against the rest of the world! If he does not, he might work out an amusing theory making the Referendum the ultimate test of reality. That, at least, would be a definite method of utilizing the experience of others, such as is at present lacking. We act quite inconsistently in sometimes submitting to the superior delicacy of the expert's senses, and sometimes rejecting it. A room full of unmusical or inartistic people would hardly dispute about tones or colors with a single musician or painter, but an assembly of non-sensitives would probably deny that Macbeth saw a ghost (though who more qualified than Macbeth to see the ghost of Banquo?). The color-blind, perhaps because they are in a minority, do not dispute the objectivity of colors they cannot see, but upon what logical principle should we be less forbearing towards those who claim to see the ultra-violet and infra-red rays of the spectrum, or the luminosity of a magnetic field? — In short, just as the excluding value of non- conformity was impaired in the first case by the possibility of genuine hyperaesthesia in the individual, so in the second it is impaired by the possibility of collective hypersesthesia. And just as in the first case conformity did not exclude error, owing to the possibility of complex hallucination, so it fails in the second, owing to the possibility of collective hallucination.
(3) The third criterion at first seems more valuable — until we recollect that every new fact and every new experience is in some degree out of harmony with and contradictory of our previous experience. Would it not be strange, then, to allow our own inexperience, and the stupidity of our ancestors to exercise an absolute censorship over the growth of knowledge? Besides, it so happens that in most cases when "universal experience" is appealed to, its voice is self-contradictory. (What right have we, e.g. to reject countless traditions in order to prove that miracles are "contrary to experience"?)
But perhaps Mr. Ritchie does not contend that any one of his criteria is singly sufficient as a test of reality and proposes to employ them collectively. But if so, should he not show some probability that they will always, or even normally, tend in the same direction? And even if they did, that would establish, not the collective theoretic certainty of criteria, each of which was individually fallible, much less a necessary basis for metaphysical inferences, but only a sort of practical probability, which it might be convenient to act upon. Thus the boasted rationality of the real reduces itself to this: upon Mr. Ritchie's own showing rationality is not an ultimate test, but resolvable into the three criteria he mentions, and in the end their value turns out to be practical!
Yet it may be that humbling the pretensions of this pseudo-rationality does good service in drawing attention to the commonest and most
