) Earthquakes, Volcanoes
and Tsunamis: An Anatomy of Hazards. Skandia
America Corporation, New York, 392 pp.
Stoddard, E.R. (1968) Conceptual Models of Human
Behauiour in Disaster. Texas Western Press, El
Paso, Texas, 53 pp.
DISASTERS VOLUME 15 NUMBER 3
David Alexander
Torry, W.I. (1979a) Anthropological studies in
hazardous environments: past trends and
new horizons. Current Anthropology 20(3),
517-540.
Torry, W.I. (1979b) Hazards, hazes and holes:
a critique of The Environment as Hazard and
general reflections on disaster research.
Canadian Geographer 23(4), 368-383.
Turner, B.A. (1976) The development of
disasters: a sequence model for the analysis
of the origin of disasters. The Sociological
Reuiew 24(4), 753-774.
Turner, B.A. (1979) The social aetiology of
disasters. Disasters 3(1), 53-59.
U. S , National Research Council (1987) Con-
fronting Natural Disasters: A n lnternational
Decade for Natural Hazard Reduction. National
Academy Press, Washington, D.C., 60 pp.
U. S. National Research Council (1989) Reducing
Disasters’ Toll: the United States Decade for
Natural Disaster Reduction. National Academy
Press, Washington, D.C., 40 pp.
Waddell, E. (1977) The hazards of scientism: a
review article. Human Ecology 5(1), 69-76.
Wallace, A.F.C. (1956) Human Behauiour in
Extreme Situations. Disaster Study no. 1,
Disaster Research Group, National Academy
of Sciences. Washington, D.C., 35 pp.
White, G.F. (1945) Human Adjustment to Floods:
A Geographical Approach to the Flood Probletti in
the United States. Research Paper no. 29,
Department of Geography, University of
Chicago, 225 pp.
White, G.F. (1973) Natural hazards research. In
Chorley, R.J., ed., Directions in Geography,
Methuen, London, 193-216.
David Alexander
D e p a r t m e n t of Geology a n d
Geography
University of Massachusetts
Amherst, M A 01003
U.S.A.