While RWA is associated with group prejudice in various ways, that was not its specific focus.  It is the conformity with authority that is its core focus.  The willingness to discriminate against or attack members of a stigmatized group is one dimension in which this is expressed.

We now turn to a second factor that is also strongly associated with group prejudice.  However, unlike RWA, it has group prejudice as its central focus.  That factor is known as Social Dominance Orientation (SDO).  Because it, too, is correlated with political conservatism, it provides further evidence that conservatism functions as a form of identity politics, and gives further evidence of how this works.  SDO is also part of a more comprehensive theory which focuses attention on how hierarchical structures and values endure, while rationales may change--or even directly contradict what people actually believe.  This last is a crucial point when addressing the persistent fictions that "real conservatives" support balanced budgets, states rights, "judicial restraint," etc., despite repeated evidence to the contrary.

SDO Introduced

SDO is considered to be "the degree to which individuals desire and support group-based hierarchy and the dominance of `inferior' groups by `superior' groups." It is an orientation that encompasses a whole range of group biases, based on age, gender, race, ethnicity, nationality or religion. It is one aspect of Social Dominance Theory (SDT), a theory that places individual SDO into a larger context which serves to explain the maintenance and perpetuation of hierarchically-structured societies.  This includes factors contributing to SDO, as well as results following from SDO.

SDT argues for a common basis for a wide range of group prejudice, while not denying the role of specific factors as well.  It also argues for an asymmetry between dominant and subordinate groups.  While prejudice can go both ways, prejudice favoring dominant groups over subordinate groups is the defining rule in how hierarchical societies are structured.  It has pervasive realworld consequences.

SDO is measured using the SDO scale, developed during the 1980s and 1990s by Jim Sidanius, Felicia Pratto and colleagues.  Their book Social Dominance: An Intergroup Theory of Social Hierarchy and Oppression is the source for the vast majority of information in this post--though some extrapolations are mine.  The SDO scale went though a number of refinements, but contains similar statements, words or phrases to be responded to, such as "winning is more important than how the game is played," "to get ahead in life it is sometimes necessary to step on other groups of people," and "social equality between groups."

SDO has been found to correlate with negative attitudes toward low-status groups such as: women, poor people, Blacks, Hispanics, Asians, Democrats, Hispanic civil rights groups and Black civil rights groups. It has been found to correlate with positive attitudes toward high-status groups such as: whites, Republicans, business executives, and politicians.

SDO has also been found to correlate with a number of legitimating ideologies or beliefs, including: racism/ethnic prejudice, sexism, nationalism, cultural elitism, patriotism, political conservatism, Protestant work ethic, and the rejection of noblesse oblige. It also correlated with attributing poverty to the internal failings of poor people, and rejecting the role of external conditions.

Social Dominance Theory: Basic Outline

SDT begins with the empirical observation that surplus-producing social systems have a three-fold group-based hierarchy structure: age-based, gender-based and "arbitrary set-based," which can include race, class, caste, ethnicity, religious affiliation, etc. Age-based hierarchies invariably give more power to adults and middle-age people than children and younger adults, and gender-based hierarchies invariable grant more power to men than women, but arbitrary-set hierarchies--though quite resilient--are truly arbitrary. There is no pre-set formula for what groups will become dominant.

SDT is based on three primary assumptions:

  1. While age- and gender-based hierarchies will tend to exist within all social systems, arbitrary-set systems of social hierarchy will invariably emerge within social systems producing sustainable economic surpluses.
  2. Most forms of group conflict and oppression (e.g., racism, ethnocentrisim, sexism, nationalism, classicism, regionalism) can be regarded as different manifestations of the same basic human predisposition to form group-based hierarchies.
  3. Human social systems are subject to the counterbalancing influences of hierarchy-enhancing (HE) forces, producing and maintaining ever higher levels of group-based social inequality, and hierarchy-attenuating (HA) forces, producing greater levels of group-based social equality.

Four factors contribute to SDO:  group status, gender, socialization and temperament.  In turn, SDO affects acceptance of influential "legitimating myths" (LMs)--whether hierarchy enhancing (HE) or hierarchy attenuating (HA).  LMs are defined as "values, attitudes, beliefs, causal attributions, and ideologies that provide moral and intellectual justification for social practices that either increase, maintain, or decrease levels of social inequality among social groups." The HE-LMs provide support for social polcies that in turn sustain group-based social hierarchies through three mechanisms:  individual discrimination (personal decision-making), institutional discrimination (rules and procedures with biased outcomes, intentional or not) and behavioral asymmetry (various ways that subordinate groups participate in their own subjugation).

SDO And Liberal/Conservative Ideology

SDT, and the correlation between SDO and conservatism, is strongly compatible with the fundamental relationship between liberalism, egalitarianism, and social progress extending over the past 500+ years--all of which conservatism has opposed.

