of mind, is by a careful and patient comparison and observation of facts, and by a cautious induction of inferences.

4. Having noticed the process of analysis, it may be proper in this place to make a few observations on the nature of synthesis, as contrasted with it. Analysis simply signifies, unless it be in geometry, where its meaning is somewhat peculiar, a resolution of any combination, whether of matter or of thought, into its component parts. There is no science in which we have more beautiful illustrations of this process than in chemistry. A single experiment in which any substance is decomposed, and resolved into its constituent elements, impresses its nature and its use upon the mind more effectually than any other mode whatever. Here we see the several substances existing in a compound, which no mechanical force can separate, passing through a certain process by which they are as perfectly removed from each other as if they had never been combined. It is by an analogous process, also called analysis, that we must proceed in every branch of philosophy in ascertaining the simple and general laws on which the complicated phenomena of the universe depend. Before we can come

*The word is derived from avaλów, resolvo: in mathematics it denotes that process by which a proposition is traced up, through a chain of necessary dependence, to some known operation, or some admitted principle. It is alike applicable to the investigation of truth in a theorem, or the discovery of the construction of a problem. Analysis, as its name imports, is thus a sort of inverted form of solution. Assuming the hypothesis advanced, it remounts step by step, till it has reached a source already explored. The reverse of this process constitutes synthesis or composition, which is the mode usually employed for explaining the elements of science. Analysis, therefore, presents the medium of invention, while synthesis naturally directs the course of instruction."

to a conclusion, for example, as to the universality of the law of gravitation, we must previously be satisfied as to the universality of the facts on which it is founded; if the facts are precisely similar in similar circumstances, to the utmost extent of human observation, the most rigid induction entitles us to refer them to the same cause, whose operation is as general as its effects. Analysis, then, is to be considered as affording the means, in certain cases, and in reference to some sciences, of examining more fully the phenomena of nature, and of making the inductions of its laws more perfect.

The method of synthesis, as its name imports, is the opposite of analysis. As the latter consists in the solution of any thing, whether corporeal or mental, into its first elements, so the former, agreeably to its etymological signification, simply means the act of joining together, or combining into one, substances which were before separate. Of this process, also, chemistry affords us the most striking illustrations. When any two bodies, such, for example, as water and common salt, are united together in consequence of chemical attraction, so that neither of them can be separated from each other by any mechanical force, the chemist denominates the process, which has accomplished this result, synthesis.

Having obtained by analysis a knowledge of the general laws on which the phenomena of nature depend, we may safely proceed to carry on our reasonings in the way of synthesis, as to the effects resulting from any given combination of them. One of the chief advantages of this method is, to confirm the convictions

obtained by analysis. The peculiar use of the latter, is to aid our inquiries in the acquisition of knowledge; of the former, to facilitate the attainment of our designs in its communication. In proceeding according to the method of synthesis, we must be careful, in the first place, that our propositions be expressed in terms that are clearly and distinctly understood; that the words of our definitions be clearer than those they are brought to explain. Of what importance this rule is to reasonings in general, and especially to the synthetic mode of ratiocination, a slight acquaintance with the controversies that have engaged mankind, and with the way in which they have been conducted, will suf ficiently convince us, Even when the matter in dispute does not consist in a verbal ambiguity, the obscure and ill-defined language employed, exciting ideas that are only remotely, if at all, connected with the subject, diverts the mind from the avowed object of inquiry, and renders it far more difficult of clear apprehension, than at the commencement of the discussion. There is, therefore, no source of fallacy more dangerous, because more apt to be unperceived, than this, and against which it behoves every lover of truth to be constantly on his guard.

There is a second rule which we ought to observe in proceeding according to the way of synthesis; and that is, that the general truth on which our reasonings are founded be so clear and convincing, as to force the assent of any reasonable man. The general point which we assume as the basis of our discussions must be self-evident, otherwise all our reasonings must be fallacious.

The third, and only other rule which I shall at pre

sent notice, as of essential importance in synthesis, is, that all the consequences we deduce be necessarily implied in the premises from which we set out. That it is only by rigidly prescribing this rule to ourselves, in our synthetic reasonings, we can hope to keep free of error, is very obvious; and yet, owing to misapprehension, or to ignorance, or to inattention, it is often left out of view, and a chain of reasoning is formed, the strength of which is greatly impaired by the slender ties by which many of its parts are held together. We cannot expect to escape this source of fallacy without clearness and accuracy of thought, and without cultivating habits of close attention.

5. The next mode of expression in use among philosophers, of which I shall take notice, is, the explanation of a phenomenon. It is generally supposed to be the peculiar advantage of philosophy, that it enables its votaries to ascertain the causes of things. How far this notion is correct, I shall not here stop to determine. It may be observed, however, that when we are said to explain any fact or appearance of nature, we mean no more than that we shew it to be necessarily included in some phenomenon or fact already known, or supposed to be known, and we consider one phenomenon as the cause of another, when we conceive the existence of the latter to depend on some power residing in the former. According to this observation, the limits of human knowledge are fixed, the real object of philosophy is ascertained, and we are the more likely to prosecute our inquiries with success, when we are previously aware of the extent to which it is allowed us to carry them.

6. When a fact which has no other evidence of its

reality than the explanation which it affords of certain appearances, is assumed for the sake of explaining these appearances, it is called an hypothesis *. That extreme fondness in the human mind for assigning the causes of the phenomena with which it is surrounded, has, in all ages, been the occasion of substituting hypothetical reasonings for patient experiment and observation. It is, besides, more flattering to the pride and vanity of man, as well as more agreeable to his indolence, to be able to account for phenomena by an ingenious hypothesis, than by an appeal to facts. And, accordingly, there is no lesson which he is more slow to learn, than that mere ingenuity can go but a very little way in explaining the phenomena of the universe; that before he can acquire sound knowledge, he must become the scholar of experience, and carry her along with him into all his researches, and that without the humility requisite to make this prac tical attainment, the more lofty the exertions of his genius, the more injurious will they prove to the progress of real science.

"I would," says Dr. Reid, in a letter to Lord Kaimes,"I would discourage no man from conjec turing, only I wish him not to take his conjectures for knowledge, or to expect that others should do so. Conjecturing may be an useful step even in natural philosophy. Thus, attending to such a phenomenon,

Newton simply defines that to be an hypothesis which is not deduced from an observation of facts: while he maintains that conjectures should have no place in experimental philosophy. "Quicquid enim ex phænomenis non deducitur, hypothesis vocanda est. Et hypotheses seu metaphysicæ, seu physicæ, seu qualitatum occultarum, seu mechanicæ, in philosophia experimentali locum non habent."