Chapter V: Part II (1)

ON CURTIUS’ CHRONOLOGY OF THE INDO-GERMANIC LANGUAGES.

In a former Lecture on the “Stratification of Language” I ventured to assert that wherever _inflection_ has yielded to a rational analysis, it has invariably been recognized as the result of a previous _combination_, and wherever _combination_ has been traced back to an earlier stage, that earlier stage has been simply _juxtaposition_.

Professor Pott in his “Etymologische Forschungen” (1871, p. 16), a work which worthily holds its place by the side of Bopp’s “Comparative Grammar,” questions the correctness of that statement; but in doing so he seems to me to have overlooked the restrictions which I myself had introduced, in order to avoid the danger of committing myself to what might seem too general a statement. I did not say that every form of inflection had been proved to spring from a previous combination, but I spoke of those cases only where we have succeeded in a rational analysis of inflectional forms, and it was in these that I maintained that inflection had always been found to be the result of previous combination. What is the object of the analysis of grammatical inflections, or of Comparative Grammar in general, if not to find out what terminations originally were, before they had assumed a purely formal character? If we take the French adverb _sincèrement_, sincerely, and trace it back to the Latin _sincerâ mente_, we have for a second time the three stages of juxtaposition, combination, and, to a certain extent, inflection, repeated before our eyes. I say, inflection, for _ment_, though originally an independent word, soon becomes a mere adverbial suffix, the speakers so little thinking of its original purport, that we may say of a stone that it falls _lourdement_, heavily, without wishing to imply that it falls _luridâ mente_, with a heavy, lit., with a lurid mind.

If we take the nom. sing. of a noun in Sanskrit, Greek, or Latin, we find that masculine nouns end frequently in _s_. We have for instance, Sk. +veśa-s+, Gr. οἶκο-ς, Lat. _vîcu-s_. These three words are identical in their termination, in their base, and in their root. The root is the Sk. +viś+, to settle down, to enter upon or into a thing. This root, without undergoing any further change, may answer the purpose both of a verbal and a nominal base. In the precative, for instance, we have +viś-yâ-t+, he may enter, which yields to a rational analysis into +viś+, the root +yâ+, to go, and the old pronominal stem of the third person, _t_, he. We reduplicate the root, and we get the perfect +vi-viś-us+, they have entered. Here I can understand that objections might be raised against accepting us as a mere phonetic corruption of +ant+ and +anti+; but if, as in Greek, we find as the termination of the third pers. plur. of the perfect ᾶσι, we know that this is a merely phonetic change of the original +anti+,[29] and this +anti+ has been traced back by Pott himself (whether rightly or wrongly, we need not here inquire) to the pronominal stems +ana+, that, and +ti+, he. These two stems, when joined together, become +anti+,[30] meaning _those_ and _he_, and are gradually reduced to ᾶσι, and in Sanskrit to +us+ for +ant+. What we call reduplication has likewise been traced back by Pott himself to an original repetition of the whole root, so that +vi-viś+ stands for an original or intentional +viś-viś+; thus showing again the succession of the three stages, juxtaposition, +viś-viś+, combination +vi-viś+, inflection, the same, +vi-viś+, though liable to further phonetic modification.

Used as a nominal base the same root +viś+ appears, without any change, in the nom. plur. +viś-as+, the settlers, the clans, the people. Now here again Professor Pott himself has endeavored to explain the inflection +as+ by tracing it back to the pronominal base +as+, in +asau+, _ille_. He therefore takes the plural +viś-as+ as a compound, meaning “man and that;” that is to say, he traces the inflection back to a combinatory origin.

By raising the simple base +viś+ to +viśa+, we arrive at new verbal forms, such as +viś-â-mi+, I enter, +viś-a-si+, thou enterest, +viś-a-ti+, he enters. In all these inflectional forms, the antecedent combinatory stage is still more or less visible, for +mi+, +si+, +ti+, whatever their exact history may have been, are clearly varieties of the pronominal bases of the first, second, and third persons, +ma+, +tva+, +ta+.

Lastly, by raising +viś+ to +veśa+, we arrive at a new nominal base, and by adding to it the stem of a demonstrative pronoun _s_, we form the so-called nom. sing. +veśa-s+, οἶκο-ς, _vicu-s_, from which we started, meaning originally house-here, this house, the house.

In all this Professor Pott would fully agree, but where he would differ, would be when we proceed to generalize, and to lay it down as an axiom, that all inflectional forms _must_ have had the same combinatory origin. He may be right in thus guarding against too hasty generalization, to which we are but too prone in all inductive sciences. I am well aware that there are many inflections which have not yielded, as yet, to any rational analysis, but, with that reservation, I thought, and I still think, it right to say that, until some other process of forming those inflections has been pointed out, inflection may be considered as the invariable result of combination.

It is impossible in writing, always to repeat such qualifications and reservations. They must be taken as understood. Take for instance the augment in Greek and Sanskrit. Some scholars have explained it as a negative particle, others as a demonstrative pronoun; others, again, took it as a mere symbol of differentiation. If the last explanation could be established by more general analogies, then, no doubt, we should have here an inflection, that cannot be referred to combination. Again, it would be difficult to say, what independent element was added to the pronoun +sa+, he, in order to make it sâ, she. This, too, may, for all we know, be a case of phonetic symbolism, and, if so, it should be treated on its own merits. The lengthening of the vowel in the subjunctive mood was formerly represented by Professor Curtius as a symbolic expression of hesitation, but he has lately recalled that explanation as untenable. I pointed out that when in Hebrew we meet with such forms as _Piel_ and _Pual_, _Hiphil_ and _Hophal_, we feel tempted to admit formative agencies, different from mere juxtaposition and combination. But before we admit this purely phonetic symbolism, we should bear in mind that the changes of _bruder_, brother, into _brüder_, brethren, of _Ich weiss_, I know, into _wir wissen_, we know, which seem at first sight purely phonetic, have after all been proved to be the indirect result of juxtaposition and combination, so that we ought to be extremely careful and first exhaust every possible rational explanation, before we have recourse to phonetic symbolism as an element in the production of inflection forms.

