This is Part 4 of a series on prestige competition and its broader contemporary effects. For other parts, click below.
The Prestige Complex
Hunter-gatherer studies suggest we are well-suited to defending our equal standing within small bands in which little to no resource accumulation occurs. In larger and more complex social arrangements, however, it seems the dispositions and psychological “mechanisms” (I use the term loosely and without any intent to imply the rather precise definition employed by some philosophers of science) that underlie this remarkable capacity to enforce egalitarianism are prone to misfiring in ways that allow, and perhaps even positively contribute to, the reemergence of hierarchy. To make adequate sense of this, we’ll need to get a better handle on how these mechanisms operate, both individually and in conjunction.
There are two psychological stories we need to tell here: the story of those who attain superior social status and the story of those who enable or allow others to do so. Boehm’s dual-story, as we’ve seen, is that we have an innate drive toward dominance and that, at some level of social and economic complexity, others will simply be unable to keep that drive in check. The psychological evidence, however, paints a rather more complicated picture.
No account of the psychological underpinnings of egalitarianism would be even half-complete without the phenomenon broadly called “inequity aversion.” This comes in a few distinct levels of sophistication in the animal world: from a relatively simple tendency to withdraw cooperation when cheated in a dyadic interaction, to an insistence that the reward/work ratio of others with whom one is not directly interacting be no greater than one’s own reward/work ratio, to, in humans, a capacity to experience guilt or shame when one’s reward/work ratio is higher than others. This last, most recent version—an intolerance of even personally advantageous inequality—was crucial to the stability and success of our ancestral hunter-gatherer groups.
Now, the successful deployment of inequity aversion in a prestige context requires an ability to make accurate social comparisons. We do not assess our prestige in vacuo but relative to others within a relevant comparison class. And it is by and large our peers—those to whom we are most similar or with whom we have the most regular contact—who constitute this comparison class. This tendency to treat comparisons with peers as more salient than broader social comparisons is known to psychologists as the “local dominance effect” (LDE)—so called because local social information tends to dominate more distant or general social information when it comes to a number of socially relevant self-assessments. The LDE is thought to be largely responsible for a number of other well-studied phenomena, such as the tendency of students to have higher academic self-concepts when attending low-ranking vs. high-ranking schools, a finding often described as the “big fish little pond effect.”
Much of the group-level interplay between egalitarianism and hierarchy, I think, is due to the joint operations of inequity aversion and one or more other features of our shared human psychology. The LDE is one such feature, and I’ve previously mentioned habituation (hedonic adaptation) as another. One further feature—as far as I know unnamed in the literature—was largely tacit in the brief account I sketched in The Architecture of Rage, but I’ve since come to suspect that it’s actually incredibly important and deserves much more explicit attention.
Social comparisons can be either upward (in which we compare ourselves to individuals with higher status) or downward, and a fair amount of research suggests that upward social comparisons tend to be regarded by the comparison-maker as far more affectively and motivationally salient. In a study of social comparison effects in a social media context, for example, Vogel, et al. (2014) found that upward comparisons caused marked reductions in state self-esteem, while downward comparisons caused no significant change in either direction. Similarly, Boyce, Brown, and Moore (2010), in a study on the effects of income rank on life satisfaction, found that the number of people within a comparison group earning more than a focal individual had on average 1.75 times more influence on that individual’s subjective income rank (how much they felt they were earning, relative to others) than the number of people earning less.
Let us call this simply the upward comparison dominance effect (UCDE). It might be an artifact of a more general “negativity bias,” but it would have likely served us well in our nomadic foraging past when there would have been very little resource accumulation and people would have been under continuous pressure to prove their value to the groups from which they benefited. In such a context, when a neighbor receives consistently more of a valued resource, it likely means not just that you aren’t getting as much but that you aren’t getting enough. If this resource is game meat, the risk of not getting enough is, of course, starvation or malnutrition; if the resource is prestige, the risk is social exclusion and perhaps even expulsion. It would have made a lot of sense, then, to pay particularly close attention to those better off or better regarded than oneself. Little wonder why extant hunter-gatherers are so adamant about—and successful at—enforcing equality, even with respect to nonmaterial resources like prestige.
One likely consequence of this UCDE is that even individuals with relatively high social status will pay a disproportionate amount of attention to those with still higher status than themselves. Indeed, there is evidence that higher social status actually upregulates sensitivity to disadvantageous inequality (Hu, et al., 2014; Hu, et al., 2016). This is apt to result in intensified social striving among comparatively high-status individuals for what ought, on a purely economic-utilitarian analysis, to be diminishing returns.
