The Growth Of Language

( Originally Published Early 1900's )

WE have looked at man fashioning the first implements and weapons and houses which were ever made; we now turn aside and ask what were the first of those immaterial instruments, those "aëriform, mystic " legacies which were handed down and gradually improved from the time of the earliest inhabitants of our globe ? Foremost among these, long anterior to the "metallurgic and other manufacturing skill," comes language. With us, to whom thought and speech are so bound together as to be almost inseparable, the idea that language is an instrument which through long ages has been slowly improved to its present perfection, seems difficult of credit. We think of early man having the same ideas and expressing them as readily as we do now; but this is not the case. Not, indeed, that we have any reason to believe that there was a time when man had no language at all, but it seems certain that long ages were necessary before this instrument could be wrought to the fineness in which we find it, and to which in all the languages we are likely to become acquainted with, we are accustomed. A rude iron knife or spear-head seems a simple and natural thing to make. But we know that before it could be made iron had to be discovered, and the art of extracting iron from the ore; and, as a matter of fact, we know that thousands of years passed before the iron spear-head was a possibility, thousands of years spent in slowly improving the weapons of stone, and passing on from them to the weapons of bronze. So, too, with language; simple as it seems at first sight to fit the word on to the idea, and early as we ourselves learn this art, a little thought about what language is will show us how much we owe to the ages which have gone before.

To begin with, then, let us try and consider what language really is. The first thing which it is very important we should realize is that writing has nothing to do with the formation of language; for writing is a comparatively late invention, and came long after languages had gone through the chief stages of their growth, and was never meant for any other purpose than to convey to the eye the idea of sound. Language itself belongs to sound only, and appeals to no other sense than the sense of hearing. This everybody will agree to at once, for it is no more than saying that people who cannot read or write have still a language; and, of course, three or four centuries ago there were comparatively few individuals among all the inhabitants of Europe who could write; even now there are hundreds of languages in the world which have never been committed to writing. The observation would indeed be scarce worth the making, but for the necessity of a precaution against thinking at all of the look of words and not of their sound. And now, say we take any word and ask ourselves what exact relationship it holds with the thought for which it stands. " Book "—no sooner have we pronounced the word than an idea more or less distinct comes into our mind. The thought and the sound seem inseparable, and we cannot remember the time when they were not so. Yet the connection between the thought and the sound is not necessary. In fact, a sound which generally comes connected with one idea may if we are engaged at the time upon a language not our own enter our minds, bringing with it an idea quite unconnected with the first. Share and chère, feel and viel, are examples in point; and the same thing is shown by the numerous sounds in our language which have two or more quite distinct meanings, as for example ware and were, and (with most people) where too; and rite and right, and wright are pronounced precisely alike; therefore there can be no reason why one sound should convey one idea more than another. In other words, the idea and the sound have an arbitrary, not a natural connection. We have been taught: to make the sound "book" for the idea book, but had we been brought up by French parents the sound "livre" would have seemed the natural one to make. So that this wondrous faculty of speech has, like those other faculties f which Carlyle speaks, been handed down on impalpable vehicles of sound through the ages. Never, perhaps, since the time of our first parents has one person from among the countless millions who have been born had to invent for himself a way of expressing his thoughts in words. This is alone a strange thing enough. Impossible as it is to imagine ourselves with-out speech, we may ask the question What should we do if we were ever left in such a predicament ? Should we have any guide in fitting the sound on to the idea ? Share and chère, feel and viel among these unconnected notions is there any reason why we should wed our speech to one rather than another ? Clearly there is no reason. Yet in the case which we imagined of a number of rational beings who had to invent a language for the first time, if they are ever to come to an understanding at all there must be some common impulse which makes more than one choose the same sound for a particular idea. How, for instance, we may ask, was it with our first parents? They have passed on to all their descendants for ever the idea of conveying thought by sound, and all the great changes which have since come into the languages of the world have been gradual and, so to say, natural. But this first invention of the idea of speech is of quite another character.

