8/10/2019 Kohn Nat08- Disgreements on Group n Ind selection (1).pdf

1/4

If biologists have learnt onething about evolution over the

past 40 years, it is that naturalselection does not work for the

good of the group. The defininginsight of modern Darwinism isthat selection sees individuals andacts on them through the genesthey embody. To imagine otherwise, genera-tions of students have been warned, is to fallinto a naive error definitively exposed as suchin the mid-1960s.

Yet group selection the idea that evolu-tion can choose between groups, not just theindividuals that make them up has a higherprofile today than at any time since its appar-

ent banishment from mainstream evolutionarytheory. And it gets better press, too. This is inpart owing to the efforts of David Sloan Wilson,of Binghamton University in New York, whoargues that the dismissal of group selection wasa major historical error that needs to be recti-fied. And it does not hurt that he has been joinedby Edward O. Wilson, the great naturalist andauthority on social insects. They and many oth-ers have worked to reposition group selectionwithin the broader theme of selection that actssimultaneously at multiple levels.

The remarkable feature of the debate is notthat neither side has managed a conclusive vic-

tory; it is that they seem, from the outside, toagree on so much. Everyone agrees that group

selection occurs, says Andy Gard-ner of the University of Edinburgh,

UK. Yet Gardner and his colleaguesStuart West and Ashleigh Griffinhave trenchantly criticized1DavidSloan Wilsons arguments on thissubject a critique to which DavidSloan Wilson responded by initiat-

ing a lengthy debate in the community underthe heading If the theorists cannot agree.

There is widespread agreement that groupselection and kin selection the post-1960sorthodoxy that identifies shared interests withshared genes are formally equivalent. Ifyoure talking about altruism, the mathematicsthat gives you kin selection is also the math-

ematics that if you slice it another way givesyou group selection for altruism, says AndrewBourke, who studies social insects at the Univer-sity of East Anglia, UK. To some researchers, thismakes the choice between the two frameworksa matter of perspective. Others doubt that theidea of group selection adds much, if anything,to the understanding obtained from kin selec-tion theory. And for theorists such as Gardnerand his colleagues, debates about group selec-tion are largely semantic.

But others see deeper conceptual disagree-ments, which, according to Samir Okasha, a pro-fessor of philosophy at the University of Bristol,

UK, stem from underlying theoretical assump-tions about how we should carve up nature.

Okasha, who has written a well-regarded bookon the subject2, reels off a string of questions.

How should a group be defined? How shouldaltruism and selfishness be defined? Does theselfish-gene view of evolution conflict with theidea that units of selection can act at differenthierarchical levels? Does group selection requirethat groups engage in a form of reproduction,as individuals do, or just that groups must havediffering levels of productivity?

To clarify that last issue, Okasha distinguishesbetween two kinds of multi-level selection,known in the trade as MLS1 and MLS2. MLS1focuses on how the division of a populationinto groups affects the frequencies of differ-ent types of individuals in the population. The

classic example would be altruism: a popula-tion divided into groups may reward altruismmore than an undifferentiated one would. MLS2focuses on the frequencies of the groups them-selves. The fittest groups in the MLS1 senseare those that contribute the most individualsto the next generation; the fittest groups in theMLS2 sense are those that contribute the mostgroups.

It is MLS1 that is implicated in the questionat the heart of debates over levels of selection how altruism can spread through a population.And it is MLS1 that can be taken as equivalentto kin selection. Interest in MLS2 processes

has grown more recently, as the question ofwhat levels of selection there are has become

A.N

UPSHAH/NATUREPL.COM;M.S

PENCERG

REEN/AP

THE NEEDS OF THE MANYThe idea that natural selection acts on groups, as well as individuals, is a source of unending

debate. Marek Kohnreports on what the two sides disagree about and why it matters to them.

296

NATURE|Vol 456|20 November 2008NEWS FEATURE DARWIN 200

8/10/2019 Kohn Nat08- Disgreements on Group n Ind selection (1).pdf

2/4

entwined with that of how those levels evolved;how the advent of multicellularity, say, createdan organism level above the level of the cell.

Insuch major evolutionary transitions, smallerunits are integrated into larger ones, be it repli-cating molecules into simple cells or organismsinto societies. For a stable new class of entity toemerge, selfish tendencies among their con-stituent units had to be suppressed at the grouplevel.

