Homo reciprocans

Political Economy and Cultural Evolution

by Cosma Rohilla Shalizi

English-speaking social science, especially economics, is dominated by a tradition going back to Adam Smith and the other late 18th- and early 19th-century British political economists and historians of civil society. It focuses on individuals, and sees their acts and choices as primary. Larger entities — markets, states, institutions, cultures, and classes — are shorthand ways of speaking about patterns in the acts of many individuals.

Partly because they lend themselves to precise, mathematical expression, individualist theories have proven theoretically insightful, practically useful, and surprisingly powerful. They are also basically unrealistic. The standard individual economic agent, Homo economicus, has been called a "hedonistic sociopath." He also has no culture at all, and is far too smart. Nobody, not even exponents of "rational choice" theories, is much like Homo economicus, which is good for humanity, but bad for those theories.

There is another social science tradition, going back to Herder, Hegel, and other German contemporaries of Smith, which evades these problems by focusing on collective entities like cultures and classes: these entities, proponents say, are real, and they do things to people; indeed, they shape the people who belong to them in fundamental ways. This isn't much of an alternative, however, because it amounts to saying that cultural effects are produced by — culture. This is like saying that opium puts people to sleep because it possesses a "dormative virtue." This is a dilemma for social scientists: do they invoke incredible creatures like Homo economicus, or vacuous entities that don't really explain anything?

Though most find Homo economicus an implausible caricature of human behavior, for want of any replacement, he has had to do. However, one of the most exciting developments in the social sciences in recent years is the emergence of someone to take his place — called Homo reciprocans by Samuel Bowles, a professor of economics at the University of Massachusetts at Amherst and a member of SFI's coordinating committee for its Keck Foundation evolutionary dynamics program. Bowles described "reciprocans" during his talk "Social Organization and the Evolution of Norms" at SFI's May 1999 Science Board Symposium which focused on "Humans and Other Social Animals."

Perhaps the most striking way to introduce this character is with some results from experimental economics. Take public goods games: the experimenter gives his subjects some money and explains that they can choose, separately, how much to keep and how much to contribute to a common pool, which will be, say, doubled, to pay for a benefit in which all will share equally. The payoffs are such that contributing nothing maximizes one's individual gains. In such "collective action" situations, Homo economicus contributes nothing, and hopes to exploit everyone else. In real-life experiments, however, few, often less than a half of the subjects, start out doing this. When given the opportunity, experimental subjects can be surprisingly determined to punish those who cheated them, even at considerable cost to themselves. More surprisingly, this is true even on the last round of the game, when they couldn't hope that punishing the cheaters now would change their behavior in the future. Homo economicus, by contrast, realizes that punishing cheaters under these conditions is, like contributing to the common pool, a pure waste of money, and so refrains from doing so.

H. L. Mencken once defined conscience as "the inner voice that warns us somebody is looking," and social scientists and biologists have often interpreted apparently generous acts as self-interest in disguise. But consider the results of the "ultimatum game," with two players, an experiment that has been carried out in over one hundred studies in twenty countries with highly consistent results. The experimenter picks a player at random, hands him a wad of cash, and tells him to divide it between himself and the other player. The second player can either accept the offer, in which case they split the pot as agreed, or reject, in which case both get nothing. But Homo economicus, playing against another economicus, offers only one cent, which is accepted. Most people offer between forty and fifty percent, and routinely reject offers of less than a third, even in one-shot games (where there's no chance for retaliation), even when the pot amounts to several months' earnings. That people make large offers is striking enough, but what really rules out Homo economicus is that people reject quite substantial offers in order to punish others for not cooperating, even when it costs a lot to do so.