I recently stumbled across a surprisingly eloquent statement of this sentiment in, of all places, the comments section of a YouTube video:
I’m an international student who went to an ivy league school (aka I’m from an upper-middle-class background), and attend a top 30-in-the-world ranked university for postgraduate studies. I feel bad for inequality but I’ve never been a social liberal– I don’t do drugs, I care for myself, I am conservative in my actions, thoughts and behaviors. Nonetheless I feel the same ‘ressentiment’ that is being described in this talk [by Age of Anger author Pankaj Mishra]. The fact of the matter is that every step up the hierarchy of society you go up, you see that there is another that exists above you. Unless you are a billionaire, literally, this hierarchy will be felt as a real force of unfairness in your life, including me and other upper-middle-class educated people.
This is, of course, no serious evidence of the UCDE (see the previously cited studies for that), but it is an interesting example of how the insecurities generated by upward comparisons, in conjunction with inequity aversion, can affect even those of comparatively high social status.
The UCDE is one important component of our explanation for the reemergence of hierarchy. In truth, there are likely two motivational systems here operating in parallel: (1) a primitive “hedonic treadmill” in which resources like prestige, material wealth, and political influence are continually pursued because any gains made in terms of them are quickly habituated to (and any losses, however trivial, are felt nearly as painfully as they would be by a low-status individual); and (2) this newer system of inequity aversion and selective upward social comparison, which can, paradoxically, result in more desperate resource competition among those who already have, by any objective measure, an extremely safe amount.
Common Leftist sociopolitical narratives—that only the have-nots are motivated by fairness concerns while the haves are motivated only by greed (i.e., the hedonic treadmill system)—neglect the role of this second motivational system. This is, I think, to their great detriment. We need to take more seriously the plausible hypothesis, disturbing as it is, that our scrambles up the social ladder may be reactive and fear-driven as much as proactive and appetite-driven. This knowledge, counterintuitive as it may seem, actually opens up new strategies for curbing the most egregious inequalities, for it recasts their perpetuation as at least in part a matter of perceptual error and not just intractable moral vice.
Now, what about the second story—the one about those who allow certain of their fellows to attain superior status? In the last section, we noted that the first steps toward formal group leadership tended to be taken through the prestige system, such that valued contribution to the group becomes a prerequisite for the accordance of greater political power. In times of intertribal conflict especially, it can be in everyone’s best interest to allow a decisive, militarily experienced leader to take charge of the group decision-making process.
Here too, though, I think self-interest alone is explanatorily inadequate and that perceptual errors are crucial to a more complete picture. Tribal societies are larger than hunter-gatherer bands and chiefdoms larger still. We are adept at enforcing equality—material, political, and social—in small cooperative groups, but in more complex societies the suite of dispositions and competencies that subserve this capacity may yield contrary results at the societal level.
Of particular importance here is the local dominance effect. Enforcing equality in a very large group may be logistically impracticable in the absence of certain kinds of formal government, but it may still be easily doable within neighborhood subgroups. There are four principal kinds of social comparisons available to one under these circumstances: (1) upward comparisons between oneself and others in one’s local group; (2) upward comparisons between one’s local group and other groups; (3) downward comparisons between oneself and others in one’s local group; and (4) downward comparisons between one’s local group and other groups.
The LDE and UCDE jointly predict that the first comparison type ought to be the most emotionally and motivationally salient and the fourth the least. The problem with this, as far as the preservation of egalitarianism goes, is that inequalities (advantageous or disadvantageous) between local groups will be less attention-demanding than inequalities within local groups, and when people care more about intragroup inequality than intergroup inequality, the result can be a “crab bucket effect” whereby individuals are prevented from bettering their status (and thereby reducing overall inequality) by the equalizing actions of their local groupmates. This can have the unintended effect of reinforcing society-level class boundaries and thereby permitting intergroup inequality to increase even further.
The LDE disposes us to worry less about some ascendant aristocracy far away than whether that aristocracy taxes my neighbor and me equivalently.
ODT as Sister Theory
I hope I’ve given you some compelling sense of how this suite of psychological traits—inequality aversion, habituation, the local dominance effect, and the upward comparison dominance effect—can yield either egalitarian or hierarchical social arrangements under different material circumstances. I’ve recently discovered that my thinking has a number of interesting points of contact with psychologist Marilynn Brewer’s Optimal Distinctiveness Theory, and I want to take some space to briefly discuss these.
According to Optimal Distinctiveness Theory (ODT), the demands of group living have selected in us competing drives for “assimilation” (i.e., inclusion) and “differentiation” (i.e., distinction). Our preferred group affiliations reflect attempts to optimally satisfy these two drives. In brief, we want to belong to groups that are sufficiently distinct from other groups. There is a sort of oscillatory relationship between our needs for inclusion and distinction, such that the more one of them is satisfied, the more the other becomes activated. The following figure, from Leonardelli, Pickett, & Brewer (2010, linked above), neatly illustrates this dynamic:
Brewer’s theory has been tested in a wide variety of contexts and seems to be about as well-evidenced as anything in the domain of social psychology can hope to be at this point in the field’s development. All due disclaimers about the replication crisis and the complexity of human motivation aside, I think Brewer has adumbrated a real phenomenon here. I’m not sure, however, that she’s quite characterized accurately.
