not lie without the field
ordinarily covered by the word superstition. For our purposes, therefore,
it is necessary to enlarge this definition. This may be done by
emphasizing the first component part of the word, and introducing into it
the notion of what has been left over, or of survival, made familiar by
the genius of Edward B. Tylor. In these lingering notions we have
opinions respecting relations of cause and effect which have resulted as
a necessary consequence from past intellectual conditions. A
superstition, accordingly, I should define as a belief respecting causal
sequence, depending on reasoning proper to an outgrown culture. According
to this view, with adequate information it would be possible to trace the
mental process in virtue of which arise such expectations of futurity,
and to discover the methods of their gradual modification and eventual
supersession by generalizations founded on experience more accurate and
extensive. Yet it is not to be assumed that in each and every case such
elucidation will be possible. In all human conduct there is an element
which cannot be designated otherwise than as accidental; this uncertainty
appears to be greater, the reaction against the natural conditions less
definite, the more primitive is the life. It is impossible to forecast in
what manner a savage may be impressed by an event of which he can note
only external conditions, or how his action may respond to the
impression. One may guess what opinion an augur would form concerning the
appearance of a single eagle or raven; but it would be labor lost to
attempt to conjecture the manner in which the imagination of the observer
would explain a flight of these birds, or what complicated rules augural
art might evolve to guide the interpretation.

This accidental quality, and the arbitrariness with which phenomena are
judged to be ominous, will be visible in the numerous "signs" here
recorded. At first sight, it may be thought that extreme folly is their
salient quality. Yet if we take a wide view the case is reversed; we are
surprised, not at the unintelligibility of popular belief, but at its
simplicity, and at the frequency with which we can discern the natural
process of unsystematic conjecture. Such judgments are not to be treated
with derision, as subjects of ridicule, but to be seriously examined, as
revealing the natural procedure of intelligence limited to a superficial
view of phenomena.

This consideration