« ÎnapoiContinuă »
trace is left in English of substantives present, cantans, is used in the Romance
languages only as an adjective, for inAlthough, as I said, it might seem stance, "une femme souffrante," &c. c more plausible to look on the modern Here, then, we see, again, that in anaparticiple in English as originally an lytical languages the participle present adjective in ing, such popular phrases can be supplanted by the oblique case as a-going, a-thinking, point rather to
of a verbal noun. the verbal substantives in ing, as the Let us now look to a more distant, source from which the modern English yet to a cognate language, like Bengali. participle was derived.
“I am going We there find + that the so-called infiniis in reality a corruption of “I am tive is formed by te, which te is at the a-going," i.e. “I am on going," and the same time the termination of the locaparticiple present would thus be traced tive singular. Hence the present, Kaback to a locative case of a verbal ritechi, I am doing, and the imperfect, noun. ?
Karitechilám, I was doing, are mere comLet us remember, then, that the place pounds of achi, I am, úchilám, I was, of the participle present may, in the with what may be called a participle progress of dialectic regeneration, be present, but what is in reality a verbal supplied by the locative or some other noun in the locative. Karilechi, I do, case of a verbal noun.
means “I am in doing,” or “I am Now let us look to French. On the a-doing.” 3d of June, 1679, the French Academy Now the question arises, does this decreed that the participles of the pre- perfectly intelligible method of forming sent should no longer be declined.2 the participle from the oblique case of a
What was the meaning of this decree ? verbal noun, and of forming the present Simply what may now be found in indicative by compounding this verbal every. French grammar, namely, that noun with the auxiliary verb “to be,' commençant, finissant, are indeclinable supply us with a test that may be safely when they have the meaning of the applied to the analysis of languages participle present, active or neuter; hut which decidedly belong to a different that they take the terminations of the family of speech ? Let us take the masculine and feminine, in the singular Bask, which is certainly neither Aryan and plural, if they are used as adjectives. nor Semitic, and which has thrown out But what is the reason of this rule ? a greater abundance of verbal forms Simply this, that chantant, if used as a than almost any known language. Here participle, is not the Latin cantans, but the present is formed by what is called the so-called gerund, that is to say, the a participle, followed by an auxiliary oblique case of a verbal noun, the Latin verb. This participle, however, is formed cantando corresponding to the English by the suffix an, and the same suffix is a-singing, while the Latin participle used to form the locative case of nouns. If, then, we examine the verb,
For instance, mendia, the mountain ; Ci.Garnett's paper on the formation of words mendiaz, from the mountain; mendian, from inflected cases, Philological Society, vol. iii, No. 54, 1847. Garnett compares the Welsh
in the mountain ; mendico, for the sake yn sefyll, in standing, Ir. ag seasamh, on
of the mountain. In like manner standing, the Gaelic ag sealgadh. The same etchean, in the house ; ohean, in the bed. ingenious and accurate scholar was the first to propose the theory of the participle being formed from the locative of a verbal
4 M. M.'s Essay on the Relation of the ? Cf. Egger, Notions Elémentaires de Gram- Bengali to the Aryan and Aboriginal Lanmaire Comparée, Paris, 1856, p. 197. “ La règle guages of India. Report of the British Assoest faite. On ne declinera plus les participes ciation for the Advancement of Science, 1847, présenta.” (B. Jullien, Cours Supérieur, i. pp. 344-45. Cf. Garnett, 1.c. p. 29. p. 186.)
5 See Inchauspe’s Verbe Basque, published 3 Diez, Vergleichende Grammatik der Ro- by Prince Louis-Lucien Bonaparte. Bayonne, mapischen Sprachen, ii. p. 114.
as many have failed before and after erorten niz, I fall ;
him, if imagining that what has been hiz, thou fallest;
found to be true in one portion of the da, he falls;
vast kingdom of speech must be equally
true in all. This is not so, and cannot we see again in erorten a locative, or, as
be so. Language, though its growth it is called, a positive case of the verbal
is governed by intelligible principles substantive erorta, the root of which
throughout, was not so uniform in its would be eror, falling ;l so that the
progress as to repeat exactly the same indicative present of the Bask verb does
phenomena at every stage of its life. As not mean either I fall, or I am falling,
the geologist looks for different characbut was intended originally for “I am teristics when he has to deal with in the act of falling,” or, to return to the London clay, with Oxford clay, or with point from whence we started, I am a
old red sandstone, the student of lanfalling. The a in a-falling stands for
guage, too, must be prepared for different an original on. Thus aright is on rihte
, formations, even though he confines away is on veg, aback is on bæc, again is
himself to one stage in the history of on gegen, among is on gemang, &c.
