Mr. Max Muller's 'Science of Thought'

Review of:
The Science of Thought (by F. Max Müller, 1887)

Part 1: April 1888

It is with some misgivings that I venture to make a few observations on Mr. Max Müller's last and most important work. The number of those who are entitled to discuss seriously the results of the labour of a lifetime devoted to studies of the most arduous and special kind must be very small, and it certainly does not include me. His work, however, though founded on the author's study of language, goes far beyond mere philology and connects it with so many subjects of the most general interest that I wish to make some observations upon it, especially because it appears to have attracted less general attention than its extraordinary interest seems to call for.

Mr. Max Müller considers that he has opened and followed up to a great extent a track which will ultimately lead to the solution of all the fundamental philosophical problems which have been considered insoluble for thousands of years and have led many people to regard metaphysics as a disease of the mind.

Such an assertion in the mouth of an unknown man would hardly deserve attention, but when a man who, after most successful studies lasting over nearly half a century, tells us that, by following a road on which he has travelled for a considerable distance, 'philosophy would have and could have no longer any secrets, it would cease to exist,' it appears, to say the very least, worth while to try to understand the grounds of his opinion.

I will try to give a sketch of his principal positions and of the grounds on which they rest, and will make a few remarks which they suggest. My attempt to do so is justified by a passage in the work itself.

‘It is no easy task (he observes) to attempt to give in a few words a true abstract of Kant's philosophy; yet if we wish to gain a clear view of the . . . movement of human thought we must be satisfied with short abstracts. . . . Whole pages, nay, whole volumes, must here be represented by one or two lines, and all that is essential is that we should not lose sight of the salient points in each system.'

The Science of Thought fills 618 pages, and though the whole of it is logically connected together, I think that the force and interest of the argument may possibly be increased by compression, as it requires somewhat careful reading to follow the author through the mass of proofs and illustrations which he has accumulated, and the collateral though related subjects into which from time to time he diverges.

The best means of summarising his work will be to collect into one view what I conceive to be his main propositions so as to show their logical connections, and then to examine each of them successively with its proofs. These propositions seem to me to be as follows:—

1. Thought is an operation which proceeds in every reasonable man. Thought, cogito, means co-agito, a word which is nearly equivalent to a Sanscrit term which Mr. Max Müller says means 'working-within.' Also 'a thing,' the most general term in language, means ‘a think.' The subject by which this operation is conducted or in which it proceeds Mr. Max Müller calls 'the Ego as personating the Self—an obscure phrase which he does not explain, but which seems not to mean more than the word 'man.' The operation he calls Mind, including under that word sensation, perception, conception, and naming, as well as various modes of combining and separating the results of those processes. [See especially p. 64 and following]

2. Thought is identical with language, the only difference between them being that language is, and that thought is not until it is uttered as language, made audible or visible by means of external signs. Hence the history and science of thought are identical with the history and science of language.

3. There are four stages in the formation of language which may be separately named and thought of, but which are no more separable in fact than a substance and its qualities. These are sensation, perception, conception, and naming. An act of imagination is necessary to convert mere sensation into perception; and an act of generalisation to convert perceptions of the same sort into conceptions. [I call them conceptions, and not, as Mr. Max Müller does, concepts, because the word ' concept' jars on my ear as a technical metaphysical expression. Such a phrase as 'I have no conception what he means' is perfectly natural. 'Conceit' would be less objectionable in sound than concept, but would not be understood; no one is likely to confound conception, the act or process, with conception, the result. Mr. Max Müller himself uses 'sensation ' to express both the process and the result. He does not talk of pleasure or pain as 'sensates.’] The representation of the conception by a sound is naming, and names are language. This represents what may be called the anatomy of language.

4. Though sensation is an essential part of thought without which it cannot exist, yet sensation alone cannot account for all our thoughts. The proper way of accounting for the whole arrangement of our thoughts is by Kant's theories as to space, time, and the categories. His philosophy arrives from a different point of view at the same results as Mr. Max Müller.

5. Language is the specific difference which distinguishes men from animals, and disproves the theory that men were developed out of animals. This is not inconsistent with the theory of evolution rightly understood, though to some extent it is inconsistent with Darwin.

6. The history of language shows that the languages spoken by the most important nations of modern Europe may all be derived from about 800 roots, expressing 121 conceptions which are turned into cognate words by the application of prefixes, suffixes, affixes, &c.—the parts of speech devised by grammarians and referable to and confirmed by the categories established by Kant and other philosophers.

7. Language is subject to the diseases of mythology and metaphor, the only cure for which is definition. By this means it is capable of such improvements as would reduce all human knowledge to the clearest and simplest form possible in the nature of the case.

These are the principal matters comprised in Mr. Max Müller's great work. I will try to develop them somewhat more fully and make a few observations upon each, though it is obvious that in a matter of such magnitude hardly any one is entitled to speak with authority.

1. The first of the propositions which I have stated is rather assumed than proved by Mr. Max Müller, and indeed it stands in no need of proof, for it is little more than a definition of the word 'mind,' and a description of the process of thinking.

The view that sensation is a part of the process, and indeed the foundation on which the whole edifice of thought is built, may to some persons appear paradoxical. It appears to me to be the fundamental truth of all rational speculation on these subjects, and indeed to mark the point of union between Mr. Müller and the most popular and influential school of philosophy of the day, that of John Mill and his disciples and adherents. Nothing is more characteristic of Mr. Müller than the way in which he holds fast by sensation and refuses to proceed a single step without its support, although the theory that in sensation itself there is a mental element is equally characteristic of him. He invariably insists that the mind is not a mere looking-glass, that in thought it is not a mere passive recipient of impressions from without. I do not believe that any one ever did hold these views. The metaphor about the looking-glass could be used by reasonable persons only as a vivid way of denying what Mr. Max Müller denies himself, the possibility of thought without sensation. Indeed, I may go a step further— I do not believe that any one who has ever watched children or noticed a picture or a tune can possibly doubt that we learn to see and to hear as we learn to speak and to walk. The conception of mind as a process in which sensation, perception, conception, and naming take place will in these days be denied by few to be correct, and most people will be ready to agree that the processes described are simultaneous and inseparable. A name is, no doubt, in some cases consciously imposed, such as the words 'ohm' &c., used in reference to electricity, but when this is done the thing named is always previously known by some more elaborate and less convenient name. Dog is a name as well as Argus. Manuscript or book is a name as well as the Science of Thought.

In restricting his list of mental operations to perception and conception Mr. Max Müller is very moderate. His knowledge of language enables him to make ceaseless efforts to simplify it. Some of his observations on the extent to which, according to his views, condensation might be carried in the matter of philosophical terms deserve gratitude which can hardly be exaggerated.

‘I believe it would really be the greatest benefit to mental science if all such terms as impression, sensation, perception, intuition, presentation, representation, conception, idea, thought, cognition, as well as sense, mind, memory, intellect, understanding, reason, soul, spirit, could for a time be banished from our philosophical dictionaries, and not be readmitted till they had undergone a thorough purification. ... I deny that there are any such things as soul, mind, memory, intellect, understanding, and reason, or that the conscious monon [why not man or men?] can be said to be endowed with them, whether in the shape of separate faculties or useful instruments.’[P. 18]

He does not, however, object to the use of such words 'as the names of certain modes of action of a self-conscious monon,' or man. [A good instance of an early objection taken to the existence in men of these numerous subsidiary beings is to be found in an extract from Richter's Levana, given in Max Müller's German Classics, ii. 632; Richter's ‘siebenjährige Tochter behauptete, wenn die Seele im Kopfe wieder Arme und Beine und einen Kopf hätte, so müsste in diesem wieder eine Seele wohnen und diese hatte wieder einen Kopf und so immer fort.' In Jeremy Taylor's Ductor Dubitantium the will and the conscience are always conceived of as subordinate human beings with rights, duties, reasons, consciences, and wills of their own.]

2. The second proposition affirms the identity of thought and language, and draws the inference that the history and science of the two must be identical. The assertion and illustration of this proposition occupy a large part of the book. Mr. Müller carefully examines the views upon this subject of a number of philosophers. He says that, amongst the Germans, W. von Humboldt, Schelling, and Hegel held his view, [P. 46.] but that of English writers no one has held it without qualification except Archbishop Whately. This inquiry is most interesting, but I cannot follow it here, neither can I enter upon the still more interesting and curious inquiry which he makes into the reasons which may have induced these writers to hesitate before admitting his views. The matter has given rise to a great controversy in Nature and elsewhere as to the truth of the assertion itself, a good deal of which I have read, and I will confine myself to giving the reasons which lead me to agree absolutely with Mr. Max Midler's opinion, and to pointing out one of the inferences which follow from the admission of its truth.

