it. On the one hand there is the
necessity for attributing soul to everything. These mythologies of
atom-souls, molecule-souls, this hatred and love which are the inner
aspects even of the simple facts of attraction and repulsion among the
elements, fit better into the nature-philosophy of Empedocles and
Anaxagoras than into ours. The main support, indeed the sole support, of
this position is that this world of the infinitely little cannot be
brought under control as far as its "soul" is concerned. Thus we can
impute "a soul" to it without danger. On the other hand, there is a
difficulty which made itself felt even in regard to Spinoza's system. All
bodily processes must have psychical processes corresponding to them, said
Spinoza. Conversely, all ideas in their turn must have bodily processes.
To the system including all bodily processes corresponds the sum-total of
psychical processes. This sum-total we call the soul. And in its entirety
it is the _idea corporis_. If "soul" were really nothing more than this,
the theory of parallelism might be right. But it is more than this. It
rises above itself, and becomes also the _idea ideae_; it is
self-consciousness and the consciousness of the ego; it makes its own
thought and the laws of it, its feelings and their intensity--its
experiences in short--a subject of thought. How does this fit in with
parallelism? Wundt himself, the most notable modern champion of
parallelism, admits and defines these limits of the parallelistic theory
on both sides.

Furthermore, the theory of parallelism, notwithstanding its opposition to
materialism, must presuppose that localisation of psychical processes of
which we have already spoken, and to which all naturalism appeals with so
much emphasis. Because of the fact that particular psychical functions
seem to be limited to a particular and definable area of the brain-cortex,
or to a spot which could be isolated on a particular convolution, it
seemed as if naturalism could prove that "soul" was obviously a function
of this particular organ or part of an organ. According to the theory of
parallelism this does not follow. It would assert: "What in one aspect
appears to be a psychical process, appears in another aspect to be a
definite physiological process of the brain." Yet it is clear that in
order to gain support for the doctrine of mutual correspondence,
parallelism has also the same interest in such localisation. For this is
the only method