Brush 1996 Dynamics of Theory Change

Dynamics of Theory Change in the Social Sciences: Relative Deprivation and Collective
Violence
Author(s): Stephen G. Brush
Source: The Journal of Conflict Resolution , Dec., 1996, Vol. 40, No. 4 (Dec., 1996), pp.
523-545
Published by: Sage Publications, Inc.
Stable URL: http://www.jstor.com/stable/174456
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.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Sage Publications, Inc. is collaborating with JSTOR to digitize, preserve and extend access to
The Journal of Conflict Resolution
This content downloaded from

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
All use subject to https://about.jstor.org/terms
Dynamics of Theory Change
in the Social Sciences
RELATIVE DEPRIVATION AND COLLECTIVE VIOLENCE
STEPHEN G. BRUSH
Department of History and Institute for Physical Science and Technology
University of Maryland at College Park
The extent to which theories in the social sciences are accepted or rejected on the basis of empiric
can be shown only by a detailed analysis of specific cases. The author examines the reception
scientists in the 1970s and early 1980s of T. R. Gurr's theory of collective violence based on the
relative deprivation. The history of this theory may be considered an example of definite progres
science: a hypothesis widely accepted at one time has been tested and rejected, thus making room
development of alternative hypotheses. But although Gurr and other advocates of the theory have ab
it in its original form following the mostly negative results of empirical tests, many social
(especially psychologists) have continued to cite it favorably. Slightly less than half of the unf
citations have been supported by references to empirical evidence.
W hat does it mean to be scientific? For social scientists, whose status has lo
problematic and whose jobs are threatened by financial pressures at some inst
this is more than an academic or philosophical question. It is also a question
provokes heated controversies within disciplines such as sociology and anthr
between those who favor quantitative empirical methods or mathematical mo
those who insist that a qualitative descriptive or hermeneutic approach is m
appropriate.1
1. A useful analysis of this controversy is given by Rosenberg (1988), who demonstrates th
scientists must choose between the two approaches (which he calls naturalistic and interpretative)
as a philosopher he avoids making that choice himself.
AUTHOR'S NOTE: The research for this article was done while the author was a member of the School
of Social Science, Institute for Advanced Study, Princeton, New Jersey, during the 1992-1993 academic
year. This case was suggested to me by James B. Rule, to whom I am indebted for valuable discussion and
for comments on an earlier draft. I thank Ted Robert Gurr for information provided in a telephone interview.
Useful ideas emerged in discussions with Gene Bums, Stephen Cole, Larry Laudan, Frederick Suppe, and
Michael Walzer. My research was supported by the Andrew W. Mellon Foundation and by the General
Research Board of the University of Maryland. Address correspondence to the Department of History and
Institute for Physical Science and Technology, University of Maryland, College Park, MD 20742. Detailed
statistical results may be found in an extended version of this article, available from the author, or on the
World Wide Web at http://Carnap.umd.edu:90/chps/faculty/Brush/relative_deprivation.
JOURNAL OF CONFLICT RESOLUTION, Vol. 40 No. 4, December 1996 523-545

? 1996 Sage Publications, Inc.
523

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
524 JOURNAL OF CONFLICT RESOLUTION
In this article, I assume that it is possible to pursue social science resear

at least comparable to, if not identical with, research in the natural scien
discuss whether this is desirable. I have chosen a particular episode from
of recent social science that seems to have some resemblance to famou
the natural sciences: a hypothesis, based on a widely accepted theoretical
developed with the help of relevant empirical data, is proposed in an explici
allows its predictions to be tested. The theoretical coherence of the hy
critically scrutinized by experts, its plausibility is compared with alternativ
ses, tests are designed and carried out, and the scientific community deliver
How does that verdict depend on the result of the empirical tests, and
depend on other factors? Do individual scientists change their views abo
or does the community view change largely through Planck's (1950
whereby younger scientists adopt the new view while older ones who sup
view just retire and die? How does this case compare with cases in the nat
where similar questions have been asked? In this way I want to contribut
effort made by historians and philosophers of science to test methodolo
against actual practice in different sciences.2
My case is the rise and fall of a theory of collective violence, ba
psychological concept of relative deprivation (the gap between what one h
one expects), developed by Ted Robert Gurr in the late 1960s. It was tested b
of empirical studies, most definitively by Edward N. Muller in the 1970s
was abandoned or substantially modified by Gurr and others who had su
relative deprivation is no longer considered the primary cause of collect
although it may be a significant contributing factor under some circum
theory has been replaced by other hypotheses involving, for example, th
dissidents to mobilize resources for action against the regime and their ra
of strategy to achieve a specific goal under particular social circumstance
It is essential to avoid the common mistake of expecting the social scien
standards that are not always met by the natural sciences. Thus one often fi
by philosophers of science and in a few popular works by scientists the stat
theories should be judged by their success in making predictions of
unknown facts. It is stated or implied that it would be unscientific to acc
a theory for nonempirical reasons. More generally, according to Karl Po
and his followers, a theory is not scientific unless it can be tested and possi
But creationist and astrological theories have been repeatedly falsified b
evidence; their advocates do not claim that they should be accorded scien
because they have been falsified, and many of their opponents reject them
that are independent of empirical tests. More to the point, in particle
electro-weak theory was accepted by many physicists for theoretical reas
was empirically confirmed, according to Steven Weinberg (1992), where
plasma theories of Hannes Alfven were rejected by most of the relevant
2. See Brush (1995) and other works cited therein. For Planck's principle, see Planck
and Hull, Tessner, and Diamond (1978).
3. A general history of this topic is presented in a monograph by Rule (1988); see also
Gumey and Tiemey (1982), Lichbach (1989), Muller and Weede (1994), and Morris and H

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Brush /DYNAMICS OF THEORY CHANGE 525
despite the confirmation of several of his predictions, even after he was aw

Nobel Prize (Brush 1990d).
Considerable confusion has resulted from ambiguous use of the word pre
and disagreement about what kind of prediction scientists should use. Acco
Popper (1959) and others, the best way to test a theory (and, at the same
demonstrate that one is being scientific) is to deduce from it predicti
phenomena that have not yet been observed. It is claimed that such predi
successful, provide better evidence for a theory than deductions or retrodi
phenomena previously known. The assumption is that a clever theorist can ad
variables or otherwise manipulate a theory to make its deductions fit any g
Thus, to keep the theorist honest, one must require the prediction of unknown
at least give more credit to prediction than to retrodiction, other things be
Writers on the methodology of the social sciences have endorsed this rule.4
as it may seem, this prediction rule does not seem to correspond to the actua
of scientists in evaluating theories. First of all, it is impossible to follow t
sciences, such as evolutionary biology, historical geology, and much of astr
where the phenomena being studied take place over such long periods of ti
significant prediction about future events could not be tested within the lifetim
predictor. Nevertheless, we do not call those fields unscientific.5 Second, sc
frequently use the word predict to mean deduce, without reference to whe
predicted phenomenon was previously known. We must therefore use the ph
prediction to specify prediction of a previously unknown phenomenon. M
scientists do not give any greater weight to novel than to nonnovel pred
although successful novel predictions may make a great impact on the public
scientists to pay more attention to the theory (Brush 1989).
I have not surveyed the linguistic usage of social scientists to determine
they mean novel prediction when they talk about prediction, but I can poin
examples relevant to the case studied here. In his influential book on relativ
tion, W. G. Runciman (1966, 7) asserted, "It is immaterial to the validity
explanation or correlation whether it was predicted beforehand or not." In a m
on theories of revolutions, Salert (1976, 18) stated that although theories sho
testable predictions, they "don't have to be predictions about the future
events. They may simply be predictions of statistical regularity." Finally, in
titled "Forecasting Internal Conflict," Gurr and Lichbach (1986) stated th
approach is actually "postdictive" because their model is tested on events
already occurred. They justified this procedure by pointing out that the "fore
based on data from an earlier period. Thus forecast, like predict, can refer to
known phenomena, as long as they were not explicitly used in making the f
4. See Berelson and Steiner (1964), Braybrooke (1987), Hutchison (1960), Social Science
Council (1954), Utter and Vanderleeuw (1991), and Zald (1991). For dissenting views, see Ru
who says prediction does not have to be novel, and Maclntyre (1971), who rejects predictivene
as a criterion.
5. Popper (1974,1978) himself contributed to the confusion by first asserting that Darwinia
is not a scientific theory because it is not falsifiable, then changing his mind (and his definition of
and declaring it a legitimate scientific theory after all.
6. This is my surmise, not a statement found in the Gurr and Lichbach (1986) paper.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
The only time novelty is clearly relevant is when the parameters of a mod
been selected to agree with previously known facts; then one has good reason
skeptical. Thus Tilly (1971, 419), in reviewing Gurr's Why Men Rebel, asserted
the variables had been reformulated during the research so "there is a serious ris
they have been adjusted to fit the data" and hence "only the application of Gurr's
determinate models ... to new cases will test them properly."
HISTORICAL SKETCH
Why do men7 rebel? In the decades and centuries before Gurr publish
proposing an answer to this question (Gurr 1970), historians and social sc
discussed a wide variety of theories of collective violence (revolutions,
riots, lynch mobs), social movements, and aggressive political participati
first half of the 20th century, American sociologists sought a unified ex
these phenomena under the rubric of "collective behavior." Such behav
disturbed the normal stable equilibrium of society, was thought to be in
socially isolated individuals acting emotionally. In some theories, instigat
others by a contagious process of "symbolic interaction" to form a diso
dangerous crowd. This viewpoint became known as the "riff-raff' or "s
earth" theory (Rule 1988). The theory of collective behavior maintained t
that any kind of rebellion, violent or not, is to be ascribed to the actions
people and represents a threat to society. Social science research on collect
and "mass society" is therefore needed to facilitate the control of violent or
violent movements.
In the 1960s, these theories of collective behavior lost their popularity among
academic social scientists at the same time that they were being employed by political
and military leaders to rationalize their policies. Riots in the black ghettos of American
cities were initially blamed on marginalized subversives or outside agitators.9 Yet
research studies had already begun to suggest that the participants in riots, like
participants in collective violence in other times and places, had the same social and
economic characteristics as the rest of the local population and had close ties with it.l?
Leftist uprisings and guerrilla movements in other countries were portrayed as part of
a worldwide Soviet-led conspiracy against the United States, yet somehow American
intervention always seemed to support repressive tyrannical regimes. The civil rights
7. Much of the social science literature, which implies that rebellions were organized and staffed
primarily by men, does not recognize a need either to verify or explain that assumption. For exceptions, see
Ferree (1980) on support for feminism and Wheeler (1975) on participation in German leftist movements.
Historians more often have recognized the role of women in rebellions; see, for example, Thompson (1971)
on the instigation of 18th-century riots.
8. See Moshiri (1991) and Rule (1988). The last phrase has been popularized by Edward Muller (1979
and other publications).
9. See especially the McCone Report (Governor's Commission on the Los Angeles Riots 1965) and
the extensive discussion of it by Fogelson (1971). McCarthy and Zald (1973) note the relation of resource
mobilization theory to the outside agitator theory.
10. See Caplan and Paige (1968), Fogelson (1971), Platt (1971), Wanderer (1968), and C. Tilly, "Queries
on Social Change and Political Upheaval in France," cited in Rule (1988, 174).

