Джон Стюарт Милль
Auguste Comte and Positivism
But though detached truths relating to the more complex order of phaenomena may be empirically observed, and a few of them even scientifically established, contemporaneously with an early stage of some of the sciences anterior in the scale, such detached truths, as M. Littré justly remarks, do not constitute a science. What is known of a subject, only becomes a science when it is made a connected body of truth; in which the relation between the general principles and the details is definitely made out, and each particular truth can be recognized as a case of the operation of wider laws. This point of progress, at which the study passes from the preliminary state of mere preparation, into a science, cannot be reached by the more complex studies until it has been attained by the simpler ones. A certain regularity of recurrence in the celestial appearances was ascertained empirically before much progress had been made in geometry; but astronomy could no more be a science until geometry was a highly advanced one, than the rule of three could have been practised before addition and subtraction. The truths of the simpler sciences are a part of the laws to which the phaenomena of the more complex sciences conform: and are not only a necessary element in their explanation, but must be so well understood as to be traceable through complex combinations, before the special laws which co-exist and co-operate with them can be brought to light. This is all that M. Comte affirms, and enough for his purpose.5 He no doubt occasionally indulges in more unqualified expressions than can be completely justified, regarding the logical perfection of the construction of his series, and its exact correspondence with the historical evolution of the sciences; exaggerations confined to language, and which the details of his exposition often correct. But he is sufficiently near the truth, in both respects, for every practical purpose.6 Minor inaccuracies must often be forgiven even to great thinkers. Mr Spencer, in the very-writings in which he criticises M. Comte, affords signal instances of them.7
Combining the doctrines, that every science is in a less advanced state as it occupies a higher place in the ascending scale, and that all the sciences pass through the three stages, theological, metaphysical, and positive, it follows that the more special a science is, the tardier is it in effecting each transition, so that a completely positive state of an earlier science has often coincided with the metaphysical state of the one next to it, and a purely theological state of those further on. This statement correctly represents the general course of the facts, though requiring allowances in the detail. Mathematics, for example, from the very beginning of its cultivation, can hardly at any time have been in the theological state, though exhibiting many traces of the metaphysical. No one, probably, ever believed that the will of a god kept parallel lines from meeting, or made two and two equal to four; or ever prayed to the gods to make the square of the hypothenuse equal to more or less than the sum of the squares of the sides. The most devout believers have recognized in propositions of this description a class of truths independent of the devine omnipotence. Even among the truths which popular philosophy calls by the misleading name of Contingent the few which are at once exact and obvious were probably, from the very first, excepted from the theological explanation. M. Comte observes, after Adam Smith, that we are not told in any age or country of a god of Weight. It was otherwise with Astronomy: the heavenly bodies were believed not merely to be moved by gods, but to be gods themselves: and when this theory was exploded, there movements were explained by metaphysical conceptions; such as a tendency of Nature to perfection, in virtue of which these sublime bodies, being left to themselves, move in the most perfect orbit, the circle. Even Kepler was full of fancies of this description, which only terminated when Newton, by unveiling the real physical laws of the celestial motions, closed the metaphysical period of astronomical science. As M. Comte remarks, our power of foreseeing phaenomena, and our power of controlling them, are the two things which destroy the belief of their being governed by changeable wills. In the case of phaenomena which science has not yet taught us either to foresee or to control, the theological mode of thought has not ceased to operate: men still pray for rain, or for success in war, or to avert a shipwreck or a pestilence, but not to put back the stars in their courses, to abridge the time necessary for a journey, or to arrest the tides. Such vestiges of the primitive mode of thought linger in the more intricate departments of sciences which have attained a high degree of positive development. The metaphysical mode of explanation, being less antagonistic than the theological to the idea of invariable laws, is still slower in being entirely discarded. M. Comte finds remains of it in the sciences which are the most completely positive, with the single exception of astronomy, mathematics itself not being, he thinks, altogether free from them: which is not wonderful, when we see at how very recent a date mathematicians have been able to give the really positive interpretation of their own symbols.8 We have already however had occasion to notice M. Comte's propensity