resemblance whatever. "Of all methods of arrangement," he says, "the most antiphilosophical seems to be the invention of this age; I mean the arranging the arts and sciences, by the letters of the alphabet in Dictionaries and Encyclopædias. With these authors the categories are A, B, C," &c. Yet these literal categories, antiphilosophical as they certainly would be, if their authors professed to give them as a scientific arrangement, still involve a resemblance of some sort, however insignificant and irrelative, to the great purposes of science. Every other arrangement in science would be still more unphilosophical, because involving no relation whatever, if, according to the principles of the Nominalist, there were no general notions, no relative feelings of resemblance,-independent of the terms of classification; but objects were first classed together, without any reason for being so classed together, more than any other objects, till the mere general term of the classification became a reason for itself; as if birds, beasts, and fishes, were not called animals, because they were previously felt to agree in certain respects; but were felt to have this relation of agreement in certain respects, because they had previously been comprehended in the one generic term animal.

With respect to the origin of the general terms themselves, as distinct from the general relative feelings which they express,I stated to you a speculation of Condillac and Dr Smith, which appears to me to be one of the most simple and beautiful speculations in the theoretical history of language. In ascribing it to these distinguished philosophers, however, I speak of it only as it is clearly developed by them,-for there are many hints of the same opinion to be found in works of an earlier date. The speculation, to which I allude, is that which supposes the proper names of individual objects to have become appellatives of a whole class, by extension from similar objects to similar, the principle, which could not fail to operate in this way, being a principle, which still continues to operate, even in the common phraseology of the most common minds,-though, by rhetoricians, whose art is, in a great measure, the art of making common things mysterious, it has been advanced to the dignity of a figure of speech.

The brief expression, or result, of the feelings of resemblance, is a general term, but when all which we feel, in our relative suggestions of resemblance, or in any other of our relative sugges

tions, is enunciated in language, it is termed a proposition, which, notwithstanding the air of mystery that invests it in our books of logic, is the expression of this common feeling of relation, and nothing more. The word animal, for example, is a general term, expressive of a particular relation of resemblance that is felt by us. A horse is an animal, is a proposition, which is merely a brief expression of this felt resemblance of a horse to various other creatures, included by us in the general term. It is the same in all the other species of relations, which we are capable of feeling. In the relation of position, for example, when we say that the planet Mercury is that which is next to the sun, our mere feeling of the local relation, that particular relative suggestion which arises on the consideration of the sun, together with its planetary attendants, -by this expression of it in words, becomes, what is termed in logic, a proposition. In the relative suggestion of degree, to say that gold is heavier than copper;-in the relative suggestion of proportion, to say, that four are to twenty, as twenty to a hundred ;in the relative suggestion of comprehension, to say, that there is a portion of heat even in the coldest snow, is to state, as a proposition, what, in the mind itself, is the mere feeling of a certain relation. In all such cases, it is very evident, that the verbal statement of the proposition does not alter the nature of the relative suggestion, or feeling of relation, which it expresses, but simply expresses to others, a relation, that must have been felt, before the proposition could be framed,-that it is not the word animal, for example, which produces the feeling of the general resemblance of those various beings, which we have classed together under that term, nor the word heavier, which makes us feel the greater pressure of a piece of gold, than of an equal bulk of copper,-but those feelings, previously existing, which have led to the verbal proposition that expresses to others those previous feelings. To insist on a distinction so obvious, seems to me, indeed, almost as if I were labouring to prove, what it would be impossible for any one to deny. But if you reflect on the influence of the doctrine of the Nominalists, with respect to general terms, as constituting all that can be said to be general in reasoning, you will perceive how necessary it is, that you should be fully impressed with the priority of the relative feeling involved in each proposition, to the proposition which expresses it, and its consequent independence of those

forms of language, which render it capable of being communicated to other minds, but do not alter its nature, as a feeling of that particular mind, in which it has previously arisen.

The proposition being only an expression of a relation of some kind or other, which has been previously felt, may, of course, be as various as the species of relative suggestions of which our minds are susceptible. There may be, as we have seen, propositions of resemblance, of order, of degree, of proportion, of comprehension -to which last class, indeed—that class which includes all the relations of a whole to its parts-the others, as I have already remarked, may, by a little effort of subtilty, be reduced; since every affirmative proposition enunciates, or predicates-to use the technical word-some quality or attribute of a subject, which may be said to form a part of the very essence of the subject itself, or, at least, of our complex notion of the subject. The one quality, of which we speak, is comprehended with other qualities in that general aggregate to which we state it to belong.