This suggests a very different view of what economic agents are actually like, and thus emerges Homo reciprocans. As Bowles puts it in an essay with his long-time collaborator and fellow U-Mass economist Herbert Gintis: "Homo reciprocans comes to new social situations with a propensity to cooperate and share, responds to cooperative behavior by maintaining or increasing his level of cooperation, and responds to selfish, free-riding behavior on the part of others by retaliating against the offenders, even at a cost to himself, and even when he could not reasonably expect future personal gains from such retaliation." This is certainly in line with empirical observations: people do produce public goods, they do observe normative restraints on the pursuit of self-interest (even when there is nobody watching), and they will put themselves to a lot of trouble to hurt rulebreakers.

Besides the fact that typically there have been no other alternatives, another objection to abandoning Homo economicus is that he is, in his own way, a reliable standard. There is just one way of being a hyper-intelligent hedonistic sociopath, but there are at least as many ways of being someone mostly inclined to follow norms as there are norms to follow. A single, definite, unambiguous prediction is, for some, superior to an endless series of "maybe" and "it could be this norm...on the other hand it could be that one" predictions. There are two ways out of this, and Bowles, characteristically, takes both.

First, which norms a given group of people follow is a factual question, which can be investigated. For example, to see whether the results of the ultimatum game are uniform across very different types of societies, Bowles, along with anthropologist Robert Boyd, experimentalist Ernst Fehr, and Gintis, have organized field experiments of the ultimatum and public goods games in a dozen simple societies around the world, including some, like the Machiguenga in Amazonian Peru, with very limited exposure to markets and other modern institutions. (No noble savages have turned up so far: the Machiguenga, in fact, are the closest approximations to Homo economicus yet discovered.)

Second, norms do not vary in arbitrary and indefinite ways; there are certain patterns which appear to be common across societies. In experimental games, subjects explain their acts by saying that self-seeking behavior would not be "fair." Fairness need not mean equality, but inequality does have to be justified somehow. They must have reasons for it; a person may be rewarded for skill or effort; for virtue; or (a surprisingly common move) because they are more than human, at the very least a different and much better kind of human. (This is borne out by the experimental results: if, for instance, you get to be the proposer in the ultimatum game by passing some test, even a trivial one, you offer less, and the other player accepts less.) Even when norms allow for inequality, they still enjoin some reciprocity; the players accept mutual, if not equal, obligations.

Readers familiar with evolutionary psychology will remember the elegant experiments of Leda Cosmides and John Tooby on "cognitive adaptations for social exchange." They showed that most people can solve certain kinds of logic puzzles when the problem is phrased as one of detecting people breaking rules, even if those same people cannot solve formally identical problems which are presented abstractly or with different subject matter. This suggests that the capability for reciprocans-type behavior is something very deeply wired into our brains.

This only makes more pressing the question, which will have already occurred to readers familiar with sociobiology, of how (if at all) Homo reciprocans can evolve and sustain itself in a population which contains some exploiters. In the presence of such exploiters, natural selection will tend to eliminate organisms which engage in unprofitable behaviors, such as helping others or engaging in costly punishments. One of the standard theories of the evolution of cooperation between relatives evades this by postulating "assortative" interactions between kin — if organisms tend to interact with their close relatives, which carry many of the same genes, then altruistic behaviors can establish themselves, even in populations which contain many exploiters. Applying such reasoning to non-kin, one of Bowles and Gintis' most recent papers shows that Homo reciprocans could have evolved during the Paleolithic era in a similar way: reciprocators could have proliferated in a population, even if their fitness was reduced by upholding norms, so long as reciprocators were more likely to find themselves with other reciprocators in well-ordered groups capable of surviving bouts of material scarcity and attacks that norm-flouting groups could not endure.

Since norms differ from place to place and time to time (sometimes even within a single society at a single time), we would like a theory that explains what determines differences in norms. Such a theory would also help close a significant gap in current individualist models. They regard agents' preferences as "exogenous," as fixed, given — and inexplicable. The question is how to make them "endogenous," to bring them within our models?