The chief object, however, of my lecture on the “Stratification of Language” was not so much to show that inflection everywhere presupposes combination, and combination juxtaposition, but rather to call attention to a fact that had not been noticed before, viz.: that there is hardly any language, which is not at the same time _isolating_, _combinatory_, and _inflectional_.

It had been the custom in classifying languages morphologically to represent some languages, for instance Chinese, as _isolating_; others, such as Turkish or Finnish, as _combinatory_; others, such as Sanskrit or Hebrew, as _inflectional_. Without contesting the value of this classification for certain purposes, I pointed out that even Chinese, the very type of the isolating class, is not free from combinatory forms, and that the more highly developed among the combinatory languages, such as Hungarian, Finnish, Tamil, etc., show the clearest traces of incipient inflection. “The difficulty is not,” as I said, “to show the transition of one stratum of speech into another, but rather to draw a sharp line between the different strata. The same, difficulty was felt in Geology, and led Sir Charles Lyell to invent such pliant names as _Eocene_, _Meiocene_, and _Pleiocene_, names which indicate a mere dawn, a minority, or a majority of new formations, but do not draw a fast and hard line, cutting off one stratum from the other. Natural growth and even merely mechanical accumulation and accretion, here as elsewhere, are so minute and almost imperceptible that they defy all strict scientific terminology, and force upon us the lesson that we must be satisfied with an approximate accuracy.”

Holding these opinions, and having established them by an amount of evidence which, though it might easily be increased, seemed to me sufficient, I did not think it safe to assign to the three stages in the history of the Aryan languages, the _juxtapositional_, the _combinatory_, and the _inflectional_, a strictly successive character, still less to admit in the growth of the Aryan languages a number of definite stages, which should be sharply separated from each other, and assume an almost chronological character. I fully admit that wherever _inflectional_ forms in the Aryan languages have yielded to a rational analysis, we see that they are preceded chronologically by _combinatory_ formations; nor should I deny for one moment that _combinatory_ forms presuppose an antecedent, and therefore chronologically more ancient stage of mere juxtaposition. What I doubt is whether, as soon as combination sets in, juxtaposition ceases, and whether the first appearance of inflection puts an end to the continued working of combination.

It seems to me, even if we argue only on _à priori_ grounds, that there must have been at least a period of transition during which both principles were at work together, and I hardly can understand what certain scholars mean if they represent the principle of inflection as a sudden psychological change which, as soon as it has taken place, makes a return to combination altogether impossible. If, instead of arguing _à priori_, we look the facts of language in the face, we cannot help seeing that, even after that period during which it is supposed that the United Aryan language had attained its full development, I mean at a time when Sanskrit, Greek, and Latin had become completely separated, as so many national dialects, each with its own fully developed inflectional grammar, the power of combination was by no means extinct. The free power of composition, which is so manifest in Sanskrit and Greek, testifies to the continued working of combination in strictly historical times. I see no real distinction between the transition of _Néa pólis_, i.e., new town, into _Neápolis_, and into _Naples_, and the most primitive combination in Chinese, and I maintain that as long as a language retains that unbounded faculty of composition, which we see in Sanskrit, in Greek, and in German, the growth of new inflectional forms from combinatory germs must be admitted as possible. Forms such as the passive aorist in Greek, ἐτέθην, or the weak preterite in Gothic _nas-i-da_, _nas-i-dédjau_, need not have been formed before the Aryan family broke up into national languages; and forms such as Italian _meco_, _fratelmo_, or the future _avro_, I shall have, though not exactly of the same workmanship, show at all events that analogous powers are at work even in the latest periods of linguistic growth.

Holding these opinions, which, as far as I know, have never been controverted, I ought perhaps, when I came to publish the preceding Lecture, to have defended my position against the powerful arguments advanced in the meantime by my old friend, Professor G. Curtius, in support of a diametrically opposite opinion in his classical essay, “On the Chronology of the Indo-Germanic Languages,” published in 1867, new edition, 1873. While I had endeavored to show that juxtaposition, combination, and inflection, though following each other in succession, do not represent chronological periods, but represent phases, strongly developed, it is true, in certain languages, but extending their influence far beyond the limits commonly assigned to them, Professor Curtius tried to establish the chronological character not only of these three, but of four other phases or periods in the history of Aryan speech. Confining himself to what he considers the undivided Aryan language to have been, before it was broken up into national dialects, such as Sanskrit, Greek, and Latin, he proceeds to subdivide the antecedent period of its growth into _seven_ definite stages, each marked by a definite character, and each representing a sum of years in the chronology of the Aryan language. As I had found it difficult to treat Chinese as entirely _juxtapositional_, or Turkish as entirely _combinatory_, or Sanskrit as entirely _inflectional_, it was perhaps not to be wondered at that not even the persuasive pleading of my learned friend could convince me of the truth of the more minute chronological division proposed by him in his learned essay. But it would hardly have been fair if, on the present occasion, I had reprinted my “Rede Lecture” without explaining why I had altered nothing in my theory of linguistic growth, why I retained these three phases and no more, and why I treated even these, not as chronological periods, in the strict sense of the word, but as preponderating tendencies, giving an individual character to certain classes of language, without being totally absent in others. Professor Curtius is one of the few scholars with whom it is pleasant to differ. He has shown again and again that what he cares for is truth, not victory, and when he has defended his position against attacks not always courteous, he has invariably done so, not with hard words, but with hard arguments. I therefore feel no hesitation in stating plainly to him where his theories seem to me either not fully supported, or even contradicted by the facts of language, and I trust that this free exchange of ideas, though in public, will be as pleasant as our conversations in private used to be, now more than thirty years ago.

