Vous êtes sur la page 1sur 26

A Structural Theory of Aggression

Author(s): Johan Galtung

Source: Journal of Peace Research, Vol. 1, No. 2 (1964), pp. 95-119
Published by: Sage Publications, Ltd.
Stable URL: http://www.jstor.org/stable/423250 .
Accessed: 30/08/2013 13:22

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Sage Publications, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to Journal of Peace


This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions

Peace Research Institute - Oslo

1. Introduction aggression. For the UNESCO, war starts

This theoretical essay is concerned with 'in the minds of men' and for the FAO,
the conditions of aggression. We shall 'in the empty barrel'. The former theory
define aggression somewhat vaguely as has found no support in contemporary
'drives towards change, even against the psychology if the word 'starts' is taken in
will of others'.l The extreme forms of this the sense of 'always originates'. But what-
phenomenon are crimes, including homi- ever the external conditions leading to
cide, between individuals; revolutions, in- aggression are, they probably have to pass
cluding elimination, between groups; and through the minds of men2 and precipitate
wars, including genocide, between nations. as perceptions with a high emotive content
These forms make aggression negative and before they are acted out as aggression.
problematic, a cause of concern and pre- It should also be noticed that what psychol-
vention. But one can also turn the coin ogy can say today is not that war cannot
and look at the other face: aggression as originate in the minds of men, but that
the driving force in history, as the motiva- there is no necessity why it should do so.
tional energy that moves mountains. Hostile aggression is no inseparable part
However, we shall be mainly concerned of the innate structure of the 'minds of
with aggression in its extreme forms where men', but added to it from the outside,
it becomes a drive to hurt and harm e.g. through special socialization processes.
others because they stand in the way of Lord Boyd Orr's dictum about hunger
one's own self-assertion, and not look at serves, like the dictum about the minds
the good causes this may serve in the of men, to direct attention to a possible
aggressors' own minds. Aggression in this side-effect of the worthy activities of
sense is pervasive, important and cata- the UNESCO and the FAO. To spread
strophic, with modern technology as a knowledge and food are positive activities
multiplier. It should be studied at its in themselves - if in addition they can be
roots, at the very points where it emerges. shown to have a war-preventive effect they
The postwar period has seen relatively may be even more hallowed. But the
little research into this problem of origin, number of examples that will come to
and much research into how to control everybody's minds of hungry individuals
aggression, for instance the tremendous and nations that have not aggressed against
effort that has gone into arms control and anybody, not to mention the myriad of
disarmament research. It is as if the apoca- individuals and nations with full stomachs
lyptic display of aggression during World and barrels in store who have, makes one
War II has had a paralyzing effect. But at wonder if the correlation between hunger
the slogan level we find two of the United and aggression, if positive at all (which we
Nations' specialized agencies sometimes doubt) is high enough to be of much
making use of rather simplistic theories of significance.

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
The difficulty with these theories, like these three interpretations, and only re-
the Dollard frustration theory, is that they turn to them at length later on.
are non-structural; they do not take the Even under these very general condi-
social context sufficiently into considera- tions it seems difficult to find counter-
tion. The first theory sees man as a self- examples to the following: 1) There will
sufficient unit; he is his own cause. It does be a division of labor in the sense that the
not correspond to our feeling that position elements will not carry out the same tasks
in the social structure does matter. Granted all the time; 2) The elements will tend to
the importance of individual characteris- be rankedaccording to a number of criteria
tics, it is at least as unlikely that the evaluating their position in the system;
chance of aggression for a given individual and 3) The relative position of the ele-
is independent of social position as it is ments according to these criteria will have
that all individuals in the same position a certain stability. All we are saying by this
should display exactly the same tendency is that stratification seems to be a universal
towards aggression. phenomenon. The distance from 'high' to
The second theory locates aggression at 'low' can be reduced, the consequences of
the places in the social structure where stratification can be alleviated, but it can-
hunger is found; in that sense it is a not be declared to be non-existent. If one
structural theory. Its sufficiency is rather wants an egalitarian society where every-
in terms of the very limited aspect of the body has the same rank this must be
social structure it considers. Hunger may arrived at by such techniques as making
supply motivational energy, but it must an element that is low in one context high
be combined with a position in the social in another context (compensation) or
structure that gives a view to better life letting individuals who have high ranks in
situations and resources. This is where the one period have low ranks in the next
minds of men enter: a theory of aggression period (rotation). It cannot be done by
should combine the idea of frustration with abolishing differential ranking as such.
the idea of perceiving aggression as a For as long as there is interaction there
possible way out of the frustrating situa- will tend to develop a certain cultural
tion. similarity, and as soon as this is the case
We now turn to the construction of a the element that has more of, or is closer
theory of that kind. to, or is more in agreement with the values
of the system will rise high, and the ele-
2. The hypothesis about rank-disequilibrium ments that have little of these values and
Imagine that we have a system of ele- seem to be far from realizing them -
ments that are actors in the traditional wthether it is might and glory, power, in-
social science sense of having goals and telligence, money, beauty, health - will
being capable of directing actions towards stay low in the system.
them, and that they interact with each Thus, an interaction system is a multi-
other.3 Concretely, we are thinking of dimensional system of stratification, where
three such types of systems: the system of those who have and those who have not,
individuals found in a group and particu- those who have more and those who have
larly in a nation; the system of groups less, find, are given, or are forced into
found in a nation if the groups are so their positions. For the sake of simplicity,
homogenized and organized that they can let us deal with these criteria of rank in
be seen as actors, and the nations in an terms of two positions only - high and
international system. However, we shall low. We shall refer to them as topdog and
prefer to proceed with the general theory underdog positions (T and U). Thus, an
for a while without making references to element in a system with five rank criteria,

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
will have a profile, say TUTTU, the were also efforts made by nations and indi-
interpretation of which depends on what viduals close to the top, but short on one
kind of system and what kind of dimen- or two dimensions, to achieve the con-
sions we are referring to. It may, for figuration of the complete topdog.
instance, stand for 'high on power, low on Then it may be argued that the com-
income, high on occupation, high on edu- plete topdog is always in a state of aggres-
cation, low on ancestry' (for individuals or sion relative to, for instance, the complete
groups) or 'high on military power, low underdog. The reward structure of society
on income per capita, high on industri- is a built-in transfer of value from under-
alization, high on educational level, low dog to topdog where the latter inevitably
on past glory' (for nations), and one may gets more than his due through a process
discuss how likely the configuration is. of accumulation. Where there is stratifi-
But two configurations are beyond doubt: cation there will also be exploitation. But
the complete topdog. TTTTT, and the however important this may be, it is not
complete underdog, UUUUU, are both aggression in our sense. Aggression in-
well-known occurrences in any social volves a desire for change, and as long as
system, individual or national. We shall the topdogs are only soaking from the
refer to these two as 'equilibrated posi- structure what is their institutional 'right'
tions', since the ranks of the elements in this cannot be defined as aggression (but
these positions are in equilibrium with possibly as exploitation). They may en-
each other; they are equivalents. counter aggression from positions lower
With five dimensions and two positions on down in the structure with more aggres-
each there are 32 possible configurations, or sion, but that is reactive aggression which
in general 2" combinations when n dimen- again is different from what we are in-
sions are used in the analysis. The theoreti- vestigating: the points where aggression
cal problem is now: where in the system,for can be studied in statu nascendi.
what social types,is aggressionmost likely to accu- A number of authors have argued that
mulate and express itself? For common sense this is not where the complete underdog is
as well as social experience make us doubt located. For imagine that it was: in that
that aggression is randomly distributed on case feudal structures like the slave society4
the configurations or social positions. or classical Indian caste society5 would not
With the conceptual apparatus devel- have been stable, but would have shown
oped so far there are three possible an- a much shorter lifetime. What was typical
swers: aggression will mainly come from of these societies was not so much the
the elements equilibrated at the top (the differences between high and low - those
complete topdog), mainly from the ele- differences are also found in 'modern'
ments equilibrated at the bottom (the social structures - but the almost com-
complete underdog), or mainly from the plete correlation between the rank-dimen-
elements in rank-disequilibrium, i.e. the sions. If a person was high in one context
elements with some positions high and relative to another person, this was also
some positions low. We shall examine the the case in a number of other contexts.
three possibilities. Nevertheless, or precisely for that reason,
The complete topdog has already ob- the structures did exhibit a remarkable
tained what the system has to offer in terms stability - until, we think, some mobility
of rewards, but this by itself will not pre- was permitted so that disequilibrated posi-
vent 'much from wanting more'. Colonial tions appeared.
wars and slavery were typical examples of James C. Davies6 has summarized much
aggression from the top, although closer of the thinking around the social basis of
analysis perhaps may reveal that they revolutions:

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
On the basis of de Tocqueville and an amelioration of the situation :12 'Nations
Marx, we can choose one of these that have endured patiently and almost un-
ideas or the other, which makes it
hard to decide just when revolutions consciously the most overwhelming oppres-
are more likely to occur - when there sion often burst into rebellion against the
has been social and economic progress yoke the moment it begins to grow lighter.'
or when there has been regress. It De Tocqueville would have found solid
appears that both ideas have explan- support for his thesis in the present Negro
atory and possibly predictive value,
if they are juxtaposed and put in the revolt in the United States; it seems to con-
proper time sequence.7 firm his thesis both in space (particularly in
the North) and time (increasing rapidly
The theories of Marx, de Tocqueville and with increasing desegregation).
Davies are all dynamic. Marx located the Davies combines the two perspectives
source of revolution low down in the and produces a neat little theory where
society, in the proletariat, and predicted revolutions are supposed to occur when a
revolutions after a period of deterioration population is exposed to a de Tocqueville-
and exploitation. Davies uses the study by effect followed by a Marx-effect, i.e., an
Zawadzki and Lazarsfeld8 against this improvement accompanied by a revolution
view, and uses this description of the of rising expectations followed by a crisis
situation of the complete underdog:9 and deterioration in the level of need
... preoccupation with physical
satisfaction so that one gets 'an intolerable
survival, even in industrial areas, is gap between what people want and what
a force strongly militating against they get'. That is when the revolutions
the establishment of the community- occur, and he gives a number of well-
sense and consensus on joint political reasoned illustrations but no statistical
action which are necessary to induce
a revolutionary state of mind. Far data to corroborate his thesis.
from making people into revolution- However good these theories, singly or
aries, enduring poverty makes for in Davies's juxtaposition, are in predicting
concern with one's solitary self or when, they do not predict or explain where
solitary family at best and resignation in the social structure the revolutions or
or mute despair at worst.
other activities of aggression are likely to
This theme recurs very often in studies of arise. Davies's answer might be: exactly in
periods or places under depression,10 or the positions exposed to the Davies-effect.
in the concentration camp studies.11Most But it would hardly be difficult to show
of these studies focus on the moral or that the complete underdog has often been
mental state of the complete underdog, exposed to the ups and downs caused by
however, or the low level of social organi- economic fluctuations and changing styles
zation, and often forget the much simpler of exploitation without resorting to any
and more obvious factor that the underdog violence, whereas other sectors of the
is deprived of the resources that make society may be extremely sensitive to the
revolutions possible: ideas, visions, ac- slightest downward trend. The Davies
quaintances, weapons, social experience, theory may locate revolutions in time, like
empathy, courage necessary to imagine the Dollard theory, but neither theory
oneself as a ruler, etc. What the underdog locates them in social space.
does not have he can get, but he gets it, We then turn to our hypothesis which
we believe, precisely by changing one of is very simple:
his U statuses into a T status and con- Aggression is most likely to arise in social
verting that status into a resource for the positions in rank-disequilibrium.In a system of
dissolution of his disequilibrium. individuals it may take the form of crime, in a
To de Tocqueville aggressionwas tied to system of groups theform of revolutions,and in

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
a system of nations theform of war. But these 3. The theory about how disequilibrium works
extreme forms of aggression are unlikely to The thesis is very simple and the theory
occur unless 1) other means of equilibration to- behind it is also simple. It rests on a
wards a completetopdog configurationhave been comparison between the social situation of,
tried, and 2) the culture has some practice in say, a TU and a UU, in our terminology.
violent aggression.l3 There are three such differences that seem
We shall now present extensive com- to be decisive in this context.
ments on this hypothesis. Thus, to take up
one point immediately: in another paperl4 a. Disequilibrium means differential treatment.
we have discussed this problem from a We have assumed that rank matters in
different angle. In a world consisting the sense that the elements are treated
mainly of complete topdog and complete according to their rank. An element in a
underdog positions there will be nobody TU position will be constantly reminded
present to bridge the gap between the two. of his objective state of disequilibrium by
Hence, a conflict between the TTTTT the differential treatment he is exposed to.
and the UUUUU, would not be alleviated This will force a correspondence between
by the presence of mixed combinations his objective situation and his subjective
that could serve partly as communication perception of it - unless he cuts out inter-
bridges, partly as a reservoir of neutrals action in one direction or the other. If
who could be brought in as mediators and he does so, he is actually living in a world
in general dampen the conflict (the clas- with only one rank-dimension. But if he
sical theory of 'criss-cross'). A short- does not, disequilibrium will be a part of
coming of this theory is its neglect of ag- his phenomenological existence, and the
gressive needs located in the unbalanced idea of rectification may occur. However,
combinations. Thus, the theory we are we do not have to presuppose, for the sake
presenting here is in a sense complemen- of the theory, that an ideology of rectifi-
tary to the other theory: for the classical cation has to be fully developed, or even
criss-cross theory, disequilibrated combi- perceived at all - only that the objective
nations will fulfill an aggression-binding existence of disequilibrium will cause an
function whereas for the present structural instability in the life-style of the person or
theory of aggression disequilibrium is a the nation, and cause what is often referred
reservoir of aggression. to as an 'unstable self-image'.
Since the predictions that follow from In more sociological terms, the crux of
these two theories are contradictory there the matter is the high probability that the
is an obvious need for a meta-theory. Just disequilibrated TU will use TT as his
to indicate one possible line of thought: reference group even if UU is his member-
it is possible that the criss-cross theory ship group, whereas a complete underdog,
presupposes 1) a high degree of system UU, may not even dare to think in terms
integration in order for the communication of TT as a reference group; the complete
and mediation effects to be present and topdog will be beyond his imagination.
2) good chances for the disequilibrated The absolute deprivation of the UU may
combinations to become equilibrated in a be higher, but the TU has relative de-
legitimate way. If the degree of system privation built into his position. The
integration is low, as in the international destabilizing effect of this discrepancy will
system, and many mobility channels are produce a mobility pressure, and the thesis
blocked, the aggression effect will probably is then that if there are no open channels
predominate over the criss-cross effect. of mobility, rectification of the disequi-
We suggest this as a fascinating topic of librium will be carried out by other means.
future theoretical and empirical research. In this process two other aspects of the

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
disequilibrium situation are of major righteousness. In the kind of achievement-
importance. oriented world in which we live, claims
will be based on achievement rather than
b. Disequilibrium means resources.We have on lack of achievement - in the latter case
commented on the effects on the TU of they are usually made explicit by others
having one foot in either camp, for in- and the Welfare State is an example of
stance, of being white and poor, as nation the institutionalization of rank-compensa-
or as individual. Obviously, the position tion.
in the top camp not only creates the moti- There are probably countless examples
vation towards equilibration, but also some of this kind of self-righteousnessbased on
of the resources that will come in handy disequilibrium. One thinks of such cases
in the struggle. The nouveaurichemay be as nations trying to equilibrate crowded
green, but he is nevertheless rich and the territory to economic rank achieved by
stories about him are about his efforts to means of rapid industrialization (Ger-
obtain balance by converting money into many, Japan) or to equilibrate territory
culture and prestige. The small, over- to rank derived from a glorious past
crowded but economically developed na- (Italy). In all these cases very explicit
tion has in its economic potential the possi- references were made to the perceived
bility for conversion into military power disequilibrium - it became a part of
with subsequent territorial expansion. national ideology, so to speak. Or one may
The member of the intellectual proletariat think of real or fictional casesof people who
may be low on almost everything, and yet have felt that their high rank on one di-
have in his intellectual maturity, knowl- mension has entitled them to deviance,
edge and academic discipline, invaluable that it has made legitimate what would
tools that can be converted into power, have been illegitimate to the complete
high-ranking occupations and income. underdog. Thus, we would imagine that
And the 'white' but 'poor' nation may to many juvenile delinquents the world
draw on its kinship with white and rich comprises a bitter discrepancy between
nations to gain influence and recognition, high ranks in terms of some resources
as is the case with many Latin-American they possess, such as intelligence or mus-
nations. cular power or initiative, and low rank on
more institutionalized dimensions such as
c. Disequilibriummeansself-righteousness.Our age, income, position, education.l5 Ob-
culture seems to be more dominated by viously, one would have to use what soci-
themes of balance and adequacy than by ologists refer to as 'informal status' in addi-
the theme of compensation. We have no tion to institutionalized statuses to come
empirical backing for this, but there seem closer to a sufficient explanatory basis.
to be more cases of people and nations To summarize: it is socially guaranteed,
saying 'considering our high rank on X by the very structure of the system, that
it is right and proper that we should also the disequilibrated is never left in peace
have a high rank on Y, because that cor- with his disequilibrium unless he cuts out
responds to what is due to us' than of and closes down some interaction chan-
people and nations saying 'considering my nels. In this unstable situation he has both
low ranks everywvhereI think I am entitled the resources and the inner justification
to some compensation on at least one needed for acts of deviance. Nevertheless,
dimension'. Claims must be justified not we do not hypothesize aggression unless
only in the eyes of the others but also in 1) other means of rectification have been
the eyes of the claimants themselves; they tried and 2) the culture has some practice
must feel they are right to the point of self- in aggression.