to use the term metaphysical in cases containing nothing that truly answers to his definition of the word. For instance, he considers chemistry as tainted with the metaphysical mode of thought by the notion of chemical affinity. He thinks that the chemists who said that bodies combine because they have an affinity for each other, believed in a mysterious entity residing in bodies and inducing them to combine. On any other supposition, he thinks the statement could only mean that bodies combine because they combine. But it really meant more. It was the abstract expression of the doctrine, that bodies have an invariable tendency to combine with one thing in preference to another: that the tendencies of different substances to combine are fixed quantities, of which the greater always prevails over the less, so that if A detaches B from C in one case it will do so in every other; which was called having a greater attraction, or, more technically, a greater affinity for it. This was not a metaphysical theory, but a positive generalization, which accounted for a great number of facts, and would have kept its place as a law of nature, had it not been disproved by the discovery of cases in which though A detached B from C in some circumstances, C detached it from A in others, showing the law of elective chemical combination to be a less simple one than had at first been supposed. In this case, therefore, M. Comte made a mistake: and he will be found to have made many similar ones. But in the science next after chemistry, biology, the empty mode of explanation by scholastic entities, such as a plastic force, a vital principle, and the like, has been kept up even to the present day. The German physiology of the school of Oken, notwithstanding his acknowledged genius, is almost as metaphysical as Hegel, and there is in France a quite recent revival of the Animism of Stahl. These metaphysical explanations, besides their inanity, did serious harm, by directing the course of positive scientific inquiry into wrong channels. There was indeed nothing to prevent investigating the mode of action of the supposed plastic or vital force by observation and experiment; but the phrases gave currency and coherence to a false abstraction and generalization, setting inquirers to look out for one cause of complex phaenomena which undoubtedly depended on many.
According to M. Comte, chemistry entered into the positive stage with Lavoisier, in the latter half of the last century (in a subsequent treatise he places the date a generation earlier); and biology at the beginning of the present, when Bichat drew the fundamental distinction between nutritive or vegetative and properly animal life, and referred the properties of organs to the general laws of the component tissues. The most complex of all sciences, the Social, had not, he maintained, become positive at all, but was the subject of an ever-renewed and barren contest between the theological and the metaphysical modes of thought. To make this highest of the sciences positive, and thereby complete the positive character of all human speculations, was the principal aim of his labours, and he believed himself to have accomplished it in the last three volumes of his Treatise. But the term Positive is not, any more than Metaphysical, always used by M. Comte in the same meaning. There never can have been a period in any science when it was not in some degree positive, since it always professed to draw conclusions from experience and observation. M. Comte would have been the last to deny that previous to his own speculations, the world possessed a multitude of truths, of greater or less certainty, on social subjects, the evidence of which was obtained by inductive or deductive processes from observed sequences of phaenomena. Nor could it be denied that the best writers on subjects upon which so many men of the highest mental capacity had employed their powers, had accepted as thoroughly the positive point of view, and rejected the theological and metaphysical as decidedly, as M. Comte himself. Montesquieu; even Macchiavelli; Adam Smith and the political economists universally, both in France and in England; Bentham, and all thinkers initiated by him, – had a full conviction that social phaenomena conform to invariable laws, the discovery and illustration of which was their great object as speculative thinkers. All that can be said is, that those philosophers did not get so far as M. Comte in discovering the methods best adapted to bring these laws to light. It was not, therefore, reserved for M. Comte to make sociological inquiries positive. But what he really meant by making a science positive, is what we will call, with M. Littré, giving it its final scientific constitution; in other words, discovering or proving, and pursuing to their consequences, those of its truths which are fit to form the connecting links among the rest: truths which are to it what the law of gravitation is to astronomy, what the elementary properties of the tissues are to physiology, and we will add (though M. Comte did not) what the laws of association are to psychology. This is an operation which, when accomplished, puts an end to the empirical period, and enables the science to be conceived as a co-ordinated and coherent body of doctrine. This is what had not yet been done for sociology; and the hope of effecting it was, from his early years, the prompter and incentive of all M. Comte's philosophic labours.
