[103] Of course concepts may be something more than mere recepts known as such: they may be the knowledge of other concepts. But with this higher stage of conceptual ideation I am not here concerned.

[104] Nature, August 21, 1879.

[105] Taine, Intelligence, pp. 399, 400.

[106] Or, as we may now more closely define it, a denominated recept. A merely denotated recept (such as a parrot’s name for its recept of dog) is not conceptual, even in the lowest degree. In other words, named recepts, merely as such, are not necessarily concepts. Whether or not they are concepts depends on whether the naming has been an act of denotation or of denomination—conscious only, or likewise self-conscious.

[107] I coin this word on the pattern already furnished by “pre-perception,” which was first introduced by Lewes, and is now in general use among psychologists.

[108] Touching the power of recognizing pictorial representations among animals, this unquestionably occurs in dogs (see Animal Intelligence, pp. 455, 456), and there is some evidence to show that it is likewise displayed by monkeys. For Isidore Geoffroy St. Hilaire relates of a species of Midas (Corinus) that it distinguished between different objects depicted on an engraving; and Audouin “showed it the portraits of a cat and a wasp, at which it became much terrified: whereas, at the sight of a figure of a grasshopper or a beetle, it precipitated itself on the picture, as if to seize the objects there represented” (Bates, Nat. on Amaz., p. 60). The age at which a young child first learns to recognize pictorial resemblances no doubt varies in individual cases. I have not met with any evidence on this subject in the writings of other observers of infant psychology. The earliest age at which I observed any display of this faculty in my own children was at eight months, when my son stared long and fixedly at my own portrait in a manner which left no doubt on my mind that he recognized it as resembling the face of a man. Moreover, always after that day when asked in that room, “Where’s papa?” he used at once to look up and point at the portrait. Another child of my own, which had not seen this portrait till she was sixteen months old, immediately recognized it at first sight, as was proved by her pointing to it and calling it “Papa.” Two months later I observed that she also recognized pictorial resemblances of animals, and for many months afterwards her chief amusement consisted in looking through picture-books for the purpose of pointing out the animals or persons depicted—calling “Ba-a-a” to the sheep, “Moo” to the cows, grunting for the pigs, &c., these sundry sounds having been taught her as names by the nurse. She never made a mistake in this kind of nomenclature, and spontaneously called all pictorial representations of men “Papa,” of women “Mama,” and of children “Ilda”—the latter being the name which she had given to her younger brother. Moreover, if a picture-book were given into her hands upside-down, she would immediately perceive and rectify the mistake; and whenever she happened to see a pictorial representation of an animal—as, for instance, on a screen or wall-paper—she would touch it and utter the sound that was her name for that animal. With a third child, who was still wholly speechless at eighteen months, I tried the experiment of spreading out a number of photographic portraits, and asking him “Which is mamma? Which is papa?” &c. Without any hesitation he indicated them all correctly.

[109] By using the word “judgment” in all these cases I am in no way prejudicing the argument of my opponents. The explanation which immediately follows in the text is sufficient to show that the qualifying terms “receptual” and “pre-conceptual” effectually guard against any abuse of the term—quite as much, for instance, as when psychologists speak of “perceptual judgments,” or “unconscious judgments,” or “intuitive judgments,” in connection with still lower levels of mental operation. And it seems to me better thus to qualify an existing term than to add to the already large number of words I have found it necessary to coin.

[110] I may here remark that this possibility of receptual predication on the part of talking birds is not entirely hypothetical: I have some evidence that it may be actually realized. For instance, a correspondent writes of a cockatoo which had been ill:—“A friend came the same afternoon, and asked him how he was. With his head on one side and one of his cunning looks, he told her that he was ‘a little better;’ and when she asked him if he had not been very ill, he said, ‘Cockie better; Cockie ever so much better.’ ... ‘When I came back (after a prolonged absence) he said, ‘Mother come back to little Cockie: Mother come back to little Cockie. Come and love me and give me pretty kiss. Nobody pity poor Cockie. The boy beat poor Cockie.’ He always told me if Jes scolded or beat him. He always told me as soon as he saw me, and in such a pitiful tone.... The remarkable thing about this bird is that he does not merely ‘talk’ like parrots in general, but so habitually talks to the purpose.”

[111] Lest there should still be any ambiguity about the numerous terms which I have found it necessary to coin, I will here supply a table of definitions.

Lower recept = an automatic grouping of percepts.

Higher recept = pre-concept; or a degree of receptual ideation which does not occur in any brute.

Lower concept = named recept, provided that the naming be due to reflective thought.

Higher concept = a named compound of concepts.

The analogues of these terms are, in the matter of naming:—

Receptual naming = denotation, which includes pre-conceptual naming.

Conceptual naming = denomination.

And, in the matter of judging, the analogues are:—

Receptual judgment = automatic, “practical,” or unthinking inference.

Pre-conceptual judgment = the higher, though still unthinking, inferences of a child prior to the rise of self-consciousness.

