strong a belief of the constancy of the result as ten thousand. When a philosopher repeats an experiment for the sake of greater certainty, his hesitation does not proceed from any doubt, that, in the same circumstances, the same phenomena will be exhibited; but from an apprehension, that he may not have attended duly to all the different circumstances in which the first experiment was made. If the second experiment should differ in its result from the first, he will not suspect that any change has taken place in the laws of nature; but will instantly conclude, that the circumstances attending the two experiments have not been exactly the same.

It will be said, perhaps, that although our belief in this instance is not founded on a repetition of one single experiment, it is founded on a long course of experience with respect to the order of nature in general. We have learned, from a number of cases formerly examined, that this order continues uniform; and we apply this deduction as a rule to guide our anticipations of the result of every new experiment that we make. This opinion is supported by Dr. Campbell in his Philosophy of Rhetoric; but it seems to me to afford a very unsatisfactory solution of the difficulty. It plainly differs essentially from Mr. Hume's theory; for it states the fact in such a manner, as excludes the possibility of accounting for it by the association of ideas; while, at the same time, it suggests no other principle, by means of which any plausible explanation of it may be obtained. Granting, at present, for the sake of argument, that after having seen a stone often fall, the associating principle alone might lead me to expect a similar event, when I drop another stone; the question still recurs, (supposing my experience to have been

hitherto limited to the descent of heavy bodies)-Whence arises my anticipation of the result of a pneumatical, an optical, or a chemical experiment? According, therefore, to Campbell's doctrine, we must here employ a process of analogical reasoning. The course of nature has been found uniform in all our experiments concerning heavy bodies; and therefore we may conclude, by analogy, that it will also be uniform in all other experiments we may devise, whatever be the class of phenomena to which they may relate. It is difficult to suppose, that such a process of reasoning should occur to children or savages; and yet I apprehend, that a child who had once burned his finger with a candle, would dread the same result, if the same operation were to be repeated. Nor, indeed, would the case be different, in similar circumstances, with one of the lower animals.

In support of his own conclusion on this subject, Dr. Campbell asserts,*" that experience, or the tendency of "the mind to associate ideas under the notion of causes "and effects, is never contracted by one example only.” He admits, at the same time, that in consequence of the analogical reasoning which I mentioned, natural philosophers consider a single experiment, accurately made, as decisive with respect to a theory. It is evident that, upon this supposition, children, and the vulgar, must see two events often conjoined, before they apprehend the relation of cause and effect to subsist between them; whereas the truth is, that persons of little experience are always prone to apprehend a constant connection, even when they see a merely accidental conjunction. So firmly are they per

No. 1. p. 137.

suaded, that every change requires a cause, and so eager to discover it, that they lay hold of the event immediately preceding it, as something on which they may rest their curiosity; and it is experience alone that corrects this disposition, by teaching them caution in investigating the general laws which form a part of the order of the universe. *

From these observations, it seems to follow, that our expectation of the continuance of the laws of nature is not the result of the association of ideas, nor of any other principle generated by experience alone; and Mr. Hume has shown, with demonstrative evidence, that it cannot

*The account which is given in the Encyclopædia Britannica of the conclusiveness of a single experiment in proof of a general law of nature is, at bottom, the very same with the theory of Campbell; and therefore a separate consideration of it is unnecessary. This will appear evident from the following extract.

"EXPERIMENTAL PHILOSOPHY seems, at first sight, in direct op"position to the procedure of nature in forming general laws." (The expression here is somewhat ambiguous; but the author plainly means, in opposition to the natural procedure of the mind, in the investigation of general laws.) "These are formed by induction "from multitudes of individual facts, and must be affirmed to no "greater extent than the induction on which they are founded. Yet "it is a matter of fact, a physical law of human thought, that one "simple, clear, and unequivocal experiment, gives us the most "complete confidence in the truth of a general conclusion from it “to every similar case. Whence this anomaly? It is not an anomaly, "or contradiction of the general maxim of philosophical investiga❝tion, but the most refined application of it. There is no law more "general than this, that nature is constant in all her operations.' "The judicious and simple form of our experiment insures us (we "imagine) in the complete knowledge of all the circumstances of the "event. Upon this supposition, and this alone, we consider the ex"periment as the faithful representative of every possible case of "the conjunction."(Article Philosophy, § 57. See also (in the same volume) article Physics, § 103.)

be resolved into any process of reasoning a priori. Till, therefore, some more satisfactory analysis of it shall ap pear than has yet been proposed, we are unavoidably led to state it as an original law of human belief. In doing so, I am not influenced by any wish to multiply unnecessarily original laws or ultimate truths; nor by any apprehension of the consequences that might result from an admission of any one of the theories in question. They are all of them, so far as I can see, equally harmless in their tendency; but all of them equally unfounded and nugatory, answering no purpose whatever, but to draw a veil over ignorance, and to divert the attention, by the parade of a theoretical phraseology, from a plain and most important fact in the constitution of the mind.

In treating of a very different subject, I had occasion, in a former work,* to refer to some philosophical opinions of Mr. Turgot, coinciding nearly with those which I have now stated. These opinions are detailed by the author, at considerable length, in the article Existence of the French Encyclopedie; but a conciser and clearer account of them may be found in Condorcet's discourse, prefixed to his essay "On the application of analysis to the probability "of decisions pronounced by a majority of votes." From this account it appears, that Turgot resolved "our belief

of the existence of the material world" into our belief of the continuance of "the laws of nature;" or, in other words, that he conceived our belief, in the former of these instances, to amount merely to a conviction of the established order of physical events; and to an expectation that, in the same combination of circumstances, the same

Philosophy of the Human Mind, chap. iv. sect. 5.

event will recur. It has always appeared to me, that something of this sort was necessary to complete Dr. Reid's speculations on the Berkeleian controversy; for although he has shown our notions concerning the primary quali ties of bodies to be connected, by an original law of our constitution, with the sensations which they excite in our minds, he has taken no notice of the grounds of our belief that these qualities have an existence independent of our perceptions. This belief (as I have elsewhere observed*) is plainly the result of experience; inasmuch as a repetition of the perceptive act must have been prior to any judgment, on our part, with respect to the separate and permanent reality of its object. Nor does experience itself afford a complete solution of the problem; for, as we are irresistibly led by our perceptions to ascribe to their objects a future as well as a present reality, the question still remains, how are we determined by the experience of the past, to carry our inference forward to a portion of time which is yet to come? To myself the difficulty appears to resolve itself, in the simplest and most philosophical manner, into that law of our constitution to which Turgot, long ago, attempted to trace it.

If this conclusion be admitted, our conviction of the permanent and independent existence of matter is but a particular case of a more general law of belief extending to all other phenomena. The generalization seems to me to be equally ingenious and just; and while it coincides perfectly in its spirit and tendency with Reid's doctrine on the same point, to render that doctrine at once more precise and more luminous.

* Philosophy of the Human Mind, chap. iii.