How everyday ethics becomes a moral economy, and vice versa

Unrealistic assumptions underlying neo-classical economic theory have been challenged by both behavioral economics and studies of moral economy. But both challengers share certain features with neo-classical theory. Complementing them, recent work in the anthropology of ethics shows that economic behavior is not reducible to either individual psychology or collective norms. This approach is illustrated with studies of transactions taking place at the borders between market rationality and relationships among persons – organ donation and sex work. The paper argues that the inherent value accorded to social relations tends to resist instrumentalization and that the biases that dealing with other people introduce into reasoning are not flaws but part of the core functions of rationality. (Published in Special Issue Bio-psycho-social foundations of macroeconomics) JEL A10 D01 D63 D91 E71 Z13

What is the problem to which the ethical turn in an economic anthropology might be a response? Let me start, in the style of a cultural anthropologist, with small examples. My Uncle Dick was a Chicago-trained economist and protégé of Milton Freedman-his car bore a vanity license plate that read "Maximize." Uncle Dick was quite comfortable with the arguments of people like Gary Becker that sex work and the adoption of children should not be restricted but instead be governed entirely by market forces. In dinner table debates, he enjoyed twitting his relatives with the cool reasoning of economic thought, to demonstrate the irrationality of various ordinary practices that we took for granted. For instance, his son (also an economist today) is an ardent outdoorsman and skilled craftsman. Uncle Dick would tease him for doing his own yard work and auto repairs, on the grounds that the value of his time far exceeded what he would pay someone else to do the jobs. And then there's the matter of Christmas presents. Why give gifts? At best, you merely receive a utility more or less equal to what you gave, minus the cost of the inefficiencies involved. Given the vagaries of gift selection, moreover, you're rarely going to maximize anyone's preferences, yours or the recipient's. True, Uncle Dick would acknowledge that the good will generated might balance out the transaction costs incurred. But how do you explain this method of generating good will? Why not just say nice things to one another? And even granting that method, why should you take the price tag off the gift, and purchase wrapping paper? After all, the recipient can easily guess what the necktie or cookbook cost, and the wrapping paper is no sooner on the gift than it's removed and added to the recycling bin. A waste of effort and expense altogether. Dick would then turn to me and ask whether I leave a tip on the counter after having lunch during a cross-country roadtrip. Having paid the bill, why should I add extra since I am never going to return to that eatery, and so have no expectation of better or worse service in the future? Now Uncle Dick recognized that rational choice models are idealizations (and thus, perhaps, not directly vulnerable to empirical counter-examples). He was also a well-adjusted member of his society, and perfectly willing to go along with local custom. But I think he was inclined, in principle, to see local custom as inherently irrational precisely because it reflects the influence of other people and the contingencies of history. One of the main sources of irrationality in economic decision-making, in this view, is the effect that other persons have on the reasoning processes of the individual. Other persons are sources of pathologies like mob violence, fallacies like superstition, or simply the blinkered nature of rote behavior. Left to his or her own resources, the cognitively healthy individual should be capable of rational maximization. 2 _________________________ 2 Even as harsh a critic of modern economic arrangements as Thorstein Veblen (1994) tended to attribute certain kinds of irrational behavior to the malign influence of social pressure. His model of "conspicuous consumption" rightly claimed that certain kinds of economic behavior are motivated by competition for social status. But that criticism is itself predicated on an underlying view of human sociality as a source of distortions of what would otherwise be rational decisions based on utilitarian goals. I will argue here that, on the contrary, rationality itself is first and foremost a social phenomenon. You can't abstract other people away from the context of reasoning. Research suggests that human cognition is not particularly well adapted for ascertaining the true state of affairs or determining the best course of action. Indeed, some cognitive scientists have argued that, given the roster of built-in fallacies and biases, this is not the primary function of rationality. Rather, it is well adapted to social functions such as influencing others (Mercier and Sperber 2017). This is consistent with some recent anthropological research (summarized in Keane 2016; see also Enfield andLevinson 2006, Keane andLempert forthcoming, Sidnell 2011), which shows that self-awareness is a function of the social need to account for oneself to others. The very nature of rationality, and of the self, are such that biases such as those expressed in tipping and gift wrapping, are built into what it is to be a person because social relations are not something added onto a self that pre-exists them, but are essential contributions to, and ultimately constituents of, personhood. Drawing on recent work in the anthropology of ethics, I argue that social interaction forms the crucial link between what cognitive scientists and moral psychologists are learning, on the one hand, and what is happening at the level of social groups on the other.

3
Moral economy I will return to tipping and yardwork in a moment. But consider gift-wrapping. In contemporary middle-class American life, the purpose of gift-giving at events such as birthdays, weddings, bar mitzvahs, or Christmas is not to distribute goods. Most people would agree that it is to express and strengthen social relations, or at least to fulfill an obligation toward those relations. Economics might call this "signaling," but the question is, what is being signaled, and why does it work? In traditional, non-capitalist, gift-exchange systems, the gift is considered an extension of the giver-you give a part of yourself to another, which is what forges the bond between you (Mauss 1967). The contemporary American version of this is what many of us were taught as children, that the best gift is the one you make yourself (Carrier 1990). That's also why, in most cases, a gift of money isn't really right. But it turns out that, contrary to the common adage, it's never just the thought that counts-the monetary value of the object does enter into the consideration. The anthropologist Daniel Miller (1998) has shown that British gift givers perform careful calculations of price in choosing what to give whom-it's embarrassing to give something too expensive, insulting to give too little, in light of the particular relationship in question. Although the ideal is to give of yourself, most of us are reduced to purchasing gifts in marketplace. As a commodity, the gift is impersonal. I didn't make that necktie, and it isn't any different from the hundreds of others produced in the same mill. So how do I personalize it so that it can represent me and add to the bonds between me and you? Taking off the price tag symbolically removes it from the marketplace, and wrapping it adds a visible bit of my own labor. That is, they are signaling mechanisms that establish a symbolic link between me and my gift, and thereby, between the donor and the recipient (see Carrier 1993). Trivial though this example may seem, it is a symptom of a much larger issue (see Sandel 2012). The sociologist Viviana Zelizer (1997Zelizer ( , see also 2005 has shown how much work nineteenth century Americans put into creating barriers between home and workplace. They did so because they saw the ethics of the workplace as properly dominated by self-interested competition and rational calculation. By contrast, the domestic sphere was the place that ought to be governed by a distinct set of ethical norms. It was where the other virtues should be fostered and cultivated: sympathy (in Adam Smith's [1976] sense of the term), selflessness, and what they took to be the Christian virtues. In order to protect the domestic sphere against the corrosive force of the marketplace, all sorts of prophylactic practices were developed, such as restricting the wife to "pin money." We no longer exclude middle class housewives from handling money but many families still consider open talk about money to be tabooed in front of the children (with poignant results when it comes to paying for college: see Zaloom 2019). The little customs of gift giving reflect a similar contrast between market and domestic relations.