The rise of egalitarian values has had a profound effect on the nature of modern societies. Because we live in a nominally egalitarian society, Sidanius and Pratto note, there is a need for "plausible deniability, or the ability to practice discrimination, while at the same time denying that any discrimination is actually taking place."   This need is not limited to specific practices, it is generalized to encompass entire political philosophies.

Indeed, the connection between liberalism and egalitarianism has been the focus of fierce conservative propaganda for the past 40+ years.   The rhetoric of a "liberal media" and "liberal elite" is part of this attack on the egalitarian thrust of liberalism.  But a large part of the juice comes from the dramatic power-shift of the 1960s, when legal segregation and the legal subjugation of women were both dealt crippling blows.  

The loss of racial and gender privilege was a traumatic event--especially, but not exclusively, for those most privileged: white males.  But it was also traumatic for others who strongly identified with the existing social order, even if it did not objectively favor them, or even treat them fairly.  There are at least two mechanisms--one psychodynamic, the other cognitive-developmental--that can help explain why women and blacks could also be traumatized, which we will discuss later in this series.

For now, we can simply say that the pre-existing system, however unequal it was, was taken to be fundamentally just by a wide range of people--on the assumption, key to SDO, that some folks are better than others, and they were simply getting what was just.  Hence, any change in it was taken to be an attack on the moral order--which, indeed, is how the right has always characterized the 1960s.  (Of course, they dare not say--sometimes even to themselves--what they really think: that it was the uppity blacks and uppity women who spoiled everything.  So they talk about sex, drugs, crime and rock `n roll.  This is an example of displacement.)

We have already seen evidence of this earlier in this series--in the numbers of people who said that blacks had too much power back in 1964, for example. And we saw that this belief was strongly correlated with the hard core of conservatism--operational conservatism.

Affirmative action was a particularly fruitful flash point for a conservative counter-attack, because it provides a means to ostensibly argue for equality, and to argue against a policy that is widely identified with liberalism.  Hence, if liberals support affirmative action, the argument goes, they are clearly opposed to equal treatment.

As it turns out, SDO is particularly helpful in piercing this mythology, which the book, Social Dominance, does with some thoroughness.

On the one hand, the authors give us a dose of reality, reviewing the international literature on affirmative action. A survey of 19 job audit studies in 5 countries found evidence of discrimination against subordinate groups (in the U.S., blacks and Latinos) in every single study, despite the presence of affirmative action in one form or another. These are general audit studies, which look at how people looking for jobs are treated.  A Dutch study, looking specifically at how affirmative action programs work, found that an affirmative action rule favoring minorities outright--a standard deemed unconstitutional in the U.S. since 1977--only succeeded in producing a true level playing field, not a tilt in favor of minorities.

Thus, all the scientific evidence shows that the folk belief is false.   Affirmative action really is an egalitarian social policy, not reverse discrimination.

On the other hand, the authors examine how SDO relates to arguments against affirmative action. Not surprisingly, Pratto and Sidanius found that opposition to affirmative action correlates with SDO.  Those opposed to affirmative action are more likely to favor dominance of some groups over others, despite their egalitarian rhetoric to the contrary.  (Remember, this is a statistical finding about group attitudes. It says nothing about any specific individual.)

However, Pratto and Sidanius go further, and examine the role of specific beliefs used to justify attitudes toward affirmative action.  They found that SDO correlates with seemingly egalitarian rationales.  First, the beliefs that "Affirmative action will increase racial conflict."  Second, the belief that "Affirmative action just increases the idea that certain groups are not as good as others."

Some who offer these rationales for opposing affirmative action could be sincere--those scoring low in SDO.  But statistically, these arguments come far more often from people who agree that "in getting what you want, it is sometimes necessary to use force against other groups."  In other words, these arguments are strategies of denial and propaganda far more often than they are sincere arguments expressing the values people actually hold.

This discussion about SDO and affirmative action is really just the tip of the iceberg, in ways that Social Dominance suggests, but does not follow up.  Sidanius and Pratto are justifiably focused on laying out how their theory elucidates what see in our present-day hierarchical society with egalitarian pretensions, with occasional attention to other societies as well. By itself, this is a tremendous accomplishment.  But they say very little about how their theory can elucidate social change--both how progress is achieved, and how comes to be contained. And here, their theory has enormous explanatory potential, as their discussion of affirmative action suggests.  SDT and SDO provide a perspective, as well as analytic tools that can be brought to bear to help dissect how new hierarchicy enhancing narratives and justifications are developed and advanced in an environment that ostensibly rejects the very values they serve.