In the first place it must be understood what Mr. Max Müller means by language. He means by it significant sounds or other signs made perceptible to the senses and conveying some conception of a more or less general character. Language would thus include words spoken, words written, and significant gestures indicating any conception of a general character.

He says:—

‘Other signs may take the place of words. Five fingers or five lines are quite sufficient to convey the concept of five between people speaking different languages, possibly between deaf and dumb people who speak no language at all.’

On the other hand, the word language, according to this definition, would not apply to sounds, signs, or gestures which indicated only particular passive states of feeling such as a cry of rage, pain, or fear, a gesture of attack or defence uttered or employed either by a man or an animal.

If language is thus defined, it is, I think, impossible to suggest any real exception to what is admitted to be the general rule that thought and language are identical. Such apparent exceptions as that the deaf and dumb can think are disposed of by the terms of the definition, for such persons think by means of significant signs, though it can never be known in what way such a sign presents itself to their minds. [The way in which the deaf and dumb are taught to speak is as follows: The teacher puts the hands of the scholar one on the teacher's and one on the scholar's throat, so as to feel the movements of the muscles while the teacher speaks. The scholar's attention is also directed to the motions of the teacher's lips. The words must thus represent to the scholar's mind certain motions of the lips as associated with other motions of the muscles of the throat. The result of this sort of treatment is most surprising. A person so trained, being told that her brothers had been playing quoits, said,' Ko-its! is that right? What a very odd word!']

Two arguments only against Mr. Max Müller —and at bottom they are different forms of the same argument—appear to me to have any considerable weight.

These cases are, first, that animals perform many acts which might be the result of thought, in which case, if thought and language are identical, they would use language, which is admittedly not the case. Secondly, that in all sorts of cases in which men act rationally, they act without thinking in words. A fencer, for instance, in fencing, a sportsman in aiming his gun, a musician in playing on an instrument. Almost everyone, in a word, who does an act requiring address and rapidity of execution acts without any external use of language, though he certainly thinks and acts as he would act if he did think.

With regard to animals Mr. Max Müller's answer is that we are entirely ignorant of the minds of animals and are wholly unable to say that they think. All that can be said is that there is not the slightest reason to suppose that they use any significant sounds; vocal or otherwise. That they have senses like our own, often more acute and possibly in some cases different in kind, is proved by an immense amount of evidence. That something which as a mere cover for our ignorance we call instinct leads them to do acts which, if the result of reflection, would require much thought and experience is equally clear. [A wonderful instance of this is given by Mr. Max Müller  in the case of the emperor moth and the provision which it makes when a grub for its protection in passing from the condition of a chrysalis into that of a butterfly, by spinning a case of a very peculiar construction. How could a creature which became an orphan as an egg, and which never had any friends, know that it was to be a chrysalis and was afterwards to become a butterfly 1How could it tell what facilities a butterfly would require for getting out of the case spun by the grub for the chrysalis, or how bristles of three different degrees of stiffness and pointing in different directions would afford those facilities? (pp. 13-14.) It is as if a new-born baby was able to make a shroud suitable for its easy resurrection.] But all this shows only our ignorance and the impossibility of making any satisfactory assertions about animals and their thoughts or whatever in them does instead of thought.

In short, the case of animals proves, not that it is possible to think without language, but that it is possible to act intelligently without language by means with which mankind are not acquainted.

As for those parts of human conduct which are referred to, many of them may be explained by the fact that there are cases in which a man must not stop to think if he wishes to act properly, but must trust to an acquired habit, as people do in repeating what they have learnt by heart, or when fencing, dancing, or playing on a musical instrument. There are other cases in which a man thinks so rapidly and in so condensed a fashion that he is not aware that he thinks at all, but it does not follow that so far as he does think it is not in words. A man who is making a speech, and decides in a moment to avoid a particular statement or to change the topic on which he is dwelling, would, I think, do so because some thought, some unspoken word, forgotten as soon as the warning which it conveyed had operated, passed through his mind, no matter how quickly and in how summary a way. He would read, for instance, into the half-uttered 'stop' or 'don't' a momentary expression in the face of a person whom he wished to persuade.

The decisive test upon the whole subject, however, appears to me to be the one which Mr. Max Müller suggests: let any one try to think of a dog without using mentally the word dog, or some equivalent word, or to think the phrase, 'Cogito ergo sum,' without those words in one language or other, and he must, I think, if honest, confess that the attempt is like trying to breathe without air, or to see in the dark. It is prohibiting yourself from using the only means by which the required thing can be done.

A strong illustration of the truth of this view is to be found in one of the popular arguments against it. Thoughts, it is said, must in some cases be deeper than words, because no words can express the thoughts which are excited by particular objects. A beautiful woman, a beautiful piece of music, a beautiful view, all raise, as the phrase is, thoughts too deep for words. To ask that the thoughts so raised may be indicated in some other way than by words is, no doubt, to ask an impossibility; but if this is so, how can any one be sure that he has such thoughts? A thought which cannot be expressed or recalled to the mind, or be in any way fixed in a definite shape, is not a thought at all, but only a state of feeling; and though it is impossible to imagine a state of feeling which cannot be named, there is no state of feeling which can be adequately described. This is shown by all attempts to do so.

A lady once described to a friend her feelings on having a strong double tooth pulled out, by saying that she felt as if her head was coming off. The friend asked what it felt like to have your head come off. Pain, pleasure, hope, fear, in all their innumerable varieties, are words with which we cannot dispense, but which tell us very little. How much do we learn by being told that a rose smells sweet, or that flowers in a bedroom are often oppressive? The noblest piece of music ever written conveys no definite meaning whatever, nothing which can be called thought, because it is not sufficiently definite. It is sometimes said of a first-rate player on the violin that he can make it speak. The phrase indicates in its exaggeration the impassable limit between language or thought and mere sound, however expressive. Every one knows what is meant by the speaking of a musical instrument, but no two persons, asked what it said, would give the same answer. The beginning of 'The heavens are telling' is identical, or nearly identical, with that of 'The Lass of Richmond Hill.' Do the notes say, 'The heavens are telling the glory of God,' or,' On Richmond Hill there lives a lass, more fair than Mayday morn'? The truth is that a thought which cannot be put into words is not thought at all; it is only an attempt to think. A word which does not call up a thought is not a word but a mere noise.

The practical consequences of admitting this doctrine are the subject of the whole of Mr. Max Müller’s book, and I believe he is the first person who has ever recognised them, or set them forth in an intelligible form. In his examination of the views of different philosophers who have treated of it, and after quoting Humboldt, Schelling, and Hegel, he says, 'None of them seems to have had a suspicion how, if these words be true, all that we call philosophy will have to be put on a new footing.' [P. 45]

I do not think that any one who carefully reads the Science of Thought will be able to deny this, though there are many particular parts of the contents of the book which are open to great question, and many others in which Mr. Max Müller’s assertions can be tested by only an infinitesimal number of persons.

The principal and the strongest point in his case seems to me to be that, whatever may be said of thought, language is, at all events, a definite, permanent thing, which can be studied and discussed according to fixed rules. A very large part of what is commonly called philosophy consists of statements which it is impossible to test, and which it is often impossible to understand at all, or at least without an amount of labour probably disproportionate to the advantage to be derived from it. Both of these defects arise from the notion that indistinct feelings in the philosopher's own mind are thoughts, and that the task before him is that of devising language fitted to express them. The result frequently is the invention of a whole mass of new words and new names, or the use of old ones in question-begging senses which greatly puzzle both writers and readers, and often have no distinct meaning whatever.

If the identity of thought and language were fully understood, it would have a stronger tendency than anything else to the encouragement of plainness and simplicity in speculation, especially upon subjects which have been under discussion for thousands of years.

It would be still more useful in marking the limits of such discussion. The whole tone of them would be changed, if it were generally understood that they are discussions about words, and that they can be conducted to advantage only by definitions of the fundamental terms contained in them. They might thus in most cases be brought to an end in a reasonable time. Suppose, for instance, the subject of discussion is free will. How much more likely it is that it will be brought to some conclusion if the meaning of the two monosyllables 'free' and' will' be considered as the meaning of any other word would be than if the disputants assume without any such examination that they know what they respectively admit and deny, and appeal on the one side to their own consciousness by assertions which no one can test, or to arguments about statistics and other matters the relevancy of which is continually denied, and is impossible to be proved on the other.