The acquisition of norms and preferences (and other bits of culture) is an abiding concern for Bowles; his first book with Gintis, on the role of public education in modern America, was, precisely, a study of how people are acculturated, and a study of the ways in which what is learned is affected by economic circumstances. Acculturation is an individual-level process: we get it not from our culture, but from parents, siblings, other relatives, neighbors, playmates, colleagues, and, of course, teachers. Sometimes, as in formal schooling in developed countries, this is a very deliberate process, and people are taught certain skills, beliefs, norms and preferences, because these are economically useful. In other cases, there is a mutual influence between social and economic organization and culture: people are apt to imitate those who achieve social success.

This kind of "replicator dynamic" is actually easier to model than is deliberate instruction, using extensions of models from population biology originally developed by Marcus Feldman with Luca Cavalli-Sforza, as well as the related evolutionary models of Robert Boyd and Peter Richerson. Individual rewards in such models are affected both by material circumstances and by economic and social organization — "who meets whom, to undertake which joint activities, with what payoffs and opportunities to acquire new traits," as Bowles puts it. This structure of social interactions, Bowles shows, controls the rate and direction of the differential replication of tastes, habits and norms, whether the replication is genetic, cultural, or some combination of both.

What happens when material circumstances and social structure change? In particular, what happens when those who follow previously accepted norms are liable to fail, whereas those who break them, who adhere to some other set of norms, may be seen to prosper? Such a situation is unsustainable; ultimately, either the norms have to change, or the material situation does. Because most people are merely Homo reciprocans, and not "rebels in defense of tradition," it's safe to bet on the norms changing, but this can be a lengthy process. One very common reaction, for instance, to introducing markets, trade, and a money-economy into societies which previously didn't have such institutions is to produce fringe groups of people who are "rootless" — cut off from the older social groups, much more given to the pursuit of individual self-interest. (Bowles' interest in cultural evolution was awakened by witnessing precisely such a sequence of events as a teacher in a remote part of Nigeria in the early 1960s: "What happened in two hundred years of European history unfolded before my eyes in the course of a couple of years.")

In highly marketized societies, careful, intelligent, competitive maximizers of personal gain, unfettered by sentiment or scruple, can do very well for themselves. Certain restrictions on competition are enforced (CEOs of other software companies can't have Bill Gates assassinated), but these have very little culture-specific content, and continual efforts are made to remove any lingering specificity. It's not so much that markets make people into Homo economicus, but that they present situations which evoke behavior that resembles his, and reward it. (In experimental Prisoner's Dilemma games, subjects tend to cooperate if the game is called "Community" and defect if it's "Wall Street.")

As Adam Smith knew, the institutions of the market can only work if many people (e.g., police, judges, parents, soldiers) do not, in the line of duty, act like Homo economicus at all, but instead act more like Homo reciprocans. Balancing two different, incompatible sets of norms — one for the marketplace, the other for the home, and for relations with friends and workmates — is not an easy task, and there is a natural tendency for the balance to tilt in one direction or the other, for the domain governed by one set of conventions to grow at the expense of the other's. Nobody really knows whether this will happen in our case, or whether we'll continue our uneasy impersonation of Homo economicus, or even whether we'll hit upon new rules. Bowles quite openly hopes for new rules more conducive to humans flourishing throughout the planet.

Bowles will coordinate a workshop at the Institute next January. "Coevolution of Institutions and Preferences" will bring together economists and other social scientists to discuss the dynamics of institutional and individual behavioral evolution. The goal is to understand how social interactions — defined by markets, intergroup bargaining, firms, and other economic organizations — shape the evolution of individual preferences, and in turn how these preferences shape the evolution, and in particular, the emergence of new economic organization. As part of the Institute's continuing program on evolutionary dynamics, Bowles is planning subsequent workshops including one on the role of group formation, the distinction between insiders and outsiders, and group extinctions in evolutionary processes.

Bowles' papers are available at http://www-unix.oit.umass.edu/~bowles


Santa Fe Institute Bulletin, volume 14, no. 2 (Fall 1999), pp. 16--20