I. Introduction
While research on norms spans the humanities and human sciences, Christina Bicchieri’s social expectation account has recently risen to prominence in philosophy and serves as a touchstone and focal point for much discussion.Footnote 1 As such it will serve as our jumping off point and stalking horse in this essay. In the following section we will motivate and explain the core ideas of her account, and note some points of interest. Our discussion here will be brief, in part because Bicchieri’s view is well known, and in part because many of the other essays in this volume also discuss it. Moreover, we go into more detail below, where relevant, when discussing those aspects of Bicchieri’s account that contrast with the cognitive evolutionary perspective we endorse. Our third (and largest) section will describe the view of norms that emerges from this perspective, focusing on its picture of humans’ distinctive normative psychology, and illustrating how a better understanding of the mechanisms underlying the acquisition, performance, and transmission of norms can help shed light on how norms influence behavior, and how they are susceptible to change. Our fourth and final section will compare and contrast Bicchieri’s view with our own. We argue that while her account usefully and accurately explains the phenomena on which it focuses, it is too restricted in scope to function as a complete theory of norms and norm psychology. However, we believe that much of her social expectation account is compatible with the broader scope of the cognitive evolutionary approach, and that together they each make important contributions to a complete theory of human normative psychology.
II. Social Norms as Self-fulfilling Social Expectations
In a number of papers and books, Christina Bicchieri has developed an impressive body of work on social norms.Footnote 2 On Bicchieri’s picture, social norms are rules that govern the behavior of individuals, in turn creating group-level regularities. Not all common behaviors, or group-level regularities, indicate the presence of a social norm, however, so the core idea of the theory is that a social norm is in place only if the relevant group-level regularity is sustained by a particular cluster of psychological states found within individual actors. The most important components of this complex are two types of social expectations—social because they are beliefs about other people. The first, which Bicchieri calls empirical expectations, are beliefs about how other people will act, specifically about how they are likely to behave in a particular type of situation. The second, which she calls normative expectations, are beliefs about how other people think one should act, specifically beliefs about what a person ought to do in a particular type of situation. These beliefs may often be accompanied by the belief that others will not just disapprove of, but actively sanction those who violate their normative expectations, punishing them for failing to act as they should (or that others will approve of and actively reward those who satisfy normative expectations). Importantly, then, this definition itself does not strictly imply the existence of social sanctions and rewards, although normative expectations often lead to sanctions and rewards, empirically.
A third important component of Bicchieri’s social expectation theory speaks to the issue of motivation. Conditional conformity is another mark of social norms. On this view, in order for a group-level regularity to be properly explained as a social norm (rather than a moral norm, custom, practice, tradition, or any other group-level regularity), those engaged in the behavior must have a preference to follow the rule only if they believe those others will also follow the rule, and that those others believe that they ought to follow the rule—that is, only if both kinds of social expectations are met. Bicchieri distinguishes social norms proper from what she calls descriptive norms, which are behavioral regularities marked by preferences conditional on empirical expectations, but not necessarily upon normative expectations; intuitively, a descriptive norm is one where people engage in a behavior because they believe everyone else engages in that behavior, even if no one believes people should engage in the behavior. Thus, the expectations and beliefs that underlie social norms and descriptive norms alike are social, and the preferences leading to conforming behavior are also social—that is, dependent on how one expects others to behave. Patterns of behavior count as social norms only if they are brought about by a combination of beliefs about other people and motivations that depend upon other people. People prefer to comply with the social norm, but only on the condition that everyone else in the group is complying, and everyone else thinks it’s the right thing to do, too.
When all of these conditions are met, a stable group-level regularity is sustained by the self-fulfilling interplay of the social expectations and social preferences of the individuals that make up the group, or network. Only group-level regularities thus stabilized, on Bicchieri’s view, are properly identified as social norms. In light of this set of individually necessary and jointly sufficient conditions, we’ll call the cluster of psychological states required for the presence of a social norm a Bicchieri-cluster, and the type of behavioral regularity it produces, which she simply calls a social norm, a Bicchieri-norm.
