THE 
 
 RELATIONS OF LANGUAGE 
 
 TO 
 
 THOUGHT. 
 
r 
 
THE 
 
 RELATIONS OF LANGUAGE 
 
 TO 
 
 THOUGHT. 
 
 "WILLIAM EOSCOE BURGESS, M.A., 
 
 VICAR OF CHRIST CHURCH, LATCHFORD. 
 
 'E5o|6 5^ fxoi XP^""" *'* Tovs \6yovs KaTa<puy6vTa iv iKeifois 
 (TKvniiv TcSy ovruiv t^v a\ri6eiav. — Platonis PhcEdo. 
 
 -£2vS 
 
 WILLIAMS AND NOEGATB, 
 
 14. HENRIETTA STREET, C0\T:XT GARDEN, LONDON; 
 
 AND 20, SOUTH FREDERICK STREET, EDINBURGH. 
 
 1869. 
 

 
 HERTFORD : 
 
 PRINTED BT STEPHEN AUSTIN. 
 
TO HER DEAR MEMORY, 
 
 BY WHOSE BEDSIDE IN HER LAST ILLNESS 
 
 THIS LITTLE WORK WAS WHOLLY WRITTEN, 
 
 IT IS NOW DEDICATED 
 
 WITH REVERENT AFFECTION. 
 
 } 
 
PEEFACE. 
 
 In consideration either of the help or of the 
 hindrance that language offers to thought, it is of 
 importance to observe the natural adaptation by 
 ■which language as an instrument is brought to 
 bear upon thought as a material. I have here 
 attempted to indicate this adaptation by noting 
 the complete correspondence of the two ultimate 
 elements of language with the two grand depart- 
 ments of thought — the subjective and the objective. 
 
CONTEN"TS. 
 
 VAGE 
 
 Chap. I. — The subjective materials of thought, or ideas. 1 
 „ II. — Discrimination of the objective ----- 14 
 „ III. — The subjective, or nominal, element of lan- 
 guage - - - 26 
 
 „ IV, — The objective, or demonstrative, element of 
 
 language- - 37 
 
 „ . v. — ^Predication, and the import of propositions - 45 
 „ VI. — The original synthesis of language - - - - 63 
 
THE 
 
 RELATIONS OF LANGUAGE 
 
 TO 
 
 THOUGHT. 
 
 CHAPTER I. 
 
 THE SUBJECTIVE MATERIALS OF THOUGHT, OR IDEAS. 
 
 In this inquiry we have to deal with the materials 
 of the thought of one person, as they are capable of 
 being identified, or at least compared, with the mate- 
 rials of the thought of another. For the sake of 
 brevity, and for some reasons to be hereafter adduced, 
 I shall call these materials of thought ideas. It will 
 be understood that by this word are denoted only the 
 elementary materials': those which analysis has not 
 been able to disintegrate : — those which Locke speaks 
 of as simple ideas. For, as he says,^ "though the 
 qualities that afiect our senses are, in the things 
 
 ^ Human Understanding, Book ii., chap. ii. 
 
 I 
 
2 TEE RELATIONS OF LANGUAGE 
 
 themselves, so united and blended, that there is no 
 separation — no distance between them ; yet it is 
 plain the ideas they produce in the mind enter by 
 the senses, simple and unmixed." It is with these 
 simple ideas, which Locke proceeds to speak of as 
 " the materials of all our knowledge," that we are 
 here immediately concerned. 
 
 There may, perhaps, be such a thing as inefiable 
 thought. There may be in the miads of some people 
 ideas that have got there by other than the common 
 and demonstrable channels : ideas that are not ob- 
 jectively derived, and which therefore cannot be ob- 
 jectively demonstrated or communicated. I can only 
 say, for myself, that I am conscious of no such 
 thought ; whilst, with regard to the object of our 
 inquiry, it is plainly all one whether there be such 
 thoughts or ideas in any man's mind, or not. It 
 is equally irrelevant to affirm or to deny their 
 existence. 
 
 Eemembered or resuscitated perceptions ^ are some- 
 times spoken of as ideas, as when we speak of the 
 idea of a rose or of an orange. It is plain, however, 
 that the perception (whether actual or remembered) 
 of a rose, or of an orange, comprises many sensations, 
 
 1 I have been advised by a friend, to -whose suggestions I am greatly 
 indebted, to substitute here, and further on, the word sensation for 
 perception. The advice has been followed in all places except those in 
 ■which, as above, it is intended to denote a natural cluster of sensations. 
 I have still thought it expedient to distinguish such natural clusters by 
 the vioid perception. 
 
TO THOUGHT. 3 
 
 referable to as many ideas ; sucli, for example, as of 
 form, coloiir, scent, taste, etc. What is spoken of 
 as the idea of a rose, is only the imaging of some 
 particular rose : — particular, not necessarily as a 
 past perception, but as a present imaging of parti- 
 cular sensations, as of form, colour, etc. 
 
 We refer our perceptions to ideas ; and it is reason- 
 ably assumed that our ideas are produced by percep- 
 tions. Yet it seems expedient to distinguish between 
 the mode of perception whereby ideas are produced, 
 and that which we afterwards learn to refer back 
 again to ideas. In the acquisition of ideas the mind 
 seems to be passive and (so far as the constituent 
 sensations are concerned) unconscious. To speak of 
 sensations unconsciously received may be thought 
 absurd. Perhaps for sensation it would be expedient 
 in this emergency to substitute the word impression. 
 What we want to express is one of the elements of 
 consciousness, and our expression must therefore not 
 imply consciousness. 
 
 That ideas should be passively and unconsciously 
 acquired need excite no surprise, since the same 
 observation may be made of perhaps all our acquisi- 
 tions. Thus all volition has been traced to an origin 
 commonly spoken of as spontaneous, though auto- 
 matic would be a more suitable word : — the meaning 
 being that the highest exercise of consciousness 
 springs from an origin in which consciousness has 
 
4 THE RELA TI0N8 OF LANG TJA GE 
 
 no play whatever. A brood of young birds, as soon 
 as they are hatched, will open their mouths wide 
 when anyone approaches the nest, as if in expectancy 
 of food. We know that there can then be no such 
 conscious motive of an action which in their after- 
 lives will be performed consciously and voluntarily. 
 In such phenomena we observe that the employment 
 of acquired powers is almost an inversion of the 
 process by which they were acquired. We must 
 therefore be careful, in investigating the origin of 
 our ideas, not to expect, in that origin, any such 
 consciousness and active discrimination of sensa- 
 tions as exist and are exercised subsequently. 
 
 The process by which ideas are acquired may be 
 described as secretive rather than as discriminative. 
 Discrimination implies the cognition of particulars. 
 Yet certainly the acquisition of ideas seems to pre- 
 cede such cognition. Ideas are produced by accu- 
 mulated impressions, and they are general because 
 the cumulus is of that which is common to all. 
 
 It has been said above that revived perceptions 
 are erroneously spoken of as ideas. We have now to j. 
 add that ideas proper cannot be called up as mental 
 images. We cannot think of sweetness as we can 
 think of an orange. We cannot think of any one 
 of our ideas apart from all others. That is to say, 
 as ideas we cannot think of them at all. We can 
 only represent them in thought as they are pre- 
 
TO TE OUGHT. 5 
 
 sented in perception, as concreted in various combi- 
 nations. There is, however, this most remarkable 
 peculiarity about our ideas, that they and they only 
 have names. Do we, then, suppose that ideas are 
 now become non-existent because they cannot be 
 reproduced individually in thought ? Then, on 
 the same ground, we must annihilate perception, 
 since we are conscious of perceptions only in their 
 relation to ideas.^ We cannot — as remains to be 
 shewn — we cannot even name a perception. We can 
 only denote it by gesticulation, or, in language, by 
 applying to it the name of some one or other of the 
 ideas to which it is referable. Here, as in the whole 
 realm of consciousness, the law of relativity pre- 
 vails. In vain do we try to grasp absolutely, and to 
 
 > " As there can be no classification or recognition of objects ■without 
 perception of them ; so there can be no perception of them without 
 classification or recognition. Every complete act of perception implies 
 an expressed or unexpressed 'assertory judgment' — a predication 
 respecting the nature of the perceived entity ; and as is generally ad- 
 mitted, the saying what a thing is, is the saying what it is like — what 
 class it belongs to. . . . The instances in which, from mental distrac- 
 tion, we go on searching for something we have in our hands, or over- 
 look that which is directly under our eyes, clearly show that the mere 
 passive reception of the visual image, or group of sensations produced 
 by an object, does not constitute a perception of it. A perception of 
 it can arise only when the group of sensations is consciously co-ordinated, 
 and their meaning understood. And as their meaning can be under- 
 stood only in virtue of those past experiences in which similar groups 
 have been found to imply such and such facts, it is clear that the under- 
 standing of them — the act of perception— involves the assimilation of 
 them to those similar groups ; involves the thinking of them as like 
 those groups, and as having like accompaniments. The perception of 
 any object, therefore, is impossible save under the form either of 
 recognition or classification." — Mr. Herbert Spencer, Principles of 
 Psychology, chap. ix. 
 
6 THE RELATION'S OF LANGUAGE 
 
 treat as an independent fact of consciousness, any- 
 elementary mental phenomenon apart from its corre- 
 late. Nor, in the failure of such an attempt, will it 
 avail to merge one of the related members in the other. 
 If we thus look upon ideas as permanent results of 
 sensations, they will not be characterized by more 
 unreality than those ^^ permanent possibilities'^ of 
 sensations to which Mr. J. S. Mill reduces our be- 
 lief in the independent existence of the material 
 world.^ And certainly I find it difficult to acquiesce 
 in that utter annihilation of ideas to which some 
 have been impelled by their opposition to Realism. 
 "General Ideas," says Mr. Bain ("Mental and Moral 
 Science," p. 180), " separated from particulars, have 
 no counterpart Reality (as implied in Realism), and 
 no mental existence (as affirmed in Conceptualism). 
 Because we have a name ' round,' or ' circle,' signi- 
 fying that certain things impress us alike, although 
 also difiering, it does not follow that there exists in 
 nature a thing, of pure roundness, with no other 
 property conjoined : a circle, of no material, no 
 
 ^ " Matter, then, may be defined a Permanent Possibility of Sensation. 
 If I am asked -whether I believe in matter, I ask whether the questioner 
 accepts this definition of it. If he does, I believe in matter, and so do 
 all Berkeleians. In any other sense than this, I do not. But I aflirm 
 with confidence, that this conception of Matter includes the whole 
 meaning attached to it by the common world, apart from philosophical, 
 and sometimes theological, theories. The reliance of mankind on the 
 real existence of visible and tangible objects means reliance on the 
 reality and permanence of Possibilities of visual and tactual sensations, 
 when no such sensations are actually experienced." — Examination of 
 Sir IVilliam Hamilton's Philosophy. 
 
TO THOUGHT. 7 
 
 colour, and no size. All nature's circles are circles 
 in the concrete, each, one embodied along with other 
 material attributes ; a certain colour and size being 
 inseparable from the form. This is the denial of 
 Realism." 
 
 " Neither can we have a mental conception of any 
 property abstracted from all others. "We cannot 
 conceive a circle, except of some colour and some 
 size. We cannot conceive justice, except by thinking 
 of just actions." 
 
 These assertions must be examined more minutely 
 before we assent to them. Prima facie it would seem 
 extraordinary that the only things which have names 
 of their own should be banished from the realms of 
 both objective and subjective existence. The denial 
 of Realism, given above, will not be called in ques- 
 tion. Because we have names of things subjective, 
 it certainly does not follow that these things exist 
 objectively. But to say that they have no mental 
 existence is to stultify the whole method of lan- 
 guage. Leaving, however, for the present, the 
 evidence of language, — it appears that the denial 
 of Conceptualism is grounded upon our inability 
 to reproduce ideas, even in our own thought, with- 
 out reference to concrete things. " We cannot have 
 a mental conception" of hardness, except with re- 
 ference to some hard body; nor of roundness, except 
 with reference to some round body. This is true, 
 
8 THE RELATIONS OF LANGUAGE 
 
 if by mental conception is intended only a resus- 
 citated impression. But it must be borne in mind 
 that tbis resuscitation is as mucb an objective re- 
 presentation as if it were an original and actual 
 impression. To think of a thing is to present it 
 objectively to the mind. The resuscitation of an 
 impression can be effected only by presentation of 
 some one or other of those objective particulars which 
 have combined to produce it. It is true, then, that 
 we can have no mental conception of an idea except 
 in connection with some objective thing. But it 
 would appear also that we can have no mental con- 
 ception of an object apart from its subjective corre- 
 late. For until the idea is constituted the consti- 
 tuent sensations are unconsciously received, as in 
 early infancy. 
 