Conceptual judgment = true judgment, whether exhibited in denomination, predication, or any act of inference for which self-conscious thought may be required.

[112] See above, Chapters II. and IV.

[113] See Mental Evolution in Animals, chapter on “Imagination.”

[114] In the opinion of Wundt, the most important of all conditions to the genesis of self-consciousness is given by the muscular sense in acts of voluntary movement (Vorlesungen über die Menschen und Thierseele, 18 vol.). While agreeing with him that this is a highly important condition, I think the others above mentioned are quite as much, or even more so.

[115] See for cases of this, Animal Intelligence, pp. 410, 443, 444, 450-452, 458, 494.

[116] The following is a good example of ejective ideation in a brute—all the better, perhaps, on account of being so familiar. I quote it from Quatrefage’s Human Species, pp. 20, 21:—“I must here beg permission to relate the remembrance of my struggles with a mastiff of pure breed and which had attained its full size, remaining, however, very young in character. We were very good friends and often played together. As soon as ever I assumed an attitude of defence before him, he would leap upon me with every appearance of fury, seizing in his mouth the arm which I had used as a shield. He might have marked my arm deeply at the first onset, but he never pressed it in a manner that could inflict the slightest pain. I often seized his lower jaw with my hand, but he never used his teeth so as to bite me. And yet the next moment the same teeth would indent a piece of wood I tried to tear away from them. This animal evidently knew what it was doing when it feigned the passion precisely opposite to that which it really felt; when, even in the excitement of play, it retained sufficient mastery over its movements to avoid hurting me. In reality it played a part in a comedy, and we cannot act without being conscious of it.”

[117] Not, however, wholly so. Mr. Chauncey Wright has clearly recognized the existence of what I term receptual self-consciousness, and assigned to it the name above adopted—i.e. “outward self-consciousness.” See his Evolution of Self-consciousness. Mr. Darwin, also, appears to have recognized this distinction, in the following passage:—“It may be freely admitted that no animal is self-conscious, if by this term is implied that he reflects on such points as whence he comes or whither he will go, or what is life and death, and so forth. But how can we feel sure that an old dog with an excellent memory and some power of imagination, as shown by his dreams, never reflects on his past pleasures or pains in the chase? And this would be a form of self-consciousness” (Descent of Man, p. 83). Of course a psychologist may take technical exception to the word “reflects” in this passage; but that this kind of receptual reflection does take place in dogs appears to me to be definitely proved by the facts of home-sickness and pining for absent friends, above alluded to.

[118] In the present connection the following very pregnant sentence may be appropriately quoted from Wundt:—“Wenn wir überall auf die Empfindung als Ausgangspunkt der ganzen Entwicklungsreihe hingewiesen werden, so müssen auch die Anfänge jener Unterscheidung des Ichs von den Gegenständen schon in den Empfindungen gelegen sein” (Vorlesungen über die Menschen und Thierseele, i. 287). And to the objection that there can be no thought without knowledge of thought, he replies that before there is any knowledge of thought there must be the same order of thinking as there is of perceiving prior to the advent of self-consciousness—e.g. receptual ideas about space before there is any conceptual knowledge of these ideas as such.

[119] Sully, loc. cit., p. 376. See also Wundt, loc. cit., i. 289. He shows that this speaking of self in the third person is not due to “imitation,” but, on the contrary, opposed to it. For “a thousand times the child hears that its elders do not thus speak of themselves.” The child hears that its elders call it in the third person, and in this it follows them. But such imitation as we here find is expressive only of the fact that hitherto the child has not distinguished between self as an object and self as a subject. Only later on, when this distinction has begun to dawn, does imitation proceed to apply to the self the first person, after the manner in which other selves (now recognized by the child as such) are heard to do.

[120] Loc. cit., p. 377.

[121] Loc. cit., pp. 435, 436.

[122] Philosophical Discussions, p. 256. See also Animal Intelligence, pp. 269, 270, for the case of a parrot apparently endeavouring to recover the memory of a particular word in a phrase. In the course of an interesting research on the intelligence of spiders (Journ. Morphol., i., p. 383-419), Mr. and Mrs. Peckham have recently found that the memory of eggs which have been withdrawn from the mother is retained by her for a period varying in different species from less than one to more than two days.

[123] Sully, loc. cit., p. 377.

[124] Wundt, loc. cit., ii. 289, 290. He gives cases where such a definite memory of the moment has persisted, and elsewhere states that such is the case in his own experience. The circumstance which here was connected with the sudden birth of self-consciousness consisted in rolling down stairs into a cellar—an event which no doubt was well calculated forcibly to impress upon infant consciousness that it was itself, and nobody else.

[125] See Mental Evolution in Animals, pp. 161-165. Perez records analogous facts with regard to the infant as unmistakably displayed in the fourteenth week (First Three Years of Childhood, English trans., p. 29).

[126] Outlines of Psychology, p. 378.

[127] Vorlesungen, &c., i. 289.