A couple more comments will be useful here. First, though Bicchieri’s 2006 book is suggestively entitled The Grammar of Society: the Nature and Dynamics of Social Norms, its conceptual roots run through game theory and economics, and it draws on the resources of rational choice theory much more than linguistics (or cognitive science more broadly). Indeed, while it rejects a purely behavioral or group-level definition of social norms, the types of psychological states at its core are essentially those of common belief/desire folk psychology, or the close counterparts (expectations and preferences, for instance) typically represented in game theoretic models. That said, Bicchieri and her colleagues have recently developed the account in further detail, appealing to mental representations such as scripts and schemata,Footnote 3 social roles and positions in social networks, and traits of particular actors, such as perceived self-efficacy.Footnote 4 Related to this, she also qualifies the definition of a social norm by appeal to the useful concept of a reference network, the group of people one takes into account in one’s social expectations and preferences, for any given norm. Finally, as she and her collaborators have emphasized recently, the self-fulfilling social expectations account has direct implications for crafting policies and other kinds of interventions designed to change these kinds of group-level regularities. Bicchieri-norms are stabilized by expectations and preferences about what others are likely to do, and what they are likely to think. Thus, efforts at redirecting behavior governed by Bicchieri-norms are unlikely to succeed when they merely attempt to change personal preferences, or to correct false factual beliefs about health risks or other undesirable outcomes of the behavior in question. Rather, since a Bicchieri-norm is sustained by a Bicchieri-cluster, and the psychological states in a Bicchieri-cluster are social and other-oriented, successful interventions need to focus on what the members in the reference network believe about what the other members in the reference network believe.
III. A Cognitive Evolutionary Approach to Norms
We will return to the Bicchieri view below, but now set it aside in order to present a cognitive evolutionary approach. In this section we draw on a range of recent work coming out of the cognitive sciences characterizing human capacities for cognizing and acquiring norms, and describe what are emerging as the key psychological properties of the mechanisms that underpin these capacities. Three points before we get started: First, while our account is informed by the insights and findings of a number of theorists, it is not identical to the view endorsed by any of them; our intent is to present a picture that represents the overlap and convergence of what we take to be the most promising work. For ease of exposition, we will refer to ours as the Minimal Account, which aims to capture what an emerging consensus sees as the core features of human norm psychology.Footnote 5 Second, while we appeal to work often discussed under the heading of “empirical moral psychology,” the conception of normative psychology we work with is both broader and narrower than what might be considered the psychology of “morality.” On one hand, there is a great deal of work in moral psychology that we will not consider here, because it is not about norms or normative cognition at all, but rather is about such issues as, for instance, the identification of intentional versus unintentional behavior, character trait-based versus situation-based explanations of behavior, conceptions of the true self, how to assign responsibility and blame for implicit bias, and so on. On the other hand, the Minimal Account applies not just to putatively moral norms, but to norms in general, including norms of logic, language, epistemology, aesthetics, religion, or etiquette.Footnote 6 Finally, our characterization of the psychological underpinnings of normative behavior posited by the Minimal Account draws on a richer vocabulary than folk psychology provides, discussing mechanisms and subsystems sometimes in lieu of beliefs and desires (or belief-like and desire-like states such as expectations and preferences).Footnote 7 While departure from commonsense origins often accompanies a research program’s gains in conceptual sophistication and explanatory power, we will address some particular implications of this fact when we compare the Minimal Account to the self-fulfilling social expectation view in section four.
A. The psychology of normativity: A minimal account
We begin with normativity in general. From a logical point of view, the defining feature of a normative proposition is that it says something about what is required, allowed, or forbidden among human activities. Normative statements either specify or imply what “should” or “shouldn’t” be done, expressing a rule or making some prescriptive claim about how people ought to think, feel, judge, or behave. This feature provides a useful point of departure for making sense of the relationship between the logic of norms, on the one hand, and the psychology of norms, on the other. First, it allows us to distinguish normative from normal. A behavior does not count as norm-governed, and a pattern of behavior does not count as realizing a norm, simply in virtue of the fact that it happens to be normal, statistically speaking, within a population. Second, it allows us to distinguish good and bad from right and wrong. As Shaun Nichols, points out, toothaches and natural disasters are bad, but they aren’t wrong; in such cases no rule has been violated or transgression committed.Footnote 8 In contrast, if a person steals a few dollars from the tip jar at a coffee shop, or keeps showing up at dinner parties empty handed and without the requisite bottle of wine, they are not just acting badly, they are doing something they shouldn’t do. In both cases the person’s action breaks a rule, written or otherwise, in virtue of which it can properly be evaluated as wrong. Thus, in our sense, normative psychology is the psychology of oughts, and so part of the psychology of rules.Footnote 9
Consequently, it is also the psychology of compliance and enforcement. Indeed, from a psychological point of view, a distinctive, defining feature of norms concerns approval and disapproval, and the use of punishments and rewards to influence behavior. Again, a behavior is not normative in our sense merely in virtue of being typical. If reward and punishment play no role in explaining why a given pattern of behavior is statistically normal within a population, this suggests the regularity occurs for reasons other than the presence of a norm prescribing it, and perhaps that the individuals in the population themselves are indifferent to whether anyone actually engages in the behavior. This amounts to the claim that in such a case there is no sense, at least within the confines of causal, naturalistic explanation, in which anyone ought or ought not to engage in that behavior. In short, norms imply “oughts,” and “oughts” imply punishment and reward.
