Politics in a State of Nature

William A Edmundson

"Aristotle thought we are by nature political animals, but the state-ofnature tradition sees political society not as natural but as an artifice. For this tradition, political society can usefully be conceived as emerging from a prepolitical state of nature by the exercise of innate normative powers. Those powers, together with the rest of our native normative endowment, both make possible the construction of the state, and place sharp limits on the state’s just powers and prerogatives. A state-of-nature theory has three components. One is an account of the native normative endowment, or “NNE.” Two is an account of how the state is constructed using the tools included in the NNE. Three is an account of the state’s resulting normative endowment, which includes a (purported) moral power to impose duties of obedience. State-of-nature theories disagree about the NNE. For Locke, it included a “natural executive right” to punish wrongdoing. Recent social scientific findings suggest a quite different NNE. Contrary to Locke, people do not behave in experimental settings as one would predict if they possessed a “natural executive right” to punish wrongdoing. Moral reproof is subject to standing norms. These norms limit the range of eligible reprovers. The social science can support two claims. One, is that the NNE is (as Aristotle held) already political. The other is that political authority can be re-conceived as a matter of standing—that is, as the state’s unique moral permission coercively to enforce moral norms, rather than as a moral power to impose freestanding duties of obedience."

bs_bs_banner Ratio Juris. Vol. 26 No. 2 June 2013 (149–86) Politics in a State of Nature WILLIAM A. EDMUNDSON* Abstract. Aristotle thought we are by nature political animals, but the state-ofnature tradition sees political society not as natural but as an artifice. For this tradition, political society can usefully be conceived as emerging from a prepolitical state of nature by the exercise of innate normative powers. Those powers, together with the rest of our native normative endowment, both make possible the construction of the state, and place sharp limits on the state’s just powers and prerogatives. A state-of-nature theory has three components. One is an account of the native normative endowment, or “NNE.” Two is an account of how the state is constructed using the tools included in the NNE. Three is an account of the state’s resulting normative endowment, which includes a (purported) moral power to impose duties of obedience. State-of-nature theories disagree about the NNE. For Locke, it included a “natural executive right” to punish wrongdoing. Recent social scientific findings suggest a quite different NNE. Contrary to Locke, people do not behave in experimental settings as one would predict if they possessed a “natural executive right” to punish wrongdoing. Moral reproof is subject to standing norms. These norms limit the range of eligible reprovers. The social science can support two claims. One, is that the NNE is (as Aristotle held) already political. The other is that political authority can be re-conceived as a matter of standing—that is, as the state’s unique moral permission coercively to enforce moral norms, rather than as a moral power to impose freestanding duties of obedience. 1. Introduction The idea of a pre-political state of nature is almost the defining feature of modern theories of political authority. By “modern” I mean “not ancient even if centuries old,” and I share the general view that Thomas Hobbes and John Locke are its seminal figures. State-of-nature theorizing repudi* I thank Dorota Mokrosinska and Emily Crookston for the opportunity to present this to the Academy Colloquium on Political Obligation and Legitimacy of the State, at the Koninklijke Nederlandse Akademie van Wetenschappen, in Amsterdam, in June 2011. Jerry Postema gave a helpful commentary and I benefited from the remarks of other participants, all of whom I should mention, but Dave Estlund, John Simmons, Govert den Hartogh, Doug MacLean, and Massimo Renzo were particularly forceful about what I had yet to do. Irit Samet, Jerry Gaus, Luciano Venezia, Bas van der Vossen, Ekow Yankah, Andrew I. Cohen, and George Rainbolt valuably commented on a later draft. © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden 02148, USA. 150 William A. Edmundson ates the older, Aristotelian notion that a human is by nature a political animal: zōon politikon. Political philosophy in the modern tradition sees political life not as natural but as an artifice. For this tradition, political society can usefully be conceived as emerging from a pre-political state of nature by the exercise of certain innate normative powers. Those powers, together with the rest of our native normative endowment––or “NNE”–– make possible the construction of the state, and at the same time place sharp limits on the state’s just powers and prerogatives. State-of-nature theories, also known as social-contract theories, have three components. One is an account of the native normative endowment, the NNE. Two is an account of how the state is constructed using the tools and materials included in the NNE. Three is an account of the state’s resulting normative endowment. This resulting normative endowment typically––and, many say, essentially––includes a (purported) general moral power to impose duties of obedience. This latter power has no original in the state of nature, and so one of the self-imposed tasks of modern political philosophy is that of explaining how such a power might have arisen. The moral power to covenant is the nearest antecedent, and so modern political philosophy has viewed the consent of the governed as pre-eminently––and even uniquely––capable of legitimating political states. But universal consent to conform to whatever the state might (even within bounds) command is nowhere to be found in the historical record. The persistent difficulty of refurbishing the idea of consent has led a significant number of philosophers working in the modern tradition to conclude that states cannot possess the moral power that they distinctively claim to have: a power to impose general duties of obedience by mere decree, typically by some process of legislation. Any political philosophy must question the legitimacy of the state; but stateof-nature theorizing seems inexorably to bend in the direction of philosophical anarchism, which denies the legitimacy of modern states, or at the extreme, its very possibility. Bluntly, modern political philosophy would have us dwell “on the edge of anarchy,” as a leading contemporary Lockean has titled a book. State-of-nature theories disagree about the NNE. In Hobbes’s (1996) account, the state of nature is one of “perpetual Warre,” and the political state is an “artificial man” constructed by individuals using a minimal toolbox of normative elements: a permission to do whatever one believes will aid one’s survival, and a power to alienate such permissions by covenant. Locke, differing from Hobbes, saw the pre-political state of nature as one not of warfare, but of inconvenience. This was in part because Locke did not have as dark a view of human nature, but in part because, as between persons, the moral background Locke assumed was more constraining. Moreover, the individual’s moral endowment, on Locke’s account, included a “natural executive right” to punish moral © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 151 wrongs (Locke 1952, 4–7, 17–9). Rawls excluded personal desert from the “original position,” his refurbishing of the state of nature. In each case, the NNE is not treated as a matter of empirical investigation and discovery, but rather as one of reflective adjustment to the other two components of the theory: the account of the process by which a state is constructed, and the specification of the powers and boundaries of the resulting construction. The state-of-nature tradition is committed to the view that political obligation is fundamentally voluntaristic. It rejects the classical notion that the state stands in loco parentis, and that political obligation is nonvoluntaristic in much the same way that filial obligation is nonvoluntaristic. For the state-of-nature tradition, voluntary consent is the “gold standard” by which state claims to legitimate authority are to be judged. This sets the bar rather high, and reaching it requires that the state structure be such that it can command universal, free assent, at least at some level of abstraction. That state-of-nature theories in the modern tradition are “liberal” in this sense is no accident. Part of their attraction is precisely that they underwrite not any and every kind of state, but only liberal states. For the modern tradition, Hobbes had to have gone wrong somewhere, because the state he authorized was too authoritarian. This result was unacceptable, and the problem had to lie in the way Hobbes had chosen to populate the NNE. In Locke’s state of nature, moral liberty is significantly more restricted than in Hobbes’s. Individuals are morally permitted to do “as they list” only within the confines of natural rights of non-interference held by all. These rights include a right to one’s body and to acquire ownership of things. Individuals also possess a “natural executive right” to punish wrongdoing wherever and whenever they think it right to do so. It is by the tacit or explicit transfer of individual natural executive rights to the sovereign that the passage to the civil state is made; and thus are the “inconveniences” of a state of nature overcome, and a more nearly optimal condition achieved. By funding the state’s authority with nothing beyond the accumulation of transferred, individual rights, Locke assured that the state’s legitimate authority was confined within the same outer boundary as the natural executive right. As Robert Nozick (1974, 134–42) pointed out, however, Locke’s amendments darkened as much as they brightened the prospects for peaceful (if inconvenient) existence in a state of nature. The natural executive right is the culprit; for in its exercise, mistakes and excesses are inevitable. These mistakes and excesses are themselves wrongs that invite counterpunishment. Although, in Nozick’s view, what impends is an escalating spiral of vendetta violence, what is even more likely is an under-use of natural liberty. Since all possess a right to veto, as wrongful, each other’s free doings, fewer things will be done. One person’s innocent novelty can Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 152 William A. Edmundson appear to another as a threatening and even harmful encroachment upon the just peace of a customary status quo. The conclusion Nozick drew was that there is no natural executive right, not one held individually anyway (Nozick 1974, 139).1 Undaunted, Nozick pressed on in the construction of a minimal, nightwatchman state; but the normative toolbox he used to do this did not include a natural, individual right to punish wrongdoing. Nozick suggested replacing the individuallyheld natural executive right with a jointly-held “right to have a say in the ultimate determination of punishment.” He thought this opened “another avenue” toward justifying the minimal state, because “something more can be said, given the unclarity about how rights to punish operate in a state of nature.” Unfortunately, and uncharacteristically, that “something more” involved him in an equivocation. Nozick: To the extent that it is plausible that all who have some claim to a right to punish have to act jointly, then the dominant [protective] agency will be viewed as having the greatest entitlement. (Nozick 1974, 139; emphasis in original) As he was quick to point out, there is still a gap between having the greatest entitlement and having an entitlement; but the concession overlooks the more basic problem, which is that an individual right to have a say in punishment is not an individual right to punish, or even that there be punishment. 2. Repugnant Consequences, and an Odd Result There is a revealing oddity in the way Nozick’s continuation of the state-of-nature story treats the natural executive right. Why exclude a normative element from the state of nature on the ground that its inclusion would have repugnant consequences? Unless there are repugnant consequences, there is no reason to leave the state of nature. A state-of-nature apology for the state has to walk a fine line. The pre-political condition has to be represented as sub-optimal, for otherwise there is no reason to leave it. But it must not be depicted as incoherent or normatively repugnant, for that would call the larger account itself into question. Hobbes posited a state of natural liberty in which individuals had no moral claims against one another (except perhaps those specifically covenanted to). Their moral relationships among themselves were therefore unintuitively brutish. Locke could not––as we cannot––accept Hobbes’s unconstrained depiction of natural liberty. Locke made two corrections. The first, giving individuals a richer array of natural rights with a coordinately restricted scope for natural liberty, seemed a more accurate portrait of nature, but it still left all 1 Horacio Spector pressed me to defend the interpretation of Nozick I state in the text; but my argument does not hinge on it. © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 153 at the normative mercy of wrongdoers. The second correction, giving individuals executive rights, made the normative gains meaningful; but at a cost––in terms of inconvenience––that still made exiting nature attractive. As Nozick pointed out, the second correction must itself be corrected in order to avoid depicting the state of nature as a field of endless moral combat. But notice that the accumulation of successive corrections tends to reduce the state of nature to a simulacrum of civil society, in which event it can do no (or vanishingly little) explanatory or justificatory work. It is already an embarrassment to the state-of-nature tradition in political philosophy if the pre-political normative condition can only be defined by calculating how well it suits the needs of one’s favored theory. The only way this enterprise can earn respect is by drawing upon independent sources for the account of the natural, pre-political condition. And if indeed it is nature we are talking about, shouldn’t political philosophy be eager to hear what the social and biological sciences have to say on the matter? If what is sought is the foundation of