At the end of the summer (August 29-September 2), I will be in New Orleans at the convention of the American Political Science Association. I will be on a panel entitled "Darwin, Justice, and Politics: The Evolution of Human Reciprocity." The other panel members are Steven Forde (the University of North Texas), Tamler Sommers (the University of Houston), Alex Schulman (Duke University), and Barry Weingast (Stanford University). The panel will be organized around a new book by Samuel Bowles and Herbert Gintis, A Cooperative Species: Human Reciprocity and Its Evolution (Princeton University Press, 2011). The panel meets Saturday, September 1, at 10:15 am.
This book is a good survey of the evidence and arguments for explaining the evolution of human cooperation through evolutionary game theory. It is also a good statement of the view favored by Bowles and Gintis--the the crucial mechanisms for the evolution of human cooperation include the group selection through warfare of strong reciprocity and parochial altruism.
I see three problems in this book. First, there is some conceptual confusion in their idea of strong reciprocity. Second, there is a gap in their evolutionary history, because the initial appearance of strong reciprocity is explained with only two paragraphs of speculation. Third, there is a second big gap in their evolutionary history--the past 7,000 years--that is covered in only a few paragraphs.
The first problem is the confusion in the very idea of strong reciprocity, which Bowles and Gintis identify as the disposition to cooperate without any expectation of future reciprocation and to punish defectors at some cost to oneself.
If there is no expectation of reciprocation, then why call this strong reciprocity?
Bowles and Gintis argue that while Robert Trivers' "reciprocal altruism" is not real altruism, but actually "enlightened self-interest," strong reciprocity is real altruism. But they also indicate that this is a conditional altruism, because the strong reciprocator will not cooperate if most others are not cooperating. So it seems that there is some self-interest here--the strong reciprocator expects his cooperation to be reciprocated--but this contradicts their definition of strong reciprocity as free from any expectation of future reciprocation.
Bowles and Gintis insist that the strong reciprocator bears the costs of punishing defectors without any expectation of any payoff. But then they also say that strong reciprocators will not bear the great costs of punishing if others are not punishing defectors. Moreover, they indicate that punishing defectors, or even just threatening to punish them, motivates defectors to become cooperators, and thus costly punishment has a payoff, because it sustains cooperation.
The fundamental problem here is that Bowles and Gintis take for granted the idea of "altruism" as originated by Auguste Comte--the idea that there's a strict separation between self-interest and selflessness, and that morality is identified as utterly selfless behavior. Comte coined the word "altruism" (altruisme in French) as a term for "the good of others" in contrast to egoism. If one accepts this, then prudence is no longer a moral virtue.
In her argument for "the virtue of selfishness," Ayn Rand rejected the morality of altruism--living for others and denying one's own good--as a denial of life.
Although this idea of altruism has been a pervasive and unexamined assumption of modern social science, biology, and moral philosophy, I doubt it. As naturally social animals, we extend ourselves into others for whom we feel some attachment. Thus, our concern for ourselves includes a concern for others who have some connection to us. Consequently, our other-regarding dispositions are extensions of our self-love. This also means that absolutely disinterested and universal love is impossible. That is, I think, the idea behind Aristotle's "friendship" (philia) and Adam Smith's "sympathy."
The incoherence in the idea of altruism is indicated by Bowles and Gintis in their edited book Moral Sentiments and Material Interests in two footnotes (33). They concede that any voluntary, intentional act of altruism must "increase the subjective utility of the actor," and thus, "if one truly cares about others, it may be self-interested to sacrifice on their behalf, even though it is manifestly non-self-regarding to do so." So now it seems that true altruism might be "self-interested"!
The other two problems in this book are related to this problem of altruism, and they concern the two big gaps in its evolutionary history. The first gap is the initial appearance in human history of strong reciprocity. Bowles and Gintis admit that they can only speculate about this, and their speculation consists of only two paragraphs (197-98). They assume that strong reciprocity originated as "a small behavioral modification of either kin-based altruism or reciprocal altruism." Notice that this concedes that strong reciprocity is evolutionarily rooted in kin selection and reciprocal altruism. What they say here is similar to what James Q. Wilson in The Moral Sense says about "attachment or affiliative behavior" (44, 127-28), which corresponds to Aristotle's "friendship" and Smith's "sympathy," and which was picked up by Darwin in his account of the "moral sense."
The second big gap in this book is the 7,000 years of human history from the early Holocene to the present. Bowles and Gintis cover this in only a few paragraphs (3-4, 111-12). Of course, it would be unfair to criticize them for not giving us a complete history of human civilization in one book. But the point here is that explaining this history is crucial for their general argument.
For example, they indicate in various passages the importance of trade throughout human evolution from prehistoric foragers to the present (2, 79, 94-99, 101, 114, 134). But they don't elaborate their view of trade in human history. What would they say about Matt Ridley's argument in The Rational Optimist that Smith in The Wealth of Nations was right to see "the propensity to truck, barter, and exchange" as the ground for explaining the uniqueness of human cooperation? Does this explain the commercial/industrial revolution of the past 200 years? What would they say about Deirdre McCloskey's argument that this great revolution came from intellectual changes favoring the "bourgeois virtues"?
Here is where Bowles and Gintis need to turn to the history of political philosophy, particularly the early modern history of liberalism. They do refer a few times to Smith. But they seem to favor "the Adam Smith of The Theory of Moral Sentiments" over "the Adam Smith of The Wealth of Nations" (1, 44-45, 79, 199-200), and thus they overlook the connection between markets and morals.
Explaining the commercial/bourgeois revolution would force Bowles and Gintis to clarify their ambivalent and ambiguous position on market societies. As former Marxist economists, they often denigrate the moral culture of market societies. And yet they have been involved in some cross-cultural research that shows that primitive societies that engage in some market exchange have stronger norms of fairness and trust than those with little market integration. Recently, in the Boston Review (May/June 2012), Bowles and Gintis have argued that markets promote morals. But in this book, they say nothing about this.
Another point about the second gap in their history is the issue of violence. Group selection through warfare is a fundamental element of their argument in explaining the evolution of "parochial altruism." They don't say anything, however, about the apparent decline in violence in human history, particularly in the last few centuries. The evidence for this has been surveyed by Steven Pinker in The Better Angels of Our Nature, and he argues that a major reason for this decline in violence is the cultural transformation brought by modern liberalism. Would Bowles and Gintis agree? In their Boston Review essays, they imply their agreement with this. But this is not clear in their book, which reflects their reluctance to embrace Darwinian liberalism. (One can also see this in Gintis's response to my essay on "Darwinian Liberalism" at the Cato Unbound website.)
If a great book is one that raises great questions, then this is a great book.
The many posts on related topics would include my posts on Pinker and Ridley. Some others can be found here, here, here, here, here, here, here, here, and here.
No comments:
Post a Comment