Response to Edward O. Wilson's, "The Biological Basis of Morality," _The Atlantic Monthly_ April, 1998 by David Mertz No Copyrot April 1998 I. NATURALISTIC FALLACIES, REDUX. ------------------------------------------------------------------------ The fundamental problem with Wilson is that he can conceive of far too few things. Explanations and understandings come at a great variety of levels, and in a great variety of guises. There seem to be two crucial aspects of such plurality which Wilson misses. One simple insight of most moderately sophisticated philosophy of science is that sciences explain in myriad orthogonal levels. The true stories told by physics (whether by today's physics, or by some future physics which corrects today's mistakes) neither contradict nor affirm the true stories of sociology, or even chemistry (themselves, of course, subject to the same fallibilist caveats). One type of truth may speak at one level of explanation, and another truth at quite a different level. For a long time, the reductionist project epitomized in Lavoisier's remark of "tell me only the position and velocity of each particle in the present, and I can predict all future events" held sway in the hopes of scientists. That is, it was a common belief of the 18th century (and some of the 19th) that every scientific explanation could be given the solid ground in being a consequence of the fundamental truths of the basic stuff of the universe. As remarked below, Wilson seems not really much to have left 18th century hopes behind, albeit with biology serving as a higher level proxy for an ultimate physics. At a somewhat different level-though still not precisely that at which I will criticize Wilson-philosophers are in the habit of distinguishing between a "context of discovery" and a "context of justification". In simple language, what historically, psychologically and accidentally leads to a particular scientific hypothesis is not at all the same thing that makes a hypothesis true. A particular thought process of a scientist, for example, may be interesting to that scientist's biographers, or even generally to historians and philosophers of science, but the hypothesis itself, once committed to concrete enunciation, must still stand by its empirical verifiability. The frequent subtext of this distinction is a defense of scientific explanations against the critics of scientists. For example, particular scientists may be nasty and small-minded types who develop their scientific theories out of unpleasant prejudices; but the theories themselves, once they come into this light-of-day no longer carry a taint of their origin. A simple concrete example is Karl Pearson's pivotal work in statistics (the correlation coefficient or the chi-square, for example). Pearson himself was motivated in part by a wish to prove the innate inequality of human intelligence (with a bit of a racial bent here). He was wrong in his motives, and the correlation coefficient is just as sound for that. With this digression, let's return to where Wilson goes astray. His wish is to "[search] for an origin of ethical reasoning that can be objectively studied." All fine and well, as far as it goes. Wilson's own explanation leans more heavily on biology than I would wish, but the general framework is unassailable: ethical beliefs evolve out of the nature of the human organism, and through the cultural and social development of such biologically specific beings. Wilson characterizes this, Strong innate feeling and historical experience cause certain actions to be preferred; we have experienced them, and have weighed their consequences, and agree to conform with codes that express them. Let us take an oath upon the codes, invest our personal honor in them, and suffer punishment for their violation. Certain romanticist fantasies aside, no one could seriously deny the general outline of this explanation. The only problem for Wilson is that his explanation has not the slightest relevance to ethics. The biological or historical origin of ethical belief says nothing one way or another about its veracity or moral weight. Yes, we believe such-and-such because of our genes (for example), but that doesn't tell us anything about the moral correctness of the belief. One can easily enough see the connection (and the differences) here with the context-of-discovery versus context-of-justification distinction. Humans have discovered ethics, in a sense, because of our evolutionary history; but our first obligation in evaluating ethical systems is to disregard-or at least bracket-this origin. Contra Wilson, you just can't get an ought from an is! An easy way, to my mind, of seeing the problem, is by moving it slightly over. Let us ask, for a moment, not about ethical beliefs, but about mathematical ones. There is no question that human's ability to engage in higher mathematics has roots in our particular biological and cultural evolution. Certain Darwinian demands upon our distant ancestors gave us the kinds of brains which, after a few thousand years of cultural shaping and pressures, would produce the theorems of number-theory, logic and calculus. A number of interesting stories could be told here, some of them perhaps true, all of them ultimately subject to the weight of empirical testing, about why humans, but not other animals, have written theorems. Perhaps certain social relations of hominids required a kind of abstraction from individual family relations which ultimately led to a general capability for abstraction. Or perhaps the manual dexterity of humans, and their early tool use, required an abstract thinking about parts which could make