A few days ago, the latest issue of The Philosophical Review arrived (yes, I actually subscribe to the print edition), and I saw Anthony Skelton’s review of the third volume of Terence Irwin’s gargantuan The Development of Ethics. (The three volumes, published between 2007–2009, amount to some 2500 pages!) Although I was aware of the existence of these books, I knew nothing specific about their content. I was gratified to learn from Skelton’s review that one of Irwin’s major aims in these books is to make a historical exploration and defense of what he calls “Aristotelian naturalism,” the teleological, eudaimonist, realist view which “identifies virtue and happiness in a life that fulfills the nature and capacities of rational human nature” (Irwin 2007, 4). The Development of Ethics traces the fortunes of Aristotelian naturalism from its first articulation by Aristotle through 2300 years of philosophical dialectic. Since I would count myself as an Aristotelian naturalist, this makes Irwin’s project interesting to me (though where I would find the time to read a 2500 page work of philosophy I have no idea). I was struck by Skelton’s casual description of Aristotelian naturalism as a form of egoism (2015, 280). I would agree that it is, but I think of this assessment as being at least somewhat controversial. Bernard Williams in Ethics and the Limits of Philosophy, for example, insists that Aristotle is no egoist. I don’t find Williams’s comments persuasive, but the point is that the question is arguable. All this started me thinking about Aristotelian egoism and its rationale, and led me ultimately to a startling problem for Aristotelian egoism. The problem is startling, to me anyway, because I had thought that Aristotle’s fundamental argument for his conception of the human good is essentially egoistic and could not be otherwise. I have also long thought that no system of ethics can be anything but egoistic if it is to have a ghost of a chance of being true. To see such longstanding views seriously undermined is startling, but it is also refreshing and rewarding to clarify and deepen one’s understanding of one’s views. Let us see in what way Aristotle is an egoist, what his argument is for his view of the human good, and where I now see a problem for his egoistic conclusion. Egoism is the view that the only reason to do anything ultimately is to confer some benefit on the agent. This rules out, as reasons for action, such things as that God said, that your mother said, that it’s the law, that it’s just the right thing to do, and that it’s required by social norms or intuitions. That is, these are ruled out as ultimate reasons. The mere fact that your mother said you should do something is not a reason to do it, according to egoism. Of course, if you want to please your mother or if you want to avoid being punished by her or if you think she has good judgment and has your best interests at heart, then her say-so can become a reason indirectly. But then her say-so is not your ultimate reason for acting. By this standard, Aristotle is an egoist. Along with every other Greek philosopher so far as I can see, he simply takes for granted that one should act to promote one’s own good and has no other reason for acting. This shows up in his eudaimonism. After arguing in Book I of the Nicomachean Ethics that reasons for action are structured teleologically and all aim at a grand, final end, he declares that it is uncontroversial that the final end is happiness. It is clear that he means the personal happiness of each agent. The difficulty, he says, is to know precisely in what happiness consists, and he proceeds in the remainder of Book I—and really in the remainder of the NE—to develop his eudaimonistic conception of happiness. Of course, Aristotle is not one of those bad egoists like Epicurus who have trouble explaining why you shouldn’t lie, cheat, and steal. Again along with every other Greek philosopher—except Epicurus this time—Aristotle is a good egoist, the kind who doesn’t have this problem. He and they avoid it by including virtuous action as a constitutive element in happiness. You can’t be happy by lying, cheating, and stealing, because to do these things is already to wreck your happiness. What distinguishes Epicurus and his numerous modern successors is that they identify the human good with something other than virtue, something like pleasure or long life or physical well-being or desire satisfaction. Thus they make virtue only instrumentally good. Since the human good (the final end) is, say, pleasure, everything else is good only to the extent that it is useful for obtaining pleasure. As a result they have a problem explaining why one should still be virtuous even in circumstances where one could get more pleasure by being vicious. I believe it is because Aristotle is a good egoist that Williams doesn’t want to allow that he is an egoist at all. If so, I think this is misguided. Egoism should be defined in terms of what is fundamental to it—the primacy of self-interest—not by whether one has trouble explaining why we shouldn’t lie, cheat, and steal when we can do so to our advantage. How does Aristotle link virtue with personal benefit? How does he derive his conception of the human good? He does so by the famous ergon argument of Book I, chapter 7. Basically the argument is that whatever has a function (in Greek, ergon) thereby has standards of its good built in to the function. The function of a flute player is to play the flute; a good flute player is one who plays well. The function of the eye is to see; a good eye sees well. Now, if a human being per se also has a function, then we can similarly derive standards of what makes a good human being. Aristotle decides that the distinctive function of the human being is reason (since it is what most fundamentally distinguishes us from all other creatures) and accordingly that the human good lies in the excellent active employment of the rational