[128] In the above sketch of the principles which are concerned in the development of self-consciousness, I have only been concerned with the matter on the side of its psychology, and even on this side only so far as my own purposes are in view. Those who wish for further information on the psychology of the subject may consult Wundt, loc. cit.; Sully, loc. cit., and Illusions, ch. x.; Taine, On Intelligence, pt. ii., bk. iii.; Chauncey Wright, Evolution of Self-consciousness; and Waitz, Lehrbuch der Psychologie, 58. On the side of its physiology and pathology Taine, Maudsley, and Ribot may be referred to (On Intelligence, Pathology of Mind, Diseases of Memory), as also a paper by Herzen, entitled, Les Modifications de la Conscience du moi (Bull. Soc. Hand. Sc. Nat., xx. 90). An Essay on the Philosophy of Self-consciousness, by P. F. Fitzgerald, is written from the side of metaphysics. On this side, also, we are met by the school of Hegel and the Neo-Kantians with a virtual denial of the origin and development of self-consciousness in time. Thus, for instance, Green expressly says:—“Should the question be asked, If this self-consciousness is not derived from nature, what then is its origin? the answer is, that it has no origin. It never began because it never was not. It is the condition of there being such a thing as beginning or end. Whatever begins or ends does so for it, or in relation to it” (Prolegomena to Ethics, p. 119). To this I can only answer that for my own part I feel as convinced as I am of the fact of my self-consciousness itself that it had a beginning in time, and was afterwards the subject of a gradual development. “Das Ich ist ein Entwicklungsprodukt, wie der ganze Mensch ein Entwicklungsprodukt ist” (Wundt).

[129] “Of all the neolithic implements the axe was by far the most important. It was by the axe that man achieved his greatest victory over nature” (Boyd Dawkins, Early Man in Britain, p. 274).

[130] Galton, Tropical South Africa, p. 213. The author adds, “Once, while I watched a Dammara floundering hopelessly in a calculation on one side of me, I observed Dinah, my spaniel, equally embarrassed on the other. She was overlooking half a dozen of her new-born puppies, which had been removed two or three times from her, and her anxiety was excessive, as she tried to find out if they were all present, or if any were still missing. She kept puzzling and running her eyes over them, backwards and forwards, but could not satisfy herself. She evidently had a vague notion of counting, but the figure was too large for her brain. Taking the two as they stood, dog and Dammara, the comparison reflected no great honour on the man.” As previously stated, I taught the chimpanzee “Sally” to give one, two, three, four, or five straws at word of command.

[131] The boy’s name was Ernest, and was thus called by all other members of the household. As I could not find any imitative source of the dissimilar name used by his sister, this is probably an instance of the spontaneous invention of names by young children, which has already been considered at the close of my chapter on “Articulation.” Touching the use of adjectives by young children, I may quote the following remark from Professor Preyer:—“A very general error must be removed, which consists in the supposition that all children on first beginning to speak use substantives only, and later pass on to the use of adjectives. This is certainly not the case.” And he proceeds to give instances drawn from the daily observations of his own child, such as the use of the word “heiss” in the twenty-third month.

[132] We shall subsequently see that at this stage of mental evolution there is no well-defined distinction between the different parts of speech. Therefore here, and elsewhere throughout this chapter, I use the terms “noun,” “adjective,” “verb,” &c., in a loose and general sense.

[133] I have seen a terrier of my own (who habitually employed this gesture-sign in the same way as Preyer’s child, namely, as expressive of desire), assiduously though fruitlessly “beg” before a refractory bitch.

[134] Many dogs will significantly bark, and cats significantly mew, for things which they desire to possess or to be done. For significant crying by children, see above, p. 158.

[135] For the case of the ape in this connection see above, p. 126. I took my daughter when she was seven years of age to witness the understanding of the ape “Sally.” On coming away, I remarked to her that the animal seemed to be “quite as sensible as Jack”—i.e. her infant brother of eighteen months. She considered for a while, and then replied, “Well, I think she is sensibler.” And I believe the child was right.

[136] Or, if any opponent were to suggest this, he would be committing argumentative surrender. For the citadel of his argument is, as we know, the faculty of conception, or the distinctively human power of objectifying ideas. Now, it is on all hands admitted that this power is impossible in the absence of self-consciousness. Will it, then, be suggested that my daughter had attained to self-consciousness and the introspective contemplation of her own ideas before she had attained to the faculty of speech, and therefore to the very condition to the naming of her ideas? If so, it would follow that there may be concepts without names, and thus the whole fortress of my opponents would crumble away.

[137] See pp. 81-83, where it is shown that even in cases where conceptual thought is necessary for the original formation of a name, the name may afterwards be used without the agency of such thought—just in the same way as actions originally due to intelligence may, by frequent repetition, become automatic. At the close of the present chapter it will be shown that the same is true even of full or formal predication.