Consider ODT in the context of our hominin past subsequent to the evolution of pair-bonding (Chapais, 2008). At this point in our history, we see vastly increased levels of intergroup toleration, with both males and females regularly dispersing from their birth families to establish themselves in new groups. I speculated earlier about a process of negotiation by which extant group members seek new members with the abilities and dispositions to contribute to the good of the group (e.g., hunters who are both skilled and generous) and prospective new members seek groups within which they feel they can acquire the most prestige (and other resources gated by prestige) for their contributions.
The needs of identifying and successfully joining such groups, I suspect, played a considerable role in shaping the two drives Brewer characterizes. However, Brewer’s evolutionary account of these drives is essentially group-selectionist in character and therefore differs notably from mine. As she and colleagues put it:
Some groups survive and function better than others. Among other things, effective cooperation is constrained by group size. On the one hand, groups that are too small may engage strong obligation but the benefits of shared resources are limited. The advantage of extending social interdependence and cooperation to an ever wider circle comes from the ability to exploit resources across an expanded territory and buffer the effects of temporary depletions or scarcities in any one local environment. On the other hand, expansion, comes at the cost of increased demands on obligatory sharing and regulation of reciprocal cooperation. Both the carrying capacity of the environment and the capacity for distribution of resources, aid, and information inevitably constrain the potential size of cooperating social networks. Thus, effective social groups cannot be either too small or too large. To function, social collectives must be restricted to some optimal size—sufficiently large and inclusive to realize the advantages of extended cooperation, but sufficiently exclusive to avoid the disadvantages of spreading social interdependence too thin. It was this structural requirement for effective group living that formed a backdrop for the development of optimal distinctiveness theory. (Leonardelli, Pickett, & Brewer, 2010, p. 65)
For Brewer, then, optimal distinctiveness is really tracking group size. My interpretation, however, is that Brewer’s two drives evolved (or, more likely, were exapted and refined) primarily to help people place themselves in groups within which they can optimize their prestige/contribution ratio. Now, group size could certainly be a relevant variable here—larger groups may be harder to compete within, while small groups can only offer so many resources to even high-prestige individuals—and we may well be cuing into it when making decisions about which groups to affiliate or identify with, but I strongly suspect it’s not the only—or even the most prominent—variable we’re cuing in on.
One reason for this suspicion is that the needs for inclusion and distinction can be optimally satisfied in numerous ways that do not actually alter the size of the focal group. The canonical ODT story is that when assimilation needs are satisfied and differentiation needs activated the individual will seek to join (or found) a minority group or subgroup; but a number of other strategies can apparently satisfy activated differentiation needs just as well. These include (1) identifying with a group that defines itself against the mainstream (Hornsey & Jetten, 2004, pp. 252-253) ; (2) perceptually enhancing the distinctiveness of one’s group (Ibid., pp. 253-254); (3) role specialization within a group (Ibid., pp. 254-255; see also Vignoles, et al., 2002); (4) identifying with a group that normatively prescribes individualism (Hornsey & Jetten, 2004, pp. 255-256); (5) seeing oneself as a loyal but nonconformist group member (Ibid., pp. 256-257); and (6) seeing oneself as more normative than other group members (Ibid., pp. 257-258).
Since all these strategies for satisfying differentiation needs leave the focal group numerically unchanged, they are harder to account for on Brewer’s assumption that optimal distinctiveness is a matter of ideal group size—at least without invoking a variety of auxiliary mismatch hypotheses.
How does my account fare by comparison? I understand differentiation needs as needs for prestige beyond the minimum necessary to secure a baseline level of group membership. Strategies (3) and (6) are rather straightforward bids for prestige from one’s fellow local group members, (6) being the more outwardly competitive of the two. Strategy (5) is a combination of prestige elicitation (loyalty is a valued trait to advertise) and a reluctance to accord prestige to rival group members. Strategy (4) reflects a preference for groups with minimal prestige policing (remember, the context for each of these strategies is a state of high satisfaction of the need to belong and high activation of the need to stand apart; under such conditions, then, it makes sense that one would be less tolerant than normal of artificial limitations on the prestige one can earn).