It was with a view to this that he undertook that wonderful systematization of the philosophy of all the antecedent sciences, from mathematics to physiology, which, if he had done nothing else, would have stamped him, in all minds competent to appreciate it, as one of the principal thinkers of the age. To make its nature intelligible to those who are not acquainted with it, we must explain what we mean by the philosophy of a science, as distinguished from the science itself. The proper meaning of philosophy we take to be, what the ancients understood by it – the scientific knowledge of Man, as an intellectual, moral, and social being. Since his intellectual faculties include his knowing faculty, the science of Man includes everything that man can know, so far as regards his mode of knowing it: in other words, the whole doctrine of the conditions of human knowledge. The philosophy of a Science thus comes to mean the science itself, considered not as to its results, the truths which it ascertains, but as to the processes by which the mind attains them, the marks by which it recognises them, and the co-ordinating and methodizing of them with a view to the greatest clearness of conception and the fullest and readiest availibility for use: in one word, the logic of the science. M. Comte has accomplished this for the first five of the fundamental sciences, with a success which can hardly be too much admired. We never reopen even the least admirable part of this survey, the volume on chemistry and biology (which was behind the actual state of those sciences when first written, and is far in the rear of them now), without a renewed sense of the great reach of its speculations, and a conviction that the way to a complete rationalizing of those sciences, still very imperfectly conceived by most who cultivate them, has been shown nowhere so successfully as there.
Yet, for a correct appreciation of this great philosophical achievement, we ought to take account of what has not been accomplished, as well as of what has. Some of the chief deficiencies and infirmities of M. Comte's system of thought will be found, as is usually the case, in close connexion with its greatest successes.
The philosophy of Science consists of two principal parts; the methods of investigation, and the requisites of proof. The one points out the roads by which the human intellect arrives at conclusions, the other the mode of testing their evidence. The former if complete would be an Organon of Discovery, the latter of Proof. It is to the first of these that M. Comte principally confines himself, and he treats it with a degree of perfection hitherto unrivalled. Nowhere is there anything comparable, in its kind, to his survey of the resources which the mind has at its disposal for investigating the laws of phaenomena; the circumstances which render each of the fundamental modes of exploration suitable or unsuitable to each class of phaenomena; the extensions and transformations which the process of investigation has to undergo in adapting itself to each new province of the field of study; and the especial gifts with which every one of the fundamental sciences enriches the method of positive inquiry, each science in its turn being the best fitted to bring to perfection one process or another. These, and many cognate subjects, such as the theory of Classification, and the proper use of scientific Hypotheses, M. Comte has treated with a completeness of insight which leaves little to be desired. Not less admirable is his survey of the most comprehensive truths that had been arrived at by each science, considered as to their relation to the general sum of human knowledge, and their logical value as aids to its further progress. But after all this, there remains a further and distinct question. We are taught the right way of searching for results, but when a result has been reached, how shall we know that it is true? How assure ourselves that the process has been performed correctly, and that our premises, whether consisting of generalities or of particular facts, really prove the conclusion we have grounded on them? On this question M. Comte throws no light. He supplies no test of proof. As regards deduction, he neither recognises the syllogistic system of Aristotle and his successors (the insufficiency of which is as evident as its utility is real) nor proposes any other in lieu of it: and of induction he has no canons whatever. He does not seem to admit the possibility of any general criterion by which to decide whether a given inductive inference is correct or not. Yet he does not, with Dr Whewell, regard an inductive theory as proved if it accounts for the facts: on the contrary, he sets himself in the strongest opposition to those scientific hypotheses which, like the luminiferous ether, are not susceptible of direct proof, and are accepted on the sole evidence of their aptitude for explaining phenomena. He maintains that no hypothesis is legitimate unless it is susceptible of verification, and that none ought to be accepted as true unless it can be shown not only that it accords with the facts, but that its falsehood would be inconsistent with them. He therefore needs a test of inductive proof; and in assigning none, he seems to give up as impracticable the main problem of Logic properly so called. At the beginning of his treatise he speaks of a doctrine of Method, apart from particular applications, as conceivable, but not needful: method, according to him, is learnt only by seeing it in operation, and the logic of a science can only usefully be taught through the science itself. Towards the end of the work, he assumes a more decidedly negative tone, and treats the very conception of studying Logic otherwise than in its applications as chimerical. He got on, in his subsequent writings, to considering it as wrong. This indispensable part of Positive Philosophy he not only left to be supplied by others, but did all that depended on him to discourage them from attempting it.