 A corresponding fact in language may be observed 
 here with more advantage than under its proper 
 heading in chap, iv., viz., that the nearest approach 
 made to the denotation of the subjective is in the 
 pronoun of the first person, which is yet of de- 
 monstrative, that is to say objective, origin. / and 
 thou and he are only modifications of demonstrative 
 elements, 'conventionally appropriated to the distinc- 
 tion of this and this and this. Everyone may observe 
 that children, when they begin to talk, do not 
 immediately adopt the form that has been appro- 
 priated to the expression of pure subjectivity. They 
 
TO TSOUGHT. 9 
 
 prefer to speak of themselves by their own proper 
 names — by using the pronoun of the third person — or 
 frequently by the objective form of the first — ^' me 
 did it." " The objective cases of pronouns, as well as 
 of nouns," says Dr. Donaldson, ^ " are always older 
 than the subjective. This appears from the fact that 
 there are many nouns which have no subjective case 
 (for instance, all neuter nouns), but no one, so far as 
 we know, which has the nominative only. It might 
 also be inferred, from a priori considerations, that it 
 must be so. All things are to us parts of an external 
 world, and must needs be spoken of as such long 
 before the mind of man can invest the not-me with 
 the powers of agency and will which we experience 
 in ourselves. TVe feel that even the spot on which 
 we stand, and which is for the moment identified 
 with our description, for we are the here, is neverthe- 
 less a not-me, and is spoken of as something without, 
 as an object, and therefore must continue to be called 
 one, till language begins to assume a logical structure." 
 To this we add, that even when this structure is 
 assumed, it is only by a logical fiction that we 
 suppose ourselves able to denote the subjective 
 immediately, that is, independently of an objective 
 reference. 
 
 If we divest the Platonic idea of the gross Realism 
 with which it is overlaid by those who give a literal 
 1 New Cratylus, p. 132. 
 
10 TEE RELATIONS OF LANGUAGE 
 
 interpretation to Plato's figurative language : if, 
 moreover, we take into consideration the philosophical 
 exigencies of his time, we can understand how he 
 could describe the world of ideas as having more 
 certainty, fixity, and permanence, than the world of 
 perceptions. "We can see that there is a sense in 
 which this is true : that, apart from the fact that 
 such intrinsic values of certainty and permanence 
 are both indicated and confirmed by the names which 
 ideas bear, as mintage- warrant, so to speak — they 
 are contrasted, as ultimate elements, with perceptions 
 as fleeting phases of composition and dissolution. 
 There would appear also to be a difierence between 
 sensations or perceptions, and ideas, analogous to that 
 between food and vital energy. As "the life is more 
 than meat," so is the idea more than sensation : the 
 connection in each case being a process of assimilation 
 which we are not able to trace from end to end. 
 
 It is here to be noted that the Realism which 
 professed to be founded upon the Platonic doctrine, 
 seems to have arisen from a misapprehension of 
 Plato's analysis of cognition into the particular and 
 the universal. In forgetfulness that an analysis of 
 thought must disclose its elements, rather than 
 classify its varieties, " Plato was supposed to have 
 divided cognition into two kinds or classes, a par- 
 ticular and a universal kind ; and not into two 
 elements — a particular and a universal element.'' 
 
TO TEOUQET. 11 
 
 So says Prof. Ferrier/ and adds : " The question im- 
 mediately arose, what is the nature of the existences 
 which correspond to these classes of cognition ? In 
 regard to the particular class there was little or no 
 
 1 Institutes of Metaphysics, sub. Prop. yi. I do not know upon 
 what grounds Professor Farrier imputes to Conceptualism the division 
 of our cognitions " not into elements of cognition, but into cognitions 
 — not into distinct factors, but into distinct kinds, of knowledge — a 
 particular kind, called sometimes intuitions, and a universal kind, 
 called usually conceptions." This is, undoubtedly, a serious error ; 
 and it is proved, he says, against the Conceptualists " by the considera- 
 tion that in the estimation of Conceptualism our particular cognitions 
 precede the formation of o\ir general conceptions, which they could not 
 do unless they were distinct and completed." For my own part, 
 guarding against such an error, I desii'e to represent the constituents 
 of ideas, not as sensations or perceptions, much less as cognitions, but 
 as impressions unconsciously received. I am aware that this word too 
 might be pressed to yield results uncontemplated by me ; but I have 
 preferred it solely on account of its exclusion of consciousness. I 
 would say that impressions unconsciously received constitute ideas, and 
 that in the light of these ideas, the same sort of impressions do subse- 
 quently become cognitions. And, if I mistake not, physiology furnishes 
 us with a striking analogy in the fact of organs changing, or greatly 
 modifying, their functions, upon the completion of their preliminary 
 stage. 
 
 The error referred to is, however, a very serious one. It is the 
 error of supposing that the subjective element of thought can be knovra 
 and treated objectively— an error, therefore, to which all are prone who 
 attempt the consideration of that element. Because to treat of a thing 
 you must first present it objectively to your mind and represent it ob- 
 
 Iectively in your language. And thus we find even Professor Ferrier 
 limsclf accounting for the usual oversight of the subjective element of 
 all cognition " by the operation of the law of familiarity, and the fact 
 that the ego is no object of sensible experience." Now, these ciixum- 
 stauces might account for the oversight of certain objective phenomena 
 in favour of certain others ; but the supposition of such a rivalry be- 
 tween object and subject must have occurred in a momentary lapse into 
 this very error of treating the subjective objectively. Moreover, if the 
 ego appears in consciousness notwithstanding that it is not a sensible 
 object, the fact of its not being a sensible object cannot be adduced to 
 account for an undue share of attention, or consciousness, enjoyed by 
 it. For, to use our author's own words (Prop. iii. § 8), " the minimum 
 tcible pel- se, consisting of subject and object, is only accidentally but 
 not essentially enlarged by augmenting the objective factor. Popu- 
 larly considered, the universe plus me is greater than a grain of sand 
 plus mc. But this difference is altogether trivial, and of no account in 
 philosophy." 
 
12 THE RELATIONS OF LANGUAGE 
 
 difficulty .... But what kind of existences cor- 
 respond to the universal cognitions ? That was the 
 puzzle. If the analysis of cognition be a division 
 into kinds, and if the particular cognitions are dis- 
 tinct from the universal, and have their appropriate 
 objects — to wit, particular things — the universal cog- 
 nitions must, of course, be distinct from the parti- 
 cular, and must have their appropriate objects." 
 
 This is probably a true account of the origin of 
 Realistic exaggerations, and it is but another de- 
 scription of the same error — an error which is barely 
 escaped even by the acute writer here quoted — to say 
 that it is ^n attempt to present the subjective ob- 
 jectively. 
 
 If ideas are produced from sensations (or rather 
 impressions) by a process of which we have no intel- 
 lectual consciousness, and therefore no remembrance, 
 it is perhaps not very wonderful that by many 
 they should be considered as innate. Nor will it 
 be thought a very material question whether they be 
 innate or not, if the only alternative is that of their 
 being produced by impressions involuntarily and 
 imconsciously received. But it must be borne in 
 mind that the mischief of the old doctrine of innate 
 ideas consisted not so much in the assigning of a 
 wrong origin to genuine ideas as in the credit af- 
 forded to spurious ideas by a supposed hyperphysical 
 origin. 
 
 It may be requisite here to advert to the fact that 
 
TO TB OUGHT. 13 
 
 no new ideas ever accrue to tlie race or to the indi- 
 vidual in their conscious experience. The emperor 
 Augustus is said to have thought it strange that he 
 could not invent a new word. The reason is that 
 he could not discover a new idea. Let the idea be 
 produced, and doubtless the word will soon be forth- 
 coming. Unphilosophically it might be thought 
 that many new ideas had cropped up in the course 
 of scientific investigation and discovery. Take elec- 
 tricity, for example, as one of the things most un- 
 heard of and unthought of by our forefathers. Now 
 there is not one of the astonishing modes of its 
 manifestation that is not easily referable to some 
 one or other of our old ideas, as of light, heat, 
 motion, etc. If it be said that these are only its 
 effects, but that we have, of electricity itself, an 
 idea that was utterly unknown until recently, we 
 reply that of electricity itself nothing whatever is 
 known. "We know it, as we know everything else, 
 solely by its effects : solely by its operation upon 
 ourselves. We say, moreover, that we cannot even 
 name it : that we can denote it only by a word that 
 refers its operations to some one or other of our 
 ideas : — that we cannot even predicate its existence : 
 that, generally, to say of any object that it exists, or 
 that it IS, is to say nothing at all : that we can 
 predicate nothing of any object beyond some one 
 or other of its effects upon ourselves. 
 
14 THE RELATIONS OF language: 
 
 CHAPTER II. 
 
 DISCRIMINATION OF THE OBJECTIVE. 
 
 Things emerge into consciousness only in re- 
 ference to ideas : — ideas only through, the medium 
 of things. That we know the perception only in 
 reference to the idea is countervailed by the admis- 
 sion already made, that we cannot reproduce our 
 ideas in thought : that we cannot call them to mind, 
 except as concreted into perceptions : as resuscita- 
 tions of the concrete. There still remains the con- 
 sciousness of a contrast — call it by what name we 
 will. We are undoubtedly conscious of the contrast 
 denoted by the two words subject and object; and 
 there seems no reason why the practical, or what is 
 called the common-sense, view of the matter should 
 come into collision with the philosophical view. JPfo 
 man is interested in anything of which he is not 
 conscious. Now if I am conscious of a non-ego, of 
 object as distinct from subject, of matter as distinct 
 from mind, res mea est : — this consciousness is to me 
 reality in the only proper sense of the word, and to 
 the fullest extent of that sense. If not, let philo- 
 
TO TEOUGET, 15 
 
 gophers quarrel about the reality of the objective as 
 long as they please : — res mea non est. The objec- 
 tive element of cognition is, however, universally 
 recognised. The contrast between it and the sub- 
 jective has been identified with that between ac- 
 tively acquired and passively received sensation. 
 Prof. Bain makes the distinction coincident with 
 that between passive feeling and the putting forth 
 of muscular energy. It will perhaps be only making 
 the same statement in other words if we say that 
 the distinction is that between our ideas and our 
 sensations.^ 
 
 If we do not, at the first glance, see the connection 
 between our recognition of objective existence and 
 action, or the putting forth of muscular energy, let 
 us consider the means whereby we test that which 
 is called real objective existence. Is not our con- 
 viction of it exactly in proportion to the response 
 that is made to our muscular exertion ? A vision 
 appears before my eyes, and yet there remains a 
 question whether this thing exists to anything else 
 besides my passive vision. "Is this a dagger that 
 I see before me?" Plainly it is not sufficient that 
 I see it. My seeing it is indeed sufficient evidence 
 of its existence as a phenomenon ; but I desire to 
 
 ' " It is to be borne in mind that there is a radical difference be- 
 tween the sensation and the recollection of the sensation, or what is 
 properly termed the idea. This fundamental and unerasible difference 
 relates to the sense of objective reality, which belongs to the sensation, 
 and not to the idea." — The Senses and the Intellect, p. 323, Note. 
 
16 TBB RELATIONS OF LANGUAGE 
 
 place it, if possible, under another category, that of 
 so-called real objective existence. To this end I 
 must try to grasp it, and so ascertain whether it 
 resists my pressure or not ; that is to say, whether 
 or not it is such a thing as to respond to muscular 
 feeling. If not, then, however it may still haunt 
 my vision, it is supposed to be a thing of nought, — 
 a purely subjective phenomenon.^ It would there- 
 fore appear that that which we call objective is 
 made known to us chiefly (if not solely) by the 
 response made to muscular exertion. Power may 
 be put forth, as in delirious contests with imaginary 
 antagonists ; but the responsive energy failing, that 
 is to say, failing the resistance, we get no notion of 
 objective existence. The delirium over, we are willing 
 to admit that the antagonism of which we were 
 conscious was a mere subjective illusion. And it is 
 observable that, as those objects have usually been 
 regarded with the most suspicion of unreality which 
 are the least palpable, so, on the other hand, sup- 
 posed immaterial things have been denoted by the 
 names of the most impalpable of material things. 
 Take, for example, the notable instance of spirit, 
 
 1 " Come, let me clutch thee. 
 I have thee not, and yet I see thee still. 
 Art thou not, fatal vision, sensible 
 To feeling as to sight ? or art thou hut 
 A dagger of the mind, a false creation. 
 Proceeding from the heat-oppressed brain ? " 
 
 Macbeth. 
 
TO THOUGHT. 17 
 
 whicli in all languages is denoted by the names of 
 wind, air, or breath. In the former case, failing the 
 resistance of such things, we are inclined to doubt 
 their existence — that is, their being out of our own 
 consciousness. 
 
