stress on anything recollected unless it be something due to careful examination, and unless we have, by rule (sec. 136), exhausted all the relevant material to be found in the memory, the stores of which are useful for conduct, but not for eliciting truths. Such a method, I hold, is scientific, fruitful and well-based. Such a method, I claim, is the quintessence of the scientific method as applied in the physical sciences. Instruments and mathematical treatment may supplement it, but not displace it. These latter only give polish and precision to the great truths otherwise obtained. [Let the student observe that the mere neglect of metaphysics or theology will not assist him in the discovery of truth.]

2. THE USE OF HYPOTHESES.

In his Comparative Psychology, Lloyd Morgan tells the world that "Psychologists make, or should make, no claim to any monopoly of knowledge in the subject they study; their province is mainly to systematise that knowledge" (p. 44). Fortunately for us, this author's practice is not in accord with his theory, for his conception of focal and marginal consciousness forms a valuable contribution to psychology. Nevertheless there is incalculable mischief in his assertion, however hedged round. What would be thought of a physicist or an astronomer who mainly systematised knowledge without seeing that it was gathered at first hand by competent specialists or by himself, or who gave a locus standi to "the plain man of shrewd insight"? The idea is monstrous. A psychology which mainly busied itself with systematising the conclusions of "those who are not professed psychologists" might as well relieve Sisyphus of his task. The one is as likely to be successful as the other.

Underlying the statement I am criticising, there is an unpleasant truth. Unhappily, psychologists have been too anxious to systematise that which they had not previously examined. They leaned fatally to the opinion that truth could be sifted from popular notions as is sand by means of a sieve. The student must recognise once for all that if he is to be on the scientific plane he must make a claim on behalf of psychologists to a "monopoly of knowledge," and that he must not attempt to systematise what has not been procured through the application of scientific methods. It has been the bane of psychologists that they have tacitly assumed that facts of consciousness do not require to be collected with the disciplined care which other sciences employ. I have advisedly said "tacitly" assumed, because few men have spoken out boldly as Lloyd Morgan does. The mischief has lain in unthinkingly proceeding along the wrong path, in giving elaborate explanations of popular fictions, and in not deliberately recognising that in psychology, as in physics, unbridled speculation is criminal waste.

It is always dangerous to make unqualified statements. Instead of condemning speculation outright, it would be perhaps better to pronounce sentence against it when its excursions are not rigorously limited. Few can object when "speculation is but the play of the imagination along the fringe which borders our knowledge" (Lloyd Morgan, Comparative Psychology, 1894, p. 323). But then we must insist upon a reasonable interpretation which shall not make the fringe equal to the robe to which it

belongs. Speculation thus interpreted is not only innocent; it is an imperative necessity. If men formed hypotheses about the things on the borders of knowledge, they would never go far astray, and often return laden with trophies. Let me, however, define this term. "By an hypothesis we understand an assertion which is employed as a principle of explanation, though its correctness is yet unproven" (Volkmann, Lehrbuch, 1894, i, p. 103). I should say, then, that accepting the above definition, hypotheses are legitimate when they apply to the fringe which borders our knowledge. In thus limiting their uses one seems not only to be cutting at the root of ordinary psychological speculation, but to be contradicting weighty authorities. According to Jevons, study without hypothesis is not scientific. "The true course of inductive procedure is that which has yielded all the more lofty results of science. It consists in Anticipating Nature, in the sense of forming hypotheses as to the laws which are probably in operation; and then observing whether the combinations of phenomena are such as would follow from the laws supposed" (Principles of Science, 1877, p. 509). The fallacy underlying the preceding statement must be exposed. Where, the student should ask himself, do we find our hypotheses? Do they come to us as divine intuitions? Can we obtain them, or any new truth, by a course of abstract reasoning? Or is it not rather that we act on the basis of one observation, and note "whether the phenomena are such as would" be in accordance with that observation? This must obviously be the case, for no compounding of zeros can force the gates of nothingness. If that be so, Jevons' statement implies that there are no hidden facts which are not amenable to casual and superficial observation. That, as we shall see, is an indefensible position. There is no guessing which can take us farther than the fringe which borders our knowledge. We may set up the wildest theory, and yet it will be composed of what is known, its extravagance arguing no fresh knowledge. Take, for instance, a passage from Volkmann, which well illustrates the length to which fanciful. speculation or adaptation may go: "The presentational mass, which is helplessly sinking, meets with the freely-rising apperception mass, is gripped and firmly held by that, and placed before the ego as an object to be viewed" (Lehrbuch, 1895, ii, p. 206). This pretentious psychic mechanism to which we are here introduced is but a copy of material mechanisms, and the hypothesis underlying its use implies that analogy holds the key to the whole realm of the unknown, while the truth is that the master facts of a science are due to deliberate research. It is for this reason that one can scarcely detect a single sound brick in the elaborate Herbartian structure-unsophisticated observation of a scientific type has been scouted, and hypotheses were powerless to discover the new facts.