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
4. MAoreabout rank-dimen^^.nsand possibilities. The units may be positively
disequilibria coupledin the sense that if one rises so does
The time has now come to be more speci- the other, and that they will also follow
fic about the rank-dimensions. These are each other on the way down. Then, again,
concerned with the most crucial things of they may be negatively coupled: the rise of
life, the matters for which people live and one is the fall of the other.'7 And, thirdly,
die.'6 But they differ tremendously, and they may not be coupled at all. Different
should be analyzed from at least three economic systems provide examples of all
angles: 1) is it possible, both in theory three. Concretely, the most dramatic ex-
and in practice, for one element to change ample of negative coupling in the case of
position at all, 2) how does the change of nations is the dimension of area: one
one element affect the position of the other nation's gain will have to be the loss of one
elements, and 3) what kinds of disequi- or more other nations unless the game can
libria are most important? be made 'variable-sum' through the ex-
Where the first problem is concerned ploration of outer space (or as it was in
the traditional distinction made in sociol- the period of the great discoveries). And
ogy between ascribed and achieved di- the same applies to property at any given
mensions is useful. The ascribed position point of time: one person's loss is somebody
is known at birth, the achieved position is else's gain, unless the game is changed
what the individual himself makes out of through such factors as destruction or crea-
his life-situation. However, this conceals tion of property or is stretched out in time.
the important distinction between dimen- Correspondingly, the dimensionsmay be
sions that are indelible in the sense that the coupled positively, negatively, or not at all:
element cannot escape from it (age, sex, the rising of a unit on one dimension may
race, primogeniture, family, ethnicity) and imply its rising on another dimension, or
perhaps even visible (age, sex, race) - and its fall, or it may imply nothing at all.
the 'delible' dimensions like nationhood More age means more power - at least
or ecological background which the indi- the power to influence elections through
vidual can move away from - even voting, and so on.
though they are known at his birth. In general it is obvious that dimensions
Achievement may also be so conditioned with negative coupling between units are
by the matrix of ascribed dimensions, as in the dangerous ones, and this may lead
most societies today (skill is not known at one to speculate how many they actually
birth, but the possibility or impossibility are. With the concept of a nation tied to
of demonstrating it if it exists is known), the idea of a territory, usually contiguous,
that mobility becomes illusory except for international politics is dangerous pre-
persons with the particular mixture of cisely because of the simple metric quality
good and evil that makes for mobility. of that dimension. Today national identity
The most aggression-provoking case is is linked to territory in a rigid way - but
probably the half-open dimension of un- one can think away that condition even
fulfilled promises, but the completely though only the Jews and some others
closed channel will also serve to accumu- have had long experience in territory-free
late aggression unless all channels are national existence. Thus, one could ima-
closed. gine each nation as consisting of a non-
The second problem is more interesting. negotiable geographical core surrounded
Imagine a system with two elements and by a periphery that could be exchanged
one dimension only: we are interested in for other goods or even lost with no ir-
how the position of one affects the position reparable damage done to national integ-
of the other. In principle, there are three rity and identity. Or one could imagine

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
a world where all nations were like the Finally, we turn to the problem of
Jewish nation prior to the birth of Israel - distinguishing between disequilibria. To
with a central authority, but no territory- say that disequilibrium matters is not to
where the nations mix much like strata in say that all kinds of disequilibrium matter
a society. equally much. A typology of disequilibria
But even though it may serve the pre- is needed, and we shall discuss this from
vention of aggression to reduce the salience three angles, two of them formal and one
of competitive dimensions and increase the of them substantive.
importance of the cooperative dimensions First of all, there is the obvious dimen-
an important caveat should be inserted sion of degree of disequilibrium. Imagine
here. World economy can probably be that we introduce a 'middledog', call him
structured in such a way that wealth be- M. We define 'degree' of disequilibrium
comes a cooperative dimension so that no as internal distance between the ranks,
unit's loss can be another unit's gain (if in which would make TTM less disequili-
no other way, by the simple method of brated than TTU. Generally, the ex-
taxation of such gains). But even if na- perience based on data from persons in
tions rise together on a dimension of disequilibrium tend to show that the
prosperity there is one aspect of this dimen- effects of disequilibrium do not show up
sion that will remain competitive forever: unless there is a considerable amount of
not absolute prosperity, but relative pros- disequilibrium present.18The consequence
perity. One nation's gain of the no. 1 is a tendency towards J-shaped relation-
position, regardless of the absolute value, ships: deviation tendencies stay at the base
is another nation's loss. And this gives rise level for equilibrium and low degrees of
to a major question that can only be disequilibrium, and then rise quickly for
decided on empirically. What matters high degrees of disequilibrium.
most, absolute or relative position? What Secondly, there is the problem of dis-
is most important to a competitive pupil, equilibrium profile. The sum of internal
to improve his grades or his relative posi- distances in the combinations TTU, UUT
tion? If he improves his grades but slides and TMU are the same, viz., 4, so we
back in relative rank, how does he look would expect more of an effect than for
at the net balance of his achievement? the combination TTM where the sum is 2.
And what about the nation accustomed But do we expect the sameeffect with this
to a no. 1 position in military power that crude measure of internal distance? A
is bypassed by another nation even if it has priori, one might argue that TTU is more
also itself increased its military capability ? desperate about his low status and UUT
Probably absolute positions are of para- more proud about his high status, so that
mount importance only in systems with the former is the more aggressive, and then
little or no interaction or in extreme cases one might argue just the other way
(e.g. below subsistence level). With a round.19 One might say that TMU is
high level of interaction we know of exposed to a particularly deviation-gener-
nothing in social science literature to dis- ating mixture, and one may say that his
prove the idea that concern about relative aggression will be neutralized precisely
position on a rank-dimension will increase, because of the complex social structure in
and since increase in interaction is a wthich he finds himself. In other words,
general trend in the world community this one should leave this problem to the data;
should make dimensions more, not less their richness, provided they are good,
competitive, and conflicts for the reason will probably by far outdo even a good
we are discussing more, not less probable theoretical imagination.
in the future. Thirdly, there is the problem of which

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
dimensions. Obviously, not all disequi- against the attacks of the overachiever.
libria even with the same profile will have Thus, the two will be pitted against each
the same deviation-generating effect. The other, probably with the strongest aggres-
effect will depend on the salience of the sion potential for any possible pair of
dimensions and of the disequilibrium. combinations. In more concrete terms, the
There is also the important suggestion underachiever may possibly be identified
made by Jackson20 that the achievedv. as a member of the extreme right and the
ascribeddistinction may be used here. Let overachiever as a member of the extreme
us compare these two patterns with the left;22 the former will keep his privileges
same internal distance, based on three since that is the high basis he has in the
dimensions: system, the latter will deny the system of
Dimension: ascribed achieved ascribed privileges or change them to his own ad-
Pattern 1 high low high vantage. Thus, we disagree with Jackson
Pattern 2 low high low that outwardly directed aggression should
necessarily be less probable for the under-
Of the three mechanisms we have men- achiever, only that it may come as a re-
tioned in the section on 'how disequilibrium action to the aggression of the overachiever
works' this distinction should affect the and be less spontaneous.
third mechanism, the norm about justice, It is interesting in this connection to
in particular. The unit with the second compare the state of the Negroes in the
pattern is an overachiever relative to his USA and in Brazil. Comparisons are very
ascribed case; the unit in the first pattern often to the effect that there is less dis-
is an underachiever, and the overachiever crimination or prejudice in Brazil, but
more than the underachiever will feel that statistical data do not appear to demon-
he deserves a fairer deal, at least in an strate this. Rather, the impression is that
achievement-oriented culture. there are more Negroes in higher positions
According to Jackson one might predict in the USA than in Brazil. We shall only
extrapunitivity for the overachiever and suggest an interpretation in terms of the
intrapunitivity for the underachiever;21 present theory. If we have a nation where
the overachiever will blame society for race and position are equilibrated we
constraining him, the underachiever will would predict a very low level of aggres-
blame himself for doing less than society sion, and hence more of a tendency to ex-
or the system might expect from him. But press oneself in accordance with the pre-
the overachiever is in the difficult situation dominant ideology of our time, the ide-
that he is low on ascribed dimensions ology of racial equality. Thus, there will
where mobility, by definition, is impos- be no laws prescribing segregation and
sible. Hence, he will have to fight like the little overt prejudice, for rank-equilibrium
educated Negro, not for white status, but is built into the social structure.
for the elimination of race as a rank-di- On the other hand, imagine a nation
mension. Thus, outward aggression takes where many Negroes are high in social
a form other than simple fight for mo- position and many whites are low. In that
bility and scarce value; it may be a fight case we would predict formal and informal
about the definition of value. barriers to secure segregation, and as a
The underachiever, e.g. the white high- barrier against aggression resulting from
class who is low on education or income, the disequilibrium. In fact, the way to
might be more motivated to direct ag- obtain 'racial equality' will be through the
gression to promote his own mobility. But suppression of race as a dimension at all -
just as likely is aggression due to a desire and here as elsewhere the role of ideology
to keep what he has - perhaps precisely and perception is probably tremendous.

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
Thus, to the extent that these models are such concepts may be meaningful and
approximations to Brazilian and US reality even clear rank-connotations in some
the USA seems to be much further along periods and not in others. One procedure
the road towards racial equality than for finding which rank-dimensions to use
Brazil, only she is in a transitional period would be to peruse the writings and
with tremendous potentials of aggression speeches of leading statesmen to see what
that Brazil has not yet really entered. The dimensions they make reference to, and
revolt of the Negroes in the USA is like more particularly what disequilibria they
the revolt of the colonies in this age of make explicit reference to. Another proce-
anti-colonialism: these are both efforts to dure for the contemporary world would
eliminate dimensions, by making all citi- be to make use of opinion polls where a
zens 'first class', or all nations 'indepen- sample could be introduced to a large
dent'. variety of rank-criteria and asked to rank
the criteria in terms of salience (for in-
5. Methodology stance by means of paired comparisons or
The time has now come to look at the the technique of double alternatives). One
thesis from a methodological point of view. could then pay particular attention to
We shall discuss first procedures for testing answers given by the elite.
the hypothesis, secondly some other pro- As to aggressivenessthere is a continuum
perties of the theory. from the idea of changein the structure of
The hypothesis relates two variables the system via the overt expression and
that describe elements: degree of rank- propagation of such ideas and actions de-
disequilibrium on the one hand and degree signed to change the social structure, to
of aggressivenesson the other. For individ- real aggression, i.e. efforts to pursue such
uals the former is relatively easily meas- ideas even against the will of other ele-
ured. A considerable part of social science ments - and even at the expense of other
ingenuity has been invested in the devel- elements. The extremes are, as mentioned,
opment of good indicators of basic rank- called crimein the individual case, revolution
dimensions. Most authors that have made or internalwar in the group case, and external
use of this kind of thinking have trichoto- war in the case of nations.
mized the dimensions in 'high', 'medium' This means that there is a large variety
and 'low', tabulated them against each of indicators. So far the only thing that
other and used the distance from the has been explored systematically in the
diagonal of agreement as the measure of literature has been the individual case
disequilibrium. One can also use rank- with attitudinal indicators, and the hypo-
order and perhaps get a more sensitive thesis has been well confirmed.24 The
measure, but it is only meaningful in logically next step would be to use be-
relatively limited systems with a high havioral indicators. Thus, the theory
interaction pitch. would predict disproportionately great
For nations it should be possible to do rank-disequilibrium among radical politi-
the same. Increasingly good statistics are cians, among criminals (especially those
available for most dimensions of interest.23 who are engaged in crimes against pro-
Data are actually likely to be too good, perty), among leading revolutionaries and
not too bad - since only gross discrimi- leaders of change-oriented groups in gene-
nations are necessary to test the theory. ral, and among very bellicose nations.
But the difficulty lies in finding which For nations attitudinal indicators could
rank-dimensions to use. Obviously, this be based on national polls and aggregate
will vary with time and place. Christian measures, or on elite polls, or on what the
or not, monarchy or not, socialist or not; elite has expressed without being solicited