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Brush/DYNAMICS OF THEORY CHANGE 527
movement and protests against American participation in the Vietnam war w

by many academics not as irrational crowd actions against society but as rat
justified attempts to make government responsive to the needs and interest
citizens."
American social scientists in the 1960s developed several alternative theori
collective violence. Smelser (1962), following the sociological tradition of P
(1937), argued that collective behavior, though representing a deviation from
tional or normal social behavior, could be analyzed using similar categories
norms, organization, and situation-combined with structural strain and cond
ness. Olson (1965), starting from the perspective of economic theory, deve
theory of "rational choice" based on individual self-interest, which explained
type of grievance often invoked to explain collective violence usually does no
such action. Tilly (1969; see also Rule 1988, ch. 6), on the basis of a historical
of violent events in France, argued that collective violence is a normal par
struggle for power among contending groups in a society; it occurs not when
are especially discontented but when they calculate that action will be success
benefits that exceed costs. Tilly became the leading advocate of the resource
zation theory, which was the major challenger to relative deprivation theory
1970s (Snyder and Tilly 1972; Tilly, Tilly, and Tilly 1975; Tilly 1978). In their di
ways, these theorists moved away from a conception of collective violence
unpredictable consequence of random irrational behavior toward a view of
as only a special case of rational-and thus presumably predictable-goal-
social behavior (Rule 1989).
In the 1960s, too, biological and psychological explanations of social and political
behavior were becoming popular again after a period in which cultural or environ-
mental explanations of behavior had been dominant.12 Some scientists thought that
nothing more than a fundamental aggressive instinct was needed to account for the
prevalence of violence in human and animal species.'3
Ted Robert Gurr, then an assistant professor of political science at Princeton
University, considered this type of explanation too facile and looked instead to
psychology for an understanding of the situations in which aggressive behavior occurs
(Gurr 1968, 1970). He argued that the potential for collective violence depends on the
discontent of members of a society; discontent is a result of a perceived gap between
what they have and what they think they should have. This gap is called relative
deprivation and provides the motive for action through the psychological "frustration-
aggression" mechanism.
11. See Gamson (1975). The ambivalent views of social scientists toward research intended to protect
the stability of the social order are shown in their reflections on "Project Camelot" (Horowitz 1967). An
extreme example of the interaction of personal feelings of grievance with research on protest movements
is the paper by Law and Walsh (1983, 124); the first author, one of the founders of the protest movement
against a dangerous "Toughman Contest" (a series of fights in which "the only rules are that there should
be no biting and kicking" and which allowed contestants of greatly unequal weights and physical condition
to be matched against each other), intentionally helped to create the events being studied.
12. See Berelson and Steiner (1964) and Degler (1991). For examples relevant to political science, see
Coming(1971), Dahl (1961), Davies (1963), Pennock and Chapman(1977), Schwartz (1974), Somit (1968),
and Stone (1981).
13. See Ardrey (1966) and Lorenz (1966).

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
528 JOURNAL OF CONFLICTRESOLUTION
Before 1970, the concept of relative deprivation (RD) was already being
used by social scientists.14 The idea that political revolutions stem from fr
expectations (originated by Aristotle and succinctly expressed in de Tocq
writings on the French Revolution) had been revived in the 1950s and 1960s
why instability followed modernization and protest followed apparent pro
phrase "revolution of rising expectations" became a cliche that endures to
It was visually summarized in the "J-curve" of Davies (1962, 6) by showin
sudden drop in "actual need satisfaction" causes it to lag much farther b
continually growing "expected need satisfaction," thus creating "an intol
between what people want and what they get."16 Conversely, RD was cons
explain why severe hardship during the depression of the 1930s did not
revolution (Maclntyre 1980).
Gurr built on these earlier works but went beyond them in one crucial r
insisted on developing a body of precise, testable hypotheses. Early in Why M
Gurr (1970, 15-17) wrote:
The concepts, hypotheses, and models of causes and processes developed in the fo
chapters are not intended as ends in themselves. Intellectually pleasing filters t
which to view and categorize the phenomena of a disorderly world are not know
Systematic knowledge requires us to propose and test and reformulate and r
statements about how and why things happen. We know enough, and know it well e
only when we can say with some certitude not just why things happened yesterd
how our actions will affect what happens tomorrow....
The kind of theory aimed at in this study is an interrelated set of general, fal
hypotheses that satisfy causal or concomitant relationships between independen
dependent variables....
The central scientific criterion for theory is that it be subject to empirical assessm
Four attributes of theory that facilitate its assessment are its falsifiability, defi
clarity, identification of relevant variables at various levels of analysis, and appli
to a large universe of events for analysis. The first two are necessary conditi
assessment, the others desirable.
Gurr suggested that RD could be measured by surveys, such as the one c