Guided by similar assumptions, researchers like Richerson and Boyd, Chudek and Henrich, and Gelfand and Jackson summarize bodies of evidence supporting the existence of “a suite of genetically evolved cognitive mechanisms for rapidly perceiving local norms and internalizing them”.Footnote 10 We refer to this complex of traits as the norm system, a more-or-less integrated, hierarchically organized collection of functional capacities and subsystems. This work suggests that some parts of the system are genetic adaptations that develop reliably across a wide range of different cultural environments, while others are highly local, typically acquired via imitation and social learning, but also sometimes through individual trial and error, and/or personal deliberation and reasoning. The combination of both innate (that is, genetically inherited) structure and social learning allows our Minimal Account of the norm system to accommodate the fact that while norms in general are a ubiquitous part of human social life in all cultures, the content of particular norms varies greatly from culture to culture, and changes over time within cultures. An appeal to genetic adaptation is able to explain why human individuals are universally capable of identifying, acquiring, and performing some set of norms or other, but it falters when it comes to explaining diversity and variation of norms across cultures; innate capacities for learning do not by themselves explain the particular content that an individual comes to learn, even if those learning mechanisms are themselves targeted at a specific and well delineated domain.Footnote 11 In the same way, appeals to innate psychological machinery explain why humans universally share the ability to learn some language or other, but it does not explain why anyone speaks the particular languages she speaks, rather than others.Footnote 12
Sripada and Stich provide a high-level model of the core psychological mechanisms comprising an individual’s norm system.Footnote 13 That model depicts a key distinction between those mechanisms responsible for the identification and acquisition of local norms, on one hand, and the performance of norms that have been acquired, on the other. Mechanisms on both sides of this divide have some aspects that are purely cognitive and representational, as well as other aspects that are affective and motivational. And while motivational features may include obvious and powerful passions of righteous anger and contemptuous disgust, they also include more subtle motives. For instance, motivational capacities within the acquisition subsystem may influence attention, shaping what we find interesting, salient, and relevant in the behavior and interactions of other people. Even children do not need to be cajoled to notice social rules, nor are they passive learners that always have to be actively or explicitly taught them. Those as young as three years old appear motivated to attend to cues indicating that a behavior is normative, and to draw inferences about the rule that is governing it.Footnote 14
But the affective and motivational aspects involved in following and enforcing norms are particularly important for our purposes, largely because they give acquired social rules their distinctive motivational force, or normative force for short. Our Minimal Account incorporates the claims that this normative force is (1) intrinsic, as opposed to instrumental, that it is (2) two-pronged—both self- and other-oriented—and that it (3) can be quite strong. More fully, when an individual genuinely acquires a norm, rather than merely becoming aware of, or simply cognitively grasping, a social rule, the rule is represented in the database of her norm system, and thereby inherits the motivational features associated with norm-guided behavior. Thus, to fully acquire a norm is to develop an intrinsic motivation to perform it, rather than following or enforcing it out of instrumental motivation connected to some other end, such as avoiding punishment.Footnote 15 And normative motivation in this sense is not just non-instrumental, but also two-pronged: a person who has acquired a norm is thereby motivated to obey simply because it is “the right thing to do,” but she is also motivated to enforce the norm on others, punishing those who violate it, and often forming longer-standing reactive attitudes toward transgressors. This is not to say that a person’s behavior is always in accord with the norm, or that she actually enacts the punishments or rewards she deems appropriate, but merely that this piece of her psychology generates an impulse to do so. Intrinsic motivation is not irresistible motivation, and both of these types of normative motivations, like any others, can be suppressed or superseded by other, more powerful motivations. That said, the force of both compliance and enforcement motivations can be, and often is, quite powerful, sometimes overriding self-interested goals. This explains many examples of personal sacrifice for the sake of a group or a cause, including, in extreme cases, martyrdom.Footnote 16 Aside from strength, the quality of normative motivation can differ along other dimensions as well, some of which have to do with details of situations to which a norm might apply, some having to do with the norm itself, and some perhaps with which emotion the norm draws on, for instance anger, disgust, guilt, shame, outrage, spite, and so forth.Footnote 17
