Williams Huxley's Evolution
Williams Huxley's Evolution
Williams Huxley's Evolution
Let us understand, once for all, that the ethical progress of society depends, not on imitating the cosmic process, still less on running away from it, but in combating it. Thomas H. Huxley
Thomas Huxley first presented Evolution and Ethics as his Romanes Lecture in 1893. The published version and Prolegomena first appeared in 1894. They have been favorites among anthologists and
George C. Williams is professor of biology at Queens University, Kingston, Ontario K7L 3N6. This article is contribution 693 from the Department of Ecology and Evolution, State University, Stony Brook, New York 11794. It is a condensed version of a manuscript to be published by Princeton University Press in a proposed volume to include Huxleys Evolution and Ethics, associated Prolegomena and notes, and other materials. An earlier version was presented at a Symposium of the American Association for the Advancement of Science in May, 1985. T h e author completed most of the work on the manuscript while a Fellow at the Center for Advanced Study in the Behavioral Sciences at Stanford, California, in 1982. A number of people were kind enough to read it and offer advice. Foremost among these are Paul W. Sherman, David Sloan Wilson, and Douglas J. Futuyma.
383
384
ZYGON
have excited comment ever since. In my view most of the better known comments (e.g., Dewey 1900;Julian Huxley 1947; Waddington 1971) would have been better left unsaid. Their main errors have been ably rectified by D. Daiches Raphael (1958) and Anthony G. N. Flew (1967), who make it clear that Thomas Huxleys views were far more tenable than those of twentieth-century critics. I think Flew (1967, 50) fails in one respect. He accuses Huxley of going too far in recognizing a negative moral value in nature rather than mere moral neutrality. I also criticize Huxley on this point, but in the opposite way. I will attempt to show that Huxley did not go, far enough in his condemnation of the evolutionary process. We now have a far deeper understanding of evolution than Huxley had, and far more information on what this process has wrought. Huxley, for instance, understood nothing of forces favoring the development of nepotism or of parent-offspring and other sorts of conflict in nature. He knew little of the prevalence and violence of infanticide and cannibalism. There are several important themes in Huxleys essay. One deals with the evolutionary origin of human ethical motivations and attitudes, and what he calls the apparent paradox that ethical nature, while born of cosmic nature, is necessarily at enmity with its parent (Huxley 1894, viii). This topic is ably discussed by Peter Singer (1981) and Richard Alexander (1987), with whom I am mostly in agreement. Although his paradox is not a matter that can be ignored, I will deal here mainly with Huxleys more prominent theme, his moral evaluation of nature, succinctly summarized: Thus, brought before the tribunal of ethics, the cosmos might well seem to stand condemned. The conscience of man revolted against the moral indifference of nature, and the microscopic atom should have found the illimitable macrocosm guilty (Huxley 1894, 59). Condemnation and guilt are dubious verdicts for mere moral indifference. Huxley clearly implies something worse, and the inconsistency makes the statement vague and disappointing. I would prefer something more along these lines: Thus, brought before the tribunal of ethics, the cosmos stands condemned. The conscience of man must revolt against the gross immorality of nature. I attribute the defects of Huxleys version to two factors. The first is that despite his accomplishments Huxley was, after all, only an Englishman and no doubt afflicted with the English habit of understatement. Secondly his grasp of natural selection was inferior to that of Charles Darwin and a few other nineteenth-century biologists, (Leonard Huxley 1900,12,372; Mayr & Provine 1980,133). No one of
386
ZYGON
its genes to future generations to the maximum possible extent. This summary of current understanding of natural selection is, of course, grossly oversimplified, but perhaps adequate here. More detailed discussions are available (Dawkins 1976; Williams 1984; Williams n.d. In press). The principles of population genetics as developed by R. A. Fisher, J. B. S. Haldane, and Sewall Wright around 1930 clarify the process by which the genetic record is kept and elucidate the nature of that record. It is a store of information about what has succeeded at transmitting genes in the past and a prediction as to what will succeed in the future. An important implication of this advance, especially important for social behavior and its moral implications, was first recognized by William Hamilton (1964). T h e survival and reproduction of a relative are partly equivalent, in evolutionary effect, to ones own survival and reproduction. Hamilton proposed an expansion of Darwins concept of fitness to include any ability to get ones own genes represented in future generations. It does not matter whether these genes are present in ones own cells or present, for genealogical reasons, in relatives. He termed this broader concept inclusive fitness. His key formulation was that assistance to a relative will be favored if the benefit to the relative, times the coefficient of relationship, exceeds the cost to the donor. Consider the problem of maximizing the fitness of individual 5 (Figure 1). T h e distribution of her special set of genes in herself and her relatives is shown by the stippling. Her fitness is measured by her ability to get these genes passed on to future generations. A useful device is to consider such questions as: What is the probability that an arbitrarily chosen gene in the focal individual ( 5 )is also present in her father ( I )? In sexual reproduction half of ones genes come from ones father. So the probability is .5, and this is the coefficient of relationship of 5 to 1 . T h e gene would be favored if it tended in any way to make 5 behave towards I as if his genes were half as important as hers. Her fitness will be enhanced, up to a point, by a willingness to jeopardize her own well-being to increase his. The extent of the self-sacrifice favored would be determined by the .5 coefficient and the cost-benefit relationship, with costs and benefits measured as effects on the likelihood of transmitting genes. On the assumption of exactly equal prospects for future reproduction and an absolute certainty that I is 5s father, natural selection would favor her risking her life up to a .5 probability of loss, if she can surely save him by taking the risk, and if he will surely die if she does not. With conditions otherwise similar except that it is nephew 7 who is endangered, .25 would be themaximally tolerable risk.
FIG. 1-Hypothetical pedigree diagram, applicable to any outcrossed diploid organism with separate sexes. Males are represented by squares, females by circles, mating by horizontal lines connecting males and females, offspring by verticals from the horizontals. Stippling shows the proportion of the genes in each individual that are necessarily identical with those in 5 . H e r fitness is measured by her ability to pass these genes to future generations. TJnstippled regions represent random samples of genes o n which 5 can have only random effects.