Suppose, again, that the history of all important words were to be made known; and that the degree to which they originally were, or in the course of time came to be, metaphors were fully understood, what a flood of light would this throw upon all sorts of controversies! Fifty or perhaps even forty years ago Coleridge was a great name in English speculation. In all Carlyle's writings there is no more striking description than that which depicts him as sitting in the character of a giver of oracles at Highgate, entreating mankind to prepare themselves for his work on the Logos (which never was written) by grasping the fundamental and all-important distinction between the Reason and the Understanding, which, says Carlyle, you could never understand. How much trouble it would have saved to him and to others to learn that Reason and Understanding are only two metaphors which describe mental operations respectively as 'counting,' and resisting or standing up against external facts until you can conceive their relations and connections. So that the distinction is as important as one by which I was puzzled as a boy, the distinction namely between the Tully who was so much admired in the last century and the Cicero to whom my admiration was directed on similar grounds in the present.

This, however, is a matter to which I shall return in a later part of this article.

3. Mr. Max Müller's third proposition is that which gives what I have called the anatomy of thought and language.

There are, he says, four stages in it: sensation, perception, conception, naming. Practically they are inseparable and simultaneous. But they can be conceived of and named separately. I have already made one or two observations on this subject, in considering what I have remembered as his first proposition. I may add to what I have already said, that the proposition to be correct must be confined to human beings. Mr. Max Müller would, I think, admit himself and even insist that animals possess both sensation and perception, which, as he says, imply some power of generalisation. The evidence to each man that animals feel and perceive is, if we except the evidence given by language, precisely the same as the evidence that other men besides himself feel and perceive. That perception is not distinguishable except in name and theoretically from conception seems equally plain. In perceiving a tree or any other natural object, we combine into one an immense number of things which might be separately named and thought of, and what is this but an early stage of conception? The same thing might be said of the perception of a leaf or a grain of dust. It thus seems impossible to separate, and therefore not expedient to distinguish, the two processes. The power of naming seems to be the point at which a plain recognisable difference between men and animals comes in. For this reason I should prefer for Mr. Max Müller's percept and concept to use the word ‘idea.' It is noticeable that he has very little occasion to speak of percepts in the course of his book. Indeed, there is nothing to say of percepts, as he calls them, except that they mark an ideal step in the history of a name.

4. What I have stated as the fourth proposition, namely, that the formation of conceptions is due not merely to our senses but to certain conditions stated by Kant as those under which we think, appears hardly necessary to the main course of his argument, though it is necessary to what he says of Darwin, and though it is easy to understand the satisfaction which Mr. Max Müller feels in connecting himself so emphatically with Kant and his views. To expect him to abstain from doing so would be to show ignorance of the almost invincible attractions which the discussions lying at the basis of all philosophy exercise over all who have ever taken part in them, and are specially likely to exercise over one who celebrated Kant's centenary by publishing an English translation of Kant's greatest work in a form as little difficult to be understood as the nature of the case allows.

The impression left on my mind by a careful study of Mr. Max Müller 's book is that, if he is right in his account of the part of Kant's philosophy with which he has to do, it makes no difference at all to the Science of Thought whether it is true or false, for the essence of it is only this, that without sensation thought is impossible, but that as soon as we use our senses we arrange our thoughts with reference to time and space and also with reference to certain lists or categories under one or more of which all our thoughts about our sensations may be arranged, and that neither time nor space nor any of these categories or lists can be referred to experience, because without them no experience would be possible.

That this is true as a general description of human thought and sensation no one disputes or ever did or could dispute, though of course the lists or categories may be differently named and numbered. Kant recognises twelve, Aristotle ten, Mill four, and Schopenhauer only one, but did any sane human being doubt that in all our thinkings time and place are always to be found more or less distinctly, or that our thoughts, if they are not to be chaotic, must be capable of some classification; that, for instance, it is one thing to think of the quantity of water in the sea (πόσον) and another to think of its quality as salt or fresh, green, blue, or transparent (πόίον)?

I cannot believe that any sane person ever disputed this statement or any part of it, except the assertion that as time and space and the categories are formative of experience they cannot be derived from it. The answer to the question whether this is so or not depends upon the meaning of the word experience. Mr. Max Müller, like some other writers, sometimes writes as if he thought that a fact learnt by experience must be learnt by degrees. He argues, for instance, that as soon as we understand what is meant by the assertion that two straight lines cannot inclose a space we assent to it at once and are not strengthened in our assent by any amount of specific evidence as to particular straight lines. It seems to me as reasonable to say that we do not learn by experience that a particular piece of paper is blue or red because after once looking at it carefully we are as sure of the fact as if we had it always under our eyes. He also leaves unnoticed facts from which many people infer that our conceptions of both space and time are acquired gradually. I think anyone accustomed to the proceedings of children will agree with me in saying that for a considerable time their movements show a complete unconsciousness of the nature of space. The young man born blind who was couched by Cheselden in the last century learned to see by very slow degrees. He said that 'all objects seemed to touch his eyes, as what he felt did his skin.' Moreover, 'he knew not the shape of anything, nor any one thing from another, however different in shape or magnitude.' [Quoted by Mill on Bailey; review of Berkeley's theory of vision. Dissertations and Discussions, ii. 110-12.] Many of a young child's proceedings give a similar impression. It is by moving about in different directions that it learns what space means.

To say that space is formative of or essential to these experiences appears to me to be true only in the sense in which every object is formative of and essential to our experience of it. Unless the particular piece of paper on which I am writing at this moment were before my eyes, I should have no experience at all of it, and it is essential to and formative of such experience as I have. Our experience of space is derived from seeing its contents, and noticing their positions in it and their distance from ourselves and each other, of which we are warned by slight differences of colour, the meaning of which it takes much experience and reflection to learn.

The truth, I think, is that the word 'experience' is something of a snare, and that it would be better to use instead of it, in reference to this matter, 'sensation' and 'inferences from sensation.' This would show how narrow and unimportant are the differences between (e.g.) Kant and Mill on these subjects. Kant, as interpreted by Mr. Max Müller, would strenuously contend that thought and language rest ultimately upon sensation, and Mill, I think, would have admitted that men are not mere passive recipients of impressions in sensation.

Those who say we get the idea of space from experience, and those who say that space is a form of sensuous intuition 'given ' in sensation, both appear to me to mean that without sensation space could not be known, and that sensation makes it known.

To speak of anything as being 'given' in sensation instead of being learnt by experience seems to me to be what Mr. Max Müller would call mythology. That is to say, in order to explain sensation itself and to avoid the admission that the nature of space is perceived by the senses as much as colour, it resorts to a fabulous process of gift, a conception subsequent to that of property and transfer and wholly inapplicable to the origin of the process of perception. I can perceive no difference at all between the combined action of light, touch, taste, and smell, which makes us aware of an orange, and the combined action of light, touch, and what has been called the muscular sense, which makes us aware of space, nor have I ever been able to see what you add to the assertion that the geometrical amount of space is true by the assertion that its truths are necessary. A necessary truth has always appeared to me to be no more than a common truth encumbered with an unnecessary and almost unmeaning epithet. When it is said that food is necessary to life or an eye to sight I clearly understand what is meant, namely that if a man have no eyes he cannot see, and that if he has no food he cannot live. If all that is meant by a necessary truth is a truth the knowledge of which is necessary to other knowledge, I admit that the truths about space, time, and the categories are necessary truths; but this is not the sense in which the word is used by Mr. Max Müller. If I understand him aright he means by a necessary truth a truth of which the negation is inconceivable. ['Dr Whewell's real position was that an a priori, or better a necessary, truth is a proposition the negation of which is not only false but inconceivable' (p. 585). This position Mr. Max Müller appears to accept. He gives a more elaborate account of the matter (pp. 597-601) which is not so shortly summed up, but which appears to me to involve Dr. Whewell's principle.]