Here we are brought to the threshold of that impenetrable mystery, " the beginning of things," and here we must pause. We recognize this faculty of speech as a thing mysterious, unaccountable, belonging to that supernatural being, man. There must have been and must be in us a something which causes our mouth to echo the thought of the heart; and originally this echo must have been spontaneous and natural, the same for all alike. Now it is a mere matter of tradition and instruction, the sound we use for the idea; but at first the two must have had some subtle necessary connection, or how could one of our first parents have known or guessed what the other wished to say ? Just as every metal has its peculiar ring, it is as though each impression on the mind rang out its peculiar word from the tongue.' Or was it like the faint tremulous sound which glasses give when music is played near ? The outward object or the inward thought called out a sort of mimicry, a distant echo not like, but yet born of the other on the lips. These earliest sounds may perhaps still sometimes be detected. In the sound flo or flu, which in an immense number of languages stands connected with the idea of flowing and of rivers, do we not recognize some attempt to catch the smooth yet rushing sound of water ? And again, in the sound gra or gri, which is largely associated with the notion of grinding, cutting, or scraping,' there is surely something of this in the guttural harshness of the letters, which make the tongue grate, as it were, against the roof of the mouth. And if we see reason to think that these primitive sounds are not always the closest possible imitations of the things they should express, we need not be surprised, since we notice the same fact with regard to our other ways of conveying ideas. In expressive actions, we only half imitate the motion we intend should be followed. We say "go," and dart out our hand, half to show that the person is to go in the direction we point out, or that he is to keep away from us; and half, again, with the object of expressing to him rapid motion by the quickness of our own movement. So with the first words. The names of animals, for instance, did not attempt to mimic the sound which the animal makes as children call a dog a " bow-wow," and a lamb a " baa " but they were, as we have said, something like echoes upon the tongue of the combined effects the animal produced to sight and hearing.

We may suppose the first created man to have immediately, by this quick, spontaneous faculty of his, found words for every object which met his eye and reached his ear, as Adam is described naming each one of the animals among whom he lived; but even when furnished with a long vocabulary to represent things and belongings of things (which we call attributes or adjectives) and motions of every sort, he would still want a number of other words which could not by any possibility spring directly from the picture formed in his mind. All languages are full of words which by themselves do not and cannot awaken definite thoughts in us. All adverbs, conjunctions, prepositions, particles, belong to this class. To, as, but, when, do not as mere words and taken by themselves convey any meaning; but adjectives, verbs, and substantives, hard, beat, hand, do convey an idea when taken quite alone. So that after the first man had got his list of sounds which were real echoes of his thought, he had to get another series of sounds which were echoes of nothing by themselves, but were useful for joining the others together, showing the connection in which they stood one to another.

This, then, was the second stage of language, the making of what we may call the meaningless words, for they are meaningless when taken quite alone. This second stage was probably a much slower one than the first. The making of meaning sounds might have gone on with any degree of rapidity, provided man started with the word-making faculty within. But there seems reason to believe that every meaningless word has arisen out of some word which once belonged to the real " echo " class; that to, for instance, with, by, and, have descended from older roots (now lost), which if placed alone would once have conveyed as much idea to the mind as pen, ink, paper, do to us. It is impossible to show what all our meaningless words have come from; and we have not even the intention of bringing forward all the reasons on which this opinion is grounded it would be too wearisome. But we may notice one or two instances of how even at the present day this process of changing meaning into meaningless words is still going on. Take first the word even, which we used a moment ago. "Even at the present day." Here even is an adverb, quite meaningless when used alone, at least as an adverb; but if we see it alone it becomes another word, an adjective, a meaning word, bringing before us the idea of two things hanging level. "Even if " is nonsense as an idea with nothing to follow it, but " even weights" is a perfectly clear and definite notion, and each of the separate words even and weights give us clear and definite notions too. It is the same with just, which is both adverb and adjective. " Just as " brings no thought into the mind, but " just man " and just and man separately or together, do. While or whilst are meaningless; but 'a while,' or ' to while ' to loiter are full of meaning. In each case the meaningless word came from the meaning word, and was first used as a sort of metaphor, and then the metaphorical part was lost sight of. The English not is meaningless, and just as much so are the French pas and point in the sense of not ; but in the sense of footstep, or point, they have meaning enough. Originally il ne veut pas meant, metaphorically, " he does not wish a step of your wishes," "he does not go a footstep with you in your wish"; il ne veut point, "he does not go a point with you in your wish." Nowadays all this metaphorical meaning is gone, except to the eye of the grammarian. People recognize that il ne veut point is rather stronger than il ne veut pas, but it never occurs to them to wonder why.