If my interpretation be correct, the progress of science should bear me out,* as indeed it does. The serious study of fact is continually going hand in hand with tentative speculation "along the fringe which borders

* See, however, Rigg, The Place of Hypothesis in Experimental Science, 1887.

our knowledge." Workers here and there wrest from nature trivial secrets. These secrets accumulate, and men now and then tentatively combine a few of them. As the store of knowledge assumes considerable proportions, so larger and larger generalisations are ventured upon in every direction. At last, when general statements of a far-reaching nature abound, men rightly venture to speculate as to the broadest generalisations possible. Already the chief notions are in the air; already those engaged in the search feel that they are approaching a solution. After many minor attempts some one, a Newton or a Darwin, sometimes comes to the front and completes the structure. As knowledge progresses, new speculations and observations "along the fringe which borders our knowledge" are readily suggested and easily verified. No stupid guess is possible to the man who has ascertained facts to go upon; and when he goes astray, the clashing of his theory with his existing store soon brings him back to the right path. Progress, though slow, is hence certain and reliable when men shun the use of large hypotheses. In the absence of well-ascertained facts, everything is changed. A guess, under these conditions, has no inherent plausibility, and what is as bad, any attempt at verification in an unexplored realm can only end in failure either acknowledged, or, as is more usual, disguised. Our intellectual sense of equilibration forsakes us when it has no general facts to assist it. Thus large hypotheses in neurology are not only worthless, but vicious, because it is only by the accumulation of facts and generalisations that solutions of neural difficulties can be reached at all. Sciences cannot be, and never have been, guessed at. Jevons' mistake was a plausible one. In the physical sciences so huge a body of organised observations has been accumulated that the guesses of men of learning are at once permissible and easily verified. Where, on the other hand, as in psychology, the stock of genuine observations is infinitesimal, these guesses will be unreasonable, and their verification will consist of an immensely protracted process, equal to the building up of the science itself. Indeed, Jevons supplies his own antidote. He says of the alchemists, of whom Newton was one, that "Many of them were men of the greatest acuteness, and their indefatigable labours were pursued through many centuries. A few things were discovered by them, but a true insight into nature now enables chemists to discover more useful facts in a year than were yielded by the alchemists during many centuries" (ibid, p. 505). The alchemists, like many present-day psychologists, indulged in large hypotheses, and large hypotheses are useless, are impedimenta in more senses than one.

I, therefore, side with Bacon (d. 1626) who says: "The subtilty of nature is far beyond that of sense or of the understanding: so that the specious meditations, speculations, and theories of mankind are but a kind. of insanity, only there is no one to stand by and observe it" (Novum Organum, ed. 1893, bk. 1, x); and who draws this admirably prophetic picture of some modern thinkers: "When any one prepares himself for discovery, he first inquires and obtains a full account of all that has

been said on the subject by others, then adds his own reflections, and stirs up, and, as it were, invokes his own spirit, after much mental labour, to disclose its oracles. All which is a method without foundation and merely turns on opinion" (ibid, bk. 1, lxxxii). Bacon has been roundly attacked for his views, which are all too sound. Even Newton has not escaped chastisement. His celebrated pronouncements against hypotheses are discounted on the assumption that he himself gained his great successes by their means. The censure on him is wholly inapplicable, for his great generalisation was no more self-evolved than that of Darwin. No man of his attainments could have been far wrong in his surmises, and the accumulations of organised knowledge were so extensive that verification was within easy reach. Newton focussed the learning of his time. When he goes beyond that, as in his conjectures concerning general stellar problems, he is only a wordy theologian. Hence, regarding the term hypothesis as implying a supposition which is not preceded by exact study, we can heartily agree with Newton's strong and unmistakable language. "Whatever is not deduced from the phenomena, is to be called an hypothesis; and hypotheses, whether metaphysical or physical, whether of occult qualities or mechanical, have no place in experimental philosophy. In this philosophy particular propositions are inferred from the phenomena, and afterwards rendered general by induction" (Mathematical Principles, trans. by A. Motte, 1729, ii, p. 392). Compare this hesitancy with the impatience of a modern writer: "Give us theories, theories, always theories!" (Baldwin, Mental Development, 1895, p. 38); or compare it with the speculative deductions of the learned Whewell: an Art of Discovery is not possible. At each step of the progress of science, are needed invention, sagacity, genius-elements which no Art can give. We may hope in vain, as Bacon hoped, for an organ which shall enable all men to construct scientific truths, as a pair of compasses enables all men to construct exact circles" (Philosophy of the Inductive Sciences, 1847, p. vii. See also his Novum Organon Renovatum, 1858).