On this class of our relative suggestions, therefore, that which involves the feeling of the relation of the parts comprehended to the comprehending whole-it will be necessary to bestow a little fuller illustration, that you may understand clearly the nature of the process of reasoning-that most important of all our mental processes-which logicians and metaphysicians have contrived to render so obscure, but which is in itself nothing more than a series of felt relations of this particular class in the instances which I selected before, of a house and its apartments; a tree and its stems and foliage; a horse and its head, and limbs, and trunk; the relation which I have termed the relation of comprehension, or comprehensiveness, is so very obvious, that a mere allusion to it is sufficient, without any commentary. In these cases, the parts, which together form the whole, are truly substances, that admit of being separated, and can as easily be conceived to exist separately as together.

But substances are not conceived by us, only as composed of certain elementary substances, which constitute them, by their mere juxta-position in apparent contiguity, and which may exist apart, after division. They are also conceived by us, as subjects of qualities, which co-exist in them, and which cannot exist apart, or, in other words-for the qualities of substances, as perceived

by us, are nothing more-they are capable of affecting us as sentient beings, directly or indirectly, in various ways. A flake of snow, for example, is composed of particles of snow, which may exist separately; and this composition of separate particles in seeming coherence is one species of totality; but the same snow, without any integral division, may be considered by us as possessing various qualities, that is to say, as capable of affecting us variously. It is cold, that is to say, it excites in us a sensation of chillness; it is white, that is to say, it produces in our mind a peculiar sensation of vision, by the light which it reflects to us ;it has weight—is of a certain crystalline regularity of figure-is soft or hard, according as it is more or less compressed-liquefiable at a very low temperature-and my conception of snow is of that permanent subject, which affects my senses, in these various ways. The conglomerated flakes, in a snow-ball, are not more distinctly parts of the mass itself, which we consider, than the coldness, whiteness, gravity, regular form, softness or hardness, and ready fusibility, are felt to be parts of our complex notion of snow, as a substance.

When I think of cases, in which the relation is of a substance to parts that are themselves substances-as when I say, that a room is a part of a house, or that a tree has branches-it is quite evident, that in these very simple propositions I merely state the relation of parts to a comprehending whole. But is this statement at all different in kind, when I speak in the common forms of a proposition, of the qualities of objects, when I say, for example, that snow is white, man capable of reasoning, the wisest of mankind still fallible? Do I not merely state one of the many qualities, comprehended in that totality of qualities, which constitutes the subject as known to me? I do not, indeed, divide a mass into integral parts; but I divide a complex notion into its parts; or at least separate from that complexity a quality, which I feel to belong, and state to belong, to that whole complex notion, from which I have detached it. It is as it were a little analysis and synthesis. I decompose, and, in expressing verbally to others the mental decomposition which I have made, I combine again the separated elements of my thought —not, indeed, in the same manner, for the analytic process is as different as matter is from mind-but with the same feeling of agreement or identity, which rises in the mind of a chemist, when

he has reduced to one mass the very elements into which he had previously transmuted the mass, by some one of the analyses of his wonderful art.

What, then, is reasoning-which is nothing more than a number of propositions, though of propositions consecutive in a certain order—but a continued series of analytic operations of this kind, developing the elements of our thought? In every proposition, that which is affirmed is a part of that of which it is affirmed, and the proposition, however technical in language, expresses only the single feeling of this relation. When I say snow is white, I state one of the many feelings which constitute my complex notion of snow. When I say man is fallible, I state one of the many imperfections, which, as conceived by me, together with many better qualities, constitute my complex notion of man. These statements of one particular relation are simple propositions, in each of which a certain analysis is involved. But, when I reason, or add proposition to proposition in a certain series, I merely prosecute my analysis, and prosecute it more or less minutely, according to the length of the ratiocination. When I say man is fallible, I state a quality involved in the nature of man, as any other part of an aggregate is involved in any other comprehending whole. When I add, he may therefore err, even when he thinks himself least exposed to error, I state what is involved in the notion of his fallibility. When I say, he therefore must not expect that all men will think as he does, even on points which appear to him to have no obscurity, I state that which is involved in the possibility of his and their erring even on such points. When I say, that he therefore should not dare to punish those who merely differ from him, and who may be right even in differing from him, I state what is involved in the absurdity of the expectation, that all men should think as he does. And when I say, that any particular legislative act of intolerance is as unjust as it is absurd, I state only what is involved in the impropriety of attempting to punish those who have no other guilt than that of differing in opinion from others, who are confessedly of a nature as fallible as their own.

In all this reasoning, though composed of many propositions, there is obviously only a progressive analysis, with a feeling, at each step, of the relation of parts to the whole, the predicate of