There are so many of these curious examples that one is tempted to go on choosing instances; but we confine ourselves to one more. Our word yes is a word which by itself is quite incapable of calling up a picture in our minds, but the word is or " it is," though the idea it conveys is very abstract, and, so to say, intangible as compared, for instance, with such verbs as move, beat nevertheless belongs to the meaning class. Now it happens that the Latin language used the word est "it is" where we should now use the word " yes"; and it still further happens that our yes' is probably the same as the German es, and was used in the same sense of it is as well. Instead of the meaning-less word " yes " the Romans used the word est " it is," and our own ancestors expressed the same idea by saying "it." Still more. It is well known that French is in the main a descend-ant from the Latin, not the Latin of Rome, but the corrupter Latin which was spoken in Gaul. Now these Latin-speaking Gauls did not, for some reason, say est " it is " for yes, as the Romans did; but they used a pronoun, either ille, "he," or hoc, "this." When, therefore, a Gaul wanted to say 'yes,' he nodded, and said he or else this, meaning " he is so," or " this is so." As it happens the Gauls of the north said ille, and those of the south said hoc, and these words gradually got corrupted into two meaningless words, oui and oc. And, as is well known, the people in the south of France were especially distinguished by using the word oc instead of oui for "yes," so that their " dialect " got to be called the langue d'oc, and this word Languedoc gave the name to a province of France. But long before that time, we may be sure, both the people of the langue d'oui and those of the langue d'oc had forgotten that their words for "yes" once meant "he" and "this."

We can from these instances pretty well guess the way in which the second vocabulary of meaningless words was formed. Man must have begun speaking always in a metaphorical way. Instead of saying "on the rock," or "under the rock," he perhaps said "head of rock " and "foot of rock," and the words he used for head and foot may have got corrupted and changed; so that the older form might remain for the meaningless words on, under, and a newer form come to be used for head and foot. Just as the Frenchman never knows that his oui and il are both sprung from the same Latin ille. Nor, again, does the ordinary German recognize in his gewiss, "certainly," the same word as his past participle gewissen, 'known'; nor the ordinary Englishman reflect that the adverb "ago" is de-rived from " agone," an old past participle of the verb " to go."

When people first began using sounds to express ideas, it would seem that a single sound was used to express each separate idea. Or, putting it differently, we may say that the earliest words were words of one syllable only. As man must have been in want of an enormous number of these simple sounds, he soon began ringing as many changes as possible upon each, so that with every sound went a whole family of others which were very like it, and meant to stand for ideas similar to the idea expressed by the first sound. Now the most important part of every sound, as far as the meaning goes, are the consonants which compose it, and even at the present day, if we keep the consonants of any word the same, and alter the vowel or diphthong, we get a fresh meaning closely connected with the first ( fly, flee, swing, swung, &c.). It was in this way that a great many words arose connected into a class by the consonants remaining unaltered and connected together also by the thread of a common idea. As in "swing," "swung," "swang," we have three different ideas expressing different tenses or times in which the action of swinging took place, and at the same time we have the central idea of swinging connected with the consonants sw-ng: just so, if in some primitive language the consonants of-l expressed the central idea of flowing, " might have stood for the verb to flow," " flu " for the substantive, river, and "fla" for some adjective or attribute common to flowing water, bright perhaps, or soft.

Even with quite modern and cultivated languages which are not, of course, the best for studying the early history of human speech-we may trace the way in which the consonants remain the same, or slightly changed, while the vowels alter, as when we recognize the German knecht in our knight, raum in our room ; or, again, the Italian padre in the French père, tavola in table, &c. Such comparisons as these show us that English and German, French and Italian, are closely connected. But where the connection between languages is very distant, and the farther we have to go back, the more have we to divide our words into their composing syllables so that we are going backwards towards the root-sounds of language, and these as we have said, are single syllables, of which the most constant parts are the consonants. Here our knowledge stops. Of all the changes which were rung upon any particular arrangement of letters, ` of-1 " say, we cannot possibly deter-mine which was the first, " ßô," " flû," " flâ," or any other. What we actually find in any language, the most primitive even, is the existence of these root consonant-sounds expressing some general idea, the idea of flowing, or whatever it may be. Probably, as we have said, this ringing the changes upon a particular sound may have gone on with any degree of rapidity, have been almost simultaneous with the power of speaking itself, which power was, we know, simultaneous with the creation of man. So that we may practically speak of man as starting with these root sounds, which would express not a particular, but a general idea. Sometimes it is not at all easy to trace the connection between the different words which have been formed from one of these general roots. From a root, which in Sanskrit appears in its most ancient form, as má, "to measure," we get words in Greek and Latin which mean " to think," and from the same root-comes our ".man," the person who measures, who compares, i.e., who thinks, also our moon, which means "the measurer," because the moon helps to measure out the time, the months. So, too, our crab is from the word creep, and means the animal that creeps. But why this name should have been given to crab rather than to ant and beetle it is impossible to say.