tools. Within an historical, rather than evolutionary, time frame, we might probably make some remarks about the mathematical requirements of societies which produce and distribute agricultural surpluses. Whatever the stories Wilson or I initially imagine, some such stories are no doubt true. It just is not, however, the causal history of human mathematical thought which makes Wiley's proof of Fermat's Last Theorem true! What makes the proof true, is the soundness of its steps and the coherency of its conceptual objects. In a Kantian sense, any thinking being able to comprehend Wiley's proof would be equal in establishing the veracity of the proof. Such beings might be angels or pixies, or they might be the odd life forms evolved on Mars or Ganymede according to quite different evolutionary pressures. It is the lifestyle of primitive hominids which precipitates the ability of humans to develop the proof, but the proof is equally veridical to any thinking being, whatever its particular evolutionary history. Just as much as evolution (biological or cultural) puts humans in the habit of realizing mathematical truths, it also puts us in the habit of believing some mathematical falsities. Some interesting work on conceptual mistakes (or a very roughly "mathematical" sort) has been done by Tversky and Kaneman. One fairly good example is a story "problem" like the below (told in rough paraphrase from memory): Jane lives in an upper-middle class suburb of Chicago. She is a board member of the Chicago ACLU, and volunteers on weekends at a drug-rehabilitation center. By occupation, is Jane more likely to be: - (a) A lawyer - (b) A doctor - (c) A criminal defense lawyer - (d) An accountant The punchline to this example is that more respondents respond to such a "problem" with the answer (c) than with the answer (a). However, given that (c) is a strict subset of (a), and the question is of probability of set membership, it is simply a mistake of mathematics to select (c) given the range of options. The explanation which Tversky and Kaneman give for this type of thinking concerns humans' tendency to cluster ideas together as part of cognition. This is sometimes called stereotypic thinking (without the negative connotation the word stereotype has in other contexts). Individual concepts are not thought individually, but rather in series of relations of proximity with other ideas. The concepts of "ACLU board member", "volunteer at a drug-rehabilitation center", and "criminal defense lawyer" are ones many people hold fairly close together as part of a stereotype. Certain attributes we find in an entity (including an hypothetical one) trigger associations with other attributes, and affect our judgments about the likelihood of those other attributes. At times, this way of thinking leads to plain logical mistakes. It is easy enough to imagine an evolutionary story of human mental evolution which traces conceptual thinking from cruder to more sophisticated abilities to group and associate like ideas together. This associative ability might well have served the competitive advantage of hominids the greater its development. In fact, it might well be this sort of conceptual development which, in a general sense, allows humans to think all the complex thoughts about mathematics, logic and science which we do. On the other hand, under this scenario, it is this very same developmental history which inclines us to a certain class of conceptual errors. Still, for all that, however deeply built into our genes such fallacious tendencies might well be, they are still errors. And so it is with ethics. And so also we can see the relevance of orthogonality. Ethics (or mathematics) just simply explains at a different level than does its causal history. It is not uninteresting to inquire into the causal history of ethics in humans, but when one does so one is doing something different from ethics. To remark as I do is to take no stand on the proper basis of ethical thought-or indeed that it has one. Ethical dictates might have a transcendental basis and thereby operate at an orthogonal level to causal histories of ethics. On the other hand, ethical dictates might have strictly what Wilson oddly calls an empiricist basis[1] and likewise be orthogonal to their causal histories. Carrying the matter further, ethics need have no real veridical basis at all to remain orthogonal with the causal history of ethics. The truths of astrology are few to none, but however few the truths be, the statements of astrology should not be judged by the causal/evolutionary history of astrological thought. By that standard, we should little be able to distinguish astrology from behavioral psychology, both certainly originating in a facility to generalize and comprehend human behaviors. Astrology (and perhaps ethics) is false, but Wilson stands in the wrong position from which to make this judgement. --------------------------------------------------------------------- [1] A better word would probably be constructivist', or perhaps 'positivist' (in the sense of legal positivism). Empiricism, to me, always seemed to denote the doctrine that concepts have their whole basis in precepts, and derivatively that empirical scientific inquiry is the best way to go about acquiring knowledge. It's opposite is rationalism (precepts coming out of concepts), not transcendentalism. Transcendentalism, at least as Kant has it, would try to walk the line by intoning that "concepts without precepts are empty, but precepts without concepts are blind." Of course, 'empiricism', with its modern resonances of brass-tacks science, is a word with which Wilson would rather claim