Appeal to the norm system, and the normative force it confers on acquired norms, surely will not explain every instance of cooperation or collective action, of people acting in accordance with social rules, or of people sanctioning those who disobey them. One person may follow a social rule out of an instrumental desire to avoid the punishments that come from violating it, while another person may follow the same rule for its own sake, regardless of the instrumental, practical value of doing so. But we maintain that this Minimal Account of normative psychology, and the research on which it draws, provides indispensible explanatory purchase in understanding how individuals dole out social rewards and punishments, and how they respond to the rewards and punishments doled out by others. It also fits into a larger picture about the group-level characteristics of norms, and factors that influence how those change over time. We turn to this next.Footnote 18
B. Norm psychology, cultural evolution, and social change
To clarify how this Minimal Account might contribute to a broader picture of social change, we begin by showing how the norm system helps bind individuals to groups, getting them in sync with the social arrangements that structure life within any given group. The Minimal Account appeals to social learning to explain how an individual comes to have the particular norms represented in her norm system, suggesting that, as in the case of language, the acquisition mechanism of an individual’s norm system intuitively and automatically “soaks up” norms from her local environment.Footnote 19 The plausibility of this claim comes from the commonalities in the developmental trajectory of individuals’ normative capacities that appear to hold across cultures, as well as from patterns in the distribution of particular norms throughout populations. Like languages, norms exhibit common patterns of within-group similarity and between-group difference.Footnote 20
The Minimal Account thus depicts human normative psychology as “expecting” certain kinds of cues and regularities in an individual’s social world, from which it will be able to glean information about the particular norms that prevail locally. These cues manifest in other people and their interactions with each other, as well as within the individual herself, and the most salient behaviors will be group-level regularities stabilized by reward and punishment. These pockets of the social world provide information about what kinds of actions are forbidden, permissible, and required, and, in the episodes of social learning that fix on them, the acquisition machinery of a person’s norm system makes (perhaps innately constrained) inferences about what the rule is that governs the observed interaction (as well, perhaps, as what follows from that rule).Footnote 21 When she acquires those norms, a person becomes attuned to her culture and its social arrangements; her own normative sensibility harmonizes with the general normative framework of the group of which she is a member, with perhaps some fine-grained calibration determined by the particular set of social roles she occupies in it. Due to the normative force exerted by the norms she has acquired, she not only typically acts in accordance with them, but typically enforces them as well. Thus, by obeying rules and punishing transgressors, she makes her individual contribution to the stability of the collective, supporting the durable structure of the often long-standing social arrangements of her culture.
Or so it might go in a “perfect” world. On the highly idealized picture just described, the norm system would successfully bind individuals to their groups, but seamlessly. The (somewhat unsettling?) vision just sketched suggests individuals who are perfect learners getting totally enculturated into a society, flawlessly performing its norms, making them (the individuals) pristine models for the next wave of norm learners, and turning them (the individuals again) into impeccable members of a society that is almost completely static, and whose norms thus remain, but for the occasional exogenous perturbation, largely unchanged as they are passed from one generation of individuals to the next. The reality is of course (happily?) messier, but this picture, like any good idealization, gets some key things right. Many group-level regularities are far from ephemeral; social arrangements and the interconnected clusters of norms that govern them can be quite stable over time, even doggedly resistant to change. Moreover, modeling work in evolutionary game theory strongly suggests that one of the most important stabilizing factors is punishment.Footnote 22 Via the kind of feedback loops just described, when the members of a group enforce norms and punish transgressors, the aggregated effect can render interconnected clusters of norms endogenously stable. Situations that fit this description are in equilibrium, or in an evolutionarily stable state, and the mechanisms that keep them in homeostasis also make them robust in the face of a range of external influences—including deliberate attempts to change them.Footnote 23