What was being protected, and what was the anxiety? One answer is the moral economy. This old concept was revived by the historian E.P. Thompson (1971) to make sense of social strife in eighteenth century Britain. The basic idea is that inequality was accepted as legitimate as long as the gentry accepted their paternalistic obligations to the peasantry according to prevailing norms of fair prices and the laborer's entitlement to subsistence in times of scarcity. As the sociologist Marion Fourcade summarizes the argument, when market forces expanded into traditional communities, "more was at stake than grain prices: time-honored norms, customary duties and communal solidarities had come under threat, too" (2017: 662). Central to the tradition of moral economy was the idea that membership in a community was itself a sufficient basis on which to claim economic rights (Götz 2015, Scott 1976, Thompson 1971. One didn't have to be a "productive member of society" in order to do so. Carrying this idea forward in time, then, what the domestic sphere represented for nineteenth century Americans was a domain in which membership in the family should, in theory, supervene on strictly monetary calculation. Each household would be a small moral economy preserved within a larger world of unconstrained market forces. We might, then, see nineteenth century middle class Americans as fighting a rear guard action against the ongoing rationalization of the marketplace and its extension to all social relations. Now, for all its insights, there are two difficulties with the moral economy model as laid out here. One is that it tends to treat so-called traditional societies as highly consensual-as Scott writes, "woven into the tissue of peasant behavior, . . . whether in normal local routines or in the violence of an uprising, is the structure of a shared moral universe, a common notion of what is just" (1976: 167). 3 Moral economy, in this view, is more or less static, only subject to disruption when external forces enter the scene-a view with which wielders of the Silicon Valley motto "move fast and break things" might agree, but that ethnographers and historians have shown to be mistaken. But, as Susana Narotzky points out, in her comments on an earlier version of this paper, moral economies will always involve struggles around category-making. Even people on the Indonesian island of Sumba, whose social organization at the time of my fieldwork (Keane 1997) exemplified the so-called "gift economy" described by Marcel Mauss, grappled with the relative indeterminacy of categories in practice. They recognized that a gift _________________________ 3 Nor are so-called "traditional societies" necessarily prone to seeing themselves in communitarian terms. For instance, the highlanders of Indonesia's Sulawesi described by Tania Li (2014) treat even their own spouses and children as more or less autonomous economic agents. that is reciprocated too quickly verges on purchase and sale (amounting to a rejection of social bonds)-but to fail to reciprocate in a timely manner verges on theft. Where a transaction fell in any given instance might be fiercely disputed. Moreover, Sumbanese felt the appeal of alternative possibilities. The very same people who loudly extolled the virtues of gift exchange and the social bonds it sustains also found themselves sometimes chafing at the relentless pressure of kinship obligations. They would even recall mythical ancestors whose superhuman powers included the ability to wed without making marriage payments to their affines-a transgressive power people could find attractive. Although these kinds of tensions, conflicts, and contradictions are most evident in communities undergoing rapid social transformation (e.g. Donham 1999, Foster and Horst 2018, Robbins 2004, Hirschkind 2006, Schielke 2015, they are endemic features of any social world. Beyond that, however, to the extent we are concerned with the contemporary industrialized world, this criticism, an empirical question about so-called traditional societies, needn't concern us here. The second difficulty is that it portrays moral economy as fundamentally conservative, and largely located in the past, or at least, to be fast disappearing (Mauss 1967, Polanyi 1957a, 1957b; but see Narotzky 2016), and, by implication, ourselves as moving toward an ethically neutral-and increasingly rational-economic future. Indeed, like neo-classical economics, the moral economy approach tends to accept that the economic and moral dimensions of human life are inherently distinct, that the former is corrosive of the latter, and therefore, that the boundaries between them should be controlled. It treats ethical life and economics as relatively distinct spheres of action that operate only completely different principles. Economic rationality may be celebrated as an efficient means of satisfying human desire or criticized as reducing humans to objects. Moral economies can in turn be valued for sustaining the common good and collective well-being, or decried as inefficient, and dictated by a cultural order that is imposed upon people by social forces external to them, artificially constraining their real desires. Tthis dichotomy tends to treat the social as a source of irrationality. This view would invite us to agree with the rational choice theorists that a properly functioning modern economy is one in which the inefficiencies and biases imposed by custom, sentiment, and ethics have been eliminated, and, perhaps, that history is moving us in that direction.
