science reduces physical
objects, or in the field of sensuous experience. Atoms can be reduced
into positive and negative electrical elements and these may, perhaps
do, imply a structure of ether that again invites further analysis and
so on ad infinitum. None of the hypothetical constructs carry with
themselves the character of being ultimate elements unless they are
purely metaphysical. If they are fashioned to meet the actual problems
of scientific research they will admit of possible further analysis,
because they must be located and defined in the continuity of space and
time. They cannot _be_ the points and instants of modern mathematical
theory. Nor can we reach ultimate elements in sensuous experience, for
this lies also within a continuum. Furthermore, our scientific analyses
are dependent upon the form that our objects assume. There is no general
analysis which research in science has ever used. The assumption that
psychology provides us with an analysis of experience which can be
carried to ultimate elements or facts, and which thereby provides the
elements out of which the objects of our physical world must be
constructed, denies to psychology its rights as a natural science of
which it is so jealous, turning it into a Berkeleyan metaphysics.

This most modern form of rationalism being unable to find ultimate
elements in the field of actual science is compelled to take what it
can find there. Now the results of the analysis of the classical English
psychological school give the impression of being what Mr. Russell calls
"hard facts," i.e., facts which cannot be broken up into others. They
seem to be the data of experience. Moreover, the term hard is not so
uncompromising as is the term element. A fact can be more or less hard,
while an ultimate element cannot be more or less ultimate. Furthermore,
the entirely formal character of the logic enables it to deal with equal
facility with any content. One can operate with the more or less hard
sense-data, putting them in to satisfy the seeming variables of the
propositions, and reach conclusions which are formally correct. There is
no necessity for scrutinizing the data under these circumstances, if one
can only assume that the data are those which science is actually using.
The difficulty is that no scientist ever analyzed his objects into such
sense-data. They exist only in philosophical text-books. Even the
psychologists recognize that these sensations are abst