This hiatus in M. Comte's system is not unconnected with a defect in his original conception of the subject matter of scientific investigation, which has been generally noticed, for it lies on the surface, and is more apt to be exaggerated than overlooked. It is often said of him that he rejects the study of causes. This is not, in the correct acceptation, true, for it is only questions of ultimate origin, and of Efficient as distinguished from what are called Physical causes, that he rejects. The causes that he regards as inaccessible are causes which are not themselves phaenomena. Like other people he admits the study of causes, in every sense in which one physical fact can be the cause of another. But he has an objection to the word cause; he will only consent to speak of Laws of Succession: and depriving himself of the use of a word which has a Positive meaning, he misses the meaning it expresses. He sees no difference between such generalizations as Kepler's laws, and such as the theory of gravitation. He fails to perceive the real distinction between the laws of succession and coexistence which thinkers of a different school call Laws of Phaenomena, and those of what they call the action of Causes: the former exemplified by the succession of day and night, the latter by the earth's rotation which causes it. The succession of day and night is as much an invariable sequence, as the alternate exposure of opposite sides of the earth to the sun. Yet day and night are not the causes of one another; why? Because their sequence, though invariable in our experience, is not unconditionally so: those facts only succeed each other, provided that the presence and absence of the sun succeed each other, and if this alternation were to cease, we might have either day or night unfollowed by one another. There are thus two kinds of uniformities of succession, the one unconditional, the other conditional on the first: laws of causation, and other successions dependent on those laws. All ultimate laws are laws of causation, and the only universal law beyond the pale of mathematics is the law of universal causation, namely, that every phaenomenon has a phaenomenal cause; has some phaenomenon other than itself, or some combination of phaenomena, on which it is invariably and unconditionally consequent. It is on the universality of this law that the possibility rests of establishing a canon of Induction. A general proposition inductively obtained is only then proved to be true, when the instances on which it rests are such that if they have been correctly observed, the falsity of the generalization would be inconsistent with the constancy of causation; with the universality of the fact that the phaenomena of nature take place according to invariable laws of succession.9 It is probable, therefore, that M. Comte's determined abstinence from the word and the idea of Cause, had much to do with his inability to conceive an Inductive Logic, by diverting his attention from the only basis upon which it could be founded.
