New Scientist 15 November 1976 Sociobiology: the art of storytelling Ludwig von Boctalanfly, a Founder of general systems theory and 2 holdout against the neo-Darwinian tide, often argued that ‘netural seleotion must fail as a cormprehen- sive theory because it explains too much—a paradoxical, but perceptive statement, In 1969 he wrote: “If selection is taken 23 an axiomatic and a priori principle, it is always possible to imagine cuxiliary hypotiescs—unprived and by ature unprovable—to make it work in any special case . Some adaptive value can always he construed or imagined, “I think the fact that a theory so vague, so insufficiently vorifiable and so fer from the criteria otherwise applied in ‘hard’ science, has become a dogma, can only be explained 01 sociological grounds. Society and science have heen so steeped in the ideas of mochaniom, utilitarieniem, and the economic concept of free competition, that instead of Goa, Selection was enthroned as ultimate reslity.” Similarly, the arguments of Christien fundamentalism ‘used to frustrate me until I realised that there ere, in principle, no counter cases. and that, on this geound alone, the theory is bankrupt, The theory of natural selection is, fortunately, in much better straits. It could bo invalidated as a general cause of evolutionary change. (If, for example, Lamarckiaa inherit- ance were true and general, then adaptation would arise so rapidly in the Lamarckian mode that natural selection would be powerless to create and would operete only t0 eliminate.) Moreover, its action and efficacy have beea demonstrated experimentally by 60 years of manipulation within Drosophila bottles—not to maittion several thousand Stephon Jay Gould is professor of geology, Magoum of Comparative Zoology, Harvard ___Too many sociobiological explanations of behaviour come into the category of “Just So™ storiesi they may be plausible but are less than rigorously supported by solid evidence years of success by plant and animal breeders, ‘Yet in one area, unfortunately a very large part, of evolutionary theory and practice, astaral selection haz operated like the fundamentalist's God—he who maketh all things. Rudyard Kipling asked how the leopard got its spots, the rhino ite wrialled akin, He called his answers “Just So stories". When evolutionists study. incividval adaptations, when they try to explain form and behaviour by reconstructing history and assessing carrent utility, they also tell just-s0 stories—and the azent is natural selection, \Virtuosity in invention replaces testability ae the criterion for accepiauce. This is the procedure that inspired yon Bertalantiy’s complaint. It is also the procedure that has given evolutionary biology a bad name among many experi mental scfentisis In other disciplines, We should heed their disquiet, not dismiss it with a claim that they understand neither natural selection nor the special procedures of his- torical science, This style of storgtelling might yisld acceptable’ anewers if we could be sure of two things: frst, that all bits of morpiclogy and behaviour arise as direct results of natural selection, and socondly, that only ene seloctive explanation exists for each bit, But, as Darwin insisted voeiterousiy, and-contrary fo the mythology about him, there is mach more to evolution than natural selection. (Darwin was a consistent pluralist who viewed natural selection as the most important agent ef evolutionary change, but who accepted a range of other ageats and specified ‘the condi- ions of their presumed effecziveness. In chapter seven of the Origin of Species (sixth edition), for exemple, he attributed the cryptic colouration of a fatfish’s upper surface to natural selection and the migration of its eyesNe Scientia to inbéritance of adquired characters. He continually lie sisted that he wrote his two-volame Variation of Animals and Plats Under Domestication (1880), ‘with its Lamerekian hypothesis of pangenesis, primarily to lus trate the effect of evolutionary factors ether than matiiral selection, In a letter to Nature in 1860, he used the sharpest nd most, waspish Ianguage of his ‘life to castigate Sir Wyville Thompson for carveaturing his theory Ly ascribing all evolutionary change to natural selection.) : Sluice all theories cite God in their support, and since Darwin ‘comes close to this status among evolutionary Diolegisis, the panselectionists of the mocern synthes’s tended to remzke Darwin in their image. But we now reject this rigid yersfon of natural sclection and rant a major role to other’ evolutionary agents (genetic drift, fixetion of neutral mutations, fer example), We ius also recognise’ that many features arise indirectly es develoy. mental consequences of other features dhiactly’ subject to natural selection Mercover, and perligps. most’ im: portantly, there are a ‘multitude’ of potential ‘selective explanations