The most intense debates, however, continue to address the same questions that troubled Darwin, who emphasised that natural selection is not the only mechanism for evolutionary change. For him, sexual selection explained many seemingly non-adaptive features of the living world. Geographical separation – as of the famous finches on the different Galapagos islands – are another, as separated populations will gradually drift apart by chance for random variation quite apart from any other adaptive pressures. And what does one mean by adaptation anyhow? Ultra-Darwinians insist, for instance, that all the subtle variations that can be found in the banding patterns of snail shells are adaptive.
Such rigid adaptationism famously came under attack by Gould and Lewontin at a meeting in 1979 commemorating the centenary of Darwin’s death, at which they described it as a ‘Panglossian paradigm.’ They pointed out that some seemingly functional aspects of an organism may be the accidental consequences of some quite other feature – as with, in their example, the spandrels (or spaces) between the arches which support the dome of Venice ’s San Marco. The spandrels are covered by glorious mosaics, which gives the illusion that they were designed to carry them, but architecturally they have no structural function. In Darwinian terms they are not adaptive. So too with biological features that might seem at first sight adaptive – such as the pattern of bands on a snail’s shell – but are accidental consequences of the chemistry and physics of shell construction.
Gould enraged many in his audience by devoting much of his allotted time to architectural rather than biological discourse, using it metaphorically to attack the determinism implicit in adaptationist theory. In the tea-break that followed his talk, Arthur Cain, a Liverpool evolutionary geneticist, denounced Gould as a Marxist, declaring that, as different shell colours camouflage the snails more or less effectively against predation by thrushes depending on whether the snails were in woodland or in open fields, it followed that every aspect of the patterning of the snail shells he studied had an adaptive function.
It was Gould and Elizabeth Vrba who gave the name to yet another potential contributor to evolutionary change: exaptation. An exaptation is some feature of an organism origin for one function that may also serve as the basis for another. The favourite example is that of feathers, believed to have evolved among small dinosaurs as a means of regulating body temperature, but which also enabled flight in the precursors of modern birds. For Gould such exaptations are an indicator of the chance or opportunistic nature of evolutionary processes, relying as they do on multiple mechanisms. If, as Gould repeatedly insisted, one could ‘wind the tape of evolution back’ to the pre-Cambrian or some other remote geological period and let it run forward again, it is very unlikely that sentient mammals like humans would appear. That is, the evolutionary future is unpredictable.