granted, that in every syllogism, considered as an argument to prove the conclusion, there is a petitio principii." Accordingly he allows, that "no reasoning from generals to particulars can, as such, prove any thing; since from a general principle you cannot infer any particulars but those which the principle itself assumes as foreknown." "But this is in fact to say, that nothing ever was or can be proved by syllogism, which was not known, or assumed to be known, before." All real accession to our knowledge, then, must be contained in the general proposition, the major premiss. But (I. 249,) "whence do we derive our knowledge of the general truth? No supernatural aid being supposed, the answer must be, by observation." "Now all which man can observe are individual cases. From these all general truths must be drawn, and into them they may be again resolved; for a general truth is but an aggregate of particular truths; a comprehensive expression, by which an indefinite number of individual facts are affirmed or denied at once." Thus "general propositions are merely registers of such inferences already made, and short formulæ for making more. The major premiss of a syllogism, consequently, is a formula of this description; and the conclusion is not an inference drawn from the formula, but an inference drawn according to the formula; the real logical antecedent, or premisses, being the particular facts from which the general proposition was collected by induction." But if the syllogism be only an explication of what already exists in the premisses, or a test of such an explication, its office must be an altogether dependent and secondary one, and it cannot take any part in the original investigation of truth. For if Truth consists of an aggregate of facts, and if the syllogism neither collects nor aggregates the facts, clearly nothing is left for it beyond examination of the bearings and consequences of truths already elsewhere obtained. Mr. Mill accordingly assigns to the syllogism altogether a subordinate place in the system. "Its function," says he, (I. 261,) "is interpretation," and its chief use is, that it affords "a set of precautions for correctly reading the general propositions or records of facts."

To obtain general propositions, therefore, as well as particular facts, we must resort to the other branch of Inference; namely, Induction. "What Induction is, therefore, and what conditions render it legitimate, cannot but be deemed the main question of the science of logic the question which includes all others."

It is this operation (I. 352,)" by which we infer that what we know to be true in a particular case or cases, will be true in all cases which resemble the former in certain assignable respects. In other words, Induction is the process by which we conclude that what is true of certain individuals of a class, is true of the whole class; or that what is true at certain times, will be true under similar circumstances at all times." At first sight, it might seem as if Mr. Mill's own criticism on the dictum de omni et nullo, would apply to this definition : "What," says he, (I. 235,) " do we learn by being told, that whatever can be affirmed of a class, may be affirmed of every object contained in the class? The class is nothing, but the objects contained in it; and the dictum de omni merely amounts to the identical proposition, that whatever is true of certain objects, is true of each of those objects. If all ratiocination were no more than the application of this maxim to particular cases, the syllogism would indeed be, what it has so often been declared to be, solemn trifling." Now, as the dictum de omni et nullo is evidently only an inversion of Induction, as above defined, all this would seem to be true of the definition. But in order to understand what is really meant here, we must bear in mind the position above quoted, that a general truth is a mere aggregate of particular truths, whence it will follow, that a class is an aggregate of particu lar individuals. And as Particularity is here taken abstractly, as mere Diversity, there results from this postulate a principle of classification, founded not on affinity, but on diversity; for an aggregate of particulars contains no relation of affinity, but only of difference.

It is true, that where we find certain facts associated, we are inclined to suppose an affinity among them; - but this is because we (instinctively at least) suppose their being associated to depend on some internal affinity. But the Inductive Theory consistently avoids any such suppositions, as hypothetical.

Accordingly, "every class is a real kind, which is distinguished from all other classes by an indeterminate multitude of properties not derivable from another;" "while, on the contrary, differences that are merely finite and determinate, like those designated by the words white, black, or red, may be disregarded if the purpose for which the classification is made does not require attention to those particular properties." (I. 171, 167.) But a distinction in class, which we

may make or not, at our pleasure, we may acknowledge or not, when made by another; and thus where the diversity is definite, no valid distinction, according to these principles, can be made. This, however, it is unnecessary to consider, until it be shown that a definite diversity can be found in nature; — that is, that in any case, the differences between two things can be exhausted. Unless we draw a line somewhere, and declare that certain differences may be disregarded, as unimportant, (a proceeding utterly unwarrantable on the principles of this system,) it is evidently impossible to come to the end of the differences between any two acorns, or oak-leaves, or any other two things in nature. The most minute examination would only widen the field and complicate the problem. All difference, then, must be difference of class; and, as (I. 93,) no two things are the same, every object in the universe must form a class by itself: that is, Classification is impossible, except as a matter of arbitrary convention," a relation," (I. 162,) "grounded not upon what the predicate con-notes, but upon the class which it denotes," that is, upon the proper name, or what we have agreed it shall stand for, "and upon the place which, in some given classification, that class occupies relatively to the particular subject."