language, the inflectional. And if he This must suffice as an illustration of steps beyond this, the most modern the principle that what is real in modern stage, then to apply indiscriminately to
, formations must be admitted as possible the lower stages of human speech, to the in more ancient formations, and that agglutinative and radical, the same tests what has been found to be true on a
which have proved successful in the insmall scale
may be true on a larger scale. flectional, would be like ignoring the I speak thus cautiously, because there difference between aqueous, igneous, and is much in the science of language to
metamorphic rocks. There are scholars tempt us to overstep these, the legiti- who are incapable of appreciating more mate limits of inductive reasoning. We than one kind of evidence. No doubt may infer from the known to the un
the evidence on which the relationship known in language tentatively, but not of French and Italian, of Greek and positively. It does not follow, even Latin, of Lithuanian and Sanskrit, of within so small a sphere as the Aryan Hebrew and Arabic, has been estafamily of speech, that what is possible blished, is the most satisfactory'; but in French is possible in Latin, that what such evidence is possible only in inflecexplains Bengali will explain Sanskrit.
tional languages that have passed their Still less would it be safe to treat all the period of growth, and have entered into languages of the world as if they were the stage of phonetic decay. To call but modifications or repetitions of San- for the same evidence in support of the skrit. Mr. Garnett, in excellent homogeneousness of the Turanian lanpaper on the participle, has traced similar phenomena in a much larger from the nature of the case, it is impos
guages, is to call for evidence which, number of languages, and he has even sible to supply. As well might the geoendeavoured to show that the original logist look for fossils in granite ! The Indo-European participle, the Latin Turanian languages allow of no grammatiamans, the Greek Túntwv, the Sanskrit
cal petrifactions like those on which the bodhat, were formed on the same prin- relationship of the Aryan and Semitic ciple :—that they were cases of a verbal families is chiefly founded. If they did, noun. In this, however, he has failed, they would cease to be what they are ;
I Cf. Dissertation critique et apologétique they would be inflectional, not agglutisur la Langue Basque, (par l'Abbé Darrigol). native. Bayonne. He takes dravat as a possible ablative,
If languages were all of one and the likewise s'ds-ut, and tan-vat (sic). It would be special tenses, nor would the ablative be so impossible to form ablatives in åt (as) from appropriate a case as the locative, for taking verbal bases raised by the vikaranas of the the place of a verbal adjective.
same texture, they might be unravelled, ample evidence to confirm the views no doubt, with the same tools.
which I then expressed. My excellent they are not—and this is admitted by friend, the Bishop of Melanesia, of whom all—it is surely mere waste of valuable it is difficult to say whether we should time to try to discover Sanskrit in the admire him more as a Christian, or as a Malay dialects, or Greek in the idioms scholar, or as a bold mariner, meets in of the Caucasian mountaineers. The every small island with a new language, whole crust of the earth is not made which none but a scholar could trace back of lias, swarming with Ammonites and to the Melanesian type.
“ What an indiPlesiosauri, nor is all language made “cation,” he writes, “ of the jealousy and of Sanskrit, teeming with Supines and suspicion of their lives, the extraordiPaulo-pluperfects. If we compare the nary multiplicity of these languages extreme members of the Polynesian dia- “affords! In each generation, for aught lects, we find but little agreement in “I know, they diverge more and more; what may
be called their grammar, and "provincialisms and local words, &c. many of their words seem totally dis- “perpetually introduce new causes for tinct. But, if we compare their nume- "perplexity." rals, we clearly see that these are com- I shall mention to-day but one new, mon property; we perceive similarity, though insignificant, cause of change in though at the same time great diversity. the Polynesian languages, in order to We begin to note the phonetic changes show that it is difficult to over-estimate that have taken place in one and the the multifarious influences which are at same numeral, as pronounced by different work in nomadic dialects, constantly islanders; we thus arrive at phonetic changing their aspect and multiplying laws, and these, in their turn, remove their number. the apparent dissimilarity in other words The Tahitians, besides the metaphowhich at first seemed totally irrecon- rical expressions, have another and a cilable. But mere phonetic decay will more singular mode of displaying their not account for the differences between reverence towards their king, by a custhe Polynesian dialects, and, unless we tom which they term te pi. They cease admit the process of dialectic regenera- to employ, in the common language, tion to a much greater extent than we those words which form a part or the should be justified in the Aryan and whole of the sovereign's name, or that Semitic families, our task of reconcilia- of one of his near relatives, and invent tion would become hopeless. Will it be new terms to supply their place. As all believed that since the time of Cook five names in Polynesian are significant, and of the ten simple numerals in the lan- a chief usually has several, it will be guage of Tahiti have been thrown off seen that this custom must produce a and replaced by new ones? This is, considerable havoc in a language. It is nevertheless, the fact.
true that this change is only temporary,
as, at the death of the king or chief, the Two was rua; it is now piti,
new word is dropped, and the original Four was ha; it is now maha,
term resumed. But it is hardly to be Five was rima; it is now pae,
supposed that after one or two generaSix was one ; it is now fene,
tions the old words should still be reEight was varu ; it is now vau.1
membered and be reinstated. Anyhow, I tried in one of my former lectures it is a fact, that the missionaries, by to explain some of the causes which in employing many of the new terms, nomadic dialects produce a much more
give them a permanency which will defy rapid shedding of words than in literary the ceremonial loyalty of the natives. languages, and I have since received Vancouver observes (Voyage, vol. i. p. 1 United States Exploring Expedition, under
135) that at the accession of Otu, which the command of Charles Wilkes. Ethnography
took place between the visit of Cook and Philology, by H. Hale, vol. vii. p. 289.