Two points will allow us to begin complicating this picture. First, as noted above, part of the messiness of the actual world is that there is cultural variation in norms, social arrangements, and group-level behavioral regularities. The norms and arrangements that organize one group can differ dramatically from the norms and arrangements that organize another group, and yet both can be endogenously stable. A second way in which reality departs from the idealized picture is that the norms and social arrangements of groups do, in fact, change over time, shifting from one stable state to another. Together these points show that there is more than one way to be stable (some of which are more adaptive, collectively efficient, and cooperative, some of which are less so). Stability is a property that can be realized by many different configurations of norms; there are multiple stable equilibria, and punishment and the operation of human normative psychology can in principle stabilize any of them. And since there are so many ways to be stable, appeal to norms, punishment, and the operation of human normative psychology alone will be inadequate to fully explain either variation or change. While it will be a key, perhaps necessary, factor in any viable explanation, the Minimal Account can’t by itself explain why those stable configurations of norms that have actually been realized have actually been realized, while other possible stable configurations have not. Nor can it fully explain, by itself, the fact that norms and social arrangements change over time, or how. This requires further resources.Footnote 24
We maintain, however, that theories of cultural evolution in general can deliver these goods. This shouldn’t be controversial; “evolution” means “change,” and so theories of cultural evolution are theories of cultural change.Footnote 25 In general, such theories use the concepts and models of biological evolution and evolutionary game theory to understand changes in the frequencies and distribution of cultural variants, where cultural variants are understood as behavior-affecting bits of information that are acquired socially.Footnote 26 Variants are typically different kinds of socially transmitted ideas, values, beliefs, preferences, skills, habits, norms, and so on. These are subject to competition in the face of cultural selective pressures of differing strengths and from assorted origins. Cultural variants are understood to have varying levels of cultural fitness depending on how likely they are to be copied and transmitted to others, and thus spread across populations and between generations.Footnote 27 When quantified, these properties can be represented in the models, which are used to calculate the changes in the frequencies of cultural variants that will occur under various conditions. This literature is thriving and continues to grow, but for now we will briefly mention a few representative and important ways that the psychological features of individuals, including features of their norm systems, can influence the transmission and spread of cultural variants. Since norms are cultural variants themselves, these psychological features are relevant for attempts to influence social change, or intervene in the evolution of social arrangements.
C. Transmission and learning heuristics
Unlike in our idealized story, humans are not perfect learners, and cultural transmission is reliable but not extremely so, and it does not always lend itself to precise, high-fidelity copying. This produces noise and variation, though some of this is counterbalanced by the fact that humans are not equal-opportunity learners. Rather, social learning capacities, including those responsible for the acquisition of norms, are guided by a number of transmission and learning heuristics. When added to humans’ hypertrophied (compared to other social animals) tendency to imitate each other, these heuristics are features that facilitate a group’s collective ability to more effectively and efficiently produce adaptive cultural variants. These heuristics go some way in “correcting” for the effects of noisy transmission or sloppy individual learning.Footnote 28 For example, a prestige heuristic makes us more likely to adopt norms and other cultural variants from those with the greatest success and status within our group, or within some subculture or reference group with which we are concerned.Footnote 29 Similarly, a conformity bias makes us more likely to imitate the most common behaviors, and, in the case of norms, acquire those performed by most of our peers, or aspirational peers in the in-group we wish to join.Footnote 30 Together, these exert systematic influence on how the frequencies of norms in a population can change over time. It is worth noting that conformity and prestige heuristics give norms and other cultural variants a cultural fitness boost independently of the content of those variants, and independently of whether or not the norms are just, fair, utility maximizing, and so forth. Intuitively, the idea is that messengers matter, and these heuristics exert an influence on the spread of norms based solely on who performs them. Nevertheless, they still directly affect how norms are transmitted, influencing which are more likely to be copied and acquired by other members of the population.