by Hadley Cantril (1965), who asked respondents in 14 countries to comp
11-point scale their past, present, and future situations with their desired
Comparing the national averages with his own estimate of the "magnitude o
in these countries (omitting the United States), he found a rank-order corr
0.59 between RD measured circa 1960 and turmoil in the period between
1965 (Gurr 1970). Of the countries found to have greater than the median
relative deprivation-Dominican Republic, India, Poland, Brazil, Nigeria, an
ama-all but one, Poland, also had greater than the median amount of turm
In a footnote, Gurr (1970, 65) stated that "data for the United States ar
from this comparison on grounds that its high levels of turmoil in the 19
attributable to the segmental RD of black Americans, which are inaccurate
14. See Geschwender (1964) and Runciman (1966, 1989).
15. According to Calvert (1970, 112), this was the "dominant non-Marxist sociological"
time.
16. Although the theories of Davies and Gurr often were lumped together in citations, Davies (1973)
distinguished his theory from Gurr's by its emphasis on the time dimension.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
sented in any summary national measure of RD." The implication was tha
Americans were more likely to rebel than Whites because they had stronger
of relative deprivation, though elsewhere in his book Gurr called attention
other variables that would have to be controlled to establish such a direct correlation.
But Gurr's use of aggregate measures of deprivation as indicators of individual motivation
was widely regarded as a serious weakness of this theory, along with his failure to show
exactly how individual motivations are translated into social movements.
Why Men Rebel won the Woodrow Wilson Prize for the best book on political
science in 1970 and was widely praised in reviews. Even those critics who disagreed
with Gurr's theory applauded his effort to formulate it as a set of propositions that
could be tested. 7 The publication of Gurr's book at the same time as a number of
other works advocating a similar viewpoint seemed to provide an effect of mutual
reinforcement. 1
Within 6 months of the publication of his book, Gurr was asked by the American
Political Science Review to referee a manuscript that presented results contrary to his
RD theory. Edward N. Muller, a political scientist at SUNY-Stony Brook, had studied
the racial disturbances in Waterloo, Iowa, in the late 1960s, using the Cantril Self-
Anchoring Striving Scale recommended by Gurr. Muller (1972, 954) found that the
potential for violence depends on a "low degree of trust in political authorities"
combined with a "high degree of belief that the use of violence by dissident groups in
the past has helped their cause.... Once these factors have been taken into account,
relative deprivation does not seem to matter." If society wants less violence, it would
be more effective to have honest police than more welfare programs.
Gurr wrote in his referee's report that if the author's results turned out to be
generally valid, his own theory would have to be fundamentally revised. He now sees
this episode as the crucial event that led him to downgrade the importance of relative
deprivation as a cause of collective violence."9
The numerous rebellions, riots, and civil disturbances in the 1960s coincided with
the establishment of surveys and archives to collect data on such events and an
explosive growth in the number of social scientists interested in analyzing the data.
Hence there seemed to be many opportunities to test Gurr's theory. One of the most
frequently cited tests was an analysis by Clark McPhail, a sociologist at the University
of Illinois, of 10 reports on five riots in American cities (Watts in Los Angeles, 1965;
17. See Feierabend (1971), Marx and Wood (1975), and Silvert (1971). A decade later, Greene (1984,
216) called relative deprivation the most often cited theory of revolution and identified Gurr as its leading
theorist.
18. See Davies (1973); Feierabend, Feierabend, and Nesvold (1969, 1973); Feierabend, Feierabend,
and Gurr (1972); Fogelson (1971); Huntington (1968); Pettigrew (1971); Runciman (1966); and Forster and
Greene (1970). Relative deprivation was invoked so frequently that it was called a "sociological workhorse"
by Young (1970). Wilhelm ( 1970, 82) complained that the concept was being used "to bridge the gap between
the positively elated exclamations by academic observers and many Negro leaders [about the supposed
progress made by Negroes in the United States] and the mounting dissatisfaction coming especially from
the Negro masses" because, according to Wilhelm, their conditions had not improved. Thus, he argued, the
"rising expectations" thesis distracts attention from legitimate grievances. Similarly, Wasbum (1974, 333)
suggested that preoccupation with variables such as RD "deflects attention from objective structural
conditions which might precipitate violence and from the social organization and social-political conse-
quences of protest movements."
19. Ted Robert Gurr, telephone interview by the author, 18 December 1992.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Omaha, 1966; Detroit, 1967; Milwaukee, 1967; Newark, NJ, 1967). He conclude
the RD model "receives scant empirical support" and should be rejected (McP
1971, 1064). Instead of relative deprivation, the factor showing the highest corre
was the opinion of Blacks (not necessarily based on personal experience) tha
were being mistreated by police.20
Other studies of the 1960s ghetto riots in American cities yielded only weak s
or refutations for the RD model. Spilerman (1970, 1971) compared the distur
in several cities and found, after controlling for the size of the non-White popu
no significant correlation between relative deprivation (measured by aggrega
cators) and the amount of violence. A survey of residents of Cleveland's poor
months after the 1966 riot obtained mixed results (Bowen et al. 1968), as did
after the 1967 Detroit riots (Crawford and Naditch 1970). In both cases, the resea
decided that additional factors were needed to nail down a connection between
deprivation and propensity toward violence. Miller, Bolce, and Halligan (1977)
concluded that the J-curve theory failed when data on family income and years of
schooling for individuals rather than averages for groups were used.
Multinational comparative studies of the kind used by Gurr himself continued to
support a correlation between political instability and aggregate measures of depriva-
tion (Feierabend, Feierabend, and Nesvold 1973), but more detailed local or longitu-
dinal studies of individual countries began to cast doubt on the efficacy of individual
discontent as a cause of collective violence. The publications most often mentioned
for providing evidence against the RD model were those of David Snyder and Charles
Tilly (1972) for violence in France during the period from 1830 to 1960 and of
Alejandro Portes (1970, 197 la, 1971 b) for leftist radicalism in modern Latin America.
Portes (1971b, 26) interpreted "the popularity of the frustration-radicalism hypothe-
sis ... as a partial result of the post-factum self-legitimation of successful revolution-
ary movements." It is not surprising that those who have already participated in a
revolution claim that their action was motivated by economic deprivation or some
other grievance. The claim is not evidence of their real motivation, especially if others
with similar grievances did not revolt. Many of these studies were subsequently
considered inadequate tests of the RD model because they did not directly measure
the perceived discrepancy between one's "just desserts"-a legitimate expectation of
reward rather than a desired but unrealistic goal-and one's actual level of attainment
and because they measured only the respondents' attitudes toward collective violence,
not their actual participation. These gaps were filled in an extensive series of studies
in New York City, West Germany, and Costa Rica by Muller and his collaborators.
Muller found that although there is a linear correlation between just desserts frustration
(JDF) and aggressive political participation (APP), this relation is entirely due to other
intervening variables. In particular, he concludes that APP is produced by a belief in
20. McPhail (1971) cites the report of Flaming (1968) as the primary source forthis conclusion. Flaming
himself admits that his study cannot establish the motives for participation in riots because he did not measure
attitudes of participants before they were arrested. His data indicate that Black arrestees blamed economic
deprivation almost as often as police mistreatment for their anger. For other research indicating that rioters
blamed police brutality more than deprivation, see Fogelson (1971) and Ladner et al. (1981).

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
the efficacy of action and doubts about the legitimacy of the political syste
1977, 1979, 1980; Muller and Jukam 1983).
Although Muller's research persuaded Gurr and others that relative depriv
not the primary cause of collective violence, there was some retreat in the 198
the flat statements in the 1970s that the RD model had been completely ref
view of Muller and other researchers seems to be that relative deprivation
contributing factor that interacts with or amplifies other factors.21 Even thou
usually fails to explain as much of the phenomenon that it is intended to predict

investigators had hoped it might... it is evidence of the theoretical power of the n
that researchers return to it again and again in the hope that by making tec
adjustments to its measurement and small conceptual refinements in its use,
empirically realize its theoretical promise. (Sanders and Tannenbaum 1983, 49)22
Remarkably, RD theory became a flourishing research topic among soci

chologists just as political scientists and sociologists were abandoning it. This
interest in a concept originating in their field seems to have begun with Faye
review (1976) of empirical evidence for several propositions derived from
model (Olson and Hazlewood 1986). Although she cites the studies of Mulle
Portes (197 Ib), Snyder and Tilly (1972), and others as evidence against the m
finds many more studies to support it. Although psychological research is m
directed toward correlating RD with feelings of discontent and resentment t
occurrences of collective violence and thus overlaps only partly with the
domain of the political scientists, there still seems to be a sharp disagreement
psychologists and political scientists (and especially sociologists, as will be
the following section) about the causes of protest movements.23
The replacement of RD by resource mobilization (RM), social-process, stru
world systems, and new social movement theories as the dominant explan
among political scientists and sociologists for collective violence has been
nied by a methodological shift away from quantitative or predictive hypothesi
toward qualitative or descriptive explanations of social phenomena.24 Thus pr
21. Ted Robert Gurr, telephone interview by the author, 18 December 1992; see DeFron
Ferree and Miller (1985), Kent (1982), Klandermans (1984), Midlarsky (1988), Midlarsky an
(1985), Muller (1985), Muller and Seligson (1987), Zimmerman (1983), and Zurcher and Snow
22. One such refinement proposed by several researchers was to distinguish between the tw
RD defined by Runciman (1966): "egoist" (failure to meet personal expectations) and "fraterna
of one's group relative to other groups). Some studies suggested that fraternal RD was more lik
to collective violence than egoist RD, and earlier failures to find correlations between RD and
violence could be ascribed to the fact that only egoist RD had been measured (Geschwend
Geschwender 1973; Guimond and Dube-Simard 1983; Dub6 and Guimond 1986).
23. A good survey of psychological research may be found in the proceedings of a 1983 sy
(Olson, Herman, and Zanna 1986). The disagreement about the cause of protest movements is d
the article by Dub6 and Guimond (1986) presented at that symposium. It should be noted that t
most of the comprehensive research of Muller (1979), which some political scientists and soci
consider the best refutation of RD theory. Crosby, Muehrer, and Loewenstein (1986) express dis
with the progress of their search for a simple predictive model.
24. See Cohen (1985), Halebsky (1974), Jenkins and Shock (1992), Maclntyre (1973), Mo
Mueller (1992, pref.), Muller (1992), Tiemey (1988), Tilly (1984), and Zald (1991, 1992). One
so far as to deny that any causal model for political violence can exist, because any factor pro
cause must itself be a result of earlier social change (Terwilliger 1971). Laslett (1992) criti
attempts to be scientific as masculinist "scientism" (see also Ferree 1992).