 The question is still admissible, is this muscular 
 test the sole test ? Admitting that we are not fully- 
 convinced of objective existence until it is found to 
 respond to our muscular feelings, may we not say 
 that muscular feeling alone would fail to convince 
 us ? We might perhaps come to this conclusion if 
 we were to limit muscular action by the instances of 
 its more conspicuous manifestation, as in walking, 
 handling etc. But it must be remembered that the 
 most important of even our ocular perceptions are 
 the result of the muscular activity of the eye. Thus 
 infants, until they acquire the faculty of adjusting, 
 or focusing, the eye, seem to have no idea of distance. 
 
 "We thus arrive at the meaning of that which was 
 perhaps originally the only means, and is even now 
 an important adjunct, of objective denotation : — viz. 
 gesticulative indication : — pointing with the hand or 
 finger. This is first the application of the muscular 
 test by grasping the object with the hand, or feeling 
 its resistance to the finger. It is then the recogni- 
 tion of the object as responding to such a test ; and, 
 lastly, it is the demonstration of it to others by the 
 mere tentation or suggestion of the test. 
 
18 THE RELATIONS OF LANGUAOE 
 
 Note 1. — It will be interesting here to observe the 
 connection that subsists between words that denote 
 graspinff, taking^ showing, the hand, finger, etc. In 
 the Semitic languages the same base (a demonstra- 
 tive) supplies the material for expressing " to show," 
 '' to seize," "hand," etc. Of the Aryan languages, 
 we find, in the Gothic, the first idea expressed in the 
 demonstrative yaind (compare our yonder), the next, 
 viz. " to seize " in hinthan, the third in handus, 
 "hand." In Grreek, we find he^la, " the right hand " 
 connected with a root which branches out into heUvv- 
 fii, " to show," and S6')(o/xat,, " to take." The Greek 
 -Xelp and Latin Mr may also here be adduced. Their 
 forms are explained below. In Latin, we have (pre)- 
 hendo, which Donaldson attributes to the Germanic 
 element of Latin speech, and connects with the Um- 
 brian pronominal form hont, and Gothic yaind; 
 therefore, also, with hand-us, "hand." But the 
 Latin manus is connected with /jltjvvq), as Be^ta with 
 BeLKvv/jit. With /jbrjvvo) and SeUvu/Mt, compare monstro 
 and mdico. 
 
 Bopp, indeed, connects Goth, handus, our " hand," 
 with the Sanskrit verb can, Icedere, ferire, occidere 
 (Glossarium Coinparativum, sub v. can). "We need 
 not, however, consider this derivation as contrary to 
 that mentioned above, because, if the same demon- 
 strative root that expressed primarily "to stretchout 
 the hand " can be split up to express the two ideas 
 
TO TEOUGET. 19 
 
 of shoiving and taking, as we see it has been split up, 
 it requires no great stretch of the imagination to 
 suppose a third purpose of stretching out the hand, 
 and therefore a third idea to be expressed by another 
 ramification of the same root, viz. the purpose and 
 idea of " striking." And the parallelism of the 
 Semitic root extends to this also, since we find the 
 Arabic hadda with much the same meaning. 
 
 I will here add Pott's remarks, quoted by Donald- 
 son [New Cratylus^ p. 269), upon these relations. 
 He says : — 
 
 " Handus is immediately connected with hinthan, 
 capere,^ which we also find in the isolated, and, I 
 might almost say, Germanized form ■prehendo. Grimm 
 {^Gr. ii., p. 35) is quite right in also referring to this 
 stem hunds, canis, 'the catcher,' qui capit /eras. 
 In this, too, we see that in the whole stem d is really 
 nothing but an affix, of which Grimm (ii. p. 231) 
 has very fully treated, for hunds, with the usual 
 changes, but without d, is found in the Greek kvv6<}, 
 Latin canis, Sanskrit cvaii (gen. abl. cunas, dat. 
 cun-e, instrum. cun-a, locat. cuni, nom. cud, accus. 
 cvan-am). We find the same stem in the Homeric 
 form 'yev-To S' ifidadXrjv, which points to an ancient 
 form yev-eiv, instead of eXelv." 
 
 " Buttmann is unquestionably right in comparing 
 yevTo immediately with eXero, just as the Mn\\o. 
 
 ' " Hente, which so often occurs in Chaucer, is the same word." 
 
20 TEE RELATIONS OF LANGUAGE 
 
 form KevTo for KeXeTO is adduced from Alcman. It 
 is this transition from n to I which prevents us 
 from recognising the stem hinihan in the Greek lan- 
 guage. We find the same stem with r for I in the 
 Sanskrit hri {caper e), to which belongs hasta {manus) 
 Latin hir, Greek ;)^et/a and alpeevv ; also, with an addi- 
 tion of j», Kdp7ro<i (the wrist), carpus, apTrd^eiv, Gothic 
 hreiban, greifen (see Grimm, ii. p. 45) ; nay, as it 
 appears, also in the Sanskrit kara (manus), and con- 
 sequently the whole wide-extended stem kri, the 
 general signification of which (Jacere) cannot be the 
 original one. The stem hri, as we must infer from 
 the letter h, which is always a later one, cannot 
 represent an original form, but we must always seek 
 for this in kri, which therefore corresponds to kri, 
 facere : ' That which is identical, as far as the letters 
 are concerned, cannot be diverse as concerns the 
 meaning' (see Grimm, ii. p. 76 foil.). The stem of 
 manus is difierent, though it is probably connected 
 with the Sanskrit pani, the hand, and with the Greek 
 fjbrjvva), monstror 
 
 Buttmann is also our authority for assigning to 
 both heLKVVfii and Se')(Ofj,ac a root BeK, with the com- 
 mon notion of stretching out the hand (Se^la) for 
 the purpose either of pointing or of receiving. 
 
 As Sa/cryXo? is referable to this common root Se/c, 
 and the Latin digitus is also plainly of kindred origin, 
 — the idea of seizing and holding being involved in 
 
TO TEOTJGBT. 21 
 
 tlie root witli tlie idea of pointing and showing, — it 
 appears too that our English word finger may be 
 traced through a parallel course. "Finger," says 
 Mr. Graham in his Book about Words, " is connected 
 with the German fang en, * to take hold of,' and is a 
 relation of our word fang, i.e. the tooth with which 
 certain animals hold their prey. In the Germanic 
 view of the word it is the instrument with which 
 we take hold. On the other hand, the Greek 8a«- 
 Ti;Xo9, from heUvvfjuL, I show or point out, appeared 
 in Latin as digitus : passed into Italian as dito, and 
 into French as doigt. The Romance view of the 
 word would then be a pointer or indicator, and the 
 Teutonic a holder or catcher." This is given by Mr. 
 Graham, as an example, under the heading of " dif- 
 ferent views of the same idea." It will be seen, how- 
 ever, that there is here no such difference: — the 
 succession of ideas being the same in the history of 
 haicTvXo<i, digitus, and finger : the ideas, viz. of 
 seizing, holding, showing. The Semitic languages 
 present precisely the same succession of ideas in this 
 word. In Hebrew yS^il? ^ finger, and V^l!^ a ra- 
 pacious animal, are both derived from a root which 
 signifies to seize, to ratine ; a modified form of which 
 (viz. D5V) signifies also to reach out to. See, under 
 these words, Gesenius, who, however, refers the 
 Hebrew word for finger to a meaning of the root 
 which is perhaps more proximate, viz. to dip : — 
 
22 THE RELATIONS OF LANGUAGE 
 
 " the forefinger," lie says, " being so called from its 
 being dipped into things," — but which appears to 
 me neither so suitable nor so warrantable as that to 
 which I have here referred the word. It is, by the 
 way, a somewhat remarkable coincidence, that, in 
 Hebrew, one and the same root should be made use 
 of for the denotation ai fingers and rapacious animals, 
 and in English one and the same root be used to de- 
 note both fingers andjangs ! 
 
 Note 2. — According to the celebrated Theory of 
 Yision which owes its origin to Bishop Berkeley, the 
 eye cannot see distance. This seems to be placed 
 beyond all doubt ; but it still appears to be in dis- 
 pute whether or not distance can be perceived by the 
 eye. This distinction, between the sense of sight 
 and perception by the eye, is admirably indicated 
 by Mr. J. S. Mill in his criticism of Mr. Samuel 
 Bailey's " Review of Berkeley's Theory of Yision " 
 ("Westminster Eeview," Oct. 1842). Prof. Bain, 
 however, still asserts {Mental and Moral Science, 
 Bk, ii. Ch. vii. 5), that " the meaning, or import, of 
 Distance, is something beyond the experience of the 
 eye ;" and again ( The Senses and the Intellect, p. 
 367) that " distance cannot be perceived by the eye, 
 because the idea of distance, by its very nature, im- 
 plies feelings and measurements out of the eye — the 
 locomotive and other moving members." These asser- 
 
TO THOUGET. 23 
 
 tions, however, are based upon a definition of distance 
 which involves some sort of bodily locomotion. It is 
 added : — " If our notion of distance did not reveal 
 to us the fact that by so many steps, or by a certain 
 swing of the arm or bend of the body, we should 
 make a definite change in the appearance of the ob- 
 ject, it would not be a notion of distance. There 
 might be an ocular eflfect, but not a revelation of 
 distance. Granted that the eye is very distinctly 
 affected by every change in the remoteness of a 
 visible object from six inches to a mile, that it re- 
 cognizes a variation of impression aU through this 
 interval, this woiJd not answer the question — How 
 far is the object removed at each step." 
 
 To me it appears that this is not the question. 
 The main question is whether any notion of external 
 existence can be derived through the eye. And 
 since Prof. Bain places the distinction between the 
 subjective and objective in the contrast between 
 passive sensation and muscular exertion, and since 
 his special purpose here is to connect the perception 
 of distance with muscular activity, he has all that 
 he wants in the conscious variations of adjustment 
 of the eye for various ocular perceptions. We must 
 be careful not to confuse the perception of distance 
 with its measurement. For purposes of measurement 
 such bodily movements obviously suggested them- 
 selves, and their use is sujQ&ciently indicated in such 
 
24 THE RELATIONS OF LANGUAGE 
 
 names of lineal measure as foot, pace, mile (or thou- 
 sand paces), 5/;aw, etc. Now, the locomotive mea- 
 surement will certainly confirm the ocular percep- 
 tion ; but it seems to me that our idea of distance 
 may be traced solely to our consciousness of the 
 eye's various adjustments for various perceptions. 
 Anyhow, if, as Professor Bain admits, these adjust- 
 ments are made by muscular exertion of which we 
 are conscious, what, I would ask, is this fact of con- 
 sciousness if it be not an idea of distance ? 
 
 The word remoteness suggests locomotion, and so 
 does the word far, which denotes simply going. 
 How far is it ? means How much going will bring 
 me to it? And therefore Professor Bain may 
 well say that the eye's varied afiection by various 
 changes of distance supplies no answer to the ques- 
 tion " How far ?" But, as before observed, this is 
 a question of the measurement of distance, not of 
 its perception : — it being admitted that for purposes 
 of measurement locomotive efibrt has been found 
 the most convenient, perhaps the only practical, 
 means. 
 
 In justice to Mr. Bain, I quote here from Tlie 
 Senses and the Intellect as follows : — "The muscular 
 sensibility of the dead strain, or of Eesistance, can 
 scarcely occur in the eye, there being nothing to 
 resist its movements but its own inertia. What is 
 called straining the eye (which happens ia close and 
 
TO TEOTTGET. 25 
 
 mimite vision) is not the same thing as straining the 
 arms in the support of a heavy weight. Hence of 
 the three primary sensibilities of muscle — Resistance, 
 Continuance, and Speed — two only belong to the 
 ocular muscles. Accordingly the eye, with all its 
 superiority in giving the mind the pictorial array of 
 the extended world, cannot be said to include the 
 fundamental consciousness of the object universe, the 
 sense of Eesistance. There is a certain kindred 
 sensibility in the common fact of muscular tension ; 
 but it is by association, and not by intrinsic suscep- 
 tibility, that the power of vision impresses us so 
 strongly with the feeling of the Object world." 
 
 Nevertheless, there seems to me no sufficient 
 reason for thus dissociating the muscular action of 
 the eye from the consciousness that is associated 
 with other musular exertion. Besides the eye's own 
 inertia, there is the determination and limitation of 
 its muscular activity by the configuration of the ob- 
 ject. This, though more subtle, is yet closely akin 
 to Resistance, and seems to me to be equivalent 
 to grosser resistance for the production of objective 
 consciousness. 
 
26 THE RELATION'S OF LANGUAGE 
 
 CHAPTER III. 
 
 THE SUBJECTIVE, OR NOMINAL, ELEMENT OF 
 LANGUAGE. 
 
