of propositions involving general terms. But the observations he has made, (admitting them in their full extent,) do not in the least affect the question about the necessity of signs, to enable us to speculate about such propositions. The vague use which metaphysical writers have made of the word conception, (of which I had occasion to take notice in a former chapter,) has contributed in part to embarrass this subject. That we cannot conceive universals in a way at all analogous to that in which we conceive an absent object of sense, is granted on both sides. Why then should we employ the same word, conception, to ex press two operations of the mind which are essentially different? When we speak of conceiving or understanding a general proposition, we mean nothing more than that we have a conviction, (founded on our previous use of the words in which it is expressed,) that we have it in our power, at pleasure, to substitute, instead of the general terms, some one of the individuals comprehended under them. When we hear a proposition announced, of which the terms are not familiar to us, we naturally desire to have it exemplified, or illustrated, by means of some particular instance; and when we are once satisfied by such an application, that we have the interpretation of the proposition at all times in our power, we make no scruple to say, that we conceive or understand its meaning, although we should not extend our views beyond the words in which it is announced, or even although no particular exemplification of it should occur to us at the moment. It is in this sense only, that the terms of any general proposition can possibly be understood; and therefore Dr. Reid's argument does not, in the least, invalidate the doctrine of the Nominalists, that, without the use of language, (under which term I comprehend every species of signs,) we should never have been able to extend our speculations beyond individuals.

That in many cases, we may safely employ in our reasonings general terms, the meaning of which we are not even able to interpret in this way, and consequently, which are to us wholly insignificant, I had occasion already to demonstrate, in a former part of this section.

IV. Inferences with respect to the use of language as an instrument of thought, and the errors in reasoning to which it occasionally gives rise. In the last section, I mentioned Dr. Campbell as an ingenious defender of the system of the Nominalists; and I alluded to a particular application which he has made of their doctrine. The reasonings which I had then in view, are to be found in the seventh chapter of the second book of his Philosophy of Rhetoric; in which chapter, he proposes to explain how it happens, "that nonsense so often escapes being detected, both by the writer and the reader." The title is somewhat ludicrous in a grave and philosophical work; but the disquisition to which it is prefixed contains many acute and profound remarks on the nature and power of signs, both as a medium of communication, and as an instrument of thought.

Dr. Campbell's speculations with respect to language as an instrument of thought, seem to have been suggested by the following passage in Mr. Hume's Treatise of Human Nature. "I believe, every one who examines the situation of his mind in reasoning will agree with me, that we do not annex distinct and complete ideas to every term we make use of; and that in talking of government, church, negotiation, conquest, we seldom spread out in our minds all the simple ideas of which these complex ones are composed. It is, however, observable, that notwithstanding this imperfection, we may avoid talking nonsense on these subjects; and may perceive any repugnance among the ideas, as well as if we had a full comprehension of them. Thus if, instead of saying, that, in war, the weaker have always recourse to negotiation, we should say, that they have always recourse to conquest; the custom which we have acquired, of attributing certain relations to ideas, still follows the words, and makes us immediately perceive the absurdity of that proposition."

How we can use words with propriety, though without attaching any particular idea or signification to them. In the remarks which Dr. Campbell has made on this passage, he has endeavored to explain in what manner our habits of thinking

and speaking gradually establish in the mind such relations among the words we employ, as enable us to carry on processes of reasoning by means of them, without attending in every instance to their particular signification. With most of his remarks on this subject I perfectly agree; but the illustrations he gives of them, are of too great extent to be introduced here; and I would not wish to run the risk of impairing their perspicuity, by attempting to abridge them. I must, therefore, refer such of my readers as wish to prosecute the speculation, to his very ingenious and philosophical treatise.