This appears to me to be open to an objection which may be thrown into many forms and illustrated in many different ways, but which can be very shortly stated. It makes mankind judges not only of what is, but of what might have been, and thus appears to me to exaggerate the human powers. If we ascribe the origin of space to God, how can we possibly say what God could have done? If we do not see our way to ascribing it to anything or anybody, what more can we possibly say of it than that it is? The proposition ‘Whatever is is' is useless. The proposition 'Whatever is' (except A, B, C, and D) might have been something else appears to me to be doubtful in the extreme, incapable of being proved, and highly objectionable because it affords to uncandid persons an opportunity to dispense with the proof that common and popular opinions are true by calling them ' necessary truths' which require no proof. [The late Professor Clifford denied the absolute truth of geometry, with unquestionable sincerity, but on grounds which I do not pretend to explain. I think he held that space had a definite shape, such as not to admit of the existence of ideally straight lines. Whether he thought there was any place where space stopped, and how, if he did, he conceived of it, I do not pretend to know, but it is easy to imagine a limit beyond which there was no object capable of being perceived, no light, no electricity, no air. Between such a space and no space at all (for space is known to us only by its contents) I do not profess to distinguish, neither does Mr. Max Müller, though on grounds from which I think I differ (see pp. 614, 615).]

There is one more of Mr. Max Muller's utterances about Kant on which I will say a word. Some expressions in the Science of Thought seem to show that in one cardinal point Mr. Max Müller differs from him, I think rightly. According to him, one great object of Kant's Critique is to solve the problem approached by Locke, Hume, and Berkeley, and, as he considered, not solved by them, of the nature of human knowledge, or, what was to him much the same, of 'reason, pure and simple.' With this solution, Mr. Max Müller professes himself to be perfectly satisfied, and yet he uses language which to me at least conveys the impression that he differs from it in an essential particular. Kant, he says, admitted 'that the raw material of our sensations and thoughts is given to us from without, not from within.'[ P.132] He maintained in consequence that our sensations must have a substantial cause which was from without, in the shape of 'substances of which our sensations are supposed to tell us the attributes.' In a word he held with what is sometimes called ‘common sense,' against Berkeley, and in order to do so he appealed to what he called transcendental considerations, that is to say, considerations which, though made manifest by and in sensation, are independent of and antecedent to it. This view seems to me to involve the admission that a necessary truth may assert contradictory nonsense, for to assert that sensation is the foundation of all thought, and that our sensations necessitate us to believe that they are caused by a 'thing,' of which they tell us absolutely nothing, seems to me contradictory, while the expressions 'substance' and the like appear to me either to be nonsense, sounds without meaning, or at least to be the names of things which do not exist invented in order to satisfy the imagination—in Mr. Max Midler's phrase, if they are not nonsense they are mere mythology.

I think that Mr. Max Müller ought to agree in this, for the following reason. He says (p. 133) that Kant was 'much more successful against Locke and Hume than against Berkeley.'

This must be a delicate way of saying that Kant was successful against Locke and Hume and not against Berkeley, for the sort of contest in which they were engaged is one in which there are no degrees in success and no medium between success and failure.

Again he says (p. 448),'We cannot enter here on the question whether there is such a thing as a substance different from its attributes. Language does not take cognisance of these refinements, but follows the' vulgus;' and after a reference to Berkeley he proceeds: 'Philosophically there is much to be said for this,' &c. This is a similar admission.

Again, his remarks on 'fundamental metaphor' (pp. 327, 495, &c.) do not exactly say, but distinctly suggest, that as we attribute unity to external objects by thinking of them more or less as living, so we attribute substance to groups of sensations—I will call them percepts as a little peace-offering—merely for our own convenience. In a word, I suspect Mr. Max Müller of being a Berkeleyan, like myself, on this particular matter. [The word 'substance' seems to me to have two meanings: (1) Anything regarded as independent of other things (and as capable of being touched). (2) The parts of anything which are important for the purpose for which it is used or applied as distinguished from what is 'immaterial' (a most expressive word), as when we speak of the 'substance' of a book or of an argument; so you might speak of a German mark as being substantially equal in value to an English shilling, because the difference in small sums is unimportant, being a fraction of a farthing, the price of 1 1/2 grain of silver.]


Part 2: May 1888

5. The fifth proposition attributed by me to Mr. Max Müller is that language is the specific difference between men and animals, and that it disproves the development of men out of animals, and it contains some remarks upon Mr. Darwin. This subject is one on which it would indeed be presumptuous for almost any one, much more for me, to interpose, but I do not think that an unlearned person shows disrespect for great names, and men between whom and himself there can on their own subjects be no rivalry or comparison, when he offers what occurs to him on their discussions.

Mr. Max Müller considers that language creates an impassable barrier between men and animals, that in the whole range of our experience there is not to be found the faintest trace of any approach to speech on the part of any animal whatever, though there is evidence in endless quantity and variety of conduct on their part which if human would show a high degree of intelligence. Dog stories and elephant stories are so common and so wonderful that they have almost ceased to be interesting, and such an instance of instinct as the one already referred to of the Emperor Moth produces despairing bewilderment. What makes the matter more marked is the fact that the difficulty does not lie in producing articulate sounds, as is proved by parrots and some other birds, nor does it lie in the circumstance that men alone use sounds made with the mouth for purposes of communication. Most animals do so.
[Mr. Max Müller quotes (p. 175) ten remarkable lines from Lucretius, which, he says, contain ' all that can be said on the possible transition from the cries of animals and our own cries of pain and joy to articulate language.' The lines quoted are v. 1054-60 and 1082-6, and occur in one of the most beautiful parts of one of the most beautiful of poems, but I do not think Lucretius suggested any sort of 'transition from the cries of animals to language.' He is arguing against the arbitrary invention of language:
'Deinde putare aliquem turn nomina distribuisse
Rebus, et inde homines didicisse vocabula prima,
Desipere est' (1039-41).
And his argument is, in a word, if it is natural for animals to communicate by sound, of which he gives many examples, why then, he proceeds, should not men 'Dissimiles alia atque alia res voce notare'? Lucretius did not believe in the possibility of the development of new genera from cross-breeding:
'Sed neque Centauri fuerunt, nec tempore in ullo
Esse quennt' (875-6).]

The peculiarity of language is that it consists of generalisations or abstractions which are signified externally by sounds and reached as thought by mental processes of which, says Mr. Max Müller, Kant gave the true account after Locke, Hume, and Berkeley (to mention no others) had investigated the matter imperfectly. Of such mental processes animals show no signs whatever; the inference is that there is nothing in them from which language could be developed or evolved. Mr. Max Müller  says [P.153] that he has such 'a belief in Darwin's intellectual honesty that I should not have been surprised at his giving up his theory of the descent of man from an ape or one kind of animal if he had been acquainted with Kant's Critique of Pure Reason;' and he charges Darwin's many disciples [p.152] with 'being unhistorical, that is, of being outside the great and continuous stream of the history of philosophy or having neglected to pay that attention and respect to their predecessors which they deserve,' because they have entirely neglected what he regards as the discoveries of Kant.

Upon this subject I will make only such remarks as are not unbecoming one who occupies the seat of the unlearned.

1. In the first place the whole subject appears to me to have lost all interest by the general discredit into which the Biblical account of the creation has fallen. If God did not create man, what does it matter how men originated?

2. In the next place it appears to me to be hopeless on this subject to get beyond conjecture and hypothesis. The definite evidence which I have seen in favour of all theories about it leaves reasonable doubt enough as to the truth of all of them to empty every gaol in the kingdom.

3. I think Mr. Max Müller 's assertion does, if it is made out, raise a difficulty in the way of Mr. Darwin's view as to the descent of man, and that it requires a more distinct answer than, according to his account, it has ever received. The amount of mental effort which must have preceded language must, according to Mr. Max Müller, have been enormous. The amount of thought which went to the formation of the simplest language must have been very considerable; but the process, once set going, is rapid, and it is difficult to believe that if it had begun in the days of Argus Mr. Max Müller would not have taught Waldmann at least four or five of the (see p. 9) chief European languages, and that Mr. Matthew Arnold would not have enabled Geist to give lectures, not at present unneeded by dogs, on sweetness and light. It is hardly conceivable that significant roots should have been in any way reached by the habit of uttering cries or interjections like those of animals. I have not space to give any account here of what Mr. Max Müller says upon these ways of accounting for the origin of language. He calls them the ' bow-wow' and 'pooh-pooh' theories, considers them in many parts of this and other works, and, I think, makes it perfectly plain that though, to an infinitesimal extent, human language may consist of imitations of such sounds, millions of ages of barking or howling would have no tendency at all to enable a dog or a wolf to invent significant roots. [Mr. Max Müller  seems to admit that uhu, the German for an owl, may be an imitation of hooting, but he says neither γλαύξ nor owl can be derived from any such sounds. The vulgar derivation of 'owl' I thought was 'howl,' as the word was formerly spelt. The h remains in 'howlet.'] If he has done nothing else, Mr. Max Müller appears to me at least to have set this matter at rest once for all.