 "En analysant les langues les plus anciennes, on 
 Toit peu k peu s'effacer les limites des categories gram- 
 maticales, et I'on arrive a une racine fondamentale 
 qui n'est ni verbe, ni adjectif, ni substantif, mais qui 
 est susceptible de revetir ces diflferentes formes. Est- 
 ce k dire que dans I'etat primitif il n'y eut aucune 
 division des parties du discours ? Non certes. La 
 racine indivise, reunissant en puissance les roles 
 divers que les progres de la reflexion ont depuis 
 separes, n'a jamais existe a I'etat abstrait. L'idee 
 s'est exprimee tout d'abord avec son cortege de de- 
 terminatifs et dans une parfaite unite." — Eenan, 
 Langues Scm., Liv. i. Chap. 3. 
 
 There are two ultimate elements disclosed by the 
 analysis of language, viz. nominals and demonstra- 
 tives. We are here concerned with the former. We 
 may say briefly that the nominal is the main element 
 of nouns both substantive and adjective, and verbs. 
 These have all a demonstrative element in their 
 composition. Demonstratives, on the other hand, 
 
TO TEOUGET. 27 
 
 have no admixture of the nominal element. Of 
 purely demonstrative origin are pronouns (personal, 
 relative, and interrogative), conjunctions, preposi- 
 tions, and most adverbs. 
 
 These results may be startling to those who are 
 not familiar with the processes by which they are 
 attained. There seems at first sight an impene- 
 trable veil of mystery overhanging the array and 
 mutual relations of verb, substantive, adjective, pro- 
 noun, preposition, etc. It is with diflB.culty that we 
 are brought to believe that there is no essential 
 difference between nouns, whether substantive or 
 adjective, and verbs. ^ It is equally difficult to see 
 how one and the same element can underKe all the 
 other parts of speech. 
 
 It is doubtful whether a logical analysis would 
 ever have arrived at these results independently of 
 the material or philological analysis. Perhaps not, 
 since we find such a profound thinker as Mr. J. S. 
 Mill building up a system of logic upon the suppo- 
 sition that words are, not what philology shews their 
 ultimate elements to be, the names of ideas, but the 
 names of things. This view is undoubtedly justifi- 
 able. In dealing with things as they are, it is some- 
 times mischievously irrelevant to consider what they 
 
 1 For a very lucid exposition of this part of our subject, the reader 
 is referred to Mr. James Mill's Analysis of the Fhemmena of the 
 Human Mind, Chap. ir. 
 
28 TEE RELATIONS OF LANGUAGE 
 
 have sprung from. Logic has to deal with language 
 as the perfected instrument, rather than as the new- 
 born product of thought. In like manner, a system 
 of morals need not recognize the ultimate analysis 
 of human motives. Our voluntary actions have been 
 traced to an elementary stage wherein a spontaneous 
 discharge of energy has been determined by acci- 
 dental circumstances. We may admit this, and yet 
 not take it into consideration in our dealings with 
 men as men. I shall, however, endeavour to shew 
 that words — i.e. such of them as are names — were 
 originally names of ideas ; and, agreeing with M. 
 Renan, that they were immediately transferred to 
 things : — so immediately as to leave no space for 
 their historical existence as names of ideas — I hope 
 to shew also the importance of recognizing that ex- 
 istence as an elementary fact. 
 
 TVe must first observe the grand distinction be- 
 tween nominals and demonstratives. "There are," 
 says Bopp,^ speaking more particularly of the Indo- 
 European or Aryan family of languages, " there are 
 two classes of roots : from the one, which is by far 
 the most numerous, spring verbs, and nouns (sub- 
 stantive and adjective) which stand in fraternal con- 
 nection with the verbs, not in the relation of descent 
 from them, not begotten by them, but sprung from 
 the same shoot with them .... From the second 
 1 Comp. Grammar, Vol. i. p. 96. 
 
TO THOJJGET. 29 
 
 class spring pronouns, all original prepositions, con- 
 junctions, and particles. We name them prono- 
 minal roots, because they all express a pronominal 
 idea, which, in the prepositions, conjunctions, and 
 particles, lies more or less concealed. JS^o simple 
 pronouns can be carried back, either according to 
 their meaning or their form, to anything more 
 general, but their declension-theme (or inflective 
 base) is at the same time their root." 
 
 The next thing to be observed is that the nominal 
 roots are aU alike predicative-. " Analyse any word 
 you Hke," says Prof. Max MuUer,^ "and you will 
 find that it expresses a general idea peculiar to the 
 individual to which the name belongs. "What is the 
 meaning of moon ? — the measurer. What is the 
 meaning of sun ? — the begetter. What is the mean- 
 ing of earth ? — the ploughed. The old name given 
 to animals, such as cows and sheep, was pam, the 
 Latin j)ecu8, which n^^dccL's, feeders,.''^ 
 
 The same author goes on to say, " The fact that 
 every word is originally a predicate : — that names, 
 though signs of individual conceptions, are all, with- 
 out exception, derived from general ideas, is one of 
 
 • Science of Language, Yol. i. p. 382. In the first sentence of the 
 passage quoted above, 1 find a difficulty in the words " a general idea 
 peculiar to the individual to which the name belongs." This sounds 
 like a contradiction in terms. It must, I think, be intended that, even 
 in what are called proper names, as .sun, moon, etc., we find only 
 general ideas appropriated to individuals. This is true of all proper 
 names, as is shown above. 
 
30 THE RELATIONS OF LANGUAGE 
 
 the most important discoveries in the science of 
 language." 
 
 It is certain that if any objects could be imme- 
 diately named, we might expect the sun and moon 
 to have names of their own. But we find they have 
 not. We might expect prominent mountains and 
 large rivers to have names immediately appropriated 
 to them. But the analysis of the names of all such 
 conspicuous objects discloses roots that express general 
 ideas. Such a theory as that of Adam Smith is thus 
 shewn to be groundless. He says : — " The assigna- 
 tion of particular names to denote particular objects, 
 that is, the institution of nouns substantive, would 
 probably be one of the first steps towards the forma- 
 tion of language. Two savages who had never 
 been taught to speak, but had been bred up remote 
 from the societies of men, would naturally begin to 
 form that language by which they would endeavour 
 to make their mutual wants intelligible to each 
 other by uttering certain sounds whenever they 
 meant to denote certain objects. Those objects 
 only which were most familiar to them, and which 
 they had most frequent occasion to mention, would 
 have particular names assigned to them. The par- 
 ticular cave whose covering sheltered them from the 
 weather : the particular tree whose fruit relieved 
 their hunger : the particular fountain whose water 
 allayed their thirst, would first be denominated by 
 
TO THOUGHT. 31 
 
 tlie "words cave, tree, fountain, or by wTiatever other 
 appellations they might think proper, in that primi- 
 tive jargon, to mark them. Afterwards, when the 
 more enlarged experience of these savages had led 
 them to observe, and their necessary occasions ob- 
 liged them to make mention of, other caves, and 
 other trees, and other fountains, they would naturally 
 bestow upon each of those new objects the same 
 name by which they had been accustomed to express 
 the similar object they were first acquainted with. 
 The new objects had none of them any name of its 
 own, but each of them exactly resembled another 
 object which had such an appellation. It was im- 
 possible that those savages could behold the new 
 objects without recollecting the old ones, and the 
 name of the old ones, to which the new bore so close 
 a resemblance. When they had occasion, therefore, 
 to mention or to point out to each other any of the 
 new objects, they would naturally utter the name of 
 the correspondent old one, of which the idea could 
 not fail at that instant to present itself to their 
 memory in the strongest and liveliest manner. And 
 thus those words, which were originally the proper 
 names of individuals, became the common name of a 
 multitude." 
 
 To all this Professor MuUer well replies by an 
 analysis of the names of cave and river. He says : — 
 "A cave in Latin is called antrum, cavea, spelunca. 
 
■ / 
 
 32 TEE RELATIONS OF LANGUAGE 
 
 Now antrum means really the same as internum. 
 Antar in Sanscrit means hetiveen and within. Antrum, 
 therefore, meant originally what is within the earth 
 or anything else. It is clear, therefore, that such a 
 name could not have been given to any individual 
 cave, unless the general idea of .being within, or in- 
 wardness, had been present in the mind. This 
 general idea once formed, and once expressed by the 
 pronominal root an or antar, the process of naming 
 is clear and intelligible. The place where the savage 
 could live safe from rain and from the sudden attacks 
 of wild beasts, a natural hollow in the rock, he would 
 call his within, his antrum; and afterwards similar 
 places, whether dug in the earth or cut in a tree, 
 would be designated by the same name. Let us 
 take another word for cave, which is cavea or caverna. 
 Here, again, Adam Smith would be perfectly right 
 in maintaining that this name, when first given, 
 was applied to one particular cave, and was after- 
 wards extended to other caves. But Leibnitz would 
 be equally right in maintaining that in order to call 
 even the first hollow cavea, it was necessary that the 
 general idea of hollow should have been formed in 
 the mind, and should have received its vocal ex- 
 pression cav." 
 
 The only thing I object to here is the encourage- 
 ment that seems to be given to the opinion that all 
 things named took their names from the first thing 
 
TO TB OUGHT. 33 
 
 named. It is perhaps only inadvertently that Prof. 
 Miiller says, in a sentence immediately preceding 
 those quoted above, " Adam Smith is no doubt right, 
 when he says that the first individual cave which is 
 called cave gave the name to all other caves." As 
 well might it be said, that the first man who records 
 his vote at an election gives the cue to the other 
 voters. The one element of the actual word cave 
 being a predicative root (therefore general) and the 
 other element demonstrative (therefore particular), it 
 is but a truism to say that the first application of the 
 actual word was to a particidar object : but so also 
 was the second, and the third, and so on ad in- 
 finitum. And again : so far was it from being the 
 case that the name first given to an object was 
 suggested for all other similar objects, that (as is 
 elsewhere stated) in the earlier stages of language, 
 we find the same object denoted by a greater variety 
 of names : — that is to say, that of one and the same 
 object more than one idea was predicated. 
 
 Professor Miiller thus proceeds : — " It is the same 
 with all nouns. They all express originally one out 
 of the many attributes of a thing, and that attribute, 
 whether it be a quality or an action, is necessarily 
 a general idea. The word thus formed was in the 
 first instance intended for one object only, though 
 of course it was almost immediately extended to the 
 whole class to which this object seemed to belong." 
 
34 THE RELATIONS OF LANGUAGE 
 
 But it cannot be that the word is " intended for 
 one object only " when one of its elements is the 
 name of a general idea, and the other a demonstra- 
 tive, and therefore of even more general application. 
 It seems that the most complete view of an actual 
 word is that in which it is regarded as a predicative 
 root determined by a demonstrative, and therefore as 
 a complete predication. Whatever there is in an actual 
 word that does not belong to the predicative root 
 may be traced, more or less directly, to the demon- 
 strative element. A word so considered is virtually 
 a proposition, in which the idea is predicated of the 
 perception. I do not now call a cave a cave, or a 
 spade a spade, because, eliminating differences, I have 
 classed all my agreeing perceptions under the name 
 of the first perception as a prototype ; but I give 
 to the last, as to the first, the name which itself 
 suggests. That is to say, I predicate of the last (if 
 I can, and only because I can) that which I predi- 
 cated of the first :— not because I so predicated of 
 the first, but because, in so far as suggestive of the 
 same idea, the last perception has a claim to the 
 name quite independent of the naming of the first. 
 This is here stated as the law by which things were 
 first named, not of course as governing the process 
 by which the names of the most important ideas 
 suggested by particular objects became convention- 
 ally'^appropriated to the denotation of those objects. 
 
TO TEOJJGKT 35 
 
 Writers on the origin of language have expatiated 
 on the richness and exuberance of the speech of the 
 men of early days. They have told us of a thousand 
 names for the sun, the sea, the lion, the camel, etc.^ 
 What does all this prove, but that in those early times 
 the things themselves had no names : that the so-called 
 names which exist in such astonishing abundance 
 are only as many predications of the nameless object 
 indicated. Afterwards, no doubt, some one or other 
 of these names has been attracted from the subject 
 and has adhered to the object. Then it has served 
 first to denote the object, and, subordinately, to con- 
 note the ideas of all those other predications. 
 
 Accounting thus for the multiplicity of names of 
 one object, it occurs here to mention the fact of 
 several and dissimilar and sometimes incongruous 
 things being called by the same name. The reason 
 is to be found in some one idea to which the several 
 phenomena are in common referable. Subsequently 
 
 ^ " It is said that in Arabic there are 500 names for the lion, 200 
 for the serpent, more than 80 for honey, 400 for sorrow, and (what is 
 quite incredible, unless every periphrasis be counted a name) no less 
 than 1000 for a sword. M. de Hammer, an unimpeachable authority, 
 has, ia a little treatise on the subject, counted also .5744 words relating 
 to the camel" (Farrar, Origin of lymgutgi:). Mr. Farrar attributes 
 this to the extraordinary prevalence of onomatopoeia in the Semitic 
 languages. But this kind of nomenclature is based upon the oDsei- 
 vatioiis of only one sense. The extraordinary exuberance of ancient 
 speech is better accounted for by supposing things to be cognized in 
 various ways and degrees by all the senses, and predication made ac- 
 conlingly. At the same time, 1 suspect }&.. de Hammer of " throwing 
 the hatcliet." 
 