politics then, as Bernard Williams said about ethics, “the kind of material that one needs to consider in arriving at any sensible view [must be] the richest account available of human powers and social arrangements” (Williams 1995, 200–1). In this essay I draw on the findings of an array of disciplines, including evolutionary psychology, social psychology, primatology, and ethology, to build a case for a rival conception of the NNE. I argue that there is no possibility of exiting a pre-political state to occupy a civil state. That is because the structure of any plausible, hypothesized pre-political state is one that involves social authority, and social authority is––when you think about it––nothing other than political authority, in recognizable if rudimentary form. There is a continuum running from the simplest to the most complex human societies. There is no logical––much less chronological–– discontinuity that can serve as a point of entry for a contractualist, contractarian, or other state-of-nature theory that purports to explain or justify (or to discredit) political authority by leveraging a set of purportedly pre-political normative resources. To make this case, I begin by revisiting one of social science’s earliest systematic investigations of the place of authority in human life. 3. The Native Normative Endowment: Milgram’s Obedience Experiments The sociologist Stanley Milgram (2009) is celebrated for his “obedience” experiments at Yale during the early 1970s. In Milgram’s experimental design, subjects were recruited as participants in research on the learning process. They were each paired with another person and asked to draw lots to decide who would play the role of “teacher” and who the role of “learner.” In fact, the drawing was a sham, and the subject always was cast Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 154 William A. Edmundson as “teacher.” The “learner” was in fact not another subject, but an actor. “Teacher” and “learner” were put in separate rooms, but not before the “learner” stated, within the “teacher’s” hearing, that he had a heart condition, which the lab-coated experimenter was heard to dismiss as insignificant. The experimenter then had the “teacher” take a seat at an impressive-looking console fitted out with a row of switches. The “teacher” was asked to administer a word-recall test to the “learner,” and to administer shocks of increasing severity to the “learner” for each error. The console panel indicated that the rightmost switches administered shocks of dangerous intensity. The “learner” made regular errors, and emitted audible cries of distress and protest as the intensity of the sham shocks increased. “Teachers” who expressed concern about the “learner’s” suffering were simply asked by the experimenter to continue. In a variety of conditions, a substantial majority of subjects continued to administer shocks at levels that they believed were dangerous to the “learner”–– continuing to the extreme level (300 volts), by which time the “learner” had ceased to make any sounds and had, so far as the “teacher” could tell, lost consciousness or died. One surprising result was that so many would inflict what they believed to be unconsented, harmful, and even life-threatening suffering on another human being, doing so solely on the strength of assurances given by a man in a grey smock that the experiment required it. The other surprising result––more surprising in retrospect perhaps than at the time––was that Milgram and his colleagues would take such liberties with their experimental subjects. Ethical protocols in force today make it difficult to reproduce Milgram’s results exactly, but subsequent experiments in similar conditions have confirmed his central findings. Most of us as undergraduates were introduced one way or another to “the Milgram Experiment” and to Philip Zimbardo’s “Prisoner Experiment” at Stanford (Haney et al. 1973). In the Zimbardo study, undergraduate subjects were recruited and asked to simulate social conditions in a penitentiary. One group was assigned the role of guards and the other the role of prisoners. Within a matter of days, the “guards” had become so abusive of the “prisoners” that the study had to be aborted. The lesson most take from these experiments is that authority is to be distrusted, largely because we cannot trust ourselves to stand up to abuses of authority. 4. Milgram’s “Subway” Experiment Less well-known is another experiment Milgram conducted, later. Its design (1992) involved asking commuters on crowded New York subways to surrender the seats they were occupying. This was part of Milgram’s ongoing investigation of the ways of urban life. The expectation was that commuters would be unwilling to accommodate a stranger’s mere request. © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 155 It turned out that there was cooperation beyond the experimenters’ initial expectations; but what Milgram found far more surprising was the difficulty his graduate students were having in bringing themselves to make a simple request, “May I have your seat?” This reticence stood out in surprising contrast to the typical readiness of graduate students to cultivate their advisors’ favor. It also threatened to bog down the whole project, so Milgram himself took to the field to investigate. This is what he later wrote: I assumed it would be easy. I approached a seated passenger and was about to utter the magical phrase. But the words seemed lodged in my trachea and simply would not emerge. I stood there frozen, then retreated, the mission unfulfilled. My student observer urged me to try again, but I was overwhelmed by paralyzing inhibition. I argued to myself: “What kind of craven coward are you? You told your class to do it. How can you go back to them without carrying out your own assignment?” Finally, after several unsuccessful tries, I went up to a passenger and choked out the request, “Excuse me sir, may I have your seat?” A moment of anomic panic overcame me. But the man got up and gave me the seat. A second blow was yet to come. Taking the man’s seat, I was overwhelmed by the need to behave in a way that would justify my request. My head sank between my knees, and I could feel myself blanching. I was not role-playing. I actually felt as if I were going to perish [. . .]. (Milgram 1992, xxiv)2 Milgram had wanted to explore the difference between “residual rules” and “formal social norms.” Residual rules constitute the “routine grounds of everyday activity,” and are called residual because they are what is presumably left over after formal social norms are set to one side. One difference between these rather loosely defined categories has to do with explicitness. Formal rules are formulated and taught. Formal rules are thus codifiable even if not codified. Residual rules “go without saying,” and are disclosed only on those very rare occasions when they are breached. Another difference, which was of particular interest to Milgram, is the mechanism of enforcement. Those formal social norms that have been codified as laws are enforced by officials. Are residual rules then enforced by an informal “negative feedback system” administered by society at large? Milgram’s “Subway Experiment” was designed in such a way that the experimenter conspicuously violated a residual rule in order to observe the reaction of the victim and of bystanders. The residual norms were two, which were hypothesized to govern interactions on the New York subway system. One rule is that seats are allocated to the first-comer. Another is that one is not to speak to a stranger. The two rules were to be violated by the researcher, who was to ask seated strangers to give up their seats, for 2 Quoted in T. Blass. 2004, 174. At the time of these experiments the “Subway Vigilante,” Bernhard Goetz, had not yet changed forever the complexion of request-making on New York subway trains. Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 156 William A. Edmundson no reason. It was expected that the residual norms would be enforced by the refusal of the subjects to comply. Among those asked to predict the rate of compliance, the median estimate was 14% compliance. This high degree of expected noncompliance was consistent with the “negative feedback” hypothesized to constitute the enforcement mechanism of residual norms. Surprisingly, the rate of compliance was much higher: As high as 56% in the most straightforward condition, in which the experimenter simply, and without explanation, asked “May I have your seat?” Even when a patently trivial reason was added––“I can’t read my book while standing up”––37% complied outright. The “intense, inhibitory emotional reaction” felt by experimenters–– rather than anticipated noncompliance––could be recruited here as an internalized enforcement mechanism. But an internalization of what? Evidently not of a lesson learned by precept, nor, given the almost unthinkable improbability of such requests being made at all, by example. Nor could such an internalization be explained in terms of negative reinforcement for, as the experiment revealed, in fact people were surprisingly complaisant when asked. Milgram’s concluding suggestion was that some residual norms are “structural,” that is, norms whose violation is: better understood as a breach of a structure of interaction than as merely a violation of rules of equity in interaction. (Milgram 1992, 44) The structural importance of these residual rules was profounder than could be explained by appeal to roles, for the very possibility of roles seems to require that there be a deeper stratum of things “just not done”––a stratum whose content Milgram did not then try to unpack. The “intense, inhibitory emotional response” is not easily explained as something tied to a combination of two hypothesized residual norms: first-come-first-served seating and no-talking-to-strangers-on-the-subway. In a variant condition of the experiment, one experimenter conspicuously addressed another standee, an apparent stranger who was in fact a co-experimenter, with the question “Do you think it would be all right if I asked someone for a seat?” (Milgram 1992, 41). No such “intense, inhibitory emotional response” was reported to have impeded the asking of this question, which was––in the perception of others––a breach of one norm and an expression of readiness to breach another. The fact that there is no felt inhibition of enquiring about asking did not prepare Milgram and his students for the extreme difficulty of asking. One of the student experimenters proposed a further test to explore “the exact nature of this inhibitory emotion.” Milgram describes it: Harold Takoshian [a graduate student] [. . .] has suggested that the procedure be changed such that an experimenter stands before a confederate (preferably an older © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 157 woman) and bluntly ask [sic] her for her seat, which she reluctantly surrenders. She is then to stand in front of the experimenter as he makes himself comfortable in her seat. The question is whether the experimenter would feel great tension even though there is absolutely no reason for him to feel guilt. (Milgram 1992, 45 n. 4) Regrettably, this variant was never run. My hunch is that the same kind of “intense, inhibitory emotional response” would be felt (though perhaps somewhat less acutely) as in the first experimental condition. Merely to be seen to have done such a thing was mortifying, even though no one was in any ordinary sense harmed, offended, or wronged. Milgram remarked that all trace of the inhibitory emotion vanished just as soon as the experimenter left the presence of the subject and witnesses. This suggested to him that what the experimenters had been asked to do was not anything wrong––for they felt no guilt at all immediately after the fact––but something (how else to describe it?) unthinkable, or as Milgram’s mother-in-law put it, insane. Although the behavior, when described, seems perfectly innocent, the actual attempt to do it provokes a reaction akin to what social psychologist Jonathan Haidt (2001) calls “moral dumbfounding”––a strong feeling that something is clearly not to be done, though one can’t say why, as when Haidt’s experimental subjects were asked to comment on a short description of a brief, “safe” sexual interlude between an adult brother and sister. But the reaction Milgram and his students experienced wasn’t quite that. It was emotional, there were no good reasons for it, but the behavior itself wasn’t even wrong. It was . . . crazy. A different kind of constraint seems to have been at work. Compare it with another. Knauft et al. (1991) quote Marshall (1979, 363, 357): The !Kung are quite conscious of the value of meat-sharing and they talk about it, especially about the benefit of the mutual obligation it entails. The idea of sharing is deeply implanted and very successfully imposes its restraints [. . .]. The idea of eating alone is shocking to the !Kung. It makes them shriek with an uneasy laughter. Lions could do that, they say, not men. The !Kung meat-sharing norm is not a structural norm of interaction, as I interpret Milgram’s distinction. It is reasoned, consciously held, and thought of as distinguishing humans from other animals. (And, quite obviously, we regard it as conventional because we do not feel its force.) Structural norms of interaction are not reasoned, or consciously followed, and alternative norms are easily imagined but deviance is surprisingly hard. Sociologist Erving Goffman analyzed the type of predicament that Milgram and his students put themselves in. According to Goffman, in our everyday social dealings we pursue lines, which determine what sort of face we can maintain. A “line” is simply a pattern of behavior that expresses the actor’s estimation of the interaction; and “face” is the positive Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 158 William A. Edmundson value the actor implicitly claims in taking the line he believes himself to be taking. If, for example, the actor is asked by a co-worker, as they pass in a corridor, “How’s it going?” he normally would not take this as an invitation to initiate a conversation, but if asked the same thing in the break room, he would. The lines open to the actor are few, and “tend to be of a legitimate institutionalized kind” (Goffman 1967, 7). The line determines the face the actor may maintain, and if he misjudges the line, he may find himself to “be in wrong face” or “out of face.” If in the corridor the actor begins to unburden himself of a serious worry, he will run the risk that his colleague will continue on her way as though not hearing. He will be embarrassed. Or, if in the break room, the actor brushes aside his colleague’s query with the brisk “Fine. You?” he may well put her out of face. When an actor takes the wrong line, or takes a line that others misunderstand, he finds himself out of face or in wrong face and, because he “cathects his face” he is liable to feel badly. Goffman’s general description of the consequence nicely fits Milgram’s specific account: Should he sense that he is in wrong face or out of face, he is likely to feel ashamed and inferior [. . .]