Let us begin with the _First Period_, which Professor Curtius calls the _Root-Period_. There must have been, as I tried to explain before, a period for the Aryan languages, during which they stood on a level with Chinese, using nothing but roots, or radical words, without having reduced any of them to a purely formal character, without having gone through the process of changing what Chinese grammarians call _full_ words into _empty_ words. I have always held, that to speak of roots as mere abstractions, as the result of grammatical theory, is self-contradictory. Roots which never had any real or historical existence may have been invented both in modern and ancient collections or +Dhâtupâṭhas+; but that is simply the fault of our etymological analysis, and in no way affects the fact, that the Aryan, like all other languages we know, began with roots. We may doubt the legitimacy of certain chemical elements, but not the reality of chemical elements in general. Language, in the sense in which we use the word, begins with roots, which are not only the ultimate facts for the Science of Language, but real facts in the history of human speech. To deny their historical reality would be tantamount to denying cause and effect.

Logically, no doubt, it is possible to distinguish between a root as a mere postulate, and a root used as an actual word. That distinction has been carefully elaborated by Indian grammarians and philosophers, but it does in no way concern us in purely historical researches. What I mean by a root used in real language is this: when we analyze a cluster of Sanskrit words, such as +yodha-s+, a fighter, +yodhaka-s+, a fighter, +yoddhâ+, a fighter, +yodhana-m+, fighting, +yuddhi-s+, a fight, +yuyutsu-s+, wishing to fight, +â-yudha-m+, a weapon, we easily see that they presuppose an element +yudh+, to fight, and that they are all derived from that element by well-known grammatical suffixes. Now is this +yudh+, which we call the root of all these words, a mere abstraction? Far from it. We find it as +yudh+ used in the Veda either as a nominal or as a verbal base, according to suffixes by which it is followed. Thus +yudh+ by itself would be a fighter, only that +dh+ when final, has to be changed into t. We have +goshu-yúdh-am+, an accusative, the fighter among cows. In the plural we have +yúdh-as+, fighters; in the locative +yudh-i+, in the fight; in the instrumental, +yudh-â+, with the weapon. That is to say, we find that as a nominal base, +yudh+, without any determinative suffixes, may express fighting, the place of fighting, the instrument of fighting, and a fighter. If our grammatical analysis is right, we should have +yudh+ as a nominal base in +yúdh-ya-ti+, lit. he goes to fighting, +yudh-yá-te+, pass.; +(a)-yut-smahi+, aor., either we were to fight, or we were fighters; +yú-yut-sa-ti+, he is to fight-fight; +yudh-ya-s+, to be fought (p. 94), etc. As a verbal base we find +yudh+, for instance, or +yu-yudh-e+, I have fought; in +a-yud-dha+, for +a-yudh-ta+, he fought. In the other Aryan languages this root has left hardly any traces; yet the Greek ὑσμῖν, and ὑσμίνη would be impossible without the root +yudh+.

The only difference between Chinese and these Sanskrit forms which we have just examined, is that while in Chinese such a form as +yudh-i+, in the battle, would have for its last element a word clearly meaning middle, and having an independent accent, Sanskrit has lost the consciousness of the original material meaning of the _i_ of the locative, and uses it traditionally as an empty word, as a formal element, as a mere termination.

I also agree with Curtius that during the earliest stage, not of Sanskrit, but of Aryan speech in general, we have to admit two classes of roots, the _predicative_ and _demonstrative_, and that what we now call the plural of +yudh+, +yudh-as+, fighters, was, or may have been, originally a compound consisting of the predicative root +yudh+, and the demonstrative root, +as+ or +sa+, possibly repeated twice, meaning “fight-he-he,” or “fight-there-there,” _i.e._, fighters.

There is another point with regard to the character of this earliest radical stage of the Aryan language, on which formally I should have agreed with Curtius, but where now I begin to feel more doubtful,--I mean the necessarily monosyllabic form of all original roots. There is, no doubt, much to be said for this view. We always like to begin with what is simple. We imagine, as it has been said, that “the simple idea must break forth, like lightning, in a simple body of sound, to be perceived in one single moment.” But, on the other hand, the simple, so far as it is the general, is frequently, to us at least, the last result of repeated complex conceptions, and therefore there is at all events no _à priori_ argument against treating the simplest roots as the latest, rather than the earliest products of language. Languages in a low state of development are rich in words expressive of the most minute differences, they are poor in general expressions, a fact which ought to be taken into account as an important qualification of a remark made by Curtius that language supplies necessaries first, luxuries afterwards (p. 32). I quote the following excellent remarks from Mr. Sayce’s “Principles of comparative Philology” (p. 208): “Among modern savages the individual objects of sense have names enough, while general terms are very rare. The Mohicans have words for cutting various objects, but none to signify cutting simple.”[31] In taking this view we certainly are better able to explain the actual forms of the Aryan roots, viz., by _elimination_, rather than by _composition_. If we look for instance, as I did myself formerly, on such roots as +yudh+, +yuj+, and +yauṭ+, as developed from the simpler root +yu+, or on +mardh+, +marg+, +mark+, +marp+, +mard+, +smar+, as developed from +mar+, then we are bound to account for the modificatory elements, such as _dh_, _g_, _k_, _p_, _d_, _s_, _n_, _t_, _r_, as remnants of other roots, whether predicative or demonstrative. Thus Curtius compares +tar+ or +tra+, with +tras+, +tram+, +trak+, +trap+; +tri+ and +tru+ with +trup+, +trib+, taking the final consonants as modificatory letters. But what are these modificatory letters? Every attempt to account for them has failed. If it could be proved that these modificatory elements, which Curtius calls _Determinatives_, produced always the _same_ modification of meaning, they might then be classed with the verbal suffixes which change simple verbs into causative, desiderative, or intensive verbs. But this is not the case. On the other hand, it would be perfectly intelligible that such roots as +mark+, +marg+, +mard+, +mardh+, expressing different kinds of crushing, became fixed side by side, that by a process of elimination, their distinguishing features were gradually removed, and the root +mar+ left as the simplest form, expressive of the most general meaning. Without entering here on that process of mutual friction by which I believe that the development of roots can best be explained, we may say at least so much, that whatever process will account for the root +yu+, will likewise account for the root +yuj+, nay, that roots like +mark+ or +mard+ are more graphic, expressive, and more easily intelligible than the root mar.