Against this, some scholars have argued that we should understand all economies to be moral economies (Boltanski and Thevenot, 2006). Certainly the liberalization of markets has its defenders on moral grounds since the days of Adam Smith and Montesquieu (Fourcade 2017, Hirschman 1977. Georg Simmel (1990) stressed the links between money and freedom, something appreciated-at times--as far afield as Papua New Guinea (Akin and Robbins 1999). There is an implicit ethics to contemporary liberal economics, which stresses the positive value of the individual, of agency, the disenchantment of the world and freedom from hierarchies, religious beliefs, and so forth (Keane 2001, 2007, McCloskey 2006, Muehlebach 2012. Minimally, all economies differentiate and hierarchize between good and bad, high and low, legitimate or illegitimate, and the market economy is no exception-think of how present day German central bankers view the Greeks (Holmes 2009). Correlatively, assuming that market culture necessarily undermines the moral terrain prevents us from truly engaging with the complexities of the relationship between the two, as well as the complexity of the social struggles at stake (see, for instance, Marina Welker's [2014] nuanced ethnography of Corporate Social Responsibility).
Consider how economic behavior itself is an affordance for ethical action. As James Siegel (1969) observes of Muslim men in Sumatran markets, what worries them about bargaining is less the chance of accepting the wrong price than that of letting others trick them into succumbing to their desires at the expense of their faculties of reason. At risk is not financial but moral loss. Thus, 'Each successful act of exchange . . . affirms the rationality of the trader' (Siegel 1969: 250; see Keane 2008). The Japanese arbitrageurs whom Hirokazuo Miyazaki (2013) studied draw on the financial models with which they work as guides to their own ethical self-disciplining. Taking their theories of economic rationality to heart, some apply them to their own individual life decisions. Similarly, Caitlin Zaloom found that financial models of risk are crucial to how Chicago commodities traders manage their identities and jockey for status. She writes that "ascetic practices and social displays of virtue enacted in the pit describe a capitalist ethic that centers on the mastery of the self under conditions of hazard and possibility" (2004: 366). But notice that these are values that are ultimately not reducible to the instrumental purposes of maximized utilities (see Sahlins 1996): they are about becoming a certain kind of person, one whose virtues are recognizable and valued by others in social relationships.
Nonetheless, as James Carrier (2018) points out, by treating all social arrangements as moral, there's a risk of making the idea of moral economy meaningless-it becomes so all embracing that it fails to provide us with useful distinctions. Although we can't dispose of the concept altogether, moral economy needs to be freed of organicist and communitarian assumptions. One alternative is to recognize that there is indeed a moral dimension to every economic system, but that the economic theories that purport to account for them and inform their institutional and legal bases may fail to understand their own ethical implications. One reason they fail to do so is that they have a weak empirical understanding of actually existing ethics. In particular, they treat ethics as a special domain of human behavior and thought that can be localized and bracketed, a source of preferences and an externality that needn't be incorporated into economic thought. An alternative that may help us gain a better understanding of the articulation of the ethical and the economic dimensions of social existence is to turn to everyday ethics.

Everyday ethics
The anthropology of everyday ethics focuses on the ways in which people's judgements of value and character operate in the course of their daily lives (Das 2014, Keane 2016, Laidlaw 2014, Lambek 2010, Mattingly 2014. Like moral economy approaches, work in everyday ethics takes as its central premise that the fact that people live in communities with distinctive histories is an irreducible feature of human life, economic and otherwise. It's a truism to note that people's economic behavior responds to existing models, norms, and habits, but the implications may still need spelling out. How people do so is reinforced by the expectations of those around them, their shared memories, and their projects, the future as they and those around them see it. Their actions take into account and may be compelled not just to other people but all sorts of other entities considered relevant in their social reality-God, "market forces," ancestral spirits, "the economy," laws, and so forth.
To state the obvious, economic behavior does not occur in a social vacuum, nor do people start from scratch every time they decide what to do. Obvious though this may be, what we mean by "social" and its role in causal explanations are far from settled matters. The anthropology of ethics takes it as given that the sociality of people's lives with one another has certain intrinsic features that are not reducible to their lives as individuals. 4 However, unlike moral economy approaches, the anthropology of everyday ethics does not start from the collectivity-it does not take "society" as a given. Nor does it assume the existence of consensus within a community. Although concepts like culture, norm, and society remain important for most anthropologists of everyday ethics, they do not have strong explanatory value. For instance, we don't expect to find that people's actions directly reflect cultural norms.
Some of us who work on everyday ethics draw insights from cognitive science, psychology, and the sociology of interactions, as well as from ethnographic observations within communities, their histories of moral thought, and the institutions that derive from them. From cognitive science we take the same basic insights that have also informed behavioral economics. These include the existence of fundamental reasoning biases such as overjustification effects, order effects, prospect theory, emotional salience biases, situationism, time discounting, and so forth (e.g. Greene and Haidt 2002, Güth, Schmittberger, and Schwartze 1982, Kahneman and Tversky 1979, Ross and Nisbett 1991, Schelling 1978, Tomasello 1999). And, as noted above, some cognitive scientists argue convincingly that these biases are not functional flaws, but basic properties of a rationality that evolved primarily for social purposes such as forging alliances, persuading others, establishing identities, and so forth (Mercier and Sperber 2017). From developmental and moral psychology we take factors that contribute to the development of ethical intuitions and give shape to their outcome. These include empathic responses, joint attention, shared intentionality, disgust responses, reciprocity of perspectives, spontaneous sharing behavior, intention-guessing, stereotyping and categorizing, norm-seeking and thirdparty norm enforcement (Bloom 2004, Cacioppo, Visser, Pickett 2006, Haidt 2001, Hirschfeld and Gelman 1999, Kalish 2012, Wellman 1992. From the sociology and linguistics of interactions we take the analysis of turn-taking, and the dynamics of face-saving and facechallenging, conversational repair, and their metapragmatic regimentation (Enfield and Levinson 2006, Garfinkle 1967, Goffman 1967, Silverstein 1993. The psychological features just mentioned tend to operate below the level of the individual's awareness. Yet most definitions of ethics involve some degree of self-awareness: an action can't usually be considered to be properly ethical if it arises completely spontaneously, or if the people involved are utterly unaware of its normative dimensions. If we look at the sociology of interactions, we find it works along the borderline between unconscious behavior and the acts of purposeful, goal-oriented people. In this respect, what goes on in social interaction plays a crucial mediating role between the domains of individual cognitive and emotional processes, on the one hand, and on the other hand, the public world of shared values, norms, rules, and _________________________ 4 It has been argued that some social worlds, e.g. in Melanesia, are best described in terms of partible persons made up of "dividuals," ways of distributing attributes of selfhood and responsibility that are not confined to individual bodies (Strathern 1988; see also Muniesa 2018). Closer to home, the cognitive scientist Edwin Hutchins (1995) showed that agency in an American naval vessel's bridge is distributed across individuals. I thank Susana Narotzky for pushing me on this matter.