faculty. This is all pretty abstract. As Aristotle proceeds, it develops that what he is recommending is that one live one’s life through the constant, excellent, active employment of reason, letting it penetrate all areas of conduct, not just overtly intellectual areas like learning and reasoning and deliberating, but areas having to do with the passions and emotions as well. Passions and emotions cannot easily be controlled directly, of course, but we can train ourselves by repetition and exercise to develop habitually appropriate emotional responses. This is the core idea of his theory of the character virtues, such as courage, moderation, liberality, and even temper. When a person has and exercises these character virtues as well as the intellectual virtues, he has everything: appropriate action comes naturally; it feels good to do the right things; right action leads as a rule to material success, health, and well-being, but even when it doesn’t the happy person is content with the path of decency that reason dictates; he is both admired by others and comfortable in his own skin; in a word, he flourishes. The details of Aristotle’s conception of the human good are less important than the structure of his basic argument for it: The good of a thing that has a function consists in its performing that function well; biological organisms are functionally organized; so their good is to function well. We ought in principle to be able to identify the good functioning of an organism empirically, by analyzing its functional organization and operation. At a gross level, the analysis is intuitive. We know pretty well without training how to spot a thriving flower or tree in the garden. Likewise in the case of our bodies, the concept of health is precisely of this functional, empirical sort. For Aristotelian naturalism, the flourishing of a good person is like the health of a good body. Obviously there are many objections that can be made to all this and many matters of detail to address. I am just outlining the basic ideas here, so I can get on with my problem. This is a blog. I don’t imagine I’m writing a treatise on Aristotelian naturalism. Though if anyone has a particular bone or two to pick with any of this, that could make for good discussion. But there is one issue I do need to mention, concerning the status of functions. They need to be real. For the good of a thing to be derivable from its function, there needs to be a function that it has. This is controversial. Since the work of Larry Wright and Rob Cummins in the mid-1970s, it has become legitimate to take functions with ontological seriousness, especially in biology. According to this view, when biologists say that the heart is for pumping the blood, the eye is for seeing, the wing is for flying, they and we can take it literally. There are scoffers. John Searle comes to mind. On the other hand, both Ruth Millikan’s and Fred Dretske’s theories of cognitive semantics are rooted in this idea, and they have not exactly been laughed off the stage. I propose not to worry too much about this. Whatever the exact ontological status of functions, our empirical investigation of them has substantial objective constraints; that is probably enough reality for the purposes of Aristotelian naturalism. One very helpful constraint in the case of biological functions comes from the Darwinian theory of natural selection. And here at last we come to the problem I see for Aristotle’s egoism. If a trait evolved because it brings about a certain result in the life of an organism, a result which would not exist without that trait, that is evidence that the trait has the function of bringing about the result. If the eye evolved—came to exist—because of the information about the distal environment it supplied to organisms, which they would not otherwise have had, that is evidence that the eye is for supplying information about the distal environment. Wright actually makes this criterial for something having a function. Cummins does not. Either way, it is at least evidence of functionality. Most traits evolve by enhancing the fitness of individual organisms. Of two primeval flatworms, the one with the proto-eye will on average survive and reproduce more than the one without. This is individual or within-group selection. Since the 1960s, it has been firmly believed to be the only kind. If you read Richard Dawkins, that is what he will tell you. But Darwin didn’t think so, and contemporary opinion is no longer so uniformly against it as it was in the 1970s and 1980s. In particular, it is beginning to be recognized that natural selection also operates at the level of groups. For this to happen, it is necessary that groups compete as groups. But this does happen. In the human case, for example, two tribes may fight over the same foraging territory. Which tribe will be more likely to win this fight, the tribe whose members exhibit solidarity and discipline or the tribe whose members are looking out for number one? Clearly when tribes are at war, it is better to be a member of the tribe whose members exhibit solidarity and discipline. To be a member of the other tribe is to be doomed to destruction no matter how personally big and strong and brave one is. Thus where group selection pressure is significant, traits like solidarity and discipline will spread through the population. The counterargument is that although it is better to belong to the group whose members exhibit solidarity and discipline than to belong to the group whose members don’t, what is still better is to be a free rider in the former group; that is, best of all is to be surrounded by tribe members who exhibit solidarity and discipline but not to exhibit these traits oneself. To be a member of a tribe full of heroes but to be careful to let one’s other fellow members be the heroes. But this isn’t really a counterargument. It is only a statement of an opposing selective force. Group selection pressure, such as tribal