The next post in this series will examine this process more closely, and show how it furthers the argument that conservatism is not, fundamentally, an ideology, but a form of identity politics.  It is composed of narratives repeatedly refashioned to rejustify hierarchy after old positions have proved untenable. While the outward appearance of these narratives is ideological, their inner substance, function and purpose is the preservation of hierarchy, and the method is the narrative creation of fictive identities whose very being is bound to the hierarchies in whose interests they are created.

The discussion about SDO and affirmative action is exemplary of a much wider phenomena in a contemporary sense as well. There is a vast abundance of information showing that women and minorities are subject to unequal treatment (discussed in some detail by Sidanius and Pratto) but this information is routinely marginalized in public discourse. The ways in which this is done are themselves examples of how hierarchy is preserved through the institutional control of information, which is a crucial component of social power, without which organized social change is extremely difficult. The suppression of such information then allows privileged groups to portray themselves as victims of unequal treatment, and champions of equality against their "oppressors"--including, of course, liberals as oppressors of freedom-loving conservatives.

Thus, it comes to pass that narratives of "unqualified minorities" advancing are everywhere, reinforcing conservative's sense that "everyone knows" this is the way it is, while empirical studies proving the exact opposite are virtually never mentioned, even by occasional liberal commentators.  Those who actually know the facts come to feel awkward, and ashamed to simply tell the truth.  This is the social dynamic of self-policing state of repression, in which, ironically, everyone feels repressed, liberal and conservative alike, although for quite different reasons.

Supporting this process is the background assumption of group superiority embodied in SDO, as well as related attitudinal structures, to be discussed in a later post.  But the influence of SDO is not limited to dominant groups.

SDO And Consensual Legitimating Myths

A key facet of SDT is the role played by consensual beliefs shared by dominant and subordinate groups.  Indeed, Sidanius and Pratto see such beliefs as crucial in the maintenance of hierarchy.  When subordinate groups cease to believe the myths that legitimate dominant power, dominant groups are increasingly pressed to either resort to force, or grant concessions.  This happens in part because rejection and resistance by subordinate groups is echoed by dissention within dominant groups, refusing to support the use of violence.  The authors cite as examples the Algerian war of independence, and the American civil rights movement.  They point out that black acceptance of segregation was key to its survival.  While the majority of blacks never accepted a large body of white racist ideology, they did accept assumptions about social power, and did internalize a good deal of negative self-images that prevented the emergence of a mass movement for decades, until those beliefs were successfully challenged within the black community.

Although they do not mention it, a similar process took place in the early 1800s, when the relatively small free black community developed an opposition to the dominant white ideology of the time--colonization.  Colonization was supposed to "solve" America's race problem by shipping all blacks back to Africa.  It was never a serious solution, but it was politically convenient because it provided common ground for northern and southern whites in a theoretical fantasy.  The development of organized opposition within the black community, expressed in the counter-ideology of abolitionism, was the precursor to the development of white abolitionist activism.

Thus, subordinate acceptance or rejection of HE-LMs is a crucial matter.  To a very large extent, hierarchies are preserved through the subordinates' acceptance of ideological myths directly damaging to them.  Sidanius and Pratto refer to the underlying phenomena as "consensual SDO" and the related HE-LMs as "consensual" LMs, such as "consensual racism."

For example, they note the results of a 1997 Gallup Poll, which asked whether blacks in their communities had the same chance as whites to get any kind of job, to get equal education or equal housing.  The numbers believing in equal chances were:

Equal chance re jobs:
     White: 79%
     Black: 46%

Equal chance re education:
      White: 79%
     Black: 63%

Equal chance re housing:
     White: 86%
     Black: 58%

While Gallup focused on the difference in beliefs, Sidanius and Pratto call attention to (1) the large majorities of whites who mistakenly believe equal opportunity exists and (2) the large numbers of blacks who agree, despite their own direct experience, or knowledge of friends' and family's experience.  They then go on to develop the more general argument that consensual SDO--shared by dominant and subdominant groups alike--plays a dominant role in legitimating hierarchical social policies.

They report the results of a large Texas study in which attitudes towards classical racism and sexism were studied.  People were asked "Which of the following objects, statements, or events do you have a positive or negative feeling toward?"  The classical racism scale items such as "racial equality," "a Back president of the USA," "white superiority," "Mexican immigrants," etc.  The classical sexism scale included items such as "a real man should avoid showing emotion," "most feminists are a little too aggressive," "traditional sex roles," "women supervisors," "male nurses," etc.  While whites scored higher than minorities and men scored higher than women, these differences were relatively small compared to the level of agreement: 13% of the variance in racism scores was due to race, while 22% of the variance in sexism scores was due to gender.  A similar study, with similar results was done using a Soviet sample as well.

