jost_j_t_kay_a_c_thorisdottir_n_social_and_psychological_bas

protection and showing some restraint toward subordinates. (Note that human hierarchies depend much more strongly on “freely conferred deference” [Henrich and Gil-White, 2001] than on the threat of force, which plays such a large role in chimpanzee hierarchies. See Boehm, 1999,7 and Fiske, 1991, on how human authority ranking is a two-way street with mutual obligations and limitations on power.)

5.Purity/sanctity: The challenge of avoiding deadly microbes and parasites, which are easily spread among people living together in close proximity and sharing food, made it adaptive to attend to the contact history of the people and potential foods in one’s immediate environment, sometimes shunning or avoiding them. This foundation is different from the others in that its origins are in our physical nature— as omnivores—rather than in our social nature (see Rozin, Haidt, & McCauley, 2000). However, once human beings developed the emotion of disgust and its cognitive component of contagion sensitivity, they began to apply the emotion to other people and groups for social and symbolic reasons that sometimes had a close connection to health concerns (e.g., lepers, or people who had just touched a human corpse), but very often did not (e.g., people of low status, hypocrites, racists). When moral systems are built upon this foundation, they often go far beyond avoiding “unclean” people and animals; they promote a positive goal of living in a pure, sanctified way, which often involves rising above petty and carnal desires in order to prepare one’s mind and body for contact with God (see Haidt, 2006, Ch. 9).

The developmental component of moral foundations theory posits that the foundations make it easy for children to learn some virtues and hard to learn others. When we say that the foundations are innate, we do not mean that they are visible in infancy or unchanged by experience. Innate means, as Marcus (2004, p. 40) puts it, “organized in advance of experience.” The genes create the first drafts of our brains, but experience in our families and cultures then edits those drafts to produce unique individuals and divergent cultures. Haidt and Joseph (2007) considered five ways that moral knowledge might be organized in advance of experience and concluded that, whereas all five

7 Boehm, 1999, suggests that human beings are naturally hierarchical creatures who have developed mechanisms (through cultural and biological co-evolution) that enable them to band together to suppress bullying and live in egalitarian societies. If this is true, then egalitarian- ism/anti-oppression would be a good candidate for a sixth moral foundation. So far, however, we have assumed that egalitarianism is based primarily on the fairness foundation.

Moral foundations theory says that people in all cultures are born with the capacity to cultivate virtues based on all five foundations. Furthermore, people in all cultures do cultivate virtues based on the first two foundations: harm/care and fairness/reciprocity (see Brown, 1991, and Hauser, 2006, on moral universals). But as a society becomes more modern and more individualistic, the first two foundations become ever more important in daily life and in moral and political philosophy, while the last three become less important. (And after the horrors of 20th-century fascism, the concepts of ingroup, authority, and purity became particularly and deservedly suspect.) Because modernity increases the value of the first two foundations while decreasing or even reversing the value of the last three foundations, we refer to harm and fairness as the two “modern” or “individualizing” foundations, and we refer to ingroup, authority, and purity as the three “traditional” or “binding” foundations. (For a similar distinction, see Hunter, 1991, on the culture war between those who follow the “progressive” impulse and those who follow the “orthodox” impulse with regard to moral truth.)

We stress, however, that the conflict between modern and traditional moralities is really between a morality based mostly (although not entirely) on two foundations and a morality based firmly on all five foundations. The same division, we find, holds true for the conflict between socially liberal and socially conservative moralities, although we note that political identities are often complex and multifaceted. For example, many groups on the political left believe strongly in the moral importance of community (e.g., communitarians, socialists, and the anti–free-trade left) and may even have a moral identity as a righteous ingroup battling an evil outgroup. Many greens and environmentalists seem to ground part of their morality on notions of purity applied to the Earth and its ecosystems. And most libertarians and free-trade conservatives would reject purity concerns outright. So, no division, including our two versus five division, can neatly separate all people who self-identify as liberals from those who self-identify as conservatives. But we believe (and have found) that the two versus five formulation works well as a first pass.

ASYMMETRIC EXAGGERATIONS

The words “liberal” and “conservative” each refer to families of political and moral ideologies. Varieties of liberalism generally share the view that liberty and equality are fundamental political goods, and so liberals typically support individual rights and the use of government programs or changes in social institutions to extend such rights as widely (and as equally) as possible. The fundamental goods of conservatism, in contrast, are harder to define

because conservatism is generally said to arise as a reaction to the changes promoted by liberals, and those changes vary widely depending on the society being changed (Muller, 1997). Nonetheless, conservatives are typically united in their desire to conserve the status quo (i.e., in their opposition to change) and by the belief that long-existing institutions, norms, and traditions embody the wisdom of many generations and should not be tampered with lightly (Jost, Glaser, Kruglanski, & Sulloway, 2003a, 2003b). Combined with a view of human nature that is usually darker than that of liberals and a belief in the limits of human knowledge, conservatives tend to believe (as Durkheim did) that strong institutions and social constraints are necessary for children’s socialization, valuable for human welfare, and hard to replace once called into question or delegitimated (Burke, 2003/1790; Muller, 1997).