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
of RM and other theorists have less interest in Popperian-style conject

refutations25 and pay little attention to experiments designed to test their
when they turn out favorably.26 Because they have changed not only the th
criteria used to evaluate theories, one perhaps may use here the much-ab
paradigm switch.27
HISTORIOGRAPHIC METHOD
In asking why the social science community rejected the RD theor

violence, I am undertaking what historians of science call a "receptio
studies rarely have been done, except for important theories that ha
by the scientific community, such as Einstein's relativity or Darw
Normally, historians merely assert that a theory was accepted or re
the outcome of a particular experiment or argument or social factor, w
to cite more than one or two scientists to support that assertion.28
For several reasons, the reception of Gurr's RD theory of collective
a good opportunity to study the dynamics of theory change in scienc
First, if the historical sketch presented in the preceding section i
accurate, the RD theory was widely accepted at one period of time a
later time. Thus it is an example of theory change.
Second, because the problem of "collective violence" was of widesp
and practical interest in the 1960s and 1970s, a large amount of publi
available. Therefore, one can hope to draw conclusions about the response of a
scientific community as a whole, rather than from merely a few specialists. To a limited
extent, one also may be able to distinguish the responses of scholars from different
academic disciplines.
Third, the theory change occurred long enough ago to permit one to determine how
it turned out. The outcome was not simply the replacement of RD theory by a
competing theory but, following that, the emergence of a more balanced view in which
RD plays a small but significant role rather than being either of primary importance
or completely rejected. Review articles and retrospective accounts (including inter-
views with participants) thus tend to articulate the pros and cons of RD theory more
thoughtfully than works published in the heat of battle.
25. See Kiser and Hechter (1991), Lichbach (1989), and Piven and Cloward (1992). This does not apply
to Muller, whose social-psychological model remains a strong contender and is supported by empirical
evidence (Rule 1988). Greene (1984) and Zagorin (1982) assert that we will never have an adequate general
(falsifiable) theory of revolutions. Recent articles in the journal Contention question whether theories of
revolution should be expected to predict whether revolutions will occur in specific countries (Keddie 1995).
26. See Martin, Brickman, and Murray (1984). Another kind of test, which led to the conclusion that
neither RD nor RM theories would have predicted the Iranian revolution, but a theory developed by Hibbs
(1973) would have done so (Kennedy 1984), seems to have been ignored by advocates and critics of all
three theories.
27. See Jenkins and Shock (1992), Ramirez (1981), Wasbum (1974), and Zimmerman (1983).
28. Similarly, according to Stephen Cole (1992), scholars who claim that social processes influence the
cognitive content of science have failed to produce a single example in which this alleged causal influence
is demonstrated by the standards of sociological research. In his review of Cole's book, Shapin (1993) quotes
this assertion but does not refute it with any specific examples.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Fourth, the relevant publications are recent enough that the Social Science Cita
Index (SSCI) permits one to retrieve much of the relevant journal literature show
how the community responded to the theory and its empirical tests. Use of the
is supplemented by a search of other literature using more traditional methods.
Fifth, the theorist himself (as quoted in the previous section) sharpened the is
and perhaps expedited the process of theory change by formulating testable hypo
ses. Thus, even if this case is not typical of theory change in the social sciences, it m
at least provide an important example of the kind of behavior that some philosop
and some social scientists regard as scientific.
In using citations to assess the reception of a theory, it is essential to recognize t
the response of the community cannot reliably be measured simply by counti
citations. Thus, in the case of physical science and mathematics, when one consid
the authors of works published a few decades ago who have been judged by th
community by the award or nonaward of the Nobel Prize (or Fields Medal, in the
of mathematics), it is found that some important works received relatively few citat
and that some of the most frequently cited works were not important reports of n
research (Brush 1990a, 1990b, 1990c).
Rather than simply counting citations, I actually examined most of the articles
books that cited Gurr's Why Men Rebel during the 15 years after its publication to
out whether the authors accepted or rejected his theory and for what reasons (if an
In addition, as indicated later, I looked at comments on RD theory published in t
preceding half-decade (1966-1970) to establish the status of this theory just before
publication of Gurr's book and identify authors who changed their views after
publication and testing.
Because SSCI concentrates on journal articles, I have supplemented it by look
at books cited in journal articles or listed in annual bibliographies and in the sub
index of the Princeton University Library under headings such as "revolution" a
"social conflict." I also included a number of publications (especially before 197
that mentioned the application of RD theories to collective violence but did not
Gurr's 1970 book. By these methods I have increased the database by about 20%
beyond that provided by SSCI.
In addition to searching for works that cite Gurr or RD theory, I also surveyed m
of the publications indexed under "revolutions" or "riots" in International Poli
Science Abstracts, Sociological Abstracts, Public Affairs Information Service, and
bibliography published by Morrison and Hornback (1976), concentrating on wo
that appeared in 1971, 1976, and 1981. The purpose here was to see what competi
theories were being discussed and estimate the popularity of Gurr's theory relativ
those theories.
RESULTS
During the 15 years after its publication, Gurr's book Why Men R
mentioned in at least 649 publications, of which 526 were listed in the Soc
Citation Index; the remaining 123 were found by my literature search. O

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
citations, 145 were considered marginal or neutral; that is, they did n
substantive remarks that could be considered either favorable or unfavorable. After
discarding the marginal citations and self-citations, I found that 181 were favorable
and 192 were unfavorable. The proportion of favorable citations was 56% from 1971
to 1975; it declined to 42% from 1976 to 1980 and then rose to 47% from 1981 to
1985. However, the total number of citations reported in SSCI showed the opposite
pattern, rising from 187 between 1971 and 1975 to 197 between 1976 and 1980, then
falling to 142 between 1981 and 1985. (This illustrates the fallacy of using citation
counts by themselves to assess the reception of a theory.)
The number of unfavorable citations is quite high compared with other studies that
found only 3% to 14% negative citations. The difference is presumably due to the fact
that most of the community ultimately rejected Gurr's theory, whereas the other studies
involved citations of authors whose work is still accepted or disciplines in which
rejected theories are not usually cited at all.29
I was especially interested in identifying authors who changed their opinion of RD
theory and in finding out why they did so. So far I have located only seven authors
who changed from favorable to unfavorable. As mentioned earlier, Gurr himself began
to have serious doubts about his theory when he read Muller's (1972) report on the
Iowa riots. However, he did not publicly abandon the theory but instead introduced a
series of modifications giving greater importance to other factors.30 Corning (1971)
had given a mostly favorable reference to Why Men Rebel but later rejected RD theory
because he said that it explained only a small part of the variance (he did not cite
specific empirical tests) and did not explain why most relatively deprived individuals
don't rebel (Corning 1975). Firestone (1974), who had cited Gurr favorably in 1972,
presented theoretical criticisms31 of Why Men Rebel in a review but did not mention
any empirical tests. Mitchell was coauthor of a study that used Gurr's model (Noton,
Mitchell, and Janes 1974) but 6 years later was a coauthor of a paper that rejected RD
because it could not be properly measured (Bowers, Mitchell, and Webb 1980). Useem
changed his favorable earlier (1972) assessment, citing empirical evidence against it
published by Snyder and Tilly (1972), Hibbs (1973), Spilerman (1970), McPhail
(1971), Portes (1971b), and Muller (1972). He also objected that the theory ignores
violence by the state against dissidents (Useem 1975).32 Williams (1975a) cited Gurr's
theory favorably though with some reservations. Then, in a longer article published in
the same year, he stated that RD theory is inadequate to explain collective action
29. Adatto and Cole (1981) found that only 3% of the citations to Max Weber in four leading sociology
journals in the period from 1964 to 1974 were negative; Cole (1975) found that only 6% of the citations to
Robert Merton in connection with deviance research were negative.
30. Gurr and Lichbach (1986, 10) state that although Gurr's theory was supported by cross-national
aggregate data, "microlevel studies of relations between perceived deprivation and participation in collective
action provided much weaker support for the theory" (reviewed in Muller, 1980). It was revised to take
account of Tilly's RM viewpoint.
31. Firestone says Gurr doesn't explain how to tell which "value" variables (economic, political,
educational, etc.) are most significant; he relies too heavily on the frustration or aggression mechanism to
explain violence and not enough on "instrumental" (goal-seeking) behavior; he equates "aggression" with
violence; and his treatment of incentive systems and social conditions is too narrow.
32. But Useem (1976) again published a favorable though brief reference to Why Men Rebel, on the
influence of regime legitimacy on attacks against it.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
(Williams 1975b). Finally, Zimmerman shifted from a slightly favorable stanc