388
ZYGON
be expected to react to hatchlings in his nest according to his likely coefficient of relationship with such hatchlings. If he is indeed their father, the coefficient is at least .5, but, as discussed below, sexual infidelity and nest parasitism may be common. A sociobiologist would expect his behavior to be closely determined by the average coefficient during recent history for males in his position. I will mention one more of the many possible misunderstandings. Even a complete kin-selection argument would still give a grossly incomplete view of the natural selection at work on social relationships. Kinship is merely one of many important factors. T h e individuals diagrammed in Figure 1 must vary in many ways, not merely in kinship to the one I picked for special attention. She may have established mutually beneficial reciprocation with others, some perhaps unrelated. Her mate (6)would be an obvious example. Her sister-in-law ( 3 ) is also unrelated, but even if she has no other significance she has value (to 5 ) if she is helping nephew 7. Individuals also vary in the amount of benefit they would realize from a given donation. A post-menopausal mother and adolescent daughter are equally related to each other, but it obviously makes evolutionary sense to be more generous to the girl than to the old woman. Questions such as why animals concern themselves so much more with their offspring than with equally related brothers and sisters are based on the assumption that kinship, to be important in evolution, has to be the only factor. A complete answer to such a question would have to consider the relative reliability, for adults of reproductive age, of available evidence for parenthood and that for other kinds of relationship; the reproductive potential, likely survival, and needs of recipients; needs and capabilities of prospective donors; alternatives for the investment of resources; and many other factors. RECENT EVOLUTIONARY THOUGHT THE MORALITY AND OF HUXLEYS COSMIC PROCESS
I t is only twenty-five years since Hamilton ( 1 964) elucidated kinship as a factor in the evolution of social behavior, and less than that since Robert Triverss (1971) insights on conditions for the evolution of reciprocity. Prior to these developments there was no consistent way of interpreting what animals were doing with each other. This absence of theoretical guidance and the outwardly benign appearance of some social interactions made it possible to paint a morally acceptable picture of nature. Romantically inspired naturalists found wholesome lessons in the mutual grooming of monkeys, the feeding of one bird by another, the brave self-sacrifice of worker bees for colony and queen. Such observations served as parables for moral guidance in a system
390
ZYGON
behavior. Full sibs are nicer to each other than half sibs. Mother-love seems to be greater for offspring that are more nearly mature and therefore more likely to mature, but younger mothers are expected and found to be less loving than older ones (Andersson et al. 1980; Bierman & Robertson 1981). Manipulation is the second reason why one individual may provide benefits for another. A mouse can provide nutritive benefits to a cat, and when handled by the cats paws is literally manipulated, but the term normally has a metaphorical meaning. For example, threat of physical force may substitute for force itself, as when a subordinate monkey relinquishes a feeding site to a dominant. More interesting kinds of manipulation result from deception, most obvious when practiced against another species. A snapper may swim eagerly towards the jaws of an anglerfish that deceives it with its lure. Social-insect pheromones that suppress reproduction by workers are interpreted as both a kin-selected response by workers and manipulation by the queen. Manipulation may imply exploitation by a manipulator, but it may be that communication within a species is usually adaptive for both sender and receiver. A hen may use vocal signals to get her chicks to behave in ways that serve her genetic interests, but obedience will usually serve a chicks interest. T h e same might be said for most verbal exchanges between human parents and their children. Richard Dawkins (1982) argues convincingly for the prevalence of manipulation in nature. The third reason for one individual acting in the service of another is reciprocity. Whatever is given up by the donor costs it less than some expected repayment. Neither the expectation nor the repayment need be conscious o r even behavioral. The secretions of an aphid may later repay an ant that protects it from a predator. In cooperative ventures the mutual benefits may be simultaneous. Two hyenas in a cooperative attack on a wildebeest calf defended by its mother are more than twice as likely to succeed as one acting alone (Gould 1982), and one calf can be a feast for several hyenas. A cooperative hyena may therefore eat while a loner goes hungry. In application to human behavior Darwin (187 1, 163) called reciprocity a low motive. Reciprocity in nature is strictly limited by the necessity of safeguards against cheating. N o matter how great the collective benefit might be from a cooperative venture, the necessary behavior will not evolve if the benefits can be enjoyed by freeloaders. It might be that a gregarious species could reduce its rate of loss to predators by having each individual warn the others with a loud call when it sights an enemy. This would in no way assure favorable selection for calling. If the call attracts the notice of the predator it may increase the callers likelihood
George C. Williams 391 of being the victim. This cost would not be borne by a silent freeloader that merely seeks its own safety on seeing a predator. It would survive better than the callers so that sounding the alarm would not be favored. This sort of reasoning has led students of animal behavior to prefer kin selection over reciprocity as an explanation for known alarm calls, and predictions based on the kin-selection model have been borne out in recent studies. Alarm calls of various sciurid rodents are given mainly by those age-sex categories most likely to have close kin nearby, and the actual presence of individuals with a coefficient of relationship of .25 or more (grandchildren, nephews, aunts) increases the likelihood of calling (Dunford 1977; Sherman 1980). Animals sound alarms to warn close relatives. If such benefits also benefit the group as a whole, that is an incidental consequence of selection for nepotism and has no bearing on the evolution of alarm calls in gregarious animals. It could work the other way around, with alarm calls being important in the evolution of gregariousness. If my neighbors shout at their children when they see a bear approaching, it behooves me to stay close enough to hear their alarm calls. T h e more such neighbors I have, the greater my safety. I should stay with them if I can, and encourage them to stay with me. More important is the selfish herd factor formulated by Hamilton (1 97 1). T h e advantage comes from maximizing competition for the bad things in life. If I were a Christian in the Flavian Amphitheater, I would want as many Christian companions as I could muster to share the attention of the lions. Perhaps the best technical term would be the St. Zgnatius Strategy, after the Bishop of Antioch who was the earliest on record to make use of this principle. Various groups of animals seem to use it more successfully than St. Ignatius did (Altman 1974; Kaufmann 1974). As a general rule today a biologist seeing one animal doing something to benefit another assumes either that it is manipulated by the other individual or that it is being subtly selfish. Its selfishness would always be defined in relation to its single ultimate interest, the replication of its own genes. Nothing resembling the Golden Rule or other widely preached ethical principle is operating in living nature. It could scarcely be otherwise. Evolution is guided by a force that maximizes genetic selfishness. Even if it turns out that natural selection among (as opposed to within) populations is more potent than is admitted by current orthodoxy, it would make little difference for a moral evaluation of Huxleys cosmic process. If some populations are consistently better than others at maintaining themselves and giving rise to new populations, they achieve their success by causing the extinction of less favored groups. To claim that this is morally superior to natural selec-
392
ZYGON
tion at the level of competing individuals would imply, in its human application, that systematic genocide is morally superior to random murder. There is no level of inclusiveness of selected entities at which the survival of the fittest is morally acceptable. The morally acceptable goal in relation to survival has to be the fitting of as many as possible to survive (Huxley 1894, 83).
