tedly observed together. The wonder is,
then, that through oft-repeated observation of certain objects we come to
believe that we know something about the behavior of other like objects,
and the further behavior of these same ones. From the fact that I have seen
a given apple fall ten times to the ground, I infer that all the apples in
the world do the same when loosened, instead of flying upward, which, in
itself, is quite as thinkable; I infer further that this has always
been the case, and will continue to be so to all eternity. Where is the
intermediate link between the proposition, "I have found that such an
object has always been attended with such an effect," and this other, "I
foresee that other objects which are, in appearance, similar, will be
attended with similar effects"? This postulate, that the future will be
like the past, and that like causes will have like effects, rests on a
purely psychological basis. In virtue of the laws of association the sight
of an object or event vividly recalls the image of a second, often observed
in connection with the former, and leads us involuntarily to expect its
appearance anew. The idea of causal connection is based on feeling (the
feeling of inner determination to pass from one idea to a second), not upon
insight; it is a product of the imagination, not of the understanding. From
the habitual perception of two events in connection (sunshine and heat)
arises the mental determination to think of the second when we perceive the
first, and, anticipating the senses, to count on its appearance. It is now
possible to state of what impression the idea of the causal nexus is the
copy: the impression on which it is based is the habitual transition from
the idea of a thing to its customary attendant. Hence the idea of causality
has a purely subjective significance, not the objective one which we
ascribe to it. It is impossible to determine whether there is a real
necessity of becoming corresponding to the felt necessity of thought.
In life we never doubt the fact, but for science our conviction of the
uniformity of nature remains a merely probable (though a very highly
probable) conviction. Complete certainty is vouchsafed only by rational
demonstration and immediate experience. The necessary bond which we
postulate between cause and effect can neither be demonstrated nor felt.

If all experiential reasonings depend on the idea of causality, and this
has no other support than subj