not.

make such a declaration, and you will learn why we canThe difficulty will be just what you experience. To this experience language refers, and further, words have no signification. We may amuse ourselves with framing propositions such as Locke's, but we mean nothing but what our senses discover.

But why, says Locke, does no body ever think of infinite sweetness or infinite whiteness, though he can repeat the idea of sweet and white as frequently as those of a yard or a day? Because, answers Locke, only those ideas which have parts are capable, by repetition, of producing the idea of infinity. But why? Because, says Locke, with this endless repetition of ideas there is a constant enlargeBut why? Locke does not answer. I will answer for him. The words refer to our operations, and the phenomena with which we are conversant. be longer if you add to it another stick. Why? Make the experiment and you will discover. The necessity has no other reference: and hence the absurdity of using it where no phenomena arc discoverable.

ment.

This stick must

Why does no body think of infinite swectness, or infinite whiteness, though he can repeat the idea of sweetness and whiteness as frequently as the idea of a stick? Because, says Locke, to the idea of the whitest whiteness, if I add another of a less or equal whiteness, it makes no increase or enlargement of my idea. Why? Because, he remarks, if you take the idea of white, which was yielded yesterday by a parcel of snow, and join it in your mind with the idea of whiteness that is yielded to-day by another parcel of snow, the two ideas embody into one, and the idea of whiteness is not increased. But why? He The answer is, however, extremely simple,

answers not.

and shows that language has no meaning when it does not refer to sensible phenomena. Why, then, cannot one piece of snow be made whiter by the addition of another piece? Conjoin them and you will discover. This is a simple reason, but no other is so good. The term “cannot" refers to this experiment, and not to verbal reasons. They have neither authority nor significance, when they are used without a reference to phenomena.

It appears, then, there is no length which may not be increased, though there is a limit to the increase of whiteness. The process is thus announced by Locke: "Every person who has an idea of a foot, finds that he can repeat the idea; and joining it to the former, make the idea of two feet, and so on without ever arriving at an end of his increase, whether the idea so enlarged be a foot or a mile, or the diameter of the earth, or the orbis magnus.”

I would ask, however, what he enlarges? So long as he speaks of joining one foot to another, he speaks significantly; but when he talks of enlarging his idea by doubling the diameter of the carth, the process becomes verbal, and the necessity which compels us to admit the enlargement, has no existence but in the forms of language :— forms that owe their significance to their reference to phe. nomena, and become insignificant the moment they are applied where no corresponding phenomena are discoverable. To enlarge in infinitum, and to diminish in infinitum, are processes of the same character; they are words which have no archetype among sensible phenomena, and are therefore sounds significant of nothing.

In Gill's Body of Divinity is the following proposition : "Though angels have no bodies, and so are not in place

circumspectively; yet, as they are creatures, they must have a somewhere in which they are definitively."

Why must creatures have a place in which they exist? Because the words refer to sensible phenomena. This book is a creature. If you attempt to dispose of the book so that it shall exist, and still have no location, you will discover the impracticability. It will be what you experience. But when the same impracticability is predi cated of angels, it exists only in the forms of language ; forms which have no more substantiality, when the phenomena to which they allude are subtracted, than the muster-rolls of an army, when the soldiers have all deserted.

The writer proceeds with his metaphysical-discoveries: "where was a place for angels to exist in before heaven and earth were made? No where." Why? Because we are again referring to the phenomena with which we are conversant, and language can have no other reference. The writer, however, thinks his reasoning is conclusive, that the heavens or the earth must have been created before angels. Yet even this obvious consequence of the premises is authoritative only because it refers to our operations thus, you cannot mark with chalk till you have something on which to inscribe the mark. Why? Try and you will find. The difficulty in this case is not logical, but a phenomenon of nature. The phenomenon affixes to the inability a signification, but without the phenomenon the inability is verbal only: it is unmeaning.

Locke says, "number applies to men, angels, actions, thoughts, and every thing imaginable." If any proposition is inherently significant and independent of phenomena, this of Locke must be the one. Yet even this is in

debted for its significance to our operations and· èxperience. Why must apples be either one or more? Try to prevent the necessity and you will discover. The necessity depends not on the structure of language, but on the phenomenon. But why must angels be either one or more? The necessity here refers to no phenomenon, and is merely verbal. Number may be inapplicable to angels. It is a name given by us to certain sights and feels, &c.; where these exist not, number is a word divested of its signification. Suppose we were to apply numbers to darkness, insipidity, or vacuity, we should speak unintelligibly; because these objects exhibit not the sights and feels to which numbers are ordinarily applied. This illustration may assist you to apprehend that where all the phenomena with which we are conversant should be absent, the word numbers would have no signification ; hence it may not be applicable to angels. We affix it to them in compliance with the forms of language.

I have now shown, that when language forces us to admit any thing, (as in the above instance, that apples must be either one or more,) the necessity of admitting the conclusion is founded on our experience. I have also shown that when propositions have thus obtained an authoritative character, we apply them where there are no corresponding phenomena: as that angels must be either one or more; and that, in such applications, the necessity of admitting the conclusion is merely verbal, and therefore fallacious.

The present lecture has been particularly devoted to the latter elucidation. Examples of the error might be accumulated without difficulty, but I have probably stated a sufficient number and variety to show that the error enters

1

deeply into all our learning. We shall now be able to discover a reason for the great solicitude evinced by abstruse writers about names and definitions. For instance, if a mathematician wishes to demonstrate that the surface of a fish-pond is not level, it is important that the earth should be denominated a sphere; because, after this preliminary, and a suitable definition of the term, it follows that as the fish-pond constitutes a part of the circumference of a sphere, the surface of the water cannot be a straight line.

This, then, is the reason why we find amongst abstract writers so much labour in the definition of the names on which their theories are to be erected. The investigation of this subject is important to the view which I wish to present of language, and it ́constitutes the theme of our next Lecture.