However, if this view of the origin of roots has to be adopted, it need not altogether exclude the other view. In the process of simplification, certain final letters may have become typical, may have seemed invested with a certain function or determinative power, and may therefore have been added independently to other roots, by that powerful imitative tendency which asserts itself again and again through the whole working of language. But however that may be, the sharp line of distinction which Curtius draws between the First Period, represented by simple, and the Second Period represented by derivative roots, seems certainly no longer tenable, least of all as dividing _chronologically_ two distinct periods in the growth of language.

When we approach the Third Period, it might seem that here, at least, there could be no difference of opinion between Professor Curtius and myself. That Third Period represents simply what I called the first setting in of _combination_, following after the _isolating_ stage. Curtius calls it the _primary verbal period_, and ascribes to it the origin of such combinatory forms as +dấ-ma+, give-I, +dâ-tva+, give-thou, +dấ-ta+, give-he; +dâ-ma-tvi+, give-we, +dâ-tva-tvi+, give-you, +dâ-(a)nti+, give-they. These verbal forms he considers as much earlier than any attempts at declension in nouns. No one who has read Curtius’ arguments in support of this chronological arrangement would deny their extreme plausibility; but there are grave difficulties which made me hesitate in adopting this hypothetical framework of linguistic chronology. I shall only mention one, which seemed to me insurmountable. We know that during what we called the First Radical Period the sway of phonetic laws was already so firmly established, that, from that period onward to the present day, we can say, with perfect certainty, which phonetic changes are possible, and which are not. It is through these phonetic laws that the most distant past in the history of the Aryan language is connected with the present. It is on them that the whole science of etymology is founded. Only because a certain root has a tenuis, a media, an aspirate, or a sibilant, is it possible to keep it distinct from other roots. If t and s could be interchanged, then the root +tar+, to cross, would not be distinct from the root +sar+, to go. If d and +dh+ could vary, then +dar+, to tear, would run together with +dhar+, to hold. These phonetic distinctions were firmly established in the radical period, and continue to be maintained, both in the undivided Aryan speech, and in the divided national dialects, such as Sanskrit, Greek, Latin, and Gothic. How then can we allow an intervening period, during which +ma-tvi+, could become +masi+, +tva-tvi+, +thas+, and the same +tva-tvi+ appear also as +sai+? Such changes, always most startling, may have been possible in earlier periods; but when phonetic order had once been established, as it was in what Curtius calls his first and second periods, to admit them as possible, would be, as far as I can judge, to admit a complete anachronism. Of two things one; either we must altogether surrender those chaotic changes which are required for identifying Sanskrit e with Greek μαι, and Greek μαι with +mâ-ma+, etc., or we must throw them back to a period anterior to the final settlement of the Aryan roots.

I now proceed to point out a second difficulty. If Curtius uses these same personal terminations, +masi+, +tvasi+, and +anti+, as proof positive that they must have been compounded out of +ma+ + +tva+, and +tva-tva+, before there were any case terminations, I do not think his argument is quite stringent. Curtius says: “If plural suffixes had existed before the coining of these terminations, we should expect them here, as well as in the noun” (p. 33). But the plural of the pronoun _I_ could never have been formed by a plural suffix, like the plural of _horse_. _I_ admits of no plural, as little as _thou_, and hence the plural of these very pronouns in the Aryan language is not formed by the mere addition of a plural termination, but by a new base. We say _I_, but _we_; _thou_ but _you_, and so through all the Aryan languages. According to Curtius himself, +masi+, the termination of the plural, is not formed by repeating +ma+, by saying I and I, but by +ma+ and +tva+, I and thou, the most primitive way, he thinks, of expressing _we_. The termination of the second person plural might be expressed by repeating _thou_. “You did it,” might have been rendered by “thou and thou did it;” but hardly by treating _thou_ like a noun, and adding to it a plural termination. The absence of plural terminations, therefore at the end of the personal suffixes of the verbs, does not prove, as far as I can see, that plurals of nouns were unknown when the first, second, and third persons plural of the Aryan verbs were called into existence.

Again, if Curtius says, that “what language has once learnt, it does not forget again, and that therefore if the plural had once found expression in nouns, the verb would have claimed the same distinction,” is true, no doubt, in many cases, but not so generally true as to supply a safe footing for a deductive argument. In so late a formation as the periphrastic future in Sanskrit, we say +dâtâ-smaḥ+, as it were _dator sumus_, not +dâtâraḥ smaḥ+; and in the second person plural of the passive in Latin _amamini_, though the plural is marked, the gender is always disregarded.