[138] In this connection it is interesting to observe the absence of the copula. Notwithstanding the strongly imitative tendencies of a child’s mind, and notwithstanding that our English children hear the copula expressed in almost every statement that is made to them, their own propositions, while still in the preconceptual phase, dispense with it (see above, p. 204). In thus trusting to apposition alone, without expressing any sign of relation, the young child is conveying in spoken language an immediate translation of the mental acts concerned in predication. As previously noticed, we meet with precisely the same fact in the natural language of gesture, even after this has been wrought up into the elaborate conceptual systems of the Indians and deaf-mutes. Lastly, in a subsequent chapter we shall see that the same has to be said of all the more primitive forms of spoken language which are still extant among savages. So that here again we meet with additional proof, were any required, of the folly of regarding the copula as an essential ingredient of a proposition.

[139] See p. 166.

[140] Thus far, it will be observed, the case of predication is precisely analogous to that of denomination, alluded to in the foot-note on page 226. Just as instincts may arise by way of “lapsed intelligence,” so may originally conceptual names, and even originally conceptual propositions, become worn down by frequent use, until they are, as it were, degraded into the pre-conceptual order of ideation. Be it observed, however, that the paragraphs which follow in the text have reference to a totally different principle—namely, that there may be propositions strictly conceptual as to form, which, nevertheless, need never at any time have been conceptual as to thought.

[141] Logic, vol. i., p. 108.

[142] Encyclopædia Britannica, eighth edition, 1857, Art. “Language.”

[143] Of course in classical times, when there was no theological presumption against the theory of development, this alternative met with a fuller recognition; as, for example, by the Latin authors, Horace, Lucretius, and Cicero. Before that time Greek philosophers had been much exercised by the question whether speech was an intuitive endowment (analogists), or a product of human invention (anomalists); and, earlier still, astonishing progress had been made by the grammarians of India in a truly scientific analysis of language-growth. But in the text I am speaking of modern times; and here I think there can be no doubt that till the middle of the present century the possibility of language having been the result of a natural growth was not sufficiently recognized. Among those who did recognize it, Herder, Monboddo, Sir W. Jones, Schlegel, Bopp, Humboldt, Grimm, and Pott, are most deserving of mention. The same year that witnessed the publication of the Origin of Species (1859), gave to science the first issue of Steinthal’s Zeitschrift für Völkerpsychologie und Sprachwissenschaft. From that date onwards the theory of evolution in its application to philology has held undivided sway.

[144] Encycl. Brit., loc. cit. Remembering that the above was published two years before the Origin of Species by means of Natural Selection, this clear enunciation of the struggle for existence in the field of philology appears to me deserving of notice.

[145] Science of Thought, preface, p. xi.

[146] Darwinism tested by the Science of Language, p. 41.

[147] There is a difference of opinion among philologists as to the extent in which modifying constants were themselves originally roots. The school of Ludwig regards demonstrative elements as never having enjoyed existence as independent words; but, even so, they must have had an independent existence of some kind, else it is impossible to explain how they ever came to be employed as constantly modifying different roots in the same way. Moreover, as Max Müller well observes, “to suppose that Khana, Khain, Khanana, Khaintra, Khatra, &c., all tumbled out ready-made, without any synthetical purpose, and that their differences were due to nothing but an uncontrolled play of the organs of speech, seems to me an unmeaning assertion.... What must be admitted, however, is that many suffixes and terminations had been wrongly analyzed by Bopp and his school, and that we must be satisfied with looking upon most of them as in the beginning simply demonstrative and modificatory” (loc. cit., pp. 224 and 225). See also Farrar, Origin of Language, pp. 100, et seq.; Donaldson, Greek Grammar, pp. 67-79; and Hovelacque, Science of Language, p. 37. It will be remarked that this question does not affect the exposition in the text.

[148] Grundriss der Sprachwissenschaft, I. i. 77. This estimate is accepted by Professor Sayce, Introduction to the Science of Language, vol. ii., p. 32.

[149] Hovelacque, Science of Language, English trans., p. 37.

[150] This method of representation was devised by Schleicher, who carries it further than I have occasion to do in the text. See Memoirs of Academy of St. Petersburg, vol. i., No. 7, 1859.

[151] Hovelacque, loc. cit., p. 130.

[152] Sayce, Introduction, &c., i. 126.

[153] Introduction, &c., vol. i., p. 374.

[154] Ibid., vol. i., pp. 375, 376.

[155] Ibid., p. 120. See also his Principles of Comparative Philology, 2nd ed., p. ix.

[156] Sayce, Introduction, &c., i., 125, 126.

[157] Hovelacque, Science of Language, p. 130.

[158] “What we most need to note is the very narrow limitation of our present knowledge. Even among the neighbouring families like the Algonquin, Troquois, and Dakota, whose agreement in style of structure (polysynthetic), taken in connection with the accordant race-type of their speakers, forbids us to regard them as ultimately different, no material correspondence, agreements in words and meanings, is to be traced; and there are in America all degrees of polysynthetism, down to the lowest, and even to its entire absence. Such being the case, it ought to be evident that all attempts to connect American languages as a body with languages of the Old World are, and must be, fruitless: in fact, all discussions of the matter are at present unscientific” (Professor Whitney in Encycl. Brit., art. “Philology,” 1885).

