city have been
discussed in the preceding section.

[31] _Ibid._, p. 275.

[32] _Ibid._, p. 275.

[33] This lack of continuity between the cognitive function of the
nervous system and its other functions accounts for the strange paradox
in the logic of neo-realism of an act of knowing which is "subjective"
and yet is the act of so palpably an objective affair as a nervous
system. The explanation is that the essence of all deprecated
subjectivity is, as before pointed out, functional isolation. That this
sort of subjectivity should be identified with the "psychical" is not
strange, since a living organism is very difficult to isolate, while the
term "psychical," in its metaphysical sense, seems to stand for little
else than just this complete isolation. Having once appealed to the
nervous system it seems incredible that the physiological continuity of
its functions with each other and with its environment should not have
suggested the logical corollary. Only the force of the prepossession of
mathematical atomism in analytic logic can account for its failure to do
so.

[34] But it would be better to use the term "logically-practical"
instead of "subjective" with the psychical implications of that term.

[35] An analysis which has been many times carried out has made it clear
that scientific data never do more than approximate the laws and
entities upon which our science rests. It is equally evident that the
forms of these laws and entities themselves shift in the reconstructions
of incessant research, or where they seem most secure could consistently
be changed, or at least could be fundamentally different were our
psychological structure or even our conventions of thought different. I
need only refer to the _Science et Hypothese_ of Poincare and the
_Problems of Science_ of Enriques. The positivist who undertakes to
carry the structure of the world back to the data of observation, and
the uniformities appearing in the accepted hypotheses of growing
sciences cannot maintain that we ever succeed in isolating data which
must remain the same in the kaleidoscope of our research science; nor
are we better served if we retreat to the ultimate elements of points
and instants which our pure mathematics assumes and implicitly defines,
and in connection with which it has worked out the modern theory of the
number and continuous series, its statements of continuity and infinity.

[36] In other words, science assumes that ever