It may be unfair to Bopp to give a selection from his initial work. But his chief importance is in clarifying the morphology of Indo-European, and even his final presentation has long been superseded. Accordingly the views which he first presented are those of greatest interest to us. Moreover, his analysis of the conjugational system of the Sanskrit language is by no means a negligible result of four years of independent work, carried on with little guidance from predecessors. The extracts presented here indicate however that Bopp's publication of 1816 was still preliminary to the important treatments in comparative linguistics.
For in 1816 Bopp is still pursuing the course of Friedrich von Schlegel. To be sure a much greater portion of his book is devoted to the language, pp. 3-157, but as much space is given to Indic literature, primarily to translations, pp. 160-312. Bopp's chief aim is accordingly an understanding of Indic culture, not of the Indic language, let alone that of the Indo-European family. His first work then resembles a comparative grammar of the Indo-European languages less than does the monograph of Rask. The publication in 1818 of Rask's work, which had been completed earlier, may have been as beneficial to Bopp in his groping toward a comparative grammar of the Indo-European languages as it was to Grimm.
To interpret Bopp's aims from the often tedious introduction of his teacher Windischmann, Conjugational System i-xxxxvi, may also be less than flattering to the mature Bopp; but it gives us an insight into contemporary hopes for comparative linguistics and accordingly some understanding of the tremendous energy with which it was pursued. According to Windischmann, ix-x, Bopp "had resolved to treat the investigation of language as a historic and philosophic study and not to be content with understanding what was written in any given language. We may rejoice at these efforts and intentions, which from a purely human point of view deserve to be named before many others, for through intimate association with the significant signs, by which the word, this child of the spirit, expresses the deepest emotions and feelings, as it does the clearest and most definite thoughts, indescribably much of the hindrances to true self-knowledge and self-culture are dispelled." Moreover, in study of languages, such as Gothic, and their structure, there was hope, according to Windischmann, for additional means to illuminate the history of the Indic and Germanic peoples and the differing cultures of each. Such considerations led Bopp to master ever more of the Indo-European languages-Sanskrit, Avestan, Greek, Latin, Lithuanian and Gothic for the first volumes of his Comparative Grammar of 1833 -- then Slavic, Celtic and Albanian for remaining volumes, and the second edition of 1857-61. The posthumous third of 1868-70 maintains some of the initial shortcomings of the early period of comparative linguistics virtually to the time of the neo-grammarians.
One shortcoming was the almost exclusive attention to morphology. We note Grimm's similar lack of interest for phonology. Raumer's attention to phonetics had its influence only on the successors to the great pioneers
Another shortcoming is Bopp's attempt to discern the origin of inflection in separate words, particularly the verb "to be". In its crass form, this is completely superseded. Yet many publications still emerge which seek the origin of inflections, like the Germanic weak preterite, in simple verbs such as do, even though highly conservative and careful linguists, e.g. H. Collitz, Das Schwache Präteritum, Baltimore, 1912, have cited almost overwhelming evidence against such views. The early notions on the development of language, from noninflected through agglutinative to inflected, have not been discarded even today, though we probably would find little receptivity for the view that certain inflections developed because of an inherent meaning of the symbol, such as s for the second person.
Franz Bopp is often credited with providing "the real beginning of what we call comparative linguistics" (Pedersen, Linguistic Science, p. 257). In keeping with this achievement his external career was distinguished. His publication resulting from four years of study in Paris, 1812-1816, led to general recognition. After visiting London and publishing there, he became professor of Sanskrit and comparative grammar in Berlin in 1821. Teaching and publication made up the rest of his life; his publications are on the whole admirable, except for a suggestion that the Malayo-Polynesian languages are related to the Indo-European. Apart from this lapse, editions, monographs and successive editions of his grammar, with translations into English and French, made him the dominant figure in Indo-European comparative grammar throughout the first half of the nineteenth century.
By verb in the narrowest sense is meant that part of speech which expresses the connection of a topic with a property, and their relations to one another.
The verb, according to this definition, has no real meaning in itself, but is simply the grammatical bond between subject and predicate, through whose inner change and formation their mutual relations are indicated.
Under this concept there is only a single verb, namely the so called verbum abstractum, sein, esse. But also with this verb, to the extent that it is to express simply the relations between subject and predicate, we have to remove the concept of existence, which it comprehends in itself; in its grammatical determination this does not need to express the existence of the subject, because this is already expressed by the subject when we state it. Thus in the sentence: homo est mortalis, it is not the verb, est, which expresses the existence of the subject homo, but the existence is contained as the first and basic characteristic in the concept expressed by the word homo, just as the characteristic mortalis like others assumed to be known for the concept homo is associated through the copula est. In the sentence: der Gott ist seiend, the word sein represents two quite different functions. In the first it determines as grammatical bond simply the relation between the subject and the predicate; in the second it expresses the property which is added to the subject.
