e to be formed upon the utter and hopeless collapse of all the
possible arguments in favour of Theism. Having fully demonstrated that
there is no shadow of a positive argument in support of the theistic
theory, there arose the danger that some persons might erroneously conclude
that for this reason the theistic theory must be untrue. It therefore
became necessary to point out, that although, as far as we can see, nature
does not require an Intelligent Cause to account for any of her phenomena,
yet it is possible that, if we could see farther, we should see that nature
could not be what she is unless she had owed her existence to an
Intelligent Cause. Or, in other words, the probability there is that an
Intelligent Cause is unnecessary to explain any of the phenomena of nature,
is only equal to the probability there is that the doctrine of the
persistence of force is everywhere and eternally true.

As a final step in our analysis, therefore, we altogether quitted the
region of experience, and ignoring even the very foundations of science,
and so all the most certain of relative truths, we carried the discussion
into the transcendental region of purely formal considerations. And here we
laid down the canon, "that the value of any probability, in its last
analysis, is determined by the number, the importance, and the definiteness
of the relations known, as compared with those of the relations unknown;"
and, consequently, that in cases where the unknown relations are more
numerous, more important, or more indefinite than are the known relations,
the value of our inference varies inversely as the difference in these
respects between the relations compared. From which canon it followed, that
as the problem of Theism is the most ultimate of all problems, and so
contains in its unknown relations all that is to man unknown and
unknowable, these relations must be pronounced the most indefinite of all
relations that it is possible for man to contemplate; and, consequently,
that although we have here the entire range of experience from which to
argue, we are unable to estimate the real value of any argument whatsoever.
The unknown relations in our attempted induction being wholly indefinite,
both in respect of their number and importance, as compared with the known
relations, it is impossible for us to determine any definite probability
either for or against the being of a God. Therefore, although it is true
that, so far as human sci