. Felt lack of judgmental support from the encounter may take him aback, confuse him, and momentarily incapacitate him as an interactant. His manner and bearing may falter, collapse, and crumble. He may become embarrassed and chagrined [. . .]. (Goffman 1967, 8) In short, “he is presenting no useable line” (ibid., 8–9). What Milgram had asked his students to do was not even wrong. It was not a useable line. Milgram’s subway experiment seems to reveal the existence of an unnoticed residual norm that is maintained by a remarkably and unexpectedly intense inhibitory emotion, and which he was inclined to say was part of the very structure of interaction rather than a mere rule of equity. Moreover, as Milgram argued, these attitudes are not easily explained as inductions from experience or deductions from moral principles of equity or respect. They rather disclose an unarticulated substructure governing human interaction. We may regard ourselves as free to depart from structural norms of interaction, and this attitude is consonant with our other libertarian conscious beliefs. But on those rare occasions when push comes to shove, we find ourselves bumping as it were into an invisible force field. It may be surprising that our NNE includes constraints and inhibitions we do not anticipate even when the circumstances that trigger them are known in advance. Milgram never thought, “Oh, I could never bring myself to do that!” Rather, the contrary, and he could not comprehend his students’ reticence. But what is native should be deep, especially if it is the product of a million years of evolutionary history. It makes sense to regard this experiment as corroborating several things. First, it corroborates the idea that human behavior is subject not only to © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 159 constraining moral rules but also to restrictive structures of interaction that are at the same time both deeper and shallower than what we usually think of in terms of moral rules. Second, it suggests that what we seem to be morally free to do can be made unavailable to us by inhibitions that we do not expect, and whose hold over us defies easy understanding. Third, it corroborates the idea that human interaction has a natural structure that is at least in part not merely a cultural artifact (although it is subject to cultural influence). I add to these a conjecture: Moral rules exist and operate within restrictive natural structures of interaction. Some of these natural structures, like language, are obvious, and are so basic and pervasive that the very idea of being guided by a moral rule makes little sense apart from them. Others may be less obvious, though equally basic and pervasive. One objection to looking to the results of such experiments is this: Social psychology can only tell us what people think. It has nothing to tell us about what their NNE is (if there is such a thing). Observing white Southerners’ responses to the sight of an interracial couple in the mid1960s might have told us a lot about what their attitudes were, but nothing about the moral value of those attitudes. The objection reflects impatience, I think. Important moral mistakes that have been current at various places and times are well noted: slavery, gender inequality, and so forth. But morality is a domain in which massive error seems unlikely. Milgram’s distinction between basic terms of interaction and principles of equity can help us understand how local moral error is embedded in a substrate of usually unarticulated dispositions that are constitutive features of social life. Twain’s Huckleberry Finn brilliantly explores the boundary between the two: Jim was to be treated, according to the official social morality of the time, as a mere “nigger,” but at a more rudimentary level of interaction there was no choice but to treat him as a person. Absent an artificial, frantically enforced, white policy of segregated seating on public transportation, asking Rosa Parks to give up her seat would have been “not even wrong,” but crazy. 5. Milgram’s “Queue Jumper” Experiment Do Milgram’s structures of interaction also inform the enforcement of rules of equity? Another of Milgram’s experiments focused on the behavior of people standing in (or, for New Yorkers, “on”) line. Although queuing is not a universal human practice,3 it is “a social occasion and, thus, governed by general sociopsychological rules” (Milgram 1992, 48). Milgram and his graduate students wanted to discover what those rules are. They chose to 3 Milgram himself breezily adverted to “the peculiar resistance to queue formation in Latin cultures” (59). Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 160 William A. Edmundson probe the reactions of those who witnessed others cutting ahead in line. Respecting the rights of earlier queuers was considered to be a “rudimentary principle of equity,” and Milgram wanted to test the general hypothesis that being subject to a normative constraint “implies its usual complimentary side: Those already standing in line may play a role in enforcing the norm” (ibid.). An earlier researcher had suggested that, in the absence of official personnel, such as an attentive usher, clerk, or other service person, the task of enforcing the no-cutting-ahead norm was disorganized but likelier to fall to the person immediately behind the intruder. That had not been experimentally confirmed, and Milgram was attracted to the “highly visible, linear dimension” (ibid., 49) in which reactions to wrongful intruders could be observed. Without using the terms, he was investigating what later researchers now commonly call second-party and third-party enforcement. Rather than wait around to observe queue-jumping in the wild, Milgram’s students formed teams of no fewer than three, and sought out queues of six or more persons at ticket booths and other places in the City. One or a pair was to intrude between the third and fourth person in line, saying only “Excuse me, I’d like to get in here,” and to intrude “before any responses could be made,” and to then face forward. The student intruders, unsurprisingly in this setting, reported having to overcome “the inhibitory anxiety that ordinarily prevents individuals from breaching social norms” (ibid., xxvi, 55); but their reticence was not so great as to jeopardize the project. (Being a selfish jerk was evidently not an unusable line.) Another student stood wherever she might inconspicuously observe and record. The remaining team members acted as “buffers,” one or a pair of whom was to be legitimately occupying a position in line, prior to the intrusion. So, one or two intruders were to break in line immediately in front of zero, one, or two buffers, and the reaction of the queue was recorded. Buffers were to remain expressionless. What was counted as an objection took forms ranging from physical displacement of the intruder to verbal reminders and reproof to “dirty looks, hostile stares, and gestures” (ibid., 52). The absence of an objection of any of these types was regarded (mistakenly?) as manifesting a reaction of total indifference. The vocalized objections observed were “primarily of a normative character” (ibid., 61) rather than mere abuse. To avoid creating any serious ugliness, an intruder was to yield “whenever he or she was directly challenged for improper entry” (ibid., 59). A total of 129 queues were investigated. Objections to the intrusion were most frequent (91.3%) when there was no buffer and a pair of intruders, and least frequent (5%) when there was a single intruder and a pair of buffers separating the intruder from the next person legitimately in line behind the “insertion point.” © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 161 Astonishingly, the person immediately behind a lone intruder objected in only 36.4 percent of the occasions observed. Nearly two-thirds of busy, brash New Yorkers did not even direct so much as a dirty look at a flagrant intruder to get him (or her) to go to the end of the line. It is hard to believe that this was owing to indifference. It is far more plausible that the subjects were resentful but inhibited. One might put this down to fear but, strikingly, a pair of intruders was far likelier than a single intruder to draw an objection from the person immediately behind the insertion point (86.9% v. 36.4%). Why? Milgram took this as confirmation of other research suggesting that “the cost factor [. . .] played a larger role than sheer moral indignation in stimulating objections” (ibid., 56). The increment (absolute, not relative) of inconvenience was doubled in the two-intruder condition, but then the risk that an objection would lead to confrontational unpleasantness was presumably more than doubled. It was therefore not likely to have been a shift in the perceived balance of cost that released the inhibition presumably at work in the single-intruder-no-buffer condition. I conjecture that a pair of unbuffered intruders created a context in which the non-objecting next-in-line standee would be perceived as morally weak. Persons farther back in line might readily assume that a single intruder had gotten a nod of consent from the subject; but they would be less likely to make that mistake about a pair. As Milgram and other researchers had observed, “those closest to the disruptive event are felt to have a special obligation to deal with it” (ibid., 58), and it presumably would be embarrassing to be seen as failing to do so. A mere increase of the level of inconvenience is not enough, evidently, to overcome an inhibition, as we shall see in a moment. But being mindful that one has been placed in the conventionally assigned role of enforcer is disinhibiting; and moral indignation is not merely harbored but expressly directed toward an intruding pair. The cost imposed by an intrusion is presumably about the same for each person behind the insertion point, whether or not immediately behind it.4 Milgram introduced buffers into the design “to see if responsibility for removing the intruder(s) would be assumed by someone else in line in the event that the person immediately following the intruder failed to object” (ibid., 58). He found no such effect. Even though the mode of insertion was designed to make it conspicuous that what was occurring was an intrusion rather than a resumption of a place held by someone else in the line, there is the possibility that those behind the buffers took the buffers to have 4 Considered in relative rather than absolute terms, the burden will not be equal, as Bas van der Vossen has pointed out to me. If I am next in line and someone breaks in ahead of me, my expected wait will be doubled. But if ten are in line ahead of me, my wait will have only been increased by 10%. Of course what matters is not an absolute or a relative figure but the perceived difference. My reaction may be idiosyncratic, but I would feel more put upon in the latter than the former case, especially if the intruder were breaking in at the front of the line. Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 162 William A. Edmundson discretely or implicitly condoned the intrusion. But Milgram reported no efforts by subjects to disambiguate the situation. He concluded that “there was no displacement of the defensive response if it did not occur at the point of intrusion [. . .] a willingness to object to intrusions attenuates quickly with positions further down the line” (ibid., 58). Far from behaving as though they possessed a natural executive right to correct conspicuous wrongdoers, Milgram’s buffered subjects acted as if––and thus seemed to have believed that––it was none of their business, despite the fact that the wrongdoing directly burdened them to a degree at least equal to that suffered by the person immediately behind the insertion point. Those ahead in line suffered no inconvenience, and so they were potential third-party enforcers. People ahead of the intruder objected significantly less often, overall (8% v. 18.2%) than those behind, although it is unclear whether this was due to a failure to note the intrusion, or a failure to perceive that it was not condoned by those behind the insertion point, or to an inhibition, or to a failure to care. The rate of objection by those immediately ahead of the insertion point correlated closely with the rate of objection by the first subject behind, and was approximately an equal rate except in the two intruder/no buffer condition (21.7% v. 86.9%), the only condition in which the person immediately behind the insertion point was likelier than not to object at all. Persons ahead of the insertion point were the source of a quarter of all objections, but it is plausible that theirs were effectively seconds to an objection from someone behind the intruder. In only one instance (that, in the one intruder/two buffers condition) was a person ahead of the insertion point the sole objector. Milgram was interested in queues as “a classic illustration of how individuals create social order, on the basis of a rudimentary principle of equity” (Milgram 1992, 48). We humans have been fitted by evolution with highly sensitive “cheater detection” capacities, but we also seem to have evolved a set of inhibitions that restrain us from acting out our “moral anger” unless we perceive ourselves as occupying a conventionally assigned enforcement role. If people hold a Lockean natural executive right, one would expect witnesses of conspicuous wrongdoing to feel no inhibition about reproving the wrongdoer. They might think it unwise, or risky, or not worth the effort. But surely any risks of a physical altercation are greater for one standing next to an intruder, especially if there is a pair of them. Yet those subjects standing directly behind a pair of intruders were the likeliest to object, and the only ones likelier to object than not. Significantly, Milgram noted no reproof of those immediately behind the insertion point or elsewhere who elected not to object to the intrusion. In other experimental studies, people and other primates have been found to be willing to incur serious costs to retaliate against those whom they feel to be cheaters in one-to-one or other small-scale transactions (Boyd et al. 2005; Cummins 1996). Why such reticence about queue jumpers? As Gerald © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 163 Gaus (2011, 190) has put the point, “When moral emotions are aroused, apathy about whether their targets continue in their behavior is deeply perplexing.” But apathy and inhibition are different. What the evidence suggests is that a general moral permission to confront and reprove wrongdoers is not included in what people really take to be their everyday armamentarium of rights. As to a wide range of everyday misdeeds, what most people seem to feel they individually possess is not a natural executive right, but the Hohfeldian “jural opposite”: a natural executive no-right. The events that pass before us are almost always perfused with moral significance, but the ways in which we respond to them are circumscribed. Because of the breadth of scope of moral rules, the scope of individual response has had to be narrowed. As Goffman (1967, 45) expressed what I take to be approximately the same point, “The general capacity to be bound by moral rules [. . .] belong[s] to the individual, but the particular set of rules which transforms him into a human being derives from requirements established in the ritual organization of social encounters.” To insist, in the wrong way, upon another’s doing the right thing, runs counter to something that is part of what makes us human. In our everyday encounters, as we come closer to being readable as expressing moral disapproval––and not merely disclosing a taste or a preference––the scope for casual comment narrows. An onlooker’s unasked-for opinion about whether, for example, a tennis serve was in or out is tolerable even if not well considered. But if she wants to offer her opinion as to which of two drivers is entitled to a contested parking spot, she had better be careful. 