Further, even if we admit with Bopp and Curtius that the terminations of the medium are composed of two pronouns, that the _ta_ of the third person singular stands for _ta-ti_, to-him-he, that καλύπτεται in fact meant originally hide-himself-he, it does not follow that in such a compound one pronominal element should have taken the termination of the accusative, any more than the other takes the termination of the nominative. The first element in every composition takes necessarily its Pada or thematic form; the second or final element has suffered so much, according to Bopp’s own explanation, that nothing would be easier to explain than the disappearance of a final consonant, if it had existed. The absence of case-terminations in such compounds cannot therefore be used as proof of the non-existence of case-terminations at a time when the medial and other personal endings took their origin. On the contrary, these terminations seem to me to indicate, though I do not say to prove, that the conception of a subjective, as distinct from an objective case, had been fully realized by those who framed them. I do not myself venture to speak very positively of such minute processes of analysis as that which discovers in the Sk. first pers. sing. ind. pres. of the middle, tude, I strike, an original +tuda+ + a + i, +tuda+ + +ma+ + i, +tuda+ + +ma+ + +mi+, tuda + +mâ+ + +ma+, but admitting that the middle was formed in that way, and that it meant originally _strike-to-me-I_, then surely we have in the first +mâ+ an oblique case, and in the compound itself the clearest indication that the distinction between a nominative and an oblique case, whether dative or accusative, was no longer a mystery. Anyhow, and this is the real point at issue, the presence of such compounds as +mâ-ma+, to-me-I, is in no way a proof that at the time of their formation people could not distinguish between +yudh (s)+, nom., a fighter, and +yudh (am)+, acc., a fighter; and we must wait for more irrefragable evidence before admitting, what would under all circumstances be a most startling conclusion, namely, that the Aryan language was spoken for a long time without case-terminations, but with a complete set of personal terminations, both in the singular and the plural. For though it is quite true that the want of cases could only be felt in a sentence, the same seems to me to apply to personal terminations of the verb. The one, in most languages we know, implies the other, and the very question whether conjugation or declension came first is one of those dangerous questions which take something for granted which has never been proved.

During all this time, according to Curtius, our Aryan language would have consisted of nothing but roots, used for nominal and verbal purposes, but without any purely derivative suffixes, whether verbal or nominal, and without declension. The only advance, in fact, made beyond the purely Chinese standard, would have consisted in a few combinations of personal pronouns with verbal stems, which combinations assumed rapidly a typical character, and led to the formation of a skeleton of conjugation, containing a _present_, _an aorist_ with an augment, and a _reduplicated perfect_. Why, during the same period, nominal bases should not have assumed at least some case-terminations, does not appear; and it certainly seems strange that people who could say +vak-ti+, speak-he, +vak-anti+, speak-this-he, should not have been able to say +vâk-s+, whether in the sense of speak-there, _i.e._, speech or speak-there, _i.e._, speaker.

The next step which, according to Curtius, the Aryan language had to make, in order to emerge from its purely radical phase, was the creation of bases, both verbal and nominal, by the addition of verbal and nominal suffixes to roots, both primary and secondary. Curtius calls this fourth the Period of the _Formation of Themes_. The suffixes are very numerous, and it is by them that the Aryan languages have been able to make their limited number of roots supply the vast materials of their dictionary. From +bhar+, to carry, they formed +bhar-a+, a carrier, but sometimes also a burden. In addition to +bhar-ti+, carry-he, they formed +bhara-ti+, meaning possibly carrying-he. The growth of these early themes may have been very luxuriant, and, as Professor Curtius expresses it, chiefly _paraschematic_. It may have been left to a later age to assign to that large number of possible synonyms more definite meanings. Thus from φέρω, I carry, we have φορά, the act of carrying, used also in the sense of _impetus_ (being carried away), and of _provectus_, i.e., what is brought in. Φορός means carrying, but also violent, and lucrative; φέρετρον, an instrument of carrying, means a bier; φαρέτρα, a quiver, for carrying arrows. Φορμός comes to mean a basket; φόρτος, a burden; φορός, tribute.

All this is perfectly intelligible, both with regard to nominal and verbal themes. Curtius admits four kinds of verbal themes as the outcome of his Fourth Period. He had assigned to his Third Period the simple verbal themes ἐσ-τί, and the reduplicated themes such as δίδω-σι. To these were added, in the Fourth Period, the following four secondary themes:--

(1) πλέκ-ε-(τ)-ι Sanskrit +lipa-ti+
(2) ἀλείφ-ε-(τ)-ι „ +laipa-ti+
(3) δείκ-νυ-σι „ +lip-nau-ti+
(4) δάμ-νη-σι „ +lip-nâ-ti+.

He also explains the formation of the subjunctive in analogy with bases such as +lipa-ti+, as derived from +lip-ti+.

Some scholars would probably feel inclined to add one or two of the more primitive verbal themes, such as

limpa-ti _rumpo_
limpana-ti λαμβάνε(τ)ι

but all would probably agree with Curtius in placing the formation of these themes, both verbal and nominal, between the radical and the latest inflectional period. A point, however, on which there would probably be considerable difference of opinion is this, whether it is credible, that at a time when so many nominal themes were formed,--for Curtius ascribes to this Fourth Period the formation of such nominal bases as

λόγ-ο, intellect, = +lipa-ti+
λοίπ-ο, left, = +laipa-ti+
λιγ-νύ, smoke, = +lip-nau-ti+
δάφ-νη, laurel, = +lip-nâ-ti+--

the simplest nominal compounds, which we now call nominative and accusative, singular and plural, were still unknown; that people could say +dhṛsh-nu-más+, we dare, but not +dhṛsh-ṇú-s+, daring-he; that they had an imperative, +dhṛshṇuhí+, dare, but not a vocative, +dhṛshṇo+? Curtius strongly holds to that opinion, but with regard to this period too, he does not seem to me to establish it by a regular and complete argument. Some arguments which he refers to occasionally have been answered before. Another, which he brings in incidentally, when discussing the abbreviation of certain suffixes, can hardly be said to carry conviction. After tracing the suffixes +ant+ and +tar+ back to what he supposes to have been their more primitive forms, +an-ta+ and +ta-ra+, he remarks that the dropping of the final vowel would hardly be conceivable at a time when there existed case-terminations. Still this dropping of the vowel is very common, in late historical times, in Latin, for instance, and other Italian dialects, where it causes frequent confusion and heteroclitism.[32] Thus the Augustan _innocua_ was shortened in common pronunciation to _innoca_, and this dwindles down in Christian inscriptions to _innox_. In Greek, too, διάκτορος is older than διάκτωρ; φύλακος older than φύλαξ.