There are many kinds of conservatives and liberals in the United States today, but Jost (2006) argues that as a first cut, a simple one-dimensional spectrum of left–right or liberal–conservative does a surprisingly good job of arranging political attitudes and orientations. People in the United States freely talk about how liberal or conservative they are, usually without adding qualifications (e.g., distinguishing social from fiscal domains). What do they mean when they make such ratings? What moral values and virtues are associated with these terms, or with the “culture war” that is sometimes said to be raging between the two sides?

To find out, we created the Moral Foundations Questionnaire (MFQ), which asks participants to rate how relevant each of 20 concerns (four for each of our five foundations) is to their moral judgments and decisions. Item examples include whether or not another person: cared for someone weak or vulnerable (harm), ended up profiting more than others (fairness), put the interests of the group above his/her own (ingroup), showed a lack of respect for legitimate authority (authority), and was able to control his or her desires (purity). Data from six samples totaling more than 5,000 respondents have supported our basic political hypothesis with these relevance ratings: participants who had earlier identified themselves as liberal (below 4 on a 1–7 scale) rated items related to the two modern foundations as being more relevant to their moral judgments than items related to the three traditional foundations, whereas participants who had identified themselves as conservative (above 4) rated all five foundations as being equally relevant to their moral judgments (Graham, Haidt, & Nosek, 2008; Haidt & Graham, 2007).

After our first two studies, we added a second section to the MFQ asking participants to agree or disagree with specific moral statements. Four different types of moral statements were used: normative ideals (e.g., “It can never be right to kill a human being” for harm), hypothetical scenarios (e.g., “If I were a soldier and disagreed with my commanding officer’s orders,

I would obey anyway because that is my duty” for authority), positive virtues (e.g., “Chastity is an important and valuable virtue” for purity), and, because government policies are often the focus of moral debates, statements about laws or policies (e.g., “When the government makes laws, the number one principle should be ensuring that everyone is treated fairly” for fairness). Data from four samples totaling more than 3,000 respondents have shown the same pattern as did the relevance ratings. Figure 15.1 shows this basic pattern, for all 5,392 online participants who took either version of the MFQ. In all samples we have looked at, we have found that scores on the harm and fairness subscales slope downward as participants get more conservative, whereas scores on the ingroup, authority, and purity subscales slope upward.

Figure 15.1 Moral foundations (relevance and statements averaged together) across the political spectrum. N = 5,392. The two modern foundations are indicated in solid lines, which are higher for liberals, and the three traditional foundations are indicated with dashed lines, which are higher for conservatives.

To quantify accuracy, we took the difference between each participant’s ratings made as a typical liberal and those made as a typical conservative (their moral stereotype difference) and compared this to the actual mean differences between liberals and conservatives who were answering as themselves in this study. This difference score gave us a “moral stereotype inaccuracy” score for each participant. Figure 15.2 graphs these scores by ideology and shows how much participants exaggerated liberal–conservative moral differences. The results suggest three conclusions:

1.Participants at all points on the spectrum hold moral stereotypes that are accurate in direction. On average, participants correctly guessed

Figure 15.2 Moral stereotype inaccuracy across the political spectrum. Higher values indicate greater exaggeration of liberal–conservative differences, measured by comparing predicted/stereotyped differences against actual differences in self-reports. The units on the Y axis are the same 0–5 scale shown in Figure 15.1.

that liberals would give the highest ratings to harm and fairness items, whereas conservatives would give higher ratings than liberals on ingroup, authority, and purity items.

2.There was some support for the ideological polarization hypothesis, but only on the three traditional foundations (see Figure 15.2): those on the extreme left and right exaggerated ingroup, authority, and purity differences more than did moderates. This exaggeration was largely symmetrical across the political spectrum.

3.There was also support for the moral foundations prediction that liberals would be least accurate. As Figure 15.2 shows, liberals exaggerated the most overall, and they were particularly inaccurate about items related to the harm and fairness foundations. This asymmetrical exaggeration consisted almost exclusively of liberals guessing that conservatives would be less concerned about matters of harm and fairness than they actually were.

Do these data suggest a “rigidity of the left” hypothesis? No, but they do suggest the possibility of a “moral color-blindness” of the left, particularly the extreme left. When thinking about why conservatives generally oppose gay marriage, immigration, and stem cell research, for example, liberals simply cannot see any moral reasons. They are therefore free to assume the worst—that conservatives are really motivated by homophobia, racism, xenophobia, and ignorant fear of new technologies. If conservatives are motivated by such immoral forces, it must therefore be the case that conservatives don’t care very much about moral concerns such as harm to innocent victims, the rights of oppressed people, or justice more generally. If one believes this, and one is asked to complete the MFQ as a “typical conservative” would, then one would give inaccurately low ratings to items related to the two modern foundations.