D. Epidemiology-inspired psychological stickiness
Another psychological factor that can influence the transmission and cultural fitness of norms—and one that also responds in part to features other than the content of the variants in question—is what we will call psychological stickiness. This is the degree to which a variant easily “meshes” with a variety of features of cognitive machinery other than capacities dedicated specifically to social learning. The general epidemiological approach to culture has been pioneered and developed by Dan Sperber and colleagues, and fruitfully applied to the study of religion.Footnote 31 The idea can be extended to norms as well: to the extent that a norm is salient, easy to identify, and easy to remember, it will also be more easily acquired and transmitted to others.Footnote 32 Moreover, a norm might engage psychological mechanisms in addition to those core components of the norm system, and in doing so it may get a further cultural fitness boost. Other psychological mechanisms that have been shown to boost the fitness of a norm include particular emotions, such as disgust,Footnote 33 narrative capacities, or the way a norm is embedded in a recognizable kind of plot or narrative,Footnote 34 close association with attention-grabbing, button-pushing, supernormal stimuli,Footnote 35 or really anything else that increases the norm’s salience, ease of comprehension, retention, and transmission by making it attractive and intuitive to some piece of our psychological repertoire. Indeed, meshing with the other norms that an individual has already acquired could itself make a new candidate norm psychologically stickier for that person, and if other individuals in the group are similarly normatively attuned, the new candidate norm will be stickier throughout the population, thereby receiving a fitness boost. In this way, norms influence the selective environment for other norms.Footnote 36
E. Ancient sociality and tribal sociality
While transmission and learning heuristics and psychological stickiness are general factors that can affect the spread of any kind of cultural variant, recent research suggests an important division between two broad families of psychological systems that humans have for navigating different kinds of social exchanges. In virtue of this, these systems are more directly relevant to norms and normative behavior, but in importantly different ways. The idea is that human capacities for interacting with each other have two distinct strata, and while both can help facilitate cooperation and coordination, each has been differently shaped by its own distinct evolutionary history, and those differences remain visible in the ways each influences social interaction today. The ancient social instinctsFootnote 37 are those that we share with many other social animals, which are responsible for our interactions with family and friends—conspecifics with whom one shares blood relations or regular patterns of interaction. These more ancient capacities operate according to principles associated with kin selection and (often reputation-based) reciprocal altruistic solutions to cooperative dilemmas. This complex of psychological traits also includes the elements of our status psychology that are based on dominance.
Tribal social instincts, by contrast, are evolutionarily recent, and include uniquely human features. As their name suggests, they are responsible for our interactions with members of tribal-sized groups, defined in abstract, symbolic terms, whose size far outstrips that of a circle of family members, friends and acquaintances (even acquaintances merely by reputation). These include many features of human normative psychology, as described above, but other components as well. One such is a set of capacities related to monitoring tribal membership. A mark of these is a heightened sensitivity to tribal boundaries, or to the symbolic markings that people use to signal what groups they belong to, as well as what station or roles they occupy within those groups, and thus what norms and beliefs they have likely acquired. These capacities also have motivational features, typically shaping differences in the way individuals behave toward ingroup and outgroup members.Footnote 38 Indeed, there has been a great deal of work recently exploring how ingroup favoritism and outgroup bias manifest, especially with respect to other aspects of social instincts and cooperation. Some of this work suggests that these heuristics emerge rather early in development, driving noticeable differences in norm enforcement and reputation management.Footnote 39 In a series of papers, Carsten De Dreu and colleagues have shown that this kind of tribalism runs deep, and that the effects of oxytocin on social interactions bifurcate along ingroup and outgroup lines on a number of different dimensions, including empathy, conformity, and cooperative and competitive tendencies.Footnote 40
Another aspect of our tribal psychology that is relevant to social change and the transmission of norms is the human disposition to give deference based on prestige, rather than just physical dominance.Footnote 41 This can manifest in many behavioral ways associated with status and status-related behavior, which includes, as mentioned above, differentially imitating highly prestigious individuals and adopting norms they evince and endorse.Footnote 42
Finally, tribal social instincts are thought to include a suite of uniquely human emotions that evolved in tandem with our normative, tribal membership, and prestige capacities. Examples of emotions that have been hypothesized to fit this description include empathy, which allows us to share feelings and cognitive states with others, shame and guilt, which can be seen to function as internalized enforcement mechanisms that make a person more likely to obey norms she has acquired, and loyalty and pride, which can emotionally bind one to her tribe. Alternatively, Kelly argues that disgust, an emotion that is uniquely human but relatively ancient and not initially social, nevertheless came to play several roles in our more modern tribal social psychology, including producing stigmatizing, dehumanizing aversions toward the members, norms, and values of particularly loathed outgroups.Footnote 43