36 TEE RELATIONS OF LANGUAGE 
 
 this one idea may become more or less completely 
 paramount in one class of phenomena, and in another 
 class may be more or less completely countervailed 
 by other ideas. Thus, if we take the word '^proha- 
 tion " first in the sense in which we are now accus- 
 tomed to use it, viz. as equivalent to the word trial, 
 and then in the sense of proof, in which sense it is 
 found in Shakespeare, we shall find a common idea, 
 the recognition of which will at least enlighten us as 
 to the nature of ^' proof ." The Hebrew word harak 
 means to " bless ;" but occasionally it means also to 
 " curse." At least, it may be said that occasionally 
 it would admit of this rendering. What, it has 
 been asked, is the use of a language the meaning of 
 whose words is so vague? The fact, however, is 
 that usage defines the signification. The Hebrew 
 word in question means primarily to hieel, then to 
 kneel for invocation, which may be either of a bless- 
 ing or of a curse. If the significations had not been 
 arbitrarily settled, there might be precisely the same 
 sort of vagueness in the use of our own words invocor 
 tion and imprecation, prejudice and prepossession, wis- 
 dom and sophistry, and many others which are now 
 recognized as being wide as the poles asunder. 
 
TO THOUGHT. 37 
 
 CHAPTEE IV. 
 
 THE OBJECTIVE OR DEMOJ^ STRATI VE ELEMENT OF 
 LANGUAGE. 
 
 The complete exposition of this part of our 
 analysis would present it under these four heads : — 
 
 (1) Pronouns, so-called, whether demonstrative, 
 personal, possessive, relative, or interrogative. 
 
 (2) Articles. 
 
 (3) Adverbs, prepositions, conjunctions, and in- 
 terjections. 
 
 (4) The complement of the nominal element in 
 verbs and nouns. 
 
 Under the first head we must here, for brevity's 
 sake, assume that which is no longer a matter of 
 doubt amongst philologists, viz. that relative and in- 
 terrogative pronouns are only modifications, in form 
 and meaning, of the simple demonstrative. It is 
 true that Bopp, the prince of philologists, seems to 
 care very little for the reduction of relative and in- 
 terrogative to demonstrative bases, and, indeed, refers 
 the Latin hie, hcec, hoc to a Sanskrit relative base. 
 He -ridicules also the idea of the Greek relative being 
 
38 THE RELA TIONS OF LANG UA GE 
 
 a modification of tlie definite article. I think, how- 
 ever, that the opinion of more modern philologists 
 would go rather with Donaldson, who asserts that 
 here " Bopp began at the wrong end," and adds 
 that " the relative, like all subordinate meanings of 
 pronouns, springs from its original demonstrative 
 force." 
 
 I will only here remark upon the increased facility- 
 arising to the analysis of thought from the pre- 
 liminary analysis of language. Without this ad- 
 vantage one might easily imagine — and I believe it 
 is usually imagined — that there is a more thorough 
 difference between demonstrative and relative pro- 
 nouns than between a demonstrative pronoun and a 
 noun. There is a logical subtlety, that passes almost 
 into a mystery, in the relative use of the pronoun. 
 
 The analysis of the personal pronouns discloses 
 no fundamental differences between them. I and 
 thou and he are only conventional distinctions be- 
 tween this and thi8 and this. The adjectival forms 
 called possessive pronouns require no consideration 
 here. 
 
 (2) Articles. — If the actual word (noun or verb) 
 does always contain a demonstrative element, it 
 would appear that the prefixed demonstrative called 
 an article is a superfluity. I think we may affirm 
 that it could hardly have been called into requisition 
 80 Ions: as the demonstrative element of the actual 
 
TO THOUGHT. 39 
 
 word was understood and consciously retained. For 
 example, when the Romans said vivat rex, they made 
 a conscious expression of the demonstrative in both 
 words. From both of these words, in modern 
 French, the demonstrative element has fallen away. 
 Its place has been supplied by the article. We 
 find now tive le roi. I conclude that the reason 
 why the Latin language could dispense so com- 
 pletely with the article, and to so great an extent 
 with the personal pronoun also, is to be found in 
 the supposition that the demonstrative element of 
 its nouns and verbs still retained its significance so 
 long as the language was spoken. 
 
 Modern European languages have made a dis- 
 tinction between the definite and the indefinite 
 article ; expressing the latter by some modification 
 of the demonstrative base that has supplied the 
 name of the first numeral. Now the name of the 
 first numeral is in the first place an intense demon- 
 strative. " One " is " this," so intensely marked 
 as to distinguish the one thing from all others. But, 
 from the very fact of this distinction, it had re- 
 ference to all others. Its meaning was relative to 
 all other demonstrable things, because it excluded 
 them all. Then it assumed a relation to other 
 things as implying them ; and so further it became 
 the name of the first numeral as the initial term of 
 an indefinite series. And so we account for the 
 
40 THE RELATIONS OF LANGUAGE 
 
 fact that things supposed to be one are not spoken 
 of as one. We do not speak of a sun or 0)ie sun, 
 nor of a God or one God, — at least, not as expressing 
 our belief in their real existence. But we speak of 
 the sun ; and if we do not also speak of the God, as 
 the Greeks did, it is only because the word God has 
 come to be regarded as a proper name. If we were 
 to speak of one sun, or a sun, or of one God, or a 
 God, we might be supposed to imply a belief in a 
 plurality of such existences. 
 
 For the most part, however, no such relation is 
 consciously intended in the use of the indefinite 
 article ; but it may be referred to that necessity 
 which seems to have existed from the beginning of 
 language, viz., that a demonstrative should enter 
 into combination with the subjective element for the 
 formation of a name. Thus avOpcoiro^ and homo are 
 synthetical forms to which corresponds, in English, 
 not the word " man," but the analytical form " a 
 man." For in Greek, the demonstrative element of 
 the noun had not so completely lost its significance 
 as not to have the force of our indefinite article ; 
 though it seems not to have availed for definite de- 
 notation. Yet even for the former purpose we find 
 the demonstrative rt? usually employed where we 
 should employ a or an. 
 
 We have seen already that the distinction of the 
 noun adjective from the noun substantive is only a 
 
TO THOUGHT. 41 
 
 syntactical contrivance. The function of the noun 
 substantive being chiefly denotative (however exten- 
 sive its connotation may be) — that of the noun ad- 
 jective is attributive, or predicative. We here ob- 
 serve further that the noun substantive is preceded 
 by the article, which is wanting to the attributive. 
 A word ordinarily used as an attributive may indeed 
 have the definite article prefixed, but it then passes 
 over to the denotative side, along with nouns sub- 
 stantive. As far as the Greek language is concerned 
 " there is no adjective or even participle," says Dr. 
 Donaldson,^ " which may not become a substantive 
 if it only has the definite article prefixed : if, in a " 
 word, it has that accompaniment which is necessary 
 for the conversion of a substantive, as the name of a 
 quality or attribute, into the name of a particular 
 thing." 
 
 It is usual to treat of the article as subservient to 
 the discharge of the functions of the noun. I would 
 reverse this consideration, and would represent the 
 noun substantive as serving the purposes of denota- 
 tion, and as doing this the more efiectually in pro- 
 portion to the increase of its connotation. It is thus 
 seen vice versa, as a satellite of the article. The at- 
 tributive, on the other hand, has not been so at- 
 tracted from its own proper sphere. It has not 
 fallen away to a subservient office as the name of a 
 
 » New Cratylus, § 294. 
 
42 THE RELATIONS OF LANGUAGE 
 
 thing, but occupies its own proper domain as tlie 
 name of an idea. Wc do not find it in the retinue 
 of a demonstrative. In other words — caret ple^ 
 rumque articulo nomen predicativum. 
 
 Now if, according to this rule of Greek Grammar, 
 the predicate usually lacks the article, it is because 
 the predicate is of the subjective idea, which cannot 
 have an objective denotation. On the other hand, 
 the subject (so called) is always an objective pheno- 
 menon, to the denotation of which the objective or 
 demonstrative element of speech is essential. This 
 is the ordinary case of the proposition which refers 
 things to ideas. Let it be observed further that 
 when things are predicated of things, or rather 
 names of names, we have what are called convertible 
 or reciprocating propositions ; and in this case, in 
 Greek, the article is either (and for the most part) 
 prefixed to both subject and predicate, or else is 
 wanting to both. See, upon this point, Middleton 
 On the Greek Article, Part. i. Chap. iii. It seems 
 plain that it can never occur to predicate one, thing 
 of another — one phenomenon of another. "We can 
 never say " This (one thing) is this (another thing)." 
 And therefore a proposition in which the article is 
 prefixed to both subject and predicate must give in- 
 formation not of things, nor yet of ideas, but only 
 of names. It is not giving a name to a thing, 
 because this would be to refer the thing to an idea, 
 
TO THOUGHT. 43 
 
 and the predicate (the name of the idea) would in 
 this case lack the article. But it is the denoting 
 of one name by another name ; and therefore, if one 
 of these names appears in combination with the de- 
 monstrative element, so will the other : — if one lacks 
 the demonstrative, so will the other. 
 
 (3) Frejyositions, Conjunctions, Adverbs, Inter- 
 jections. — Prepositions and conjunctions express the 
 relations of things to things. Therefore if things 
 are capable only of demonstrative denotation, their 
 mutual relations cannot be otherwise denoted. True 
 prepositions and conjunctions are found to be of de- 
 monstrative origin. 
 
 Adverbs are, in the first place, simply demon- 
 strative — such as so, tJien, thus. But it is often 
 convenient to express the manner or quality of 
 actions by a reference to ideas ; and so we get such 
 adverbs as truly, softly, perforce, etc. 
 
 In Interjections we seem to find the most rudi- 
 mentary stage of demonstrative speech. By many 
 this remark would be made to include all speech. To 
 me it seems that philology owes nothing to what is 
 called the Interjectional Theory of Language, except 
 a considerable addition to that error through which 
 she has waded in search of truth. This theory 
 accounts well enough for the vocal method of de- 
 noting things ; but it has failed to give, as it can 
 hardly be expected to give, an account of the no- 
 menclature of ideas. 
 
44 THE RELATIONS OF LANGUAGE 
 
 (4) Tlte complement of the nominal element in Verbs 
 and Nouns. — Under this head it will be sufficient to 
 say, in addition to what has been adduced here and 
 in the preceding chapter, that it is admitted by 
 all, that nominal roots are never found alone and 
 naked, as actual words. The other constituent 
 of an actual word is a demonstrative. This consti- 
 tution is requisite for the functions of such words 
 as are now names of things. It was only by 
 a combination of indication and predication that 
 names were ever given to things objective ; al- 
 though now, in most cases, the consciousness of the 
 predicative element in such names is whoUy lost. In 
 must be remembered, however, that when the name 
 of a thing is used as a predicate, it is in virtue of 
 its predicative element, to the exclusion of its de- 
 notation. 
 
TO TEOTIGHT. 45 
 
 CHAPTEE Y. 
 
 ON PREDICATION. 
 
 The "words that are formed from nominal roots 
 (viz. nouns, substantive and adjective, and verbs) as 
 distinguished from the pure demonstratives (viz. 
 articles, pronouns, prepositions, conjunctions, and 
 some adverbs), seem to be really propositions. For 
 in all cases the reference of the nominal root is de- 
 termined by the demonstrative element of the actual 
 word. "We find no naked nominal roots. Actual 
 names have all a demonstrative element. A word, 
 then, is a proposition in which the idea of the 
 nominal root is predicated of the demonstrative ele- 
 ment. This is, however, a purely intuitive predica- 
 tion. For example, whatever may be the import 
 of the nominal root of the word " tree" he who 
 should use this word would express that a certain 
 object referred itself to a certain idea. And if he 
 uses the same word in reference to certain other ob- 
 jects, it is because these other objects refer them- 
 selves to the same idea. It is easy to see how such 
 \miform predications concerning things would at 
 once become names of the thing's themselves. 
 
46 TUE RELATIONS OF LANGUAGE 
 
 It must not be supposed that, when we speak of 
 intuitive predications, we mean such as correspond 
 with the realities of things. Intuitive predications 
 correspond with intuitions, and, as these may be 
 fallacious, the predications founded upon them may- 
 be false. But their falsity is relative not to the in- 
 tuitions upon which they are founded, but to those 
 that take place subsequently. An object dimly seen 
 may be referred to the idea which underlies the 
 word " tree." Seen more closely, the object is 
 found to be a house or a windmill. Yet the word 
 " tree " applied to the first phenomenon is as cor- 
 rectly applied as are the words " house " or " wind- 
 mill " to the others. So, again, the sun might very 
 well be referred to an idea of motion, and it is sup- 
 posable that such an idea might furnish a name for 
 the sun. A word so formed would be a true predi- 
 cation of the phenomenon as presented to those who 
 should so predicate of it, though a very different 
 phenomenon is presented to those who afterwards 
 observe and compare more carefully. 
 