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
by social scientists, e.g. in articles and the theory does not predict overt aggres-
speeches. Thus, it should be possible to sion as an immediate outcome of disequi-
test the idea that there is a disproportion- librium (other alternatives are tried out
ately great desire for change in disequilib- first), the test of the hypothesis should
rated nations. And the behavioral indica- provide for a time lag between disequi-
tors should not be too difficult to find with librium and aggression, and more so for
the growing experience social science gains the behavioral and extreme forms than
with 'statistics of deadly quarrels'.25 for the attitudinal and more tranquil
A major virtue of the hypothesis is that forms. A great deal of experimentation
it should be valid both synchronically and with time cuts and lags will probably be
diachronically. It should point out the necessary before a pattern can emerge.
likely aggressors at any given point in In this process 'proof' by illustration
time, as was discussed in the introduction. should not be accepted. One can always
But given one element, an individual or a 'prove' by the usual library hunt for ten
nation and its history, the hypothesis should cases in favor and one against (to show
help us locate the periods of aggression in that one is not biased), but we have pre-
that history: the theory is dynamic. Thus, ferred to postpone the presentation of such
there is a piece of common folklore about data till they can be evaluated against a
radicalism and youth belonging together background provided by data amenable
which gains in perspective if we apply this to statistical treatment.
theory: the youth may be at the period One may ask: why not systematically
of institutionalized disequilibrium with study all wars during, for instance, the
education high and most other dimensions last one hundred and fifty years to find in
low, the period of student riots and demon- how many cases the aggressing nation can
strations. This is the age where change be said to have been disequilibrated ?
seems logical, inevitable and desirable and There are two objections to this primitive
the attitude is known to taper off somewhat procedure. First of all, a percentage of
with age. The folklore interpretation is in disequilibration among nations that ag-
terms of growing wisdom, 'he who is not gress says nothing if it cannot be compared
a radical in his youth does not have a meaningfully with the corresponding per-
good heart, but he who is not a conserva- centage for nations that do not aggress.
tive when he becomes older does not have One is not interested in the percentage of
a good brain'. Our interpretation is in belligerent nations that are disequilibrated,
terms of growing equilibrium more than but in the percentage of disequilibrated
growing wisdom, and hence decrease in nations that engage in aggressive acts. But
motivation due to decrease in relative apart from the trivial point about rules
deprivation.26 The condition is that the for percentaging there is another objection:
society permits this equilibration, that the danger of running into a tautology.
there is structural provision for it - other- Given a war and an aggressor it is only
wise the result will be a society with pock- a question of time and research imagina-
ets of accumulated disequilibrium that tion to dig up a rank-disequilibrium, for
may one day burst and yield a revolution. the number of dimensions along which a
For diachronical tests of the theory the nation can be ranked is legion. Thus, tlhe
system could be observed through a period methodology will make the theory non-
and the period divided into time-chunks falsifiable in practice. Undoubtedly, a test
of, say, from one to five years. For each of the theory will require much speculation
time-chunk the elements should be rated about relevant disequilibria, but the
according to degree of disequilibrium and strength of the theory will depend on the
degree of aggressiveness displayed. Since ratio between number of types of disequi-

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
libria and number of aggressive acts they action.Imagine that a disequilibrated per-
can explain. The lower the ratio the more son brings a disequilibrated group to
powerful the theory; if each aggressive act power in a disequilibrated nation. Would
should be explained by its own kind of one not expect the motivational energy
disequilibrium we would still have a dis- stored in these disequilibria to multiply
equilibrium theory, but a far less powerful rather than to add up, and to result in a
one. particularly spectacular form of aggres-
A major advantage of this theory, in sion? Again the case of Hitler's Germany
addition to its dynamic character, is that comes to mind, but also, perhaps, the case
the hypothesis is meaningful across dif- of Castro's Cuba, and in general, new na-
ferent levels of social organization. Indi- tions high on economy and with a more
viduals as well as groups and nations can or less revolutionary leadership.
be ranked in terms of rank-dimensions The methodological implication of this
and aggression, although there will be is not only the necessity of independent
problems in connection with the choice of tests at all levels as pointed out above, but,
indicators.27Thus, the theory is a theory in addition, the need to explore patterns
in general conflictology, not in any of the of causality and interaction.
special fields concerned (social psychology,
sociology, international relations, not to 6. Discussion
mention history). Let us now imagine that there is some-
There are three aspects of this cross- thing to this reasoning, at least in the sta-
level nature of the theory. tistical sense that disequilibrium in rank
First of all, the theory points to a basic accounts for a major part of the variation
isomorphismbetween different levels of in aggressive behavior. We can probably
social organization. Disequilibrium is rela- do this with some justification partly
tively easily identifiable at all levels, and because some excursions into data seem
its consequences are spelt out in the hypo- to corroborate the theory, and partly be-
thesis. The hypothesis is about systems of cause the contrary idea of placing the
interacting units and three interpretations peak of the probability distribution for
are given; to the extent that the hypothesis aggressiveness a priori on the points of
is verified these three levels of organization equilibrium seems to lack theoretical justi-
will have to be isomorphic. fication. The question is whatfollows from
Secondly, there is the possibility of a this, can anything be said in more con-
causal connectionbetween the levels. We crete terms about specific arrangements
shall not explore this, only indicate some of systems, specific structures. With no
possibilities. Disequilibrium at one level value-implication to the effect that aggres-
can lead to disequilibrium at another level: siveness is an absolute evil, it is both theo-
a highly disequilibrated individual may retically and practically important to
become the leader of a completely under- know something about conditions that
dog group and lead it into disequilibrium promote or do not promote aggressiveness.
by giving it power, property or education. We shall try to spell out some implications
The group, in turn, may lead the nation of this kind, mainly for the international
into disequilibrium. But we are more in- system, since this is our major concern.
terested in the cases where disequilibrium
at one level expresses itself as aggression a. The consequencesof economicdevelopment
at another level, as perhaps may be argued Economic development is a major issue
to be a description of the case of Hitler and of our times, for most people on earth
Nazi Germany. probably the major issue. The motives
Thirdly, there is the problem of inter- for giving technical assistance are inter-

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
esting.28 Economic and technical aid are nations are often nations in more or less
defended (against the attacks from those similar stages of economic (and even
who want to give less or nothing) on such social) growth and hence will be competing
diverse grounds as: 1) it buys us friends, for exactly the same goals, the same values,
who (a) will not attack us themselves (the the same markets, the same friends.
direct insurance argument) and (b) will But the argument is rather that econo-
support us if somebody else attacks us (the mic development will deter the poor from
indirect insurance argument); 2) it in- joining together to rob the rich, for when
creases our military power by securing they receive aid they will be deprived of
bases etc. in return; 3) it increases our the motive. The argument rarely specifies
political power by securing UN votes etc. how this is going to happen in detail, and
in return; 4) it increases our economic one reason may be that the idea is prob-
power by tuning the market to our pro- ably wrong. Imagine a world with four
ducts or otherwise; 5) if we do not do it salient rank-dimensions, (or three, or two
the others will do it and gain 1, 2, 3 and/or for that matter) and that aid is given from
4; 6) for humanitarian reasons, all human TTTT nations to UUUU nations, devel-
beings shall be guaranteed a minimum oping some of them into TUUU nations.
standard of living; 7) for reasons of In all probability this would be a more,
justice: there is an equitable distribution not a less dangerous world to live in -
of economic goods that should at least be ceterisparibus. All theoretical reasons men-
approximated to; 8) because it promotes tioned in the beginning of this article, the
world peace. We are concerned with the sense of self-righteousness, the access to
last argument. resources of different kinds, the internal
Some would say it follows directly from strains due to differential treatment in
one or more of the first seven arguments, different interaction contexts would oper-
which may be true but is certainly not ate. As a matter of fact, the following
self-evident. Then it may be said that the development seems much more likely:
transfer of technical economic assistance that a group of TUUU nations join their
requires a kind of international coopera- newly gained forces and resources in
tion and even superstructure that by itself making an organization for revolution in
will be conducive to a safer world com- the international community. Another
munity. But the argument is usually inter- version of this would correspond to a
preted so as to mean something else: pattern for revolutions discussed in the
statesmen talk about the North-South following section: one or a few heavily
conflict being of equal or even greater disequilibrated nations mobilize the com-
importance than the East-West conflict, plete underdog nations against the com-
alluding to the conflict between rich and plete topdog nations according to the eight-
poor nations. This may be true in more point scheme to be presented below.29
than one sense, but it does not follow that In other words, and to put it bluntly:
economic development is the remedy. Of economic development per se will probably
course, it is a tautology to say that in a create more, not less rank-disequilibrium
world where economic value has been and hence be conducive to more, not less
evenly distributed (if such a world is imagin- aggression. It is unnecessary to add that
able) there will be no conflict because this does not imply that it cannot be
of differences in wealth. That leaves us justified on other bases, or that economic
with the conflicts due to similarities in development and technical assistance in
wealth: the desire to be high in a relative particular cannot have peace-building
sense, not only an absolute one, and the functions for other reasons, e.g. by con-
highly important factor that equally rich tributing to international superstructure.

