A Decade of System Justification Theory:
Accumulated Evidence of Conscious and Unconscious
Bolstering of the Status Quo
John T. Jost
Department of Psychology, New York University
Mahzarin R. Banaji
Department of Psychology and Radcliffe Institute for Advanced Study,
Harvard University
Brian A. Nosek
Department of Psychology, University of Virginia
Most theories in social and political psychology stress self-interest, intergroup conflict, eth-
nocentrism, homophily, ingroup bias, outgroup antipathy, dominance, and resistance.
System justification theory is influenced by these perspectives—including social identity
and social dominance theories—but it departs from them in several respects. Advocates of
system justification theory argue that (a) there is a general ideological motive to justify the
existing social order, (b) this motive is at least partially responsible for the internalization
of inferiority among members of disadvantaged groups, (c) it is observed most readily at
an implicit, nonconscious level of awareness and (d) paradoxically, it is sometimes
strongest among those who are most harmed by the status quo. This article reviews and
integrates 10 years of research on 20 hypotheses derived from a system justification per-
spective, focusing on the phenomenon of implicit outgroup favoritism among members of
disadvantaged groups (including African Americans, the elderly, and gays/lesbians) and
its relation to political ideology (especially liberalism-conservatism).
KEY WORDS: ideology, system justification, intergroup relations, implicit bias
There is a cluster of related theories that are by now so prevalent in social
science that they strike the contemporary reader as self-evidently true. Although
these theories are by no means indistinguishable, they share a set of common fea-
tures, including the tenets that groups serve their own interests, develop ideolo-
Political Psychology, Vol. 25, No. 6, 2004
0162-895X © 2004 International Society of Political Psychology
Published by Blackwell Publishing. Inc., 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington Road, Oxford, OX4 2DQ
881
gies to justify those interests, have strong preferences for members of their own
kind, are hostile and prejudicial toward outsiders, and are conflict-seeking when-
ever it helps to advance their partisan interests and particularistic identities. For
the sake of classification—and in order to contrast them with our own approach—
we refer to these as “group justification” theories (see also Jost & Banaji, 1994).
They hold that people are driven by ethnocentric motives to build ingroup soli-
darity and to defend and justify the interests and identities of fellow ingroup
members against those of outgroup members. Such theories may contain one or
more of the following specific assumptions:
Similar others are preferred to dissimilar others. (Allen & Wilder, 1975;
Brewer, 1979; Tsui, Egan, & O’Reilly, 1992)
Prejudice is a form of hostility directed at outgroup members. (Adorno,
Frenkel-Brunswik, Levinson, & Sanford, 1950; Allport, 1954; Brown,
2000b; Pettigrew, 1982)
Intergroup relations in society are inherently competitive and conflict-
ridden. (Bobo, 1988; Sherif, 1967; Sidanius & Pratto, 1999)
Intergroup behavior is driven primarily by ethnocentrism and ingroup
favoritism. (Brewer & Campbell, 1976; Brewer & Miller, 1996; Sumner,
1906; Tajfel & Turner, 1986)
Prejudice, discrimination, and institutionalized oppression are inevitable
outcomes of intergroup relations. (Sidanius & Pratto, 1993)
Members of dominant groups strive to impose their hegemonic will on
members of subordinated groups. (Fiske, 1993; Sidanius & Pratto, 1999)
Members of subordinated groups first seek to escape the implications of
group membership by exercising individual exit and mobility options.
(Ellemers, Wilke, & van Knippenberg, 1993; Hirschman, 1970; Tajfel,
1975)
When individual exit/mobility is impossible, members of subordinated
groups engage in identity enhancement strategies of resistance and com-
petition. (Scott, 1990; Spears, Jetten, & Doosje, 2001; Tajfel & Turner,
1986)
In coping with chronically threatened social identities, members of sub-
ordinated groups typically express stronger levels of ingroup favoritism
than do members of dominant groups. (Leach, Spears, Branscombe, &
Doosje, 2003; Mullen, Brown, & Smith, 1992)
Political ideology mirrors/group membership individual and collective
self-interest and/or social position. (Centers, 1949; Downs, 1957; Olson,
1971; Sidanius, Singh, Hetts, & Federico, 2000)
882 Jost et al.
A sense of injustice is triggered by violations of relative standards or
established fairness norms. (Deutsch, 1985; Gurr, 1970; Taylor &
Moghaddam, 1994; Walker & Smith, 2002)
In the social scientific imagination, it is as if the advantaged are relentlessly
looking to cash in on their dominance and the disadvantaged are proud revolu-
tionaries-in-waiting. Both types of groups are seen as primarily self-interested,
and overt conflicts of interest are assumed to be endemic.