Nor can it be admitted that the nominal suffixes have suffered less from phonetic corruption than the terminations of the verb, and that therefore they must belong to a more modern period (pp. 39, 40). In spite of all the changes which the personal terminations are supposed to have undergone, their connection with the personal pronouns has always been apparent, while the tracing back of the nominal suffixes, and, still more, of the case-terminations to their typical elements, forms still one of the greatest difficulties of comparative grammarians.[33]

Professor Curtius is so much impressed with the later origin of declension that he establishes one more period, the fifth, to which he assigns the growth of all compound verbal forms, compound stems, compound tenses, and compound moods, before he allows the first beginnings of declension, and the formation even of such simple forms as the nominative and accusative. It is difficult, no doubt, to disprove such an opinion by facts or dates, because there are none to be found on either side: but we have a right to expect very strong arguments indeed, before we can admit that at a time when an aorist, like ἔδεικ-σα, Sanskrit +a-dik-sha-t+ was possible, that is to say, at a time when the verb +as+, which meant originally to breathe, had by constant use been reduced to the meaning of being; at a time when that verb, as a mere auxiliary, was joined to a verbal base in order to impart to it a general historical power; when the persons of the verb were distinguished by pronominal elements, and when the augment, no longer purely demonstrative, had become the symbol of time past, that at such a time people were still unable to distinguish, except by a kind of Chinese law of position, between “the father struck the child,” and “the child struck the father.” Before we can admit this, we want much stronger proofs than any adduced by Curtius. He says, for instance, that compound verbal bases formed with +yâ+, to go, and afterwards fixed as causatives, would be inconceivable during a period in which accusatives existed. From +naś+, to perish, we form in Sanskrit +nâśa-yâmi+, I make perish. This, according to Curtius, would have meant originally, I send to perishing. Therefore +nâśa+ would have been, in the accusative, +nâśam+, and the causative would have been +nâśamyâmi+, if the accusative had then been known. But we have in Latin[34] _pessum dare_, _venum ire_, and no one would say that compounds like _calefacio_, _liquefacio_, _putrefacio_, were impossible after the first Aryan separation, or after that still earlier period to which Curtius assigns the formation of the Aryan case-terminations. Does Professor Curtius hold that compound forms like Gothic _nasi-da_ were formed not only before the Aryan separation, but before the introduction of case-terminations? I hold, on the contrary, that such really old compositions never required, nay never admitted, the accusative. We say in Sanskrit, +dyu-gat+, going to the sky, +dyu-ksha+, dwelling in the sky, without any case-terminations at the end of the first part of the compound. We say in Greek, σακέσ-παλος, not σάκοσ-παλος, παιδοφόνος, not παιδαφόνος, ὀρεσ-κῷος, mountain-bred, and also ὀρεσί-τροφος, mountain-fed. We say in Latin, _agri-cola_, not _agrum-cola_, _fratri-cīda_, not _fratrem-cīda_, _rēgĭfugium_, not _regis-fugium_. Are we to suppose that all these words were formed before there was an outward mark of distinction between nominative and accusative in the primitive Aryan language? Such compounds, we know, can be formed at pleasure, and they continued to be formed long after the full development of the Aryan declension, and the same would apply to the compound stems of causal verbs. To say, as Curtius does, that composition was possible only before the development of declension, because when cases had once sprung up, the people would no longer have known the bases of nouns, is far too strong an assertion. In Sanskrit[35] the really difficult bases are generally sufficiently visible in the so-called Pada, cases, _i.e._, before certain terminations beginning with consonants, and there is besides a strong feeling of analogy in language, which would generally, though not always (for compounds are frequently framed by false analogy), guide the framers of new compounds rightly in the selection of the proper nominal base. It seems to me that even with us there is still a kind of instinctive feeling against using nouns, articulated with case-terminations, for purposes of composition, although there are exceptions to that rule in ancient, and many more in modern languages. We can hardly realize to ourselves a Latin _pontemfex_, or _pontisfex_, still less _ponsfex_ instead of _pontifex_, and when the Romans drove away their kings, they did not speak of a _regisfugium_ or a _regumfugium_, but they took, by habit or by instinct, the base _regi_, though none of them, if they had been asked, knew what a base was. Composition, we ought not to forget, is after all only another name for combination, and the very essence of combination consists in joining together words which are not yet articulated grammatically. Whenever we form compounds, such as _railway_, we are still moving in the combinatory stage, and we have the strongest proof that the life of language is not capable of chronological division. There was a period in the growth of the Aryan language when the principle of combination preponderated, when inflection was as yet unknown. But inflection itself was the result of combination, and unless combination had continued long after inflection set in, the very life of language would have become extinct.

I have thus tried to explain why I cannot accept the fundamental fact on which the seven-fold division of the history of the Aryan language is founded, viz., that the combinatory process which led to the Aryan system of conjugation would have been impossible, if at the time nominal bases had already been articulated with terminations of case and number. I see no reason why the earliest case-formations, I mean particularly the nominative and accusative in the singular, plural, and dual, should not date from the same time as the earliest formations of conjugation. The same process that leads to the formation of +vak-ti+, speak-he, would account for the formation of +vak-s+, speak-there, _i.e._, speaker. Necessity, which after all is the mother of all inventions, would much sooner have required the clear distinction of singular and plural, of nominative and accusative, than of the three persons, of the verbs. It is far more important to be able to distinguish the subject and the object in such sentences as “the son has killed the father,” or “the father has killed the son,” than to be able to indicate the person and tense of the verb. Of course we may say that in Chinese the two cases are distinguished without any outward signs, and by mere position; but we have no evidence that the law of position was preserved in the Aryan languages, after verbal inflection had once set in. Chinese dispenses with verbal inflection as well as with nominal, and an appeal to it would therefore prove either too much or too little.