for each feature, There. ic no such“ thine. neture ae a cclfevident and unambiguous str, ce When we examine the history of favoured. stériés for any particrlar adaptation, we do not trace a tale uf increas ing truth “as -uile story’ replaces the last, bit ‘rathe® a chronicle of shifting fads and Fashions, Whied Newtonian mechanical explanations were ding high, :G_ Sitenson wrote (in 1961), “The problem of the pelycasauy dorsal fin - ._. Seems excentially solved by Romer’s deniensiration that the regression relatfonship of fin area to body volume is appropriate to the Functioning of the fin as a temperature regulating mechanism.” Simpson's framess seins almost amusing since now—a mere 15 years later with behavioural Stories in vorae—most palacontologists fee! equally sure ‘that the sail was primarily « device for sexnal display. (Yes, I know ‘the litany: It might have performed bath functions. But this too isa story.) On the other side of the same shift in fashion’ a event atticle on functional endothermy in some large beetles had this to say about the why of it all: “It is possible that the increased power and soeed of terrestrial locomotion’ asso. ciated with a medost elevation of body temporatures tay offer reproductive advatitaues by increasing the effective: ‘Ress of intraspecitie agressive behaviour, particularly he: tween males.” This conjecture reflects no evidence drawn from the beetles themselves, only the aarrent fashion in ‘selective stories, We may be conddent that the came data, tollected 18 years ago, Would have inspired a speculation ‘about improved design and mechanical advantage,” Most work in sociobiology hae beea done in the mode ‘of adaptive stoxyielling based on the optimising character ‘and pervasive povrer of natural selection. AS euch, its weak nhesses of methodclogy are thoze thet have plagued s0 Much of evolutionary theory for more than z century. Sociobiole- ‘gists have anchored their stories in tho basis Darwiviaa notion of selection as individual reproductive ‘success Sociobiclogists have broadened their range oF selective stories by invoking concepts of inclusive Atmos’ end Kin Selection to solve'{successtully T think) the vexatious prob Jem of altruism-previously the greatest stumbling block ‘nian theory of social behaviour.’ (Altruistie acts fe the cement uf stable societies. Until we couldexplain zppareat acts of self-sacrifice as potentially beneficial to the genetic fitness of secrificers themselves—propagation ‘of genes through enhanced survival of kin, for example the prevalence of altruism blocked any Darwinian theory of secial behaviour.) Thus, kin selection has broadened the range of permis- able stories, but it hae not alleviated any methodological Aificulties "inthe process of storytelling itself. Vou Sertalanfiy's objections stil! apply, if anything with greater ‘oree, bescuse behaviour is geuerally more plastic and 16 November 170 5st More difficult to specity and homologise than smoxphologi. Sociobiologists wre still telling speculative stories, ell hitching without evidence to one potential ster emang, many, still using mere consistency with natural seledtios 9.8 criterion of acceptance, 4 David Barash, for example, tolls" the following story abou moantain Bluebirds. (Tt is, by the way, @ perfectly plausible story that may well be trie, T only wish co erty Cise its assertion without’ evidence or test, usin consis, teney with natural sclection as the sole criterion for useful speculation.) He reasoned that a male bird might be more sensitive to intrusion of other males before exes are laié than after (when he can be certain that his genes are i, Side). So Barash studied two nocts, making tne observa: tloas at 10-day intervals, the first before the exis were laid, the last two after. For each period of observation, he monnted a stuffed male near the ‘nest while the male occupant was out foraging. When the male returnel hie counted aggressive encountere with both model and female {At time one, males ia both nests were quite ausreseyh towards the model and less, but still substantially aseres. sive towards the female as'well. At-time two, after pgs had been’ laid, males were less aggressive to models aad “scarcely aggressive to females at all, At time three, males rere still lesa aggressive towards iiodels, and not averse sive at all towards females, é Is consistency enought Barash concludes that he has established consistency with natural selection and nced dé no more: “These resus are consistent with the expectations of evolutionary theory. ‘Thus ageression foward an intreding male (the. mode) would clearly be especially advantageous eariy in the breeding season, when territories and nests. aro normally defended .. . The initial, aggressive response to the mated female is also adaptive in that, given a situation sigcesting a high probability of adultery (that is, the