Thus equipped with his fixed root and the various words formed out of it, man had enough material to begin all the elaborate languages which the world has known. And he continued his work something in this fashion. As generation followed generation the pronunciation f words was changed, as is constantly being done at the present day. Our grand-mothers pronounced " Rome," room, and " brooch," as it was spelt, and not as we pronounce it—" broach." And let it be remembered, before writing was invented, there was nothing but the pronunciation to fix the word, and a new pronunciation was really a new word. When there was no spelling to fix a word, these changes of pronunciation were very rapid and frequent, so that not only would each generation have a different set of words from their fathers, but probably each tribe would be partly unintelligible to its neighboring tribes, just as a Somersetshire man is to a great extent unintelligible to a man out of Yorkshire. The first result of these changes would be the springing up of a number of "meaningless" words. "Head of rock " and " foot of rock " would sink to equivalents of "over" and "under," when of two names for had and foot one became obsolete as a noun, and was only used adverbially. What had originally meant, metaphorically, head of rock and foot of rock might come to be used for over and under the rock when new words had arisen for head and foot, in exactly the same way that the word ago has become a " meaningless" word to the Englishman of to-day.

The next step was the joining of words together. In one way this process may have begun very early. Two " meaning" words, as soon as formed, might be joined together to form a third idea ; just as we have " anthill," which is a different thing from either " ant " or " hill." But there are other ways of joining words more important in the history of language than this. There is the joining on of the " meaningless" words. Although we always put the meaningless qualifying word before the chief word, and say " on the rock," or " under the rock," it is more natural to man, as is shown by all languages, to put the principal idea first, and say " rock on," " rock under," the idea rock being of course the chief idea, the part of the rock, or position in relation to the rock, coming after. So the first step towards grammar was the getting a number of meaningless words, and joining them on to the substantive, " rock," " rock-by," " rock-in," " rock-to," &c. So with the verb. The essential idea in the verb is the action itself, the next idea is the time or person in which the action takes place; nd the natural thing for man to do is to make the words follow that order. The joining process would give us from love, the idea of loving, " love-I," "Love-thou," „love-he,"&c., and for the imperfect " love-was-I,” „love-was-thou,” " love-was-he," " love-was-we," " love-was-ye," "love-was-they," for perfect " love-have-I," " love-have-thou," " love-have-he," &c. Of course, these are merely illustrations, but they make the mode of this early joining process clearer than if we had chosen a language where that process is actually found in its purity, and then translated the forms into their English equivalents.

We have now arrived at a stage in the formation of language where both meaning and meaningless words have been introduced, and where words have been made up out of combinations of the two. We see at once that with regard to meaningless words the use of them would naturally be fixed very much by tradition and custom, and whereas there might be a great many words standing for ant and hill, and therefore a great many ways of saying ant-hill, for the meaningless words, such as under and on, there would probably be only one word. The reason of this is very plain. While, all the separate synonyms for hill, expressed different ways in which it struck the mind, either as being high, or large, or steep, or what not, for under and on being meaningless words not producing any picture in the mind, only one word apiece could very well be used. While under and on were meaning words, meaning, perhaps, as we imagined, head of, or foot of, there would be plenty of synonyms for them; but only one out of all these would be handed down in their meaningless forms. It is important to remember this, because this accounts for all the grammars of all languages. As a matter of fact, every one of those grammatical terminations which we know so well in Latin and Greek, and German, was originally nothing else than meaningless words added on to modify the words which still retained their meaning. We saw before, that it was much more natural for people tò say "rock-on" or "hand-in" than "on the rock" or "in the hand " though, of course, our arrangement of the words seems the most reasonable to us because rock and hand were the most important ideas and came first into the mind, while on, in, &c., were only subsidiary ideas depending upon the important ones. If we stop at rock or hand without adding on and in, we have still got something definite upon which our thoughts can rest, but we could not possibly stop at on and in alone, and have any idea in our minds at all. It is plain enough therefore that, though we say "on the rock," we must have the idea of all the three words in our mind before we begin the phrase, and therefore that our words do not follow the natural order of our ideas; whereas rock-on, hand-in, show the ideas just in the way they come into the mind. It is a fact that all case-endings arose from adding on meaningless words to the end of the word, the noun or pronoun. Das Weib, des Weib-es, dem Weib-e; hom-o, hom-finis, hom-ini : the meanings of case-endings such as these cannot, it is true, be discovered now, for they came into existence long before such languages as German or Latin were spoken, and their meanings were lost sight of in ages which passed before history. But that time when the terminations which are meaningless now had a meaning, and the period of transition between this state and the state of a language which is full of grammatical changes inexplicable to those who use them, form distinct epochs in the history of every language. It is just the same with verb-endings as with the case-endings —ich bin, du bist, really express the I and thou twice over, as the pronouns exist though hidden and lost sight of in the -n and -st of the verb. In the case of verbs, indeed, we may without going far give some idea of how these endings may be detected. We may say at once that Sanskrit, Persian, Greek, Latin, French, Italian, Spanish, German, English, Norse, Gaelic, Welsh, Lithuanian, Russian, and other Slavonic languages are all connected together in various degrees of relationship, all descended from one common ancestor, some being close cousins, and some very distant. Now in Sanskrit " I am " is thus declined:

