consciousness of their existence, and then the nature of things operates upon it so as to make corresponding impressions which bring the original ones into cognizance - a gratuitous and not even plausible hypothesis.

The doctrine of a twofold origin as here set forth seems to be a part of his strange theory of a pre-established harmony, or at least chimes in with it the ideas and maxims are in the mind, while objects and events totally unconnected with them but completely correspondent are existing and happening outside. Nothing can be more totally destitute of evidence. It is a pure fiction.

The first part of these remarks will apply with little or no modification to Kant's cognitions d priori.

By cognitions rigorously interpreted he must mean either knowledge itself or the propositions in which such knowledge is affirmed, and in either case the objections urged against Leibnitz are valid against him. There is, indeed, another interpretation—a third meaning-brought forward in defence or explanation of the cognitions in question; an interpretation which would resolve Kant's doctrine into a mere assertion of certain modes of procedure which are natural to the mind, and are called forth by the exercise of the senses on appropriate occasions, at various periods in after life.

The discrimination of modes of mental action from general propositions, which I insisted upon in

the preceding letter, will enable me to show, when the occasion arrives, in what sense the plea is urged and how far it is available.

I have in the preceding argument treated these innate principles and à priori cognitions solely as general propositions, without regard to the character of the facts comprised in them, and have endeavoured to show that, from their very nature as such, they must be posterior to a knowledge of the individual facts which they comprehend; that, without such knowledge, no principles, maxims, or cognitions of any kind can exist.

But it is not all general propositions which, in the theories before us, are maintained to be innate principles or à priori cognitions. It is only those which are characterised by necessity and universality: attributes (it is alleged) not to be discovered by experience or perception but furnished by the mind itself.

The examination of the doctrine here intimated will occupy the two next letters, after which I shall enter upon the consideration of cognitions in their second character, in which, emerging from the condition of maxims or general propositions, they claim to be regarded as modes of mental procedure.

IN taking up the subject mentioned at the close of my last letter, I must draw your attention to the circumstance that in the passages quoted from Leibnitz and Kant there are two assertions made respecting innate principles and cognitions à priori : first, that they are independent of experience and even of all impressions on the senses; secondly, that they owe the necessity and universality which distinguish them from other propositions to the mind itself. More extraordinary assertions never saw the light.

A sort of haze seems to envelope some of the terms here employed, particularly the words experience and necessity. To the latter, I shall come by-and-by: at present I have to do with the former.

Experience is evidently of various kinds, sometimes it is simple and sometimes complex. When it is simple and has reference to external objects, it is the same thing as perception through the organs of sense. Perception is, indeed, a more

comprehensive word, for it may be used of only a single quality, whereas "experience," in common usage and in the simplest cases, denotes the perception of two or more qualities in connexion with each other, or what is appropriately termed a fact, of which a proposition is the verbal expression. We cannot, in ordinary language, be said to know the colour red by experience. We know it from perceiving or having perceived it, but we might, with great propriety, be said to know by experience that blood is of that colour. So, in common parlance, we learn by experience that ice is cold, that steel is hard, that metals are expanded by heat; or we may resort to the wider term and say we perceive them to be so.

I have introduced these remarks for the purpose of showing that the question "whether an external fact is learned from experience," is virtually identical with the question "whether it is learned from perception."

Kant himself seems to admit the same thing and to draw a similar distinction between the two phrases, when he says, in a passage already quoted, that cognitions à priori are independent of experience and even of all impressions on the senses.

Bearing in mind these considerations let us examine how far innate principles and à priori cognitions can be properly characterised as being thus independent. Since however a separate examination of the instances given or referred to, both by

Leibnitz and by Kant, would only weary you by a double commentary, I will direct my remarks chiefly to the latter author, with the bare intimation that they will in substance apply to his great predecessor.

For the purpose in view I will take the proposition that two straight lines cannot inclose a space*, which, according to Kant's assertion, must be an à priori cognition or judgment; and as such must be independent of experience or even of any impression on the senses.

But here I stumble, as I have no doubt you will do, at the very threshold; for it is plain that in order to form such a judgment you must have learned through your organs of sense what a straight line is, what the act of inclosing is, and what a space is. You must also have before you two definite straight lines, either parallel to each other or inclining to each other; and in either case you inevitably perceive that they do not inclose a space, just as clearly as you perceive that they are straight lines, not crooked or curved, and that they are black or coloured.

This proposition has been discussed in reference to the same part of philosophy by Dr. Whewell, Mr. John Mill, Sir John Herschell, and other writers; and on finding myself going over the same ground, I had thoughts of substituting some other proposition; but as what appears in the text was written without advertence to their dissertations, and my treatment of the question differs in several respects from that of any of my predecessors, I think it the best way to let the passage stand as originally penned.