There are, however, some remarks to be made which tend to show that the neglect of Kant with which Darwinians are charged is hardly so great a fault as it looks. Mr. Max Müller certainly does minimise the distinction between men and animals, he narrows the gulf so far that to the imagination at least the transition seems much less impassable than it otherwise would. His Rubicon becomes a brook. He admits that a time existed when men had no language at all. He finds the origin of language, or at least suggests as a possible and probable origin of it, certain sounds by which he supposes dumb herds of gregarious men to have accompanied their various labours. When they dug, they made a noise like ‘Khan;' when they rubbed, something like 'Mard,' and so on; and it is a strong pledge both of his candour and of his courage that Mr. Max Müller points out all this, enlarges upon it,[Pp.174-8] and nevertheless frankly insists on the distinction between the cries of animals and the roots of language as the foundation of his case.

There are some other points on which he is not so explicit, or which he does not think it worth while to explain. The last words of his review of Kant's philosophy[P.151] are: 'That without which no experience, not even the simplest perception of a stone or a tree, is possible cannot be the result of repeated perceptions. And we may add as a corollary: all percepts are conceptual.' Surely beasts have perceptions, and, therefore, as I have already said, conceptions, up to a certain extent. Surely, too, as far as we can judge, they are as much aware of space and time as we are, or, to use Kant's rather heavy language, space and time are with them, as with us, 'fundamental forms of sensuous intuition.'[P.188] We learn, moreover, that even if the human mind was ultimately evolved from a mollusc 'the category of causality . . . works in the mollusc. Dogs, therefore, à fortiori, must know of the categories,' and if so the transcendental side of truth and knowledge must be open to animals as well as to men. This appears to be a strange conclusion, and to weaken the barrier which, according to Mr. Max Müller, Kant established against Darwin by proving that there is a transcendental side to human knowledge which affords, as I understand him, a root for language and thought.

With regard to Mr. Max Midler's general relation to Darwin and the doctrine of evolution he remarks with obvious truth that it is absolutely impossible for a student of language to be anything else but an evolutionist, and he seems to me to prove with almost superfluous wealth of illustration that his own particular study affords by far the best attested case of evolution to be met with anywhere. The evidence which he produces in every page of his book of the gradual change of words and formation of language for very many centuries and of the regular way in which the changes take place is overwhelming. The curious thing is to see that in the presence of modern scientific opinion he is almost as nervous about being suspected of being unsound about evolution as a clergyman used to be, say, fifty years ago about being suspected of sympathy with German philosophy or an Oxford graduate of, say, thirty years ago or less of being a 'damned intellectual.'

6. The proposition which I have numbered 6 is in one respect the most important of Mr. Max Müller's statements. It is certainly the most surprising of them, and it is one on which I cannot affect to do more than to repeat accurately what he says.

It is developed at full length in several chapters of his book, namely, the 5th, 6th, 7th, and 8th. The different points do not admit of being very distinctly separated from each other, and the whole matter is treated with such a wealth of seductive detail that any abridgment of the main doctrine appears almost too dull and barren to print. I think, however, the following few sentences give an accurate account of the principal matters which Mr. Max Müller considers himself to have established.

His first point is that, on the fullest examination, all the words in the Aryan languages may be reduced to two component parts, one of which he calls the root and the other a demonstrative element which takes the form of a suffix, affix, or prefix connected with the root and limiting its meaning, and which in later times were represented by terminations of declensions and conjugations.

The root always expresses a conception. The demonstrative elements, he thinks, 'must be considered as remnants of the earliest and almost pantomimic phase of language, in which language was hardly as yet what we mean by language, namely logos, a gathering, but only a pointing.' The roots, he says, are the ultimate facts of the science of language, as to the origin of which only conjecture is possible. The number of roots in any cluster of languages is, he says, estimated at about 1,000.[P.183] And the result of a long and most curious inquiry (chapter vii.) into the roots of Sanscrit is that there are about 800 of them, which express 121 conceptions and no more. Of these 121 conceptions he gives a list, pp. 622-32. It consists entirely of verbs, most of which describe very simple and primitive acts—to dig, to plat, to crush, rub or smoothe, to smear, to scratch, to fall, to cut, to join, to fight, to cook, &c. It must be observed that he reckons as one conception verbs having closely analogous meanings, or it is perhaps more proper to say that, as a conception is not itself a word, he in many cases indicates the conception by several words, as, 59, reach, strive, race, have; 106, weary, waste, slacken. On the other hand, far the greater number of the conceptions are represented by more roots than one. There are, for instance, sixteen roots which mean bum, sixteen which mean speak, twenty-nine which mean shine, and 45 which mean inarticulate noise, each root varying in the sort of noise represented.

Of these roots and conceptions Mr. Max Müller says: 'There is no sentence in English of which every word cannot be traced back to the 800 roots and every thought to the 121 fundamental concepts.' The number of English words is said to be 250,000. [I have not space to give any account of the process by which the roots are said to have been gradually changed into words. Mr. Max Müller  calls it the process of phonetic change, and says that it consists of changes which take place in different languages according to fixed and known rules like what is called Grimm's law. It is, no doubt, a matter of much interest and importance, though rather in a narrow way. On Grimm's law see Lectures on Science of Language, ii. 216; also Science of Thought, p.353, &c.]

The manner in which this result has been reached is as follows. The roots themselves are verbs. The other parts of speech were all derived from them by the addition of the demonstrative elements already mentioned. Thus the primitive men who were the authors of language ' if they wished to distinguish the mat as the product of their handiwork from the handiwork itself they would say "Platting-there;" if they wished to encourage the work they would say "Platting-they or you or we." . . . How some of these elements came in time to be restricted to certain meanings, such as here, there, he, thou, I, it, &c., we cannot tell.' Enough, however, remains to indicate that all the parts of speech were produced by the various demonstrative elements which were applied to different roots, and which thus, to use Mr. Max Müller 's language, applied to the primaeval conceptions all the various categories of Aristotle.

From the very first, he tells us, men spoke in sentences, even when they used single roots. Thus 'Dig!' pronounced in a loud voice would be equivalent, according to the tone and the circumstances in which it was uttered, to 'Friends, let us dig,' or, 'Dig ye, O slaves!’ In accordance with this view Mr. Max Müller considers that the Imperative Mood was probably the first of all parts of speech. Nouns substantive, he supposes, arose thus:—All nouns were originally significant. Thus a wolf, which in Sanscrit is Vrika, comes from the root VRASK, which means to tear, and so meant tearing—a tearing thing. This shows that all substantives were originally adjectives as well.

Moreover, every root would be capable of many different meanings and applicable to all sorts of different things. ‘"I shake" might mean I shake a tree, or I am in a state of shaking, i.e. I tremble, or I shake by him, i.e. I am shaken by him.'

Connected with the Sanscrit root DHÛ to shake, we have not only Sanscrit dhûti dust, dhûma smoke, but Greek θνμος, not so much what is shaken as what is itself in a constant state of commotion and activity. It is possible that the same root may account for the Greek θαύμα, originally the feeling of wonder and astonishment. Then what causes that feeling, a wonder or miracle?

These few specimens show with what rapidity, what variety of meaning, what extraordinary profusion, the different words which would from time to time be needed would be supplied out of a very small original stock. Mr. Max Müller suggests that each root would be applied by a few demonstrative elements to each of the different categories of language. The original root ' shake' would make, e.g., shaking—a substantive—smoke, dust, the wind. These are words in the first category of ούσία, or substance. As an adjective, 'tremulous,' in the category of ποίον; as a verb, intransitive, transitive, or passive in other categories, and in each of these different categories it might have as many meanings and be connected with as many kinds of shaking as the original root in each of its forms. Eight hundred roots, each adaptable to ten categories, gives at once 8,000 words, and when we remember that each root may have several meanings, this number must be again multiplied by the number of all these meanings; and if we take into account derivatives of these roots and their various combinations, the number of possible words is seen at once to be practically infinite. It is indeed as large as the number of combinations of the letters of the alphabet.

