less "conscious of their own existence" than the protozoa? Have we any right to assert either that the brute has an "ego," or that the plant has no sensation? The line which separates living from not living is at present as clear as that which separates man from brute, but when Mr. Wallace talks of "sensation" and the "ego" of conscious and unconscious life, he passes into the region of metaphysics, where he is as little competent to lead as we are willing to follow.

III.

MENTAL EVOLUTION IN MAN.1

THIS is an essay, almost a first essay, in comparative psychology-that is to say, it is an attempt to apply to the mental and moral nature of man the method which has been so fruitful of results when applied to physiology and morphology. It is not an attempt to merge psychology in physiology, but to compare the psychological facts of which man is immediately conscious with facts supposed to be similar in other living things.

The first difficulty here is one with which Mr. Romanes does not directly concern himself, though for Auguste Comte it seemed to make all psychology impossible-the difficulty, namely, of collecting the facts which are to be dealt with. "It requires art and pains," says John Locke, "to set the mind at a distance from itself, and make it its own object." Comte says in effect, the thing is impossible. No "art and pains" can help us. For, ex hypothesi, the observer and the observed are

1 Mental Evolution in Man. By George J. Romanes, M.A., LL.D., F.R.S. Kegan Paul, Trench, Trübner & Co.

identical, and the observer is only in the judicial attitude necessary for observation when nothing is going on-i.e. when there are no facts to observe. Hume had said almost the same :

"Should I attempt to experiment on my own mind, 'tis evident this reflection and premeditation would so disturb the operation of any natural principles as must render it impossible to form any just conclusion from the phenomenon."

Hence Comte's conclusion is, psychology is no science, but a branch of physiology.

But this view Herbert Spencer will have none of. Between the physiological order and the psychological there is a barrier which at present is absolute :

'Physiology (we are told) cannot properly appropriate subjective data, or data wholly inaccessible to external observation." It "ceases to be physiology when it imports into its interpretations a psychical factor, a faculty which no physical research whatever can disclose or identify, or get the remotest glimpse of. . . . Psychology under its subjective aspect is a totally unique science, independent of, and antithetically opposed to, all other sciences whatever."

In fact, a purely physiological psychology is as impossible as a subjective account of somnambulism. It is only psychology so far as it is false to its physiological method.

All this, we take it for granted, Mr. Romanes would allow. Psychological facts must be primarily and directly known in the consciousness of the individual, even if, with Mr. Sully, we allow that

For

all introspection is retrospection. But how then about the comparative method? Nobody knows any mind but his own, and not much of that. as he tries to look back to the earlier stages the words of memory become inarticulate and its characters illegible. And how does he know that other men have minds, or that they are at all like his own? And what about the familiar process of "tossing about psychological babies," as Dr. Martineau calls it, "and wringing from them ambiguas voces as to how they feel"? The psychology of other people's children only looks more legible than the recollection of one's own childhood, because we think we read there what we really read into it.

We are not here raising a difficulty which Mr. Romanes is not fully aware of. His answer would be, that other people's minds are to us neither subjects, like our own is, nor objects, like that which we can study physiologically, but ejects. The word is borrowed from Professor Clifford, and is used to mean an inference from acts or movements observed, which acts or movements, in our own experience, have been associated with certain psychological facts, and are therefore assumed to be a guarantee for the presence in others of similar facts. In the case of our fellowmen this assumption is undoubtedly made, and if it is not always justified, it is frequently confirmed

by what our fellow-men tell us by means of a language which we have come to understand. But the farther we get back in the history either of the individual or the race, the larger the assumptions which we have to make, and the smaller the help that language can give us. It is no use catechizing a child as to its psychical condition, though we may watch its acts and the growth of its language, and assume that these reflect the growth of its " mind.”

When, however, in our loyalty to the comparative method, we attempt to extend our inquiry to nonhuman animals, whose conversational powers are more limited than those of a child, we find ourselves making larger drafts on our original assumptions. Before, we were only guilty of "automorphism," interpreting other people's acts by ourselves; now we are guilty of the far graver crime of "anthropomorphism "—¿.e. interpreting in terms of man the acts and movements of creatures which are not human. Mr. Samuel Butler extends the same method to plants :

"In its own sphere,” he says, “a plant is just as intelligent as an animal, and keeps a sharp look-out upon its own interests, however indifferent it may seem to be to ours."

We are not anxious to maintain the opposite in either case, but it is well to remind ourselves that it is an assumption, and that the assumption becomes greater, and the conclusions less scientific, as we move away from the individual consciousness