George C. William 393 reduce nutritive reserves on which the mother depends for survival to the next breeding season. Thus successive young are ultimately in conflictwith each other,just like litter mates. Only mammals can have a weaning conflict, but analogous strife can be found in all groups in which parents provide services for developing offspring. Joanna Burger (1981) describes the lively dissention over termination of parental feeding of young in herring gull colonies. Conflict between mates o r potential mates is often bellicose. Recent accounts of reproductive behavior in wild animals are tales of sexual intrigue full of deception, desertion, strife, and sometimes lethal violence. Conflict arises because of basic physiological differences between males and females in most species (Trivers 1972). A female may need to mate with a male to produce offspring, but one mating may be enough to fertilize all the eggs she has o r all she can produce in a breeding season. Mating once with the best. male available can be a better strategy than mating once with the best and once with the second best. The female can be expected to try to optimize her choice of male and the timing, locality, and all other circumstances that might influence the ultimate success of the eggs to be fertilized. For the male, reproductive success may be largely a matter of how many females he can inseminate. He can be expected to make use of every available opportunity to fertilize eggs and to seek out or try to produce such opportunities. The common outcome is for males to spend much time and effort in mating attempts that females just as persistently avoid. A tank full of guppies is a scene of endless conflict between males and females. Such is the power of romantic self-deception that many people find such displays attractive. Mating conflict exacts a high cost for survival and well being in many species (Daly 1978). The possibility of cuckoldry gives another dimension to male-female conflict for species in which fertilization is internal and males contribute to the care of offspring. A female songbird may be fertilized, willingly o r unwillingly, by a neighbor rather than her apparent mate, or her mate may die or desert. Her best strategy is then to behave as if she has a full complement of fertilizable eggs and allow herself to be courted and won by a suitable male. He could still sire a fraction of her clutch but would waste part of his parental investment on another males offspring and raise fewer of his own than he otherwise might. The result is an evolutionary arms race between a females ability to deceive and a males ability to detect deception. Many details of courtship in birds have recently been interpreted as a dialog based on male suspicion and female representations of fidelity. An essential part of courtship may be a males effort to sequester and monitor his mate long enough to minimize the possibility of cuckoldry. Evidence for such
394
ZYGON
interpretations is provided by swallows (Beecher & Beecher 1979), bluebirds (Gowaty 1982; Power & Donner 1980), and doves (Lumpkin 1983). There may be benefits to a female in undetected adultery with one or more of her mates rivals. A female mouse can perhaps make the rival less likely to kill her young later on (Elwood & Ostermeyer 1984). Adultery can also be a safeguard against a mates sterility. If a male bird has a 10% probability of being partly o r entirely sterile, a strictly monogamous mate has a 10% probability of losing all o r part of her brood. Insemination by two males would reduce the hazard to 1%. That females often copulate with males other than their mates has been shown by surgically sterilizing males. About half of the mates of sterilized redwing blackbirds studied by Thomas A. Roberts and James J. Kennelly (1980) produced fertilized eggs, and similar observations have been made in many other species (McKinney, Cheng, & Bruggers 1984). These observations need not indicate adaptive bet hedging by females, a possibly costly tactic. Actual or suspected adultery may expose a female to rejection or even violent attack by her mate (Barash 1980), and it may be that much of the illegitimacy results from rape. Even a brief absence of her mate may make a female vulnerable to rape by neighboring males in the goose colony studied by Mineau and Cooke (1979). An unguarded female mallard may be attacked so persistently by gangs of males that she drowns (Barash 1977).Joanna Burger and C. G. Beer (1982) observed 162 males that succeeded in mounting unwilling females in a gull colony, and many more that did not succeed. Apparent rape has been documented in eighty-one species of birds (McKinney, Cheng, & Bruggers 1984). In some insects it would appear that females mate only as a result of male violence (Parker 1979),and in some sharks it appears that mating takes place only after injurious attack by males (Pratt 1979). Rape occurs in turtles (Berry & Shine 1980) and newts (Verrell 1982), and homosexually in a parasitic worm. T h e male victim may later find it difficult to fertilize a female with his own sperm (Abele & Gilchrist 1977). The concept of rape would most clearly include mating as a result of threat or actual violence by males. A more theoretically useful definition would include any circumvention of mate-choice mechanisms used by females. By this definition it would be of common occurrence in plants (Janzen 1977; Willson & Burley 1983). In animals it should include the use of deception and stealth. A subordinate male frog may hide near a calling dominant. A female attracted to the caller may be intercepted and mounted and have her eggs fertilized by the unwanted
George C. William
395