 One use of an isolated word is, therefore, to make 
 others conceive what we conceive. But " names are 
 intended not only to make the hearer conceive what 
 we conceive, but also to inform him what we be- 
 Keve." (Mr. J. S. Mill, System of Logic, Bk. i. Ch. 
 ii.) Mr. Mill has previously said that "in every 
 act of belief two objects are in some manner taken 
 
TO THOTTGET. 47 
 
 cognizance of ; that there can be no belief claimed 
 or question propounded which, does not embrace two 
 distinct (either material or intellectual) subjects of 
 thought, each of them capable or not of being con- 
 ceived by itself, but incapable of being believed by 
 itself. I may say, for instance, 'the sun.' The 
 word has a meaning, and suggests that meaning to 
 the mind of anyone who is listening to me. But 
 suppose I ask him whether it is true : whether he 
 believes it ? He can give no answer. There is as 
 yet nothing to believe or to disbelieve." 
 
 This is true in reference to the present state of 
 language. But it was plainly not so when these 
 words were first framed and first used. Then the 
 word " sun " expressed an assertion concerning a 
 certain object : an assertion which may or may not 
 be acquiesced in by another. Suppose two of these 
 first framers or users of language to come for the 
 first time upon one of the great fresh-water lakes of 
 America. Suppose one of them to apply the word 
 salt to this object — a word which has been so com- 
 monly applied to another similar object, the sea, as 
 to have now become one of its names. The other 
 replies by using the word " sweet " with reference to 
 this present object. The first one doubts, and is 
 convinced only by tasting the water for himself. 
 So long as words possessed, as we find they origin- 
 ally did possess, a predicative meaning, in addition to 
 
48 THE RELA TTONS OF LA NG UA GE 
 
 their denotation, so long did they express, even when 
 isolated, something that might be believed or disbe- 
 lieved. When this meaning is merged in the deno- 
 tation, then it may require the help of another word 
 to express that which itself implies. Thus, if I say 
 " the sun shines," the proposition is certainly pleo- 
 nastic. The assertion is made of this object, that it 
 shines. But this is already asserted in the word 
 sun} " The sun " is a complete predication, of 
 which the is the subject. For, as has before been 
 observed, external objects were not primarily named. 
 They were only indicated. Afterwards they appro- 
 priated certain predications as their own names, 
 which thereupon lost their predicative meaning. 
 Thus, "the sun" no longer means "this shines," 
 but, having become in the first place merely the 
 denotative name of a certain object, has afterwards 
 been so completely transmuted as that it may now 
 be truly said of it that it offers nothing to belief or 
 disbelief. Therefore, in utter unconsciousness of what 
 we have already asserted in saying " the sun," we 
 proceed to the pleonasm (admitted to be necessary) 
 " the sun shines." ISTor even here do we stop ; for 
 if anyone should say, " The sun-shine is splendid," 
 it is only a repetition of " shine, shine, shine," — re- 
 calling the methods of barbaric speech. 
 
 1 See Bopp, Glossarium Comparativum, sub. t. 
 
TO TE OUGHT. 49 
 
 After all, wherein lies the difference between a 
 proposition, as it is usually recognized, and a simple 
 name? The name contains within itself, as we 
 have seen, both subject and predicate; as when 
 I say " the sun," I mean that this thing at which 
 I am looking and pointing shines. But, it will be 
 said, you want the coj)ula — the substantive verb — 
 and even in such a bare proposition as " this shines " 
 the copula is impKed : — the amplified form being, 
 "this is shining." Moreover, it will be said, "this" 
 is a pronoun, used jyro nomine, and therefore you 
 must supply the nomen, and say, in full, "this sun is 
 shining." But to this we reply, that this whole 
 theory of the copula is erroneous and mischievous, 
 as has been abundantly shown by Mr. Mill himself. 
 When, let us ask, would the need of a copula be 
 felt, and in what form would it be likely to appear ? 
 In the first place, so long as the predicative element 
 of words was consciousl}^ retained, there would be 
 no need of a copula. But suppose the original pre- 
 dication to a great extent lost in the denotation. 
 Notwithstanding this loss, predications are stiU 
 made of this denotable thing, and thereby its con- 
 notation goes on continually increasing. The word 
 therefore which originally predicated a single idea, 
 now connotes a multitude. Who can number the 
 sand of the sea ? Let him enumerate the ideas 
 connoted by its name. But here we observe that 
 
 4 
 
50 THE RELATIONS OF LANGUAGE 
 
 such connotation of a word overwhelms its special 
 predication. Connotation is subservient to denota- 
 tion ; and therefore a word whose connotation is so 
 increased is found in attendance upon a demonstra- 
 tive. Thus when we speak of " the sun " and " the 
 sea," by the words sun and sea we do but denote 
 more precisely and unmistakably the objects that 
 are vaguely denoted by the demonstrative " the." 
 Now if we wish to predicate of a merely denotable 
 object the bundle of ideas contained in the word 
 " sun " or in the word " sea," what happens ? We 
 find that the denotable object is denoted by a demon- 
 strative, and that the bundle of ideas is also denoted 
 by (or, as we have expressed it, in attendance upon) 
 a demonstrative. The proposition, then, will contain 
 two demonstratives; and it is conceivable that the 
 second demonstrative is the origin of the copula. 
 That the copula may be superseded, even now, by 
 a reduplicated demonstrative is proved by such a 
 usage as that of the Latin ecce. To assert that 
 this thing at which I point is a man, it is not 
 necessary to say Sic est homo, but only Ecce homo} 
 
 1 I quote the following remarks from one of Mr. Findlater's Notes 
 to the New Edition of Mr. James MiLl's Analysis of the Phenomena of 
 the Human Mind, Speaking more particularly of the usages of non- 
 Aryan languages, he says: — " Of the substitutes put in place of the 
 substantive verb, by far the most common are pronouns, and particles 
 indicating position. Thus ia Coptic, the descendant of the ancient 
 Egyptian, the demonstrative pe, ' this,' after a noun singular mascu- 
 line, or te when the noun is feminine, is equivalent to is ; and vie 
 'these,' after a plural, to are. In the ancient hieroglyphic monuments 
 
TO THOUGST. 51 
 
 If tlie import of the nominal element of some 
 words lias been forgotten, and has become merged 
 in the demonstrative element, this latter has also lost 
 its conscious existence in many cases. The nomina- 
 tive case sign has required an article as its exponent, 
 and the obKque cases have called in the help of pre- 
 positions. 
 
 To return to the former case, the merging of the 
 
 the function of the substantive verb is performed by the same means. 
 Even in the Semitic langTiages, ■which have substantive verbs, pronouns 
 are habitually used instead of them ; so that I I, or I he, stands for 
 / am, and we we, or ice they, for we are. ' Thou art my King ' (Ps. 
 44, b) is in the Hebre-w ' Thou he my king.' ' We are the servants of 
 the God of heaven' (Ezra 5, 11) is in Chaldee 'We they servants, 
 etc' ' I am the light of the world,' is in Arabic ' I he the light of the 
 world.' Although such modes of expression are foreign to the Indo- 
 European languages, even they furnish abundant evidence of the pre- 
 dicative power of pronouns and particles." 
 
 The assertion of this last sentence is unfortunate. The purpose 
 is, to set aside the predicative power of the copula. If this pre- 
 dicates anything, it predicates existence. Herein lies its mischief. In 
 order to set this aside, you say that it is only adventitiously that the 
 copula predicates anj-thing : — that it is itself only a demonstrative. 
 Then you subvert the whole of this good doctrine by sanng that the 
 copula, even when so reduced to a demonstrative, is yet predicative. 
 We must hope this is an oversight on the part of Mr. Findlater, who 
 goes on to say : — 
 
 " If any word required to have inherent in it the peculiar affirma- 
 tive power attributed to verbs, it is the word yes. Accordingly Home 
 Tooke derives it from the French imperative a-yez : forgetting, or not 
 knowing, that the Anglo-Saxon gese or yea (cognate with the Sanscrit 
 pronoun yd) was in existence long before the French ayez. The fact 
 is, that Eng. yen, Ger. ja, and the corresponding words in the other 
 European languages, are oblique cases of demonstrative pronouns, and 
 mean simply 'in this (manner),' or 'thus.' The Italian si (yes) is 
 from Ijatin sic (thus) ; the Proven9al oc is from the Latin hoc ; and 
 the modem Fr. oiii was originally a combination of Iwc illo, and passed 
 through the stages of ocil and ciil into its present form." 
 
 " The consideration of these and a multitude of similar phenomena 
 suggests that the Sanscrit as-mi, Gr. cimi, Lat. sum (for es-um), Eng. 
 am, may have had for its root the demonstrative pronoun sa, and meant 
 primarily 'that (or there) as to me.' " 
 
52 THE RELA TIONS OF LANG UA GE 
 
 predicative in the denotative clement of words — 
 that is to say, the transfer of names from ideas to 
 things — we observe that, however such a change 
 may appear subversive of the original scheme of 
 language, its advantages are obvious. First, it sup- 
 plied the means of discoursing concerning objects 
 that could not be immediately indicated — as, for in- 
 stance, of the sun at midnight, or of the sea in an 
 inland country. Secondly, it enabled language to 
 rise above merely intuitive predications. So long 
 as ideas only were named and things only indicated, 
 predications could be made only of individual things. 
 Of each thing you might make hundreds of predi- 
 cations, but the subject of each being no otherwise 
 defined than by actual and gesticulative indication, 
 you might as well make so many predications of 
 as many different things. The demonstrative 
 element could not supply what was wanted, yiz. 
 a nucleus around which continually accumula- 
 ting observations might cluster : a common sub- 
 ject of an indefinite number of predications. I 
 might point with my finger to the sun, and make 
 concerning it as many predications as ever have 
 been or will be made; yet so long as the subject 
 of them all is only this thing to which I point 
 with my finger, the assertion of all my observa- 
 tions will be like water spilt on the earth, that 
 cannot be gathered up again. To give this thing 
 
TO THOUGHT. 53 
 
 a name of its own seems now such, an obvious ex- 
 pedient that we are apt to overlook its importance. 
 And it may appear strange that a name could be 
 furnished to the thing only by a transfer to it of 
 the name of some one or other of the ideas to which 
 it was most impressively referable. This, however, 
 we must for the present overlook in considering 
 the importance of the step by which things objec- 
 tive acquired names of their own, independent of, 
 however derived from, the names of ideas. What 
 I now assert of the sun or of the sea is an asser- 
 tion not merely concerning that to which I point 
 with my finger, for that is only a momentary in- 
 tuition, but concerning all that vast handle of ob- 
 servations that have been made about each of these 
 objects, and recorded in the name of each. 
 
 If now I make an assertion concerning an object, 
 my assertion is not necessarily an intuitive predica- 
 tion. If it were, I should still say that the sun 
 moves round the earth, for so it certainly stiU ap- 
 pears to my mere intuition, just as it did to that of 
 Moses or of David. But the assertions I now make 
 about things, however initiated by my own intuitive 
 observation, are both limited and extended by my 
 belief of many other assertions that have been ini- 
 tiated by other observations, whether of myself or 
 of others. My predications are now not only of that 
 which I conceive, but of that which I believe. 
 
54 TEE RELATIONS OF LANGUAGE 
 
 The jjroccss seems to have been this : — First, ideas 
 only are named ; things objective are indicated by de- 
 monstrative gesticulation and demonstrative speech. 
 Then from the constant predication of some one 
 idea of some one object, the transfer of the name 
 of the idea to the object (as the name of salt to the 
 sea). Then the loss of the special significance of 
 that name ; such loss, however, being infinitely 
 more than compensated by the infinite connotation 
 of which such a name becomes capable. 
 
 Through all this, the import of propositions re- 
 mains the same. A proposition is still the reference 
 of some thing (or bundle of things) to some idea (or 
 bundle of ideas). Because the object is now no 
 longer merely denoted : because by the expedient of 
 naming it we have given to our denotation an inde- 
 finite capability of connotation : because its actual 
 connotation is continually increasing ; — it still re- 
 mains true that every predication we can make of 
 such an object is a referring it to our ideas. It 
 matters not whether our predications are common 
 and trite, or whether they originate with ourselves : 
 whether we say only that the sun is warm and 
 bright, or whether Gralileo and Newton predicate of 
 the sun things which none before them had predi- 
 cated : — it matters not whether our assertions are 
 proclamations or only records of discovery ; they are^ 
 still references of things to ideas. 
 