Establishing priorities

Charles Darwin saw that what was good for thegroup might not be good for the individual. InThe Descent of Man (John Murray, 1871), heobserved that although a high standard ofmorality gives but a slight or no advantage toeach individual man and his children over theother men of the same tribe, it would cer-tainly give an immense advantage to one tribeover another. However, by the 1960s, manynaturalists and biologists had come to seethe relationship between what is good for theindividual and what is good for the group asunproblematic. They assumed that individualswould subordinate their own interests to those

of the species.Some, though, had doubts. And when the latenaturalist Vero Wynne-Edwards argued explic-itly that natural selection acted on groups andthat individuals restrained their consumptionand reproduction for the good of the commu-nity3, his book provoked a small but powerfulbacklash. An under-appreciated alternativemodel for the same phenomena had recentlybeen proposed by the theorist Bill Hamilton,then at the London School of Economics.Hamilton argued that the spread of altruisticbehaviour, which Wynne-Edwards assumedwould be selected because of its benefits to the

group at large, would in fact be determined bythe degree to which altruists shared genes with

their beneficiaries. A quartet of the critics metin Oxford and coined the term kin selectionto distinguish Hamiltons theory from Wynne-Edwardss group selection.

Kin selection was easily and powerfullyexpressed in mathematics, and became the neworthodoxy, its ascendancy cemented in placeby its forceful and compelling popularizationin Richard Dawkinss The Selfish Gene (OxfordUniv. Press, 1976). Group selection might bepossible in theory, these thinkers allowed, butit could be ignored in practice. As the theorist

George C. Williams declared in his 1966 bookAdaptation and Natural Selection (PrincetonUniv. Press) a rallying call against group selec-tion that had great influence inAmerica the higher levelsof selection are impotent.

With the focus on altruism,disadvantageous by definition,the impotence of the higherlevels seemed insurmount-able. Evolutionary modelsdeveloped by John MaynardSmith, the most junior mem-ber of the Oxford quartet, drove home the point

that it would take just one selfish individual tospoil things for everybody. Cheats would inevi-tably multiply at the altruists expense. Selectionfor characteristics that benefited a group at theexpense of individuals would be opposed byselection within the group, and selection at thatlower level would win.

In their recent writing on the subject4, Wilsonand Wilson accept the first part of this proposi-tion, but challenge the second. Sometimes selec-tion between groups is weak and sometimes itis very strong, they argue; the balance betweenlevels of selection should be evaluated case bycase. Drawing on Darwins original insight, they

boiled down the principles underlying the altru-ism question to three short sentences. Selfish-

ness beats altruism within groups. Altruisticgroups beat selfish groups. Everything else iscommentary.

Gardner, Griffin and West distinguishbetween the old group selection of Wynne-Edwards and this new version, embeddedwithin a framework of multi-level selection4.David Sloan Wilson, however, just sees onelong argument since the 1960s. Group selec-tion was originally rejected on the groundsthat it was implausible rather than on evidenceit didnt happen, he says. But the evidence has

changed. It can now be accepted on the best ofevidence, says Wilson, not that it always hap-pens, but it happens some of the time.

Wilson highlights experi-ments on microbes as beingtotally definitive in thisregard even if the experi-menters dont always presenttheir results that way. Ben-jamin Kerr, of the Universityof Washington in Seattle, andhis colleagues have shown,for example, that prudent

viruses, which do not over-exploit their bacte-

rial hosts, prevail over rapacious ones whentheir ability to disperse and find new prey isrestricted5. Wilson and Wilson say that the studyelegantly establishes the plausibility of Wynne-Edwards hypothesis without citing Wynne-Edwards or the term group selection.

Kerr is happy to affirm the connection.Wynne-Edwards observed that groups wouldevolve to conserve their resources for the benefitof their descendants if those descendants lived inthe same place or as he put it2, fall heirs to thesame ground. Kerr recalls that in 1964 MaynardSmith discussed the prospects for group selec-tion on this basis using a model that involved

imaginary mouse colonies in haystacks

6

. Ourexperiment was similar in spirit to Maynard

The adaptation of

individual organisms

is an amazing result.

It should be treasured

rather than abused.

Andy Gardner

297

NATURE|Vol 456|20 November 2008 DARWIN 200 NEWS FEATURE

8/10/2019 Kohn Nat08- Disgreements on Group n Ind selection (1).pdf

3/4

Smiths intent, says Kerr: to elucidate the con-ditions favouring the type of restraint Wynne-Edwards wrote about. The intent is indeedsimilar in that the model and the experiment areattempts to put Wynne-Edwardss verbal argu-ments into a more rigorous form. But whereasKerr feels that Wynne-Edwards had a point,Maynard Smith actually wanted to dispose ofan argument he regarded as silly. The haystackmodel established the view that conditions per-mitting group selection were highly restrictiveand unlikely to be encountered in real life.

