many of the other influencing circumstances will have changed as to vitiate the experiment.":

1

The foregoing considerations suffice to show the utter inadequacy of the inductive method, in the narrower sense of that expression, as a means of solving the class of problems with which Political Economy has to deal, arising from the impossibility of employing experiment in economic inquiries under those rigorous conditions which are indispensable to give cogency to our inductions. But if Political Economy and social studies generally are placed at this serious disadvantage as compared with the various branches of physical research; on the other hand, as I shall now proceed to show, the former studies enjoy in their turn advantages peculiar to themselves,— advantages which, if duly turned to account, may perhaps be found to go some considerable way towards redressing the balance.

§ 2. Let us endeavour to realize the position of a speculator on the physical universe at the outset of physical inquiry. The most striking feature of the situation would be the extraordinary variety and complexity of the phenomena presented to his gaze, contrasted with the absence of any clear indication of the causes at work, or the laws of their operation. He would find himself in the midst of a mighty maze, possibly not without a plan, but offering to the student no apparent clue by which to thread its intricacies. No wonder

1 6

'System of Logic,' book iii. chap. x. § 8; and see for a fuller discussion of the same question, book vi. chap. vii. of the same work,

66

that in presence of such a problem the primitive thinker should have yearned for some comprehensive and all-explaining principle, and should have directed his efforts at once and by whatever means to supply this capital requirement. "For the human mind," says Bacon, strangely strains after and pants for this, that it may not remain in suspense, but obtain something fixed and immovable, on which as on a firmament it may rest in its excursions and disquisitions"-some ultimate force, some paramount and all-pervading principle, by intellectual deductions from which light may be let in among the confused and jarring elements of the world. Accordingly, it was to the attainment of some such 'Atlas for their thoughts,' that the efforts of the earliest thinkers were invariably directed. Nor were they wrong in the importance they attached to the possession of such a stand-point: only unfortunately they mistook the means of securing it, and instead of proceeding by sap and mine, endeavoured to carry the position by a coup de main. Each thinker made his guess. According to one, the ultimate principle was water; according to another, air; according to a third, number; and so the game went on through long ages; till at length the truth began to dawn that, as our knowledge of physical causes and laws-even of their existence-comes to us exclusively through observation of their physical effects, it is by way of those effects through the study of physical phenomena that the approach to the former must be made, if

1 De Aug. Scien. lib. v. cap. iv.

made at all; in other words, it began to be seen that the inductive method was the only method suitable, at all events at the outset of inquiry, to physical investigation. This truth, recognised and acted on at intervals by a few here and there, was at length proclaimed by Bacon in language which arrested the attention of the scientific world, and has become a portion of the heritage of mankind. But the point to be attended to here is that the necessity for the method of induction as the path to physical discovery arose entirely from the fact that mankind have no direct knowledge of ultimate physical principles. The law of gravitation and the laws of motion are amongst the best established and most certain of such principles; but what is the evidence on which they rest? We do not find them in our consciousness, by reflecting on what passes in our minds; nor can they be made apparent to our senses. That every particle of matter in the universe gravitates, each towards the rest, with a force which is directly according to the mass, and inversely according to the square of the distance, or that a body once set in motion will, if unimpeded by some counter force, continue for ever in motion in the same direction and with unimpaired velocity, these are propositions which can only be established by an appeal to the intellect; the proof of all such laws ultimately resolving itself into this, that assuming them to exist, they account for the phenomena. They are not the statement of any actual experiences, but, in the words of Mr. Herbert Spencer, 'truths drawn from our actual experiences, but never

66

Men culled," says

presented to us in any of them.' Dr. Whewell, "the abstract rule out of the concrete experiment; although the rule was in every case mixed with other rules, and each rule could be collected from the experiment only by supposing the others known." 1 And what is true of the laws of gravitation and of motion is true equally of all the ultimate principles of physical knowledge. Thus the undulatory theory of light, the theory of the molecular constitution of matter, the doctrine of vis inertia, all alike elude direct observation, and are only known to us through their physical effects.

The inductive method, therefore, in the narrower sense of the expression, formed the necessary and inevitable path by which, having regard to the limitation of the human faculties, physical investigation was bound, in the outset of its career, to proceed. I say in the outset of its career; because, so soon as any of the ultimate laws governing physical phenomena were established, a new path by which to approach physical problems would at once be opened. The inquirer would have secured that 'Atlas for his thoughts' for which the earlier speculators sighed; and the method of deduction—incomparably, when conducted under the proper checks, the most powerful instrument of discovery ever wielded by human intelligence-would now become possible. What, accordingly, we find in the history of the most important physical sciences, is this--a long period of laborious inductive research, during which the ground is prepared and the seed

1 Whewell's History of the Inductive Sciences,' vol. ii. p. 26.

sown, terminating at length in the discovery,-most frequently made at nearly the same time by several independent inquirers, of some one or two great physical truths; and then a period of harvest, in which, by the application of deductive reasoning, the fruits of the great discovery in the form of numerous intermediate principles connecting the higher principles with the facts of experience are rapidly gathered in. Thus the progress of mechanical science was slow, notwithstanding what had been done by Archimedes and the ancients, till the primary dynamical principles were established by Galileo and his contemporaries; but these once firmly seized and the deductive process applied to the premisses thus obtained, a crowd of minor discoveries in mechanics, hydrostatics, and pneumatics, all involved in the more fundamental principles, followed in rapid succession.' It is thus that most of those middle principles, the axiomata media of physical science, have been arrived at. But it is not in the discovery of axiomata media only that the potency of the deductive process has been exemplified. In combination with induction it has frequently been the means by which the highest physical generalizations have been reached. Of this the most eminent example is the law of gravitation itself, arrived at by Newton in the main by way of deduction from the dynamical premisses supplied by the discoveries of Galileo. In effect the problem, as it came to the hands of Newton, had assumed nearly this form-to find a force which, in conjunction and in conformity with the laws of 1 'History of the Inductive Sciences,' book vi. chaps. iii.—-vi.