It seems to me therefore, that simply through lack of a completely abstract verb, a verb which embraces the concept of existence in itself is used in most languages for the sake of a grammatical bond; and there might well be a language, which is not without a totally meaningless copula, through whose inflection or inner change the relations between subject and predicate might be expressed. In Sanskrit there are two verbs which correspond to the verb esse, namely asti and bhavati. Whether both are exchanged with one another equally frequently, and although the first is replaced by the latter in the tenses lacking to it, nonetheless for both synonyms a fine difference must exist, which may possibly have been distinguished more sharply originally. Nonetheless it seems to me from observation of the use of both, and from comparison of the substantives and adjectives derived from the roots of both, to emerge clearly that asti almost alone expressed grammatical union, that bhavati however is primarily used when existence is to be expressed. From the root bhū come the words bhāvana, svajambhū, prabhu, bhūtam, bhavān, etc., all of which point to existence. From the root as one can hardly find a noun derived other than the participle sat and its negative asat. The following verse from the Bāgavat Gita can probably not be translated faithfully into any language:
Nāsatō vidjatae bhāvō nābhāvō vidjatae satah.
The relation of the subject with its predicate is not always expressed through a special part of speech, but is unexpressed; and the relations and secondary determinations of meaning are indicated through the inner change and inflection of the word itself that expresses the attribute. The adjectives inflected in this way make up the sphere of verbs in the usual sense.
Among all the languages known to us, the sacred language of the Indians shows itself to be one of the most capable of expressing the most varied relations and connections in a truly organic manner through inner modification and forming of the stem syllable. But disregarding this remarkable capability of modification, occasionally it is pleased to incorporate the root of the verbum abstractum in which case the stem syllable and the incorporated verbum abstractum divide the grammatical functions of the verb.
Among the languages which are of common origin with the Old Indic we have to admire the capability of indicating the most varied determinations of relationship, most of all in the Greek. In the conjugation of the verbs it not only follows the same principle as the Sanskrit, but the inflections by which it expresses the same relations are exactly the same; and it combines in the same tenses and in the same way the verbum abstractum with the stem syllable.
The Roman language agrees with the Indic no less than does the Greek, and one could hardly find in it a relation expressed by an inflection which is not common to it and Sanskrit. In the conjugation of verbs however the combination of the root with an auxiliary verb has become the prevailing principle for it. In this combination however it does not express a part of the relation, which is to be defined, through inflection of the stem syllable, as this is the case in Indic and in Greek, but the root remains totally unchanged.
It is the purpose of this essay to show how in the conjugation of the Old Indic verbs the definitions of relationship are expressed through corresponding modifications of the root, how at times however the verbum abstractum is combined with the stem syllable to one word, and stem syllable and auxiliary divide the grammatical functions of the verb; to show how the same is the case in the Greek language, how in Latin the system of combination of root with an auxiliary has come to be dominant, and how only in this way the apparent difference of the Latin conjugation from that of Sanskrit and Greek arose; finally to prove, that in all the languages which stem from Sanskrit or from a mother language in common with it, no definition of relationship is indicated by an inflection which is not common to them and that original language, and that apparent exceptions only arise from the fact that either the stem syllable is combined with the auxiliaries into one word, or that from participles the tempora derivativa which are customary already in Sanskrit are derived, in the fashion as verba derivativa can be formed from substantives in Sanskrit, Greek and many other languages.
Among the languages that stand in closest relationship with Sanskrit I recognize especially Greek, Latin, Germanic and Persian. It is remarkable that Bengalese, which surely has undergone the least foreign admixtures among the New Indic dialects, does not agree in its grammar nearly so completely with Sanskrit as do the above-mentioned languages, while on the other hand it attests a far greater number of Old Indic words. Yet new organic modifications have not taken the place of the Old Indic inflections, but after their meaning and spirit have gradually vanished, their use also diminished, and tempora participialia (among which I do not understand periphrastic forms like the Latin amatus est) replaced the tenses which were formed in Sanskrit through inner change of the stem syllable. Similarly in the New Germanic languages, several indications of relationship are expressed through periphrasis, which in Gothic were designated by inflections that were already used in Sanskrit and Greek.
In order to show in its full light the truth of these principles which are extremely important for the history of languages, it is necessary to become acquainted above all with the conjugational system of the Old Indic languages, then to survey and compare the conjugations of the Greek and the Roman, the Germanic and Persian languages, whereby we will see their identity, but will also recognize the gradual and graded destruction of the simple speech organism and observe the striving to replace it by mechanical combinations, from which an appearance of a new organism arose when their elements were no longer recognized.