presence of the model near the female) and essuming that replacemert females are-availeble, obtaining a new mate would erhancs the fitness of males... The decline in male-female augtes, sivencss during incubation and fledgling stages could be attributed fo the impossibility of being euckolced aftor the eggs have beon laid... The results are cousistent with an evolutionary Interprezation. In addition, the tetm ‘adultery 4 mnblushingly employed in this letter without quotation \ marks, as I bolieve it reficcts a trae analogy to the human concept, in the sense of Lorenz, It may also be prophesied that continued application ef a. sinilay evolutionacy approach will eventually shed considereble light on various human foibles as well” en Gonsisteat, ves, But what about the obylous alternative, dismissed without test in a line by Barash: male returns at times vo and three, approaches the model a few times encounters no reaction, mutters to himself: the. avian equivalent of “it's that damned stuffed bird again?” and ceases to bother. And why not the ovident tests expise a male to the model for the first time ajter the eves are Tela, Q We have been deluged in recent years with sociohie- logical stories, Some, like Barash’s are plausible, if ensup Borted. For many others, T can only confees my satuition of extreme unlieeliness, to say che Ieast—for adaytive aid genetic arguments about why fellatio and cunnilingus are More common among the upper classes, or why miele pam handlers are more successful with females and people-whe are eating than with males and-people who are net cating, Not all sociobiology proceeds in the mode of srorytelling for individual cases, It rests on firmer methodelogica! fround When it seeks broad correlations cevoss taxonomic Ines, az between reproductive strategy and distribution of resources, for example, or when it can make testable, ‘Quantitative predictions 2s in Bob Trivers and Hope Hare's9 Now Scientist 16 November 1976 work on haplodiploidy and eusociality in Hymenoptera. Here sociobiology has had and will eontimne to have sc. cess, And here I wich it well, Por it represents aut extensivit of basic Darwinism to a realm where it should apply. Sociobiological explanations of human behaviour ei counter two special difficulties, suggesting that a Darwinian, model may be generally inapplicable in this case. © First we have very little direct evidence about the gonetics of behaviour in humans; and we know no wey t) obtain it for the specific behaviours that figure most prominently in sociobiological speculation—aygression and. conformity, for instance. With our long generations, it is very dificult 10 amass much data on heritability. More importantly, we cannot (ethically, that’ is) perform the kind of breediag oxperi ments, in stendardised en viroaments, that would yield, the required information. Thus, in dealing with Iumans, sociobiologiste rely even, more heavily than usual om speculative story telling. At this point, the political debate engendered by socio- biology comes appropriately to the fore, For these spect- lative stories about human behaviour have broad tapli- cations and proscriptions for social policy—end this true quite apart from the intent or personal politics of the storyteller. Tatent aad usage are very different things; the latter marke poli cal and social influence, the former is gossip ox, as best, sociology. ‘The “common _ political charaeter and effect of these stories lies in the direction historically taken by. nativistic arguments about kuman behaviour’ and capabilities—a defence of existing social arrangements as part of our Biology, : Jn Taising this point, T do not act to suppress truth for fear of its political consequences. ‘Truth, as we understarid it, must always be our.primary criterion, We liye, because ‘wo must, with all manner of unplossant bielogical truthis— death being the most pervasive and ineluctable. 1 complain because sociobiological stories are riot truth, rather they are unsupported speculations wit political clout (again, [ ‘must emphasise, quite apart from the intent of the story- teller). All science is embedded in cultural contexts, and the Tower the ratio of data to social iaaportanee, the more science reflects the context. In stating that there is politics in sociobiology, I do not criticise the seieatists inyolyed in it by claiming that an unconscious politics has intruded into a supposedly objec tive enterprise. For they arc behaving like all zood scien- tists—as human beings in a cultural context, 1 only asle for 4 more explicit recognition of the context —and, spect fcally, for more attention to the evident impact of specula {ive suciobiological stories, Yor example, when the Dew ‘York Times suns a wooklong front page series on women ahd their rising achievements and expectations, spends the fist four days documenting their progress