Although it is very hard to establish a new and correct view of a subject which for more than two thousand years has been handled by innumerable writers, and which, moreover, does not receive additions through the growth of experience, yet this must not deter me from presenting to the thinker for examination the following attempt of this kind. An inference is that operation of our reason by virtue of which, through the comparison of two judgments a third judgment arises, without the assistance of any knowledge otherwise obtained. The condition of this is that these two judgments have _one_ conception in common, for otherwise they are foreign to each other and have no community. But under this condition they become the father and mother of a child that contains in itself something of both. Moreover, this operation is no arbitrary act, but an act of the reason, which, when it has considered such judgments, performs it of itself according to its own laws. So far it is objective, not subjective, and therefore subject to the strictest rules. We may ask in passing whether he who draws an inference really learns something new from the new proposition, something previously unknown to him? Not absolutely; but yet to a certain extent he does. What he learns lay in what he knew: thus he knew it also, but he did not know that he knew it; which is as if he had something, but did not know that he had it, and this is just the same as if he had it not. He knew it only _implicite_, now he knows it _explicite_; but this distinction may be so great that the conclusion appears to him a new truth. For example: All diamonds are stones; All diamonds are combustible: Therefore some stones are combustible.