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
b. The conditionsfor revolutions lectuals. Both sides will develop ideologies
According to the theory presented the that make symbiosis less likely, as is so
recipe for a revolution should be relatively easily observed in most countries in Latin
clear. A revolution needs leaders and fol- America. Thus, a climate for the emer-
lowers, and traditionally the leaders seem gence of revolutionary leadership is cre-
to come from somewhere high up in the ated, unless the rulers are clever enough
tertiary sector of society whereas the fol- to co-opt the intellectual proletariat by
lowers come from somewhere low down giving them something that tastes of
in the primary and secondary sectors. power, e.g. by paying them for writing
Thus, what is needed is first of all a suf- recommendations.
ficient amount of built-in disequilibrium The revolutionaries may be able to sway
in these social positions. If the point of a sufficient number of followers to do the
departure is a feudal system, for instance footwork for them, but we are concerned
a newly independent, formerly colonial with the structural conditions for auto-
territory, then one way of arriving at this matic supply, not with special conditions.
would be as follows:30 Thus, one simple formula is to copy what
1. Create universities and other insti-
is mentioned above lower down, and in
tutions of higher learning so as to the primary or secondary sectors of society:
turn out a sizeable number of in- 3. Institute mass educationwith a com-
tellectuals who feel they have a key pulsory base and easy access to
- their high level of education - educational follow-up institutions
not only to their own well-being, of various kinds especially so as to
but to the welfare of the whole permit autodidactic leaders to
society, e.g. economists, physi- emerge. The factor of self-right-
cians. eousness will probably work more
2. Make few positions available so that strongly for the autodidact than
the high educational status will for the formally trained person,
not be translated into the kind of especially if he is high on what he
instrumentality that gives power. has learnt himself and low on
Regardless of the economic situ- formal schooling. This contributes
ation, this intellectual proletariat to an explanation of the role of
is a proletariat in the sense of not typographers in social revolutions:
having access to the machinery their work brings them close to
they know (or think they know) a source of rank-disequilibrium
how to turn. They are forced into through studies.
other positions, and these posi- 4. Make no otherchanges,which means
tions will call for subsidiary capac- that the recent rise in education
ities (accounting, typing, low ad- is not accompanied by any cor-
ministration) and the disequilib- responding rise in economy or
rium is created. power.
It should be noticed how easily such an Again, it should be noticed how easily this
intellectual proletariat is created: in the is done. Mass education, like mass medi-
age of technical assistance and internation- cation, costs little compared to building
al fellowships it takes little time to turn dams or irrigation schemes or the creation
out university graduates, but it still takes of a sector of heavy industry. Also, like
much time to tune an administration to an mass medication it can bring quick results
efficient utilization of their skills. The and cause disequilibria. The disequili-
intellectuals will probably oversell their brium caused by raising the hygienic
products precisely because they are under- standard without a corresponding rise in
bought and underdemanded, and they the economic basis is well known - to this
will be feared, envied and hated by their can be added the effects of a free education
rank opposites, the powerful non-intel- market without a corresponding freedom

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
in the markets of economics also be argued that the two factors are
(goods) and
politics (power).31 rather similar. They are both themes of
Let us then add to these four conditions frustration, and more than that: a boom
four more, and we should have a relatively followed by a depression or repression is
good set of predictors: likely to create rapidly a high number of
5. A patternof boomfollowed by depres- disequilibria at critical points in the social
sion or repressionas mentioned structure. But disequilibria are also created
earlier - the pattern made ex- by rapid economic growth, so the argu-
plicit by Davies in his article. ments in the preceding section about the
6. Contactbetween the two (or more) consequences of economic development
disequilibrated groups, between for external war can also be turned into
the tertiary high and the primary arguments for internal war. In other words,
or secondary low in disequilib- there is little inspiration in our theory for
rium. Urbanization provides the
medium for such contacts, and is anyone who might want to stop 'com-
a rapidly increasing resource on munism' nationally or internationally
the world level. through a policy of technical assistance
7. An ideology that does not have to and economic aid. What is wrong with
explain the past or present nor to that theory is that it confuses the social
predict or prescribe the future, situation in nations or groups equilibrated
but has to provide a kind of se- at the top with all the disequilibrium
mantic bridge over the social
states they have to go through on their
distances within the group urging
change. This function of ideology, way up from a complete underdog situ-
to provide a revolutionary group ation.
with emotive symbols that are
easily applied and have the same c. On the numberof elementsin a system
reference for those who use them,
is the more necessary the greater All our reasoning so far has been with
the social distance within the no reference whatsoever to the number,
group. N, of elements in a system. This simple
8. A charismatic leader. The functions variable, number, is rarely used for other
of personification and centraliza- purposes than data analysis in social
tion are not easily satisfied without science; here we shall try to point out one
a leader. To say that he should theoretical implication of number. We
have charisma is probably a
choose a simple world: it is ranked ac-
tautology since the proof of his
charisma lies in his ability to be a cording to two dimensions and in a
leader and sway people into random way so that 1/4 of the elements
action. But a personality with are TT, 1/4 are TU, 1/4 are UT and 1/4
appeal across social distances is are UU. The question is: what difference,
indispensable. if any, does it make if this world has 4, 40,
Any one of these conditions may serve as 400, 4,000 or 4 million elements?
a spark to ignite the motivational energy We expect the drastic demands for
stored in the disequilibrium mentioned in reallocations to come from the TU and
points 1-4 - and all four together should UT positions. But there is a long distance
be more than sufficient. We would, as between making a demand, a request and
mentioned, believe more in disequilibrated open aggression, and the probability of
rank as a source of revolutions than in the aggression will also depend on the number
Davies factor, for a completely balanced of alternatives. With increasing N the
underdog group is so psychologically and number of combinations for a given ele-
ideologically conditioned as to absorb the ment increases very rapidly. There is no
vagaries of economic cycles. But it may need to enter into the mathematics of the

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
combinatorics since they reflect nothing of elements throws the elements against each
substantive interest except the extreme other mercilessly: all comparisons involve
rapidity with which the number of sub- some element of strain, there is no refuge
systems that can be formed increases in the relaxed atmosphere of the complete
with N. peer, the real equal. With only two ele-
Some of these subsystems may be bel- ments it would be still worse, especially if
ligerent coalitions that increase the prob- they were posited against each other in
ability of aggression. But there is another TU and UT positions, one being high in
kind of subsystem which is more important power and low in culture and the other
in this context. To discover it one only one being just the opposite. It is easy
has to recognize one way in which ten- enough to see where the difficulties dis-
sions are alleviated for disequilibrated appear: only at the point where all ele-
individuals in intra-national systems: they ments have coalesced into one.
form subsystems that dissolve the disequi- If there is stability of the kind under
librium. A group of 100 TUs isolating discussion in systems consisting of one
themselves from the rest of the system will element on the one hand and systems con-
no longer see themselves as TUs; the sisting of many (hundreds, thousands,
dimensions are negated in the homogene- millions) on the other hand, then some
ity of the groups. Thus, one university conclusions can be drawn. Nations owe
professor may feel entitled to more power, their stability, if they have any, to 1) a
to the power that corresponds to his wis- large number of individuals, and 2) a
dom - together with 99 others he may sufficiently complicated social structure
be more concerned with minute differences in terms of all kinds of rank-dimensions to
in wisdom within the group than gross prevent a simplification to a very low
differences between his position and other number of groups. Of course, stability
positions in the society. is enhanced by obliterating the rank-
If the idea is simply that people and dimensions, by making them irrelevant,
groups and nations like feeling superior as during a foreign occupation - but this
or at least not inferior, then the impli- is a rare occasion and when it does not
cations are equally simple. A TU for N = obtain there is strength in number alone.
4 will have one other element to associate Similarly the international community
with (UU), whereas the TU will have 19 should have its points of stability for the
others for N 40 (10 UU and the remaining one world state as well as for a world
9 TU). The UT are excluded because consisting of a large number of nations.
their profiles lie over the profile for TU With the present trends the world is
at one point. The gain in sources of grati- probably movirfg into the in-between
fication is conspicuous and continues: 1, region; for instance with two federations
19, 199, 1,999, ... On the other hand, as developing out of the NATO-alliance,
soon as these subsystems are formed they (one North-American and one European),
may serve to reduce effectively the number of the Warsaw alliance (Soviet and Eastern
of elements until one has a supersystem of Europe), Asia (for or against China),
four elements, where the elements are Africa (Arab and bloc) and Latin America
complicated organizations bearing the (Spanish and Portuguese) one is left with
TT, TU, UT and UU characteristics. a dozen nations and all kinds of rank-
Thus, the higher the number of ele- profiles. Formal political union is not
ments, the higher the probability of finding necessary for such effects to be demon-
some kind of organizational insulation strable: even now European nations play
against the strains produced by a disequi- a counterpoint to the US; Eastern Euro-
librium position. The world with four pean nations (singly more than combined)