6. Isn’t It All Relative? The admitted non-universality of queueing does not contradict the main point, which is that permissions to punitively answer conspicuous wrongdoing are not generally held by all. Queuing is an instance of a more general type of human practice, sharing. All human cultures have rules that determine which are the resources that members are to share, which are not, and how sharing is to proceed. Hunter/gatherers share “big meat,” but not little berries. The focus of this essay is the NNE and the existence and nature of Milgram’s structures of interaction, not the precise rules of equity in dealing. That the latter are variable from time to time and place to place is not to the present point. The point of the objection goes deeper, however. Distinguishing structures of interaction and rules of equity does not address the possibility that both are very recent products of civilization, not ancient adaptations. The phenomena of reticence and restraint that Milgram and Goffman describe are consistent with the psychology that Norbert Elias describes. Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 164 William A. Edmundson Precisely this is characteristic of psychological changes in the course of civilization: the more complex and stable control of conduct is increasingly instilled in the individual from his or her earliest years as an automatism, a self-compulsion that he or she cannot resist even if he or she consciously wishes to. (Elias 2000, 367) This “blindly functioning apparatus of self-control” functions by putting deviant behavior on the far side of “a wall of deep-rooted fears,” and becoming “as it were, ‘second nature’ ” (ibid., 368, 369). But the process that Elias describes as having produced this formation goes back only to mediaeval feudalism. What then could justify the claim that an explanation has to invoke prehistorical evolutionary conditions? This is a serious objection. The roots of the inhibiting constraints at work are not an artifact of Western culture, because they can be traced much farther back, even to the societies of our primate ancestors and cousins, as I will try to show. A more concrete but equally serious objection is that there are places where queue-jumping regularly does provoke a reaction among all and sundry. Israel, I am told, is like this. In Seattle and Berlin, I am also told, a spirit of civic seriousness flourishes to such a degree that jaywalkers are liable to be admonished by passers-by.5 In these instances, citizens exercise what William Godwin (1793, 198) called a “republican boldness” in judging and reproving others. The main point is unaffected: Permissions to criticize conspicuous wrongdoing cannot be presumed to be held by all. The composition of the normative endowment of the typical individual in various places and times may of course be more or less extensive than what should be posited as part of the NNE. As Larry Arnhart (2004) has pointed out—by way of deflecting Freud’s claim that cultural variation refutes any biological basis of the incest taboo—what is at issue is the existence of a predisposition that is subject both to cultural variation in its detailed implementation, and even to cultural override in extreme circumstances. It is a matter, as Aristotle said, of what holds “for the most part” (cf., e.g., An. Post. I. 87b19–26; Metaph. VI. 1026b27-1027a28; NE I. 1098a25-1098b8). Given the traumatic background of the founding of the state of Israel, and its precarious situation, it is unsurprising that a vigilant norm might emerge, appointing each a moral censor, having responsibility for overseeing all. 7. Fehr and Fischbacher’s Study People who reprove or otherwise punish those who wrong them are second-party enforcers. But a wide range of social norms depends upon enforcement by those who are not personally wronged and have nothing 5 On successive four-day visits to Berlin and to Frankfurt in August, 2011, I witnessed a fair amount of crossing against the light, but no enforcement, informal or otherwise. I am told that Munich is less tolerant. © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 165 to gain by acting as enforcer (Bendor and Swistak 2001; Sripada and Stich 2006). They depend on third-party enforcement. Third-party enforcement is the center of a pair of related puzzles. Why would anyone bother to enforce a norm if not directly wronged or threatened by a breach? And if a norm is serious enough to count as a moral norm, how can a failure to enforce it, in a proper case, not itself count as a wrong act, especially where second-party enforcement is unavailing? I am mentally prepared to be upbraided for jaywalking. But I would be flabbergasted were anyone to upbraid me for failing to upbraid another person for jaywalking. Such a thing is unheard of even in places renowned for their civic seriousness. Experimental economists Ernst Fehr and Urs Fischbacher recently conducted some of the first laboratory studies of the specific phenomenon of third-party enforcement. They remark that “if only second parties imposed sanctions, a very limited number of social norms could be enforced because norm violations often do not directly hurt other people” (Fehr and Fischbacher 2004b, 264). What they have in mind is shirking and other kinds of wrongdoing that does not involve an identifiable victim. Cooperative norms are typically of this kind. Cooperative norms address coordination problems, that is, problems that are too big for individuals to solve alone. Many such problems are not so big that every last member of the group has to contribute to the solution. For example, a successful hunt for large game––what anthropologists call “big meat”––will sometimes require those with lesser skills to beat the bushes to flush game in the direction of the spearmen. Quite often one or two “beaters” more or less will not matter. But the spearmen in hunter/gatherer societies customarily share the kill with all. Those who eat without having put in the effort expected of them are free-riders. Unless the group instills a cooperative norm, the temptation to free-ride on the efforts of others will be unchecked, and that in turn will eventually leave the group and its members worse off. But, from an evolutionary perspective, how can such norms have taken hold? Cooperation is “altruistic” in the sense that the cooperator often incurs a cost while garnishing no additional benefit. Natural selection operating solely at the individual level would seem to favor the shirkers and free-riders. But, as numerous studies have demonstrated, people do observe fairness norms more often than one would expect if they were entirely selfish: In that sense, many of us do behave altruistically. Even very small children recognize unfair behavior and strongly disapprove of it. Fairness looks to be both innate, and impossible to explain as an adaptation, at least at the level of individual selection. How reliably do people act fairly where free-riding isn’t punished? That is, where they need not worry about third-party punishment? Observing how subjects play the Public Goods Game is one way to find out. In a standardly conducted Public Goods Game, groups of four isolated, Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 166 William A. Edmundson non-interacting subjects are given equal initial endowments of 20 units, equivalent to cash (Camerer and Fehr 2004). Participants do not communicate and are isolated from each other (cf. Ostrom, Walker, and Gardner 1992). Each has the option of keeping the entire endowment or contributing it or a portion of it to a common pool. Each earns a dividend of 40% on the assets of the common pool, and earns this irrespective of contribution. Cooperating is the socially optimal strategy, since each point contributed yields a 40% return. But contributing nothing is the individually optimal strategy, at least in a one-shot trial, since a free-rider can keep her entire initial stake and reap a 40% return on the investments of others. If all four contribute their entire stakes, each nets eight units at each round. If no one contributes, the net return will be zero. The Public Goods Game is especially interesting insofar as it could model a wide range of phenomena, from hunting mastodons to rooting out corruption. Fehr and Gächter (2000) found that punishment tended to induce a convergence toward full cooperation in a ten-round Public Goods Game, while in the no-punishment condition defection became the norm. Gächter, Renner, and Sefton (2008) found that punishment promotes average net earnings more robustly in fifty-round than in ten-round iterations of the standard Public Goods Game. Nonetheless, defections were widespread in the final rounds in these and other studies. Even though many players were willing to play “pro-socially” in the opening rounds, and persisted in doing so for a few rounds, in the face of defections, in the end, individuals were willing to cooperate only on the condition that cooperation was an enforced norm. Evolution is not confined to one level of selection, fortunately for us. An innate sense of cooperativeness within a group will enhance the chances of its members’ genes being passed along. Groups of people that cooperate, treat each other fairly, and don’t shirk, are more cohesive, and make better hunters and warriors. But group selection is an eligible explanation of the prevalence of cooperative norms only if it supposes there is robust thirdparty enforcement. Because third-party enforcement is costly to the enforcer, it too is altruistic in the evolutionary-biological sense, and altruistic sacrifices to benefit the group are generally detrimental to individual fitness. Given familiar game-theoretic assumptions, the altruists, including the third-party enforcers, in an isolated group should become extinct. Sober and Wilson (1999, 55–71) explain how an altruistic trait or norm might evolve without third-party enforcement, but they acknowledge that prevalent third-party enforcement can make altruism a paying proposition in terms of individual fitness. The better explanation of the evolution of human cooperation appeals to both individual- and group-level selection. Fehr and Fischbacher’s studies are an unprecedented attempt to compare patterns of third-party to second-party enforcement. One study is a variation of the Dictator Game, in which player A unilaterally divides a © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 167 windfall sum between himself and one other player, B. A third subject, C, was included, and asked to specify “deduction points” to be withdrawn from the holdings of dictators in various outcomes of the Dictator Game. Another study gave third-party subjects a similar opportunity to deduct from the holdings of participants in a Public Goods Game. In both studies the deductions were modestly costly to these third-party “enforcers,” but were leveraged by a 1 to 3 ratio: One unit of expenditure by an enforcer would inflict three units of loss upon a wrongdoer. Monetary loss is an attractive proxy: It is quantifiable, ethically usable, and in the middle of the range the anthropologist Hoebel described, running from “the curled lip, the raised eyebrow, the word of scorn or ridicule, the rap on the knuckle and refusal to invite back to dinner, through economic deprivation, physical hurt, prolonged social ostracism, through imprisonment or exile to the ultimate in social ostracism––execution” (Hoebel 1954, 15). It is significant that at the outset the subjects were told of the form of the game and their respective roles in it. One hypothesis was that dictators in the Dictator Game would be punished in proportion to their selfish deviation from a fifty-fifty split: the presumed “distribution norm.” Another hypothesis was that a lone defector in the Public Goods Game would be punished for violating the presumed “cooperation norm.” Fehr and Fischbacher’s hypotheses were confirmed, to an extent. Roughly 60% of the potential third-party enforcers sanctioned a greedy dictator in the Dictator Game. Just under half (45.8%) of the third-party enforcers deducted from the gains of unilateral defectors in the Public Goods Game, and a fifth (20%) sanctioned pairs of defectors––a surprising result insofar as these defectors were to get only reduced payoffs anyway. In summary, Fehr and Fischbacher found that a large percentage of subjects are willing to enforce distribution and cooperation norms even though they incur costs and reap no economic benefit from their sanctions and even though they have not been directly harmed by the norm violation [. . .]