The essential thing for the student to remember is that the chief facts of every science are only obtainable by the close observer after laborious research.* Let me give one illustration. What are dreams? Seeming to resemble the waking imagination, we forthwith guess that they are the result of vivid imaginings. Quantities of books had been written on the subject, based on speculation and occasional observation. Yet proper light only began to be thrown on the problems of dream-life when men methodically and for a considerable period observed their own dream-states. Excepting such contributions as those of Giessler and Schwartzkopff, most of the books are almost superfluous. Let me now state the dream facts as they appear to me (ch. 10). (1) The muscular and sensory tones are lowered, and the position of the various parts of the body is unknown (sec. 19, 1st and 2nd conclusion). (2) The characteristic pictures spring usually out of the

* Men are sometimes said to stumble upon important discoveries. In such cases it is the preparedness of the discoverer which accounts for the discovery.

dark or closed-eye field of vision, and are not imaginings at all. (3) Most of the dreams are initiated and maintained from without, resembling waking life and not waking imagination. (4) The amount of possible effort is largely reduced, whence follows confused thinking, (5) impossibility of strenuous thought, (6) inappropriate recollection, (7) a strong tendency for appetites, expectations, doubts, hopes and fears to actualise themselves, (8) the continual creation of a setting to each picture, giving it an air of reality, and (9) the fact that the happenings are compounded out of most recent events and those immediately passed. Thus points (2) and (3) argue afferent or outer influences, and the other points follow from the lack of strenuousness. Now, by what rational calculus could one have jumped from imaginative seeing to the retinal pictures which are plainly due, in part at least, to new circumstances. And, suppose we had not previously ascertained the fact, how could we have discovered the arbitrary creation of settings to the dream-pictures? The two essential facts of dream-life are afferent, or outward, influences and a drop in strenuousness. Yet the waking imagination has nothing to do with either of those factors: it tends away from outer impressions and implies considerable force of thought. Where, we may ask accordingly, lies the difference between the methods of ancient philosophy and those of current psychology? And who would ever confound scientific with philosophic procedure? The traditional philosophic method is as barren as the syllogism which it has produced. It may, therefore, be laid down for the student's guidance that scientific progress depends on gaining new classes of facts, that such can only be acquired by painfully close and methodical observation in the first instance, and that they are not obtainable by employing hypotheses which go beyond well-ascertained facts and established generalisations. As Tyndall puts it: "The thing to be encouraged here is a reverent freedom -a freedom preceded by the hard discipline which checks licentiousness in speculation" (Scientific Use of the Imagination, 1872, p. 33).

Adam, L'Imagination dans la Découverte Scientifique, 1890; Boirac, La Méthode Expérimentale, 1898; Naville, De la Place de l'Hypothèse dans la Science, 1876; Naville, Les Conditions des Hypothèses Sérieuses, 1877; Naville, Les Principes Directeurs des Hypothèses, 1877; and Venn, The Use of Hypotheses, 1878.

3.-APPROACHES TO THE STUDY OF PSYCHOLOGY.

To understand the human frame, we require to know its constitution to the minutest part, together with its reactions when stimulated: a microscope and a dissecting knife, with a battery, might be an adequate equipment for this task. As matters stand, these tools are found to be inadequate, and secondary means are, therefore, resorted to in addition. We study the development of the embryo, animal characteristics, evolutionary traits, and cases of disease or malformation. We also stain the tissues with preparations which affect only certain parts; we cut nerve bundles, as we might cut strings, and notice which of the nerves degenerate as a consequence; and we experimentally alter or remove structures or parts of structures.