towards social gtuality, devotes the last day to patential limits upon this Progress, and advances sociobiological stories as the on'y ‘argument for potential limits—then we know that-these are Stories with consequences; “Sociologists believe that women ‘will continue For soma'years to achieve greater parity with mon, both in-the work place and in the howe. But an uneasy sense of frustration and pessimism is growing among some advocates of Full female equality in the face of mounting éonservative opposition. Moreover, even some staunch feminists: are reluctantly reaching the conclusion that women’s aspirations mey'ultimately be limited by inherent biological differences that will forever leave men the ominant sex” (Hew York Ties, 50 November, 1977). The article then quotes two social scientists, exci with a story, First, “If you define dominance as who ovcapies for” mal roles of ‘responsibility, then there is no suciely where niales are not dominant, When something is so anirersal, ‘the probability is—as reluct tant as I am to say it—that there is some quality of the organism that leads to this condition.” Secondly, “Tr may mean that there never vyill be full parity in jobs, that women will always pro dominate in the caring tastes like, teaching and social work. and in the life sciences, while men will pr vail in those requiring more aggression —business and polities, for exacnple—and in the ‘dead’ sciencos Tike physica” © ‘Secondiy, the standard foundation "of Darwinian Just'so stories does not apply to humans. That Foundation is the implication: if adap- tive, then genetic—for the inference of adaptation is usually the only basis of a selective story, and Dar. winism is 2’ theory of genetic diange and varia. : . ‘ton. in vopulations. ‘Much of, human behaviour je'clearly adaptive, but the problem for sociobiology is that: humans have developed an altetnative, non-genetic system. to support and transmit adaptive behaviour—cultural evolution, (An edaptive be: haviour does‘ ndt_tetiuire’ zenetic input and Darwinian selection for its Origin and maintenance'in humans; it may. arise by trial-and ervor in a few individuals whe do not Giffer genetically from their groupmiates in any way relevant to this behaviour epread by learning and imitation, and stabilise across generations by value, custom and tradition.) Moreover, cultural transmission ig far more powerful in potential’ speed and spread than natural selection—for cultural evolution operates in. the “Lamarckian” mode by inheritanes through custom, writing and technology of characteristics acquired by Datei luuman activity in each generation, ‘Thus, the existence of adaptive behaviour in huiaons says nothing about tte probability of a genetic basis for it, or about the operation of natiral selection, ‘Take, for example, Trivers's concept of “reciprocal altruism’. The phenomenon exists, to ¢ sure, and it is clearly adaptive In. honest moments, we all ackbowiedge that many of our “altruistic” acts are performed in the hope and expectation of future reward. Can anyone imagine a stable society with: out bonds of reciprocal obligation, But structural necessi ties do not imply direct genetic coding, (AIL human be: haviours are, of course, part of the potential range pi mitted by our geaotypé— but socicbiological speculetions posit dicetz natural selecifon for specitic behavioural traits. As Benjamin Franklia said: “Bither we hang together, or388 New Sdenist, 16 Novertber 1978 assuredly we will all hang separately.” 4 Tho grandest goal—I de not say the only goal—ef huitian secobiology must Fail in the face of these difficulties. ‘That goal is no less than the reduction of the behavioural (i deol most of the social) sciences to Darwinian theory. Edyate Wilson presents a visio of the hnman sciences ahriaking in their indepeadeat domain, absorved on one le by neurobiology and on the other by sociobiology. Bat this vision cannot be fulfilled, for the reason cited above, Although we can ideatify adaptive behaviour in humans, we cannot tell if it is genetically based (while rauich of it must arise by fairly pare cultural evolution). ‘Yet the reduction of the human sciences to Darvinism requires the genetic argument, for Darwinism is 2 theory about genetic change in populations. All else is analogy and metaphor, My crystal ball shows the human sociobiologists retreai+ ng to a fallback position—indeed it is happening alzea They will argue that this fallback ig as powerful as thor original position, though it actually represents the unravel Jing of their fordest hopes. ‘They will argue! yes, indeed, ‘we cannot tell whether an adaptive behaviour is genctically coded or not. But it doesn’t matter, The same adaptive com straints apply whether the behaviowr evolved dy cultural or Darwinian routes, and biologists have identified ‘and explicated the adaptive constraints, (Steve Emlen tells me, for example, that, some Indian peoples’ gather food ti accordance with predictions of optimal foraging strategy— a theory developed by ecoloxisis.