to a slightly unfavorable view of RD (1980), mentioning the negative eviden
arguments published by Freeman (1972), Salert (1976), Marx (1972), Muller (
and Tilly (1975).33
I classified the unfavorable citations according to whether they mentioned e
cal evidence against the RD theory. (Very few of the favorable citations men
empirical evidence for it but simply used the theory to account for various phen
The proportion of unfavorable citations supported by empirical evidence was
47%; it exceeded 50% only in the later 1970s.
The most often cited empirical evidence against RD theory was that report
McPhail (1971) in an analysis of 10 reports on ghetto riots in American cities
1960s (see the "Historical Sketch" section), cited 22 times. The various publi
by Edward Muller and his collaborators, which provided the most definitive e
against the theory according to Gurr and to the previous assessment of Rule
were cited almost as often (19 times) as were the works of Charles Tilly an
collaborators (19 times).
Because about half of the unfavorable publications mentioned no empirical
for rejecting the theory, and many others gave nonempirical as well as em
reasons, it is necessary to provide a summary of the nonempirical reasons to und
why the community rejected the RD theory. I have classified the nonempirical
into a number of categories that, although obviously subjective and partly overla
seem to give a reasonable description of the criticism. In each case, I cite one
examples of that kind of criticism, usually authors who present the criticism
detail.
Two nonempirical reasons were each mentioned even more often than th
often cited empirical studies:
1. The measurement of RD is indirect and unreliable [mentioned 31 times],
the concept itself is hard to use [4]; the data do not really tell us about the mo
of participants (Gurney and Tierney 1982; Wilson 1973). The theory does n
clear the reference group relative to which deprivation is perceived [3]. Gurr
use appropriate methods for measuring the magnitude and kind of violence, w
consequently mismeasured (Drew 1974; Kramnick 1972) [8].
2. The theory suffers from an individual/collective gap [29]: it fails to expl
individual feelings of deprivation are transformed into collective action (Gurn
Tierney 1982; Feagin and Hahn 1973). Some critics, who reject reduction
principle [11], deny that this gap could ever be closed: social phenomena can n
explained satisfactorily in terms of individual psychology but must always i
factors peculiar to the structure of society; this makes it doubtful that any
theory of collective violence is possible (Skocpol 1976).
Several other reasons for rejecting Gurr's theory were given less often:
3. The critic prefers resource mobilization theory (Tilly 1973) [18] or ano
competing theory [13]. A theory explaining collective violence in terms of mo
33. But Zimmerman's (1983) rejection of RD theory was only temporary; his assessment i
favorable.
34. Subsequently, numbers given in brackets refer to the number of times a reason is given.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
536 JOURNAL OF CONFLICTRESOLUTION
is superfluous because in any society there is always a prevalence of

adequate to fuel rebellion [8]; what needs to be explained is how disconten
organize their actions in a particular society and whether the governmen
suppress them (McCarthy and Zald 1976; Oberschall 1973). In general
slighted by RD theory (Zagorin 1973) [11].
4. The theory attempts to explain violence against government and society
to deal with the role of government violence that may be more importan
situations (Berman 1976) [12]. A related complaint is that the RD perspect
a bias in favor of stability: it assumes that the lack of violence, whether or n
enforced by state authorities, is "normal," but the behavior of dissidents is i
or abnormal and thus in need of explanation [9]. There is too much stress
of emotional or irrational factors, not enough on collective violence as a
rational goal-seeking activity (Barkan 1980) [6]. Gurr's book is thus seen a
telling political and military authorities "how-to-keep-them-down" (Papck
5. There are exceptions to the RD hypothesis: RD sometimes does not
rebellion, and rebellions sometimes happen when there is no RD [13]. More
the individuals who participate in rebellions are less deprived than those
society who are passive (Corning 1975). This sounds at first like an empiric
but often turns out to be based on only a few individual cases, ignoring th
evidence that Gurr urges as the primary support for his theory. Similarly, R
are accused of the postfacto fallacy: they start from a revolution that actual
and work back to its supposed causes, but they cannot predict from given
whether a revolution will occur (Rule and Tilly 1972). They complain that
criterion for causing rebellion (Aya 1979) [3]; people may actually be ple
others doing better, though eventually (when?) they will become angry if
get their share (Hirschmann 1973).
6. A positive correlation between relative deprivation and the amount o
does not prove that the former causes the latter; it could mean that peop
deprivation as a result of their participation in collective violence (Portes
we don't know if violence is cause or effect [7].
7. RD theorists give no adequate account of the time developmen
processes [7]; they fail to measure the relevant variables over significant t
(Rule and Tilly 1972; Davies 1973), ignoring past history that might prov
dents [4] and later history showing the consequences [5] of collective viole
1976; Terwilliger 1971). They also ignore external factors such as the actio
countries (Kerbo 1978) [5].
Aside from empirical or nonempirical evidence, another dimension that
be especially important in this case is the disciplinary allegiance of social s
35. Conversely, Gurr's theory could be used by a critic of U.S. counterinsurgency policy i
argue that escalating the repression of dissidents (e.g., by extensive bombing as advocated by
would lead to more violence in response, contrary to cost-benefit calculations based on a ra
theory (Schultz 1979).
36. In those cases where these affiliations were not clearly stated in the publication itself, I u
reference books such as the National Faculty Directory and Dissertation Abstracts Internat

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Of those whose training or current positions could be identified, approximately

of the authors who cited Gurr's Why Men Rebel [ignoring marginal citations] w
political scientists, 30% were sociologists, 15% psychologists, and the remaining
distributed among history, economics, and other fields. I divided the period from 1
to 1985 into three 5-year periods and counted the number of authors within each pe
who commented favorably or unfavorably.
For political scientists, about 75% were favorable from 1971 to 1975, declining
about 50% from 1976 to 1980 and remaining at that level from 1981 to 1985. Th
figures may be considered an upper limit to the proportion of political scientists
favored RD theory because presumably those who rejected it might not cite it at
especially in the 1980s, when the theory was no longer a major contender. The t
number of (favorable or unfavorable) citations by political scientists was about 11
year during the 1970s, dropping to about 6 per year in the early 1980s.
Sociologists followed a somewhat different pattern: the percentage of favorab
citations declined from 45% between 1971 and 1975, to 30% between 1976 and 19
and to 20% between 1981 and 1985. The total number of citations was about 8
year during the 1970s and almost as high in the early 1980s.
Psychologists, though providing a much smaller number of citations, were m
more enthusiastic about Gurr's theory, maintaining a uniform level of nearly
favorable comments. There was a substantial increase in the total number of both
favorable and unfavorable citations in the early 1980s. As mentioned earlier, many of
these psychology publications were not concerned with collective violence so much
as with nonviolent protests and other forms of discontent.
Finally, thanks to the Biographical Directory (American Political Science Associa-
tion 1988), I could determine the birth years of many of the political scientists. This
information could be obtained much less completely from several sources for econo-
mists, historians, sociologists, and psychologists, so those groups were lumped to-
gether. The data are not yet complete enough to support any weighty conclusions, but
they suggest that support for RD theory was somewhat more likely to come from those
political scientists born before 1936 and less likely from those born later (especially
sociologists). This would be consistent with Planck's (1950) principle only if one
regards RD as the "old" theory while RM and other competitors are seen as "new"
theories appealing to the younger generation. But contrary to that interpretation,
perhaps the most remarkable (though very tentative) finding is that nearly all those
born in the 1950s support RD theory.
My survey of the literature on revolutions and riots indicated that only a small
fraction of this literature cited Gurr's work or indeed mentioned any theoretical ideas
about the causes of collective violence at all. Most publications focus on describing a
particular event or the experience of a particular country rather than explaining the
phenomena in general terms. If one gave due weight to the non-American literature,
one would probably find that Marxist or neo-Marxist theories are more frequently
mentioned (though often without a specific citation that would be retrieved through a
citation index) than RD theories.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
CONCLUSIONS
Walter Laqueur (1977, 10) wrote that the net result of applying polit
collective violence is negative: "no truly scientific, that is predictive or
theories have emerged."37 In part, this was because Laqueur believed
theory is even conceivable and that, in particular, RD cannot be measur
because no quantitative index can reflect the qualitative aspect of huma
I would argue on the contrary that Gurr's theory was predictive and that
by empirical tests is part of the progress of science, not a failure. If r
show that a plausible explanation of a phenomenon is not in fact satisfacto
the way for other hypotheses to be developed and tested.
The reasons for rejecting RD as the primary cause of collective violen
empirical and nonempirical. The fact that many social scientists (inclu
gave nonempirical reasons for rejecting it does not mean that the
"unscientific." In the physical sciences, too (as noted earlier), a theory m
on theoretical grounds without regard to its empirical tests or accepted f
reasons before its empirical predictions have been tested.
But two of the nonempirical objections suggest that theories in the s
may be judged by harsher standards than those in the natural scienc
criticism that an assumed causal element such as RD is not directly me
not be considered a fatal flaw in physics or chemistry, where theories ba
properties of atoms were successfully pursued long before atoms cou
observed; even today, theories based on unobservable quarks are wide
Second, the fact that collective phenomena are reductively explained
individual properties, far from being a defect, would be considered a virt
science; thus, statistical mechanics has been acclaimed for explaining
phenomena such as the condensation of gas to liquid in terms of the
molecules.
More generally, a model in physics is not expected to take account of all factors
that might possibly be involved; on the contrary, it is widely recognized to be more
effective to focus on one or two important factors when first investigating a problem.
Yet numerous critics of RD theory dismissed it for failing to include additional factors,
without any attempt to show that those factors are as important as the ones Gurr did
include.
Planck's (1950) principle does seem to govern this case in one respect: I found only
a handful of social scientists who actually changed their views about RD on the basis
of either empirical or nonempirical reasons. For the most part, the supporters stopped
publishing on the subject (or continued to support it); those who rejected RD, or
adopted an alternative theory, had not supported RD in earlier publications. But the
37. Lichbach (1989) has a similar pessimistic conclusion about the research on inequality and political
conflict.
38. Compare with Rosenberg's (1952, 58) statement, in connection with Milton Friedman's defense of
"unrealistic assumptions" in economic models, that "the idealized character of assumptions through the
most successful theories of physical science, has long ago settled the question of whether unrealistic
assumptions are permissible in predictively successful scientific theorizing."