warfare, selects for pro-group traits like solidarity and discipline; individual selection pressure selects for selfish traits like abandoning one’s fellows when the going gets dangerous. Both are always operating, and each tends to drive out the other. If there are never any wars, then selfish traits will inexorably spread through the population by individual selection pressure in the way just described. But if wars are frequent, they will tend to be won by the group with the most robust pro-group membership, and pro-group traits will spread through the population at the expense of the selfish ones. Which process will predominate, group selection or individual selection, depends on conditions and can change with conditions. That’s all I will say about this interesting topic. To find out more, and for a thorough and convincing argument for the reality of group selection in case you’ve read Dawkins lately and don’t believe it, see two papers by David Sloan Wilson and Edward O. Wilson, “Rethinking the Theoretical Foundations of Sociobiology” (2007) and “Evolution ‘for the Good of the Group’” (2008). Henceforth let us accept for the sake of argument that group selection operated (and operates) in human evolution and that we have pro-group traits of some kind. It doesn’t really matter for my purposes what they are exactly. I suggested “solidarity” and “discipline” without defining them. Whatever the traits are, they will be ones that economists tell us are irrational, like tipping in restaurants and voting in political elections and punishing wrongdoers. (Need to be reminded why punishment is irrational? Suppose you have been assaulted by some random person you are unlikely ever to meet again. Then the harm is done and will not be undone by having the malefactor spend time in jail. True, if he commits assault and gets away with it, he’ll be encouraged to do it again. But almost certainly not to you, so you have no interest in discouraging him. The rational thing is to save your time and effort: forget it and let his next act of aggression be someone else’s problem.) I hope the problem with Aristotelian egoism is now coming clear. The ergon argument says the good consists in functioning well. This means that our functions as human beings, whatever they are, set the terms of what makes us good human beings. This argument is safe as long as its conclusion is restricted to what makes us good. This is simply the logic that says if the function of a flute player is to play the flute, a good flute player is one who plays the flute well. The trouble is that, as the argument is employed, it goes further. It draws conclusions about what makes us happy. About what makes for our well-being. About what is good for us. From the putative fact that we have the function of reasoning, it is concluded that a good human being reasons well. That is the safe part. But it is also concluded that it is good for a human being to reason well. This assumes that what makes us good instances of our kind is also good for us. But for creatures with pro-group traits, this is not necessarily true. The honey bee that stings an invader and thereby kills itself is being a good honey bee. (“Do be a do-bee.” Sorry, I couldn’t help it.) But its action is not good for it! It costs it its life. That is the nature of pro-group traits; they are good for the group, not the individual. Aristotelian naturalism takes for granted an individualistic metaphysics of human beings that group selection theory implies is false. If it were true, then we could indeed conclude from the fact that something makes one a good human being that it is good for one. This is the implicit premise of eudaimonism: that to be a good human being is to thrive, to flourish, to be happy, to function well as an individual. But with pro-group traits, this is not necessarily any truer of human beings than it is of honey bees. Time to wrap up. As should be clear, I don’t see the ergon argument as the problem. I believe it is sound. The problem is that an unstated assumption of metaphysical individualism—the assumption that all our human traits are pro-individual—led to the (as it turns out) unwarranted conclusion that the ergon argument supports eudaimonism. Well, largely it does, of course. We aren’t honey bees and don’t have particularly many pro-group traits. But I believe we have some, and they are important. To the extent we do, eudaimonism is false. More amazingly, egoism is false. We actually have a reason, in the ergon argument, to do something that does not benefit us. What I would say in all earnestness to a honey bee, if it could deliberate about its actions, is that the most important thing in life is to be a good honey bee. To be a scurrilous honey bee who lets some other worker sting the invader is to live a bad life as a honey bee. I hope there is something intuitive about this. To run away from the fight to save itself is to be a bad bee. I think that is objectively true. To the extent that we have pro-group traits, it turns out, to my astonishment, that it is true (in a much more limited way, of course) for us too. WORKS CITED
- Cummins, Robert. 1975. “Functional Analysis.” Journal of Philosophy, 72: 741-764.
- Irwin, Terence. 2007. The Development of Ethics: A Historical and Critical Study, Vol. 1, From Socrates to the Reformation. Oxford University Press.
- Skelton, Anthony. 2015. “Review of Terence Irwin, The Development of Ethics: A Historical and Critical Study, Vol. 3, From Kant to Rawls, Oxford University Press, 2009.” The Philosophical Review, 124: 279–286.
- Williams, Bernard. 1985. Ethic and the Limits of Philosophy. Harvard University Press.
- Wilson, David Sloan and Edward O. Wilson. 2007. “Rethinking the Theoretical Foundations of Sociobiology.” The Quarterly Review of Biology, 82: 327–348.
- ———. 2008. “Evolution ‘for the Good of the Group’.” American Scientist, 96: 380–389.
- Wright, Larry. 1976. Teleological Explanations. University of California Press.