[Note: This is consistent with some data presented (but not analyzed) earlier in this series--the degree to which conservatives and liberals agree on social spending. It is not surprising from a social science perspective, but it does go strongly against common beliefs, which assume a far more unified group perspective than actually exists on virtually any subject.]

Sidanius and Pratto go on to provide data from their studies, showing correlations between consensual SDO and LMs on the one hand, and opposition to redistributive social policies on the other.  Political conservatism was one of the LMs in their study.  This is important to note, since it reminds us once again that RWA and SDO are present across all political lines.  They are simply present more intensely among conservatives.

More significant is the different functions of these factors among different groups.  One example is what's called outgroup favoritism--the tendency of subordinate groups to prefer the dominant group over their own.  This tendency runs counter to the tendency toward ingroup favoritism found in classic ethnocentrism, for example.  In contrast, amongst dominant groups, the two tendencies reinforce.  Outgroup favoritism among blacks has declined significantly since the pre-Civil Rights era, and has even vanished by some measures.  But a basic comparative relationship remains unchanged, as Sidanius and Pratto note, reporting the results of one large study: "[I]ronically, the more that dominants (Whites) believed U.S. society to be free of racial discrimination, the more they actually showed ingroup bias in favor of their racial ingroup. The opposite pattern was found among the two groups of subordinates. [Latinos and blacks]"

This finding is worth reflecting on a bit.  As Sidanius and Pratto show in some detail, there is overwhelming evidence that the U.S. is not free of racial discrimination.  So the tendency to believe it is free of discrimination is a tendency toward denial, for both dominants and subordinates.  Yet, if it were the case, then the tendency among subordinates would be straightforwardly rational--the absence of discrimination would remove the need for stronger group ties, and make positive attitudes toward others a net good, since such feelings would generally be reciprocated from a larger and wider range of people.

OTOH, the tendency among dominants doesn't make the same sort sense in terms of the non-discriminatory fantasy--unless one adds the sorts of beliefs that surround affirmative action.  That is, increased ingroup favoritism makes perfect sense if one feels that there is no discrimination, but minorities are getting ahead via false claims that it persists.  This is, of course, precisely the sort of narcissistic self-pity that is cultivated to keep the white backlash alive decades after the Civil Rights Movement.

The above offers yet another perspective on why conservatives are more cohesive, and have stronger feelings of group superiority, as well as persecution.  Furthermore, the fact that whites have higher SDO, but that cross-racial consensual SDO is stronger than SDO differences between races, helps illuminate how SDO contributes to political conservatism as a form of identity politics centered around whiteness, but not limited to whites.  The same goes for maleness, Christianity, and heterosexuality.


At one point, Altemeyer examined SDO and RWA together. His paper summarizing this comparative investigation provides a concise comparison of the two factors in their starkest terms.  Altemeyer wrote:

    "Right-wing authoritarians...seem to be highly prejudiced mainly because they were raised to travel in tight, ethnocentric circles, and they fear that authority and conventions are crumbling so quickly that civilization will collapse and they will be eaten in the resulting jungle.  In contrast, High SDOs already see life as `dog eat dog' and--compared with most people--are determined to do the eating."
Altemeyer was describing the core logic behind the two factors, expressed in terms of the extremes.  While relatively few people actually think this way either purely or consciously, the core logic influences us profoundly in many ways, even if we would forcefully reject such a bald statement of what it ultimately stands for.

The two factors correlate moderately with each other, as one might expect given that RWA includes authoritarian aggression, defined as "A general aggressiveness, directed against various persons, that is perceived to be sanctioned by established authorities."  Their interactions appear to be quite complex, as preliminary studies using different sorts of sample populations show different patterns of causal influences.  But one thing is strikingly clear: between the two of them, they account for the biggest chunk of group prejudice.  While other factors may play some role, no factor that is unrelated to them is anywhere close to them in level of importance.

This is also why I have chosen to highlight them in this series.  To the extent that conservatism functions as a group identity, defined in opposition to inferior others, the most logical and economical explanation for it would rely on the same mechanisms that are at work underlying group prejudice in general.  This assumption receives additional support from the fact that both RWA and SDO are correlated with political conservatism.

However, they are not the only psychological factors that have been found to be associated with conservatism.  After the next post--which, as promised, expands on the example of affirmative action discussed above--the following post in this series will take a broader look at the most comprehensive effort so far to integrate all the findings out there into a single model.  When it was published, it caused quite an uproar, which was, in its own way, as illuminating as the study itself.


The underlying material in this diary comes from the book, Social Dominance: An Intergroup Theory of Social Hierarchy and Oppression by Jim Sidanius and Felicia Pratto.

Previous posts in this series:

Originally posted to Paul Rosenberg on Sun Mar 05, 2006 at 11:25 AM PST.

Also republished by Psychology of Conservatives and Liberals.

Your Email has been sent.