Roots, Mr. Max Müller says, are ultimate facts in the history of language. We cannot pretend to say how they originated. He tells us 'an illustrious philological society at Paris' passed a resolution 'never to admit a paper or allow a discussion on the origin of language;' nevertheless he offers a conjecture on the subject which he regards as probable. Herds of men, as yet dumb, were, he thinks, accustomed to carry on various works in company with each other. They made one noise when they dug, another when they platted, and so on; and these sounds were the original roots, for in virtue of their character as men they possessed the power, not merely of making these sounds, but of attaching a significance to them. It is impossible to do justice to this view strictly, or even to distinguish it from the views held by Darwin, to which it has a considerable similarity. The discussion of it is a most interesting part of the Science of Thought.[See great part of chap, vi., and especially pp. 290-307.]

Be this as it may, and assuming that Mr. Max Müller  is right in his main statements of fact, which rest on abundant definite evidence, let us look for a moment at the general consequences which follow. It becomes in the first place clear that the question which occupied Locke, Hume, Berkeley, and Kant as to the nature of abstract terms has been solved. Whatever may be the inference from the fact, the fact seems to be proved, if Mr. Max Müller is right, that the oldest words to be found are abstract words, and express general conceptions, conceptions relating in all cases to acts done by the persons who used the words. This is certainly an infinitely more definite intelligible conclusion, and one far better capable of being tested than those of the old-fashioned inquiries in question, which were essentially arbitrary and scholastic. Mr. Max Müller  is, I think, entitled to say that his researches are more favourable to the part of Kant's views with which he has to do than to his antagonists', and that philology goes to show that both Kant's categories and Aristotle's form a fair framework for the classification of all possible thoughts, besides showing that in the earliest times about which we have any means of knowledge the thoughts of men were conversant with general terms and not with names arbitrarily imposed on particular objects perceived by the senses.

Besides this philosophical inference, which he appears to me to be fairly entitled to claim as being founded upon facts observed with strenuous labour and capable of being verified by others, his argument suggests another which is not new but which is certainly highly important in reference to all philosophical discussion. All our words, and therefore all our thoughts, arise from some 800 roots, and a much smaller number of conceptions, say 121, which those roots suggest, but all these conceptions refer to human acts of the commonest kinds. All our thoughts, therefore, are either the names of such common acts or are metaphors derived from them. The verb 'to be' is derived from roots meaning breath, grow, dwell, and time. The most awful terms of religion, God, the soul, the most ecstatic expressions of passion, the most ardent poetry, the most trivial commonplaces, are as much made out of these roots as the human body is made out of lime, water, and a few other ingredients equally familiar. [Mr. Max Müller  by way of example reduces to its roots the following phrase from a leading article in the Times: 'Every Englishman is entitled to his grievance, as may be proved out of Magna Charta and the Bill of Rights,' p. 655.]

The enormous importance of this result, which was to a considerable extent anticipated conjecturally by Locke, but which Mr. Max Müller has placed in the clearest possible light by a vast accumulation of evidence, is self-evident, but I shall say what I have to say about it in connection with the last proposition, which is as follows:

7. Language is subject to the diseases of mythology and metaphor, the only cure for which is definition.

Notwithstanding his ardour about philology, Mr. Max Müller is no blind worshipper of words. The following sentences condense into a few lines remarks which recur in every part of the book.[ P.581] 'The advance of true philosophy depends here ' (he had been giving reasons for discarding the use of the word 'species'), 'as everywhere, on a true definition of our words. [His objection to 'species' is that it means only ειδος, likeness or varying shape, whereas 'genus’ refers to the common descent of individuals from one original pair or stock. Whether you call a man and a monkey members of the same species depends on the degree of similarity which you ascribe to them. Whether they belong to the same genus depends on the question whether they have common ancestors. The first question is obviously very much a matter of taste, and cannot be profitably discussed. The second is a question of fact dependent on evidence. The 'human mind,' says Mr. Max Müller, is the 'origin of species.’ Mr. Max Müller is quite hard upon 'species.' It may have had a good character in its youth, but now, alas, 'as applied to natural history, species is a myth, that is a spurious and deceitful word, and that species must go into the same limbo as Titans and Gorgons. ... It is dead, and must be struck out of the dictionary of philosophy.' Poor species! De mortuis nil nisi bonum.] They want constant defining, refining, correcting and even removal, till in the end the most perfect language will become the most perfect philosophy.' We need not be afraid of words. They have in them nothing but what men put into them, and when in the course of ages they lose their contents and signify nothing, they ought to share the fate of' species.' This manly straightforward way of looking at the subject is well suited to the frankness with which Mr. Max Müller states the diseases of language, which indeed are easily connected with its essence. These are metaphor and mythology, or metaphor running into mythology. The steps appear to be as follows:—

First, no word is or can be an adequate representative of anything. Roots are inadequate because they are of necessity vague, being all conceptions or verbs, and they are, moreover, applicable to many subjects. Nouns substantive are still more inadequate, because they name given objects by reference to isolated and not always very characteristic qualities. A wolf does much else besides tearing, though he is named 'the tearer.' Gold might just as well be called 'the heavy' or 'the ductile ' as 'the shining.' The very act of perception or conception, as has been already remarked, bestows upon the thing perceived or conceived a species of independence and unity which, so far as we know, does not belong to it. 'We never see a species or handle a species,' says Mr. Max Müller.[P.571] The same is true of a substance, yet all language applies to things in themselves, and fails to recognise what Mr. Max Müller  seems to be disposed to believe on the subject with Berkeley and against Kant. Language, therefore, is radically inadequate. It is never co-extensive with what it names, even when it is confined to simple matters and is examined with the closest scrutiny which the mind can apply. As words multiply and knowledge increases this defect involved in its very essence becomes more and more apparent and produces effects more and more strongly marked. Metaphor comes in and things come to be named not by their own qualities but by their resemblances to other things. In Locke's words '" imagine," "apprehend," " comprehend," "adhere," "conceive," "instil," "disgust," "disturbance," "tranquillity," are all words taken from the operations of sensible things, and applied to certain modes of thinking.' From these words and metaphors we easily proceed to the invention of nouns substantive which depend upon them. The mind, the reason, the understanding, the will, the imagination, the fancy, the conscience, &c., are conceived of as definite existences or powers. As Mr. Max Müller points out, this, even in modern times, is only one step from the personification of particular metaphors. The Goddess of Reason was worshipped in the course of the French Revolution by crowds which were not aware that reason etymologically means only counting; probably no single man in them knew that the steps from the word to the goddess were all well-known illustrations of a process nearly universal and explanatory of a large part of human language and philosophy.

The path in older and simpler times from words to imaginary beings was wider and more quickly trodden. I will merely indicate the direction which it is said to have taken. Not only were things invested with substantiality, but they were invested with personality. The pieces of wood and iron or stone which were called a spear were soon said to be ferocious and pitiless. The Hindoo and many other pantheons were rapidly peopled by beings made of metaphors and personified names of things falsely supposed to exist. If Mr. Max Müller is right, the dawn by itself was the occasion of masses of mythology; clouds, storms, dogs, cows, were appropriated to it under an indefinite number of names. Jupiter, Juno, Venus, and all their innumerable relations, colleagues, and associates were produced in the same way. Is it not written with all imaginable particulars in the latter half of vol. ii. of his Lectures on the Science of Language? Of course in all this there is a great deal which may be right or wrong, fanciful or rational. The indications from which Mr. Max Müller draws his inferences appear to an unlearned eye somewhat slender evidence, and in the nature of the case the matter must be doubtful, and derivations15 must admit of great difference of opinion. [Inter alia, compare Mr. Max Müller's derivation of ' beefeater' from 'buffetier' with the article on the word 'beefeater' in Dr. Murray's great dictionary, which holds that beefeater means (1) One fed upon beef; (2) A member of the guard established by Henry VII. 'Beefeater'—beef + eater (cf. O.E. hlafasta, lit. loaf-eater, a menial servant). 'The conjecture that sense 2 may have had some different origin, e.g. from "buffet," sideboard, is historically baseless. No such form of the word as " buffetier" exists, and "beaufet," which has been cited as a phonetic link between buffet and beefeater, is merely an 18th century bad spelling, not so old as beefeater.' Sir Alfred Lyall's essay in (I think) the Fortnightly Review on the 'Religion of an Indian Province' sets forth views as to the origin of mythology, and in particular as to the readiness with which conspicuous and even notorious people are turned into gods, which show a strong probability that Mr. Max Müller has given the Dawn &c. rather more than their due.]