subordinate (Forester 8c Lykens 1986). Subordinate male sunfish may lurk about the nests of dominants, sometimes with the color and behavior of females. They may then dart into the nest of a spawning pair, release a cloud of semen, and flee any attack by the dominant male (Dominey 1980; Gross 1979). Similar phenomena in diverse species are reviewed by Anthony Arak (1984). Related phenomena occur in species with internal fertilization. For example, James Farr (1980) showed that by a combination of stealth and speed a male guppy may sometimes circumvent attempts at rejection by females. Even in species in which mating does not normally occur without female consent, an incapacitated female may be quickly inseminated. Male fruit flies mate more readily with anesthetized females than with active ones, and fish hybrids can be produced routinely by presenting males of one species with anesthetized females of another (Bowden 1969). Besides adultery and rape, just about every other kind of sexual behavior that has been regarded as sinful can be found abundantly in nature. Brother-sister matings are the rule in many species (Hamilton 1967). Masturbation is common in mammals (Beach 1964), and especially common in juvenile male marmosets whose mothers are in heat (Hershkovitz 1977). Males may mount obviously pregnant females or other inappropriate objects (S. A. Altmann 1962). Homosexual behavior is common in a wide variety of vertebrates (Beach 1978), and perhaps homosexual preference (Weinrich 1980). The killing of other members of the same species is a frequent phenomenon in a wide variety of forms and contexts. Simple cannibalism is the commonest and can be expected in all animals except strict vegetarians (Polis 1981). It is a general rule among fishes. Aquarium keepers find that speedy separation of young from adults is needed to prevent consumption of the young in many species. T h e phenomenon is widespread in nature, an extreme example provided by the walleye (Cuff 1980). Stomachs of large walleye contained smaller ones, which had eaten still smaller ones, for at least a fourfold cycle of cannibals within cannibals. Conspecific destruction can also take special forms in fishes, several shown by the mottled sculpin (Downhower 8c Brown 1981). Large males are better egg guarders than small ones, and females prefer them, but if there is too great a difference in size he may find her more tempting as a meal than as a mate. Egg guarding is a stress for males, because it does not allow productive foraging. The required fast is justified only if it results in a certain minimum of hatched young. If the male gets only a single clutch of eggs and no more for a few days, he may eat them and abandon the attempt to reproduce that year. If he
396
ZYGON
gets a succession of clutches from different females, and is sufficiently hungry after a long bout of egg guarding, he may eat the last batch instead of waiting for them to hatch. The eating of eggs or young in other individuals nests is common in many fishes (Keenleyside 1972; Rohwer 1978) and amphibians (Crump 1983). Sherman (198 1) showed that about eight percent of the young in a colony of ground squirrels are killed by members of their own species. A male may raid a nest to kill and eat one of the young. A female may raid the nest of a competitor and kill all the young but not eat them. Nearly half the litters in a prairie dog colony were victimized by infanticide, sometimes by close relatives (Hoogland 1985). Cannibalism among insects is common, often the eating of young or of mates o r potential mates (Buskirk, Frolich, & Ross 1984; Eichwort 1973; ONeil & Evans 1981). Destruction of the young of rivals can be a major source of mortality in birds (Picman 1977; Trail, Strahl, & Brown 1981). Destruction of one of a pair of young, with consumption of the victim, is frequent in birds of prey (Stinson 1979), and this killing may be by a brother or sister. Valerie Gargett (1978) recorded hundreds of attacks by a larger black eagle chick against a smaller nest mate over a period of three days, when the smaller finally died of its wounds and starvation. An authoritative compendium on infanticide in the animal kingdom is now available (Hausfater & Hrdy 1984). It is now generally realized that infanticide rates of some primitive human populations were sufficient for important demographic consequences (section 5 in Hausfater & Hrdy). Mammalian infanticide may evolve as a male adaptation to female reproductive physiology because lactation may inhibit ovulation. For a males reproduction the essential resource is an ovulating female. A female deferring ovulation while nursing another males young is only a potential resource. She can be changed from potential to actual by the death of the unweaned infant. For females such young are of highest importance because with a bit more investment they can be turned into self-feeding juveniles. S o the death of the young would be a major loss for the female but only a minor gain in time for the male. Unfortunately such issues are not settled by any principles of justice or net costs and benefits, but strictly on a might-makes-right basis. Hrdy (1977a; 197713) describes dramatic examples of this sort of male-female conflict for a monkey that often lives in groups of related adult females, their young, and one temporarily dominant male. Males keep such harems only so long as they can avoid being deposed by rivals. Sooner or later some rival will succeed in defeatingany currently privileged male and in taking over his mates. When this happens there
George C. Williams
397
is an immediate conflict between the new male and any female with an unweaned infant. T h e sooner the infant stops nursing, the sooner its mother will turn into a valuable resource, and the only way to get it to stop nursing is to kill it. This is not always an easy task. The male is bigger and stronger than the female, and can kill an infant with one efficient bite into the skull, but he may faildespite repeated attempts. A females motivation to protect her infant may be greater than his desire to kill it, and it may be necessary for him to fight both the mother and one or more other relatives of the infant. Members of the harem are closely related, and a grandmother or aunt or older half-sister has half as much at stake in the infants survival as the mother has. Kin selection results in the formation of coalitions of related females in opposition to infanticidal males. In the absence of ideology, only nepotism can produce such a coalition, and never anything analogous to feminism or an urge for justice or group welfare. If infanticide by a male raises his fitness even slightly, it is in a mothers interest to have her son practice it if and when he succeeds in taking a harem from some other mothers son. There is no way for the female sex to constitute a group for which group selection could produce any kind of change. If she loses her infant, the mother quickly comes into estrus and accepts her infants killer as the father of her next offspring. Many conspecific killings result from contests over resources, a common event among insects (Parker 1979).T h e disputed resource is most often a female ready to reproduce. I n large mammals with horns or antlers, death or debilitating injury from fighting over females may claim five to ten percent of