as-mi I am. 's-mas we are.
a-si thou art. 's-the ye are.
as-ti he is. 's-anti they are.

By separating the root from the ending in this way we may the more easily detect the additions to the root, and their meanings. As is the root expressing the idea of being, existing; mi is from a root meaning I (preserved in me, Greek and Lat. me, moi, m[ich], &c.); so we get as-mi, am-I, or I am. Then we may trace this through a number of languages

connected with the Sanskrit. The important part of as-mi, the consonants, are preserved in the Latin sum, I am, from which, by some further changes, come the French suis, the Italian sono : the same word appears in our a-m, and in the Greek eimi (Doric esmi), I am. Next, coming to the second word, we see one of the s's cut out,-and we get a-si, in which the a is the root, and the si the addition signifying thou. To this addition correspond the final s's in the Latin es, French es—tu es, and the Greek eis (Doric essi). So, again, as-ti, the ti ex-presses he, and this corresponds to the Latin est, French est, the Greek esti, the German ist; in the English the expressive t has been lost. We will not continue the comparison of each word; it will be sufficient if we place side by side the same tense in Sanskrit and in Latin,' and give those who do not know Latin an opportunity of recognizing for themselves the tense in its changed form in French or Italian:

ENGLISH. SANSKRIT. LATIN.

I am as-mi sum.
thou art a-si es.
he is as-ti est.
we are 's-mas sumus.
ye are 's-the estis.
they are 's-anti sunt.

The plural of the added portion we see contains the letters m-s, and if we split these up again we get the separate roots mi and si, so that mas means most literally "I," and " thou," and hence "we." In the second person the Latin has pre-served an older form than the Sanskrit; the proper root consonants for the addition part of the second person plural, combining the ideas thou and he, from which, ye. The third person plural cannot be so easily explained.

It will be seen that in the English almost all likeness to the Sanskrit terminations has been lost. Our verb "to be" is very irregular, being, in fact, a mixture of several distinct verbs. The Saxon had the verb beó contracted from beoma (here we have at any rate the m ending for I), I am, byst, thou art, bydh, he is, and the same appear in the German bin, bist. It is, of course, very difficult to trace the remains of the meaningless additions in such advanced languages as ours, or even as Sanskrit, Latin, and Greek. Nevertheless, the reader may find it not uninteresting to trace in the Latin through most of the tenses of verbs these endings—m, for I, the first person; s, for thou, the second person; t, for he, the third person; m-s, for I and thou, we; st, for ye, thou and he, ye; nt, for they. And the same reader must be content to take on trust the fact that other additions corresponding to different tenses can also be shown or reasonably guessed to have been words expressive by themselves of the idea which belongs to the particular tense; so that where we have such a tense as

amabam I was loving.
amabas thou wast loving.
amabat, &c . he was loving.

we may recognize the meanings of the component parts thus:

ama-ba-m love-was-I.
ama-ba-s love-was-thou.
ama-ba-t love-was-he.

Of course, really to show the way in which these meaningless additions have been made and come to be amalgamated with the root, we should have to take examples from a great number of languages in different stages of development. But we have thought it easier, for mere explanation, to take only such languages as were likely to be familiar to the reader, and even to supplement these examples with imaginary ones—like " rock-on," " love-was-I," &c.—in English. For our object has been at first merely to give an intelligible account of how language has been formed, of the different stages it has passed through, and to leave to a future time the question as to which languages of the globe have passed through all these stages, and which have gone part of their way in the formation of a perfect language. Between the state of a language in which the meaning of all the separate parts of a word are recognized and that state where they are entirely lost, there is an immense gap, that indeed which separates the most from the least advanced languages of the world.

