otten mediating son of Puluga and the green shrimp _may_ bear
traces of Christian teaching. Caution is indicated.

Does Mr. Max Muller, so strict about evidence, boggle at the stone house,
the only son, the shrimp? Not he; he never hints at the shrimp! Does he
point out that one anthropologist has asked for caution in weighing what
the Mincopies told Mr. Man? Very far from that, he complains that 'the
old story is repeated again and again' about the godless Andamans. {97c}
The intelligent Glasgow audience could hardly guess that anthropologists
were watchful, and knew pretty well what to believe about the Mincopies.
Perhaps in Glasgow they do not read us anthropologists much.

On p. 413 our author returns to the charge. He observes (as I have also
observed) the often contradictory nature of our evidence. Here I may
offer an anecdote. The most celebrated of living English philosophers
heard that I was at one time writing a book on the 'ghostly' in history,
anthropology, and society, old or new, savage or civilised. He kindly
dictated a letter to me asking how I could give time and pains to any
such marvels. For, he argued, the most unveracious fables were
occasionally told about himself in newspapers and social gossip. If
evidence cannot be trusted about a living and distinguished British
subject, how can it be accepted about hallucinations?

I replied, with respect, that on this principle nothing could be
investigated at all. History, justice, trade, everything would be
impossible. We must weigh and criticise evidence. As my friendly
adviser had written much on savage customs and creeds, he best knew that
conflicting testimony, even on his own chosen theme, is not peculiar to
ghost stories. In a world of conflicting testimony we live by
criticising it. Thus, when Mr. Max Muller says that I call my savages
'primitive,' and when I, on the other hand, quote passages in which I
explicitly decline to do so, the evidence as to my views is
contradictory. Yet the truth can be discovered by careful research.

The application is obvious. We must not despair of truth! As our
monitor says, 'we ought to discard all evidence that does not come to us
either from a man who was able himself to converse with native races, or
who was at least an eye-witness of what he relates.' Precisely, that is
our method. I, for one, do not take even a ghost story at second hand,
much less anything so startling as a savage rite.