If, then, from certain apparent resemblances between a number of things, we form them into a class; and if, then, it be proposed to conclude from these resemblances, that a given attribute belonging to a certain individual among them, but not known to belong to the rest, does in fact belong to them, the proposition would be so far from identical, that, on the contrary, it would be altogether unfounded.

The problem of Induction, therefore, instead of a triviality, seems to be a hopeless puzzle. That we do infer general truths from particular experience, all will allow; but how this is even possible, on the principles here laid down, (much more the ground of it,) it is difficult to perceive. For any thing that appears, it may be a groundless prejudice.

It is of the utmost necessity, therefore, to discover some test, or evidence a posteriori, by which the wanting foundation may be supplied to generalization, and until this be done, the whole fabric of science must swing in air.

Some of the practical difficulties growing out of this defect in his principle Mr. Mill notices, though not the defect itself. The popular induction, he says, (I. 377,) "consists in ascrib

NO. II.

12

ing the character of general truths to all propositions which are true in every instance that we happen to know of;" — it is "simply a habit of expecting that what has been found true once or several times, and never yet found false, will be found true again." But this by no means follows. Thus, "from the earliest records, the testimony of all the inhabitants of the known world was unanimous on the point, that all swans are white." Yet this "cannot have been a good induction, since the conclusion has turned out erroneous." "The uniform experience, therefore, of the inhabitants of the known world, agreeing in a common result, without one known instance of deviation from that result, is not always sufficient to establish a general conclusion." And, we may add, if not always sufficient, in the absence of any test as to when it may be relied upon and when not, it can never be sufficient. Mr. Mill accordingly makes a distinction (I. 359, 369,) between a mere aggregation of cases, and a real induction namely, that the facts must not only be brought together, but, moreover, that "the connecting link must be some character which really exists in the facts themselves, and which would manifest itself therein if the conditions could be realized which our organs of sense require."

But this is saying, in other words, that our classification must not be conventional or accidental, but founded in the nature of things: a direct contradiction to the notion of Classification before mentioned. This contradiction is necessarily inherent in the system; for the problem of Science is to generalize particulars, and this the Inductive theory renders impossible. But let us see what the "connecting link" must be. As it is to be a character "really existing" in the facts, it must be something common to all of them. All community of attributes and all identity in principle being ab stracted, (as the exclusion of prejudice and hypothesis demands,) we have nothing left whereby to group objects, except their position in Time and Space. Avoiding the question, whether even these relations do not presuppose an ulterior principle of affinity among particulars so related, it is true, that in all our experience we find things and events occurring in a certain order, in Time and Space; - every object occupies a certain space, and every event happens in a certain time, whilst other characters may be abstracted without destroying them. One or both of these characters, then, must be the link which we seek. Kant takes both :- Mr.

For this

Mill selects position in Time; that is, Succession. selection he gives no reason; but evidently he is restricted to it by his postulate, and his negative principle of classification. Extent in Space is present and continuous; - extent in Time, on the contrary, implies diversity and succession. "We take no note of Time, but by its loss," for Time is the abstract form of Change. Things are connected, then, as being transient and successive, and this is their only general and common character: the only one which we can say really exists in all things.

66

"Of all truths relating to phenomena, the most valuable to us are those which relate to the order of their succession;" but "among all those uniformities in the succession of phenomena, which common observation is sufficient to bring to light, there are very few which have any, even apparent, pretension to this rigorous indefeasibility; and of those few, one only has been found capable of completely sustaining it." This is succession in Time, or, as he calls it, Causation. "Between the phenomena which exist at any instant, and the phenomena which exist at the succeeding instant, there is an invariable order of succession;""to certain facts, certain facts always do, and, as we believe, always will, succeed. The invariable antecedent is termed the cause; the invariable consequent, the effect." Upon the universality of this truth depends the possibility of reducing the inductive process to rules; " and this notion of Cause is "the root of the whole theory of Induction." (I. 395.) Now, as we are in search of" rigorous universality," and as the theory of Cause is introduced to give such universality to the results of Induction, it is above all necessary, as our author (I. 411,) remarks -that the succession itself should be universal and unconditional, and unless it is so, it cannot have any title to the name of Cause. Uniform experience, therefore, he says, is not suffi cient to establish the fact of Causation. But if so, it is clearly incapable of being established at all, on the Inductive theory. No conceivable method, no variation nor comparison of experiments, can ever establish the unconditionalness of a succession in Time; for to do so, the experience must be coextensive with all Time. It does not follow, because a particular succession has hitherto been invariable, that it will henceforth continue to be so. Even that we suppose or guess that it will be, is not accounted for. At all events it is a mere hypothesis.