. We also found that sanctions by second parties directly harmed were much stronger than third-party sanctions, indeed strong enough to make norm violations unprofitable, whereas the sanctions of a single third party were not. Thus, in the context of our experiment, more than one third party is needed to enforce the norm. However, this condition is probably met frequently in real life. Therefore, taken together, our results suggest that altruistic third-party sanctions are likely to be powerful enforcers of social norms. (Fehr and Fischbacher 2004b, 85) Given Fehr and Fischbacher’s results, at first it may look as though the Lockean hypothesis, that there is a natural executive right, is borne out rather than my own. After all, almost two-thirds of the subjects decided that they would impose a sanction of some sort, and incur a cost in doing so. Moreover, in a subsequent study a similar readiness was found in Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 168 William A. Edmundson fifteen different, geographically and culturally diverse populations (Henrich et al. 2006). Why would all these people act this way unless they believed they enjoyed something like a natural executive right to punish wrongdoing? Notice, though, that these subjects were primed to enforce by having been authorized by the experimenters to do precisely that (cf. Fehr and Fischbacher 2004a). Without such priming and authorization, there may have been much less punishing in evidence. Unfortunately, it is not obvious how to design experiments like these without tipping off the would-be third-party enforcer to the fact that enforcement was among her permissible options. But putting that option in the hands of third-parties is to obliterate the possibility of discovering whether the third-party subjects believe they come to the table, as it were, already possessing such permission. When the men and women in the lab coats start assigning roles, watch out. That was one of the lessons of the Milgram “Obedience” experiments. What may be the more remarkable result is that over one-third of the third-party players were unwilling to expend even the minimum amount (CHF 0.3 = $0.24) to sanction a breach of the presumed norm, even after having been given $8 just for showing up, and having been primed to understand that their only active role in the proceeding was to “assign deduction points.” Fehr and Fischbacher’s results indicate that enforcement by lone-ranger third parties fails to make deviance unprofitable: “In the context of our experiment, more than one third party is needed to enforce the norm.” They suggest that this is not going to be a problem for group-selection theory because “this condition is probably met frequently in real life.” So, the conclusion that “altruistic third-party sanctions are likely to be powerful enforcers of social norms” assumes, at the least, that third-party permissions to enforce are distributed in a way sufficient to keep the norm viable. But, as I have argued, that distribution cannot be Lockean. Not only is a wide diffusion of responsibility a notoriously iffy way of ensuring that someone takes responsibility (Darley and Latané 1968; Hudson and Bruckman 2004), but the forces Nozick pointed out constantly threaten to spur an escalating spiral of retribution and counterretribution (Denant-Boemont et al. 2007; Nikiforakis 2008). What is needed is what was present from the outset in all of the experiments discussed: An authority, i.e., some certain one or some certain body of persons morally permitted, and expected, to enforce and to deputize enforcers of cooperative social norms. 8. Hermann’s Findings on “Antisocial” Punishment Benedikt Herrmann and colleagues have reported findings that indicate other shortcomings of diffuse, lone-ranger enforcement (Herrmann, Thöni, © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 169 and Gächter 2008). These researchers ran iterations (ten rounds) of the Public Goods Game. To get a cross-cultural perspective, the researchers recruited 1120 university students from sixteen different cities (“pools”), from Zürich (site of the Fehr and Fischbacher study) to Chengdu (China, home of Foxconn, assembler of the iPad), from Muscat (the capital of Oman, on the Arabian Peninsula, between Yemen and the United Arab Emirates) to Minsk (the capital of Belarus, a former nuclear power), and twelve other places. The trials were run under two different conditions: one with and one without punishment. In the punishment condition, after each round, each of the four subjects could, anonymously and simultaneously, assign up to ten “deduction points” among his partners. As in the Fehr and Fischbacher study, for each deduction point expended, the player on the receiving end lost three points. (In the non-punishment condition, each player kept all his earnings at each round.) The researchers considered punishment of one who contributed less, by the hand of one who contributed more, to be prosocial. In contrast, instances of punishment of one who contributed more by one who contributed less counted as antisocial. The terminology is defensible. Those who punish lesser contributors are engaged in a personally costly effort to achieve a better overall result. Numerous studies support the idea that the establishment and stability of cooperative norms (beneficial or otherwise) depends upon a readiness to engage in costly punishment of noncooperators (e.g., Boyd and Richerson 1992). On the other hand, costly punishment of those who contribute more is a puzzle. Previous studies (e.g., Henrich et al. 2006) had encountered a number of non-confused subjects who rejected “super-fair” proposals in runs of the Ultimatum Game (“Here, you take sixty and I’ll take forty . . . . No deal!?”). The Herrmann study was designed in part to find out more about this puzzling phenomenon. Herrmann and colleagues did indeed find antisocial punishment at work, but only in certain places. These are some of the explanations they considered: “Revenge is a likely explanation for antisocial punishment in most participant pools” (Herrmann, Thöni, and Gächter 2008, 1364). A player who has been on the receiving end of prosocial punishment should have a pretty good idea which of his fellow-players had judged that he hadn’t been pulling his weight, but not all the antisocial punishers were exacting revenge. Low contributors might also view high contributors as do-gooders who have shown them up. Punishment may therefore be an act of “do-gooder derogation” [. . .]. Similarly [. . .] people for various reasons might be suspicious of others who appear too generous. Normative conformity, a desire and expectation to behave as all others do, is part of human psychology [. . .] and may lead to the punishment of all deviators, cooperators, and free-riders alike. (Herrmann, Thöni, and Gächter 2008, 1364) Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 170 William A. Edmundson Anybody seen as being aggressively “goody-goody” might incur the kind of “antiauthoritarian sanctioning” that Boehm (1999, 52) documents as prevalent across hunter-gatherer cultures. Just as a proficient hunter will be scrutinized for signs of becoming an “upstart” aspiring to domination (1999, 43–7), so also anyone who too readily and conspicuously makes outsized donations to a common pool might come under suspicion and invite “moralistic aggression” (Trivers 1971) in return. There was also a noticeable amount of punishment that was neither prosocial nor antisocial: that is, instances of punishers punishing contributors of an amount equal to their own! The authors of the study put this down to the 3-to-1 hurt-to-cost ratio intersecting “a strong taste for dominance, a competitive personality, or a desire to maximize relative payoffs” (Herrmann, Thöni, and Gächter 2008, 1364). None of these impulses or proclivities can affect the game unless the players are given a stack of “deduction points” to play and, inevitably, a cue that they are authorized to use them. In the non-punishment condition, cooperation deteriorated over the course of the ten rounds “almost everywhere” (Herrmann, Thöni, and Gächter 2008, 1365). In terms of nearing or staying near the socially optimal level of cooperation, then, what came of handing each player in the punishment condition what was, in effect, an “executive right” to punish? The presence of a punishment opportunity had dramatic consequences on the achieved cooperation levels [. . .]. Cooperation was stabilized but at vastly different levels [. . .]. The higher antisocial punishment is in a participant pool, the lower is the average cooperation level in that participant pool. (Ibid., 1364) Wasn’t the introduction of the punishment option a prosocial innovation, nonetheless, overall? The Lockean assumption is: Yes, it is. But Herrmann and colleagues are more cautious: [T]he presence of a punishment option had at least a weakly significant cooperation-enhancing effect in 11 participant pools [but] the change in cooperation [. . .] was not significant in the other five [. . .]. Thus, the cooperation-enhancing effect of a punishment opportunity cannot be taken for granted. This finding stands in contrast to previous results [. . .] where punishment always increased cooperation. (Ibid., 1365) What accounts for this anomaly? Darwinian accounts of cooperation depend upon the efficacy of costly punishment. Hermann and colleagues have an answer: The reason for this result is related to antisocial punishment: The higher antisocial punishment was in a participant pool, the lower was the rate of increase in cooperation in the P[unishment] experiment relative to the N[o-punishment] experiment. (Ibid.) So the anomaly is explicable as the effect of antisocial punishment. But then the question becomes: © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 171 What explains the large differences in antisocial punishment and hence cooperation levels? Punishment may be related to social norms of cooperation [. . .] strong norms of civic cooperation might act as a constraint on antisocial punishment. (Ibid., emphasis added) Would it help if diffuse, decentralized, Lockean punishment of anti-social punishers were possible as a further response? The problem would simply be displaced. (I will say “displaced” rather than speak of “the problem of anti-social second-order third-party punishment.”) Cinyabuguma, Page, and Putterman (2006) added a second-order round of punishing. They found that those who punish free-riders are subject to “perverse” second-order punishing, as if to mirror the perverse punishing of high-contributors in the first round. They also found that adding the second round did not significantly increase cooperation and welfare. What should also be borne in mind is that the line between moralistic aggression and pretensions to dominance is a fine one. One man’s moralistic aggressor is another’s aggressor, simpliciter. “Norms of civic cooperation” are needed to assure that third-party punishment is patently distinguishable from private aggression. The hypothesis I am defending is that punishment––as distinct from mere aggression and from mere retaliation––is embedded in certain constraining structures of interaction in all human societies. These structures constrain not only antisocial but also strategic and prosocial punishment. Although “prosocial” and “antisocial” have precise definitions here, the underlying concepts are sharply contested: Selectively constraining antisocial punishment only is not a likely option. Without such constraining structures, beneficial social cooperation on any but the briefest, most primitive scale is hard to envisage. The design of this and other studies reveals this structure only indirectly, insofar as each of them implicitly deputizes each participant to engage in what the structure of interaction would otherwise inhibit. The Herrmann study backhandedly acknowledges much of this when it concludes that the “detrimental effects of antisocial punishment on cooperation (and efficiency) [. . .] provide a further rationale why modern societies shun revenge and centralize punishment in the hands of the state” (Herrmann, Thöni, and Gächter 2008, 1367). But these detrimental effects are unlikely to have been borne patiently until modern societies put on the clamps. Rather, restrictions on moralistic aggression are far likelier to have had a long evolutionary history. This species of interactional structure is an adaptation that is part of our human nature. It is, in short, what makes us the political animals we are. 9. Punishment: Direct, or Repertorial? None of the conditions in the studies that have been carried out to date offer the subjects the option of engaging in what could be called Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 172 William A. Edmundson “repertorial” rather than “direct” punishment (cf. Darley and Latané 1968). Repertorial punishment can be defined as any conduct by a subject that is intended to inform someone else, perceived as having a greater authority, of anti-social conduct that the subject could be excused for directly punishing on her own. This kind of conduct is known by a variety of pejoratives: “snitching,” “tattling,” “ratting on,” “squealing,” and so on. It is a species of gossip, with the difference that it is addressed to someone clearly authorized (and likely) to punish a wrongdoer directly. Direct punishment, as I define it, is any action intended by the actor to make a wrongdoer conscious that his conduct has been judged wrongful. This is a minimum, which may take any number of symbolic forms (Feinberg 1965) that normally will involve imposing an additional unwanted cost on the wrongdoer. The costs and risks associated with direct punishment normally are greater than those that go with repertorial punishment, but in an experimental setting they could be equalized. One obvious, testable hypothesis is that subjects are more willing to engage in repertorial than direct punishment if given the option by an experimenter who is clearly empowered to administer direct punishment. Gossip, the genus to which repertorial punishment belongs, is vital to social life; even though the popular image of gossip is a negative one. Gossip is disdained as a malicious, vulgar, and chiefly feminine pastime. In fact, gossip is essential to the maintenance and refinement of social and moral norms (Merry 1984). That gossip should have such a role is not surprising if standing to administer direct punishment is constrained in the way I have been suggesting. Wrongdoers can be expected to try to avoid any observation likely to incur direct punishment, so it should not be a surprise if standing to “administer” repertorial punishment is more widely diffused and by that measure more nearly Lockean. Each of us has a “natural repertorial right” to discuss moral wrongdoing, although there are restrictions here, too. Only those closest in kinship or friendship may inform the one wronged by adultery of what everyone else is talking about. In some cultures, gossip itself is believed to have an occult power to visit harm upon its subject (Shweder et al. 1998, 129). Via gossip, knowledge of wrongdoing can make its way to those who do have clear standing to administer direct punishment. In tribal societies, a council of elders might fill this role. These social/political feedback loops have probably been at work as long as humans have had the power of speech. Members of our nearest living relative species, the chimpanzees, groom one another to bond, to reconcile, and to pass the time. It has been conjectured that speech arose to fill the grooming role for us hairless primates (Dunbar 1996). “Nit-picking” is what we do as we form and reinforce social bonds. We find common ground in ceaseless dialogue about what absent others have done, are doing, and are likely to do. Gossip not only transmits information to those permitted to act on it: Its circula© 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 173 tion is an important means of consensus building. Hunter/gatherer bands typically have plenty of time to socialize, and when conditions are favorable to gathering into “super-bands,” there would be an extended opportunity for cross-band exchanges of gossip as well as for the ceremonial rites that are more often thought of when the subject is culture (Cashdan 1989). Chimpanzee politics began as and remains a business of male coalitionbuilding (de Waal 2007). For us humans, it might not be extravagant to suggest that our politics began when gossip made it possible for the weaker members of groups to contribute, repertorially, to the enforcement of moral norms. 