David Sloan Wilson also points to a study7

by Paul Rainey and Katrina Rainey that lookedat bacteria cooperating to form a mat on thesurface of a liquid medium. The mat is madeby mutants that produce larger than normalquantities of a sticky polymer, a trait calledwrinkly spreader (WS). Wilson and Wilsonsay that selection between groups maintains theWS trait, even though the individuals makingthe polymer incur costs that the other bacteriadont have to pay. But Rainey, who works at theNew Zealand Institute of Advanced Study atMassey University in Auckland, says that thisis not what the experiment showed. In fact,

given our experimental design, this possibilityis more or less ruled out. Kin selection providesthe simplest explanation for the evolutionaryemergence of the WS type. After all, the factthat daughter cells are physically stuck to theirprogenitors tends to keep things in the family.

Paul Rainey does not dispute that kin selec-tion is equivalent to multi-level selection of theMLS1 type. Indeed, he has described the emer-gence of the mats as a triumph of selection act-ing at the higher level.8He is quick to point out,however, that this is not group selection of theMLS2 type.

In Raineys eyes, David Sloan Wilsondoes

not provide an adequate solution to the problemthat MLS2 selection needs groups to reproduce

as individuals. Wilson proposes that selectionmay act on groups identified by shared traitsrather than kinship, and that such groups maybe small and ephemeral. Rainey considers suchmodels useful for explaining how cooperation ismaintained, but doubts that they can account foradaptation at the level of the group, and is espe-cially sceptical that they can explain the evolu-tion of group-level adaptations during a majorevolutionary transition such as the integration ofcells into multicellular organisms. That requiresan explanation for how selection shifts its focus

from individuals within groups to groups them-selves; from MLS1 to MLS2. Theorists have notpaid much attention to this transition, Raineysays. Samir Okasha refers tothis as a grey area. I think it is agreat black box!

Rainey thinks that his matwork could help to open upthat black box, and that theindividuality groups need forMLS2 selection might developin such systems. Cells thatcheat by enjoying the benefits of being in matswithout contributing the chemical that holds

them together may in fact have a role at a higherlevel. Because they can swim away and give riseto new colonies, they have the potential to act asa kind of germ line for colonies that otherwisecannot reproduce. If cheats can work like thisthey could give the mats the individuality thatis needed if natural selection is going to operateat the colony level.

There is a problem here. For a cheat to founda colony, a fresh mutation is needed WS hasto be rediscovered. Rainey, though, is confidentthat selection can impose design on the process.From the outset, selection will work to optimizethe rates at which cheaters arise groups with

too many fall apart. Rainey predicts that withthis toe-hold in the process, evolution will

produce systems that can switch between matforming and swimming stages by changing geneexpression, not waiting for de novo mutation.

Rainey affirms that as natural selectionfavours the evolution of cheats whenever co-operation has a cost, selection will act on anypopulation that contains cooperative types atboth the individual and the group level. Buthe cautions that this should not make groupselection the default explanation for the evolu-tion of the higher level. His reservation echoesa complaint made by Bourke. I dont like what

I see as an almost systematic downplaying ofthe evidence in favour of kin selection, Bourkesays. There are lots of other people out there,

working on vertebrates rightthrough to microbes, who findevidence for kin selection.

Bourke states the orthodoxcase firmly. Everything isgene selection, he declares.When populations are struc-tured into groups, either kin-or group-selection ideas may

need to be applied. But it is still gene selection.What matters is the kind of trait being selected

for, and the structure of the population it is in.If a group-level selection effect is seen, it will bebecause of relatedness a nonrandom associa-tion of genes. I dont see that labelling it groupselection adds anything to that insight, whichwas essentially Hamiltons. Kerr, by contrast,regards kin selection and group selection notas alternative views but as potentially comple-mentary perspectives. Looking at the same sys-tem in different ways can often provide deeperunderstanding, he says.

Although semantic and conceptual confusionare in ample supply, on their own they dont sus-tain the controversies over group selection. Core

beliefs are at stake, and powerful intuitions areat work.

R.A.CLEVENGE

R/CORBIS;SCIMAT/SPL;M.QUINTON/GETTYIMAGES

Our species is the

primate equivalent of

a beehive or a single

organism.

David Sloan Wilson

298

NATURE|Vol 456|20 November 2008NEWS FEATURE DARWIN 200

8/10/2019 Kohn Nat08- Disgreements on Group n Ind selection (1).pdf

4/4

From an orthodox point of view, group-selec-tion ideas get in the way of an appreciation ofthe power of natural selection. Gardner says thatin principle he is equally happy with kin selec-tion and multi-level selection, but in practice heprefers to work with the former. It is closer to thebasic theory of adaptation, connected throughHamilton to Ronald Fisher, one of the architectsof the modern synthesis that brought togetherpopulation biology and evolutionary theory inthe middle of last century9. It keeps the focus onthe constantly striving individuals that populate

Darwins vision.And the idea of adaptation is to be treated

with respect, not carelessly applied at the level ofthe group, where it doesnt really work. I guessthe reason I fell in love with biology rather than,say, physics, is that you have this purely mechan-ical process natural selection that gives riseto purpose and function and the appearance ofdesign in the living world, Gardner says. Theadaptation of individual organisms is an amaz-ing result, and nothing like it occurs elsewherein the natural sciences. It should be treasuredrather than abused.

