general laws of the tides may be established, and previsions which shall be almost entirely correct may be founded upon these laws. This is what is meant, or ought to be meant, when sciences which are not exact sciences are spoken of. Astronomy was a science before it became an exact science. It has only been exact since it has explained not only the planetary motions, but also their perturbations.

The science of the tides is not yet an exact science, not by a radical impossibility in its nature, but because it is very difficult to establish the derivative uniformities with precision. 'The science of human nature is of the same kind.' It is far from being of the same exactness as our present astronomy, but there is no reason why it should not be a science like that of the stars, or even such as astronomy was, while as yet its calculations included only the principal phenomena, and not the perturbations.

1. The phenomena with which this science is conversant being the thoughts, feelings, and actions of human beings, it would have attained the ideal perfection of a science, if it enabled us to foretell how an individual would think, feel, or act throughout life, with the same certainty with which astronomy enables us to predict the places and the occultations of the heavenly bodies. It needs scarcely be stated that nothing approaching to this can be done. Hence, even if our science of human nature were theoretically perfect, that is, if we could calculate any character as we can calculate the orbit of any planet, from given data, still, as the data are never all given, nor ever precisely alike in different cases, we could neither make positive predictions, nor lay down universal propositions.'

,1

But the approximative generalizations are sufficiently exact for practical life; that which is only probable when it is affirmed of individuals taken at hazard, is certain when it is affirmed of the conduct of the masses; and therein lies the utility of psychology.'

Thus the aim of psychology is fixed: its object is the phenomena of mind. Its character is determined; it is (or may be)

1 Mill's Logic, ed. 1856, p. 421.

2 Logic, b. vi. ch. iii.

a science; not exact, but approximative, and sufficient for practical purposes. Let us now see the method of it.1

Two entirely opposite schools have contributed to its deviation from the right way,—on one side, that of Auguste Comte; on the other, that of German metaphysics. Mr. Mill writes thus of the former :

M. Comte claims for physiologists alone the scientific knowledge of intellectual and moral phenomena. He totally rejects psychological observations properly so called, the internal consciousness. He thinks that we have to acquire our knowledge of the human mind by observing others. How can we observe and interpret the mental operations of others without previously knowing our own? He does not tell us this. But he considers it evident that the observation of ourselves by ourselves can teach us only very little concerning feelings, and nothing on the subject of understanding. 'It is not necessary,' adds Mr. Stuart Mill, 'to refute a sophism at length, whose most surprising part is, that it should have imposed on any one. Two answers may be made to it:-1. M. Comte may be referred to the experience as well as to the writings of the psychologists as a proof that the mind can not only be conscious of more than one impression at a time, and even perceive a considerable number (according to Hamilton), but even lend them all attention. 2. It might have occurred to M. Comte that it is possible to study a fact by means of memory, not at the instant in which we perceive it, but the moment after, and this is, in reality, the mode by which we acquire the best of our knowledge of intellectual actions. Besides, in fact, we know what passes in ourselves, whether thanks to consciousness or thanks to memory, but in a direct way in both cases, and not (as happens about what we do in a state of somnambulism) by their results.' This simple fact destroys the entire argument of M. Comte. Everything of which we have direct consciousness we can observe. The successions, therefore, which obtain among mental phenomena, do not admit of being deduced from the physiological laws of our nervous organization; and all real knowledge of them must continue, for a long time at least, if not for ever, to be

1 Logic, loc. cit.

sought in the direct study, by observation and experiment, of the mental successions themselves. Since, therefore, the order of our mental phenomena must be studied in those phenomena, and not inferred from the laws of any phenomena more general, there is a distinct and separate science of mind.

The relations, indeed, of that science to the science of physiology must never be overlooked or undervalued. It must by no means be forgotten that the laws of mind may be derivative laws resulting from the laws of animal life, and that their truth therefore may ultimately depend on physical conditions. . . . But on the other hand, to reject the resource of psychological analysis, and construct the theory of the mind solely on such data as physiology at present affords, seems to me as great an error in principle, and an even more serious one in practice. Imperfect as is the science of mind, I do not scruple to affirm that it is in a considerably more advanced state than the portion of physiology that corresponds to it; and to discard the former for the latter appears to me an infringement of the true canons of inductive philosophy.