10. Sabini and Silver’s Reflections on the Milgram Obedience Study Two of Milgram’s students, John Sabini and Maury Silver, were intrigued by the inhibition of impulses to deliver justified moral reproach. They offered an explanation of “why the moral nature of reproach inhibits” (Sabini and Silver 1982, 37). They had observed that the moral tincture of reproach is part of the reason people often feel inhibited when witnessing wrongdoing. Moreover, this inhibition can operate to prolong a person’s cooperation in what she believes to be seriously wrongful. Sabini and Silver diagnosed the results of Milgram’s “Obedience” study this way: Perhaps one difficulty that subjects had in seeking to withdraw is a consequence of the face-to-face relationship between the subject and the experimenter: Invoking a moral norm important enough to justify disobedience implies that the experimenter ought not to order them to continue. Thus, any justification they might have offered for refusing to continue would have invoked an explicit or implicit condemnation of authority. (Sabini and Silver 1982, 38, emphasis original) In their view, it is precisely the moral wrongness of continuing that made it difficult to refuse. Several of the subjects even apologized for the seeming rudeness of their raising questions about the propriety of “going on.” We treat the subject’s explanation seriously: The making of a moral reproach is rude [and] this rudeness is a source of inhibition against reproaching another. Yet this explanation is unsatisfying, not because it is wrong, but because it is absurd to be concerned with courtesy when another’s life is at stake. (Ibid., 39) Sabini and Silver resist the natural inference that the inhibition is not merely one of courtesy. It is moral. They do correctly characterize the inhibition as one of standing, akin to legal doctrines that circumscribe the category of persons who will be allowed to bring suit regarding a matter to which they are not properly related. Their resistance is correct, though, if I am right that the inhibition is an aspect of the structure of interaction, rather than a moral rule in the more ordinary sense. Ambiguity remains. Were the subjects inhibited by the knowledge that their non-acquiescence would be perceived as moral reproof, or were they Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 174 William A. Edmundson rather inhibited by their perception of the experimenter as an authority figure? Or both, to varying degrees? Milgram in fact had run a condition involving two “teachers,” one of whom was an actor, the other the subject. The actor-“teacher” was at first assigned the role of simply recording data, but the experimenter absented himself on a pretext and left the two “teachers” to proceed in his place, but without any specification of the shock levels. The actor-“teacher” took it upon himself to propose that the shocks be increased with each wrong answer from the “learner,” and instructed the subject-“teacher”––whose prior role had been to administer the shocks––to proceed accordingly. There was “a sharp drop in compliance [. . .] only a third as many subjects followed the [actor-“teacher”] as follow the experimenter” (ibid., 97). In a further variation, the actor“teacher” brushed aside the objecting subject and administered escalating shocks personally. In this condition, several of the subjects stood up to defend the helpless “learner.” They felt free to threaten the [actor-“teacher”] and were not reluctant to criticize his judgment or personally chastise him; their attitude contrasts sharply with the deferential politeness subjects invariably displayed in other [conditions], when an authority was at the helm. (Ibid., 98) From this, one might infer that in this, as in all earlier experimental conditions, the status of the experimenter as authority was the crucial inhibitor, rather than the moral rebuke implicit in non-cooperation. But Milgram’s general remarks were ambivalent: Although the aim of the [variant conditions] was to strip the commands of any authoritative source, it was almost impossible to do this in a completely effective manner. There were many traces of derived authority even when the experimenter was absent. The over-all situation had been defined by the authority, as well as the idea of administering shocks [. . .]. Authority was hovering in the background and had created the basic situation in which the participants found themselves. (Ibid., 96–7) One of these “traces of derived authority” (though perhaps not one that Milgram had foremost in mind) stemmed from the fact that the experimenter had not expressly or even implicitly authorized doing real harm to the “learner,” or proceeding over the “learner’s” objections. As Milgram noted, “In refusing to go along with the [actor-‘teacher’], most subjects assume [sic] that they are doing what the experimenter would have wanted them to do” (ibid., 98). The resisting subjects, in other words, felt specifically authorized, in the circumstances, to act in ways that might imply moral rebuke. This is perfectly consistent with, and corroborative of, Sabini and Silver’s thesis that, unless authorized, we humans are inhibited in making communications that are interpretable as expressing or implying moral reproach. © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 175 11. An Evolutionary “Just-so” Story: How the Hominin Got Its Politics Any group that is effectively governed by moral rules, correct or otherwise, is also one that is a political society, in the sense that the enforcement of those rules is implicitly recognized as a matter over which the group has authority in the first and last instance. In short, any society that is moral is also political. The present argument for this claim is that interactions so structured are likely to have been adaptive for groups, going back to the long-ago period when the hominin line diverged from the great apes, and possibly much earlier. Morality itself is an adaptation, and for morality to work, its enforcement had to be regulated, that is, morality was adaptive only where it was politically constrained, in the sense I have defined. The constraint, internalized as an inhibition, is highly unlikely to have been a late-coming adjunct or adjustment to already entrenched, prior dispositions to be angered by wrongdoing. Morality is not a merely cultural phenomenon that has been handed down to us, so to speak, in the way that other kinds of cultural artifact are handed down, such as myths, rituals, and creeds. Of course, the specifics of moral doctrine are transmitted culturally. This can easily be inferred from the fact of cultural variation. Every human society regulates violence, sexuality, and pair-bonding, just as every human society possesses a language. A language is in essence a set of rules that determine what are and what are not meaningful utterances, and what those meanings are. Humans possess an innate capacity to learn language. Human linguistic capacity is an adaptation. The capacity is not transmitted culturally, though the particular vocabulary and grammar of any language has to be learned. Language learning is possible only because the capacity to learn does not have to be taught. How could it be? Similarly, humans have a moral capacity. This capacity is no more an item of cultural transmission than a linguistic capacity is. Of course, the details of a group’s moral code, like a particular vocabulary and grammar, have to be taught. But both language and morality have a “deep grammar” that is invariant; and the capacity to speak and to engage socially are evolved human capacities that have a biological basis. Infants do not need to learn to respond positively to their mothers’ smiles any more than they need to be taught to babble (Mikhail 2011; Hauser 2006). The innate deep grammar of human morality disposes us to react and interact in certain ways and not in others. Certain expressions and gestures have a universal significance, a moral meaning. Certain configurations and relationships universally have moral significance, and are pregnant with moral meaning. These are evolved traits that, in social circumstances, express themselves in each of us, to some degree or other, because they are literally “part of our DNA.” The vast variety of human expression and association that we experience is a matter of changes played on a fundamental Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 176 William A. Edmundson scale that, though itself also changeable, is so much a part of human nature that it would be a matter of enormous difficulty and hazard to alter it deliberately––though, over time, alter it will. Humans could, conceivably, be bred to be unresponsive to each other’s facial expressions. Some human individuals, as a matter of fact, are indifferent. But that does not mean that a readiness to respond to such expressions is not part of human nature. An individual’s NNE reflects the basic grammar of moral interaction. Part of that consists of attunement to aspects of the exogamy configuration, such as the kinship and affinity status of an object of sexual interest. Part of it consists of a readiness to cooperate with others on fair terms. Part of it consists of a general readiness to conform to the norms of one’s group. Part of it consists of a readiness to disapprove of those who ignore norms of the group. If there is an NNE at all, whatever its exact content, it will reflect a restriction of the use of moralistic aggression to enforce group norms, just as it will restrict any other type of aggression against others of the group. This presupposes, and does not in the least tend to deny, that “moral anger” is an appropriate emotion. It is. As Aristotle put it, “The man who is angry at the right things and with the right people [. . .] as he ought, when he ought, and as long as he ought, is praised” (Aristotle 1994, IV.5 1125b–1126a). Our present concern is the “as he ought, when he ought” part. We are by nature “rule-following punishers,” and an aversion toward antisocial conduct has been observed in infants (Hamlin et al. 2007). Having an aversion and being prepared to act on it are two different things, however. The two can be contrasted in several ways. Simply being in a state of conscious aversion to someone else’s wrongful conduct is not taxed with any obvious cost, or none over and above the psychological cost attached to being averse to anything else. Of course, if one is victimized by another’s wrongful conduct, failing to act on one’s aversion to being victimized creates the risk of suffering the cost of being victimized again. But not all wrongs are wrongs that have a discrete and identifiable victim. In such cases, any measure taken to avenge, suppress, or discourage that wrong will be costly to the individual, though beneficial to the group. The individual propensity to administer costly punishment is a variable trait, and the variation is to a high degree heritable (Wallace et al. 2007). Negative moral judgments involve the emotions (Prinz 2007); but the administering of costly punishment also correlates with activity in the right lateral prefrontal cortex, suggesting that it is not simply a visceral response, but one mediated by higher cognitive functions (Knoch et al. 2010). Punishing is not without its pleasures (de Quervain et al. 2004; Singer et al. 2006), as well as its advantages: A punisher may gain a reputation as a formidable person, and so gain in esteem. But calculating whether the cost of costly punishing is worth expending is, itself, costly. These considerations, in themselves, suggest two things. © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 177 One is that it would be individually beneficial to be relieved of the cognitive burden of having to calculate, case by case, whether acting upon one’s aversion to wrongdoing unrelated to one’s direct interests is worth the cost. Suppose, for example, that an exasperated parent disciplines her refractory child in a way that is wrongful, but short of threatening major physical or emotional harm. You witness this in a public place, a supermarket aisle for example. You have a healthy aversion to the conduct. But what do you do? If you are like me, you are uncomfortable even thinking about it. It would be easy if one were able to say to oneself, “It’s none of my business.” And it is easy, because (extreme cases aside) it isn’t one’s business. A parent’s (limited) prerogative to discipline a child assigns the rest of us a manageable role. It may be for the parent’s own parent, or spouse, to say something, but