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
to the USSR; in Asia a China-non-China will pass before the UT, high on education,
division is meaningful; in Africa the Sa- starts wondering why he should be less
hara is a dividing region and in Latin well off than the TT, not to mention the
America the poor could well be pitted TU - he may find comfort in his top
against the rich. position for a while, and then start
worrying about his low position.
d. On the numberof dimensions The question is now whether there is a
It looks as if one may use the same argu- limit to this kind of reasoning set by the
ment about the stability in one and many, number of dimensions. A world with a
but not in few, for the number of dimen- very high number of salient dimensions
sions. Of course, the number of dimen- would permit more flexibility; the number
sions is a more volatile characteristic of would serve as a cushioning against the
the system than the number of elements. effects of disequilibrium. For each new
But it is not operationally meaningless, as dimension gives a possibility upwards, and
indicated in the section on methodology. hence a source of gratification, especially if
One will have to count the number of different nations climb on different dimen-
criteria of ranking on which there is a sions. To take an example: sports competi-
degree of consensus above a pre-established tions and 'contests' as to which nation
level. has the highest number of Nobel laure-
Imagine that there is only one such ates both define positions that can be
dimension - for instance per capita highly gratifying and compensating. Our
income. This does not mean that conflicts hypothesis, then, is that a U position may
and aggression will disappear, but that loom high in the national conscience if it is
one source of aggressive behavior, arising surrounded by one or a couple of T posi-
from disequilibrium (e.g. with size of tions but not if it is surrounded by very
territory) does not exist. The 'haves' will many such positions. Then it may even
be envied by the 'have-nots' but quite be turned into a point of pride: look at
possibly be more safe in their topdog what we have been able to do in spite of
position than they would have been if that handicap (the national ethos of the
there were some more dimensions avail- small nations in North-West Europe,
able. For on these other dimensions the Norway for instance, illustrates the case).32
underdogs might rise and get into the Again, if the U is surrounded by nothing
web of disequilibrium. No doubt, there because the system is seen as one-dimen-
are exceptions to this. Topdogs have al- sional, then the hypothesis is that an
ways and everywhere known the impor- attitude of acquiescence will be more
tance of stretching out a ladder of com- likely, for there is no disequilibrium pres-
pensation to the underdogs, for instance ent to provoke restlessness.
in the form of an ideology that promises The consequence of this is a prediction
salvation in a transcendent existence (reli- of stability in extremely monolithic and
gion) or in this world ('die orgiastische extremely pluralistic cultures - in the
Chiliasmus'). But these ladders are not former there will be one criterion, in the
rank-dimensions in the sense that they latter many, for pluralism is precisely a
lead to identifiable positions where one is multi-faceted basis for evaluation. Again,
treated differently by the whole society. it may look as if the world is presently
Thus, to institute mass education and give moving towards the in-between region.
knowledge to the masses with the idea that Communications and international organ-
'this will satisfy them, they will think less ization make greater areas relevant for
of getting property' is both unpsychologi- each other - but at the same time the
cal and unsociological. Not much time richness in total cultural variation is

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
broken down. But what could be called the criss-cross theory. And thus world
the world consensus is probably richer history unfolds itself, and makes aggres-
today than the narrow military-economic sion as lasting as rank-disequilibrium, that
basis of evaluation prevalent during the is, as long as human society exists. But
period of colonialism. that does not imply, of course, that the
Combined with the preceding section forms of aggression will necessarily be as
this means roughly the following: the world dangerous as is the case today, nor that
is neither in the simple monolithic world- social change will always cause as much
state, nor in a very pluralistic system of rank-disequilibrium as it does today.
numerous and small nations - say a In a sense the whole theory is located
couple of thousand. The world is in- somewhere between purely system-oriented
between and in a situation where there is theories of aggression (like the theories of
considerably less than maximum protec- Richardson processes) and purely element-
tion to be found in the formation of sub- oriented theories (attributing aggression
systems. We are then thinking not only of to national character). According to the
the subsystems where nations that feel theory it is the rank-balance within a unit
similar keep together, but also of cultural that counts, not the rank-balance between
subsystems where a nation regards as units. But this balance within is defined
salient only some dimensions. She may with reference to the balance between and
pick the dimensions where she ranks high is meaningless except in a system context.
- and receive gratification. But, un- In fact, as indicated in the reasoning about
doubtedly, she may also pick precisely the formation of subsystems, the system
the dimensions that present her with a reference is indispensable if one is to
maximum internal disequilibrium - so decide whether an element is in equilib-
richness and variety in number of dimen- rium or not.
sions give a potential for containment of
aggression, not a fool-proof recipe. 7. Somepolicy implications
Not much can be said in terms of policy
* * *
implications of a principle so general as
In general, one may say that this per- the principle under discussion, and not
spective implies a number of connections much should be said as long as the prin-
between change and conflict, betwteen the ciple has not been confirmed for a variety
causes of disequilibria and their aggressive of strategically placed tests. Nevertheless,
dissolution. Social change is structural; it it is interesting to speculate as if this had
may, for instance, introduce new and already been done and to look further at
usually disequilibrated rank-combinations. the implications.
Or it may distribute new resources more First of all, the principle and a value
evenly, which is another way of saying premise against aggression are compatible
that some complete or nearly-complete with a large variety of possible worlds.
underdogs will rise on one or more dimen- Among these worlds should be mentioned,
sions due to mass education, prosperity, 1) thefeudal orderconsisting of the complete
universal suffrage or, on the national level, topdog and the complete underdog alone,
to such factors as freedom from colonialism 2) the pluralistic world consisting of a large
or economic development. The result is variety of nations, possibly several thou-
disequilibrium with consequent aggression sands or based on a large variety of
until more equilibrated combinations of criteria of ranking and achievement, and
rank-sets are achieved - or a precarious 3) the unitary world where independent
balance between equilibrated and dis- national actors have disappeared from the
equilibrated combinations as described by scene or there is only one dimension of

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
evaluation. In all other worlds, according preted negatively. That is, to take the
to the theory, unbalanced configurations most important example in the world just
exist and are bound to harbor aggression now, if a nation which is the world's
- as long as one assumes the existence of largest in terms of population, one of the
rank-dimensions and values worthy of largest in terms of area and the most
being pursued. In general ideological de- 'glorious' in terms of history undertakes
bate these three worlds are usually con- the transition from a semi-colonial status to
sidered so disparate that one of them would independence and autonomy through the
exclude one or more of the others; the the- catharsis of a revolution and then under-
ory points to what they have in common. goes a dramatic socio-economic change
Secondly, a general implication well that alters its economic status towards that
known from debates on economic devel- of an industrial nation: if that nation is
opment is the following one: if change is voted out of existence as a political entity
one-dimensional only from the position of by being denied membership in the club
the complete underdog, aggression is likely of nations and more particularly by being
to be its effect; for aggression to be avoided denied access to the club of great powers,
a necessary condition is multi-dimensional then the hypothesis of this article would be
change. This calls for a tremendous very wrong if the predicted aggression did
amount of political ingenuity.33 One na- not occur within a reasonable time-span.
tion leaps economically ahead of the Whether that nation adds to its territory
others 'with whom it is natural to compare or seeks de facto political control through
oneself'. The idea of compensation and other media the principle does not predict,
egalitarian concepts of social justice would but the principle would predict that if due
prescribe some kind of mobility to the political status is given to it before aggres-
other nations so that the first nation 'does sive tendencies have had too much of a
not become too strong'. Thus, one maxim habit-forming effect, then its neighbors
of statesmanship will have it that federal should have less to fear.
capitals should be placed in one of the It is not difficult to look through national
poorer member States and State capitals statistics and to discover disequilibria and
at least not in the richest or biggest city make predictions about the future behavior
of the State. Within a well-integrated na- of some nations in terms of aggressive be-
tion this is probably non-consequential, havior. A major formula of aggression will
the forces created are not unleashed be- probably be precisely the nation favored
cause of the strong bonds between the by nature, culture, talent or an economic
constituent elements. But in a loosely aid benefactor that makes a jump ahead
integrated group of individuals or nations of its neighbors so that all three mechan-
the principle would predict aggressive isms of disequilibrium are set into mo-
consequences unless the system generates tion. This makes those economic policies
enough value to permit mobility that is to the effect that some nations should be
not at the expense of the others. developed first, as has been advocated by
The prescription based on the principle some,34 particularly dubious from this point
would rather make increase in political of view. And since parallel development
influence an immediate concomitant of of many nations on one dimension is also
economic growth. This would be in agree- an implication of the principle one is not
ment with achievement conceptions of stuck with the elitist formula of parallel
social justice, but lest it should lead to a development of one nation on many
feudal world order that many people would dimensions.
rule out on the basis of other value pre- Clearly, if aggression had less dangerous
mises, the principle should rather be inter- consequences, as it may have in a world

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
where arms are controlled and alternative principle than the opposite principle, if
methods of conflict management better economic development is to take place
developed, one could focus on the positive without too many deviations into war and
aspect of aggression and leave the nega- warlike processes. And throughout the
tive side of the coin, the damage done article a number of indications given
unto others, aside. For disequilibrium is a should make it clear that the implications
reservoir of initiative that serves evolution of an equilibrium principle are not neces-
unless it is frustrated. But in the present sarily on the 'conservative' side, or neces-
world there are probably better theoretical sarily on the 'radical' side - whatever
reasons to favor some kind of equilibrium these terms may mean.

* This is a revised version of a paper presented during spring and summer 1964 at the Circolo
Turati, Milano, Facolta delle scienze politiche, Universita di Torino, Polemological Institute,
University of Groningen, Danish Conflict Research Society in Copenhagen, the study group in
conflict and peace research, Lund University, the study group under the Scandinavian Summer
University in Aarhus, and at a plenary session of the Scandinavian Summer University in
Bergen, here published as PRIO publication no. 1-1. Deep gratitude is expressed for all the good
ideas received during these discussions - particularly to Mr. Bengt Hoglund of the study group
in Lund. The study is an outcome of a grant from the Aquinas Foundation, New York, and from
the Norwegian Research Council for Science and the Humanities, and serves as a theoretical
basis for a series of empirical investigations.