Every language that is now spoken on the globe has gone through the stage of forming meaningless words, and is there-fore possessed of words of both classes. They no longer say " head-of-rock " or " foot-of-rock," but " rock-on " and " rock-under." But there are still known languages in which every syllable is a word and where grammar properly speaking does not exist. For grammar, if we come to consider it exactly, is the explanation of the meaning of those added syllables or letters which have lost all natural meaning of their own. If each part of the word were as clear and as intelligible as " rock-on " we should have no need of a grammar at all. A language of this sort is called a monosyllabic language, not because the people only speak in monosyllables, but because each word, however compound, can be split up into mono-syllables which have a distinctly recognizable meaning. "Ant-hill-on " or " love-was-I," are like the words of such a language.

The next stage of growth is where the meaning of the added parts has been lost sight of, except when it is connected with the word which it modifies ; but where the essential word has a distinct idea by itself, and without the help of any addition. Suppose, for instance, through ages of change the "was-I" in our imaginary example got corrupted into "wasi," where wasi had no meaning by itself, but was used to express the first person of the past tense. The first person past of love would be "love-wasi," of move "move wasi," and so on, "wasi" no longer having a meaning by it-self, but "love " and " move " by themselves being perfectly understandable. A language in this stage is said to be in the agglutinative stage, because certain grammatical endings (like " wasi") are merely as it were glued on to a root to change its meaning, while the root itself remains quite unaffected, and means neither less nor more than it did before.

But, as ages pass on, the root and the addition get so closely combined that neither of them alone has a distinct meaning, and the language arrives at its third stage. It is not difficult to find examples of a language in this condition, for such is the case with all the languages by which we are surrounded. All the tongues which the majority of us are likely to study, almost all those which have any literature at all, have arrived at this last stage, which is called the inflexional. For instance, though we were able to separate " asmi" into two parts—" as " and " mi "—one expressing the idea of being, the other the person "I," this distinction is the refinement of the grammarian, and would never have been recognized by an ordinary speaker of Sanskrit, for whom " asmi " simply meant "I am," without distinction of parts. In our " am" the grammarian recognizes that the " a " expresses existence, and the "m " expresses I; but so completely have we lost sight of this, that we repeat the "I" before the verb. Just the same in Latin. No Roman could have recognized in the "s " of sum " am " and in the "m " " I "; for him sum meant simply and purely " I am." " I " was no more separable in his eyes than the French êtes (Latin estis) in vous êtes, is separable into a root " es," contracted in the French into " ê," meaning are, and an addition " tes " signifying you. This, then, is the last stage upon which language enters. It is called inflexional, because the different grammatical changes are not now denoted by a mere addition to an intelligible word, but by a change in the word itself. The root may indeed remain and be recognizable in its purity, but very frequently it is unrecognizable, so that the different case or tense endings can no longer be looked upon as additions, but as changes. Take almost any Latin substantive, and we see this: homo, a man, the genitive is formed, not by adding something to homo, but by changing homo into hominis, or, if we please, adding something to the root hom—which has in itself no meaning.

Thus, to recapitulate, we discover first two stages which language went through before it presents itself in any form known to us: what we call the meaning words came into existence, and then out of these were gradually formed the meaningless words.

These stages were in the main passed through before any known language came into existence, for there is no tongue which is not composed in part of words from the meaningless class: though at the same time it is a process which is still going on, as where even and just the adjectives become even and just the adverbs, or where the French substantives pas and point take a like change of meaning. Then after the meaningless words have been acquired come the three other stages which go to the making of the grammar of a language, stages which can be traced in actual living languages, and which have been called the monosyllabic, the agglutinative, and the inflexional stage. With the last of these the history of the growth of language comes to an end. It happens indeed, sometimes, that a language which has arrived at the inflexional stage may in time come to drop nearly all its inflexions. This has been the case with English and French. Both are descended from languages which had elaborate grammars the Saxon and the Latin; but both, from a mixture of languages and other causes, have come to drop almost all their grammatical forms. We show our grammar only in a few changes in our ordinary verbs the past tense and the past participle, use, used

buy, bought, &c. ; in other variations in our auxiliary verb, and by changes in our pronouns I, me, ye, you, who, whom, &c. ; and by the "'s" and "s" of the possessive case and of the plural, and by the comparison of adjectives. The French preserve their grammar to some extent in their pro-nouns, their adjectives, and in their verbs. But these are cases of decay, and do not find any place in the history of the growth of language.

From this we pass on to examine where the growth of language has been fully achieved, where it has remained only stunted and imperfect.