"In consequence of these circumstances," says Dr. Campbell, "it happens that, in matters which are perfectly familiar to us, we are able to reason by means of words, without examining, in every instance, their signification. Almost all the possible applications of the terms (in other words, all the acquired relations of the signs,) have become customary to us. The consequence is, that an unusual application of any term is instantly detected; this detection breeds doubt, and this doubt occasions an immediate recourse to ideas. The recourse of the mind, when in any degree puzzled with the signs, to the knowledge it has of the things signified, is natural, and of such subjects perfectly easy. And on this recourse, the discovery of the meaning, or of the unmeaningness, of what is said, is the immediate effect. But in matters that are by no means familiar, or are treated in an uncommon manner, and in such as are of an abstruse and intricate nature, the case is widely different." The instances in which we are chiefly liable to be imposed on by words without meaning are, (according to Dr. Campbell,) the three following:

First, Where there is an exuberance of metaphor.

Secondly, When the terms most frequently occurring denote things which are of a complicated nature, and to which the mind is not sufficiently familiarized. Such are the words, government, church, state, constitution, polity, power, commerce, legislature, jurisdiction, proportion, symmetry, elegance.

Thirdly, When the terms employed are very abstract, and

consequently of very extensive signification.* For an illustration of these remarks, I must refer the reader to the ingenious work which I just now quoted.

.

To the observations of these eminent writers, I shall take the liberty of adding, that we are doubly liable to the mistakes they mention, when we make use of a language which is not perfectly familiar to us. Nothing, indeed, I apprehend, can show more clearly the use we make of words in reasoning than this, that an observation which, when expressed in our own language, seems trite or frivolous, often acquires the appearance of depth and originality, by being translated into another. For my own part, at least, I am conscious of having been frequently led, in this way, to form an exaggerated idea of the merits of ancient and of foreign authors; and it has happened to me more than once, that a sentence, which seemed at first to contain something highly ingenious and profound, when translated into words familiar to me, appeared obviously to be a trite or a nugatory proposition.

The effect produced by an artificial and inverted style in our own language, is similar to what we experience when we read a composition in a foreign one. The eye is too much dazzled to see distinctly.

The deranged collocation of the words in Latin composition, aids powerfully the imposition we have now been considering, and renders that language an inconvenient medium of philosophical communication, as well as an inconvenient instrument of accurate thought. Indeed, in all languages in which this

"The more general any word is in its signification, it is the more liable to be abused by an improper and unmeaning application. A very general term is applicable alike to a multitude of different individuals, a particular term is applicable but to a few. When the rightful applications of a word are extremely numerous, they cannot all be so strongly fixed by habit, but, that, for greater security, we must perpetually recur in our minds from the sign to the notion we have of the thing signified; and, for the reason aforementioned, it is in such instances difficult precisely to ascertain this notion. Thus, the latitude of a word, though different from its ambiguity, hath often a similar effect." - Philosophy of Rhetoric, vol. ii. p. 122.

latitude in the arrangement of words is admitted, the associations among words must be looser than where one invariable order is followed; and of consequence, on the principles of Hume and Campbell, the mistakes which are committed in reasonings expressed in such languages will not be so readily detected.

Languages, being controlled by popular use, are not adapted for scientific purposes. The errors in reasoning to which we are exposed, in consequence of the use of words as an instrument of thought, will appear the less surprising, when we consider that all the languages which have hitherto existed in the world, have derived their origin from popular use; and that their application to philosophical purposes was altogether out of the view of those men who first employed them. Whether it might not be possible to invent a language which would at once facilitate philosophical communication, and form a more convenient instrument of reasoning and of invention than those we possess at present, is a question of very difficult discussion, and upon which I shall not presume to offer an opinion. The failure of Wilkins's very ingenious attempt towards a real character and a philosophical language, is not perhaps decisive against such a project; for not to mention some radical defects in his plan, the views of that very eminent philosopher do not seem to have extended much further than to promote and extend the literary intercourse among different nations. Leibnitz, so far as I know, is the only author who has hitherto conceived the possibility of aiding the powers of invention and of reasoning, by the use of a more convenient instrument of thought: but he has nowhere explained his ideas on this very interesting subject. It is only from a conversation of his with Mr. Boyle and Mr. Oldenburgh, when he was in England, in 1673, and from some imperfect hints in different parts of his works, that we find it had engaged his attention. In the course of this conversation, he observed, that Wilkins had mistaken the true end of a real character, which was not merely to enable different nations to correspond easily together, but to assist the reason, the invention, and the memory. In his writings, too, he somewhere speaks of an alpha