necessary to proceed by observation, by experiment, and then by induction,
but with prodigious mistrust of induction. Induction consists in drawing
conclusions from the particular to the general, from a certain number of
facts to a law. This is legitimate on condition that the conclusion is not
drawn from a few facts to a law, which is precipitate induction, fruitful
in errors; but from a very large number of facts to a law, which even then
is considered as provisional. As for metaphysics, as for the investigation
of universal law, that should be entirely separated from philosophy itself,
from the "primary philosophy" which does not lead to it; it has its own
field, which is that of faith: "Give to faith what belongeth to faith." In
the main he is uninterested in metaphysics, believing them always to
revolve in a circle and, I do not say, only believes in science and in
method, but has hope only from knowledge and method, an enthusiast in this
respect just as another might be about the super-sensible world or about
ideas, saying human knowledge and human power are really coincident, and
believing that knowledge will support humanity in all calamities, will
prolong human life, will establish a new golden age, etc.

Moreover, let there be none of that eternal and unfounded fear that
knowledge will cause the disappearance of the religious feeling. With
profound conviction and judging by himself, Bacon said: "A little
philosophy inclineth a man's mind to atheism, but depth in philosophy
bringeth a man's mind about to religion." Such is true philosophy,
"subordinate to the object," attentive to the object, listening to the
voices of the world and only anxious to translate them into human language:
"that is true philosophy which renders the voices of the world the most
accurately possible, like an echo, which writes as if at the dictation of
the world itself, adding nothing of its own, only repeating and
_resounding_."