ART. I.-EXAMINATION OF MILL'S PHILOSOPHY
OF NECESSARY TRUTH, AND OF CAUSATION.

A System of Logic, Ratiocinative and Inductive; being a con-
nected view of the principles of Evidence, and the methods of
Scientific Investigation. By JOHN STUART MILL. New York:
Harper and Brothers.

MR. MILL'S elaborate treatise has already been made the subject of a careful and extended notice, in our columns; with especial reference, however, to the principles of logic, and the theory of induction, which it develops. We propose now, to offer some observations upon a different aspect of the work, in the form of an examination of the metaphysical theories embodied in it and as we must not unnecessarily prolong an effort, which we fear will be somewhat trying to the patience of our readers, we plunge at once, with only these few prefatory words, into the midst of the subject we have indicated.

One of the principal characteristics of Mr. M.'s work, viewed as an exposition of metaphysical theories, is its earnest opposition to the commonly received doctrine of necessary truths. In this doctrine Mr. M. is an unbeliever; and he enters into a very extended and profound examination of the subject, with the design of showing that all the truths most generally alledged as examples of our necessary belief, are in fact only generalizations from a wide induction of familiar facts. Among these, he includes the truths relating to quantity; and he points out, with a patience and ingenuity of effort which are almost amusing, the process by which we arrive at the general conclusion that two units and one unit, are in all cases identical with three units. The discussion is at length made to turn, by consent of himself and his oppo

VOL. VIII.

21

nents, upon the question whether the truth that two straight lines can not enclose a space, is a necessary inference from our conception of the nature of straight lines, or a simple generalization from our experience. The advocates of the former view, (among whom Prof. Whewell of Cambridge stands prominent,) maintain that we derive this truth not from any examination of external nature, but from consciousness; and that we perceive intuitively, the impossibility alledged in the statement of it.

This appeal to consciousness Mr. Mill evades, by a view which certainly has much plausibility. He maintains that the nature of lines and angles is in itself so entirely simple, that we easily form imaginary pictures, corresponding accurately to the lines and angles which exist in nature. Hence he concludes that the intuition to which his opponents appeal, is nothing more than the inspection of an imaginary diagram. We form such a picture in the mind, examine it, and pronounce upon it, by a process which is as real an induction from particulars of experience, as it could be were the diagram drawn upon paper.

To the other principal argument of Mr. Whewell, that the affirmation is seen by every one to be necessarily true, inasmuch as we can not even conceive of two straight lines which shall enclose a space, Mr. M. replies in an argument of several pages, not by denying, but only by seeming to deny, it. The substance of his reply is that many things have been at various periods pronounced inconceivable, which have been afterwards ascertained to be actual facts, such for instance as the existence of antipodes, &c.; leaving his readers to believe, what he hesitates to affirm, that the conception of two straight lines which shall enclose a space, may hereafter become as conceivable and familiar a one as that just mentioned. Had Mr. Mill maintained that the truth in question stands in his opinion entirely on a level with the examples he gives, in which the alledged inconceivableness of the thing denied was a mere prejudice arising from the novelty of the subject, we should have felt bound to attempt a refutation of his views. But, while he is careful not to intimate even an actual doubt of the truth of his opponent's position, that the necessity of the axiom is universally perceived, we can not think any refutation necessary. By Mr. Mill's virtual concession, the truth does not stand on the same ground with those he has adduced. It is a necessary truth; the supposition of its falsity is not, in Mr. M.'s opinion, an admissible one in the nature of things. To what purpose is it then to refer to instances in which novelties have been hastily and erroneously pronounced inconceivable? To test the case, let us suppose that such an exception should be alledged to exist in the case of this axiom, as is found in some others of the general truths of observation; what would our author say of the allegation? It is a law that all matter expands

in heating and contracts in cooling, to which water—within the limits of 38 to 32-forms an isolated exception; it expands in cooling and contracts in heating. Suppose it now announced that two straight lines after crossing one another at an exceedingly minute angle, do actually meet again and enclose a space. The angle, from its minuteness, would be incapable of ordinary observation, and of course equally incapable of any imaginary representation; yet who does not feel that the assertion of such a discovery would be simply ridiculous? Plainly the case is inconceivable, even for purposes of argument. To say that two lines which meet, diverge, and then meet again, are straight lines, is to say what every one instantly revolts at, without the smallest respect to any degrees, limits or exceptions whatever. Such a decision too, is the more remarkable, as it is made not on the ground of experience, but rather, in opposition to it. So far as our senses teach us any thing on the subject, they teach that lines which are to all appearance straight, do meet after touching. The straightest objects in nature, if placed side by side, and held up to the sun, will admit light through visible spaces. Indeed, if we did not feel absolutely able to correct our experience by our necessary convictions, we should be forced to the belief that two straight lines may enclose spaces, since lines whose divergence from absolute rectitude we can not possibly detect, do actually enclose minute spaces. In such instances all experience gives way instantly to the necessary conviction that the fact of enclosing a space is, a priori, conclusive-a conviction which overrules all observation.

