eliminate from it the data of the problem with which you start, and
that you can never do, any more than you can stand on your own
shoulders or outstrip your own shadow.... In one word, to
constitute the reality of the outward world--to make possible the
minimum of knowledge, nay, the very existence for us of molecules
and atoms--you must needs presuppose that thought or thinking self,
which some would persuade us is to be educed or evolved from
them.... To make thought a function of matter is thus, simply, to
make thought a function of itself[5].'

From this reasoning there can be no escape; and it is more rational for
a man to believe that colour exists as such in a flower than, after
having plainly seen that such cannot be the case, forthwith to
disregard the teaching of this analogy, and to imagine that any apparent
evidence of mind as a result of matter or motion can possibly be
entertained as real evidence.

Remembering, then, that from the nature of this particular case it is as
impossible for mind to prove its own causation as it is for water to
rise above its source, it may still be well, for the sake of further
argument, to sink this general consideration, and to regard such
spurious evidence of causation as is presented by Materialism, without
prejudice arising from its being _prima facie_ inadmissible.

Materialists, as already observed, are fond of saying that the evidence
of causation from neurosis to psychosis is as good as such evidence can
be proved to be in any other case. Now, quite apart from the general
considerations just adduced to show that from the peculiar nature of
this case there can here be no such evidence at all--quite apart from
this, and treating the problem on the lower ground of the supposed
analogy, it may be clearly shown that the statement is untrue. For a
little thought will show that in point of fact the only resemblance
between this supposed case of causation and all other cases of
recognized causation, consists in the invariability of the correlation
between cerebral processes and mental processes; in all other points the
analogy fails. For in all cases of recognized causation there is a
perceived _connexion_ between the cause and the effect; the antecedents
are physical, and the consequents are physical. But in the case before
us there is no perceived, or even conceivable, connexion between the
cause and the effect; for the cause