be no mere contribution to philosophical literature. It must be no mere bringing together of materials for some other hand to arrange. How fond most of the contributors to science are of taking this view of their own labours! Modest people! As if any one would thank a mason who should say to him—"There, sir, are the stones; you can now build your house for yourself!" It must embrace every essential part of philosophy, thoroughly digested, and strictly reasoned out as a harmonious and consistent whole. It must show the exact point where every opinion and every controversy in philosophy takes off from the tap-root or main trunk of the great tree of speculation. The disputants themselves never
an idea which is involved, either directly or indirectly, either proximately or remotely, in the conception of 7+5, the subject of the proposition.
The fact is, that all propositions expressing necessary or a priori truths are analytic or resolvent, and that many of them are, at the same time, ampliative. They are ampliative—that is to say, they add something to our knowledge—whenever the predicate is wrapped up in such profound latency on the subject as not to appear to be involved in it all. In such cases, a new conception appears to be added to the subject of the proposition; but the truth is, that this new conception has not been really added to the subject, but has merely been forced out of it by a strong intellectual pressure. This latency of the predicate in the subject seems to have been the circumstance which misled Kant in his account of the synthetic judgments a priori—and in his attempt to show that all such propositions (including the geometrical axioms) did not depend for their necessity on the principle of contradiction. In regard to his deduction of the categories of the understanding, which he sets forth as synthetic judgments a priori, our present limits permit merely this remark to be made, that these are either not necessary a priori principles of intelligence, or, if they are so, the criterion of contradiction must be their test.