At the end of the five periods which we have examined, but still before the Aryan separation, Curtius places the sixth, which he calls the Period of the Formation of Cases, and the seventh, the Period of Adverbs. Why I cannot bring myself to accept the late date here assigned to declension, I have tried to explain before. That adverbs existed before the great branches of Aryan speech became definitely separated has been fully proved by Professor Curtius. I only doubt whether the adverbial period can be separated chronologically from the case period. I should say, on the contrary, that some of the adverbs in Sanskrit and the other Aryan languages exhibit the most primitive and obsolete case-terminations, and that they existed probably long before the system of case-terminations assumed its completeness.

If we look back at the results at which we have arrived in examining the attempt of Professor Curtius to establish seven distinct chronological periods in the history of the Aryan speech, previous to its separation into Sanskrit, Greek, Latin, Slavonic, Teutonic, and Celtic, I think we shall find two principles clearly established:--

1. That it is impossible to distinguish more than _three_ successive phases in the growth of the Aryan language. In the first phase or period the only materials were roots, not yet compounded, still less articulated grammatically, a form of language to us almost inconceivable, yet even at present preserved in the literature and conversation of millions of human beings, the Chinese. In that stage of language, “king rule man heap law instrument,” would mean, the king rules men legally.

The _second_ phase is characterized by the combination of roots, by which process one loses its independence and its accent, and is changed from a full and material into an empty or formal element. That phase comprehends the formation of compound roots, of certain nominal and verbal stems, and of the most necessary forms of declension and conjugation. What distinguishes this phase from the inflectional is the consciousness of the speaker, that one part of his word is the stem or the body, and all the rest its environment, a feeling analogous to that which we have when we speak of _man_-hood, _man_-ly, _man_-ful, _man_-kind, but which fails us when we speak of _man_ and _men_, or if we speak of _wo-man_, instead of _wif-man_. The principle of combination preponderated when inflection was as yet unknown. But inflection itself was the result of combination, and unless it had continued long after inflection set in, the very life of language would have become extinct.

The _third_ phase is the inflectional, when the base and the modificatory elements of words coalesce, lose their independence in the mind of the speaker, and simply produce the impression of modification taking place in the body of words, but without any intelligible reason. This is the feeling which we have throughout nearly the whole of our own language, and it is only by means of scientific reflection that we distinguish between the root, the base, the suffix, and the termination. To attempt more than this three-fold division seems to me impossible.

2. The second principle which I tried to establish was that the growth of language does not lend itself to a chronological division, in the strict sense of the word. Whatever forces are at work in the formation of languages, none of them ceases suddenly to make room for another, but they work on with a certain continuity from beginning to end, only on a larger or smaller scale. Inflection does not put a sudden end to combination, nor combination to juxtaposition. When even in so modern a language as English we can form by mere combination such words as _man-like_, and reduce them to _manly_, the power of combination cannot be said to be extinct, although it may no longer be sufficiently strong to produce new cases or new personal terminations. We may admit, in the development of the Aryan language, previous to its division, three successive strata of formation, a _juxtapositional_, a _combinatory_, and an _inflectional_; but we shall have to confess that these strata are not regularly superimposed, but tilted, broken up, and convulsed. They are very prominent each for a time, but even after that time is over, they may be traced at different points, pervading the very latest formations of tertiary speech. The true motive power in the progress of all language is combination, and that power is not extinct even in our own time.

[Footnote 1: This Lecture has been translated by M. Louis Havet,
and forms the first fasciculus of the Bibliothèque de l’École des
Hautes Études, publiée sous les auspices du Ministère de
l’Instruction Publique. Paris, 1869.]

[Footnote 2: See Benfey, _Ueber die Aufgabe des Kratylos_,
Göttingen, 1868.]

[Footnote 3: Theokritos, xvii. 9.]

[Footnote 4: In my essay _On the Relation of Bengali to the Aryan
and Aboriginal Languages of India_, published in 1848, I tried to
explain these plural suffixes, such as +dig+, +gaṇa+, +jâti+,
+varga+, +dala+. I had translated the last word by _band_,
supposing from Wilson’s Dictionary, and from the Śabda-kalpa-druma
that _dala_ could be used in the sense of band or multitude.
I doubt, however, whether _dala_ is ever used in Sanskrit in that
sense, and I feel certain that it was not used in that sense with
sufficient frequency to account for its adoption in Bengali. Dr.
Friedrich Müller, in his useful abstracts of some of the grammars
discovered by the _Novara_ in her journey round the earth
(1857-59), has likewise referred +dal+ to the Sanskrit +dala+, but
he renders what I had in English rendered by _band_, by the German
word _Band_. This can only be an accident. I meant _band_ in the
sense of a band of robbers, which in German would be _Bande_. He
seems to have misunderstood me, and to have taken _band_ for the
German _Band_, which means a ribbon. Might +dala+ in Bengali be
the Dravidian +taḷa+ or +daḷa+, a host, a crowd, which Dr.
Caldwell (p. 197) mentions as a possible etymon of the pluralizing
suffix in the Dravidian languages? Bengali certainly took the idea
of forming its plurals by composition with words expressive of
plurality from its Dravidian neighbor, and it is not impossible
that in some cases it might have transferred the very word +daḷa+,
crowd. This +daḷa+ and +taḷa+ appears in Tamil as _kala_ and
_gala_, and as Sanskrit _k_ may in Sinhalese be represented by _v_
(+loka+ = _lova_), I thought that the plural termination used in
Sinhalese after inanimate nouns might possibly be a corruption of
the Tamil _kala_. Mr. Childers, however, in his able “Essay on the
formation of the Plural of Neuter Nouns in Sinhalese”
(_J. R. A. S._, 1874, p. 40), thinks that the Sinhalese _vala_ is
a corruption of the Sanskrit +vana+, forest, an opinion which
seems likewise to be held by Mr. D’Alwis (l.c. p. 48). As a case
in point, in support of mv own opinion, Mr. Childers mentioned to
me the Sinhalese _malvaru_, Sanskrit +mâlâ-kâra+, a wreath-maker,
a gardener. In Persian both _ân_ and _hâ_ are remnants of decayed
plural terminations, not collective words added to the base.]