concepts. This is important because it helps us understand just how social norms do and do not affect actual behavior. In other words, it is by looking at social interactions that we can begin to grasp how the collective norms of moral economy and the behavior of individuals interact.
The anthropology of everyday ethics treats ethics as a dimension of ordinary activities, focusing on people's general propensity for judgment and evaluation in light of noninstrumental values. We define the ethical as an orientation to other people and activities that are undertaken for their own sake rather than as a means toward some further goal. It includes people's fundamental intuitions about how they should live, and how they should live with one another. Two important preconditions for everyday ethics are the propensity for interactions with people marked by empathy and cooperation, and the tendency to make evaluations and judgments of people and their actions on non-instrumental terms (Keane 2016: 45). Having said that, we should be clear that purely intuitive judgements, such as those described by much of the psychological literature, are not in themselves necessarily ethical. Nor do they necessarily give rise to ethics. However, they provide raw material for ethical life. They are what we call "affordances," things that can be taken up for a purpose when relevant. At the heart of our approach is this premise: humans tend to value social relations for their own sake. Although relationships obviously can be instrumentalized, this is a secondary process, parasitic on the first. The value of social relations is so fundamental to the very formation of the self that is cannot be wholly explained in terms of some other goal, such as maximizing utilities. Let me briefly sketch some of the research findings that support this claim.
The urge to be sociable for its own sake appears very early in infants. For example the infant spontaneously gestures toward things in his or her surrounding well before beginning to speak. Chimpanzees and humans both know how to point out objects to others. But chimps only do so in order to get the thing being pointed to-in other words, the gesture is purely instrumental and self-centered. By contrast, human children point to things simply in order to direct another person's attention to them. They do so just for the sake of sharing orientation toward something with another person, a way of being in company with one another (Tomasello 2009). By the age of eighteen months, toddlers are also spontaneously helping others. They have developed the cognitive capacity to grasp what someone else wants, and to recognize when that person is thwarted and thus in need of assistance. Studies of infants and toddlers-people too young to be fully marked by cultural training-show they have a strong orientation toward other persons for their own sake, an ability to displace their attention and feelings away from the self, and a propensity to help and cooperate with others. All this is not necessarily cheerful and positive. For instance, they also show signs of bias both for and against other people, excluding and stigmatizing others as well as shoring up in-group solidarities. But even in this negative form, people are prone to evaluating one another, and their actions, on other than utilitarian grounds. 5 Overall, we see children are motivated to engage in activity without instrumental goals beyond the activity itself. The most obvious form this takes is play. But we also see an instinct for conformism. Before school age, children have already developed intuitions about norms. _________________________ 5 Evolutionary psychologists may object that the infant's orientation to others is explained by the long period of dependence on caregivers. Even if we grant this origin story, we should not conflate causes (this came about because of X) with consequences (this is done in order to X). Humans did not develop bipedalism in order to run the New York Marathon. This is especially evident in the early development of norm-seeking and third party norm enforcement in childhood. Third part norm enforcement-for instance, objecting to a distribution of candy in which someone other than oneself has been treated unfairly-shows that children value norms as such. What we are finding is (1) children value (some) relationships as goods in their own right, (2) they accept norms because they value those relationships. In other words, social relationships are a core human value that cannot be reduced to instrumental explanations.
We can see the famous ultimatum game in this light (Güth, Schmittberger, and Schwartze 1982). In this experiment, one person is given a sum of money and told to split it with someone else. The second person can either accept the offer or reject it, in which case both individuals get nothing. It turns out that if you offer a sum that falls below a certain threshold, around onethird of the total, the recipient will tend to reject it. This flies in the face of the rational calculation of utility, according to which one should accept even a very small sum, since it is still better than nothing at all. One interpretation of this result is that people value fairness for its own sake enough that they would rather forgo monetary gain than be treated unfairly. In this view, it's money that is the "externality." Moral psychologists tend to stop at this point, limiting their explanations to individual psychology, for example, by claiming it shows people have an innate sense of fairness. But fairness in this context only makes sense as a judgment about social relations. It's not the quantity of money that makes a distribution unfair, it's what that money says about how another person values me, something that makes no sense outside of the context of the relationship between us, and their impact on my sense of self.
As I mentioned, it's hard to make sense of the idea of ethics without reference to people making decisions about alternative courses of action about which they are more or less aware. People need to know what they are doing, if their acts are to count as ethical. And most of the psychological findings, taken in isolation, concern processes that operate below the level of awareness, often automatically. How, then, does awareness of actions and their ethical value come into play? The missing piece here is social interaction. The individual's behavior characteristically anticipates the perspectives of the other persons with whom he or she is engaged. People respond to one another, defining and redefining who they are and what they are doing over the course of time. They do so for one another. This is true of the moment-bymoment scale of a conversation (consider how you might defuse a tense moment by reframing what you have said as "just kidding"), as well as the long term scale of the life span. The result is the ongoing work of constructing a shared sense of reality, of intersubjectivity which is continually in the process of being built up, reshaped, or undermined.