The explanation for strategies (1) and (2) is a bit more involved, in that it deals with intergroup, rather than intragroup, comparisons. Because one’s identity under conditions of high assimilation is tightly bound up with the identity of the group, one may become more sensitive to prestige differentials between groups than within one’s group (high assimilation, in other words, can counteract the LDE). Essentially, one’s focal prestige arena jumps up a level of granularity, the competition staged now between groups rather than between co-localized individuals.
A second reason why I think prestige competition provides a better explanatory context for ODT than group size optimization per se is that minority group membership is associated with higher self-esteem (Leonardelli, Pickett, & Brewer, 2010), even when such groups face discrimination by majority groups (Gray-Little & Hafdahl, 2000; Crocker & Major, 1989). Now, because minority groups provide a better balance of inclusion and distinction, ODT predicts that people will tend to prefer to be in such groups rather than in majority groups. But mere preference satisfaction needn’t be accompanied by increased self-esteem; that it appears to be requires some further explanation.
Now, the orthodox ODTheorist is not without possible replies here. She may, for example, appeal to Terror Management Theory (TMT) and claim that minority group membership increases self-esteem because the cooperative efficiency of smaller groups better ensures the survival of their members. TMT, though, holds that the principal function of self-esteem is the reduction of death anxiety; consequently, while it would predict that increases in self-esteem tend to cause decreases in death anxiety, it’s not clear that it could explain why minority group membership—even if it does reduce death anxiety—appears to upregulate self-esteem. Either the reduction in death anxiety causes the increase in self-esteem—a reversal of the expected causal arrow—or self-esteem mediates the relationship between minority group membership and reduced death anxiety—in which case the relationship between minority group membership and self-esteem remains unexplained.
More general concerns attend the theoretical background of TMT, which I find at best incomplete, conceptually confused, and difficult to situate satisfactorily within the evolutionary framework in which ODT was formulated. For self-esteem as TMT conceptualizes it to be an evolved countermeasure to an awareness of our own mortality, ancestral death anxieties must have been sufficiently impairing as to offset their obvious survival benefits in the form of increased threat vigilance and avoidance of risky behaviors and situations. To say that I’m skeptical that this was the case would be an understatement. I don’t, however, want to get too bogged down in this largely tangential dialectic. I refer the curious reader to Leary and Schreindorfer (1997) for a constructive big-picture critique of TMT.
On my view, one of the key functions of self-esteem is to track one’s perceived relative prestige within a comparison group (a perspective consistent with Sociometer Theory) and to modulate one’s behavior accordingly. When our prestige is perceived to be relatively high, our self-esteem is high and we grow bolder and more assertive within the group—something the !Kung San foragers knew well and actively suppressed. Remember, though: Not all social comparisons count equally. The UCDE, in conjunction with Sociometer Theory, predicts that upward social comparisons will have a much stronger (negative) effect on self-esteem than the positive effect of downward social comparisons, and that, ceteris paribus, we ought to feel best in (and thus prefer) social groups with fewer upward comparison targets. Of course, such groups will tend to be relatively small—i.e., minority groups.
Much more deserves to be said here, but doing so would carry us too far off track. Hopefully, I’ve made at least a prima facie plausible case for situating ODT within a prestige competition context. Our motivation to seek optimal distinctiveness is not, I think, driven merely or primarily by considerations of group size but by considerations of what we might call “prestige potential.” That is, it’s not that we seek minority group membership per se, or because such groups functioned better in the ancestral environment; it’s that we seek membership in groups in which we feel we can attain the highest relative prestige for our contributions, and it just so happens that the most desirable groups will tend to be fairly small, since they will contain on average fewer people of obviously higher status than ourselves.
We’ve now taken a closer look at some of the key psychological players that underpin our prestige-seeking behavior and facilitate group-level transitions between egalitarianism and hierarchy. To recap, these are: (1) inequity aversion; (2) hedonic adaptation; (3) the local dominance effect; and (4) the upward comparison dominance effect. We might say, collectively, that they constitute something like a “Fairness Complex.” This is not a domain-specific “mental module” of the sort orthodox evolutionary psychologists might posit, but a loose assemblage of quirks and competences already present, if only in simplified form (e.g., coalitional psychology, habituation, negativity bias) in our pre-hominin ancestors. Note that it is a general system capable of regulating competitive behavior not just in the realm of social prestige, but in more material arenas as well. If, however, we add self-esteem (as conceptualized in Sociometer Theory) and its associated emotions—pride, shame, guilt, etc.—into the mix, then we have what we may properly call the “Prestige Complex.”
No doubt my characterization is incomplete, but I think I’ve mapped enough of the complex’s major contours and contexts to be able to speculate intelligently, if still roughly, about how it operates—and especially about how its affective and behavioral outputs are likely to vary in response to novel environmental inputs. It is precisely this speculatory task that will occupy the next, and most important, part of this series.