An important claim here is that, like the newer emotions and the addition of prestige to our status psychology, neither the human norm system nor the sensitivities to group membership and tribal boundaries are merely refinements to, or elaborations of, the ancient social instincts we share with other animals. They are a new thing under the sun. One happier result of this is that humans can cooperate with each other in many different ways, on a number of different scales. Ancient and tribal instincts provide different tricks to identify others in the group as likely cooperators and, more importantly, to detect and sanction defectors.Footnote 44 It has been posited that together these can collectively act as a “moral hidden hand,” or a source of pro-social behavior and psychological stickiness that influences the spread and evolution of norms. Recall that on its own punishment can stabilize any norm: the useful and the pointless alike, as well as the just and unfair, the cruel and the kind. The moral hidden hand, however, can act as one of the pressures that drive social change toward the better—or at least the more prosocial—by giving a cultural fitness boost to norms that lead individuals to act for the good of the group, paying personal costs for the sake of others. Thus, norms that activate our feelings of empathy, our sense of fairness, or our aversion to gratuitous harm, receive a transmission advantage over those that do not. This, in turn, acts as a gentle but persistent selection pressure favoring more equitable and compassionate social arrangements over the long run, because the norms that prescribe such arrangements are more likely to “mesh” well with the range of human cooperative instincts.Footnote 45
A less happy but equally interesting upshot is that these two sets of instincts can also be at odds with each other, and the resulting struggles can manifest both within individuals, and, collectively, at the level of groups. In other words, “These new tribal social instincts were superimposed onto human psychology without eliminating ancient ones favoring friends and kin. This resulted in an inherent conflict built into human social life.”Footnote 46
Finally, a controversial but compelling hypothesis posits that another distinct and important process driving the cultural evolution of norms and social change is cultural group selection.Footnote 47 Darwin himself famously endorsed the basic idea: “A tribe including many members who, from possessing in a high degree the spirit of patriotism, fidelity, obedience, courage, and sympathy, were always ready to aid one another, and to sacrifice themselves for the common good, would be victorious over most other tribes; and this would be natural selection.”Footnote 48 Correctly understood, this is a macro form of cultural evolution, and concerns competition and cultural selection between different clusters of norms, practices, and other cultural variants. When one “tribe”—a group bound together by shared norms, values, and other cultural institutions—competes with another, the tribe equipped with the more efficient, effective, and advantageous cluster of variants will typically win, and its technology, norms, and social arrangements will spread at the expense of the defeated tribe’s.Footnote 49
IV. The Social Expectations Account Meets The Cognitive Evolutionary Approach: Two Ways of Putting the “Social” in Social Norms
Both Bicchieri’s account and the cognitive evolutionary account distinguish normative behavior from merely normal behavior, and both accept a common broad notion of what a norm is, namely, a rule of behavior that has both individual- and group-level properties.Footnote 50 Moreover, while group- or population-level regularities are central to each account, both reject the idea that norms are to be accounted for purely at the group or population level. Rather, both identify norms by appeal to psychological characteristics of the individuals that make up the group. But each does so by appeal to different kinds of psychological structure, resulting in different explanations for the stability of the relevant group-level regularities. The social expectations account defines norms in terms of what we called the Bicchieri-cluster: empirical expectations about what others will do, normative expectations about what they think should be done, and conditional preferences to comply if others do. The cognitive evolutionary account appeals to a suite of systems that includes mechanisms for the social learning of practices of enforcement (including rewards), along with other special mechanisms that make up the human norm system, as specified by the Minimal Account.
Other than the expected terminological differences (which we suspect are reconcilable), the key contrasts between the two views flow from what they have to say about motivation and acquisition. On our view, human psychology is equipped with a distinctive, specifically normative kind of motivation. This motivation is intrinsic, driving an individual to follow and enforce norms for their own sake, regardless of instrumental or conditional reasons. On this account, the motivation to comply with or enforce an acquired norm is, in the relevant sense, asocial, or independent of social reasons. This seems to be in direct contrast to the social expectations account, which defines norms in terms of an individual’s social, other-oriented beliefs and preferences. In this way, it defines norms in contrast to such behavioral regularities as customs, practices, and traditions. This strikes us as an odd feature of the account, since it seems obvious to us, and natural to say, that customs, practices, and traditions are themselves governed by norms, that is, by culturally acquired, intrinsically motivating, socially enforced, rules that specify how people ought to behave.
