"the idiosyncracies of an individual mind," one appeals to these neighbors for confirmation. It is hard to see when or how these creatures of an individual mind should have won sufficient independence to comment on their parent's shortcomings. For the rest, it is not very clear whether it is the constructing self or the constructed self which would be condemned by their dissent.
Turning from this account of scientific method to the question of its limit as an instrument for acquiring knowledge, the author's answer is unhesitating: it has no absolute limit. A theory of knowledge and a description of scientific method must be identical, and if an attempt at the former leads us to posit certain 'unknowables,' it sets up 'fetishes' that a poetic nature may feel to be sublime, but which science can only find amusing (pp. 14-25). The author is inclined to regard the metaphysician as wedded to these "unknowables," and as sharing with the theologian the ridicule of having invented and fostered illusions.
Now as a rule there is no one more afraid of 'unknowables' than is the metaphysician, and if sometimes he embraces one it is with a sort of despair. But this despair springs from the contemplation of difficulties the author can hardly be said to have faced; and the mistake of having yielded to it is immediately seized upon by succeeding schools as a point of departure. Only, the spectacle of an Aristotle or of a Kant stumbling into pitfalls he himself has been foremost in pointing out, presents to the student of history a matter for reflection rather than merriment. The ease with which a detected error may disguise itself in a new formula, and pass for sound doctrine, should make one cautious about congratulating oneself on having escaped from an old mistake. This caution the author seems scarcely to have felt in the midst of the enthusiasm with which he expels certain "fetisches of the metaphysician" from the "temple of science."
There is space for only one example; but we may take one that is characteristic of the type of thought that our author represents. Consider, e.g., the relation of the discredited 'thing-in-itself' to the accepted 'fact of observation' which serves as material for scientific construction. In placing the former beyond the range of possible experience, it is generally felt that we have robbed ourselves of any terms that might serve to describe it. Not only do we not know the historical thing-in-itself, but we have no way of explaining what it is that we do not know—hence the emptiness of the conception. It may seem that nothing could be safer from such an objection than a bare 'fact of experience.' Yet if this term is to stand for that from which scientific construction starts, as from a datum, it is clear that it cannot owe any part of its original meaning to the 'classifications' that science effects as part of the structure. If however, we try to describe a fact in terms that do not get their meaning from just such a classification, we shall find ourselves helpless. It need scarcely be said that 'sense-impression' is no such term. To leave the 'fact' a limit to the process of tearing our structure down again, is to make the very