[159] Introduction, &c., i. 120.

[160] Ibid., i. 116.

[161] “The number of separate families of speech now existing in the world, which cannot be connected with one another, is at least seventy-five; and the number will doubtless be increased when we have grammars and dictionaries of the numerous languages and dialects which are still unknown, and better information as regards those with which we are partially acquainted. If we add to these the innumerable groups of speech which have passed away without leaving behind even such waifs as the Basque of the Pyrenees, or the Etruscan of ancient Italy, some idea will be formed of the infinite number of primæval centres or communities in which language took its rise” (Sayce, Introduction, &c., ii. 323).

[162] Life and Growth of Language, p. 259.

[163] Ibid., p. 262.

[164] I may add that the hypothesis admits of corroboration from sources not mentioned by its author. For Archdeacon Farrar wrote in 1865:—“The neglected children in some of the Canadian and Indian villages, who are left alone for days, can and do invent for themselves a sort of lingua franca, partially or wholly unintelligible to all except themselves;” and he quotes Mr. R. Moffat as “testifying to a similar phenomenon in the villages of South Africa (Mission Travels).” He also alludes to the fact that “deaf-mutes have an instinctive power to develop for themselves a language of signs,” which, as we have seen in an earlier chapter, embraces the use of arbitrary articulations, even though in this case the speakers cannot themselves hear the sounds which they make.

While this work is passing through the press an additional paper has been published by Dr. Hale, entitled, The Development of Language. It supplies further evidence in support of this hypothesis.

[165] Wundt, Vorlesungen, &c., ii., 380, 381.

[166] Sayce, Introduction to Science of Language, ii, 13.

[167] The difference of opinion in question seems to arise from individual prepossessions with regard to the ulterior question whether or not the aboriginal roots of all languages must have been polysyllabic. For my own part, and for the reasons already given, I can see no presumption in favour of the view that primitive languages must all have presented the “polysinthetic genius.”

[168] Histoire des Langues Semitique, p. 138.

[169] Etymological Dictionary, p. 746.

[170] See Max Müller, Science of Thought, p. 332.

[171] Ibid., p. 404.

[172] Ethnologische Forschungen, ii., s. 73, et seq. He here quotes Varro to the effect that the roots of Latin amount to about a thousand.

[173] Language and the Study of Language, p. 256.

[174] Sayce, Introduction to the Science of Language, ii., p. 4.

[175] Geiger, Ursprung der Sprache, s. 16.

[176] Sayce, loc. cit., ii. p. 6.

[177] Wedgwood, Etymol. Dict., p. iii.

[178] Farrar, Origin of Language, p. 53.

[179] Science of Thought, p. 439.

[180] Science of Thought, p. 549.

[181] Science of Thought, pp. 551, 552.

[182] Ibid., pp. 551, 552.

[183] “The Aryan languages are the languages of a civilized race; the parent speech to which we may inductively trace them was spoken by men who stood on a relatively high level of culture” (Sayce, Introduction, &c., i. 56). “The primitive tribe which spoke the mother-tongue of the Indo-European family was not nomadic alone, but had settled habitations, even towns and fortified places, and addicted itself in part to the rearing of cattle, in part to the cultivation of the earth. It possessed our chief domesticated animals—the horse, the ox, the goat, and the swine, besides the dog: the bear and the wolf were foes that ravaged its flocks; the mouse and the fly were already domestic pests.... Barley, and perhaps also wheat, was raised for food, and converted into meal. Mead was prepared from honey, as a cheering and inebriating drink. The use of certain metals was known; whether iron was one of them admits of question. The art of weaving was practised; wool and hemp, and possibly flax, being the materials employed.... The weapons of offence and defence were those which are usual among primitive peoples, the sword, spear, bow, and shield. Boats were manufactured and moved by oars.... The art of numeration was learned, at least up to a hundred; there is no general Indo-European word for ‘thousand.’ Some of the stars were noticed and named; the moon was the chief measurer of time. The religion was polytheistic, a worship of the personified powers of nature” (Whitney, Language and the Study of Language, pp. 207, 208). For a more detailed account of this interesting people, see Poescher, Die Arier.

[184] “Unsere Wurzeln sind die Urwurzeln nicht; wir haben vielleicht, von keiner einzigen die erste, ursprüngliche Laut-form mehr vor uns, ebensowenig wohl die Urbedeutung” (Geiger, Ursprung der Sprache, s. 65). And this opinion, so far as I know, is adopted as an axiom by all other philologists.

[185] “It is impossible to bring down the epoch at which the Aryan tribes still lived in the same locality, and spoke practically the same language, to a date much later than the third millennium before the Christian era” (Sayce, Introduction, &c., ii., p. 320).

[186] This fact alone would be sufficient to dispose of what I cannot but consider, from any and every point of view, the transparent absurdity of the doctrine that “the formation of thought is the first and natural purpose of language, while its communication is accidental only” (Science of Thought, p. 40). Such a “purpose” would imply “thought” as already formed; and, therefore, the doctrine must suppose a purpose to precede the conditions of its own possibility.