} But it does matter, It makes all the difference in the world whether human beliaviours develop ana stabilise by caltural evolution or, by direct Darwinian selection for genes influencing specific adaptive actions, It makes a reat difference because, cultural and Darwinian evolution difer: profoundly in the three major areas that embody what evolution, at least as a quantitative science, is all about 1. Rate, Cultural evolution, ag a “Lamarekian” process, can proceed orders of magnitude more rapidly ‘tien Dal winian evolution, Natural selection continues ‘its worl within Homo sapiens, probably at characteristic rates for change in Jarge, fairly stable populations, but Lhe poe! of cultural, evolution hassatwarfed its influence (alteration in frequeney of the sickling gene ». changes in. modes OF ‘communication aind transportation , Consider what'we have’ Gone in the past SOD), years, all without the shehtest evidence for any change ir the power of the Inman bral, 2. Modifiabitity, Complex traits of cultural evolution cen be altered rapitiiy; Darwinian change is ‘imted to. mach slowor rates of spread of alleles by natural selection. 5. Dif usability. Shice traits of enltural evolution een be trancmitted by imitation and inouleation, evolutionary pat terms inelude fvequent and complex aiastomosis ariong branches. Darwinian’ evolution is a process of cortinuous Aivergence and ramification, I believe that the future will bring mutual iMlumisation between two vigorous, independent disc’plines Darwinian theory and cultural history. This is « good thing, joyously to be welcomed. But there will he no reduction of ho human sciences to Darwinian theory and the research programme of humaa socobiology will fail, The name, of course, may survive, Th is an irony of history that more monts are judged successful it their label sticks, though the emerging Content of a discipline may lie cloter to kat opponents originally “advecnted. “Modem geology, Jo: cxamole, is an even Dlend of Lyell’s strict umformh tarienisn: and the claims of eatastrephiste, But we call the hybrid doctrine by Lyell’s name, ; T welcome the coming failure of -reductinistie opas Because it will lead us to recogaise human complexity at its proper level..For consumption by Phne's miliions, may colleague Bob Trivers maintamed: “Sooner or tcter, polity fal science, Taw, economics, peychology, psychialiy, and anthropology will all be branches of sociebiology” (Time, 1 August, 1977, p 54). It's one thing to coniecture, as Twonld allow, that common features among independently de- yoloped Iegal systems might reflect adaptive constraints and, might be exphicated usefully with some biological ‘analogies. Tt is quite another to statg, es Bob ‘Trivers cit, that the entire legal profession, amoag others, will be sub. sumed as mere epiphenomena'of Darwinian processes, T road Trivers’s statement the day after I had sung in a full production of Berlioz’s Requiem. And 1 remembered the viseeral reaction 1 had experienced upon hearing the four brass choirs, finally amalgamated with the 10 tympani inthe massive din preceding the great Tube mirun the spine tingling and the inveluntary tears that almost pro. ‘Yeuted me from singing. L tried to analyse st in the terms of Wilson's conjecture—reduction of behaviour lo neuro: biology én the one hand end sociobiology on the other And T realised that this conjecture might apply to my experience. My reaction had been physiological and, as a good mechanist, 1 do not doubt that its neurological foxrda. tion can he ascertained, Twill also not be surprised to learn thet ‘the reaction has something to do with adaptation (emotional overwhelming to cement group coherence. in the face of danger, to tell a story), But I also realised that these explanations, however “true”, could never caphire the meaning of that experience. And 1 sey this not to espouse mysticism or incompre- hensibility, but merely to assert that the world. of human bebavionr is too complex and multifarious to be uiilocked by any simple key. I say this to maintain that this richness =f anything—is both our hope and our essence. g ‘Be i vem oar wl se peatedin Slay ope mea ce Und Rt NARS Socd nn el, isaly Generar nts hee Binders for Bind your magazines in the Easibinder Attrectively bound in Blue Balacron with the title Gold blocked on the spine, the Easibinder isdesigned to hold 13 copies of NEW SCIENTIST. ‘The Fasibinder opens flatforoasyreference and copiescan be removed anc replaced with ease. The price £2.70 inclusive. ‘Overseas postage 25p each extra. Please allow 3/4 weeks for fulfilment of order, L.A Uxie Steet Landen VIBTSZ ecco = eens INCE at foretoee P.0,/Chaqus value. £2,70each fo: NEWSCIENTIST ns... Nae ADDRES bare,