1 This is different from standard definitions in the field, e.g. the famous definition given by
Dollard that aggression is any 'sequence of behavior, the goal-response to which is the injury of
the person toward whom it is directed'. This definition is also used in the standard work by
Berkowitz, L., Aggression:A Social PsychologicalAnalysis (New York: McGraw-Hill, 1962). But
we agree with Klineberg when he writes (The HumanDimensionof InternationalRelations,New
York: Holt, Rinehart and Winston, 1964, p. 11) that 'The question of universality of aggres-
sion is further complicated by considerable difference in the definition of the term itself. One
writer, for example, refers to the original meaning of aggression as a tendency to go forward
or approach. This is regarded as instinctive, whereas the inborn or instinctive nature of hostility
has never been demonstrated. Another describes it as the will to assert and to test our capacity
to deal with external forces, and it is this, rather than hostility, that is a fundamental character-
istic of all living beings' (p. 10). But universality or fundamentality still leaves us with the problem
of where or for whom aggression in this broad sense is most pronounced, and with the problem
of under what conditions aggression expresses itself as hostility. We use aggression somewhat in
the sense of 'self-assertion', but only insofar as this self-assertion implies an effort to change social
relations, i.e. no longer to comply with existing conditions.
2 Klineberg, op. cit., pp. 7-17.

3 For an analysis of the concept of interaction see Galtung, J., 'Expectations and Interaction

Processes', Inquiry,1959, pp. 213-34.

4 See, for instance, the brilliant analysis in Tannenbaum, F., Slave and Citizen, The Negro
in the Americas(New York: Knopf, 1947). It took many years before the rank-disequilibrium of
some, and complete underdog status of most US Negroes led to the mutinies of the 19th century.
5 See, for instance, the analysis in Barth, Fredrik, 'The System of Social Stratification in Swat,
North Pakistan', in Leach, E. (Ed.), Aspectsof Castein SouthIndia, Ceylonand North-WestPakistan
(Cambridge: Cambridge University Press, 1960), p. 113.
6 Davies, J. C., 'Toward a Theory of Revolution', American SociologicalReview, 1962, pp. 5-19.
7 Op. cit., p. 6.
8 Zawadzki, B. and Lazarsfeld, P. F., 'The Psychological Consequences of Unemployment',
Journalof SocialPsychology,1935, pp. 224-51.
9 Davies, op. cit., p. 7.

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
10 See, for instance, Ch. 5 in Arnold Rose, Indaginesull'integrazione socialein duequartieridi Roma
(Roma: Istituto di Statistica, 1959), pp. 60-69; or such classics as Angell, R. C., The Family
Encountersthe Depression(New York: Scribner, 1936) or Mirra Komarovsky, The Unemployed
Man and his Family (New York: Dryden, 1940).
11 Kogon, E., The Theoryand Practiceof Hell (New York: Berkley Medallion Books, 1958),
Chapter 23, 'The Psychology of the Prisoners'. (Title of the original: Der SS Staat)
12 Quoted from Davies, op. cit., pp. 5 ff.
13 The condition of rank-disequilibrium is, of course, not a necessary condition for aggression.

Aggression may arise for other reasons. And it is hardly a sufficient condition either - perfect
relationships between variables are rarely if ever found in the social sciences - but we shall
argue later that for high levels of disequilibrium aggression seems to be a very probable con-
14 RankandSocialIntegration (Oslo: Peace Research Institute, stencil 10-2, 1963), to be published
in Berger, Zelditch, Anderson, SociologicalTheoriesin Progress(Boston: Houghton Mifflin Co.,
15 Work is under way at the Peace Research Institute in Oslo to test this implication of the

hypothesis. So far it is based on more intuitive impressions of delinquents, and some investigations
purporting to show that juvenile delinquents more than others in comparable positions are marked
by some of the characteristics of the entrepreneur,
viz., initiative, intelligence, energy, ability to take
risks and to see possibilities, etc.
16 The following list may be useful as a reference. We have presented two sets of variables

for nations, depending on whether the variables are 'analytic' (based on statistical information
about individuals) or 'global', i.e. sui generis.


1. Age (adults v. adoles- Age as a nation

cents and children) population pyramid
2. Sex. (men v. women) J
3. Family ('good' v. 'bad') Alliances ('good' v. 'bad')
4. Primogeniture ('first born') Doyen in a group of nations
5. Race (Caucasoid, Mon- rates of racial composition Dominantculture
goloid, Negroid, etc.)
6. Ethnicity(Gentiles v. Jews, rates of ethnical composition Dominantculture
emigrants v. immigrants)
7. Ecology(urban v. rural) rateof urbanization
8. Geography(center v. peri- Centralv. peripheralnation
9. Nation (for individuals
only - all national vari-
ables relevant as context)
10. Education(degree) rate of literacy Educationalstructure
11. Occupation, split into Occupationalstructure
a. sector (tertiary, secon- I-rate in primaryoccupations
dary, primary) 1-rateof populationin low occu-
b. position (high v. low) pations Socialstructure
12. Ideology('right' v. 'wrong') rates of believers Dominantideology
Stage of economicgrowth
13. Income, property (rich v. per capita income = per capita GNP (development)
poor-'dispossessed'; consumption+ per capita naturalresources
bourgeoisie v. proletariat) investmentstandardof living utilization of resources(rate of
14. Power (rulers u. subjects) rate of popularparticipation Politicalstructure
15. Legality (law-abiding v. rates of criminality Legality
16. Health (well v. ill) ratesof morbidityetc., life expect- MedicalStructure

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions

17. Knowledge(those who Cognitiveculture

know v. those who do not
18. Skill (those who can v. Technicalculture
those who cannot)
19. Conviction(the 'true be- rates of true believers Ethicalculture
lievers' v. the others)
20. Taste (artists v. laymen) Estheticculture

17 This corresponds to the distinction made in game theory between cooperative and competi-
tive games. Also see Schelling, T., The Strategyof Conflict (Cambridge: Harvard University
Press, 1960), Chs. 2, 3 and 6.
18 See
Jackson (9, p. 473) or Galtung, J., Membersof Two Worlds,A SociologicalInvestigationof
Three Villages in WesternSicily (Oslo: PRIO publication no. 6-2, forthcoming), section 4.6.
19 See Jackson, op. cit.
20 Ibid., pp. 476 ff.
21 Loc. cit.
22 I am indebted to Tom Broch for this suggestion.
23 For many of the nation variables indicated above, in note no. 16, UnitedNations: Statistical

Yearbookand UnitedNationsDemographicYearbook will do. For other variables the excellent World
Handbookof Political and Social Indicatorsprepared by the Political Data Program of Yale Uni-
versity will answer many questions. But this is for the contemporary scene; since the hypothesis
should also be tested over time historical data are needed where cruder distinctions will have to
be made.
24 The main findings in the empirical investigations based on the ideas of rank-equivalence

can be summarized as follows for this and other references:

Author Aboutthe rank-equilibrated Aboutthe rank-disequilibrated

Fenchel less so unstable self-images because of differential
et al. (4) treatment

Adams (1) 'less subject to discontents or 'lack of congruency is an effective motivator

compensatory behavior of the individual'
groupperformance low group performance high
group atmosphere 'harmoni- group atmosphere less harmonious
ous, trusting cooperative

Lenski (14) less radical support Democrats more, have liberal and
left views

Lenski (16) less isolated 'social isolates more numerous'

more sociable in their motiv- 'less likely to report sociable motivations' -
ation for social participation

Goffman (6) less concerned about change 'prefer extensive change of the distribution of
of power distribution power in society' - but only when 'experience
opportunities for upward mobility are low'

Jackson (9) less frustration frustration leads to

less ambiguity 1. stress (intrapunitive) if low status more
2. radicalism(extrapunitive) if low status more

Landecker (13) in general, more conscious- less so

ness of class status

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
The focus of the research has been on the persons with rank-differences. Unfortunately, only
the very stimulating article by Adams has group data in addition to individual data, and this
leads him to the important conclusion: 'it should be noted that there is an apparent antithesis
between productivity and group euphoria, at least in settings such as the military organization'
(ibid., p. 22). In general, it should be noticed that all findings reported by these authors are
based on relatively small percentage differences. Hence, even though the phenomenon is too
consistent to be written off in any way, it is probably of secondary importance to other factors.
26 The science of peace research, when it gains momnentumand perspective, will probably
acknowledge its tremendous debts to the genius of Lewis F. Richardson even more clearly than
is the case today. Regardless of the criticism that can be raised against his work, his courage in
treating nations and wars like the physicists treat bodies in mechanics has been invaluable. It
is to be hoped that there will be many follow-up studies such as Paul Smoker's work on the mathe-
matical side (JPR, 1964, pp. 55-64) orJ. D. Singer's forthcoming work on '150 years of conflict'
on the statistical and theoretical side.
26 Similarly, one can predict at what period in a woman's life the tendencies towards partici-

pation in emancipist movements will be most pronounced. She starts as a young girl with equi-
librium between her low sex-status and her low age and probably also education, income, power
(if only in the family), etc. But as she grows older she will grow into disequilibrium between
her sex position and all the concomitants of age; hence, we would predict participation in radical
movements at a much older age than for men. Men gain equilibrium as they grow older, women
lose it - hence the difference.
27 The only author to our knowledge that has written consistently about the sociology of

international relations from the point of view of international stratification is Lagos, Gustavo,
International Stratification and UnderdevelopedCountries (Chapel Hill: University of North Carolina
Press, 1963). However, as pointed out by Amital Etzioni in his review of the book (American
Journal ofSociology, 1964, pp. 114 f.) 'the potentially promising line of examining the consequences
of status inconsistency, of a country rich but weak, poor but honest, etc., is not sufficiently devel-
oped'. For further comments see Appendix 2 in Galtung, 'Rank and Social Integration' in Berger,
Zelditch, Anderson, op. cit.
28 See article by Mari H. Ruge, 'Technical Assistance and Parliamentary Debates', JPR,

1964, pp. 77-94.