TO TEOTTGET. 55 
 
 In Mr. Mill's chapter on tlie Import of Propo- 
 sitions (Si/stem of Logic, Bk. i. Ch. v.) we read as 
 follows : — " The notion that what is of primary im- 
 portance to the logician in a proposition, is the re- 
 lation between the two ideas corresponding to the 
 subject and predicate (instead of the relation be- 
 tween two phenomena which they respectively ex- 
 press) seems to me one of the most fatal errors ever 
 introduced into the philosophy of Logic ; and the 
 principal cause why the theory of the science has 
 made such inconsiderable progress during the last 
 two centuries. The treatises on Logic, and on the 
 branches of Mental Philosophy connected with 
 Logic, which have been produced since the intru- 
 sion of this cardinal error, though sometimes written 
 by men of extraordinary abilities and attainments, 
 almost always tacitly imply a theory that the in- 
 vestigation of truth consists in contemplating and 
 handling our ideas or conceptions of things : a doc- 
 trine tantamount to the assertion, that the only 
 mode of acquiring knowledge of nature is to study 
 it at second hand, as represented in our own mind." 
 
 I must confess that I cannot see how a know- 
 ledge of nature can be other than a representation 
 of it in our minds. Nature is supposed to be one 
 thing : our mind another. A knowledge of nature 
 must involve both as elements. Mr. Mill proceeds 
 thus : — " Meanwhile, inquiries into every kind of 
 
5Q THE RELATIONS OF LANGUAGE 
 
 natural phenomena were incessantly establishing 
 great and fruitful truths on most important subjects, 
 by processes upon which these views of the nature 
 of judgment and Reasoning threw no light, and in 
 which they afforded no assistance whatever. No 
 wonder that they who knew by practical experience 
 how truths are arrived at, should deem a science 
 futile which consisted chiefly of such specula- 
 tions." 
 
 But, if it is a fatal error to suppose that a propo- 
 sition expresses only the relation between two ideas, 
 it seems to me equally fatal to suppose that it is 
 concerned only with things or objects. If conscious- 
 ness cannot be restricted to either subjectivity or 
 objectivity : if it is lighted up only by the antithesis 
 of these its correlated elements, it would seem that 
 the announcements and records of consciousness — 
 that is to say, projwsitions — must express such cor- 
 relation. What, let us ask, has been the result of 
 those " inquiries into every kind of natural phe- 
 nomena " of which our author here speaks ? The 
 process may have required the comparison of things 
 with things, but the things must surely be in our 
 own cognition before they can be brought into com- 
 parison with one another; and I suppose it will 
 hardly be asserted that the objective thing is all 
 that is involved in cognition. Moreover, the result 
 is described as the " estabKshing of great and fruit- 
 
TO TEOUGET. 57 
 
 ful truths ;" and this, to my apprehension, means 
 the true reference of things to ideas. 
 
 If we take the two propositions, " The sun moves 
 round the earth," and " The earth moves round the 
 sun," we may say that it is the result of our in- 
 quiries into natural phenomena that we now deny 
 the former and affirm the latter. But I would very 
 deferentially suggest that such inquiry was not in- 
 stituted immediately between the two things sun and 
 earth. The establishment of the truth of the second 
 proposition was founded on a correction of the old 
 notions concerning these objects. But a notion of 
 a thing is, I apprehend, a reference of it to some 
 idea or set of ideas. I have, for example, an idea 
 of motion and an idea of stability. Prima facie I 
 refer my perception of the sun to the former, and 
 my perception of the earth to the latter. If now 
 I affirm a true proposition concerning this particular 
 relation between the sun and the earth, it can only 
 be because I have made a true adjustment of these 
 objects to my ideas of motion and stability. 
 
 " The real import of by far the most numerous 
 class of propositions " is thus stated by the same 
 author, in the same chapter on the Import of Propo- 
 sitions : — "The object of belief in a proposition, 
 when it asserts anything more than the meaning of 
 words, is generally either the coexistence or sequence 
 of two phenomena. At the very commencement of 
 
58 TEE RELATIONS OF LANGUAGE 
 
 our inquiry we found that every act of belief im- 
 plied two Things. We have now ascertained what, 
 in the most frequent case, these two things are, — 
 namely, two Phenomena ; in other words, two states 
 of consciousness ; and what it is which the proposi- 
 tion aiBrms (or denies) to subsist between them, 
 namely either succession or coexistence." 
 
 " Thus,^ in the proposition. All men are mortal, the 
 word man connotes the attributes which we ascribe to 
 a certain kind of living creatures, on the ground of 
 certain phenomena which they exhibit, and which are 
 partly physical phenomena, viz. the impressions made 
 on our senses by their bodily form and structure, 
 and partly mental phenomena — namely, the sentient 
 and intellectual life which they have of their own. 
 All this is understood when we utter the word man, 
 by anyone to whom the meaning of the word is 
 known. l^ow, when we say, Man is mortal, we 
 mean that wherever these various physical and 
 mental phenomena are all found, there we have 
 assurance that the other physical and mental phe- 
 nomenon, called death, will not fail to take place." 
 
 Upon this I remark, that there is some little delu- 
 sion arising from the connotation of the subject : 
 this connotation, as I have already said, chiefly sub- 
 serving the purpose of more precise denotation. The 
 
 ^ This illustration in the original precedes the general statement I 
 have above quoted. 
 
TO THOUGHT. 59 
 
 delusive effect of it is to place the subject apparently 
 on the same footing with the predicate. In the 
 example here given, it must be considered that, 
 looking upon a man as a phenomenon that, in the 
 first place, could only be denoted or indicated, but 
 of which thousands of predications could be made 
 as the resxilt of experience and observation; the 
 word man, in this proposition, connotes the import 
 of all these predications, excepting mortality, which 
 by the proposition is now added to the collection. 
 But each of these predications, whether those im- 
 plied in the word man or this one now added, is a 
 predication concerning the merely denotable object 
 " tJiis." If, as we have supposed, the first predica- 
 tion lost its proper significance and became a name 
 of the thing indicated, and then around this petrified 
 nucleus were gathered all the ideas to which the 
 thing itself was referable ; and if now I make yet 
 another predication ; the subject, whatever now its 
 connotation, still stands in the same relation to the 
 predicate as when it was the naked and merely de- 
 notable thing "this." The relation is still that of 
 an objective phenomenon to a subjective idea. 
 
 Let us, for further illustration, take the propo- 
 sition, " The diamond is combustible." He who 
 first discovered a diamond would either simjDly de- 
 note it, or else give it a name. But any name he 
 he could give would be given only in consequence 
 
60 THE RELATIONS OF LANGUAGE 
 
 of some observed property of the object, and would 
 therefore be equivalent to a predication concerning 
 it. If he called it only a stone, this was a predica- 
 tion of a whole bundle of ideas. Or he might make 
 single predications concerning it, as 
 
 " This is hard:'' 
 
 " This is brilliant : " 
 
 " 2'his is transparent." 
 Now, whichever of these predications (or of any 
 other predications supposed to be made) shall give 
 a name to the object, will probably lose its own 
 special significance. Such a name, whatever it be, 
 is primarily denotative ; but with this denotation 
 it implies, or connotes, all the properties that have 
 been observed in that object. Notwithstanding this 
 connotation, however, the latest predication I make 
 is as much a predication of the objective thing, the 
 merely denotable object " this," as the first one was. 
 The subject of the last predication is the same as 
 of the first ; the difference (however important in 
 other respects, yet unimportant in this) being that 
 I make my later predications of the thing under a 
 name which connotes a bundle of ideas previously 
 predicated. But the result of my last experiment 
 is still this, that the thing now in my hands is com- 
 bustible. You may call it a stone or not, as you 
 please. You may deny its brilliancy or its hard- 
 ness, but the import of my present assertion is that 
 
TO THOUGHT. 61 
 
 this thing (call it by what name you please, or pre- 
 dicate of it any other observable property) is com- 
 bustible. This assertion is made, or may be made, 
 ■^thout reference to other observable properties. 
 Suppose the thing lacked the brilliancy, and pos- 
 sessed the hardness, of the diamond, or vice versa, 
 mj assertion concerning this thing remains true, 
 that it is combustible. Moreover, you may disinte- 
 grate the connotation of the name in as many ways 
 as the name connotes ideas. You may take any one 
 idea and predicate it of all the others. Do I now 
 say that there is a substance underlying, and inde- 
 pendent of, all these attributes ? No ; but I say 
 that the combination of phenomena found in such 
 a thing (the diamond, for instance) is first denoted, 
 or indicated, or demonstrated, — that the results of 
 observation are one by one predicated of it; and 
 that the latest predication is a predication of a 
 thing of which some other 'predications have been 
 made. The thing itself is still an objective pheno- 
 menon. Any predication I make of it is only a re- 
 ferring of it to some one or other of my ideas. And 
 precisely the same remarks may be made concerning 
 the other proposition, that all men are mortal, and 
 of that whole class which Mr. Mill speaks of as " by 
 far the most numerous class of propositions." What 
 I contend for is that a real proposition is not con- 
 cerned with two things (objective) or with two ideas 
 
62 THE RELA TIONS OF LANG TJA GE 
 
 (subjective) ; that it is, rather, a reference of an 
 objective phenomenon (or set of phenomena) to an 
 idea (or set of ideas). 
 
 "When we say," says Mr. J. S. Mill, in a note to 
 the chapter on Predication in his edition of Mr. 
 James Mill's Analysis of the Phenomena of the 
 Human Mind, — "When we say, John walked out 
 this morning, it is not a correct expression of the 
 communication we desire to make, that * having 
 walked out this morning,' or ' a person who has 
 walked out this morning,' are two of the innume- 
 rable names of John. They are only accidental!}^ 
 and momentarily names of John by reason of a cer- 
 tain event, and the information we mean to give is, 
 that this event has happened. The event is not re- 
 solvable into an identity of meaning between names, 
 but into an actual series of sensations that occurred 
 to John, and a belief that anyone who had been 
 present and using his eyes would have had another 
 series of sensations, which we call seeing John in 
 the act of walking out." 
 
 But the question is not about the event but about 
 the assertion of the event ; and this is resolvable 
 into (1) a denotation of the subject, the connotation 
 of whose name (John) only subserves the purpose of 
 denotation, and (2) a reference of the thing so de- 
 noted to the set of ideas involved in the words 
 " walked out this morning." 
 
TO TROJJGBT. 
 
 CHAPTER VI. 
 
 THE ORIGINAL SYNTHESIS OF LANGUAGE. 
 
 Although my purpose does not comprise any 
 speculations concerning the origin of language, the 
 views of this point which have been taken by M. 
 Renan, in so far as they extend to the sphere of our 
 present inquiry, demand consideration. In his essay 
 De VorUjinc chi langage, M. Renan repeatedly asserts 
 that the synthetical stage of language precedes the 
 analytical. Thus in Chap. YII. we read : " II n'est 
 pas douteux qu'on n'ait debute par I'expression 
 compos^e, et que I'esprit, avant de dissequer la 
 pensee et de I'exprimer partie par partie, n'ait 
 d'abord cherch^ a le rendre dans son unite. L'ag- 
 glutination dut etre le precede dominant du Ian- 
 gage des premiers hommes, comme la synthese ou 
 plutot le syncretisme fut le caractere de leur pen- 
 see." In the same paragraph, and in confirmation 
 of this assertion, he says : " Dans nos langues 
 modemes, le sujet, le verbe, ainsi que plusieurs des 
 relations de temps, de modes, et de voix, sont ex- 
 prim^s par des mots isoles et ind^pendants. Dans 
 
G4: THE RELATIONS OF LANGUAGE 
 
 les langues anciennes, au contraire, ccs idi^cs sont 
 le plus souvent accumulees dans un mot unique et 
 exprimees par une flexion. Le seul mot aniabor 
 renferme I'idee d'aimer, la notion de la premiere 
 personne, celle du futur et celle du passif. L'alle- 
 mand en disant : Ich werde gelieht werden, represente 
 ces quatre notions par quatre mots separes." The 
 sum of all which is that "I'idee s'est exprim^e 
 d'abord avec tout son cortege de determinatifs et 
 dans une parfaite unite." 
 
 It must be borne in mind that our author is in- 
 tent upon maintaining for language its position as 
 a natural process of the mind in opposition to two 
 non -natural theories : the one which rej)resents lan- 
 guage as supernaturally bestowed by God : the other 
 which represents it as arbitrarily and artificially — 
 that is to say, unnaturally — invented by man him- 
 self. As against these two theories, we agree with 
 M. Renan, that language has its roots in, and is a 
 natural process from, human nature. And if we 
 acknowledge this whole nature to be of God, it is 
 absurd (not to say profane) to recognize Him only 
 at the vanishing point of our own ken. 
 
 " Nee Deus intersit, nisi dignus vindice nodus 
 Incident," 
 
 is a rule that might be observed with advantage by 
 those who, with equal readiness, summon the Deity 
 to the first knot that occurs, and dismiss Him when 
 
TO THOUGHT. 65 
 
 thej find that they can unravel it themselves. Lan- 
 guage is of God as thought is of God, and not other- 
 wise. The analysis of language, as of thought, 
 discloses elements that can never have come under 
 man's arbitrary manipulation. They surpass his 
 synthetical power (at least his conscious power of 
 synthesis) quite as much as do the material elements 
 that are detected by chemical analysis. 
 
