What we know of the social insects, and especially the extraordinary organization of agriculture and warfare among the ants, is striking both in terms of their convergence and in the almost alien nature of these complex societies. Certainly the jointed skeletons, the compound eyes, the miniaturized clones, and apparently robotic social organization are a familiar staple of science fiction. Suppose that there are advanced extraterrestrials: will they be like us, at least vaguely humanoid, or so alien as to defy belief and perhaps even recognition, let alone communication? The majority certainly tends towards the latter opinion. It probably owes as much as anything to George Gaylord Simpson, one of the last century's great evolutionary biologists. He was a prolific writer, and among the 16 papers he published in 1964 was one baldly entitled ‘The nonprevalence of humanoids’.

Simpson's article, presented with characteristic force and intelligence, was a sustained protest against what he saw as unwarranted extrapolation and conjecture. He argued that the history of life as revealed on Earth could not possibly be taken as a useful guide to biological events anywhere else in the Universe. Simpson presented his argument carefully, acknowledging that planets suitable for habitation would probably be in fairly short supply, and he conceded further that the likelihood of life itself arising was even lower. But no matter: as Simpson remarked, the Universe is a big place, so however uncommon life was, the total number of planets with life must be quite large.