What we call "the second-person perspective" refers to the ways people account for themselves to others-justifying, explaining, criticizing, praising, and so forth (Darwall 2006, Keane 2016. Empathy, sharing, cooperation, intention-seeking, and reciprocity of perspectives all require the second-person perspective. They involve address to another person, and just as important, they (usually) take that other person as someone who can address me in turn. When people account for themselves, or when they must cooperate, or when they seek to influence actions in the future, they must in some sense understand one another. As noted above, the language of reason itself should be understood in this context, as the medium by which people try to persuade and influence one another. In short, people have to draw on a shared language of norms and values in order to be recognizable to one another. That is how moral economy, culture, and religion enter the picture. They don't directly cause behavior, but they provide the tools that make it understandable and therefore, subject to ethical judgments and downstream responses to it. An insult is different than an unintended error, an honorarium is different from a gift, which is different from a bribe-and disputes over the distinctions among categories are not confined to tax accountants and criminal lawyers. The language of self-explanation and justification through which people account for themselves to one another and try to shape the judgments they form helps endow behavior with its social consequences.
This focus on social interaction is meant to help us understand how intuitive and instinctive actions can be properly called ethical, and how public norms become part of the individual's intimate individuality. As noted, most philosophers hold that for an action or a thought to be ethical, it must be recognizable to those who make them as being ethical (e.g. Parfit 2011). The value concepts and categories that make ethics explicit and available to actors' awareness circulate in the public sphere of a community. That is, they are not simply innate responses, unconscious intuitions, or private thoughts. Rather, they are social in their very sources, their expression, and, to an important extent, their function. The primary arena in which explicit ethical concepts come into play is in justification and accusation. People explain their actions by portraying their motives and goals, and identifying the nature of what they are, for other people. When we describe ourselves as sincere or just kidding, when we make excuses ("it was an accident," or "I misunderstood you"), for instance, we are trying to persuade others to accept an ethical characterization in terms that we both recognize-something they might accept, reject, or redefine. We recognize it because we share a normative vocabulary.
Verbal interaction is the preeminent site where people may demand explicit reasons and accounts of one another. It is in response to the demands posed by talk that rationalizations and justifications arise. The natural home of argument, reasoning, and justification is not in the individual autonomous mind but in palpable social interactions, whether face-to-face or in more mediated forms (Keane 2016: 162-3). These start at a very basic level, in the collaborative work of constructing a shared reality between actors. (This shared reality need not be friendly. Even people who are quarreling have to agree that that's what they are doing.) Notice that crucial things like trust reflect "the confidence that we are interacting with people who are committed to the same definitions of reality to which we adhere (Garfinkel 1963)-and trust is crucial to simple things like accepting money. We can see this from studies of economies in which money is not trusted (Guyer 2004 on Africa, Truitt 2013 on Vietnam, Lemon 1998 on Russia).
The basic claim that the anthropology of everyday ethics makes is that ethics is not something the individual does, feels, or cognizes all by him or herself. Nor is ethics just carrying out a cultural or religious program, a set of norms to which one has been socialized. We start with the claim that people value social relationships for their own sake. Their acceptance of social norms and religious morality depend on this prior value. Moreover, even the individual's sense of self draws on his or her relations to others. We can see this in ethical concepts such as dignity, respect, humiliation, and shame. Words like these, and their equivalents across languages, point to the important role that other people play in the value that individuals accord to themselves. Ironically, one of the fiercest early critics of consumerism, Thorstein Veblen (1994), noted this, without, however, drawing the conclusion we would. As mentioned above, when he criticized the Americans of the nineteenth century's Gilded Age for conspicuous consumption, he was pointing to something important: they did not value material goods for their own sake, but for the sake of their social identities To summarize, the anthropology of everyday ethics draws our attention to the ways in which individual behavior cannot be understood in isolation from social interactions and relationships. First, interactions catalyze actors' self-awareness and prompt them to define who they are and what they are doing. Second, when people define themselves and their actions, they must use categories and concepts that are recognizable to others, which forces them to draw on existing cultural resources. Third, to the extent that they engage in reasoning, they do so not simply in order to obtain objective grounds for action, but as part of the larger activity of persuading, cooperating with, challenging, or otherwise affecting other people. These certainly can have instrumental dimensions-but they build on, and are ultimately relative to, the fundamental propensity for value judgements. How, then, does the study of everyday ethics change the way we look at economic life? Here I turn to ethnographic examples of economic activity in the ethical and legal borderlands of economic rationality.

Gift and commodity: Organ donation
One way in which the idea of moral economy takes concrete form in contemporary capitalist market relations is in the distinction between gift and commodity. This distinction is acknowledged in both law and everyday practices. We can see it at work in the humble custom of removing the price tag and wrapping the present, and, less humbly, in the treatment of gifts in the tax code or the prohibition on purchase of office. The distinction also tends to mark the limit of those things human that, as Kant put it, are ends in themselves and without price (Feinberg 2005, Keane 2001, Sandel 2012, Zelizer 2005. Hence the strictures in American law against treating as commodities certain things, such as body parts or sex. 6 How that line is delimited, however, is enormously variable across societies and their histories. It wasn't not long ago, after all, that human beings were sold on the open market in the United States. Here I want to turn to one way in which pressure is being exerted on the gift/commodity distinction in the present, a system for acquiring kidneys for transplanting. Unlike most organs, a kidney can be removed from a healthy person with no ill side effects, so a living person can donate it to someone in need. The demand for kidneys vastly exceeds the supply but American law prohibits using the marketplace for allocating this scarce resource. 7 One solution has been the nonsimultaneous, extended, altruistic donor (NEAD) chain. The modular form involves two pairs of individuals. Each pair has a patient in need of a kidney, and someone willing to donate to them. Now suppose the donor's kidney isn't compatible with the patient's immune system, and that the same situation applies to the other pair. But it turns out that the available kidney in _________________________ 6 The National Organ Transplant Act forbids the transfer of "any human organ for 'valuable consideration for use in human transplantation'" (Healy and Krawiec 2012: 660). 7 In one recent year in the US, 4,573 candidates for kidney transplant died while waiting for an organ to become available (Healy and Krawiec 2012: 651). the first pair is compatible with the patient in the second pair, and vice versa. The solution is to swap donors: the volunteer in the first pair gives a kidney to the patient in the second pair, and in return the second volunteer donates to the first patient. The problem is, matching up pairs within the constraints of time and space to carry out the transplant is extremely difficult. The NEAD solution is to establish a chain of donor-patient pairs, starting when "an altruistic donor freely gives a kidney to a patient, initiating a chain of transplants among a series of donorpatient pairs." We needn't do into the details here, but the question is, what kind of transaction is this? And what difference does it make how people define the kind of transaction it is?