adult males every breeding season (Clutton-Brock et al. 1979; Wilkinson & Shank 1976). Death from strife among neighbors tends to be recorded for any wild animal population carefully observed for a thousand hours or more (Wilson 1975).Consider for comparison the annual homicide rate of 0.0003 for Houston, apparently the most murderous of major American cities (World Almanac and Bookof Facts 1983,966). It would be necessary to keep ten Houston residents under continuous observation for three centuries to make it likely that one murder would be seen. Males of our own and most other mammalian species are more likely than females to kill o r injure others of their own kind, but competition among females may not be so much of lesser intensity as of greater subtlety. They commonly deprive each other of resources in various ways, aggressively interfere with each others courtship, attack each others young, and actively aggravate male competition when it serves their interests (Silk et al. 1981; Wasser & Barash 1983). Losses from sexual conflict may go far beyond direct results to participants. Males in combat may accidentally injure or kill females or
398
ZYGON
young in such diverse animals as seals (LeBoeuf 1974) and dung flies (Parker 1979). Bright colors useful in attracting females or intimidating rivals may attract predators. This is certainly true in many fishes (Endler 1978) and insects. Elaborate sexual and competitive behaviors also increase vulnerability either by their conspicuousness or the inattention of participants. George Schaller (1972, 243) found that predation by lions was especially high on fighting warthogs or courting reedbucks. Merlin Tuttle and Michael Ryan (1981) noted that a predatory bat used frogs mating calls to locate such prey. Other losses are less dramatic but real. In many species a large size is advantageous for males in winning females, but is deleterious in other respects. Myron Baker and Stanley Fox (1978) found that larger male grackles were especially vulnerable to a chemical eradication technique. Such observations support the common assumption that females more closely approximate the engineering optima for a species but that males compromise these optima with requirements of sexual competition. Adult male modifications may go far beyond scaling for larger size. Antlers and horns are nutritionally expensive structures, and there may be major structural modifications related to the offensive use of these weapons and to withstanding violent collisions with rivals (Schaffer 1968). Russell Landes (1980) calculations support the recently common assumption that increased effectiveness in sexual competition can increase the likelihood of extinction. The consistently higher human male rates of mortality and morbidity measure the price of features useful in sexual rivalry. Many of the unpleasant phenomena reviewed above have been described only recently. They are likely to be impressive only if unexpected as a result of the blindness of romanticism. None are really needed for the argument being advanced. The inescapable arithmetic of predation and parasitism should be enough to show that nature is morally unacceptable. This should be so even if some romantic fictions were really true, which they are not. It is sometimes claimed, for example, that predators take only what they need and avoid unnecessary killing. The excess carnage in attacks by bluefish or swordfish on fish schools are a well known example ofwasteful predation (Bigelow & Schroeder 1953). Foxes raiding gull colonies often kill far more prey than they use (Kruuk 1976). Prehistoric Indians would sometimes drive bison herds off cliffs and then only harvest foetuses. Andrew Sih (1980) and Jeffrey Lucas (1985) argue the general importance of individual optima of wastefulness. T h e survival of one organism is possible only at great cost to others. The moral message in this obvious fact has been recognized by many philosophers and humanists, despite the general prevalence of roman-
ticism. Tennyson (Zn Memoriam, 55) confessed to a confusion and pessimism about Nature when
considering everywhere Her secret meaning in her deeds And finding that of fifty seeds She often brings but one to bear,
He must have realized that one-in-fifty would be extraordinarily favorable odds for all but a small minority of the worlds species. PROSPECTS MORALITY N FOR I
AN
IMMORAL WORLD
Huxley viewed the cosmic process as an enemy that must be combated. Mine is a similar but more extreme position, based on the more extreme contemporary view of natural selection as a process for maximizing selfishness. If the enemy is worse than Huxley thought, the more urgent is the need for biological understanding. As Singer (1981, 168) stated it in precisely this context, The more you know about your opponent, the better your chances of winning. Inadequate knowledge is likely to lead to counterproductive tactics, as Donald Symons (1979) noted for feminist issues. Modern insights may help resolve Huxleys (1894, viii) paradox. How could the maximization of selfishness produce an organism often capable of advocating, and occasionally practicing, charity towards strangers and even towards animals? Huxley dealt only sketchily with this problem, but anticipated that biology would someday . . . arrive at an understanding of the aesthetic faculty . . . that makes people want a society that . . demands self-restraint; in place of thrusting aside, or treading down, all competitors, it requires that the individual shall not merely respect, but shall help his fellows; its influence is directed, not so much to the survival of the fittest, as to the fitting of as many as possible to survive (Huxley 1894, 80-82). A number of recent sociobiologists have concerned themselves with Huxleys paradox, most notably Alexander (1987). Not surprisingly, all invoke kin-selected altruism and adaptations for taking advantage of reciprocation. They attempt to show why such factors operating in stone-agesociety should produce attitudes that favor the development, in modern societies, of broadly inclusive ethical systems. A similar attempt is made by Singer (198 l),a philosopher admirably knowledgeable on modern evolutionary theory. He proposes that the human capacity for reasoned argument constantly imposes a necessity for publicly presentable justification of personal action. If the chief of the Eastern Mohawks wants to achieve a trade agreement with his Western counterpart, he must be able to argue convincingly that the agreement
I.