1
In this paper, we question these common, almost ubiquitous assumptions and
make a case for a contrary perspective. We challenge these conventionally
accepted principles not because we think that they are unhelpful or incorrect or
fail to capture the modal case, but because the many notable exceptions and devi-
ations are instructive, revealing, and helpful for creative theory-building (see
McGuire, 1997). The received view is a good story, but it is not the whole story.
We think that it needs to be supplemented with an alternative theoretical per-
spective that takes the important exceptions seriously. In this article, we further
advance a psychological theory of system justification, defined as the “process by
which existing social arrangements are legitimized, even at the expense of per-
sonal and group interest” (Jost & Banaji, 1994, p. 2). Specifically, we review 10
years of research stimulated by a system justification perspective on intergroup
relations, and we present some new data pertaining to the ideological basis of
conscious and nonconscious intergroup attitudes.
The Accumulation of Evidence Against the Received View
In recent years, evidence against the propositions listed above has been accu-
mulating, and a number of commentators have begun to express dissatisfaction
with pieces of the received view. Jackman (1994), for instance, railed against
“conflict theories” of intergroup relations and the conception of prejudice as “irra-
tional antagonism.” She suggested that, from a system maintenance perspective,
there is far more to be gained by members of dominant groups fostering cooper-
ative, even affectionate relationships with their subordinates. Her historical and
survey research shows that dominants and subordinates are highly averse to con-
flict and antagonism and generally develop collaborative relationships, even
within the context of dramatically inegalitarian institutions such as slavery. Glick
and Fiske (2001) similarly criticized Allport’s (1954) popular definition of preju-
dice as antipathy for failing to explain benevolent forms of sexism. They showed
that seemingly favorable attitudes toward women can help to sustain gender
A Decade of System Justification Theory 883
1
The assumption of universal self-interest, whether made by social scientists or lay people, may itself
contribute to system justification, insofar as it justifies self-interested behavior on the part of advan-
taged group members by suggesting that everyone—including members of disadvantaged groups—
equivalently embraces self-interest (which is not the case, as we will show).
inequality and discriminatory systems and should therefore be considered preju-
dicial, even though such attitudes are highly appealing to many women (e.g.,
Kilianski & Rudman, 1998). The weight of evidence is also mounting against the
notion that ingroup bias is a default feature of intergroup relations and that
members of low-status groups typically use a wide repertoire of identity enhance-
ment strategies. To take one example from the survey literature, Sniderman and
Piazza (1993) found in a large, nationally representative sample that African
American respondents generally accepted unfavorable stereotypes of their own
group as lazy, irresponsible, and violent. Indeed, they endorsed these stereotypes
even more strongly than European American respondents did. Experimental and
field studies have since shown that members of disadvantaged groups often hold
ambivalent, conflicted attitudes about their own group membership and surpris-
ingly favorable attitudes toward members of more advantaged groups (e.g., Jost
& Burgess, 2000; Jost, Pelham, & Carvallo, 2002). On the basis of these and
other findings, Smith and Mackie (2002) concluded that intergroup attitudes
are more complex and differentiated than the received view allows. Ingroup
favoritism and outgroup derogation may be relatively common, but they are by
no means the only reactions that people have to social groups, especially when
status and power differences are involved.