We will go through the tenses of the Indic verbs here in the sequence in which they follow one another in the Sanskrit grammars, and in the process will give as briefly and compactly as possible the reason for every change of form and depict the manner how every modification of meaning corresponds to an individual modification of the word. From this it will become clear of itself that many tenses must be explained as compounds. Since however in my assertions I cannot support myself on the authority of others, for up to now nothing has been written about the origin of the grammatical forms, I will have to support them with cogent proofs.
In the tempus praesens the meaning of the root is limited through no added secondary indication; the subject has real use of the predicate designated by the root. Also from the root, which is the common mother of all parts of speech, the tempus praesens is formed through simple addition of the designations for person. The designation for the first person is M for the singular and plural, and for the dual V; designation of the second person is S or H which is related to it; designation of the third person is T for all three numbers. The endings, or the accents of the personal designations serve to determine the numbers, not the formation and characterization of the tenses.
Example: ad, eat
|3.||atti < adti||attah < adtah||adanti|
|2.||atsi-adsi||atthah-adthah||attha < adtha|
Note. The D of the root becomes T before T and S in accordance with the rules of euphony. (end of p. 13).
In Greek, as in Sanskrit, certain random letters are added to roots, which as in Indic are maintained only in some tenses and disappear again in the others. One could, as in Sanskrit, divide the verbs into different conjugations in accordance with these, which then would largely correspond with the Indic in their characteristics. -- The first Indic conjugation adds a to the first root; thus patschati comes from patsch. With this one can compare those Greek verbs which insert e, a or o between root and designation for person. The third conjugation of Sanskrit repeats the initial letters of the root, e.g. dadāti, tischthati from dā and sthā. So in Greek dídōmi, héstēmi from da and stat. The fifth Indic conjugation adds nu to the root; e.g. sunuma 'we beget' from su. To this corresponds in Greek rhēgnumen, déiknumen, dáinumen from the roots rhēg, deik, dai. -- The eighth Indic conjugation adds u, e.g. tanuma 'we extend' from the root tan. -- The ninth conjugation adds the sylable na in Sanskrit, e.g. krināmi from kri. N is often inserted in Greek between the root and the designation for person, as in krínō, klínō, témnō, etc., from kri, kli, tem.
In order to learn to know the principle of the Latin conjugation, it is necessary that we start out from the conjugation of the auxiliary verbs, partly because of their frequent combination with the other verbs, partly because in their simpler change the principle of the Latin conjugation is easier to recognize.
The Latin language has two verbs, which are used for combination between subject and the predicate expressed by an adjective or substantive, and for the expression of their mutual relation to one another. Their stem syllables are es and fu, corresponding to the Indic roots of the same meaning as and bhu. As in Sanskrit bhavati replaces those tenses that went out of use for asti, so it happened for Latin fu. The ancients said esum; the Etruscans (=Umbrians) for sum: esume. Esu-me is like Indic as-mi and the Greek esmi, esmai. -- The praeteritum of esum is eram, with change of the s to the related r, accordingly eram for esam. Also in Sanskrit and in Greek the personal designations with A are emphasized. But the past is not expressed through this emphasis, rather through modification of the root: through replacement of the augment, through reduplication or change of the stem vowel. Eram is different from esum; its use gives its past meaning, but this modification of the meaning does not correspond to a particular modification of the root.
However much the inflections have gone out of use in other parts of speech of the Persian language, through which in Indic and the languages related to it important secondary specifications are indicated, yet especially in the inflection of verbs the close bond can be recognized which ties it to those languages whose system of conjugation we have examined. With the old Germanic dialects it affords in the principle of the change of verbs such striking agreement that for the sake of brevity I consider myself justified to place it with them in one class. In the Persian language and in all Germanic dialects, the tempus praesens is derived from the root through simple affixation of the personal designations, which are known to us from Sanskrit as from Greek and Latin. Yet these have not maintained themselves throughout, but are at times replaced through vowels, as in Greek and Roman; eventually the designation of a definite person becomes the common ending of all others, as will be clear from the following examples.
From the roots ber, luf, sok, mach, brenn there are made in Persian, Anglo-Saxon, Gothic, Frankish and Icelandic the following presents:
|2.||-- -i||-- -ast||sokj-ais||-- -ost||-- -er|
|3.||-- -ed||-- -ath||sok-eith||-- -ot||-- -er|
|2.||-- -id||-- -iath||sok-eith||-- -ot||-- -ed|
|3.||-- -end||-- -iath||sok-and||-- -ont||-- -a|