Thus, then, we have direct observation clearly established against positivism.1 Let us now see how our author combats the opposite school, the metaphysicians, German or otherwise, whom he calls, in general terms, philosophers à priori.

The dispute between the à priori philosophers and the à posteriori philosophy, he says, goes far beyond the bounds and the bearing of psychology, and is especially concentrated on the field of ontology. I have no intention of declaring myself a partisan of either, both having done much for humanity, both requiring to be known by whoever purposes to approach philosophical questions, each having largely profited by the criticisms of the other. By concentrating the question simply on the ground of psychology, we find that the difference between the two philosophies consists in the different theories which they give of the complex phenomena of the human mind.

6

Experience is not the exclusive property of one of them. They both depend on it for their materials. The fundamental difference has reference, not to the facts themselves, but to their origin.

1 See Logic, book vi. chap. iv., and Comte and Positivism, p. 67.

We may say briefly and generally that one of these theories considers the most complex phenomena of the mind as being the product of experience, whereas the other considers them as original.

A priori psychology maintains that, in every act of thought, even the most elementary, there is an element which is not given to the mind, but which is furnished by the mind, in virtue of its own faculties. The most simple of all the phenomena, an exterior sensation, requires, according to it, a mental element to be a perception, and thus to become, instead of a passive and fugitive condition of our being, a durable object exterior to the mind. The notions of extent, solidity, number, force, etc., although acquired by the senses, are not copies of impressions made upon the senses, but creations of the laws of our minds put in action by sensations. Experience, instead of being the source and the prototype of our ideas, is itself a product of the forces proper to the mind, elaborating the impressions which we receive from without; it contains a mental as well as an external element. Experience, invoked in vain to account for our mental laws, is only possible by those laws. Now, if experience does not explain experience, à fortiori it does not explain the ideas of moral, super-sensible things; experience is their occasion, but not their source.

A posteriori psychology, on the contrary, while it recognises the existence of a mental element in our ideas, and admits that our ideas of extent, solidity, time, space, virtue, are not exact copies of impressions made upon our senses, but a product of the labour of the mind, does not consider this production as the result of particular and impenetrable laws, which cannot be accounted for. It thinks, on the contrary, that that is possible. It thinks that the mental element is a fact, but not an ultimate fact. It thinks that it may be resolved into simpler laws and more general facts, and that it is possible to discover the process followed by the mind in the construction of these great ideas; in a word, that their genesis can be determined.

Let us define the difference between the two Schools of psychology by an example. The transcendentalists examine our ideas of space and time; they find that each contains in itself in an indissoluble manner the idea of the infinite. Naturally we

have no experimental knowledge of the infinite; all our ideas derived from experience are ideas of finite things. Nevertheless it is impossible to conceive of time and space otherwise than as infinite, and it is impossible to derive them from experience; these are the necessary conceptions of the mind. The à posteriori psychologist, on his side, sees clearly that we cannot think of time and space otherwise than as infinite, but he does not consider that as an ultimate fact. He sees in it an ordinary manifestation of one of the laws of the association of ideas,—the law that the idea of a thing irresistibly suggests the idea of another thing with which it has often been found by experience to be intimately united. As we have never had any experience of a point in space without other points beyond it, nor of a point in time without other points which follow it, the law of inseparable association causes us to be unable to think of any point in time or space, however distant, without immediately imagining other points yet more distant. This explains their infinitude without introducing necessity.' It may be that time and space have limits, but in our present condition we are totally unable to conceive of them. If we could reach the end of space, we should be apprised of it no doubt by some novel and strange impression of our senses, but of which we cannot at present form the very slightest idea.

'

The preceding example brings out clearly the two principal doctrines of the most advanced à posteriori psychology :

1. That the most abstruse phenomena of the mind are formed of more simple and elementary phenomena.

2. That the mental law by means of which this formation takes place is the law of association.

The most complete and scientific form of à posteriori psychology, is that which considers the law of association as the supreme principle. Its great problem is to determine, not how far this law extends-for it extends to everything: ideas, emotions, desires, volitions, etc.,-but how many mental phenomena it is capable of explaining, and how it explains them. On this part of the subject there are differences of doctrine, and the theory, like every theory in an incomplete science, progresses steadily.1

1 Loc. cit. p. 108.