theorems of geometry if they conflicted with their political interests of the ruling class may be true, but this kind of partisanship has no place in the sciences3. If anyone wishes to argue that the earth is flat or the biblical account of creation literally true, they would be well advised not to become astronomers, geographers or palaeontologists. Conversely, those who resist the inclusion of the biblical account of creation into the school textbooks of California as a "possible hypothesis"4 do so, not because they may have partisan views (which may be the case), but also because they rely on a universal consensus among scientists that it is not only factually wrong, but that no argument in its favour can claim scientific status. It is not, so far as can be seen, "a possible scientific hypothesis". To challenge the refutation of the flat-earth thesis, or of the belief that God created the world in seven days, is to challenge what we know as reason and science. There are people who are Willing to do so explicitly or by implication. If they should, by some unlikely chance, prove to be right, we as historians, social or other scientists, would be out of a job. This does not significantly reduce the scope of legitimate scientific disagreement into which partisanship can and does enter. There can be considerable argument about what the facts are, and where they can never be definitively established (as in much of history), argument may continue indefinitely. There may be argument about what they mean. Hypotheses and theories, however universal the consensus which greets them, do not have the non-controversial status of e.g. verifiable or falsifiable facts or mathematico-logical propositions. They can be shown to be consistent with the facts, but not necessarily as uniquely consistent with the facts. There can be no scientific argument about the fact of evolution, but there can be, even today, about the Darwinian explanation of it, or about any specific version of it. And insofar as the "fact" itself is trivial, when taken out of the context of the questions we ask about it and the theories we form to link it with other facts, it also remains caught up in the web of possible partisanship. The same is true even of mathematical propositions, which become significant or "inter- esting" only by virtue of the links we establish between them and other parts of our intellectual universe. Nevertheless, and at the risk of being accused of positivism, the non-controversial nature of certain state- ments and of the means of establishing it, must be asserted. Some propositions are "true" or "false" beyond reasonable doubt, though the boundaries between reasonable and unreasonable doubt will be drawn differently, within a marginal zone, according to partisan criteria. Thus most traditional scientists would probably require far stronger and more rigorously sifted evidence to establish the existence of various extra-sensory phenomena than they would to accept, e.g. the survival of some animal believed to be long extinct; and this because many of them are a priori reluctant to accept the existence of such phenomena. Conversely, as the Piltdown forgeries and other examples show, an a priori readiness to accept verification of a plausible hypothesis can seriously relax the scientist's own criteria of Validation. But this does 521