an association. In any event, empiricism proper, as near as I can tell, should be rather agnostic about which side of Wilson's ethical transcendentalism fence to fall. II. JUST-SO STORIES. ------------------------------------------------------------------------ Wilson's evolutionary tales always seem to have a bit of Kipling in them. One feels in reading his human evolutionary scenarios that, yes, the story is possible. But one never at all feels that Wilson's stories are necessarily true ones. Wilson's knowledge of anthropology, archaeology, and hominid paleontology are simply that of an amateur with an expertise elsewhere. He can speculate about hominid evolution, just as I, or anyone, can speculate about hominid evolution. But outside of the ethology of social insects, Wilson's stories are nothing special. Wilson's just-so stories are actually probably forgivable for a popular journal like The Atlantic Monthly. Here they can be taken as illustration only. The problem is that Wilson's "scientific" works are just as rife with such imaginings. Still, let's take a look at a rather crucial story within the article under discussion. It plays a somewhat pivotal, albeit somewhat covert, role in Wilson's overall argument, Imagine a Paleolithic band of five hunters. One considers breaking away from the others to look for an antelope on his own. If successful, he will gain a large quantity of meat and hide-five times as much as if he stays with the band and they are successful. But he knows from experience that his chances of success are very low, much less than the chances of the band of five working together. In addition, whether successful alone or not, he will suffer animosity from the others for lessening their prospects. It's a nice story. It explains why people should have evolved genes for cooperation and sociability-traits which are sometimes exhibited in humans, and of which I, like Wilson, approve of and would wish to encourage. On the other hand, here is another story with about as much evidence from the archeological record, Imagine a Paleolithic band of five hunter/gatherers. They hunt for small animals like rabbits which are relatively plentiful and gather nuts and berries from wild plants. One considers breaking away from the others to look for food on his own. He knows from experience that he is above average in his rate of gathering food and resources. By going on his own, he will no longer have to share his food and resources with those who gather them less well. In addition, by being successful alone, he will gain the respect and deference of the others for his great abilities. This is a less nice story, at least from the point of view of its "ethical" implications. Under a slightly different, but equally plausible, evolutionary scenario, the best-and-brightest of each generation of hominids will be attracted to solitary activity. The only members attracted to cooperation are those with the worst survival abilities-and this situation will certainly, over thousands or millions of generations, lead to the elimination of the genes of the "cooperationists." Neither of these stories is likely to be precisely accurate, nor certainly is either exhaustive. It is clear from the most basic observation that humans sometimes cooperate and sometimes compete. We are sometimes sociable and sometimes solitary. All these variations are certainly within the possible behavioral range of humans. Either simple story above-or any of the many like them-would conclude by logic that the behaviors we observe everyday could not be within the human bio-behavioral repetoire. So real stories must be a lot more complicated. The subtext of Wilson's article, and of his sample evolutionary story, is that humans in our genes have inclinations to act in ways which right-thinking people like Wilson and readers of The Atlantic Monthly find generally "ethical" to start with. Whatever its apparent logical structure, Wilson's article would not have been written or published if there were not supposed a certain consilience between the brand new ethics Wilson is "discovering" and the old-fashioned ethics which most of us educated Westerners generally believe already. If some biological story-say one with actual evidence behind it-told of the strong evolutionary pressures in favor of early hominids murdering each other at the slightest whims, Wilson would certainly not be claiming that doing as our genes "tell us" is the right way to act. The "discovery" only works if it tells us pretty much what we always knew. This is why Wilson's particular just-so story has to occur within the pages of his article. Let readers tell whatever story we might imagine ourselves, and who knows what awful conclusions a biological ethics might bring us to. But as long as the paleolithic hunters are pretty close in their behaviors and motivations to a boy-scout troop, all is well. The sort of crude Hegelian inversion of saying whatever is (in our genes) is what ought to be (in our moral systems) only seems sensible as long as Wilson provides the proper guidance in describing the is. RESOURCES ------------------------------------------------------------------------ The source Wilson article may be found at: URL: http://www.theatlantic.com/issues/98apr/biomoral.htm Links to a variety of subsequent related discussion is at: URL: http://www.theatlantic.com/issues/99nov/9911sciencewars.htm An interesting looking article by that I found by accident is, Steve Rosenthal, _How Science is Perverted to Build Fascism_. If nothing else, it is hosted at my alma mater, so I feel I should reference it. URL: http://www-unix.oit.umass.edu/~cscpo/eowilson.htm