nal, or that it must have come into existence without a cause. In this
respect, therefore, it would seem that the theory of Atheism has the
advantage over that of Theism; for while the former theory is under the
necessity of embodying only a single inconceivable term, the latter theory
is under the necessity of embodying two such terms.

[33] Mr. Herbert Spencer has treated of this subject in his memorable
controversy with Mill on the "Universal Postulate" (see _Psychology_, Sec.
427), and refuses to entertain the term "Inconceivable" as applicable to
any propositions other than those wherein "the terms cannot, by any effort,
be brought before consciousness in that relation which the proposition
asserts between them." That is to say, he limits the term "Inconceivable"
to that which is _absolutely_ inconceivable; and he then proceeds to affirm
that all propositions "which admit of being framed in thought, but which
are so much at variance with experience, in which its terms have habitually
been otherwise united, that its terms cannot be put in the alleged relation
without effort," ought properly to be termed "_incredible_" propositions.
Now I cannot see that the class "Incredible propositions" is, as this
definition asserts, identical with the class which I have termed
"Relatively inconceivable" propositions. For example, it is a familiar
observation that, on looking at the setting sun, we experience an almost,
if not quite, insuperable difficulty in _conceiving_ the sun's apparent
motion as due to our own actual motion, and yet we experience no difficulty
in _believing_ it. Conversely, I entertain but little difficulty in
_conceiving_--_i.e._, imagining--a shark with a mammalian heart, and yet it
would require extremely strong evidence to make me _believe_ that such an
animal exists. The truth appears to be that our language is deficient in
terms whereby to distinguish between that which is wholly inconceivable
from that which is with difficulty conceivable. This, it seems to me, was
the principle reason of the dispute between Spencer and Mill above alluded
to,--the former writer having always used the word "Inconceivable" in the
sense of "Absolutely inconceivable," and the latter having apparently used
it--in his _Logic_ and elsewhere--in both senses. I have endeavoured to
remedy this defect in the language by introducing the qualifying words,
"Absolutely" and "Relatively," which, although not appropriate words, are
the