29 The role of China in the follow-up of the 1955 Bandoeng conferences can probably be studied
under this perspective.
30 A study of the Cuban revolution guided by this theoretical perspective is in process at the
Peace Research Institute in Oslo. The reader may find it useful to test the ideas against standard
knowledge of this revolution and of the life history of Fidel Castro.
31 Thus, it is not surprising that turns towards the left have taken place where the rate of illiteracy
is relatively low, as in Kerala in Southern India, Cuba or Chile, which stood a fair chance of
electing into power a Marxist president, Senator Salvador Allende. The relation between the
interplay of social indicators and the type of political system is probably a good deal more com-
plicated than the famous analysis by Seymour Lipset: 'Economic Development and Democracy',
in Political Man (New York: Doubleday, 1960), pp. 45-76.
32 Norwegians love per capita statistics, for the simple reason that many indicators when divided

by a small population will make Norway rank relatively high.

33 Adoption is a system whereby a wholesale transition from (almost) complete underdog to
(almost) complete topdog position can take place. The Japanese feudal system of bestowing on
a servant 'who has proven himself' status as son of the family is a good example, and very functional
for the preservation of the feudal order precisely because it preserves rank-equilibrium. Knighting
in the middle ages probably had a similar function, whereas winning in the lottery does not, it
may create gross disequilibria. However, since equilibration will be on other dimensions than
property, aggression may not necessarily result from this type of disequilibrium.
34 See, for instance, Shonfield, Andrew, The Attack on World Poverty (London: Chatto and

Windus, 1961).

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
on rank-disequilibrium
(1) ADAMs,STUART.'Status Congruency as a Variable in Small Group Performance,' Social
Forces, 1953, pp. 16-22.
(2) BENOIT-SMULLYAN, E. 'Status, Status Types and Status Interrelationships,' ASR., 1944,
pp. 151-61.
(3) BROOM, L. 'Social Differentiation and Stratification,' in SociologyToday. (New York: Basic
Books 1959) pp. 429-41.
(4) FENCHEL,G. H., MONDERER,J. H. and HARTLEY, E. L. 'Subjective Status and the Equi-
libration Hypothesis,' JASP, 1951 pp. 476-79.
(5) FREEDMAN, R. et al. Principlesof Sociology.(New York: Henry Holt, 1956) Chs. 7, 13.
(6) GOFFMAN, I. W. 'Status Inconsistency and Preference for Change in Power Distribution,'
ASR, 1957, pp. 275-81.
(7) HODGE.'Status Consistency of Occupational Groups,' ASR, 1962, pp. 336-43.
(8) HUGHES,E. 'Dilemmas and Contradictions of Status,' AJS, 1944, pp. 353-57.
(9) JACKSON, E. F. 'Status Consistency and Symptoms of Stress,' ASR, 1962, pp. 469-80.
(10) KAUFMAN, H. F. et al. 'Problems of Theory and Method in the Study of Social Stratifi-
cation in Rural Society,' Rural Sociology,1953, p. 15.
(11) KENKEL,W. F. 'The Relationship Between Status Consistency and Politico-Economic
Attitudes,' ASR, 1956, pp. 365-68.
(12) Landecker, W. 'Class Crystallization and Its Urban Pattern,' Social Research,1960, pp.
(13) LANDECKER, W. 'Class Crystallization and Class Consciousness,' ASR, 1963, pp. 219-29.
(14) LENSKI,G. 'Status Crystallization: A Non-Vertical Dimension of Social Status,' ASR,
1954, pp. 405-13.
(15) LENSKI,G. 'Comment on Kenkel's Communication,' ASR, 1956, pp. 368-69.
(16) LENSKI,G. 'Social Participation and Status Crystallization,' ASR, 1956, pp. 458-64.
(17) MORRIS, R. T. and MURPHY,R. J. 'The Situs Dimension in Occupational Structure,'
ASR, 1959, pp. 231-39.
(18) RINGER,B. B. and SILLS, D. L. 'Political Extremists in Iran,' POQ, 1952/53, pp. 689-701.
(19) SOROKIN, P. Society,Cultureand Personality.(New York: Harper, 1947) pp. 289-94.
(20) WEBER,MAX, in Gerth, Mills. From Max Weber:Essays in Sociology.(New York: Oxford
University Press, 1946) pp. 180-95.
(21) ZALEZNIK, CHRISTENSEN,ROETHLISBERGER.The Motivation, Productivity and Satisfaction of
Workers. (Boston: Harvard University Press, 1958) pp. 56-66.

The article, which is completely theoretical with no attempts at empirical verification,
develops a structural theory of aggression between individuals, groups and nations.
The theory sees a social system as a system of units in interaction and multi-dimen-
sionally stratified according to a number of rank-dimensions. The theory tries to locate
the maximum probability of aggression against other units, and after exclusion of one-
dimensional rank or the complete topdog and complete underdog positions and ex-
cluding one-dimensional downward and upward mobility as a suitable basis of ex-
planation, settles for rank-disequilibrium as a (nearly) sufficient condition. The theo-
retical basis is the differential treatment and relative deprivation that follows from rank-
disequilibrium, the resources that the high status provides the unit with to improve his
low status, and the sense of self-righteousness that easily develops. A methodology for
testing the theory is developed, and a number of implications for the theory of the
consequences of economic development, the theory of revolutions, for the number of
nations in the international system or the number of rank criteria used are examined.
Three world orders are thought to be aggression-reducing: a feudal international order,

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions
a highly pluralistic system with many nations or rank criteria and a highly unitary
system with a world-state or one rank-criterion. The importance of economic develop-
ment in such a way that there is a multi-dimensional growth and a parallel develop-
ment between nations is emphasized.

IpaTKoe conepHaHHe.
3Ta CTaTbH pas3BsaeT cTpyKTypaJibHyio Teopuio arpeccHH Me>Ky OTAeJIbHIMH HHusIBHAaMH,
rpynnraMl, HaAIEMH C HECTOTeOpeTEIeCKOt TOKIH 3peHHH, He xeJiaH nonBLTOK o60cHOBaTL ee
TeopHH onpenejineT KaKyI)-JI160 congaJIuHyIo cIcTeMy, KiK CECTeMy 'aCTH1 BO B3saHMoAe-
CTBEH, pa3jaejieHHuix Ha HeKoJIabKO BejiWMH, COriaCHO onpejejieHHOMy KOjinmeCTBy pa3pHAOB.
Teopuf feJIaeT nonbiTKy yHa3aTb ycJIoBHH, npH KOTOpbIx c HaH6ojbIMe4 BepOHTHOCTbIO,
oAHa cEcTeMa MoReT cosepHIITb arpeccHio pOTEHBApyrot.
IcKHJIio'aH oXHOBeJIWIHHLIe paspHnbI, HJI1 noJIoReHPe BbInrpaBUIei4 H npoHrpaBmelt
CTopoH, a TaK7{e UCKJIiotaH ofHOBejmqEHHOe KOJIe6aHHe IBBepx HJIH BH13/ B KaHeCTBe noRxo-
HIiiHX KpETepIeB HJIH oG6bCHeHun, npeiJaraeTcH HapymueHe paBHOBecHH pa3pffoB, no
CyIIecTBy, ocTaTOvHbIM yCJIOBHeM uAn arpeccrn.
TeopeTwuecKHM o6ocHosaHieM aTOMy cJIy?KHT paarnsYHoe o6paineHe c 1HHnasBaME, rpyn-
naMH, HaIHHMEH, MyBCTBO6bITb o6ofAeHHbIM, B03HHKaioLIee B pe3yjbTaTe HapyeHEHH pasBHO-
BeCHf B paspwRax, B03MOHKHOCTI, IMeioiiiecIH y rpynn c BbICOKHIM noJIOHKeH1eM B OAHOMHan-
paBJIeHBI, nOBbICEHTbcBoe 6ojinee HH3Koe nojinoeHEHe B ApyroM HanpaBJieHHH, aerKo B08HH-
Kaiounee ,,QyBCTBO CaMonpaBenHOCTE<< F T. A.
ia TeopHio o cocTaBe HaLHZI B Me>KnyHapOjHOH CHCTeMe H yKa3abIBaeTCH HOJiHeTBO npHMeHreM-
hIX pa8pHAOB.
HpeAno3araeTCH, YTO Tp1 cnoco6a ycTpocCTBa nopwAKa B MHpe CH1EaIoT BepOHTHe arpeccHH:
4eoJaJIbHan MeCAyHapO 11aHcHcTeMa, MHorocTopOHHHH cclCTeMa, BHJI1oHaIo01aH HecKoJIbKo
HaLHH, HJ1H pa8prWOB, H efIuiaH cHcTeMa oHoro MHpOBOrO rOCyAapCTBa, HJIH OJHH paspHa.
HoffHepK1BaeTCH 3HaMeHHe pasBHTHT B pa3sHbIX HanpaBJeliHHx, a TaKwe sHa'eHHe napajin-
IeJIbHoro paSBEITIIH Hag11.

This content downloaded from on Fri, 30 Aug 2013 13:22:52 PM

All use subject to JSTOR Terms and Conditions