not for chance bystanders (Schoeman 1992). Another thing is that the significant variability in the inherited propensity to engage in costly punishment suggests that it is unlikely that there should be any moral norm requiring it. Propensity aside, there are large differences in people’s ability safely to administer costly punishment. A Lockean might think this an innocuous point: Locke posited a natural executive right, not a natural executive duty. That is true, but the absence of a duty means that the incidence of private, third-party punishment is likely to fall into no predictable or governable pattern. Alternatively, one might say that indeed there is a general duty conscientiously and carefully to administer proportionate, direct third-party punishment. Presumably, breaches of this duty would, in their turn, be objects of punishment. Once diffuse punishment of defectors from the norm is allowed, it would be ad hoc and artificial not to allow diffuse counter-punishment in cases of unjust, mistaken, or excessive punishment. But Denant-Boemont, Masclet, and Noussair have found that “the negative effect on contributions of permitting [diffuse] counterpunishment is greater in magnitude than the positive effect of permitting [diffuse] sanction enforcement” (Denant-Boemont et al. 2007, 152). Recent work by O’Gorman, Henrich, and van Vugt (2009) shows that, in a standard Public Goods Game, enforcement by a single, designated player improves contributions by just as much as diffuse, Lockean enforcement, and more profitably, too. Moreover, the disturbing phenomenon of anti-social punishment is absent (cf. Bowles and Gintis 2004, whose simulations do not correct for this effect). As O’Gorman, Henrich, and van Vugt summarize: [I]t appears that while participants in diffuse punishment situations attend to both their own contribution and that of the target, perhaps using their own contributions to guide their decision on whether to punish, those in the solitary punisher condition attend only to the contributions of the target, possibly focused solely on whether contributions are maximally beneficial for the group, in which case any deviation from a full contribution represents an undesirable shortfall. (O’Gorman et al. 2009, 326) Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 178 William A. Edmundson To designate a single punisher is, of course, to suppress the Lockean lone-ranger. The selective advantages to a group of avoiding a Lockean norm seem evident. Diffuse Lockean punishment is unlikely to have come down to us, having survived eons of selection. As O’Gorman, Henrich, and van Vugt express it: Observed hunter-gatherer groups adopt various mechanisms to ensure cooperation, and leadership is one such mechanism that both integrates with humanity’s primate heritage and offers a mechanism for groups to coordinate activity [. . .]. Models in economics [. . .] and evolutionary biology [. . .] indicate that evolution can favour a single punisher per social group and that the actions of this one punisher can efficiently galvanize group cooperation. This solution is particularly interesting since it lacks the second-order free rider problem [those who free-ride on the costly enforcement actions of others] [. . .] and it avoids the problem of uncoordinated over punishment. (O’Gorman et al. 2009, 323; citations omitted) A single-punisher approach is a solution only if that punisher is himself or itself subject to social control. This means that any such agent is treated as such, that is, as agent for the group as principal. 12. Moral Anger, Moral Error, and the “Moral Channel” Diffuse, Lockean, lone-ranger moral enforcement would capitalize on the emotion of moral anger felt by victims and by-standers alike. In a way, this is a strength. We know by first-hand experience that we feel these emotions. A huge variety of public entertainment excites these emotions. Without the prevalence of the proper “reactive attitudes” among a group, one could hardly say that the group lived by moral norms. But moral anger is volatile and corrosive stuff (Pettigrove 2012). Moral anger’s epistemic vices are a powerful reason to put some distance between reactive attitudes and retributive responses. It is easy to see that a group that did so would, pro tanto, enjoy a selective advantage over those that did not. There are other indications that a structure of moral interaction that inhibits the impulse to administer punishment––and effectively limits the moral permission to do so––would have conferred selective advantages. Consider how delicately attuned we are––pre-consciously, consciously, and even unconsciously––to subtle expressions of moral reproof. The neural architecture of our brains, like the brains of our primate cousins, has dedicated and automatic face-detectors, which register not only the presence of a human face but also the direction of its gaze (Emery 2000; Haxby, Hoffman, and Gobbini 2000; Baron-Cohen et al. 1995). Remarkably, this capacity has been found to be active in persons who are totally blind (Tamietto et al. 2009; de Gelder and Tamietto 2007, de Gelder 2010). Darwin (1998) was among the first to notice that types of facial expression are universal across human cultures. Despite vast cultural differences, there is a universal facial expression of contempt, a key moral emotion (viz., one © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 179 corner of the mouth is stretched obliquely) (Eckman and Heider 1988). Burklund, Eisenberger, and Lieberman have investigated the neurological effects of observing facial expressions that convey specifically social disapproval. In their studies, a socially-disapproving face is formed by “raising one side of the upper lip, lowering the inner corners of the brow in a fashion similar to that displayed when expressing ‘confusion,’ and slightly tilting or pulling the head backwards” (Burklund, Eisenberger, and Lieberman 2007, 239–40). Having previously measured subjects’ selfreported “rejection sensitivity” by using a standard questionnaire, the researchers then exposed the subjects to video clips of a sociallydisapproving face, and to facial expressions (of anger, and of disgust) not of specifically social disapproval. They found that the more rejectionsensitive subjects “exhibited greater dorsal anterior cingulate cortex activity in response to disapproving facial expressions, but not in response to anger or disgust facial expressions” (ibid., 238). Although further research is needed, the study indicates that there is a neural basis for differential receptivity even to subtle cues of specifically moral disapproval. The neural basis is likely also trans-cultural and heritable: Other research suggests that certain variations within three genes involved in central neurotransmitter systems are associated with individual differences in social sensitivity (Way and Lieberman 2010). P.F. Strawson (1968) explained the essential role that “reactive attitudes” play in moral life; and these attitudes are deeply integrated with distinctive, universally constant, and often involuntary facial expressions. (Interestingly, having experimental subjects make an angry face can get them to be as angered as they would be by recalling a wrong they had suffered, or by empathizing with a literary character who suffered unjustly! — Keltner et al. 1993.) Experiments show that we have a remarkable ability to pick out a single disapproving expression among a crowd of other faces (Hansen and Hansen 1988; Pinkham et al. 2010). Having detected a face, we are unable to resist orientating ourselves toward it, even when we try (Burnham and Hare 2007). The “moral channel” is exquisitely receptive. Recent research shows that the subtlest facial cues dramatically influence behavior, even where it is obvious to the experimental subject that there is no one really watching. In one experiment, subjects were given test booklets whose covers bore one of two configurations of dots: either two dots above a single dot or the reverse presented in the same way. We perceive the two-dots-over-one-dot figure as a face, and subjects given test booklets bearing these devices “cheated” less, though unaware of the influence (Rigdon et al. 2009). In another experiment, subjects engaged in cooperative games on a computer terminal while alone in a room. One group’s computer screens displayed a small humanoid face, while a control group’s screens did not. The former group was much less likely to breach norms of fair dealing (Burnham and Hare 2007). A picture of a pair of eyes, Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 180 William A. Edmundson posted over the “honesty” box in a university coffee lounge induces markedly greater compliance with contribution rules (Bateson, Nettle, and Roberts 2006). H.L. Mencken (1990, 231) quipped that “conscience is the inner voice that warns us somebody may be looking.” It could turn out that, in truth, conscience is the inner voice that warns us that somebody is always looking. A little social monitoring, evidently, goes a long way. As Burklund, Eisenberger, and Lieberman indicate, “social rejection has been shown to be associated with such negative outcomes as decrements in self-regulation [. . .] increases in negative affect, and increased physiological stress responses such as elevated blood pressure [. . .] and cortisol” (Burklund, Eisenberger, and Lieberman 2007, 240, citations omitted). Social disapproval is typically intended to encourage conformity, but it can also lead its target to lose self-control. When explicit expressions of reproof are presented, there is a significant risk of provoking indignation and counterreproof (Nikiforakis 2008). Goffman, on the subject of ordinary spoken interaction, put the point well: The human tendency to use signs and symbols means that evidence of social worth and of mutual evaluations will be conveyed by very minor things, and these things will be witnessed, as will the fact that they have been witnessed. An unguarded glance, a momentary change in tone of voice, an ecological [sic] position taken or not taken, can drench a talk with judgmental significance. (Goffman 1967, 33) Vendettas can be triggered this way, almost upon “any momentary cause,” and once begun they can become almost unstoppable until one side conquers the other or some diverting event or superior authority intervenes (Fermor 1958, 86–99; Chagnon 1968). So it would not be surprising if inhibitions evolved to assure that the moral channel is a “hot line” to be used only sparingly. Speaking of the Cheyenne of the North American Plains, Hoebel and Llewellyn put the point this way: “[A] close-knit culture is commonly also a culture which places restraints upon easy voicing of a grievance. Subtle relations can be shattered by noise about them; and face is likely to be precious. A grievance which does break out is likely therefore to be breaking through” what they refer to as “the social inertia” (Hoebel and Llewellyn 1941, 301, 302; emphasis in original). I have been arguing that it is not inertia, but an inhibition that discreetly checks the impulse to impute moral fault to another. One might think that inhibitions of the use of the moral channel were unneeded in earlier evolutionary periods because solitude offered the individual a ready escape. But this is not so. Part of being a social animal is being unable to cope with extended periods of isolation, especially if these are enforced. Ostracism is a particularly painful social sanction because of its dilemmatic nature. The offender seeks the company of © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 181 others, but when finding it encounters a wall of disapproving gazes or averted faces. This is painful, and involves areas of the brain associated with somatic pain (Eisenberger, Lieberman, and Williams 2003). The “cold shoulder,” maybe unsurprisingly, has also been found to influence human perception of ambient temperature (Zhong and Leonardelli 2008); and, in an animal study, social isolation has been found to affect the chemistry of the brain (Agís-Balboa et al. 2007). Is it possible that the stereotypical eye-rolling expression of disdain exploits this vulnerability? Eye-rolling involves averting one’s gaze from the object of one’s disdain and turning it toward an audience, real or imagined, that is presumed to share the eye-roller’s disapproval, as though to say, “I speak for them.” 13. Conclusion This essay is meant to be a contribution to a neo- (or as a scoffer might say, a crypto-) Aristotelian political philosophy. It points to a state of nature only in the sense that it points to (or, to the scoffer, waves in the direction of) a substantial, remarkable, and still-developing body of empirical work going on in and around the intersection of evolutionary biology, social psychology, neuropsychology, and similar fields. It does not try to fit the familiar mold of state-of-nature theorizing as Hobbes, Locke, and Nozick practiced it. Especially, it does not conceive of politics as an artificial construction layered upon or erected with a pre-political native normative endowment. The liberal tradition in political philosophy has shown a tendency to populate this endowment a priori, with a bias toward securing previously determined ideological territory. The view advanced here does not take our political circumstances as something we are regrettably “born into,” as though we might equally as well have been born out of some such circumstances altogether. The view I defend does not take up the question whether and how to bargain or reason our way back into political circumstances by using some “thin” pre-political set of notions and moral principles. It rejects the question as unreal. This view does, however, take seriously the idea that there is a native normative endowment. But it understands this endowment as a subject of continuing empirical study, and not as a set of ideological axioms or armchair postulates. Nothing I’ve said does much to support the idea that our political nature is unified or, in other words, the idea that the political virtues––much less, the virtues generally––exhibit a unity. Looking at the evidence, we see that humans are social, and sociable, but at the same time selfish and often almost instinctively anti-authoritarian. In this regard we differ not only from social insects, but also from our nearest neighbors, the chimpanzees. Chest-thumping, pant-grunting, and what we would deride as bowing and scraping are too much of what it is to be part of chimpanzee society for it Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 182 William