The axiom then differs from a truth of induction in a particular which seems decisive of debate. If induction had established the truth, it could establish also the exception; but this being absolutely out of the question, the theory of an inductive origin of this axiom seems negatived.

The argument then falls back on the consideration before stated, namely, that many things have been deemed inconceivable which were nevertheless entirely true. In connection with this argument, our author advances an explanation of the apparent necessity of some truths, which is worthy of some attention. He maintains, (Bk. II, ch. v, § 4,) that all associations of ideas which are very general, acquire a strength which renders it difficult to break them; and come at last to be regarded as necessary truths. "When we have often seen and thought of two things together, and have never in one instance either seen or thought of them separately, there is by the primary laws of association an increasing difficulty, which in the end becomes insuperable, of conceiving the two things apart."

To this view we should scarcely feel disposed to take exception, were it merely a strong statement of the doctrine of the as

sociation of ideas; but viewed as the enunciation of a philosophical truth it seems essentially inaccurate. We can not help thinking that associations of the most intimate character are frequently broken, without any such difficulty, in cases in which the elements are connected only by this habitual association of them.

No association, for example, can be more absolutely invariable, than that by which we connect the idea of time with the succession of our mental phenomena: yet few of those who have speculated upon the Divine existence, have been willing to consider the phenomena of the Infinite mind as necessarily sustaining any such relation. The association, however, is of the most absolute kind ; we have never seen, we have never thought of any event originating in human experience, which did not take place in time. So we have never conceived of a human spirit which did not sustain a certain familiar relation to space; but it is very common to withhold all positive assertion of the necessity of such a relation in our future state. We have never known any being or thing which had not its origin at some definite period; but there is no tendency among men to affirm the absolute necessity of such origin in respect to all being. It would appear then, that associations uniform, and invariable in the highest degree, may exist, without involving any such insuperable difficulty as Mr. Mill asserts, of separating the associated elements.

stance.

On the other hand, it is equally obvious upon a careful inspection of them, that it is not from such uniform association that our ideas of necessary truth originate. Probably no conception would be pronounced, by those who contend against Mr. Mill, more absolutely necessary, than that of substance; no belief more necessary than that all phenomena stand related to some subWe necessarily refer all qualities which we observe, to some substance which in itself is imperceptible. There is here no such association as Mr. M. assigns for the origin of our necessary belief. We have not been accustomed to observe substances associated with qualities, till we have learned to think the association a necessary one. Quite the contrary; the idea of substance in the world without us is a pure fiction of our conceptive faculty. We have never perceived it at all; we are satisfied that we never can perceive it. Upon Mr. M.'s theory, therefore, there exists absolutely nothing out of which this conception could be formed. The theory furnishes therefore a very insufficient account of the origin of necessary truths. In one class of instances, it seems palpably erroneous; the conditions, which it assigns as the ground of such beliefs, do not exist; there is no such association of ideas as it affirms. In the other class, in which the conditions do exist, the result does not follow; there is no such insuperable difficulty in separating the elements habitually asso

ciated, as alone would give the idea of a necessary connection between them.

Indeed Mr. Mill himself does not uniformly carry out his analysis of these elements with any consistency. When there is a question of our necessary belief that space and time are infinite, he at once points to the familiar association of ideas which explains the difficulty. "We never," he says, "saw any object without something beyond it, nor experienced any feeling without something following it. When therefore we attempt to conceive the last point of space, we have the idea irresistibly raised of other points beyond it. When we try to imagine the last instant of time, we can not help conceiving another instant after it."

Now if this were all the truth, and the "apparent infinity" of these ideas, as Mr. Mill conveniently designates it, is but the limitation arising from our weakness, the impossibility ought to be equally great in all instances where the association is equally universal. We never saw an event which was not uniformly dependent upon previous antecedents: and if such an association is all the ground which there is for our belief in the one case, there ought to be the same "apparent impossibility," when we attempt to conceive of events occurring without such antecedents. This, however, Mr. Mill is by no means disposed to admit. He maintains that any one who will carefully make the attempt, will find there is no inherent impossibility of conceiving that in other portions of the universe, events may succeed one another quite at random. Thus the same state of facts, the same unbroken association of ideas, becomes the ground of conclusions quite opposite; and authorises us to affirm the possibility of conceiving a limitation of causation, and the impossibility of conceiving any limitation of space.

The fact that the idea of necessity is apt to connect itself with established uniformities in nature, should undoubtedly induce caution in pronouncing upon necessary truths; but a vague circumstance like this affords certainly very slender ground for argument against all necessity of truth. It should induce the philosopher to examine, and to express, with the utmost care, whatever truths he propounds as necessary; and to fortify his positions by ample appeals to the individual, and the universal consciousness. But when such appeal is fully made, and consciousness universally affirms an absolute impossibility of the opposite conception, then the objection that many facts have been erroneously pronounced inconceivable, seems entirely out of place. This reference to the errors of the past, not as a reason for careful scrutiny of our reasonings, but for the purpose of forestalling conclusions which otherwise must be held to be logically established, is scientifically inadmissible.