 But, admitting all this, we cannot concur with the 
 opinion that the analysis of language can teach us 
 nothing of language as it is. Such appears to be 
 the opinion of M. Renan, who forbids the supposi- 
 tion of an elementary stage as preliminary to the 
 synthetical ; and would have us discern the elements 
 of language only in its corruption, decomposition, 
 and decay. 
 
 " It is true," says Professor Huxley,^ writing upon 
 a subject very analogous to that which here concerns 
 us, " that chemical investigation can tell us little or 
 nothing, directly, of the composition of living 
 matter, inasmuch as such matter must needs die in 
 the act of analysis ; and upon this very obvious 
 groimd, objections, which, I confess, seem to me 
 to be somewhat frivolous, ha.ve been raised to the 
 drawing of any conclusions whatever respecting the 
 composition of actually living matter, from that of 
 
 ' In a contribution to the Fortnightly Eeview, Feb. 1st, 1869, On 
 the Physical Basis of Life. 
 
66 THE RELATIONE OF LANGUAGE 
 
 the dead matter of life, which alone is accessible to 
 us. But objectors of this class do not seem to re- 
 flect that it is also, in strictness, true that we know 
 nothing about the composition of any body whatever, 
 as it is. The statement that a crystal of calc-spar 
 consists of carbonate of lime, is quite true, if we only 
 mean that, by appropriate processes, it may be re- 
 solved into carbonic acid and quicklime. If you 
 pass the same carbonic acid over the very quicklime 
 thus obtained, you will obtain carbonate of lime 
 again ; but it will not be calc-spar, nor anything 
 like it. Can it, therefore, be said that chemical 
 analysis teaches nothing about the chemical compo- 
 sition of calc-spar ? Such a statement would be 
 absurd ; but it is hardly more so than the talk one 
 occasionally hears about the uselessness of applying 
 the results of chemical analysis to the living bodies 
 which have yielded them." 
 
 For reasons perfectly analogous, we assume that 
 even in the interpretation of language, the analysis 
 of its elements must be of service, though the ele- 
 ments themselves are not language. 
 
 With characteristic indisposition to give historical 
 precedence to the simpler forms of language, M. 
 Renan attacks the theory of the primitive biliteral 
 character of Semitic roots. The arguments in favour 
 of this theory cannot be stated with more effect than 
 by M. Renan himself. I refer here both to his essay 
 
TO THOUGHT. 67 
 
 Be Vorigine du langage and to his Histoire generale des 
 langues Sdmitiques. He does not disguise tlie facts 
 that amongst the Semitic roots there are many that 
 are triliteral only by a grammatical fiction : that 
 others more really triliteral are distinguished by the 
 feebleness of one of their radical letters ; and that 
 of nearly all, if not of all, it may be asserted that 
 the comparison of roots of kindred meaning shews 
 the fundamental meaning of each to reside in two 
 of its radical letters — the third serving to modify- 
 that fundamental meaning. On the whole evidence 
 he concludes that " on est amene a se representor 
 chaque racine semitique comme essentiellement com- 
 posee de deux lettres radicales." Has one then a 
 right to suppose that these Semitic languages have 
 actually passed through such a stage ? " VoUa sur 
 quoi un esprit sage, persuade qu'on ne saurait de- 
 viner a priori les voies infiniment multiples de I'es- 
 prit humaiu, h^sitera toujoui's a se prononcer." M. 
 Renan slips away from the conclusion, like an eel 
 seized by the tail, by way of illustrating " les voies 
 infiniment multiples de I'esprit." " Comment con- 
 cevoir," he asks, "le passage de I'etat monosylla- 
 bique a I'etat trilitere ? Quelle cause assignor a 
 cette revolution ? A quelle epoque la placer ? 
 Serait-ce, comme le disaient naivement les anciens 
 linguistes, lorsque les id^es se multiplierent et qu'on 
 sentit le besoin d'exprimer plus de nuances, ou, 
 
68 TEE RELATIONS OF LANGUAGE 
 
 comme Gesenius inclinait a le croirc, au moment de 
 rintroduction do I'ecriturc ? Est-ce par hasard, est- 
 ce d'un commun accord que se fit cette innovation 
 grammaticale ? On s'arrute devant ies impossibilites 
 que pr^sentent a I'imagination de telles hypotheses. 
 Le passage de I'etat monosyllabique a I'^tat trilitere 
 est de ceux qui n'auraient pu se faire sans une tres- 
 grande reflexion." 
 
 Throughout this passage M. Renan considers 
 letters rather than sounds. The modification of a 
 primitive root may be represented by the addition of 
 a third letter ; but this modification has probably 
 taken place before sounds were represented by letters 
 at aU, — that is, before the invention of alphabetical 
 writing. It is therefore unfair to assume such an 
 abrupt transition as is implied when we speak of the 
 change from a biliteral to a triHteral system. Be- 
 sides which, are there not innumerable instances in 
 the Indo-European languages of primitive roots being 
 modified in a manner which is represented in writing 
 by the addition of one or more letters ? 
 
 The position that the synthetical stage of language 
 preceded the analytical would be more tenable if, first 
 conceding to our author that " I'idee s'est exprimee 
 d'abord avec tout son cortege de determinatifs et 
 dans une parfaite unite," we could find such unity 
 represented in the word by which such an idea is 
 expressed. If, to take 31. Eenan's example above, 
 
TO THOUGHT, 69 
 
 we could find that, however the idea conveyed by 
 the one word amabor may be analyzed into the ideas 
 conveyed by the four words I shall he loved, yet that 
 the word itself cannot be so disintegrated, — then we 
 might take this word as representing the "perfect 
 unity " in which a thought, complex to us, was re- 
 ceived and expressed by les premiers hoimnes. But 
 we find no such thing. On the contrary, we find 
 that the analysis of the word yields results com- 
 pletely corresponding with those of the analyzed 
 thought. 
 
 To say that the synthetical form is that in which 
 the most ancient languages are found, is to say no- 
 thing to this point. There appears to be a continual 
 revolution from synthetical to analytical and from 
 analytical to synthetical forms. The modifying and 
 determining elements cluster round a root, and form 
 with it one word. This device answers well enough 
 so long as each element preserves the consciousness 
 of its meaning. Let it once lose this, and then, 
 though it may still cling to the nucleus, like the 
 shell to the newly -hatched chicken, yet its place is 
 supplied by another, having a conscious meaning, 
 and commencing its duties as an independent word. 
 This, too, in process of time, may become attached, 
 die, and be thrown ofi*. Thus, if we look at the 
 synthetical form amaho, and find that the thought 
 conveyed by this word is analyzed in the early stages 
 
70 TEE RELATIONS OF LANGUAGE 
 
 of the llomance languages, we find a synthetical 
 process again taking place in the later stages. 
 
 J' aimer ai is the synthetical form of fai a aimer} 
 
 We are therefore justified, I think, in assuming 
 the historical existence of an analytical form as 
 preliminary to the synthetical amaho, although the 
 monuments of the Latin language may hear no 
 testimony to it. " The period during which, as in 
 the Proven9al dir vos ai, the component elements of 
 the old Aryan grammar maintained a separate 
 existence in the language and the mind of the 
 Aryans, had closed before Sanskrit was Sanskrit 
 or Grreek Greek. That there was such a period we 
 can doubt as little as we can doubt the real existence 
 of fern forests previous to the formation of our coal- 
 fields." (Prof. Max MiiUer.) 
 
 In like manner, if in the d of I loved we recognize 
 the auxiliary did, which appears in the analytical 
 expression / did love, we could hardly doubt that love 
 did preceded loved, even though we had not the tes- 
 
 1 " In poems of the Troubadours we find : — ' E si li platz, alberguar 
 nCa,' ' and if it pleases him, he has to lodge me.' 
 
 ' Amarai ? oc ; si li platz ni I'es gens, 
 E sinol platz, amar I'ai eissamen.' 
 ' ShaU. I love ? Yes, if it pleases her and she is kind ; and if it does 
 not please her, I will love her equally.' " 
 
 " In ProveuQal, too, the verbs aver and esser, with the preposition a 
 before another verb, were used to express the future ; as, ' ab lieys ai 
 a guerir,' 'with her I shall recover.' ' A I'advenement del qual tuit 
 an a ressuscitar,' ' at whose coming all will rise again.' ' Tern que 
 m'er a morir,' 'I fear I shall die."' (Sir G. C. Lewis, On the 
 Bomance Languages, p. 174.) 
 
TO THOTTGET. 71 
 
 timony of the Gothic preterite, in the plural of 
 which the full auxiliary has been preserved. 
 
 Mr. Farrar, in his Essay on the Origin of Lan- 
 guage, gives a chapter which is mainly a repetition 
 of M. Kenan's remarks upon this subject, adding 
 however, some very curious illustrations of his own. 
 Speaking of what he calls the holophrastic style of 
 language, in which " the entire thought or sentence 
 is produced under the form of one complex and rich 
 unity," and attempting to shew that this rich com- 
 plexity had historical precedence of simpler and more 
 analytical forms — repeating the very words of M. 
 Renan, that "it has been a fatal mistake of philology 
 to suppose that simplicity is anterior to complexity " 
 — he adduces, as examples of holop/inisis in early 
 Greek, the expressions Kara Saxpv ^eoucra and Kara 
 iriova fJb'^pl eKTja. Surely these are to be taken 
 rather as examples of analytical language : — not, 
 to use Mr. Farrar's figure, as specimens of " the 
 conglomerated jewels of old speech." But even if 
 we could find that the most ancient languages are 
 in the synthetical stage, a survey of their subsequent 
 history wiU forbid us to regard that stage as ulti- 
 mate; and, in my opinion, will afford no warrant 
 for the paradoxical assertion that in language com- 
 plexity precedes simplieity. 
 
 It is undoubtedly true that in consciousness we are 
 more familiar with synthetical combinations than 
 
72 THE RELATIONS OF LANGUAGE 
 
 with the component particulars disclosed by analysis. 
 But the synthesis has been our own work. "We have 
 had to deal with the elementary particulars first, 
 although we may have forgotten such dealings, and 
 so conclude that they were never transacted in con- 
 sciousness. Nay more — it is certain that in opera- 
 tions whereof only the synthetical result appears in 
 consciousness, we have had to deal primarily with 
 elements that are revealed to consciousness only by 
 close analysis. The most uninstructed person will 
 distinguish readily between a potato and an apple ; 
 and yet it is certain that the perception of each is a 
 very composite perception, the elements of which 
 must have been not only severally apprehended but 
 "co-ordinated," See upon this point Mr. Herbert 
 Spencer's Principles of Psychology, Chap, x., from 
 which I take the following example. 
 
 "When learning to read, the child has to class 
 each individual letter by a distinct mental act. This 
 symbol A has to be thought of as like certain others 
 before seen ; and as standing for a sound like certain 
 sounds before heard. By continued practice these 
 processes become more and more abbreviated and 
 unconscious. Presently the power is reached of 
 classing by one act a whole group of such symbols 
 — a word ; and eventually an entire cluster of such 
 words is taken in at a glance. JN^ow, were it not 
 that these steps can be recalled, it would seem ab- 
 
TO THOUGHT. 73 
 
 surd to say that when the reader, by what appears 
 almost a single cognition, takes in the sentence — 
 ' This is true,' — that he not only classifies each 
 word with the before-known like words, but each 
 letter with the before-known like letters. Yet, as 
 it is, he will see this to be an unavoidable inference. 
 For, as it is undeniable that such acts of classing 
 were performed at first, and as no time can be named 
 at which such acts were given up, it follows that the 
 entire change has arisen from their immensely in- 
 creased rapidity — from their having become automatic 
 or organic. And if this result has taken place with 
 acts of classing that were commenced so late as five 
 or six years old, still more must it have taken place 
 with those much simpler ones that were commenced 
 at birth." 
 
 No illustration could be more to the point. In 
 contemplating, on the one hand, the faculty of ex- 
 pressing our ideas in words, and, on the other, the 
 faculty of taking in such ideas by a mere glance at the 
 symbolization of their minutest elements, — it woidd 
 seem even less paradoxical to say, however untruly, 
 of the latter, what M. Renan asserts of the former, 
 that it was manifested, all at once, like Minerva from 
 1 1 :> brain of Jupiter, completely armed, equipped, and 
 furnished — " avec tout son cortege, et dans une par- 
 faite unite." 
 
LIBRftRY OF CONGRESS 
 
 Illlil|i1liii||i mil II iilirin iilij liii jlli ii'i 
 003 023 204 5 • 
 
 HEKTFORD : 
 
 IKI) BY STEPHEN AU3TIW.