As Healy and Krawiec (2012: 650) point out, the "exchange of awkward goods-for example adoptive children, gametes, human tissue, and so on-is often accompanied by a considerable amount of practical and symbolic work that signals the transaction's social meaning and dictates the basic principles by which the exchange is governed." But the NEAD chain is ambiguous, having features of both gift and commodity (or, speaking in terms of the mode of transaction, a contract), some of which come into play at different points. At the point of the original donation, the emphasis is on the moral incentives of altruism and paying it forward. And yet, unlike organs donated posthumously, the kidneys that travel along NEAD chains are not pure gifts because each donation is meant to induce a reciprocal donation somewhere further down the line (2012: 661). So, on the one hand, like posthumous organ donation, which relies on the altruistic motive of the "gift of life . . . the NEAD chain also harnesses the logic of the gift" albeit with a more collective emphasis, since it depends on "feelings of group solidarity and commitment-to keep the chains going" (2012: 656). On the other hand, unlike posthumous organ donation, the chain is based on the implicit promise of a kidney given in return.
Since that reciprocal donation has been bargained for, it resembles a contracted-for compensation (2012: 662). But there is no way to enforce this quasi-contract, since should the second donor renege, you can't have a court order them to undergo surgery involuntarily. And a system of fines risks openly commodifying the organ, since the recalcitrant donor could just treat the fine as the price of the kidney they (or the patient for whom they entered into the transaction) have received. What seems to reinforce reciprocity is encouraging participants to feel that they have, in fact, entered into a contractual arrangement. Yet NEAD has consciously avoided instituting actual contracts, for fear they will undermine the trust and moral commitment on which participation depends (2012: 659).
This worry about contracts seems to be a version of what psychologists call the "overjustification effect," and economists call "crowding out." Although the research is not conclusive, there is evidence that in certain contexts, introducing a reward or a punishment tends to debase the value of what would have been a disinterested action or something that is a good in itself, by making it seem self-interested (Tomasello 2009: 9-10, Healy andKrawiec 2012: 664-6). In addition, by treating the participants as if rewards or punishments were necessary in order to motivate them, an overly contractual approach may also signal that they cannot be trusted to do the right thing on their own initiative. In either case, the value of the action as an inherent good, or of the actor as virtuous, is challenged by how an action is defined, that is, the symbolic work to which Healy and Krawiec refer. But this work doesn't occur in a vacuum. What philosophers call "action under a description" depends on what descriptions are available in a given society at a given historical moment-acts of aristocratic honor at one time and place, Confucian virtue, Franciscan kindness, Spartan discipline, or socialist brotherhood in others. You can't effectively or even recognizably just carry out any possible action in any time and place (Keane 2016).
Thus, possible act descriptions are subject to historical forces-they are not just expressions of innate human psychology. What we can see in the case of kidney donation is that categories and the ethical norms they entail are bending under pressure. As Healy and Krawiec point out, organ exchange doesn't fit existing models for either gift or commodity, but has aspects of both (2012: 647). The boundaries between what can or cannot be a commodity, what is or is not subject to the rational calculations of the maximizer are changeable. Notice as well that the nature of the transaction is not given just by the goods in question, nor by rational choices alone. If you were to remove the social forces involved, you might well eliminate the motivations that allow the system to work as well. It is just such constraints-the "awkwardness" of donation and the ambiguity between gift and commodity, that my Uncle Dick might have considered to be exogenous biasing factors. But I want to argue that they are also constitutive of the actors' motives for entering into transactions in the first place, and the judgments that guide them. What's important about this example is that it mixes motives of gain and altruism in ways that are almost impossible to disentangle. And, moreover, the moral benefit is inseparable from the donor's relationships to other persons and their negotiations over how their actions should be defined.