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
usual assumption behind Planck's principle-that younger scientists are more

to accept a new theory-is not clearly valid here. There is some indication that yo
researchers tended to reject RD in the 1970s, although the youngest (born in the 1
favored its (partial) revival in the 1980s.
Scientists from different disciplines displayed varying responses to RD th
Political scientists were initially enthusiastic but less so after 5 or 10 years. Sociolo
gave it a rather hostile reception, but psychologists were much more favora
Psychologists and sociologists continue to disagree on whether relative deprivat
a major cause of discontent and social movements.39 If one requires that the abi
reach eventual consensus on disputed questions is a criterion for being scientific
it would appear that these disciplines can be regarded as sciences separately bu
as part of a single social science.
REFERENCES
Abeles, Ronald P. 1976. Relative deprivation, rising expectations, and Black militancy. J
Issues 32 (2):119-37.
Adatto, Kiku, and Stephen Cole. 1981. The functions of classical theory in contempo
research: The case of Max Weber. In Knowledge and society: Studies in the sociology
and present, edited by R. A. Jones and H. Kuklick, 137-62. Greenwich, CT: JAI.
American Political Science Association. 1988. Biographical directory. Washington, DC: Am
Science Association.
Ardrey, Robert. 1966. The territorial imperative. New York: Atheneum.
Aya, Rod. 1979. Theories of revolution reconsidered: Contrasting models of collective violence. Theory and
Society 8:39-99.
Barkan, Steven E. 1980. Judging the judges. Human Rights 8 (4):34-7.
Berelson, Bernard, and Gary A. Steiner. 1964. Human behavior: An inventory of scientificfindings. New
York: Harcourt Brace.
Berman, Bruce J. 1976. Bureaucracy and incumbent violence: Colonial administration and the origins of
the "Mau Mau" emergency in Kenya. British Journal of Political Science 6:143-75.
Bowen, Don R., Elinor Bowen, Sheldon Gawiser, and Louis H. Masotti. 1968. Deprivation, mobility, and
orientation toward protest of the urban poor. In Riots and rebellion, edited by L. H. Masotti and D. R.
Bowen, 187-200. Beverly Hills, CA: Sage.
Bowers, D. A., C. R. Mitchell, and K. Webb. 1980. Modelling bicommunal conflict: I. Posing the problem.
Futures 12:473-88.
Braybrooke, David. 1987. Philosophy of social science. Englewood Cliffs, NJ: Prentice Hall.
Brush, Stephen G. 1989. Prediction and theory evaluation: The case of light bending. Science 246:1124-29.
. 1990a. The most-cited physical sciences publications in the 1945-1954 Science Citation Index.
Current Contents, 14 May, 7-17.
.1990b. The most-cited physical sciences publications in the 1945-1954 Science Citation Index.
Current Contents, 15 October, 8-13.
. 1990c. The most-cited physical sciences publications in the 1945-1954 Science Citation Index.
Current Contents, 22 October, 7-16.
1990d. Prediction and theory evaluation: Alfvtn on space plasma phenomena. Eos, Transactions
of the American Geophysical Union 71:19-33.
39. One reason for this may be that RD theory is viewed as a "reduction" of sociology to psychology,
which thus tends to annoy sociologists and please psychologists.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
. 1995. Dynamics of theory change: The role of predictions. In PSA 1994: Proceedin
biennial meeting of the Philosophy of Science Association, vol. 2, edited by D. Hull, M
R. M. Burian, 133-45. East Lansing, MI: Philosophy of Science Association.
Calvert, Peter. 1970. Revolution. New York: Praeger.
Cantril, Hadley. 1965. The pattern of human concerns. New Brunswick, NJ: Rutgers Univ
Caplan, Nathan S., and Jeffery M. Paige. 1968. A study of ghetto rioters. Scientific Ameri
Cohen, Jean L. 1985. Strategy or identity: New theoretical paradigms and contemporary s
Social Research 52:663-716.
Cole, Stephen. 1975. The growth of scientific knowledge: Theories of deviance as a case study. In The idea
of social structure: Papers in honor of Robert Merton, edited by L. A. Coser, 175-220. New York:
Harcourt Brace.
. 1992. Making science: Between nature and society. Cambridge, MA: Harvard University Press.
Coming, Peter A. 1971. The biological bases of behavior and some implications for political science. World
Politics 23:321-70.
- . 1975. Toward a survival-oriented policy science. Social Science Information 14 (5):59-86.