A. Edmundson to be easily imagined how we humans branched off from a common ancestor that, presumably, was that way too. Human social life is a struggle against dominance itself, as much as it is a struggle to become as dominant as one can. The evolutionary focal point of our ambivalence toward authority can be located in the role of the male head-of-household in hunter-gatherer societies. Within the household, he dominated wife (or wives) and children alike. Beyond the household, he was never more than primus inter pares, one voice among many. Unanimity was the rule of decision at both levels, but at the household level only the male head had the vote. Majority rule occupies an uneasy place in human politics, which it can hold onto decently only if represented as deriving from some deeper unanimity: perhaps from the magnanimous dispensation of an absolute sovereign, perhaps from the willing of a deeper volonté commune, or perhaps from the mandate of sovereign reason itself. Each of these is a fiction. Nothing in a naturalistic account can guarantee that the early evolutionary environment prepared us well for our present circumstances. Political philosophy is not to be blamed for wanting to make up the deficiency. I’ve said nothing directly to support another aspect of the Aristotelian view, which is that politics is a constitutive part of the good life. Why think that living a good life requires that one “partake in coercing one’s fellow citizens?” as Bas van der Vossen has pressed the point. If that Periclean idea is an element of an Aristotelian view of what it is to be a political animal, I disown it, and in fact I think the thesis of this essay goes a way toward explaining why a condition in which not everyone partakes directly in third-party enforcement is not only better for those who can be disengaged from it, but also for the larger group. However, it is not good to be so disengaged that one is altogether silent about wrongdoing. The virtuous disposition is not one of “republican boldness” unmoored from any more specialized role within a political order. Rather it is one that accepts a division of moral labor. The division need not and in justice ought not be hierarchical in a fundamentally anti-democratic sense. But it is one that is natural to call “natural” and plausible to believe to be generally beneficial. What I have outlined defies what is now commonly known as “Hume’s Law”: There is no deriving an “ought” from an “is.” I find solace in the fact that Hume himself devoted a scant paragraph to this so-called law (more of an observation, really), and then proceeded with his “attempt to introduce the experimental method of reasoning into moral subjects” as though it had not even occurred to him as a possible objection to his own project. I also derive some small comfort from an unlikely source, “Kant’s Law”: “Ought” implies “Can.” A correct account of what duty requires has to portray it as something within our capabilities; and what our capabilities are is something that science, and experience, have to tell us. They may be © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 183 in some ways narrower, and in other ways much wider, than what we now believe. College of Law Georgia State University Atlanta, Georgia, USA E-mail: [email protected] References Agís-Balboa, R., et al. 2007. Down-Regulation of Neurosteroid Biosynthesis in Corticolimbic Circuits Mediates Social Isolation-Induced Behavior in Mice. Proceedings of the National Academy of Sciences USA 104: 18736–41. Arnhart, L. 2004. The Incest Taboo as Darwinian Natural Right. In Inbreeding, Incest, and the Incest Taboo: The State of Knowledge at the Turn of the Century. Eds. A.P. Wolf and W.H. Durham, 190–218. Stanford, CA: Stanford University Press. Aristotle. 1994. The Complete Works of Aristotle Translated into English. Ed. Jonathan Barnes. Princeton: Princeton University Press. Baron-Cohen, S., et al. 1995. Are Children with Autism Blind to the Mentalistic Significance of the Eyes? British Journal of Developmental Psychology 13: 379–98. Bateson, M., D. Nettle, and G. Roberts. 2006. Cues of Being Watched Enhance Cooperation in a Real-World Setting. Biology Letters 2(3): 412–4. Bendor, J., and P. Swistak. 2001. The Evolution of Norms. American Journal of Sociology 106: 1493–545. Blass, T. 2004. The Man Who Shocked the World. New York: Basic Books. Boehm, C. 1999. Hierarchy in the Forest. London: Harvard University Press. Bowles, S., and H. Gintis. 2004. The Evolution of Strong Reciprocity: Cooperation in Heterogeneous Populations. Theoretical Population Biology 65: 17–28. Boyd, R., et al. 2005. The Evolution of Altruistic Punishment. In Moral Sentiments and Material Interests: The Foundations of Cooperation in Economic Life. Eds. H. Gintis et al., 215–27. Cambridge, MA: MIT Press. Boyd, R. and P. Richerson. 1992. Punishment Allows the Evolution of Cooperation (or Anything Else) in Sizable Groups. Ethology and Sociobiology 13: 171–95. Burklund, L., N. Eisenberger, and M. Lieberman. 2007. The Face of Rejection: Rejection Sensitivity Moderates Dorsal Anterior Cingulate Activity to Disapproving Facial Expressions. Social Neuroscience 2: 238–53. Burnham, T., and B. Hare. 2007. Engineering Human Cooperation. Human Nature 18: 88–108. Camerer, C., and E. Fehr. 2004. Measuring Social Norms and Preferences Using Experimental Games: a Guide for Social Scientists. In Foundations of Human Sociality. Eds. J. Henrich, R. Boyd, et al., 53–95. Oxford: Oxford University Press. Cashdan, E. 1989. Hunters and Gatherers: Economic Behavior in Bands. In Economic Anthropology. Ed. S. Plattner, 21–48. Stanford: Stanford University Press. Chagnon, N.A. 1968. Ya˛nomamö: The Fierce People. New York: Holt, Rinehart and Winston. Cinyabuguma, M., T. Page, and L. Putterman. 2006. Can Second-Order Punishment Deter Perverse Punishment? Experimental Economics 9: 265–79. Cummins, D. 1996. Evidence for the Innateness of Deontic Reasoning. Mind & Language 11: 160–90. Darley, J. and B. Latané. 1968. Bystander Intervention in Emergencies: Diffusion of Responsibility. Journal of Personality and Social Psychology 8: 377–83. Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 184 William A. Edmundson Darwin, C. 1998. The Expression of the Emotions in Man and Animals. P. Ekman, ed. Oxford: Oxford University Press. (1st ed. 1872.) de Gelder, B., and M. Tamietto. 2007. Affective Blindsight. Scholarpedia 22: 3555. Available at www.scholarpedia.org/article/Affective_blindsight de Gelder, B., and M. Tamietto. 2010. Uncanny Sight in the Blind. Scientific American 5: 60–5. Denant-Boemont, L., D. Masclet, and C.N. Noussair. 2007. Punishment, Counterpunishment and Sanction Enforcement in a Social Dilemma Experiment. Economic Theory 33: 145–67. de Quervain, D., et al. 2004. The Neural Basis for Altruistic Punishment. Science 305: 1254–8. de Waal, F. 2007. Chimpanzee Politics. Baltimore, MD.: Johns Hopkins University Press. Dunbar, R. 1996. Gossip, Grooming, and the Evolution of Language. Cambridge, MA: Harvard University Press. Eckman, P., and K.G. Heider. 1988. The Universality of a Contempt Expression: a Replication. Motivation and Emotion 12: 303–8. Eisenberger, N., M. Lieberman, and K. Williams. 2003. Does Rejection Hurt? An fMRI Study of Social Exclusion. Science 302: 290–2. Elias, N. 2000. The Civilizing Process: Sociogenetic and Psychogenetic Investigations. Oxford: Blackwell. Emery, N. J. 2000. The Eyes Have It: The Neuroethology, Function and Evolution of Social Gaze. Neuroscience and Biobehavioral Reviews 24: 581–604. Fehr, E., and U. Fischbacher. 2004a. Third Party Punishment and Social Norms (Working Paper No. 106). Zürich: Institute for Empirical Research in Economics, University of Zürich. Fehr, E., and U. Fischbacher. 2004b. Third-Party Punishment and Social Norms. Evolution and Human Behavior 25: 63–87. Fehr, E., and S. Gächter. 2000. Cooperation and Punishment in Public Goods Experiments. American Economic Review 90: 980–94. Feinberg, J. 1965. The Expressive Function of Punishment. The Monist 49: 397–423. Fermor, P.L. 1958. Mani: Travels in the Southern Peloponnese. London: John Murray. Gächter, S., E. Renner, and M. Sefton. 2008. The Long-Run Benefits of Punishment. Science 322: 1510. Gaus, G. 2011. The Order of Public Reason. Cambridge: Cambridge University Press. Godwin, W. 1976. Enquiry Concerning Political Justice. Ed. I. Kranmick. Harmondsworth: Penguin. (1st ed. 1793.) Goffman, E. 1967. Interaction Ritual: Essays on Face-to-Face Behavior. New York: Pantheon. Haidt, J. 2001. The Emotional Dog and Its Rational Tail: A Social Intuitionist Approach to Moral Judgment. Psychological Review 108: 814–34. Hamlin, J. K., K. Wynn, and P. Bloom. 2007. Social Evaluation by Preverbal Infants. Nature 450: 557–60. Haney, C., W. Banks, and P. Zimbardo. 1973. Interpersonal Dynamics in a Simulated Prison. International Journal of Criminology and Penology 1: 69–97. Hansen, C., and R. Hansen. 1988. Finding the Face in the Crowd: An Anger Superiority Effect. Journal of Personality and Social Psychology 54: 917–24. Hauser, M. 2006. Moral Minds: How Nature Designed Our Universal Sense of Right and Wrong. New York: Ecco. Haxby, J. V., E.A. Hoffman, & I. Gobbini. 2000. The Distributed Human Neural System for Face Perception. Trends in Cognitive Science 4: 223–33. © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2 Politics in a State of Nature 185 Henrich, J., et al. 2006. Costly Punishment Across Human Societies. Science 312: 1767–70. Herrmann, B., C. Thöni, and S. Gächter. 2008. Antisocial Punishment Across Societies. Science 319: 1362–7. Hobbes, T. 1996. Leviathan. R. Tuck, ed. Cambridge: Cambridge University Press. (1st ed. 1651.) Hoebel, E. 1954. The Law of Primitive Man: A Study in Comparative Legal Dynamics. Cambridge, MA: Harvard University Press. Hoebel, E., and K. Llewellyn. 1941. The Cheyenne Way. Norman, OK: University of Oklahoma Press. Hudson, J., and A. Bruckman. 2004. The Bystander Effect: a Lens for Understanding Patterns of Participation. Journal of the Learning Sciences 13: 165–95. Keltner, D., P. Ellsworth, and K. Edwards. 1993. Beyond Simple Pessimism: Effects of Sadness and Anger on Social Perception. Journal of Personality and Social Psychology 64: 740–52. Knauft, B., et al. 1991. Violence and Sociality in Human Evolution. Current Anthropology 32: 391–428. Knoch, D., et al. 2010. A Neural Marker of Costly Punishment Behavior. Psychological Science 21: 337–42. Locke, J. 1952. The Second Treatise of Civil Government. T. Peardon, ed. Indianapolis, IN: Bobbs-Merrill. (1st ed. 1690.) Marshall, L. 1979. Sharing, Talking, and Giving: Relief of Social Tensions Among !Kung Bushmen. In Kalahari Hunter-Gatherers: Studies of the !Kung San and Their Neighbors. Eds. R. Lee and I. DeVore, 220–32. Cambridge: Cambridge University Press. Mencken, H.L. 1990. The Vintage Mencken. Ed. A. Cooke. New York: Vintage. Merry, S.E. 1984. Rethinking Gossip and Scandal. In Toward a General Theory of Social Control. Ed. D. Black, 271–302. New York: Academic Press. Mikhail, J. 2011. Elements of Moral Cognition: Rawls’ Linguistic Analogy and the Cognitive Science of Moral and Legal Judgment. Cambridge: Cambridge University Press. Milgram, S. 1992. The Individual in a Social World: Essays and Experiments. 2nd ed. New York: McGraw Hill. Milgram, S. 2009. Obedience to Authority: An Experimental View. Preface by P. Zimbardo. New York: Harper Perennial. (1st ed. 1975.) Nikiforakis, N. 2008. Punishment and Counter-Punishment in Public Good Games: Can We Really Govern Ourselves? Journal of Public Economy 92: 91– 112. Nozick, R. 1974. Anarchy, State, and Utopia. New York: Basic Books. O’Gorman, P., J. Henrich, and van M. Vugt. 2009. Constraining Free Riding in Public Goods Games: Designated Solitary Punishers Can Sustain Human Cooperation. Proceedings of the Royal Society 276: 323–9. Ostrom, E., J. Walker, and R. Gardner. 1992. Covenants Without a Sword: SelfGovernance is Possible. American Political Science Review 86: 404–17. Pettigrove, G. 2012. Meekness and “Moral” Anger. Ethics 122: 341–70. Pinkham, A. et al. The Face in the Crowd Effect: Anger Superiority When Using Real Faces and Multiple Identities. Emotion 10: 141–6. Prinz, J. 2007. The Emotional Construction of Morals. Oxford: Oxford University Press. Rigdon, M., et al. 2009. Minimal Social Cues in the Dictator Game. Journal of Economic Psychology 30: 358–67. Sabini, J. and M. Silver. 1982. Moralities of Everyday Life. Oxford: Oxford University Press. Ratio Juris, Vol. 26, No. 2 © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. 186 William A. Edmundson Schoeman, F. 1992. Privacy and Social Freedom. Cambridge: Cambridge University Press. Shweder, R. A., et al. 1998. The “Big Three” of Morality (Autonomy, Community, and Divinity), and the “Big Three” Explanations of Suffering. In Morality and Health. Eds. A. Brandt and P. Rozin, 119–69. London: Routledge. Singer, T., et al. 2006. Empathic Neural Responses are Modulated by the Perceived Fairness of Others. Nature 439: 466–9. Sober, E., and D. Wilson. 1999. Unto Others: The Evolution and Psychology of Unselfish Behavior. Cambridge, MA: Harvard University Press. Sripada, C. and S. Stich. 2006. A Framework for the Psychology of Norms. In The Innate Mind: Culture and Cognition. Eds. P. Carruthers, S. Laurence and S. Stich, 280–301. Oxford: Oxford University Press. Strawson, P. F. 1968. Freedom and Resentment. In P. F. Strawson, Studies in the Philosophy of Thought and Action, 59–80. Oxford: Clarendon. Tamietto, M., et al. 2009. Unseen Facial and Bodily Expressions Trigger Fast Emotional Reactions. Proceedings of the National Academy of Sciences USA 106: 17661–6. Trivers, R. 1971. The Evolution of Reciprocal Altruism. Quarterly Review of Biology 46: 35–57. Wallace, B., et al. 2007. Heritability of Ultimatum Game Responder Behavior. Proceedings of the National Academy of Sciences USA 104: 15631–4. Way, B., and M. Lieberman. 2010. Is There a Genetic Contribution to Cultural Differences? Collectivism, Individualism and Genetic Markers of Social Sensitivity. Social Cognitive and Affective Neuroscience 5: 203–11. Williams, B. 1995. Replies. In World, Mind, and Ethics: Essays on the Ethical Philosophy of Bernard Williams. Eds. J.E.J Altham and R. Harrison. Cambridge: Cambridge University Press. Zhong. C.-B. and G. Leonardelli. 2008. Cold and Lonely: Does Social Exclusion Literally Feel Cold? Psychological Science 19: 838–84. © 2013 The Author. Ratio Juris © 2013 John Wiley & Sons Ltd. Ratio Juris, Vol. 26, No. 2

RELATED PAPERS

RELATED TOPICS

Log In

Politics in a State of Nature

Politics in a State of Nature

Related Papers

RELATED PAPERS

RELATED TOPICS