Distinguishing Friends from Enemies

All sorts of things can affect one’s tendency to like or dislike other individuals. Some are unexpected. A subliminal hint of lemon odor can influence one’s evaluation of a stranger’s likeability (Li et al. 2007). Inducing a feeling of disgust–by placing a dirty kleenex nearby, for example—can boost the severity of a negative moral judgment (Schnall et al. 2008a). Allowing a disgusted individual to wash her hands will prompt her to tone down her negative moral assessment (Schnall et al. 2008b). A person’s evaluation of stimuli can even be influenced by whether he is encouraged to press down with his hands on a table top at which he is sitting or pull up on its underside (Cacioppo et al. 1993).

Other influences on our interpersonal likes and dislikes are far more obvious. If a person is threatening to harm me, or is actually intentionally harming me, then I’ll likely form a negative image of him or her. If the person is promising to act in ways that improve my welfare, or is actually intentionally acting in such ways, then I’ll likely form a positive image of him or her. This seems so obvious that one doesn’t need to cite psychology literature to back it up. And if convincing us of this truism is “the heart” of Wright’s essay, then he certainly won’t meet any resistance from this quarter.

The issue is the extent to which Wright takes this truism and tries to draw more substantial conclusions from it. One concern I have regards the role of intentionality in our attitudes towards other individuals (or groups). If X is out to harm me intentionally then I’ll likely form a negative image of X; but if X hurts me by accident then I’ll be much more forgiving. Now, something important to note about zero-sum games versus non-zero-sum games is that intentionality has nothing to do with it. Even insects and plants can play these games with each other. So long as two parties both have interests and can causally affect each other’s interests, then the possibility of a game is on. The parties do not have to encounter each other spatially or even be aware of each other’s existence: If a nocturnal insect and a diurnal insect are competing for the same foodstuff, then they are playing a zero-sum game.

We can imagine scenarios where intentions and outcomes come apart. Well-intentioned aid packages sent to distant lands can end up in the hands of corrupt local warlords, who are thereby bolstered in their capacity to victimize those very individuals whose interests the charity was supposed to advance. Similarly, someone who seeks to harm another through spreading malicious rumors, say, may end up prompting sympathy for the subject of the gossip in a manner that actually ends up benefiting him or her. Putting this in more abstract form: Sometimes two parties intend to play a zero-sum game but end up playing a non-zero-sum game, and sometimes vice versa.

An interesting question to ask is: When intentions and outcomes come apart, where do our sympathies and antipathies lie? I hazard to suggest that the answer is that they generally go along with intentions.

If I hear that some distant person’s actions are harming my welfare then I naturally won’t be too pleased; but if I learn, further, that this occurs only through a complex causal chain of which this person is ignorant—if I learn that in fact this distant person thinks quite well of me and intends me good things—then my attitude won’t suffer the distortions and failures of imagination of which Wright speaks. In other words, it is not the belief that I am in a zero-sum game with X that causes the failure of “moral imagination,” it is the belief that X intends me harm—a belief that I may have even while knowing that I am in fact playing a non-zero-sum game with X. By symmetrical reasoning: If I believe that some person Y is really out to get me, but in fact Y’s actions are inadvertently benefiting me, then this may be sufficient for me to consider Y an “enemy” and have the usual range of hostile attitudes towards him. (Of course, I will also be pleased about the benefits that I am accruing through his actions, and probably consider Y not only an enemy but also an idiot for failing to see that I profit from his hostility.) In other words, it is not the belief that I am in a non-zero-sum game with Y that causes the expansion of my “moral imagination,” it is the belief that Y intends to act for my benefit—a belief that I may have even while knowing that I am in fact playing a zero-sum game with Y.

Another point to which I would like to draw attention is an apparent assumption of asymmetry in Wright’s argument. When we believe ourselves to be engaged in a zero-sum game, Wright claims, “our mind naturally impedes clear comprehension of [our interactant’s] motivations.” The detection of zero-sumness apparently distorts perception and blocks accuracy. But why is the same point not made of our attitudes towards non-zero-sumness? When we form a flattering image of an ally—when we choose to overlook his past misdemeanors, for example–why is this not equally a kind of distortion, a kind of inaccuracy in our thinking? Why is my unflattering portrayal of my enemies any less true to reality than my flattering portrayal of my allies?

Wright may respond that both attitudes are symmetrically inaccurate, but that one kind of inaccuracy is benign and the other pernicious. Our unflattering inaccuracy with respect to our enemies stands in the way of understanding, and a lack of understanding of enemies is a pragmatically bad idea. (How much more effectively we could defeat them if we could understand them!) By comparison, our flattering inaccuracy with respect to our allies may also stand in the way of true understanding, but no great harm ensues. After all, our allies are, by definition, not seeking our harm.

However, if I understand Wright correctly, he is not advocating that we extend our moral imagination to our real enemies; he is not arguing in his essay that we should overturn our unflattering and understanding-hindering antipathy towards terrorists. Rather, his point is that we have been sucked into interpreting people who are in fact allies as enemies. Wright’s central claim is that we should overturn the distorting influence of unflattering antipathy towards that vast majority of Muslims who are in fact not our enemies at all.

I wonder, again, if the point is supposed to be symmetrical. Suppose that instead of having been misled by the media into thinking that a group of friends is really our enemy, we have been misled by the media into thinking that a group of enemies is really our ally. If our (supposed) evolved mechanisms have kicked in to provide us with a distortedly flattering view of the virtues of these people, then by parity of reasoning we should strive to rein in our moral imagination; we should overturn the distorting influence of flattering sympathy towards these individuals who are in fact not our allies at all.