Crawford, Thomas J., and Murray Naditch. 1970. Relative deprivation, powerlessness, and militancy: The
psychology of social protest. Psychiatry 33:208-23.
Crosby, Faye. 1976. A model of egoistical relative deprivation. Psychological Review 83:85-113.
Crosby, Faye, Peter Muehrer, and George Loewenstein. 1986. Relative deprivation and explanation: Models
and concepts. In Relative deprivation and social comparison, edited by James M. Olson, C. Peter
Herman, and Mark P. Zanna, 17-32. Hillsdale, NJ: Lawrence Erlbaum.
Dahl, Robert A. 1961. The behavioral approach in political science: Epitaph for a monument to a successful
protest. American Political Science Review 55:763-72.
Davies, James Chowning. 1962. Toward a theory of revolution. American Sociological Review 6:5-19.
1963. Human nature in politics. New York: John Wiley.
1973. Aggression, violence, revolution, and war. In Handbook of political psychology, edited
by J. N. Knutson, 234-60. San Francisco: Jossey-Bass.
DeFronzo, James. 1991. Revolutions and revolutionary movements. Boulder, CO: Westview.
Degler, Carl N. 1991. In search of human nature. New York: Oxford University Press.
Drew, Paul. 1974. Domestic political violence: Some problems of measurement. Sociological Review
22:5-25.
Dub6, Lise, and Serge Guimond. 1986. Relative deprivation and social protest: The personal-group issue.
In Relative deprivation and social comparison, edited by James M. Olson, C. Peter Herman, and Mark
P. Zanna, 201-16. Hillsdale, NJ: Lawrence Erlbaum.
Feagin, Joseph R., and Harlan Hahn. 1973. Ghetto riots: The politics of violence in American cities. New
York: Macmillan.
Feierabend, Ivo K. 1971. Review of Why men rebel, by T. R. Gurr. American Political Science Review
65:194-96.
Feierabend, Ivo K., Rosalind L. Feierabend, and Ted Robert Gurr, eds. 1972. Anger, violence, and politics:
Theories and research. Englewood Cliffs, NJ: Prentice Hall.
Feierabend, Ivo K., Rosalind L. Feierabend, and Betty A. Nesvold. 1969. Social change and political
violence: Cross-national patterns. In Violence in America, edited by H. D. Graham and T. R. Gurr,
632-87. Washington, DC: Government Printing Office.
. 1973. The comparative study of revolution and violence. Comparative Politics 5:393-424.
Ferree, Myra Marx. 1980. Working class feminism: A consideration of the consequences of employment.
Sociological Quarterly 21:173-84.
. 1992. The political context of rationality: Rational choice theory and resource mobilization. In
Frontiers in social movement theory, edited by A. D. Morris and C. M. Mueller, 29-52. New Haven,
CT: Yale University Press.
Ferree, Myra Marx, and Frederick D. Miller. 1985. Mobilization and meaning: Toward an integration o
social psychological and resource perspectives on social movements. Sociological Inquiry 55:38-61.
Firestone, Joseph M. 1972. Theory of the riot process. American Behavioral Scientist 15:859-81.
1974. Continuities in the theory of violence. Journal of Conflict Resolution 18:117-42.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Flaming, Karl H. 1968. Who "riots" and why? Black and white perspectives in Milwaukee. Milwau
WI: Milwaukee Urban League.
Fogelson, Robert M. 1971. Violence as protest: A study of riots and ghettos. Garden City, NY: Doubl
Forster, Robert, and Jack P. Greene, eds. 1970. Preconditions of revolution in early modem Eur
Baltimore, MD: Johns Hopkins Press.
Freeman, Michael. 1972. Review article: Theories of revolution. British Journal of Political Scie
2:339-59.
Gamson, William A. 1975. The strategy of social protest. Homewood, IL: Dorsey.
Geschwender, Barbara N., and James A. Geschwender. 1973. Relative deprivation and participation in the
civil rights movement. Social Science Quarterly 54:403-11.
Geschwender, James. 1964. Social structure and the Negro revolt: An examination of some hypotheses.
Social Forces 43:248-56.
Governor's Commission on the Los Angeles Riots. 1965. Violence in the city: An end or a beginning? Los
Angeles: Jeffries Banknote.
Greene, Thomas H. 1984. Comparative revolutionary movements. 2d ed. Englewood Cliffs, NJ: Prentice
Hall.
Guimond, Serge, and Lise Dub6-Simard. 1983. Relative deprivation theory and the Quebec Nationalist
Movement: The cognition-emotion distinction and the personal-group deprivation issue. Journal o
Personality and Social Psychology 44:526-35.
Gumey, Joan Neff, and Kathleen J. Tierney. 1982. Relative deprivation and social movements: A critical
look at twenty years of theory and research. Sociological Quarterly 23:33-47.
Gurr, Ted Robert. 1968. Psychological factors in civil violence. World Politics 20:245-78.
1970. Why men rebel. Princeton, NJ: Princeton University Press.
Gurr, Ted Robert, and Mark Irving Lichbach. 1986. Forecasting internal conflict: A competitive evaluation
of empirical theories. Comparative Political Studies 19:3-38.
Halebsky, Sandor. 1974. New perspectives on political radicalism. Journal of Political and Military
Sociology 2:113-24.
Hibbs, Douglas A. 1973. Mass political violence: A cross-national causal analysis. New York: John Wiley
Hirschmann, Albert. 1973. The changing tolerance for income inequality in the course of economic
development. Quarterly Journal of Economics 87:544-66.
Horowitz, Irving Louis, ed. 1967. The rise and fall of Project Camelot. Cambridge: MIT Press.
Hull, David L., Peter D. Tessner, and Arthur M. Diamond. 1978. Planck's principle. Science 202:717-23.
Huntington, Samuel. 1968. Political order in changing societies. New Haven, CT: Yale University Press.
Hutchison, Terence. 1960. The significance and basic postulates of economic theory. 2d ed. New York
Kelley.
Jenkins, J. Craig, and Kurt Shock. 1992. Global structures and political processes in the study of domestic
political conflict. Annual Review of Sociology 18:161-85.
Keddie, Nikki R., ed. 1995. Debating revolutions. New York: New York University Press.
Kennedy, Charles R., Jr. 1984. Multinational corporations and political risk in the Persian Gulf. International
Journal of Middle East Studies 16:391-403.
Kent, Stephen A. 1982. Relative deprivation and resource mobilization: A study of early Quakerism. British
Journal of Sociology 33:529-44.
Kerbo, Harold R. 1978. Foreign involvement in the preconditions for political violence. The world system
and the case of Chile. Journal of Conflict Resolution 22:363-91.
Kiser, Edgar, and Michael Hechter. 1991. The role of general theory in comparative-historical sociology.
American Journal of Sociology 97:1-30.
Klandermans, Bert. 1984. Mobilization and participation: Social-psychological expansions of resource
mobilization theory. American Sociological Review 49:583-600.
Kramnick, Isaac. 1972. Reflections on revolution: Definition and explanation in recent scholarship. History
and Theory 11:26-63.
Ladner, Robert A., Barry J. Schwartz, Sandra J. Roker, and Loretta S. Titterud. 1981. The Miami riots of
1980: Antecedent conditions, community responses and participant characteristics. Research in Social
Movements, Conflicts and Change 4:171-214.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Laqueur, Walter. 1977. Interpretations of terrorism: Fact, fiction and political science. Journal of
porary History 12:1-42.
Laslett, Barbara. 1992. Gender in/and social science history. Social Science History 16:177-95.
Law, Kim S., and Edward J. Walsh. 1983. The interaction of grievances and structure in social m
analysis: The case of JUST [Johnstowners United to Stop another Tragedy]. Sociological Q
24:123-36.
Lichbach, Mark Irving. 1989. An evaluation of "Does economic inequality breed political conflict?" studies.
World Politics 41:431-70.
Lorenz, Konrad. 1966. On aggression. New York: Harcourt Brace.
MacIntyre, Alasdair. 1971. Is a science of comparative politics possible? In Against the self-images of the
age, 260-79. New York: Schocken.
- . 1973. Ideology, social science, and revolution. Comparative Politics 5:321-42.
MacIntyre, Stuart. 1980. Little Moscows: Communism and working-class militancy in inter-war Britain.
London: Croom Helm.
Martin, Joanne, Philip Brickman, and Alan Murray. 1984. Moral outrage and pragmatism: Explanations for
collective action. Journal of Experimental Social Psychology 20:484-96.
Marx, Gary T. 1972. Perspectives on violence. Contemporary Psychology 17:128-31.
Marx, Gary T., and James Wood. 1975. Strands of theory and research in collective behavior. Annual Review
of Sociology 1:363-428.
McCarthy, John D., and Mayer N. Zald. 1973. The trend ofsocial movements in America: Professionalization
and resource mobilization. Morristown, NJ: General Learning Press.
. 1976. Resource mobilization and social movements: A partial theory. American Journal of
Sociology 82:1212-41.
McPhail, Clark. 1971. Civil disorder participation: A critical examination of recent research. American
Sociological Review 36:1058-73.
Midlarsky, Manus. 1988. Rulers and the ruled: Pattemed inequality and the onset of mass political violence.
American Political Science Review 82:491-509.
Midlarsky, Manus I., and Kenneth Roberts. 1985. Class, state, and revolution in Central America: Nicaragua
and El Salvador compared. Journal of Conflict Resolution 29:163-93.
Miller, Abraham H., Louis H. Bolce, and Mark Halligan. 1977. The j-curve theory and the Black urban riots:
An empirical test of progressive relative deprivation theory. American Political Science Review
71:964-82.
Morris, Aldon D., and Cedric Herring. 1987. Theory and research in social movements: A critical review.
Annual Review of Political Science 2:137-98.
Morris, Aldon D., and Carol McClung Mueller, eds. 1992. Frontiers in social movement theory. New Haven,
CT: Yale University Press.
Morrison, Denton E., Kenneth E. Hornback (with Virginia Bemis, Mark Pittman, Tracy Harris, Linda Buttel,
Deborah Hughes, Joanne White, Carla Rossi, and Deborah Godfrey). 1976. Collective behavior: A
bibliography. New York: Garland.
Moshiri, Farrokh. 1991. Revolutionary conflict theory in an evolutionary perspective. In Revolutions in the
late twentieth century, edited by J. A. Goldstone, T. R. Gurr, and F. Moshiri, 4-36. Boulder, CO:
Westview.
Mueller, Carol McClurg. 1992. In Frontiers in social movement Theory, edited by A. D. Morris and C. M.
Mueller, 3-25. New Haven, CT: Yale University Press.
Muller, Edward N. 1972. A test of a partial theory of potential for political violence. American Political
Science Review 66:928-59.
. 1977. Mass politics: Focus on participation. American Behavioral Science 21:63-86.
.1979. Aggressive political participation. Princeton, NJ: Princeton University Press.
.1980. The psychology of political protest and violence. In Handbook of political conflict, edited
by T. R. Gurr, 69-99. New York: Free Press.
. 1985. Income inequality, regime repressiveness, and political violence. American Sociological
Review 50:47-61.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Muller, Edward N., and Thomas O. Jukam. 1983. Discontent and aggressive political participation. B
Journal of Political Science 13:159-79.
Muller, Edward N., and Mitchell A. Seligson. 1987. Inequality and insurgency. American Political Sc
Review 81:425-51.
Muller, Edward N., and Erich Weede. 1994. Theories of rebellion: Relative deprivation and power
contention. Rationality and Society 6:40-57.
Noton, M., C. R. Mitchell, and F. R. Janes. 1974. The systems analysis of conflict. Futures 6:114-32.
Oberschall, Anthony. 1973. Social conflict and social movements. Englewood Cliffs, NJ: Prentice Hall.
Olson, James M., and J. Douglas Hazlewood. 1986. Relative deprivation and social comparison: An
integrative perspective. In Relative deprivation and social comparison, edited by J. M. Olson, C. P.
Herman, and M. P. Zanna, 1-15. Hillsdale, NJ: Lawrence Erlbaum.
Olson, James M., C. Peter Herman, and Mark P. Zanna, eds. 1986. Relative deprivation and social
comparison. Hillsdale, NJ: Lawrence Erlbaum.
Olson, Mancur. 1965. The logic of collective action. Cambridge, MA: Harvard University Press.
Papcke, Sven. 1971. How-to-keep-them-down. Politische Vierteljahresschrift 12:125-29.
Parsons, Talcott. 1937. The structure of social action. New York: McGraw-Hill.
Pennock, J. Roland, and John W. Chapman, eds. 1977. Human nature in politics. New York: New York
University Press.
Pettigrew, Thomas F. 1971. Racially separate or together? New York: McGraw-Hill.
Piven, Frances Fox, and Richard A. Cloward. 1992. Normalizing collective protest. In Frontiers in social
movement theory, edited by A. D. Morris and C. M. Mueller, 301-25. New Haven, CT: Yale University
Press.
Planck, Max. 1950. Scientific autobiography. London: Williams & Norgate.
Platt, Anthony M., ed. 1971. The politics of riot commissions, 1917-1970: A collection of official reports
and essays. New York: Macmillan.
Popper, Karl. 1959. The logic of scientific discovery. London: Hutchinson.
.1974. Darwinism as a metaphysical research programme. In The philosophy of Karl Popper, edited
by P. A. Schilpp, 133-43. LaSalle, IL: Open Court.
. 1978. Natural selection and the emergence of mind. Dialectica 32:339-55.
Portes, Alejandro. 1970. Leftist radicalism in Chile: A test of three hypotheses. Comparative Politics
2:251-74.
.1971a. Urbanization and politics in Latin America. Social Science Quarterly 52:697-720.
1971b. On the logic of post-factum explanations: The hypothesis of lower-class frustration as the
cause of leftist radicalism. Social Forces 50:26-44.
Ramirez, Francisco 0. 1981. Comparative social movements. International Journal of Comparative
Sociology 22:3-21.
Rosenberg, Alexander. 1988. Philosophy of social science. Boulder, CO: Westview.
. 1992. Economics: Mathematical politics or science of diminishing returns? Chicago: University
of Chicago Press.
Rudner, Richard. 1966. The philosophy of social science. Englewood Cliffs, NJ: Prentice Hall.
Rule, James B. 1988. Theories of civil violence. Berkeley: University of California Press.
. 1989. Rationality and non-rationality in militant collective action. Sociological Theory 7:145-60.
Rule, James B., and Charles Tilly. 1972. 1830 and the unnatural history of revolution. Journal of Social
Issues 28:49-76.
Runciman, Walter G. 1966. Relative deprivation and social justice: A study of attitudes to social inequality
in twentieth-century England. London: Routledge and Kegan Paul.
. 1989. Confessions of a reluctant theorist: Selected essays. New York: Harvester Wheatsheaf.
Salert, Barbara. 1976. Revolutions and revolutionaries: Four theories. New York: Elsevier.
Sanders, David, and Eric Tannenbaum. 1983. Direct action and political culture: The changing political
consciousness of the British public. European Journal of Politics 11:45-61.
Schultz, Richard. 1979. Coercive force and military strategy: Deterrence logic and the cost-benefit model
of counterinsurgency warfare. Western Political Quarterly 32:444-66.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Schwartz, David C. 1974. Toward a more relevant and rigorous political science. Journal
36:103-37.
Shapin, Steven. 1993. Mertonian concessions. Science 259:839-41.
Silvert, Kalman H. 1971. Much conflict, little power. Comparative Politics 3:447-62.
Skocpol, Theda. 1976. France, Russia, China: A structural analysis of social revolutions. Comparative
Studies in Society and History 18:175-210.
Smelser, Neil J. 1962. Theory of collective behavior. New York: Free Press.
Snyder, David, and Charles Tilly. 1972. Hardship and collective violence in France, 1830 to 1960. American
Sociological Review 37:520-32.
Social Science Research Council, Committee on Historiography. 1954. The social sciences in historical
study. Bulletin No. 64. New York: Social Sciences Research Council.
Somit, Albert. 1968. Toward a more biologically-oriented political science: Ethology and psychopharma-
cology. Midwest Journal of Political Science 12:550-67.
Spilerman, Seymour. 1970. The causes of racial disturbances: A comparison of alternative explanations.
American Sociological Review 35:627-49.
- 1971. The causes of racial disturbances: Tests of an explanation. American Sociological Review
36:427-42.
Stone, William F. 1981. Political psychology: A Whig history. In Handbook of political behavior, edited by
S. L. Long, 1-67. New York: Plenum.
Terwilliger, Robert F. 1971. Review of Why men rebel, by T. R. Gurr. Social Research 38:169-72.
Thompson, E. P. 1971. The moral economy of the English crowd in the eighteenth century. Past and Present
50:76-136.
Tiemey, William G. 1988. Much ado about something: Naturalistic inquiry and the paradigm revolution.
Journal of Higher Education 59:225-30.
Tilly, Charles. 1969. Collective violence in European perspective. In The history of violence in America,
edited by H. D. Graham and T. R. Gurr, 4-45. New York: Bantam.
.1971. Review of Why men rebel, by T. R. Gurr. Journal of Social History 4:416-20.
. 1973. Does modernization breed revolution? Comparative Politics 5:426-47.
.1975. Revolutions and collective violence. In Handbook of political science, vol. 3, edited by F.
Greenstein and N. W. Polsby, 483-555. Reading, MA: Addison-Wesley.
- 1978. From mobilization to revolution. Reading, MA: Addison-Wesley.
- 1984. Big structures, large processes, huge comparisons. New York: Russell Sage.
Tilly, Charles, Louise Tilly, and Richard Tilly. 1975. The rebellious century, 1830-1930. Cambridge, MA:
Harvard University Press.
Useem, Michael. 1972. Ideological and interpersonal change in the radical protest movement. Social
Problems 19:451-69.
.1975. Protest movements in America. Indianapolis, IN: Bobbs-Merrill.
1976. Legitimacy and government control of the production of academic social knowledge. Journal
of Political and Military Sociology 4:217-32.
Utter, Glenn H., and James M. Vanderleeuw. 1991. Election predictions: Theory and social science. Midsouth
Political Science Journal 12:114-31.
Wanderer, Jules J. 1968. 1967 riots: A test of the congruity of events. Social Problems 16:193-98.
Wasbum, Philo. 1974. Political sociology in the United States: A case study of theoretical change and
continuity. Human Relations 27:321-37.
Weinberg, Steven. 1992. Dreams of afinal theory. New York: Pantheon.
Wheeler, Robert F. 1975. German women and the Communist International: The case of the Independent
Social Democrats. Central European History 8:113-39.
Wilhelm, Sidney M. 1970. Who needs the Negro? Cambridge, MA: Schenkman.
Williams, Robin M., Jr. 1975a. Relative deprivation. In The idea of social structure, edited by L. A. Coser,
355-78. New York: Harcourt Brace.
-----. 1 1975b. Race and ethnic relations. Annual Review of Sociology 1:125-64.
Wilson, John. 1973. Introduction to social movements. New York: Basic Books.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC
Young, M. Crawford. 1970. Rebellion and the Congo. In Protest and power in Black Africa, edited b
Rotberg and A. A. Mazrui, 969-1011. New York: Oxford University Press.
Zagoiin, Perez. 1973. Theories of revolution in contemporary historiography. Political Science Quar
88:23-52.
. 1982. Rebels and rulers 1500-1660. Vol. 1. New York: Cambridge University Press.
Zald, Mayer N. 1991. Sociology as a discipline: Quasi-science and quasi-humanities. American Sociologist
22:165-87.
- . 1992. Looking backward to look forward: Reflections on the past and future of the resource
mobilization research program. In Frontiers in social movement theory, edited by A. D. Morris and C. M.
Mueller, 326-48. New Haven, CT: Yale University Press.
Zimmermann, Ekkart. 1979. Crises and crises outcomes: Towards a new synthetic approach. European
Journal of Political Research 7:67-115.
. 1980. Macro-comparative research on political protest. In Handbook of political conflict, edited by
T. R. Gurr, 167-237. New York: Free Press.
. 1983. Political violence, crises, and revolutions. Cambridge, MA: Schenkman.
Zurcher, Louis A., and David A. Snow. 1981. Collective behavior: Social movements. In Socialpsychology:
Sociological perspectives, edited by M. Rosenberg and R. H. Turner, 447-82. New York: Basic Books.