400
ZYGON
would benefit all Mohawks. Some such need led Plato to urge consideration of the welfare of all Greeks, not merely of all Athenians (Singer 198 1, 1 17). As the necessity for dealing with an ever broader range of groups arises, comparably expanded ethical systems must be advocated and followed. As Singers ever expanding circle of consideration reaches individuals of negligible relationship or likelihood of reciprocation it puts an altruist into ever more basic conflict with Huxleys cosmic process. This combat being urged, between the microscopic atom and the illimitable macrocosm may seem a bit one-sided. Is not Annie Dillard showing a pathetic megalomania when she exclaims I came from the world, I crawled out of a sea of amino acids, and now I must whirl around and shake my fist at that sea and cry shame! (Dillard 1974, 177). Are Huxley and Dillard perhaps like Jobs neighbors urging him to rebel against oppression by an arbitrarily malicious, all-powerful, and omniscient god? No, not omniscient. The evolutionary process is immensely powerful and oppressive, but unlike Jobs god it is abysmally stupid. It can reliably maximize current selfishness at the level of the gene, but is blind to future macroscopic consequences of current action. It does not have the sense to realize that mechanisms evolved for practicing unfair nepotism o r making self-seeking deals with others can be subverted in the interests of broad altruism. All through evolutionary history there have been such changes with important future consequences entirely unrelated to the selfishness that brought them about. I will mention two of many possible examples. Manipulation was designed in the service of selfishness, but can play a positive role in human society. Anyone who makes an anonymous donation of money or blood o r other resource, as a result of some public appeal, is biologically just as much a victim of manipulation as someone whose selfsacrifice serves the interests of a tyrant. In his otherwise enlightened treatment of ethics and sociobiology Singer (1981) misses the role of manipulation in philanthropy and social activism. He maintains that the prevalence of anonymous blood donation shows an altruism in human nature that defies evolutionary interpretation. It is indeed correct that truly altruistic acts cannot be favored by selection, but the ability to induce others to behave altruistically certainly can be. My second example is anatomical. The subtle and versatile engineering of the human hand was fully appreciated by Galen about 1800 years ago (May 1968). In the last million years it has been modified by selection for precise manipulation of objects, but such selection could only be brought to bear on an organ already adapted to grasping and rotation. This earlier stage was produced by selection for special sorts
of arborial locomotion, such as brachiation, which demands an ability to turn the body with a hand holding an overhead limb. The production of an organ capable of turning doorknobs was made possible as an incidental consequence of selection to be better than ones neighbors at swinging through the trees. The helping hand of the Good Samaritan and the motivation for its use raise no question on the malice o r power of natural selection. They merely show that Huxleys tenacious and powerful enemy is a mindless fool. The stupidity of the enemy is the one advantage enjoyed by those who would follow Huxleys banner into combat. It is hardly a cause for complacency. The enemy is in fact powerful and tenacious, and we need all the help we can get to overcome billions of years of selection for selfishness. If biology can aid our understanding of our enemy it could be of great value. A key insight is the need to be suspicious of attitudes that arise from the biology of family life and group loyalties. They are based on narrow genetic self-interest and must be there to serve the enemys cause, not ours. As Singer states it: Discovering biological origins for 3ur intuitions should make us skeptical about thinking of them as df-evident moral axioms. . . . Far from justifying principles that are ihown to be natural,a biological explanation is often a way of debunkng the lofty status of what seemed a self-evident moral law (Singer 1981, 70-71). In natural selection there may be a parallel to the doctrine of original sin as a partial explanation for human misconduct. This was noted by Huxley (1894, 27) and more recently by Campbell (1975, 103-104) and Harry Power (1981, 17). Human speech is so effective a form of communication that once evolved it gave rise to a system of information transfer to rival the transmission of genes in reproduction. It made elaborate teaching possible, and it enormously augmented a purely cultural evolutionary process. The mechanism of cultural evolution is fundamentally different from that of genetic transmission, but the resulting evolutionary processes have many formal features in common. The transmission of cultural information by verbal messages or direct imitation can be followed, like genes, by phenotypic effects on recipient individuals. Recognizable cultural elements (Dawkinssmemes) can wax and wane in a population, and such quantitative effects can be treated by mathematical devises not very different from those of population genetics (Cafalli-Sforza & Feldman 1981). A meme (e.g., a new song, food fad, religious cult) will spread wherever society provides a favorable environment for its spread. There is an obvious analogy with selection acting on a gene. Perhaps the natural selection of memes is a bit less deplorable than that of
402
ZYGON
genes. Human beings are not as stupid as nature. They can sometimes anticipate the social effects of adopting a meme. But ultimately the ethical problems are much the same. The one necessary and sufficient reason for a meme spreading in a human population is that it is good at getting itself spread. It is not necessary that it enhance the biological fitness or perceived well-being of its practitioners nor that it increase general prosperity. I am sure that cigarette smoking has spread more rapidly through many populations in recent decades than have many sound public health practices. Richard Dawkins (1976) identifies the natural selection of memes as another aspect of the tenacious and powerful enemy. In its recognition of the evolutionary importance of manipulation and the added potential received from language, biology can add its voice to the chorus of warnings on the dangers of propaganda. Both the ability to use propaganda and an inclination to be suspicious of it were designed by natural selection to spread selfish genes in tribal microcosms. In its boundless stupidity this evolutionary process incidentally designed machinery capable of dealing with the question of whether a message is one of help or harm for what we really want for the macrocosm. We can use abilities developed for petty intrigue to deal with sermons coming from the pulpit, activists (or sociobiologists) tracts, o r manipulation by politicians on television screens. People can now espouse remote and inclusive ideals far removed from the selfishness that gave rise to the power to do so. It is understandable that people in the novel civic environments of the last few millennia will have aspirations for the dictatorship of the proletariat, or the triumph of the master race, o r the saving of souls. Because such strivings are not directly favored by natural selection, I have hope that some such cause can provide the humane artifice that can save humanity from human nature. Modern biology also helps us identify ourselves as apt soldiers for Huxleys army and clarifies the distinction between ourselves and the enemy. Each of us at conception received a unique genotype that never existed before. Unless there is an early division of the embryo it will never be duplicated. This genotype has no significance in evolution beyond its brief and minor influence on rates of increase and decrease of the component genes. It is of the utmost personal significance because it directs individual development and controls vital functions. In these processes each gene interacts in various ways with the rest of the genotype and with environmental conditions. On the average, a gene must produce results favorable to its own replication, but the particular interactions in our own development were unique and unpredictable and may deviate markedly from average effects. We are