[187] I use the term “verbs” merely for the sake of brevity and clearness. Of course there cannot have been verbs, strictly so-called, before there were parts of speech of any kind. The more accurate statement is given in the next sentence, and is the one which I desire to be understood hereafter in the short-hand expression “verbs.”

[188] “It must be borne in mind that primitive man did not distinguish between phenomena and volitions, but included everything under the head of actions, not only the involuntary actions of human beings, such as breathing, but also the movements of inanimate things, the rising and setting of the sun, the wind, the flowing of water, and even such purely inanimate phenomena as fire, electricity, &c.; in short, all the changing attributes of things were conceived as voluntary actions” (Sweet, Words, Logic and Grammar, p. 486).

[189] As a matter of fact, and as we shall subsequently see, there is an immense body of purely philological evidence to show that verbs are really a much later product of linguistic growth than either nouns or pronouns. This is proved by their comparative paucity in many existing languages of low development (their place being taken by pronominal appositions, &c.); and also by tracing the origin of many of them to other parts of speech. (See especially Garnett’s Essays, Pritchard on the Celtic Languages, Quart. Rev., Sept. 1876; The Derivation of Words from Pronominal and Prepositional Roots, Proc. Philol. Soc. vol. ii.; and On the Nature and Analysis of the Verb, ibid., vol. iii.) Later on it will be shown that in the really primitive stages of language-growth there is no assignable distinction between any of the parts of speech. Archdeacon Farrar well remarks, “The invention of a verb requires a greater effort of abstraction than that of a noun.... We cannot accept it as even possible that from roots meaning to shine, to be bright, names were formed for sun, moon, stars, &c.... In some places, indeed, Professor Müller appears to hold the correct view, that at first ‘roots’ stood for any and every part of speech, just as the monosyllabic expressions of children do” (Chapters on Language, pp. 196, 197; see, also, some good remarks on the subject by Sir Graves Haughton, Bengali Grammar, p. 108).

[190] “Standst du dabei, als sich der Brust des noch stummen Urmenschen der erste Sprachlaut entrang? und verstandst du ihn? Oder hat man dir die Urwurzeln jener ersten Menschen vor hundert tausend Jahren überliefert? Sind das, was du als Wurzeln hinstellst, und was wirklich Wurzeln sein mögen, auch Wurzeln der Urzeit, unveränderte Reflexlaute? Sind jene deine Wurzeln älter als sechstausend, als zehntausend Jahre? und wie viel mögen sie sich in den früheren Jahrzehntausenden verändert haben? wie mag sich ihre Bedeutung verändert haben?” (Steinthal, Zeits. b. Volkerpysch. u. Sprachwiss., 1867, s. 76).

[191] Supra, p. 68, et seq.

[192] Ursprung der Sprache, s. 74. To the same effect, and from the side of psychology, I may quote Wundt:—“Oft hat man desshalb in der Sprache einen Ubergang vom Abstrakten zum Konkreten zu finden geglaubt, weil dieselbe thatsächlich zunächst umfassendere, dann individuellere Vorstellungen bezeichnet und erst zuletzt wieder die Namen individueller Objekte zu Gemeinnamen stempelt. Aber was am Anfang dieser Reihe liegt ist etwas ganz anderes als was den Schluss derselben bildet: Gemeinnamen sind wirkliche Zeichen für Allgemeinvorstellungen und Begriffe. Jene ersten Vorstellungen, welche das Bewusstsein bildet und die Sprache ausdrückt, sind nicht Allgemeinvorstellungen sondern umfassende Vorstellungen. Beides ist wesentlich aus einander zu halten” (Vorlesungen, &c., ii. 382). The passage then proceeds to discuss the psychology of the subject.

[193] Introduction, &c., ii. 5, 6.

[194] And even as regards this minority (such as “to be,” “to think,” “to do,” &c.), we must remember an important consideration on which Geiger bestows a number of excellent pages. Briefly put, this consideration is that the offspring of words are everywhere proved to have progressively changed their meanings by successive steps and in divergent lines: applying this general law to the case of roots, it follows that the oldest meaning which philology is able to trace as expressed by a root, need not be anywhere near the meaning which attached to its remoter parents: the latter may have been much less conceptual.

[195] Professor Max Müller says in one place, “The Science of Language, by inquiring into the origin of general terms, has established two facts of the highest importance, namely, first, that all terms were originally general; and, secondly, that they could not be anything but general” (Science of Thought, p. 456). Elsewhere, however, he says, “Although during the time when the growth of language becomes historical and most accessible, therefore, to our observation, the tendency certainly is from the general to the special, I cannot resist the conviction that before that time there was a pre-historic period during which language followed an opposite direction. During that period roots, beginning with special meanings, became more and more generalized, and it was only after reaching that stage that they branched off again into special channels” (ibid., pp. 383, 384). Again, in his earlier work on the Science of Language (vol. i., pp. 425-432), he argues in favour of terms having been aboriginally general. It will thus be seen that with reference to this question he is not consistent. Touching the first of his doctrines above quoted, Geiger pertinently observes that against such a conclusion there lies the obvious absurdity, that if a language were to consist exclusively of general terms, it would be ipso facto unintelligible to its own speakers; “for what hope could there be of any mutual understanding with a language comprising only such words as “to bind,” “to sound,” &c.? (Ursprung der Sprache, s. 16). Clearly, Professor Max Müller’s difficulties regarding this subject are quite imaginary, and would disappear if he were to entertain the natural alternative that there is no reason to suppose aboriginal words were exclusively restricted to being either special or general—i.e. generic.

