o be nothing like a movement of inference.
It is evident, indeed, that memory is involved in and underlies every
such transition of thought. Illusions of memory illustrate rather a
process of wrong classing, that is to say, of wrongly identifying the
present mental image with past fact, which is the initial step in all
inference. In this way they closely resemble those slight errors of
perception which are due to erroneous classing of sense-impressions. But
since the intellectual process involved in assimilating mental elements
is very similar to that implied in assimilating complex groups of such
elements, we may say that even in these simple kinds of error there is
something which resembles a wrong classing of relations, something,
therefore, which approximates in character to a fallacy.

By help of this brief review of the nature and causes of illusion, we
see that in general it may be spoken of as deviation of individual from
common experience. This applies to passive illusion in so far as it
follows from the accidents of individual experience, and it still more
obviously applies to active illusion as due to the vagaries of
individual feeling and constructive imagination. We might, perhaps,
characterize all illusion as partial view, partial both in the sense of
being incomplete, and in the other sense of being that to which the mind
by its peculiar predispositions inclines. This being so, we may very
roughly describe all illusion as abnormal. Just as hallucination, the
most signal instance of illusion, is distinctly on the border-land of
healthy and unhealthy mental life; just as dreams are in the direction
of such unhealthy mental action; so the lesser illusions of memory and
so on are abnormal in the sense that they imply a departure from a
common typical mode of intellectual action.

It is plain, indeed, that this is the position we have been, taking up
throughout our discussion of illusion. We have assumed that what is
common and normal is true, or answers to what is objectively real. Thus,
in dealing with errors of perception, we took for granted that the
common percept--meaning by this what is permanent in the individual and
the general experience--is at the same time the true percept. So in
discussing the illusions of memory we estimated objective time by the
judgment of the average man, free from individual bias, and apart from
special circumstances favourable to error. Similarly, in the case of
belief, tru