Debate is something on which science thrives. Active disputes are signs of a discipline in rude health, in which discovery piles on discovery, each new fragment of knowledge questioning the one before, until sufficient findings accumulate to decide the matter one way or the other. Or sometimes, a consensus is reached that none of the original protagonists had thought of. This consensus is always provisional. In contrast, a field in which everyone blandly agrees with everything is a field in stagnation.

That is why the publication of a paper in the Proceedings of the National Academy of Sciences (PNAS), robustly countering the identification of Homo floresiensis as anything other than a malformed human pygmy (T. Jacob et al. Proc. Natl Acad. Sci. USA doi:10.1073/pnas.0605563103; 2006), is to be warmly welcomed. The original discovery of a population of bizarre hominids that lived on the island of Flores in Indonesia until relatively recent times (P. Brown et al. Nature 431, 1055–1061; 2004) understandably caused a sensation. The idea of an extinct species of 'little people' that coexisted with modern humans gripped the public imagination, and journalists were quick to seize on the nickname of 'hobbit' to stand in for the less wieldy formal nomenclature. Hobbits, of course, are diminutive creatures that sprang from the imagination of J. R. R. Tolkien, who noted, in his novel The Hobbit, that hobbits are more elusive nowadays than they once were.

The latter-day hobbit story was spiced up considerably by the characters of the scientists who discovered it — some of whom publicly and not always politely disagreed with one another about the discovery's significance. Not to mention the lively rebuttals from many academic challengers (including but not exclusively the authors of the PNAS paper), who contend that to brand the Flores creature as a distinctive species is to create something as fictional as anything invented by Tolkien. Strong words have been exchanged. Skulduggery has been alleged. Accusations fly. This is ideal fodder for journalists whiling away the summer 'silly season'.

No one should be misled, however — the reported invocation of scientific error does not mean that an error has been committed, let alone any kind of misconduct. Is one entitled to ask whether the 2004 Nature paper was 'wrong', then? The answer is, robustly, 'yes'.

Palaeoanthropology — the study of the corporeal remains of human evolution — is a notorious arena for splenetic debate. To take the long view, it would have been surprising had the unearthing of the hobbit not led to strongly worded counterblasts, in which critics contended that the new find is really either a diseased human or an ape. Such controversies erupted in 1856 after the discovery of Neanderthal man (said to be a diseased human), and again in 1925 after the announcement of Australopithecus africanus (claimed to be a juvenile ape). The status of these creatures as distinct species, neither human nor ape, is now beyond question.

More recently, similar debates followed the discovery of Toumaï (Sahelanthropus tchadensis) and the East African taxon Kenyanthropus platyops. In the last two, as with the Indonesian 'hobbits', the debate continues — a sign that more remains to be discovered.

But there are obstacles: for two full summer seasons, no Indonesian or foreign group has dug at the Liang Bua cave on Flores. This seems to be an unhealthy by-product of the scientific controversy. The situation should be resolved so that this particular bit of palaeoanthropology can thrive again.