[196] Bunsen, Philosophy of Universal History, ii. 131.

[197] Professor Max Müller in all his works; but it is observable that his opposition to what he calls the “bow-wow and pooh-pooh theory” was more strenuous in his earlier publications than it is in his later.

[198] It is needless to say that innumerable instances might be quoted of this metaphorical change in the meanings of words, even in existing languages,—so much so, indeed, that, as Richter says, all languages are but dictionaries of forgotten metaphors. For example, there is a single Hebrew word of three letters which may bear any one of the following significations:—to mix, to exchange, to stand in place of, to pledge, to interfere, to be familiar, to disappear, to set, to do a thing in the evening, to be sweet, a fly or beetle, an Arabian, a stranger, the weft of cloth, the evening, a willow, and a raven. (See Farrar, Chapters on Language, p. 229. He adds, “Assuming that all these significations are ultimately deducible from one and the same root, we see at once the extent to which metaphor must have been at work.” For further examples of the same principle, see ibid., pp. 234, 251, 252.)

[199] Science of Thought, pp. 317, 318.

[200] Or, as Heyse puts it, many onomatopœias are not “old fruitful roots of language, but modern inventions which remain isolated in language, and are incapable of originating any families of words, because their meaning is too limited and special to admit of a manifold application” (System, s. 92, quoted by Farrar, Chapters on Language, p. 152, who also shows that words of onomatopoetic origin are not invariably sterile. When such origin is not so remote as to have become wholly obscured by a widely connotative extension, it does remain possible to trace its progeny through areas of smaller extension).

[201] “Nichtsdestoweniger bleibt es eine wichtige psychologische Thatsache, dass die Laute einen onomatopoetischen Werth haben, dass wir diesen Werth heute noch fühlen. Nur ist dieses Gefühl nicht sicher genug, um als wissenschaftlicher Beweis zu gelten, wie es denn auch bei den verschiedenen Racen verschieden ist. Die Sprachen der mongolischen Race haben zur Bezeichnung von Naturereignissen viele Onomatopöien, welche wir nicht mitfühlen. Und das ist weder zu verwundern, noch ist es ein Beweis gegen die geistige Einheit des Menschengeschlechtes. Das Gefühl wird ja vielfach durch Associationen der Vorstellungen bestimmt. Andere Associationen aber walten im Kaukasier, andere im Mongolen” (Zeits. b. Volkerpsych. u. Sprachwissen., 1867, s. 76).

[202] Introduction, &c., i., p. 108. He points out that “bilbit, glut-glut, and puls, are all attempts to represent the same sound.”

[203] Chapters on Language, p. 154.

[204] Ueber Namen des Donners, 1855.

[205] Steinthal’s Zeitschrift, &c.

[206] Professor Max Müller has argued that in the Indo-European languages the apparently onomatopoetic words signifying “thunder” are derived from the root tan, to “stretch,” and therefore were not of imitative origin. But Farrar has satisfactorily met this objection, even as regards this one particular case, by showing that even if not originally onomatopoetic, these words afterwards “became so from a feeling of the need that they should be” (Origin of Language, p. 82). See also, Chapters on Language, pp. 178-182; Heyse, System, s. 93; and Wundt, Vorlesungen, &c., ii. 396.

[207] See also Nodier, Dictionnaire des Onomatopées; and Wedgwood, Dictionary of English Etymology.

[208] Probably the explanation of this apparent inconsistency is to be found in the fact that Noiré’s special version of the onomatopoetic theory comes within easy distance of a hypothesis which Max Müller had himself previously sanctioned. This hypothesis, originally propounded by Heyse in his System der Sprachwissenschaft, is that, just as every inorganic substance in nature gives out a particular sound when struck—metal one sound, wood another, stone another, &c.—so different animals have inherent tendencies (or “instincts”) to emit distinctive sounds. In the case of primitive man this inherent tendency was in the direction of articulate speech. For my own part, I do not see that this theory explains anything; and therefore agree with Geiger, who says of it:—“Die Annahme eines jetzt erloschenen Vermögens der Sprachschöpfung und die damit zusammenhängende von einem vollkommenen Urzustande des Menschen ist eine Zuflucht zum Unbegreiflichen, und nicht weit von dem Eingeständnisse entfernt, dass es uns der Natur der Dinge nach für immer unmöglich sei, den wahren Sinn der Urwurzeln zu erkennen und den Vorgang des Sprachursprunges zu erklären. Wir würden mit einer solchen Annahme auf einen mystischen Standpunkt zurückgeführt sein, da doch schon Herder das ‘Gespenst vom Wort Fähigkeit’ bekämpft und gesagt hat: ‘Jch gebe den Menschen nicht gleich plötzlich neue Kräfte, keine sprachschaffende Fähigkeit, wie eine willkürliche qualitas occulta’” (Ursprung der Sprache, s. 24). Sayce, also, well remarks of this hypothesis, “It really rests upon an a priori conception of the origin of speech, which is neither borne out by linguistic facts nor easily intelligible.... Such a theory of language is plainly mystical” (Introduction to Science of Language, vol. i., pp. 66, 67).