Miller (1999) argued persuasively that self-interest is a product of social and
cultural norms rather than a universal “fact” about human motivation. Empirical
studies conducted by Miller and Ratner (1998) demonstrate that group member-
ships have much weaker effects on social attitudes than observers assume. With
regard to political attitudes, there is notoriously little correspondence between
indicators of self-interest (such as income, social class, and demographic group
membership) and ideology (e.g., Jost, Glaser, Kruglanski, & Sulloway, 2003a;
Lane, 1959/2004; Lipset, 1981; Sears & Funk, 1991; Sidanius & Ekehammar,
1979; Stacey & Green, 1971; Wilson, 1973). Even on issues that should be highly
relevant to considerations of self-interest, such as policies of economic distribu-
tion, research repeatedly shows that low-income groups are scarcely more likely
than high-income groups to support such policies, although they would obviously
benefit from them (Fong, 2001; Gilens, 1999; Jost, Pelham, Sheldon, & Sullivan,
2003; Kluegel & Smith, 1986). In a similar vein, Newman (2002) concluded on
the basis of her urban ethnographic work that, in defiance of current sociological
theories, “ghetto dwellers are neither the passive victims of nor the heroic resisters
against capitalist or racist exploitation” (p. 1586). Evidence against the received
view has been accumulating, and much of it is more consistent with a system
justification perspective that stresses accommodation and rationalization of the
status quo than with identity-based or interest-based theories.
Like all contemporary researchers of intergroup relations, we have been influ-
enced immensely by theories of social identification (Tajfel & Turner, 1986) and
social dominance (Sidanius & Pratto, 1999). From our viewpoint, however, these
approaches are hampered by adhering so closely to conventional assumptions of
884 Jost et al.
self-interest, homophily, ingroup bias, outgroup antipathy, and intergroup conflict.
In the case of social identity theory, Tajfel (1975) absorbed much of this frame-
work from Hirschman’s (1970) rational choice analysis of exit versus loyalty.
Other aspects may have resulted from Tajfel and Turner’s (1986) overgeneraliza-
tion of results from the minimal group paradigm in an effort to explain very dif-
ferent contexts involving longstanding inequalities between groups. With regard
to social dominance theory, assumptions of self-interest may derive from a reading
of evolutionary theory in which, among other things, ethnocentrism among
humans is seen as determined by inclusive fitness as an extension of “genetic self-
ishness” (Sidanius & Pratto, 1999, p. 27).
To the limited extent that these theories address attitudes toward the overar-
ching social system (rather than intergroup attitudes), they tend to regard the social
order as something that is imposed by one group and resisted by the other.
2
This
is their strength—because there is considerable heuristic value in making such an
assumption—but it is also their weakness. The image of intergroup relations that
results is overly self-interested and insufficiently ideological; these two criticisms
are not contradictory, because ideology is motivated by many factors in addition
to self-interest (Jost et al., 2003a). Theories of social identity and social domi-
nance fail to account for the degree to which psychological responses to the social
and political status quo are characterized by active bolstering and system justifi-
cation, especially among members of disadvantaged groups. That is, hierarchy is
maintained not only through mechanisms of ingroup favoritism and outgroup
derogation exercised by members of dominant groups, but also by the complic-
ity of members of subordinated groups, many of whom perpetuate inequality
through mechanisms such as outgroup favoritism.
To illustrate the one-sided emphasis on homophily, ingroup favoritism, and
ethnocentrism (and the corresponding neglect of outgroup favoritism), we have
listed in Table 1 several books on social identity and intergroup relations, com-
paring the number of index entries for “ingroup bias” and “ingroup favo(u)ritism”
to entries for “outgroup bias” and “outgroup favo(u)ritism.” For 11 books pub-
lished between 1981 and 2000, there were 142 index entries for ingroup
favoritism, whereas there were 12 entries for outgroup favoritism, 8 of which
came from a single chapter by Hinkle and Brown (1990). This one-sidedness is
not accidental. Prevailing theories contain a much more developed set of explana-
tory concepts around the struggle to foster positive group distinctiveness and to
favor ingroup members than around the motive to justify the status quo and the
tendency to internalize status hierarchies. Framing theories around concepts of
A Decade of System Justification Theory 885
2
On this issue, Havel (1991) wrote perceptively that “only a very generalized view (and even that
only approximative) permits us to divide society into the rulers and the ruled....In the post-
totalitarian system [the line of conflict] runs de facto through each person, for everyone in his own
way is both a victim and a supporter of the system. What we understand by the system is not, there-
fore, a social order imposed by one group upon another, but rather something which permeates an
entire society and is a factor in shaping it” (p. 144).
