etc., all of which were once directly under the control of
religion. What remains to be done is to separate in theory what has
already been separated in fact, with such additions as a more critical
knowledge may suggest as advisable.

Far more suggestive, however, than the association of religion with what
we may call the normal social forces, is its connection with conditions
that are now clearly recognised as abnormal. From the earliest times we
find the use of drugs and stimulants, the practice of fasting and
self-torture, with other methods of depressing or stimulating the action
of the nervous system, accepted as well-recognised methods of inducing a
sense of religious illumination, or the feeling that one is in direct
communion with a supernatural order of existence. Equally significant is
the world-wide acceptance--right up to recent times--of purely
pathological states as evidence of supernatural intercourse. About these
two sets of facts there can be no reasonable doubt. Over and over again
we can observe how the promptings of disease are taken for the voice of
divinity, and men and women who to-day would be handed over to the care
of the physician hailed as an incarnation of deity. In modern asylums
we find one of the commonest of delusions to be that of the insane
person who imagines himself to be a specially selected instrument of
deity. In such instances the causal influence of pathological conditions
is admitted. On the other hand, we have belonging to the more normal
type the person who claims a supernatural origin for many of his actions
and states of mind. And between these two extremes lie a whole series of
gradations. They exist in all stages of culture, and it is difficult to
see by what rule of logic or of experience one can say where the normal
ends and the abnormal begins. If we assume the inference of the normal
person concerning the origin of his mental states to be correct, it
seems difficult to deny the possibility of those of the insane person
having a similar origin, although distorted by the influence of disease.
If, on the other hand, we say the insane person is wholly wrong as to
the origin of his mental states, may we not also assume that the normal
person has likewise erred as to the cause of his emotions or ideas?

Two considerations may be urged in support of this conclusion. In the
first place, there is the fact of the fundamental identity of human
qualities under all conditions of