This, however, is a small matter. Put it at the highest, and abundant—I might almost call it oppressively abundant—evidence remains to show that between metaphor and mythology, the inference from the existence of a word to the existence of a ‘think' corresponding to it, is to the last degree unsafe. Thought and language are no doubt identical. But neither of them corresponds to definite things or thinks unless those things or thinks are either seen, heard, touched, tasted, or smelt, or unless their existence is inferred from such sensations; and the inference from what is seen to what is not seen is liable to all sorts of snares. What, for instance, can we infer from the collection of phenomena called collectively electricity? Can we infer the existence of an electric fluid? Or is the expression electric fluid only a way of recording the fact that most of the phenomena collectively called electricity accurately resemble in many ways the phenomena of water or air. Any one who wishes to see what strange questions are thus raised or raiseable will find an excellent specimen of them in the article ' Ether' (by the late Mr. Clerk Maxwell) in the last edition of the Encyclopaedia Britannica. This, in itself, is a contribution from the Science of Language to the Science of Thought sufficient to revolutionise it.

The extent to which the combined processes of dismissing myths and exposing the true character of metaphors will go in diminishing the mass of philosophical inquiries and in showing that a great many standing quarrels are quarrels about nothing at all, is shown by different illustrations given in various parts of Mr. Max Müller's works in a way which very possibly gives an exaggerated view of their importance. I will specify one or two of them; but it is necessary to resist the fascination which they exercise. One of the most striking is the speculation on the word Nature, [Lectures on the Science of Language, ii. 617.] which means 'she who gives birth, who brings forth.' Passing over earlier illustrations of the way in which nature has been personified and invested with all manner of personal attributes, we find Dugald Stewart making Nature to a great extent a synonym for God, and our very newest school of philosophers speaking of it as something which is continually occupied in selecting—a work which is by many people described in such language that it is difficult to say whether they do or do not regard Nature as a kind of she-god or at times as a kind of she-devil. Nature, again, is opposed to 'the supernatural'—a neuter adjective which, like so many neuter adjectives, neither is nor can be connected with any substantive, and which whether so connected or not is a word as ill-defined and as hard to be understood as the feminine substantive to which it is opposed. How can you discuss the neuter adjective unless you are agreed as to the meaning of the feminine substantive with which it is contrasted?

Another remarkable word is 'Nothing.' [Lectures on Science of Language, ii. 378-82.] We find the late John Sterling exclaiming: 'The thought of an immense abysmal Nothing is awful, only less so than that of All and God.' We also find that ‘it has been dragged into the domain of religious thought, and under the name of Nirvana has become the highest goal of millions among the followers of Buddha.

Is it too much when Mr. Max Müller describes these as cases in which 'language has reached to an almost delirious state, and has ceased to be what it was meant to be—the expression of the impressions received through the senses, or of the conceptions of a rational mind?' 'Nihil,' says Mr. Max Müller,' is nihilum, which stands for nifilum, i.e. ne filum, and means "not a thread or shred"'—a curious anticipation of the Joe Miller Irish definition, 'Nothing is a stocking with neither a foot nor a leg.'

The derivations of the two words, or sets of words, represented respectively by deos, Deus, and its modern derivations, and 'God and Gott,' together with the accounts which Mr. Max Müller  gives of the various senses in which they have been used, raise questions of infinitely higher importance, at which I only glance in passing; but it is impossible to read what he says without feeling that he has raised questions which will do much more to shake all received opinions than any quantity of transcendentalism and 'necessary truth' will do to confirm them. [See Science of Thought, pp. 277-83, Science of Language, ii. 316-18, 479-505, and, in Introduction to Science of Religion, pp. 200-272, correspondence about the Chinese name for God.]

As a last illustration, I will refer to the way in which he deals with the two words 'spiritualism' and 'materialism.' Spirit is breath. It is apparently used, and always has been used, as a metaphor of the appropriateness of which we have no means of judging, by which we affirm the existence in and as the chief part of man of something invisible and intangible, to which perception, thought, and the origination of voluntary motion are ascribed. Matter is materies, the proper meaning of which is building-timber, and the two words together may thus be taken to mean the perceiving subject and the objects perceived. Matter and spirit are thus correlative terms, each of which becomes contradictory if it usurps the place of the other. The meaning of matter, the reason why the word is appropriately generalised from such a thing as building-timber, is something which can be perceived by the senses, but cannot perceive, and this implies spirit, or some other word, to represent that which perceives but is not perceived. It is sometimes said, 'No thought without phosphorus.' Mr. Max Müller  would reply, 'True, but also, No phosphorus without thought,' and this, he says, is revealed to us by the whole structure of language, which has names for a number of objects perceived, and also for the perceiving subject; these last-mentioned names being metaphors pervading many languages and many nations, and used for thousands of years, thus showing that men have always felt the want of such names, and have to a great extent agreed as to the kind of names they want. So far I agree with Mr. Max Müller, if my agreement is of any importance, and I may quote a line or two of his discussion as pure Berkeleyanism: 'Matter, in the usual sense of the word, as something outside and independent of us, does not exist. Spirit, in the usual sense of the word, as something inside and independent of the world without, does not exist. There exists' (surely exist)' a perceiving subject and a perceived object.'
[Science of Thought, p. 573. A few lines lower occurs a passage which sums up his whole theory in a few words: 'Where do we find that mind and intellect? Some say in the brain. The brain is a wonderful labyrinth. I have looked into it, and examined it, but I cannot find any trace of mind or intellect, as little as I can find it where the ancients saw it—in the heart or the stomach. The brain may be a sine qua non of intellect, as the eye is of sight or the ear of sound, but as little as the eye can see and the ear hear can the brain think. I find intellect nowhere but in the products of intellect, namely in words. These I can hear and perceive, these I can trace from their present form to their most simple and natural beginnings. The whole world becomes clear and transparent as soon as I see it in words, not in sounds but in words, in living not in dead words, in words as independent of their sounds as the oyster is of its shell, in words which are thoughts as much as thoughts are words.']

I think that in this, and in the passage quoted in the note, Mr. Max Müller  goes just a little too far. Upon the evidence as it stands I should agree with him, but words are not always right. Mr. Max Müller himself shows at how many gates error may enter into them. Suppose some one succeeded in making an artificial brain—a thing made of wood and springs which could feel, speak, and think, language would be shown to be wrong, and it would be proved that matter could think, or rather that the conception of the world as consisting of invisible subjects and visible objects of perception on which language is founded is not a truth necessary or otherwise but a πρώτον ψεύδος. The question whether such a feat is possible is one which the actual construction of language as it is cannot predetermine. The utmost that can be said about it is, that there is absolutely no indication, even the faintest, to suppose that it could ever be performed. In the same way if a dog could be taught to talk in the human sense of the word, Mr. Max Müller would be refuted. Precisely the same may be said of this possibility.

The deepest and most interesting and important by far of Mr. Max Müller's applications of his doctrines is his application of them to religious belief. I must glance at it in the most cursory manner.  It is indeed only hinted at in the Science of Thought, but is more fully treated in the Science of Religion.

In part of what he says I most heartily agree. Of part he fails to convince me. His repeated attempts to establish the existence of necessary truths and some of the other elements of transcendentalism do not seem to me to be more successful than those of his predecessors in that undertaking. They look to me like desperate efforts to jump off one's own shadow, and to acquire, by asserting its existence, a warrant for the truth of some of our opinions which they can never have. In this work he hints, in a word, at faith.

'I how' in Greek (οίδα) meant originally 'I have seen, and therefore I know.' To apply such a word to our knowledge of causes, forces, atoms, and faculties would be a solecism; to apply it to God would be self-contradictory. We want another word which should mean 'I have not seen, and yet I know,' and that is —faith.’[P.609]

In the Science of Religion [Pp.14,15] he dwells on the same subject, and in his Lectures on the Science of Language [ii. 627] he describes faith as a third kind of knowledge 'which possesses evidence equal, nay superior, to the evidence of sense and reason,' and the absence of which is 'sometimes called spiritual darkness.'

To all this I am as strongly opposed as any one can be. It tempts one to say of faith what Mr. Max Müller says of species: that, in the sense he gives it, it is a false word, supplying an excuse for the dishonesty of people who substitute feeling for reason and who are arrogant enough to ascribe to those who differ from them a species of blindness.