We are afraid it must also be said, though shown only by slight indications in his fundamental work, and coming out in full evidence only in his later writings – that M. Comte, at bottom, was not so solicitous about completeness of proof as becomes a positive philosopher, and that the unimpeachable objectivity, as he would have called it, of a conception – its exact correspondence to the realities of outward fact – was not, with him, an indispensable condition of adopting it, if it was subjectively useful, by affording facilities to the mind for grouping phaenomena. This appears very curiously in his chapters on the philosophy of Chemistry. He recommends, as a judicious use of "the degree of liberty left to our intelligence by the end and purpose of positive science," that we should accept as a convenient generalization the doctrine that all chemical composition is between two elements only; that every substance which our analysis decomposes, let us say into four elements, has for its immediate constituents two hypothetical substances, each compounded of two simpler ones. There would have been nothing to object to in this as a scientific hypothesis, assumed tentatively as a means of suggesting experiments by which its truth may be tested. With this for its destination, the conception, would have been legitimate and philosophical; the more so, as, if confirmed, it would have afforded an explanation of the fact that some substances which analysis shows to be composed of the same elementary substances in the same proportions, differ in their general properties, as for instance, sugar and gum.10 And if, besides affording a reason for difference between things which differ, the hypothesis had afforded a reason for agreement between things which agree; if the intermediate link by which the quaternary compound was resolved into two binary ones, could have been so chosen as to bring each of them within the analogies of some known class of binary compounds (which it is easy to suppose possible, and which in some particular instances actually happens);11 the universality of binary composition would have been a successful example of an hypothesis in anticipation of a positive theory, to give a direction to inquiry which might end in its being either proved or abandoned. But M. Comte evidently thought that even though it should never be proved – however many cases of chemical composition might always remain in which the theory was still as hypothetical as at first – so long as it was not actually disproved (which it is scarcely in the nature of the case that it should ever be) it would deserve to be retained, for its mere convenience in bringing a large body of phaenomena under a general conception. In a résumé of the general principles of the positive method at the end of the work, he claims, in express terms, an unlimited license of adopting "without any vain scruple" hypothetical conceptions of this sort; "in order to satisfy, within proper limits, our just mental inclinations, which always turn, with an instinctive predilection, towards simplicity, continuity, and generality of conceptions, while always respecting the reality of external laws in so far as accessible to us" (vi. 639). "The most philosophic point of view leads us to conceive the study of natural laws as destined to represent the external world so as to give as much satisfaction to the essential inclinations of our intelligence, as is consistent with the degree of exactitude commanded by the aggregate of our practical wants" (vi. 642). Among these "essential inclinations" he includes not only our "instinctive predilection for order and harmony," which makes us relish any conception, even fictitious, that helps to reduce phaenomena to system; but even our feelings of taste, "les convenances purement esthétiques," which, he says, have a legitimate part in the employment of the "genre de liberté" resté facultatif pour notre intelligence." After the due satisfaction of our "most eminent mental inclinations," there will still remain "a considerable margin of indeterminateness, which should be made use of to give a direct gratification to our besoin of ideality, by embellishing our scientific thoughts, without injury to their essential reality" (vi. 647). In consistency with all this, M. Comte warns thinkers against too severe a scrutiny of the exact truth of scientific laws, and stamps with "severe reprobation" those who break down "by too minute an investigation" generalizations already made, without being able to substitute others (vi. 639): as in the case of Lavoisier's general theory of chemistry, which would have made that science more satisfactory than at present to "the instinctive inclinations of our intelligence" if it had turned out true, but unhappily it did not. These mental dispositions in M. Comte account for his not having found or sought a logical criterion of proof; but they are scarcely consistent with his inveterate hostility to the hypothesis of the luminiferous ether, which certainly gratifies our "predilection for order and harmony," not to say our "besoin d'idéalite", in no ordinary degree. This notion of the "destination" of the study of natural laws is to our minds a complete dereliction of the essential principles which form the Positive conception of science; and contained the germ of the perversion of his own philosophy which marked his later years. It might be interesting, but scarcely worth while, to attempt to penetrate to the just thought which misled M. Comte, for there is almost always a grain of truth in the errors of an original and powerful mind. There is another grave aberration in M. Comte's view of the method of positive science, which though not more unphilosophical than the last mentioned, is of greater practical importance. He rejects totally, as an invalid process, psychological observation properly so called, or in other words, internal consciousness, at least as regards our intellectual operations. He gives no place in his series of the science of Psychology, and always speaks of it with contempt. The study of mental phaenomena, or, as he expresses it, of moral and intellectual functions, has a place in his scheme, under the head of Biology, but only as a branch of physiology. Our knowledge of the human mind must, he thinks, be acquired by observing other people. How we are to observe other people's mental operations, or how interpret the signs of them without having learnt what the signs mean by knowledge of ourselves, he does not state. But it is clear to him that we can learn very little about the feelings, and nothing at all about the intellect, by self-observation. Our intelligence can observe all other things, but not itself: we cannot observe ourselves observing, or observe ourselves reasoning: and if we could, attention to this reflex operation would annihilate its object, by stopping the process observed.