Commodities and relationships: Sex and death
What NEAD shows is that certain kinds of transactions depend on the motivating power of people's ethical sensibilities, yet the social meaning of the transactions that result is not necessarily given in advance. Rather, an emergent set of economic and ethical categories is being worked out on the ground. I turn now to another example of ambiguity, emergence, and social relations in an "awkward" transaction: sex work. I draw here on the research of Kimberly Kay Hoang (2011Hoang ( , 2016 on sex work in Ho Chi Minh City, Vietnam as it developed after Doi Moi ("renovation"), the transition from socialist to market economy initiated in 1986. Hoang found that sex work in this region is stratified into a low-end sector catering to poor local men, a mid-tier for white backpackers, and a high-end that caters to overseas Vietnamese men. At the low end, there is no ambiguity. The transaction is fast and anonymous, the payment immediate and direct, in a straightforward market in orgasms. Matters become more complicated in the higher tiers of this industry, which have more open-ended temporal trajectories with multiple possible outcomes. Like the low-end sector, the mid-tier encounters begin with a clear price paid for a commodity. However, "as their relationships develop, intimate caring and sexual labors become complexly intertwined with economic relations" (Hoang 2011: 384). The complexity is introduced by the latent possibility that a short term transaction will develop into a long term relationship, even in marriage. In these higher strata, the "porous boundaries in the relations between sex workers and their clients . . . allow women to offer a variety of services that go beyond sex, in return for various forms of payment beyond money" (2011: 368-9). At this point, the role of money is transformed as both women and their clients are careful to treat it as a gift, not a payment for services (2011: 386). What is received in exchange differs, however, according to the type of client. Both backpackers (mid-tier) and overseas Vietnamese (high-tier) are often interested in more than sex, but what they seek differs. Some backpackers are looking for "authentic" experiences, others end up wanting to rescue women from poverty. The overseas Vietnamese, taking advantage of their relative wealth in Vietnam compared to their lower status in their current countries of residence, engage in conspicuous consumption, purchasing expensive liquor and other commodities in highly visible ways. Hoang remarks that the sex industry provides them with "public spaces where they can achieve a sense of dignity" that might not be readily available to them in France or the United States (2011: 394). The dignity in question, however, is not simply displayed in the purchase of those commodities. It is also found in the conspicuous disclaiming of commodification in relation to the sex workers. By insisting that transactions with women are gifts, we might infer, they are displaying their own value as persons, insofar as their relationships are also recognized as personal. 8 The conspicuous consumption that the Vietnamese clients engage in, and the dignity they wish to sustain, are examples of a value that depends on one's relationship to other people. The shift from lower to higher end transactions also reflects the value of that relationships as projected forward in time. What this research also shows is the role of time scales in making ethical categories. The short term relations, with a clear beginning and ending point, are unproblematically subject to direct monetary measurement. Higher tier sex work depends on the reciprocity of perspectives between the partners. This in turn permits the projection of their respective intentions into an imagined future (even potential marriage) with no definitive end point. Like kidney exchanges, this projection toward the hypothetical future cannot be guaranteed unilaterally or in advance by either partner. Organ donation and mid-and high-tier sex work depend on mutual trust-like money itself. Trust is not just a social value, nor is it simply a matter of individual psychology. It is both a property of social interaction and has a reflexive relationship to it (projecting from present relations to hypothetical future ones). In other words, the difference between the instrumental rationality of the lower order sex work, the buying of an orgasm, and the higher tier ones, lies precisely in the emergence of a social relationship. At that point, the direct calculation of financial values becomes obscure-it hardly goes away (the $500 bottle of brandy, for instance, takes its social meaning from its price) but it is no longer handled in direct calculations of exchange. Incommensurables come to be treated on the model of the gift, a voluntaristic (or quasi-voluntaristic) transaction between persons within an indefinitely extendable time frame. As relations become more open-ended and longterm, quantification becomes increasingly problematic. This reflects a more general principle identified by the anthropologists Jonathan Parry and Maurice Bloch (1989): people who resist _________________________ 8 Even men in traditional kinship systems who acquire their wives through the negotiation of marriage payments insist on carefully distinguishing the transaction from what happens in a marketplace. This is done by tight restrictions on the kinds of goods transacted and the ritualization of the process, among other things, which work to make the distinction highly visible-which implicitly recognizes the possibility, noted above, that transactions do not automatically fit into given categories and stay put there (Keane 1997). One of the justifications Sumbanese people offer for the marriage payments is precisely to indicate the dignity of the wife (Keane 2001(Keane , 2008. monetizing long term cycles of exchange often don't object to doing so for short term ones. The former are identified with social relations, the latter with instrumental purposes. 9 As Immanuel Kant recognized, to treat something as exchangeable with another is to depersonalize it. This is the challenge faced by any system of compensation (again, under dramatically differing historical conditions, as the old German system of wergild, or blood compensation payments, makes clear). When Kenneth Feinberg was appointed by the United States Congress to direct the compensation for families of people killed in the 9/11 terrorist attacks, he indicated the peculiar nature of the task by accepting the symbolic compensation of one dollar. The challenge he faced in carrying out his assignment was this: the Congressional mandate was to follow a purely financial form of reasoning: "eligible families and victims should receive different levels of compensation depending on the financial hardship visited upon the survivors" (Feinberg 2005: 151). Faced with this attempt to rationalize the death of a person as a loss of income, however, the claimants tended to resist. Some, for example, defined the compensation as a recognition of their loss of a family member, demanding more money "because of the victim's intrinsic moral worth" (2005: 154). Feinberg argues that these people were not motivated by the desire for financial gain. Rather, they were compelled by ethical value: "to accept a lesser payment would be insulting to the memory of the departed. The families had to ask for more. " (2005: 156). Indeed, so important was this moral dimension that seven families refused to file claims at all, even after Feinberg met with them personally to urge them to do so. As one of them said to him (echoing, I think, those who reject an inadequate payment in the ultimatum game), "Go away Mr. Feinberg. Thank you for coming, but no amount of money can replace it. " (2005: 162)." Here we may be seeing an extreme case of crowding out, money being seen not as a measure of moral value (as in the families who asked for more) but as displacing it, in a zero sum game. 10 In the context of early twenty-first century America, both possibilities were in play.

7
Economics, the self, and other people Let me now return to yardwork, tipping, and gift wrapping with which I opened. For my cousin, yardwork is a value in its own right, like play (or, for that matter, labor; see Bear 2017). When someone who could afford to pay someone else prefers to do it him or herself, they may be valuing, say, the exercise of a skill mastered, the escape from high stakes matters, the pleasure of exercise, or the appeal of living in a landscape one has shaped oneself. All of these are aspects of the activity which is an end its own right, one that has this value because, ultimately, The social relations associated with the long term are not necessarily seen as positive. Impoverished highlanders in Indonesia's Sulawesi, for instance, often prefer immediate payments for labor undertaken for others, fearing that entering into long term "gift" relations may set them on course toward a relation of servitude with the donor (Li 2014).