George C. Williams
403
all special genotypically besides having unique individual histories that give us our own personal collections of memories and experience that constitute the self. There is no conceivable justification for any personal concern with the interests (long-term average proliferation) of the genes we received in the lottery of meiosis and fertilization. As Huxley was the first to recognize, there is every reason to rebel against any tendency to serve such interest. Gunther Stent ( 1 979) objects that Dawkinss concept of rebelling against ones own genes is a meaningless contradiction. He apparently missed the relevance of major technologies (hair dyeing, tonsillectomy, etc.) based on such rebellion. They are directed at individuals perceptions of flaws in the development controlled by the genes in the special association of their own genotypes. T h e combat urged by Huxley and refined by Dawkins is against the much less personal multigenerational effects of our genes, and this is even less of a contradiction. It is what Joseph Lopreato (1981, 124) meant when he proposed that the goal of moral striving should be the ultimate negation of the commandment of natural selection which is very much in the spirit of Huxleys (1894,63) urging that we refuse any longer to be the instruments of the evolutionary process.. . . Like Huxley and everyone else, I have my opinions on what the world ought to be like and on the best ways of moving towards that goal. Huxley did not write his essay to champion his particular view of utopia, and my purpose is merely to update Huxleys message and to characterize the enemy more clearly than was possible in Huxleys time. The updated program for the betterment of the human condition is a twofold attack on the natural enemy and any institutional enemies favored by cultural evolution. In Dawkinss words (1976,215), w must . . . rebel against the tyranny of the selfish replicators. e
REFERENCES Abele, Lawrence G. and Sandra Gilchrist. 1977. Homosexual Rape and Sexual Selection in Acanthocephalan Worms. Science 197:81-83. Alexander, Richard D. 1987. TheBiology of Moral Systems. Chicago: Aldine de Gruyter. Altmann, Jeanne. 1980. Baboon Mothers and Infants. Cambridge, Mass.: Harvard Univ. Press. Altmann, Stuart A. 1962. A Field Study of the Sociobiology of the Rhesus Monkey, Macaca Mullata. Annals of the New York Academy of Sciences 102:338-435. _ __ . 1974. Baboons, Space, Time, and Energy. American Zoologzst 14:221-48. Andersson, Malthe, Christer G. Wiklund, and Helen Rundgren. 1980. Parental Defense of Offspring: A Model of an Example. Animal Behavior 28:536-42. Arak, Anthony. 1984. Sneaky Breeders. In Producers and Scroungers: Strategies .f Exploitation andParasitisrn, ed. C. J . Barnard, 154-94. New York: Chapman & Hall. Baker, Myron Charles, and Stanley F. Fox. 1978. Differential Survival in Common Grackles Sprayed with Turgitol. American Naturalist 112:675-82. Barash, David P. 1977. Sociobiology of Rape in Mallards ( A n m platyrhynchos): Responses of the Mated Male. Scknce 197:788-89.
404
ZYGON
Beach, Frank A. 1964. Biological Bases for Reproductive Behavior. I n Social Behavior and Organization Among Vertebrates, ed. W. Etkin, 117-42. Chicago: Univ. of Chicago Press. . 1978. Sociobiology and Interspecific Comparisons of Behavior. In Sociobiology and Human Nature, ed. M. S. Gregory, A. Silvers, and D. Sutch, 116-35. Chicago: Univ. of Chicago Press. Beecher, Michael D. and 1. M. Beecher. 1979. Sociobiology of BankSwallows: Reproductive Strategy of the Male. Science 205: 1282-85. Berry, James F. and Richard Shine. 1980. Sexual Size Dimorphism and Sexual Selection in Turtles (order Testudines). Oecologia 44: 185-91. Bierman, Gloria C. and Raleigh J . Robertson. 1981. An Increase in Parental Investment during the Breeding Season. Animal Behavior 29:487-89. Bigelow, Henry B. and William C. Schroeder. 1953. Fishes of the Gulf of Maine. Fkhery Bulletin, Vol. 53. Washington, D.C.: U.S. Government Printing Office. Bowden, Bradley S. 1969. A New Method for Obtaining Precisely Timed Inseminations in Viviparous Fishes. Progressive Fish Culturirt 3 1:229-30. Burger, Joanna. 198 1. On Becoming Independent in Herring Gulls: Parent-young Conflict. American Naturalist 117:444-56. ____and C. G. Beer. 1975. Territoriality in the Laughing Gull ( L . atricilla). Behaviour 55:301-20. Buskirk, Ruth, Cliff Frolich. and K. G . Ross. 1984. The Natural Selection of Sexual Cannibalism. American Naturalist 123:612-25. Camphell, Donald T . 1978. Social Morality Norms as Evidence of Conflict between Biological Human Nature and Social System Requirements. Dahlem Konferenzen, Life Science Research Report 9:75-92. Cavalli-Sforza, L. L. and Marcus W. Feldman. 1981. Cultural Transmission and Evolution: A Quantitative Approach. Princeton, N.J.: Princeton Univ. Press. Clutton-Brock, T . H., S. D. Alhon, R. M. Gibson, and F. E. Guinness. 1979. The Logical Stag: Adaptive Aspects of Fighting in Red Deer (Corvw elaphw L.). Animal Behavior 27:2 11-25. Crump, Martha L. 1983. Opportunistic Cannibalism by Amphibian Larvae in Temporary Aquatic Environments. American Naturalist 121:28 1-89. Cuff, Wilfred R. 1980. Behavioral Aspects of Cannibalism in Larval Walleye, Stizostedion uitreum. Canadian Journal o Zoology 58: 1504-507. f Daly, Martin. 1978. The Cost of Mating. American Naturalirt 112:77i-74. f Darwin, Charles. 187 1. The Descent o M a n and Selection i n Relation to Sex. London: John Murray (reprinted 1981 by Princeton Univ. Press). Dawkins, Richard. 1976. The Selfish Gene. Oxford: Oxford Univ. Press. ____. l982.The Extended Phenotype. Oxford: W. H. Freeman & Co. Day, Gordon M. 1953. The Indian as an Ecological Factor in the Northeastern Forest. Ecology 34~329-46. Dewey, John. [ 1900) 1954. Evolution and Ethics. Scientific Monthly 78:57-66. Dillard, Annie. 1974. Pilgrim at Tinker Creek. New York: Harpers Magazine Press. Dominey, Wallace J . 1980. Female Mimicry in Male Bluegill Sunfish-A Genetic Polymorphism? Nature 284:546-48. Downhower, Jerry F. and Luther Brown. 1981. The Timing of Reproduction and Its Behavioral Consequences for Mottled Sculpins, Cottus bairdi. In Natural Selection and Social Behavior, ed. R. D. Alexander and D. W. Tinkle, 78-95. New York: Chiron Press. Dunford, Christopher. 1977. Kin Selection for Ground Squirrel Alarm Calls, American Naturalist 1 1 1 :782-85. Eichwort, Kathleen R. 1973. Cannibalism and Kin Selection in Labidomera clivicollis (Coleoptera: Chrysomelidae). American Natu.ralist 107:452-53. Elwood, Rohert W. and Malcolm C. Ostermeyer. 1984. Does Copulation Inhibit Infanticide in Male Rodents? Animal Behavior 32:293-94. Endler, John A. 1978. A Predators View of Animal Color Patterns. Evolutionary Biology 11:319-64.