Avoiding the implicationsGroup-selection thinking is perceived by someas not just an abuse of natural selection but alsoa denial of it. Tellingly, Dawkins chose evolu-tionary theorist Robert Triverss version of thisargument for his recent anthology10of great sci-ence writing. Imagining that evolution worksfor the greater good looks like a way to avoid theradical implications of natural selection likeclaiming that it has little effect, for instance, onthe human mind.

For David Sloan Wilson, however, it is theindividualistic perspective that betrays influ-ences from outside science. The concept that

self-interest can be a grand explanatory prin-ciple seemed to become current in the middle

of the twentieth century, and what happened inevolutionary biology was part of that trend, hesays. Now, against the background of individu-alism and self-interest, we have the propositionthat our species is the primate equivalent of abeehive or a single organism so what are theimplications of that? Thats not science fiction;that is the growing consensus.

And this human dimension is ultimately whygroup selection arouses passions. When Darwinconsidered selection at group levels, he thoughtof social insects and of humans. When Hamilton

extended the equation derived by his collabora-tor George Price to accommodate group-selec-tion processes, he combined the discussion withspeculations about the adaptive value of warfareas a way for groups to compete and encoun-tered a reception similar to that of a cat whopresents its owner with a half-dead mouse. Butthat was more than 30 years ago. Today the ideathat warfare and altruism are two sides of thesame coin that competition between groupsby way of warfare encourages altruism withingroups is widely discussed, as promulgated11by Samuel Bowles of the Santa Fe Institute inNew Mexico.

Robert Boyd of the University of California,Los Angeles, and Peter Richerson at the Uni-versity of California, Davis, have promoted theidea that selection between cultural groups canbe highly effective in humans. Whereas genetic-group models are vulnerable to the fact thatindividuals can migrate from group to group,reducing the between-group variance on whichselection works, cultural models have the advan-tage that human migrants can adopt the culturesof the groups to which they migrate. Peopleintuitively recognize their species as groupishand realize that models of group evolution canhave implications for the understanding of

human societies.That is part of their attraction for Joel Peck,

whose modelling studies at the University ofSussex, UK, suggest that the conditions underwhich group selection could work might not beas restrictive as his late colleague Maynard Smithhad implied. Underlying his interest in the lifesciences, Peck says, is a desire to be involved inpositive social change. Group selection is oneof the topics within evolutionary biology that ismost directly about how societies change. Henotes that in the major transition that unifiedthe ancestors of the mitochondrial and nucleargenome into the eukaryotic cell, the two ances-

tors interests were almost entirely united. Aswell as being fascinating in themselves, proc-esses such as these may give us hints abouthow to unify the interests of different humanindividuals, or groups of individuals, he says.

Wilson suggests that in the light of groupselection, the human species displays unity; Peckhopes that group selection may show humanshow to achieve unity. Ultimately, their scienceis about the good of the species. Marek Kohn is a writer in Brighton, England.

He is the author ofA Reason For Everything:

Natural Selection and the English Imagination.

1. West, S. A., Griffin, A. S. & Gardner, A.J. Evol. Biol. 21,

374385 (2008).2. Okasha, S. Evolution and the Levels of Selection (Oxford Univ.

Press, 2006).3. Wynne-Edwards, V. C.Animal Dispersion in Relation to Social

Behaviour (Oliver & Boyd, 1962).4. Wilson, D. S. & Wilson, E. O.Q. Rev. Biol.82,327348

(2007).5. Kerr, B., Neuhauser, C., Bohannan, B. J. M. & Dean A. M.

Nature442,7578 (2006).6. Maynard Smith, J. Nature201,11451147 (1964).7. Rainey, P. B. & Rainey. K.Nature425,7274 (2003).8. Rainey, P. B. Nature446,616 (2007).9. Whitfield, J. Nature 455, 281284 (2008).10. Dawkins, R. Oxford Book of Modern Science Writing(Oxford

Univ. Press, 2008).11. Choi, J. K. & Bowles, S. Science318,636640 (2007).

See also pages 281 and 295, and online at www.

nature.com/darwin.

299

NATURE|Vol 456|20 November 2008 DARWIN 200 NEWS FEATURE