When a man tells me that I am blind he makes a remark which, unless he can prove it, is most offensive. In the case of actual blindness such an assertion can be readily proved in a thousand obvious ways. Nothing is easier than to persuade a blind man that you possess a sense of enormous importance which he does not. Standing by his side and holding his hand you might describe to him distant objects which he might afterwards handle, taste, and smell for himself. A man who has little or no musical ear is easily convinced that others are more fortunate, but if A and B differ as to the existence of God, A has surely a right to feel insulted if B, being neither wiser nor better than himself nor able to appeal to any argument of which both acknowledge the force, says, 'Poor fellow, you are spiritually dark, and I have faith.' It is remarkable that in this matter Mr. Max Müller is directly at issue with Kant, as he himself declares. [Science of Religion, p. 15. 'Kant most energetically denied to the human intellect the power of transcending the finite or the faculty of approaching the Infinite. He closed the ancient gates through which man had gazed into Infinity, but in spite of himself he was driven, in his Criticism of Practical Reason, to open a side-door through which to admit the sense of duty, and with it the sense of the Divine. This has always seemed to me the vulnerable point in Kant's philosophy.']

 On one, and that a most important point, I am happy to be able to agree with him as emphatically as I dissent from him on the points mentioned. He appears to me to have proved, in the clearest way and by the most satisfactory evidence, that in all ages and in all countries, and from the very earliest period of which we have any sort of records, men have been continually trying by all sorts of devices, by mythology, by philosophy, by naming objects which lie beyond sense (I refer to such names as soul, self, spirit, God, &c.), to give shape, consistency, and definite form to thoughts which can hardly have been suggested by the bare exercise of their senses, and this is a most weighty fact. Mr. Max Müller's procedure in relation to it appears to me to contrast most advantageously with other attempts to explain common beliefs on this matter which I will not further notice at present, but which it might be most interesting to examine. Mr. Max Müller  holds out hopes that if he live long enough he may publish a 'Science of Mythology,' and also a treatise of which it will be the object to answer the question who or what is the self, and how does he differ from the eyes which are but lenses, and the words which are but instruments. [Cf. p. xii and p.550] If these works are ever accomplished he will, I think, have a right to say that he has taught us more about philosophy than all other living writers put together, and I am not at all sure that the same may not be said of his achievements, whether he publishes anything more or not. Great writers on metaphysics appear to me to have been doing their utmost to abolish metaphysics ever since Locke; but Mr. Max Müller, with all his Kantism, has carted and is carting away more of them than any one else ever did.

I have only in conclusion to make some remarks on the means by which Mr. Max Müller proposes to produce the great results which he partly has produced and partly hopes to produce. It is, in one word, by Definition, and especially by Definitions of the leading terms in various branches of knowledge. Here again I feel bound to give the most complete assent to what he says. In the particular subject on which I may claim to know something, law and legislation, such a reform would do more for simplification and the reduction of unmanageable bulk than all the Codes and all the Acts of Parliament that ever were or that ever could be passed, and the sciences in which real unmistakable progress has been made are those in which there is no dispute as to the meaning of fundamental terms. The truth of this appears to me to be proved to demonstration by comparing political discussion to Euclid's elements, and by thinking what Euclid's elements would be reduced to if such an expression as 'liberty' were put without further definition amongst the definitions or axioms. I firmly believe that all or almost all the unpopularity under which Political Economy has always laboured would be removed if it were clearly understood that its object is simply to examine the means by which a person assumed to have absolute control over his own property may most effectually increase it by means not forbidden by law, and that a teacher of Political Economy has no more business as such to advise people to make the acquisition of riches the chief end of life than a physician has to advise them to make the acquisition of physical health or strength the chief object of life. It is, however, needless to insist upon the importance to all knowledge of the utmost attainable clearness and simplicity in every part of it, and this can be obtained only by definition.

An essay on definitions would be interesting in the extreme, but I will not be tempted into saying more than a very few words about them. I do not think that great philological knowledge, historical or otherwise, is essential to such work. It may, no doubt, be a great help and guide in such matters, but it may also receive too much attention.

By definition I mean the limitation of the meaning of a word already in use so as to make it include or exclude senses in which it is not intended to be used. It always involves more or less of an appeal to the reader's good faith, for it is almost impossible to define in such a way that people cannot misunderstand or misrepresent if they wish to do so, and the attempt to frame definitions which are proof against disingenuousness often involves so much intricacy that the definition does more harm than good. Instances of this in legal definitions are innumerable, as every lawyer knows. [Here is a single instance from the Indian Penal Code, s.319: 'A person is said to use force to another if he causes motion, change of motion or cessation of motion to that other, or if he causes to any substance such motion or change of motion or cessation of motion as brings that substance into contact with any part of that other's body, or with anything that other is wearing or carrying, or with anything so that that contact affects that other's sense of feeling, provided that the person causing the motion or change of motion or cessation of motion causes that motion or cessation of motion or change of motion in any one ['or more' seems to be wanted here] of the three ways hereinafter described, first, by his own bodily power, secondly, by disposing any substance in such a manner that the motion or change or cessation of motion takes place without any further action on his part or on the part of any other person, thirdly, by inducing any animal to move, to change its motion, or to cease to move.' I think the following would have been shorter, clearer, and fuller: A person is said to use force to another, whether he does so directly or by means of any animal or thing. Suppose a man receives a blow without moving, is not force applied to him? or suppose I cut a ditch in my land into which, long afterwards, someone falls, do I apply force to him?] On the other hand, the want of a definition may cause endless expense and throw confusion over great departments of law. I need only mention the words 'possession,' 'malice,' and 'evidence,' and refer to the apparently perfectly simple words of the 17th section of the Statute of Frauds and the decisions upon it, to show what I mean.

Leaving legal illustrations out of account, and looking only to matters of more general interest, the task of definition may be said to consist in choosing, amongst the great number of meanings which come in course of time to be attached to a word, not the meaning which is most closely connected with its root, but the meaning which will mark a practical distinction of importance and so make the word most useful. I will give two remarkable illustrations of the importance of doing this—the use of the words • law' and 'liberty.'

'Law,' according to Mr. Skeat, means that which lies or is placed, or is in due order.

Hence the word has come to be used in two main classes of senses, the juridical and the philosophical. Thus every Act of Parliament is a law, and we speak also of the law of gravitation and innumerable other laws relating to every imaginable kind of subject. The result of this has been to introduce confusion and misunderstanding both into law and philosophy. In law the phrase 'international law' continually occurs, though in the true legal sense it is not a department of law at all, but a branch of morality. On the other hand, philosophy has been confused by metaphors taken from the juridical sense of the word, and applied to what are called Natural Laws, which appear, at least to me, as wild as the shrieks quoted by Mr. Max Müller  from Sterling about the horrors of Nothing. Mr. Max Müller himself speaks of ‘the material world in which we live'—'a world governed without mercy by the law of gravity.' How could the formula that 'any two masses in the universe attract each other with a force which varies according to the inverse square of the distance' either govern or show mercy? I remember a zealous disciple of the late Mr. Maurice declaring that if he thought the law of gravitation was not a living law, but 'a dead tyrannous rule,’ he would 'in the strength and power of a man defy the law of gravitation,' which is exactly like proposing to defy the multiplication table unless it did its duty by its family. If the word 'law' were reserved for law as defined by Mr. John Austin, who was in this matter a disciple of Hobbes, and if instead of talking of laws of nature we spoke of formulas or rules for understanding nature, all this would be avoided, an immense mass of obscure and often heated language would be laid aside, and a variety of subjects would fall into their proper places.

The words ‘free,' 'liberty,' in the same way have become the centre of as many absurd and exaggerated statements as gathered about the Dawn in early times, according to Mr. Max Müller. Freedom is spoken of in all moods and tenses as a good in itself, as the great aim and object of all political institutions, as the only cure for its own defects, &c. &c. It is, in short, worshipped much more than the multiplication table, under the name of Reason, and in the likeness of a prostitute was worshipped in the streets of Paris. The original sense of ‘free' is again, according to Mr. Skeat, 'having free choice —acting at pleasure, rejoicing, and the word is closely connected with Sanskrit fruya, beloved, dear, agreeable.' This agrees in the closest way with the use of the word in English and in other languages, which warrants the assertion that it means as used almost nothing whatever except the absence of something conceived of as an evil. To be free from pain, from care, from disease, means only not to be in pain, not to feel care, not to be diseased. It would not in English, at all events, be natural to say that a man was free from health, from good character, or from happiness. I do not think any definition is required or could be given of the word 'liberty;' what is wanted is the realisation of its purely negative sense. It would be one of the most important correctives and sedatives to human thought that could be conceived to insist, when the word is used, upon clear explanations on the question, Who is to be at liberty from or free to do what? We should thus have a distinct test as to the goodness or badness of the proposed liberty, and be delivered from masses of nauseous rhetoric.

The Nineteenth Century, April-May 1888.