On the cognitive evolutionary account, what makes a norm social isn’t necessarily the expectations or preferences that cause anyone to conform to it or enforce it, but rather the way in which it is acquired by individuals: norms are socially learned from cultural peers and parents. This is why we have emphasized the psychological machinery dedicated to norm acquisition, and the ways in which the Minimal Account shows this machinery to be both automatic and intuitive, but also exquisitely sensitive to cues of group membership and social status. The transmission of norms from one individual to another is of enormous importance, especially since, once a norm has been acquired and represented in an individual’s norm system, that norm becomes imbued with intrinsic motivations of substantial force.
On the evolutionary cognitive account, the following state of affairs is the paradigm for norms and human normativity: social acquisition of a behavioral rule that in turn leads to intrinsic motivation to comply and sanction. However, motivation is not compliance, and intrinsic motivation is not unconditional compliance, but rather compliance conditional on learning history, rather than social or normative expectations. We understand Bicchieri’s temptation to cordon off cases involving unconditional compliance as extreme, to save such hard cases for later, and to dub them “moral norms” in order to distinguish them from the “social norms” on which her account focuses. There is a large and difficult conversation to had here, but the most pressing problem is not that we are without any way to delineate morality from other forms of normativity, or to cleanly distinguish moral norms from social norms, conventions, prudential rules, or the rest. The problem is that there are too many plausible ways to delineate the domain of morality, but no two seem to slice the pie in the same way.Footnote 51 For instance, Bicchieri’s account of social norms places them within a larger taxonomy of behavioral phenomena.Footnote 52 In it, she separates out what she calls customs, legal injunctions, and moral rules from what she calls norms (descriptive and social). Her conception of moral norms defines them as those wherein compliance is unconditional, not dependent either on social expectations or social preferences. She does not state it in these terms but this seems to entail that individuals conform to moral norms because they are intrinsically motivated to do so, and also that such intrinsic motivation only accompanies moral norms, that it is a unique and essential mark of morality. On our view, this mistakes a crucial feature common to all norms—normative force—for one that is specific to moral norms alone. In doing so, it also restricts her account so that it has few conceptual resources for explaining the cases that emerge as paradigmatic from the point of view of the cognitive evolutionary account. Culturally acquired intrinsic motivations are also conferred on all kinds of (intuitively nonmoral) norms, including epistemic norms, aesthetic norms, norms of logic, language, religion, etiquette, and so on. All of these phenomena are important, but they fall outside of Bicchieri’s conception of social norms.
That said, on our view, the narrow focus of the social expectations account is quite understandable from the point of view of the practical aims of providing actionable policy advice and diagnosing, measuring, and changing many important norms.Footnote 53 However, we would add that what is good for practical purposes is not necessarily what is good from the point of view of full theoretical understanding, and in the absence of full theoretical understanding it remains unclear what is best for practical purposes. Bicchieri-norms bracket off what the cognitive evolutionary account identifies as the most interesting, central and important aspects of human normativity, including the roles they play in generating cooperative behavior and collective action, and the psychological adaptations that evolved specifically for negotiating social environments in which norms are prevalent. Bicchieri-norms are an important subset of norm-related social phenomena at the intersection of individual psychologies and collective level regularities, but it remains unclear how much social change can be affected by focusing only on this subset. Of the idea that Bicchieri-norms count as “social norms” proper, or that they make up the core subject matter of the study of norms and normativity in general: we remain unconvinced.
Fortunately, we see no reason why Bicchieri-norms cannot still be incorporated into a theoretical perspective that is much broader in scope, which does address the study of norms and normativity in general. In many important cases of normative behavior, individuals do conform conditionally, on the basis of instrumental motivations derived from social expectations. We see no reason to reject or exclude the detailed account of those cases that Bicchieri has developed.
V. Conclusion
We end on an ecumenical note. The types of resources we have discussed in this essay will all eventually contribute to our nascent but growing understanding of norms and social change. Perhaps they will be of use in trying to produce and guide social change as well.Footnote 54 While we hold that evolutionary thought has a foundational role to play in this coming synthesis, we disagree with the claim that “nothing about norms and institutions makes sense except in the light of evolution.”Footnote 55 On the contrary, Christina Bicchieri and her collaborators on the social expectations account have provided a clear, useful, and high-resolution understanding of an important subset of norms and norm-related social phenomena. While we have highlighted the differences and limitations of that account, particularly with regard to motivation and acquisition, we are optimistic that it can ultimately be integrated into the evolutionarily based picture we have advocated. Together, they can shed more light on all kinds of norms, and on different aspects of the complicated tapestry of human normative psychology, cultural transmission, and social change.