There is little need for an elaborate refutation of a fallacy respecting which the only wonder is that it should impose on any one. Two answers may be given to it. In the first place, M. Comte might be referred to experience, and to the writings of his countryman M. Cardaillac and our own Sir William Hamilton, for proof that the mind can not only be conscious of, but attend to, more than one, and even a considerable number, of impressions at once.12 It is true that attention is weakened by being divided; and this forms a special difficulty in psychological observation, as psychologists (Sir William Hamilton in particular) have fully recognised; but a difficulty is not an impossibility. Secondly, it might have occurred to M. Comte that a fact may be studied through the medium of memory, not at the very moment of our perceiving it, but the moment after: and this is really the mode in which our best knowledge of our intellectual acts is generally acquired. We reflect on what we have been doing, when the act is past, but when its impression in the memory is still fresh. Unless in one of these ways, we could not have acquired the knowledge, which nobody denies us to have, of what passes in our minds. M. Comte would scarcely have affirmed that we are not aware of our own intellectual operations. We know of our observings and our reasonings, either at the very time, or by memory the moment after; in either case, by direct knowledge, and not (like things done by us in a state of somnambulism) merely by their results. This simple fact destroys the whole of M. Comte's argument. Whatever we are directly aware of, we can directly observe.
And what Organon for the study of "the moral and intellectual functions" does M. Comte offer, in lieu of the direct mental observation which he repudiates? We are almost ashamed to say, that it is Phrenology! Not, indeed, he says, as a science formed, but as one still to be created; for he rejects almost all the special organs imagined by phrenologists, and accepts only their general division of the brain into the three regions of the propensities, the sentiments, and the intellect,13 and the subdivision of the latter region between the organs of meditation and those of observation. Yet this mere first outline of an apportionment of the mental functions among different organs, he regards as extricating the mental study of man from the metaphysical stage, and elevating it to the positive. The condition of mental science would be sad indeed if this were its best chance of being positive; for the later course of physiological observation and speculation has not tended to confirm, but to discredit, the phrenological hypothesis. And even if that hypothesis were true, psychological observation would still be necessary; for how is it possible to ascertain the correspondence between two things, by observation of only one of them? To establish a relation between mental functions and cerebral conformations, requires not only a parallel system of observations applied to each, but (as M. Comte himself, with some inconsistency, acknowledges) an analysis of the mental faculties, "des diverses facultés élémentaires," (iii. 573), conducted without any reference to the physical conditions, since the proof of the theory would lie in the correspondence between the division of the brain into organs and that of the mind into faculties, each shown by separate evidence. To accomplish this analysis requires direct psychological study carried to a high pitch of perfection; it being necessary, among other things, to investigate the degree in which mental character is created by circumstances, since no one supposes that cerebral conformation does all, and circumstances nothing. The phrenological study of Mind thus supposes as its necessary preparation the whole of the Association psychology. Without, then, rejecting any aid which study of the brain and nerves can afford to psychology (and it has afforded, and will yet afford, much), we may affirm that M. Comte has done nothing for the constitution of the positive method of mental science. He refused to profit by the very valuable commencements made by his predecessors, especially by Hartley, Brown, and James Mill (if indeed any of those philosophers were known to him), and left the psychological branch of the positive method, as well as psychology itself, to be put in their true position as a part of Positive Philosophy by successors who duly placed themselves at the twofold point of view of physiology and psychology, Mr Bain and Mr Herbert Spencer. This great mistake is not a mere hiatus in M. Comte's system, but the parent of serious errors in his attempt to create a Social Science. He is indeed very skilful in estimating the effect of circumstances in moulding the general character of the human race; were he not, his historical theory could be of little worth: but in appreciating the influence which circumstances exercise, through psychological laws, in producing diversities of character, collective or individual, he is sadly at fault.
