« African American Beauty | Gene Expression Front Page | Nakedness is a virtue »
June 08, 2003


I mentioned in a previous post that Sewall Wright’s ‘Shifting Balance’ theory of evolution had recently been criticised. It is only fair to add that it has also been defended. Key references are:

The original critique
J. Coyne, N. Barton and M. Turelli: A critique of Sewall Wright’s Shifting Balance theory of evolution. Evolution, vol. 51, 1997, pp.643-671
M. Wade and C. Goodenough, Evolution, vol. 52, 1998, pp.1537-53
S. Peck, S. Ellmer and F. Gould, Evolution, vol. 52, 1998, 1834-39.
Reply to the counter-critiques
Coyne, Barton and Turelli, Evolution, vol. 54, 2000, pp. 306-17
Replies to the reply
M. Wade and C. Goodenough, Evolution, vol. 54, 2000, pp.317-24
S. Peck, S. Ellmer and F. Gould, Evolution, vol. 54, 2000, pp.324-27

No doubt this will rumble on.

To refresh memories about the Shifting Balance theory, Sewall Wright believed that evolution in large, randomly-mating populations would be slow and ineffective. The population would quickly evolve to an optimum equilibrium state (a peak in the ‘adaptive landscape’), but evolution would then stop, because natural selection could not cross the ‘valleys’ to other peaks. Since evolution clearly does take place, Wright proposed that the geographical substructure of populations was important. In small isolated sub-populations inbreeding and genetic drift could enable the sub-population to cross ‘valleys’ to new peaks of fitness despite adverse natural selection on the way. Successful new combinations of genes could then spread by migration or differential extinction of groups. The theory has been widely influential in American biology, notably on Ernst Mayr’s view of speciation, the Eldredge-Gould theory of punctuated equilibrium, and D.S. Wilson and Michael Wade’s theories of group selection.

On the other hand the Shifting Balance was never very popular among evolutionists in Britain. Haldane was sceptical, R. A. Fisher was hostile, and E. B. Ford reported empirical evidence against it. Fisher in particular denied that large populations would get stuck on peaks in the adaptive landscape, since in a large population there would always be numerous mutations arising to upset the equilibrium, and in any case environmental conditions (biotic and non-biotic) were constantly changing. The idea of a static ‘adaptive landscape’ was therefore misleading. Fisher also doubted whether sub-populations would remain isolated for long enough to diverge purely by genetic drift. The recent debate is largely about whether Fisher or Wright was closer to the truth.

I won’t (and can’t) answer that question, but I think the opposition between Fisher and Wright is sometimes misrepresented. Wright is represented as believing that species are divided into partly isolated sub-populations, while Fisher is represented as believing that an entire species is a single large randomly-mating population. If the disagreement is put in these terms, clearly Wright is right. But of course Fisher was well aware that species have a geographical structure, and that sub-populations are partially isolated from each other. Where he dissented from Wright was in doubting that the isolation was usually strict enough, or lasting enough, to be as important as Wright claimed. His objection is expressed in the second edition of his Genetical Theory of Natural Selection:

“The circumstance that smaller numbers, even less than 100, are sometimes found to reproduce themselves locally, does not, as has been supposed, add to the frequency of random extinction [of genes], or to the importance of the so-called ‘genetic drift’. For this, perfect isolation is required over a number of generations equally numerous with the population isolated. Even if perfect isolation could be postulated, which is always questionable, it is still improbable that the small isolated population would not ordinarily die out altogether before a period of evolutionary significance could elapse, or that it would not later be absorbed in other populations with a different genetic constitution” (Dover edition, p.10).

The underlying problem with the Shifting Balance is that populations will not diverge by genetic drift if there is, on average, more than one migrant between them per generation, regardless of the size of the populations. (The last point may be counter-intuitive, but is well-established. The effectiveness of genetic drift is inversely proportional to the size of the population, so that with larger populations a proportionately lower migration rate is needed to prevent them drifting apart. Of course, none of this applies to asexual species, mitochondria, or Y chromosomes.) Fisher considered this degree of isolation unlikely in the absence of major geographical barriers. Oddly, neither Fisher nor Wright seems to have mentioned in this context that most species have behavioural and/or morphological adaptations for dispersal at some stage in their life-cycle. Among humans, exogamous marriage rules, or capture of females from other tribes, would have the same effect.


Posted by David B at 03:24 AM

This is far from my area of expertise or interest,but given the nature of the site I will conjecture wildly and make a far-fetched analogy.

The Shifting Balance / punctuated equilibrium / spandrel theory seem more like what a libertarian would like, with innovation contingently coming from small groups.

Whereas the more rationalist, determinist, spandrel-free theory argued by Dennett in a book that said "Cranes! Not Skyhooks!" far too many times, seems rather Stalinist and excessively centralized. (I don't know where Dennett's sources stood on the "shifting balance" question; I definitely know what he thought about spandrels and punctuated equilibrium).

This fits with a common view of modern history, which I agree with -- NW Europe's disunity was actually a big advantage, since different nations were able to try different strategies, which would not likely have happened if the Hapsburgs or the Bourbons had succeeded in controlling all of Europe.

I definitely feel that China's unity since ~1300 has been a big detriment, if only because political-ideological refugees had nowhere to flee to. The refugee-importing nations (France, U.S., Holland, Switzerland, England and even Russia during one period) gained enormous talent from the exporting nations (Russia later, Austria, Spain). It's not as simple as that, of course, but unity has an enormous cost.

Posted by: zizka at June 8, 2003 07:46 PM

I think the idea that a large, unsplit population would necessarily reach some stable genetic equilibrium representing a local 'fitness maximum' makes sense only in an environment that itself does not change much (maybe sharks in the ocean, or the nautilus, would be good examples of this). Most places experience enough chaos from climate change, migrations of other animals, and population boom/bust cycles that there should be enough noise to disrupt the equilibrium and keep the gene pool diverse.

Posted by: bbartlog at June 9, 2003 09:46 AM

I'm usually more likely to side with the British school in debates like this, but that's usually because A) I'm not an expert and often operate on hunches B) the British school is usually the majority view (on both sides of the Atlantic) and seemingly less likely to be larded down with ideology.

The theory has been widely influential in American biology, notably on Ernst Mayr’s view of speciation, the Eldredge-Gould theory of punctuated equilibrium, and D.S. Wilson and Michael Wade’s theories of group selection.

While Mayr is good example, neither Gould or D.S Wilson have been "wildly influential" in any place but popular books.

Posted by: Jason Malloy at June 9, 2003 10:21 AM

I think the ideological angle could be played either way. Preference for 'small groups' versus 'large populations' could be presented as 'libertarian', but an emphasis on group selection as against individual or gene selection could be presented as socialist or collectivist. Apart from Gould, I don't know much about the politics of the American 'players'. On the British side, politics doesn't seem to have much to do with their views: Haldane was a Marxist, Fisher was conservative, more recently Maynard Smith is an ex-Marxist but still quite leftish, Dawkins is also leftish, but Hamilton (judging from his prefaces in 'Narrow Roads') would be considered by many practically a fascist, though needless to say I don't think that would be fair. Yet they all had much the same views on evolution.

Posted by: David B at June 10, 2003 02:40 AM