[Footnote 5: Stanislas Julien, _Exercises Pratiques_, p. 14.]

[Footnote 6: Endlicher, _Chinesische Grammatik_, § 122. Wade,
_Progressive Course on the Parts of Speech_, p. 102. A different
division of words adopted by Chinese grammarians is that into
_dead_ and _live words_, _ssè-tsé_ and _sing-tsé_, the former
comprising nouns, the latter verbs. The same classes are sometimes
called _tsing-tsé_ and _ho-tsé_, unmoved and moved words. This
shows how purposeless it would be to try to find out whether
language began with noun or verb. In the earliest phase of speech
the same word was both noun and verb, according to the use that
was made of it, and it is so still to a great extent in Chinese.
See Endlicher, _Chinesische Grammatik_, § 219.]

[Footnote 7: _Agglutinative_ seems an unnecessarily uncouth word,
and as implying a something which glues two words together, a kind
of _Bindevocal_, it is objectionable as a technical term.
_Combinatory_ is technically more correct, and less strange than
agglutinative.]

[Footnote 8: Professor Pott, in his article entitled “Max Müller
und die Kennzeichen der Sprachverwandtschaft,” published in 1855,
in the _Journal of the German Oriental Society_, vol. ix. p. 412,
says, in confutation of Bunsen’s view of a real historical
progress of language from the lowest to the highest stage: “So
cautious an inquirer as W. von Humboldt declines expressly, in the
last chapter of his work on the _Diversity of the Structure of
Human Language_ (p. 414), any conclusions as to a real historical
progress from one stage of language to another, or at least does
not commit himself to any definite opinion. This is surely
something very different from that gradual progress, and it would
be a question whether, by admitting such an historical progress
from stage to stage, we should not commit an absurdity hardly less
palpable than by trying to raise infusoria into horses or still
further into men. [What was an absurdity in 1855 does not seem to
be so in 1875.] Mr. Bunsen, it is true, does not hesitate to call
the monosyllabic idiom of the Chinese an inorganic formation. But
how can we get from an inorganic to an organic language? In nature
such a thing would be impossible. No stone becomes a plant, no
plant a tree, by however wonderful a metamorphosis, except, in a
different sense, by the process of nutrition, _i.e._, by
regeneration. The former question, which Mr. Bunsen answers in the
affirmative, is disposed of by him with the short dictum: ‘The
question whether a language can be supposed to begin with
inflections, appears to us simply an absurdity;’ but unfortunately
he does not condescend, by a clear illustration, to make that
absurdity palpable. Why, in inflectional languages, should the
grammatical form always have added itself to the matter
subsequently and _ab extra_? Why should it not partially from the
beginning have been created with it and in it, as having a meaning
with something else, but not having antecedently a meaning of its
own?”]

[Footnote 9: Cf. D. G. Brinton, _The Myths of the New World_,
p. 6, note.]

[Footnote 10: Julien, _Exercises Pratiques_, p. 120. Endlicher,
_Chineseische Grammatik_, § 161. See, also, Nöldeke, _Orient und
Occident_, vol. i. p. 759. _Grammar of the Bornu Language_
(London, 1853), p. 55: “In the Treaty the genitive is supplied by
the relative pronoun _agu_, singularly corroborative of the Rev.
R. Garnett’s theory of the genitive case.”]

[Footnote 11: “Time changes the meaning of words as it does their
sound. Thus, many old words are retained in compounds, but have
lost their original signification. E.g., _’k·eu_, mouth, has been
replaced in colloquial usage by _’tsui_, but it is still employed
extensively in compound terms and in derived senses. Thus, _k·wai‘
’k·eu_ a rapid talker, _.men ’k·eu_, door, _,kwan ’k·eu_, custom
house. So also _muh_, the original word for eye, has given place
to _’yen, tsing_, or _’yen_ alone. It is, however, employed with
other words in derived senses. E.g., _muh hia·_, at present; _muh
luh_, table of contents.

“The primitive word for head, _’sheu_, has been replaced by
_.t‘eu_, but is retained with various words in combination. E.g.,
_tseh ’sheu_, robber chief.”

Edkins, _Grammar of the Chinese Colloquial Language_, 2d edition,
1864, p. 100.]

[Footnote 12: Grimm, _Deutsche Grammatik_, ii. 339.]

[Footnote 13: Cf. the German Liebhart, mignon, in Anshelm, 1, 335.
Grimm, _Deutsche Grammatik_, iii. 707. I feel more doubtful now as
to _sweetard_. Dr. Morris mentions it in his _Historical Outlines
of English Grammar_, p. 219; but Koch, when discussing the same
derivations in his _English Grammar_, does not give the word. Mr.
Skeat writes to me: “The form really used in Middle English is
_sweeting_. Three examples are given in Stratmann. One of the best
is in my edition of William of Palerne, where, however, it occurs
not _once_ only (as given by Stratmann), but _four times_, viz.:
in lines 916, 1537, 2799, 3088. The lines are:--

‘Nai, sertes, _sweting_, he seide· that schal I neuer.’ 916
‘& seide aswithe· _sweting_, welcome!’ 1537
‘Sertes, _sweting_, thæt is soth. seide william thanne.’ 2799
‘treuli, _sweting_, that is soth· seide william thane.’ 3088

The date of this poem is about A.D. 1360. Shakespeare has both
forms, viz.: _sweeting_ and _sweet-heart_. Chaucer has _swete
herte_, just as we should use _sweet-heart_.”]

[Footnote 14: Diez, _Grammatik_, ii. 358. Grimm, _Deutsche
Grammatik_, i. p. 340, 706.]