It

intimation of its future constancy. This irresistible persuasion comes to us from another quarter. forms a distinct principle in the frame or workmanship of our intellectual system. It is a befitting

theme of gratitude and wonder that this instinctive faith from within, should be responded to by the unexcepted fulfilment of Nature's actual and abiding constancy from without. But the one is

not a derivative from the other.

The two are in harmony-but it is a contingent harmony.

3. The use of experience is not to strengthen our faith in the constancy of Nature's sequences—— but to inform us what the sequences actually are. We do not need to be made surer than we are already that the progressions of Nature are invariable but we need to learn the steps of each progression. As far as we can discover of the human mind, it counts, and has at all times from its earliest capacities of thought, counted on the same antecedents being followed up by the same consequents. It is not the office of experience to lesson us into this confidence. But experience is indispensable to teach us,-which be the causal antecedents and which be the consequents related to them by the tie of invariableness, in those successions that are taking place around us. Our object in the repetition of an experiment is not to be made sure that what Nature has done once in certain circumstances, she will in the same circumstances do again. But it is to ascertain what the circumstances really be which are essential to the result in question. The truth is, that in that assemblage of circumstances which precedes some certain

For

event, there may only be one or so many of them that have causal influence upon the result, and the rest may be mere accompaniments whose presence is not necessary to the production of it. It is to distinguish the causal antecedents from the merely casual ones, that an experiment has often to be varied or done over again. It is not that we ever have the least suspicion of Nature as if she fluctuated in her processes. But it is to disentangle these processes from that crowd of accessaries, wherewith they are at times beset or encompassed, that we have so repeatedly to question her. this purpose we withdraw certain ingredients from the assemblage. We supply certain others. mix them up in various proportions—and all this, not to strengthen our belief in the regularity of Nature-but to discover what the trains or successions are, according to which this regularity proceeds. We are not sure that the instinct by which we are led to anticipate the same result in the same circumstances is stronger in manhood than in infancy. But in manhood we know the result and we know the circumstances. This seems the whole fruit of experience. It teaches not the strength or invariableness of the connexion that runs through all nature—but it teaches the terms of that connexion.

We

4. And it is instructive to observe the real process of an infant's mind, during that education by which it becomes acquainted with surrounding nature. When it strikes the wooden table with a spoon, it needs not repeat the stroke for the purpose of obtaining a surer or firmer expectation of the consequent noise. That expectation is probably as

confident at the first as afterward-and it is of importance to remark, that at the outset of its experience it is quite general and indiscriminate. For instance, it would anticipate the same noise by striking the spoon against any surface whatever, as when placed on a carpet, or on the level of a smooth sandy beach. Originally it would expect the same noise by striking on a soft yielding substance that it did by striking on the hard tableand the office of experience is not to strengthen its hope of a similar result from a similar act, but in truth to correct the exuberance of that hope. It is to teach it discrimination-and how in the midst of a general resemblance, to mark those minuter differences which in fact present it with antecedents that are really different, and which should lead it to expect results that are different also. It is thus that the primary undirected and diffused expectation, of meeting again with what it met once in the act of striking with a spoon on a wooden surface, comes afterwards to be modified. It learns -not that there is a surer tie between the terms of nature's sequences than it imagined at the first -but it learns how to distinguish between the terms which are really different, though before it had vaguely confounded them. And so it is taught

with each distinct antecedent to look for a distinct consequent instead of expecting the same noise by the infliction of a stroke upon all surfaces, to expect no noise at all by a stroke upon the sand, and different sorts of noises by a stroke on different surfaces, whether wood or metal or stone or liquid.*

* This phenomenon of the infant mind will be found not only

5. Now this may explain how it is, that our faith in the constancy of nature appears to grow with experience and that, notwithstanding the obvious

to throw light on the origin and progress of our belief in testimony, but to accord with the surmises both of Dugald Stewart and Turgot, when they approximate to what we have long regarded as the true account or philosophy of the process described by the human mind in the formation of abstract and general ideas. The truth is that our disposition to generalize by noticing the points of resemblance between different objects, often takes the precedency of our disposition to specialize by noticing their points of distinction or dissimilarity-and so, at the commencement of our mental history, we are liable to confound when we ought to discriminate. This observation rightly applied will be found to correct both the philosophy of Dr. Campbell and the scepticism of Hume on the subject of human testimony.

The following sentences from Dugald Stewart and from Condorcet's life of Turgot will evince the existence of this thought in embryo, or as an undeveloped germ in the minds of both these philosophers.

"This remark becomes, in my opinion, much more luminous and important, by being combined with another very original one, which is ascribed to Turgot by Condorcet, and which I do not recollect to have seen taken notice of by any later writer on the human mind. According to the common doctrine of logicians, we are led to suppose that our knowledge begins in an accurate and minute acquaintance with the characteristical properties of individual objects; and that it is only by the slow exercise of comparison and abstraction, that we attain to the notion of classes or genera. In opposition to this idea it was a maxim of Turgot's, that some of our most abstract and general notions are among the earliest we form. What meaning he annexed

to this maxim, we are not informed; but if he understood it in the same sense in which I am disposed to interpret it, he appears to me entitled to the credit of a very valuable suggestion with respect to the natural progress of human knowledge. The truth is, that our first perceptions lead us invariably to confound together things which have very little in common; and that the specifical differences of individuals do not begin to be marked with precision till the powers of observation and reasoning have attained to a certain degree of maturity. To a similar indistinctness of perception are to be ascribed the mistakes, about the most familiar appearances which we daily see committed by those domesticated animals, with whose instincts and habits we have an opportunity of becoming intimately acquainted. As an instance of this, it is sufficient to mention the terror which à horse

strength of this principle in very early childhood. After an infant has once struck the table with a spoon and elicited the noise which it likes, it proceeds with all confidence to repeat the stroke-not on the table only, but on other substances, in expectation of a similar noise to that which had pleased and gratified it before. But it is speedily checked in this expectation. It learns that with every difference in the antecedent circumstances, there may be a difference in the result-and it further learns that there may often be real differences which escape its observation. Now the longer it has been accustomed to witness the same phenomenon in the same ostensible circumstances, it becomes the more confident that these are the only essential circumstances to the result, or at least that the ostensible circumstances always involve the essential or the real ones. Should it awake in the morning, and perceive the nurse or mother by its side and smiling over it-then were there but a moment of prior consciousness, and the recollection of what had happened yesterday, it might on the next morning open its eyes with the

sometimes discovers in passing, on the road, a large stone, or the waterfall of a mill."-Stewart's Philosophy of the Human Mind, Vol. II., p. 242-4.

M. Turgot croyoit qu'on s'était trompe en imaginant, qu'en général l'esprit n'acquiert des idées générales ou abstraites que par la comparaison d'idées plus particulières. Au contraire, nos premières idées sont tres-générales, puisque ne voyant d'abord qu'un petit nombre de qualités, notre idée renferme tous les êtres auxquels ces qualités sont communes. En nous éclairant en examinant d'avantage, nos idées deviennent plus particulières sans jamais atteindre le dernier terme; et ce qui a pu tromper les métaphysiciens, c'est qu'alors précisément nous apprenons que ces idées sont plus générales que nous ne l'avions d'abord supposé."-Vie de Turgot, p. 159. Berne, 1787.