But, before we suggest a color-blindness hypothesis, we note two limitations of our study. First, because our Internet-based sample was skewed to the liberal side, it is possible that those who said “strongly liberal” included more extremists and activists than was the case for our participants who rated themselves as “strongly conservative.” The most dogmatic and ideologically rigid conservatives might turn out to be just as morally color-blind, if we could find them and test them. Second, our study asked only about explicitly held beliefs, and it presented moral concerns cleanly separated as single items. In real life, moral concerns conflict and compete, and people make trade-offs (Tetlock, 1986). People then justify their choices and actions using convenient lofty principles, but those principles are often disconnected from the motives that really drove their decisions (Haidt, 2001, 2007). Therefore,

although liberals were fairly inaccurate about conservatives’ explicit values, it is not yet known whether liberals are also inaccurate about conservative behavior.

Nevertheless, these findings suggest that liberals don’t appreciate the extent to which issues such as gay marriage are morally conflicting for conservatives; it’s not that conservatives don’t care about fairness and equal rights, it’s that they also care about ingroup loyalty, traditional family structures, and spiritual purity. Recognizing that conservatives have a variety of moral concerns that liberals do not share and often do not recognize as legitimate moral concerns can help liberals better understand conservatives and respond to their arguments. It can also help us to improve SJT.

SYSTEM JUSTIFICATION IS PART OF (A FIVE-FOUNDATION) MORALITY

System justification theory examines the widespread motivation to rationalize the status quo and justify the existing social system, even among those who seem to be disadvantaged by the system ( Jost & Banaji, 1994). Although directly normative claims are seldom made in the literature, the implicit message is that system justification is a lamentable and ethically troubling motivation. In a two-foundation (harm and fairness) morality, system justification is problematic because it perpetuates existing inequalities and implicates people of low status as complicit in their own victimization. System justification research has largely concentrated on empirical demonstrations of the detrimental effects this motivation can have for low-status group members (e.g., lowered self-esteem, internalized inferiority; see Jost, Banaji, & Nosek, 2004, for a review). It has been proposed that despite these negative effects, system justification occurs in part because it makes people feel better by reducing their anxiety, uncertainty, and (for high-status people) guilt ( Jost & Hunyady, 2002).

We agree with the central claims of SJT that a system-justifying motive is widespread and that it cannot be explained as a by-product of ego justification or group justification. We also agree with its more specific claim that this motivational tendency is most often manifest nonconsciously and may be best observed using implicit measures ( Jost, Banaji, & Nosek, 2004). In stressing the importance of automatic and intuitive processes, SJT fits neatly with the social-intuitionist model of moral judgment (Haidt, 2001; Haidt & Bjorklund, 2008). However, although we agree with most of SJT, we see the justification of existing systems in a different light: it is a normal part of human morality, no more in need of psychological explanation than are concerns about harm and fairness.

In any culture that builds on the authority foundation, traditions and longstanding institutions are vested with moral importance and seen as embodying the collective wisdom of many generations. Changes should not be made readily or radically (see Burke, 2003/1790, on the excesses of the French Revolution; and see Sunstein, 2006, on the wisdom of Burkean Minimalism for modern law). In any culture that builds on the ingroup foundation, loyalty to the group is the supreme virtue, and criticizing it can, particularly in times of external conflict, be seen as a kind of betrayal. In any culture that builds on the purity foundation (along with authority or ingroup), the group, its leaders, its traditions, and its land may become sacred. Revisionists and reformers who want to implement changes based on the latest scientific findings or political trends treat everything as profane and open to tinkering. Their goal is usually morally worthy from a two-foundation perspective (e.g., decreasing suffering and increasing fairness), but it is often seen as sacrilegious and harmful (to individuals and to the nation) from a five-foundation perspective.

We think that a five-foundations perspective on system justification can benefit SJT in three ways: (a) it can help explain how rationalizing the status quo can reduce anxiety, even for those oppressed by it; (b) it can provide insights into potential evolutionary origins of system justification; and (c) it can provide a more complete explanation of why conservatives (and liberals) favor the policy positions that they do.

Benefi t 1: Morality as a Mechanism of Meaning

To explain why people justify the systems in which they participate, Jost, Fitzsimons, and Kay (2004) point to a long list of cognitive, social, motivational, and structural factors including uncertainty reduction, fear of equality, belief in a just world, needs for control and cognitive consistency, dissonance, reduction, political socialization, and mass media influences. Many of these factors have in common that they make people feel better about the status quo in the face of external threats and internal inconsistencies. Similarly, Jost and Hunyady (2002) point out the palliative (pain-reducing) functions of system justification as a coping strategy. But why exactly is it that justifying existing arrangements has so many psychological benefits, and why exactly are these benefits able to outweigh the many negative opposing effects that system justification has for members of low-status groups (e.g., dissonance, low self-esteem, depression, neuroticism, etc.)?

A moral foundations perspective suggests that the benefits of justifying the system are not just palliative, they are meaning-providing and can often be important for human flourishing. Happiness, health, and longevity are all correlated with income in Western nations, but this first-order correlation