[209] Encyclo. Brit., art. “Philology,” vol. xviii., p. 769.

[210] See, for instance, Farrar, Chapters on Language, p. 184.

[211] See above, pp. 138-144.

[212] See above, pp. 121, 122.

[213] See Vorlesungen, &c., ii. 394, 395.

[214] See above, pp. 132-136.

[215] Introduction to the Science of Language, ii. 302.

[216] See above, pp. 138-143.

[217] Der Ursprung der Sprache, s. 31. His own answer to the question is as follows:—“Sind die Wörter Produkte der Natur order der Willkür? Beides und beides nicht. Kein Wort hat naturnothwendig seine bestimmte Bedeutung; insofern sind sie alle willkürlich: aber keines ist zu seiner Bedeutung durch menschliche Willensthätigkeit gekommen” (ibid., s. 113).

[218] Schelling, Einl. in die Philos. d. Mythologie, s. 51.

[219] Anthropologie der Naturvölker, i., 272. See also, F. Müller, Grundriss der Sprachwissenshaft, I. i. 49.

[220] Science of Language, ii. 91, 92.

[221] Grund. d. Sprachwiss., i., 43.

[222] Ægypten, i. 324.

[223] Sayce, Introduction, &c., i. 119, 120.

[224] Science of Thought, 423-440.

[225] Sayce, Introduction, &c., i. 111.

[226] Ibid., i. 113, 114.

[227] Sayce, Introduction, &c., i. 121.

[228] Science of Thought, p. 242.

[229] Garnett, Philolo. Essays, p. 87.

[230] Ibid., 77, 78.

[231] Farrar, Origin of Language, p. 99. The passage continues, “We might have conjectured this from the fact already noticed, that children learn to speak of themselves in the third person—i.e. regard themselves as objects—long before they acquire the power of representing their material selves as the instrument of an abstract entity.” He also alludes to “some admirable remarks to this effect in Mr. F. Whalley Harper’s excellent book on the Power of Greek Tenses;” and recurs to the subject in his more recently published Chapters on Language, p. 62. I could quote other authorities who have commented upon this philological peculiarity of early pronouns; but will only add the following in order to show how the peculiarity in question may continue to survive even in languages still spoken. “The Malay ulun, ‘I,’ is still ‘a man’ in Lampong, and the Kawi ugwang, ‘I,’ cannot be separated from nwang, ‘a man’” (Sayce, Introduction, ii. 26). Lastly, Wundt has pointed out that this impersonal form of speech is distinctive, not only of early pronominal elements, but also of early forms of predication. For instance, “Die ersten Urtheile, die in das Bewusstsein hereinbrechen, subjektlose Urtheile sind, und dass die Prädikate derselben stets eine sinnliche Vorstellung ausdrücken. ‘Es leuchtet es glänzt, es tönt,’—solcher Art sind die Urtheile, die der Mensch zuerst denkt und zuerst ausspricht. Jenes Prädikat, dass sogleich bei der Wahrnehmung eines Gegenstandes sich aufdrängt, wird zur Bezeichnung des Gegenstandes selber. ‘Das Leuchtende, Glänzende, Tönende,’—solcher Art find die Wörter, die ursprünglich in der Sprache gebildet werden” (loc. cit., ii. 377).

[232] Science of Thought, p. 221.

[233] Ibid., p. 554.

[234] Ibid., 241.

[235] Sayce, Introduction, &c., ii. 25; see also to the same effect, Bleek, Ursprung der Sprache, 70-72; F. Müller, Grundriss der Sprachwissenshaft, I., i., s. 40; and Noiré, Logos, p. 186. The chief ground of this scepticism is that it is difficult to conceive how a word could ever have gained a footing if it did not from the first present some independent predicative meaning. But it seems to me that the force of this objection is removed if we remember the sounds which are arbitrarily invented by young children and uneducated deaf-mutes, not to mention the inarticulate clicks of the Bushmen. Moreover, there is nothing inimical to the pronominal theory in the supposition that pronominal elements, even of the most aboriginal kind, were survivals of still more primitive sentence-words—a supposition which would of course remove the difficulty in question. But, as explained in the text, this difficulty, even if it could not be thus met, would really not be one of any importance to my exposition.

[236] Introduction, &c., i. 117.