10 Even gamblers recognize the distinctions in question, and can exert considerable effort and expense to distinguish between monetary gain and the value of the social relations in play (Zuckerman 2018).
it contributes to the sense of self, and does so as an end, as Kant would say, not as the means to some further end, such as greater wealth or power. Of course rational choice theory has no trouble with this proposition in principle, since it can simply put that value into a black box called "utility." But it does so at the risk of petitio principii. If any possible value can be defined ex post facto as a utility, then the concept of utility loses its explanatory purchase. Turning, then, to the gift, a moment's reflection shows that although we think of gifts as spontaneous, it's obvious that in most cases they are required. You wouldn't show up at a wedding, or family Christmas gathering, or send your child to a birthday party, without a gift. But as Marcel Mauss (1967) observed long ago, although the gift is obligatory, across a wide range of societies the act of giving must be performed as if it were voluntary. Therefore the formal properties of the transaction are significant. They define the kind of action in question. Gift wrapping shows the intersection between the micro level of social interaction and the larger moral economy, a domain of relative consensus about, or at least anxiety about, the nature of social relations and the putative effects of the market on them.
Turing to tipping, Americans do this for a very selective set of workers, most commonly, taxi drivers, restaurant servers, and hair stylists. What these have in common is that they perform jobs that come close to those one would perform for oneself, or that an intimate might do for one. In other words, tipping takes place on the border of the personal. In an earlier period, jobs like these might have been carried out by servants, but ideologically, contemporary Americans tend to be uncomfortable with the servant-master relationship if it is expressed too openly. I suggest that this is why the payment we call tipping takes the form of a quasi-gift. It acts as if to define the service rendered as itself being like a gift from a quasi-friend, rather than hired labor. The form that the tip takes-normatively obligatory but performed as if it were voluntary-is a signalling mechanism. It performs work on the social interaction and in response to the ethical implications of the interaction, defining the exchange, and thus the relationship, as being of a certain kind. Where tipping is the norm, choosing not to tip can threaten your own sense of yourself as a decent person. From the interactive perspective, tipping mollifies one's discomfort with personal services. Tipping shows the need to incorporate both social interaction and shared habits into any behavioral economic approach. Individual cognition and emotion are not sufficient to explain it. As for rational choice, although the value of behaving like a decent human being doesn't always trump calculation and the maximization of utilities, it cannot be eliminated from analysis as an externality or reduced to something else-unless it is the utility in question, in which case, again, we risk circular reasoning.
Tipping, as a quasi-gift, shares with gifts the property that it must appear to be voluntary yet in fact it is obligatory. The norm has no formal enforcement mechanisms, but relies on the value of self-esteem, which in turn reflects how one appears to others, even strangers (the counterman in a roadside diner whom you'll never see again). The sense of ethical obligation is fundamentally social in nature, being are about relationships, not about material gain. Again, you might define the sense of self-esteem as a utility within the logic of maximization, but in doing so, the concept of utility begins to lose its meaningfulness.
In short, yardwork exemplifies the role that work can take among the innate values identified with the self. Gift giving works to sharpen the boundary between personal relations and the market in commodities. And tipping works to blur that same boundary. All three practices are economically inefficient, sometimes even costly. As both Feinberg (2005) and Zelizer (2005) show, people are willing to pay a steep financial price in order to maintain the distinctiveness of human identities and the social relationships that shape them.

8
Ethics and interaction: The take-away points The single most important take-away point I'd like to stress is this: we cannot understand economic behavior by treating individuals in isolation from the practices that sustain social relations and the value those relationships hold for them. Other people are not simply exogenous sources of bias through conformism or peer pressure. Nor are they vehicles of constraint, prohibition, or domination. Self-other relations show that other people are crucial to one another's own sense of motivation, self-perception, self-value, desires, and sense of obligation. Social relations are also the domain within which judgments are make explicit and the norms by which they are evaluated are brought into play-and they motivate the very act of rational argument itself. The approach from everyday ethics that I sketch out here does not assume the historical rupture narrative that the idea of moral economy tends to reproduce. It does not depend on either organicist models of cultural order -something lost in the traditional past-or an independent and ahistorical realm of essential human wants. It gives moral economy a central place in the workings of any and every economic system, regardless of historical epoch. What changes historically is not the presence or absence of morality but rather its place within the rationalizations and ideological self-justification of any given socio-economic system. Like behavioral economics, this approach rejects the assumptions of rational choice models and the calculating individual who knows his or her interests and can survey all the options. But it also departs from behaviorist economics several respects. First, rather than eliminate reason altogether, it offers a different account of reason, one that is fundamentally interactive. Second, rather than taking individual psychology as the baseline, it places individuals in social interactions. By stressing the self-other relationship, it brings to the foreground things like dignity, respect, persuasion, deceit, shame, and so forth. It takes as basic that the natural home of argument, justification, and ethical values is in social interaction. But there is a crucial scalar dimension to the model. Self-justification, persuasion, and so forth draw on the concepts and values that circulate within a larger social context. Conversely, what happens in interaction can end up exerting pressure back on those public concepts and values.
To summarize, I want to stress four points. First, social relations are a value that is built into what being a person is. Although those relations can be instrumentalized for material gain, that is a second order process parasitic on the first. As the study of child development shows, and research on the ethics of interaction reinforces, relations with other people start as ends not means, and the individual's sense of self cannot be established independently of those relations. Second, rationality is fundamentally interactive. As a tool for calculative judgment it is highly fallible. Its primary arena lies in actions with, upon, and against, other people. This also means that the "biases" that other people introduce into the reasoning process are part of what reasoning is about in the first place. Rational choice (as commonly portrayed) can therefore at best play a limited role in understanding economic activity, at least given the norms and functions of rationality that it presupposes. Third, choice is also a poor model for much economic activity, if we understand choices to be independent events in isolation from what has come before and after. Choices take place in the context of other choices made over time, including those that have been made by other people. And fourth, other people are not simply exogenous sources of irrationality that interfere with what would otherwise be rational choices. Other people are part of what gives those choices the value they have and provide the means for achieving that value.