~~
406
ZYGON
McKinney, Frank, Kimberly M. Cheng, and David J. Bruggers. 1984. SpermCompef tition in Apparently Monogamous Birds. In Sperm Competition and the Evolution o Avian Mating System, ed. Robert L. Smith, 523-45. New York: Academic Press. May, Margdret Talmadge. 1968. Galen on the Usefulness of the Parts of the Body. Ithaca: Cornell Univ. Press. Mayr, Ernst, and Will Provine. 1980. The Evolutionary Synthesis. Cambridge, Mass.: Harvard Univ. Press. Mineau, Pierre, and Fred Cooke. 1979. Rape in the Lesser Snow Goose. Behaviour 70:280-91. Oeil1, Kevin M. and Howard E. Evans. 1981. Predation on Conspecific Males by Females of the Beewolf Philanthw. basilaris Cresson (Hymenoptera: Sphecidae). Journal of the Kansas Entomologtcal Society 54553-56. Parker, Geoffrey A. 1979. Sexual Selection and Sexual Conflict. In Reproductive Competition in Insects, ed. M. S. Blum & N. A. Blum, 123-26. New York: Academic Press. Picman, Jaroslav. 1977. Intraspecific Nest Destruction in the Long-Billed Marsh Wren, Telmatodytes palustris palustris . CanadianJournal 4 Zoology 55: 1997-2003. Polis, Gary A. 1981. The Evolution and Dynamics of Intraspecific Predation. Annual Reuiew o Ecology & Systematics 12:225-51. f Power, Harry W. 1980. On Bluebird Cuckoldry and Human Adultery. American Naturalist 1 16:705-709. ___. 1981. The Question of Altruism. Sociobiology 6:7-21. and C. G. P. Doner. 1980. Experiments on Cuckoldry in the Mountain Bluebird. American Naturalist 1 16:689-704. Pratt, Harold L., Jr. 1979. Reproduction in the Blue Shark, Prionace glauca. Fishely Bulletin, U.S. 77:445-70. Raphael, D. Daiches. 1958. Darwinism and Ethics. In A Century of Darwin, ed. S. A. Barnet, 334-59. London: Heinemann Educational Books. Roberts, Thomas A. and James J. Kennelly. 1980. Variation in Promiscuity among Red-Winged Blackbirds. Wilson Bulletin 92: 1 10-12. Kohwer, Sievert, 1978. Parent Cannibalism of Offspring and Egg Raiding as a Courtship Strategy. American Naturalist I 12:429-40. Schaffer, William M. 1968. Intraspecific Combat and the Evolution of the Caprini. Evolution 22:817-25. Schaller, George B. 1972. The Serengeti Lion. Chicago: Univ. of Chicago Press. Sherman, Paul W. 1980. The Limits of Ground Squirrel Nepotism. American A s s < J c ~ ~ tion f o r the Advancement of Science, Selected Symposium 35:505-44. ___ . 198 1. Reproductive Competition and Infanticide in Beldings Ground Squirrels and Other Mammals. In Natural Selection and Social Behavior, ed. R. D. Alexander & D. W. Tinkle, 31 1-31. New York: Chiron Press. Sih, Andrew. 1980. Optimal Foraging: Partial Consumption of Prey. American Naturalist 116:281-90. Silk, Joan B., C. B. Clark-Wheatley, P. S. Rodman, and Amy Samuels. 1981. Differential Reproductive Success and Facultative Adjustment of Sex Ratios among Captive Female Bonnet Macaques (Macaca radiata). Animal Behauior 29:1106-120. Singer, Peter. 1981. The Expanding Circle. New York: Farrar, Straus, and Giroux. Speth, John D. 1983. Bison Kills and Bone Counts. Chicago: Univ. of Chicago Press. Stent, Gunther S. 1979. Introduction: T h e Limits of the Naturalistic Approach tn Morality. Dahlem Konferenzen, Life Sciences Research Reports 9: 13-2 1. Stinson, Christopher H. 1979. On the Selective Advantage of Fratricide in Raptors. Evolution 33: 12 19-225. Symons, Donald. 1979. The Evolution o Human Sexuality. New York: Oxford Univ. f Press. Trail, Pepper W , Stuart D. Strahl, Jerram L. Brown. 1981. Infanticide in Relation to Individual and Flock Histories in a Communally Breeding Bird, the Mexican Jay (Aphelocoma ultramarina). American Naturalist 1 18:72-82.
~
Notice
The Association for Religion and Intellectual Life (ARIL) announces the Sixth Coolidge Research Colloquium for individual study of intellectual-religious topics, for communal living and ecumenical dialogue at Episcopal Divinity School, Cambridge, MA, June 4-June 30, 1989. Support includes room and board as well as access to EDS, Harvard, and other Boston libraries. Resource theologians for 1989 are Rabbi Jonathan Omer-Man, the Reverend Jack Boozer, and Sr. Nancy Malone, O.S.U. APPLICATION DEADLINE: FEBRUARY 1, 1989. For more information, contact Beth and Me1 Kaiser, 815 Dolley Madison Road, Greensboro, NC 17410 (919-292-5616), or ARIL (9 19-654-5425).