197.95.143.137 on Fri, 17 Jul 2020 10:39:58 UTC

Brush 1996 Dynamics of Theory Change

Uploaded by

Copyright:

Available Formats

Brush 1996 Dynamics of Theory Change

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Brush 1996 Dynamics of Theory Change

Uploaded by

Copyright:

Available Formats

Dynamics of Theory Change in the Social Sciences: Relative Deprivation and Collective

Stable URL: http://www.jstor.com/stable/174456

This content downloaded from

RELATIVE DEPRIVATION AND COLLECTIVE VIOLENCE

JOURNAL OF CONFLICT RESOLUTION, Vol. 40 No. 4, December 1996 523-545

This content downloaded from

In this article, I assume that it is possible to pursue social science resear

This content downloaded from

despite the confirmation of several of his predictions, even after he was aw

This content downloaded from

This content downloaded from

movement and protests against American participation in the Vietnam war w

This content downloaded from

Gurr suggested that RD could be measured by surveys, such as the one c

This content downloaded from

This content downloaded from

This content downloaded from

usually fails to explain as much of the phenomenon that it is intended to predict

Remarkably, RD theory became a flourishing research topic among soci

This content downloaded from

of RM and other theorists have less interest in Popperian-style conject

In asking why the social science community rejected the RD theor

This content downloaded from

This content downloaded from

This content downloaded from

(Williams 1975b). Finally, Zimmerman shifted from a slightly favorable stanc

This content downloaded from

is superfluous because in any society there is always a prevalence of

This content downloaded from

Of those whose training or current positions could be identified, approximately

This content downloaded from

This content downloaded from

usual assumption behind Planck's principle-that younger scientists are more

This content downloaded from

- . 1975. Toward a survival-oriented policy science. Social Science Information 14 (5):59-86.

This content downloaded from

This content downloaded from

This content downloaded from

This content downloaded from

This content downloaded from

This content downloaded from

You might also like