Well, the thoughts expressed in both the last two paragraphs seem plausible enough. Who could deny the platitude that we should see our friends as friends and our enemies as enemies (and cats as cats and dogs as dogs)? And note that the pragmatic asymmetry has disappeared: Neither kind of inaccurate thought could be claimed to be benign. Seeing our friends as enemies (which is Wright’s concern) deprives us of understanding that could be enormously useful. But seeing our enemies as friends is, if anything, even more dangerous.

I should like to close with a fairly obvious thought, but it seems to me an important one. Remember that the notions of “friend” and “enemy” that are relevant here are to be defined in game theoretical terms: based not on the intentions of the parties in question but on whether they stand to benefit or suffer from the other’s pursuit of its interests. The worry is that any claim of the form “We in the West stand to benefit/suffer from Muslims pursuing their own interests” is just far too simplistic to warrant endorsement. The idea these two complex, sprawling, nebulous entities are playing one grand game is a highly doubtful proposition–even allowing for a dose of idealization. They are playing a myriad of games at many levels: numerous non-zero-sum games (that are potentially zero-sum games), many zero-sum games (that are potentially non-zero-sum games). The globalized economy ensures that the costs and benefits traded among nations and cultures is of a magnitude of complexity that challenges (and in all probability defies) comprehension. Merely drawing our attention to one interaction that appears to be non-zero-sum (that “what’s good for Muslims broadly is bad for radical Muslims … [making the West] more secure from terrorism”)—an interaction, moreover, that appears more potential than actual, since Wright’s principal complaint is that the West is not acting in a way that’s “good for Muslims”—is insufficient evidence that the relation between these two parties is in general characterized by non-zero-sumness.


Cacioppo, J.T., et al. 1993. “Rudimentary determination of attitudes: II. Arm flexion and extension have differential effects on attitudes.” Journal of Personality and Social Psychology 65: 5-17.

Li, W., et al. 2007. “Subliminal smells can guide social preferences.” Psychological Science 18: 1044-1049.

Schnall, S., et al. 2008a. “Disgust as embodied moral judgment.” Personality and Social Psychology Bulletin 34: 1096-1109.

Schnall, S., et al. 2008b. “With a clean conscience: Cleanliness reduces the severity of moral judgment.” Psychological Science 19: 1219-1222.

Also from this issue

Lead Essay

  • This month’s Cato Unbound features an essay drawn from The Evolution of God, the ambitious new book by Robert Wright, author of Nonzero and The Moral Animal. In this essay, Wright explores the relationship between “moral imagination” and the possibility of religious tolerance and social cooperation. Wright argues that moral imagination is part of our evolved mental machinery. When we see others as potentially cooperative, moral imagination is awakened to better grasp the needs and interests of partners and allies. But when we see ourselves caught in a zero-sum game with others, moral imagination, and thus sympathy and the spirit of toleration, shrinks as we prepare for a fight. Wright argues that the widespread perception that “the West” and “the Muslim world” are playing a zero-sum game is an illusion created by a misfire of moral imagination. The media’s relentless focus on the truculent acts of a small minority of Muslim extremists encourages the sense that the larger, more moderate Muslim world is much more hostile than it really is. But this sense narrows moral imagination, making it harder still to grap the possibility of cooperation and the point of toleration.

Response Essays

  • In his reply to Robert Wright’s lead essay, philosopher Richard Joyce, author of The Evolution of Morality, emphasizes the distinction between potentially and actually engaging in mutually beneficial cooperation. That “the West” could be in a non-zero-sum game with the “Muslim world” doesn’t imply it is actually in one. Moreover, Joyce argues, “non-zero-sum” and “good” do not mean the same thing from the perspective of an individual’s or group’s interest. If there is gain to be had from conflict, reason may recommend it. Tolerance and understanding are wonderful, Joyce agrees. But he finds something “unsettling” and “morally troubling” in what he takes to be Wright’s “[attempt] to justify these attitudes purely by an appeal to self-interest.” There are psychological limits to what appeals to self-interest can accomplish, and the congruence of self-interest and cooperation is far from certain in many cases. Additionally, Joyce suspects that Wright may be guilty of a weakly supported conjecture when he posits an evolved adaptation for distinguishing between zero-sum and non-zero-sum games.

  • Timur Kuran, Professor of Economics and Political Science and Gorter Family Chair in Islamic Studies at Duke University, finds insight in Wright’s account, but argues that it is insufficient to really explain the sense of conflict between many Muslims and the West. Kuran argues that displays of hostility in conformance with local expectations and social pressures can pay off handsomely. An expansive sense of possible of positive-sum relations with distant others does nothing to change the incentives that arise from collective processes at the local level. Not even suicide bombers require a false picture of zero-sum conflict. They may martyr themselves simply to bring status to their families. Wright’s neglect of the such alternative causes of cultural conflict, Kuran argues, leads him to offer advice of limited value.

  • Jonathan Sheehan, associate professor of history at the University of California at Berkeley, touches base with Blaise Pascal and reminds us that explicit game theory was first deployed as a religious argument aimed at conversion. In secular terms, a convert is a gain for one sect but a loss for another. But in religious terms, as Augustine noted, even the harsh coercion of heretics can be viewed as non-zero-sum–the heretic, whether he thinks so or not, has Heaven to gain. So, Sheehan argues, “the real stakes of the game do not matter. Or, more precisely, the nature of the game is the real stake.” To characterize the game as in fact non-zero-sum, as Wright does, is to miss the real moral and political issue about how the stakes of the game will be determined in the face of deep disagreement about what the game is. “Modern conflicts between ‘the West’ and ‘the Muslims,’” Sheehan concludes, “have less to do with misfiring mental machinery, and more to do with the absence of any recognized authority for determining the kinds of games we are playing, and which interests should count in them.”