2 Hale, p. 288.
and his own, no less than forty or fifty independent dialects, the results must of the most common words, which be very different from those which occur in conversation, had been entirely take place in Latin when split up into changed. It is not necessary that all the Romance dialects. In the Romance the simple words which go to make up dialects, however violent the changes a compound name should be changed. which made Portuguese words to differ The alteration of one is esteemed suffi- from French, there always remain a few cient. Thus in Po-mare, signifying "the fibres by which they hang together. It night (po) of coughing (mare), only the
might be difficult to recognise the French first word, po, has been dropped, mi plier, to fold, to turn, in the Portuguese being used in its place. So in Ai-mata chegar, to arrive, yet we trace plier back (eye-eater), the name of the present to plicare, and chegar to the Spanish queen, the ai (eat) has been altered to llegar, the old Spanish plegar, the amu, and the mata (eye) retained. In Te
Latin plicare, here used in the sense of arii-na-vaha-roa (the chief with the large turning towards a place, arriving at a mouth), roa alone has been changed to place. But when we have to deal with maoro. It is the same as if with the dialects of Chinese, everything that accession of Queen Victoria, either the could possibly hold them together seems word victory had been tabooed altogether, hopelessly gone. The language now or only part of it, for instance tori,
spoken in Cochin China is a dialect so as to make it high-treason to speak of Chinese, at least as much as Norman
, during her reign of Tories, this word French was a dialect of French, though being always supplied by another; such, spoken by Saxons at a Norman Court. for instance, as Liberal-Conservative.
There was a native language of Cochin The object was clearly to guard against China, the Ar.namitic, which forms, as the name of the sovereign being ever it were, the Saxon of that country on used, even by accident, in ordinary which the Chinese, like the Norman, was conversation, and this object was at- grafted. This engrafted Chinese, then, tained by tabooing even one portion of is a dialect of the Chinese as spoken in
China, and it is most nearly related to “But this alteration," as Mr. Hale the spoken dialect of Canton. Yet few remarks, “affects not only the words ,
Chinese scholars would recognise Chinese “themselves, but syllables of similar in the language of Cochin. It is, for in« sound in other words. Thus the name
stance, one of the most characteristic fea“of one of the kings being Tu, not tures of the literary Chinese, the dialect
only was this word, which means to of Nankin or the idiom of the Mandarins, "stand,' changed to tia, but in the
that every syllable ends in a vowel, either “word fetu, star, the last syllable, pure or nasal. In Cochin China, on " though having no connexion, except the contrary, we find words ending in "in sound, with the word tu, under- 'k, t, p. Thus, ten is thap, at Canton "went the same alteration-star being chap, instead of the Chinese tchi.S No “now fetia ; tui, to strike, became “ tiai ; and tu pa pau, a corpse, tia
Diez, Lexicon, s. v. llegar; Grammar, i pa pau. So ha, four, having been ? Endlicher, Chinesische Grammatik, par. changed to maha, the word aha, split, 53, 78, 96. “ has been altered to amaha, and murihú,
3 Léon de Rosny, Tableau de la Cochin“the name of a month, to muriáha.
chinie, p. 295. He gives as illustrations “ When the word ai was changed to
Annamique. dix thap
chap amu, maraai, the name of a certain pouvoir dak
tak “wind (in Rarotongan, maranai), became Sang
lam. It is equally clear that, if a radical
He likewise mentions double consonants in or monosyllabic language, like Chinese,
the Chinese as spoken in Cochin China, namely,
bl, dy, ml, ty, tr; also f, r, s. As final conbegins to change and to break out in
sonants he gives, ch, k, m, n, ng, p, t.--P. 295. exponent of the plural ; and kak, 2 Lectures on the Science of Language, i.
wonder that the early missionaries de- accent, it means what has been thrown scribed the Annamitic as totally distinct away ; pronounced with the grave cirfrom Chinese. One of them says : cumflex, it means what is left of a fruit “ When I arrived in Cochin China, after it has been squeezed out; pro“I heard the natives speak, particularly nounced with no accent, it means three; “the women : I thought I heard the pronounced with the ascending or in“twittering of birds, and I gave up all terrogative accent, it means a box on “hope of ever learning it. All words are
Thus “monosyllabic, and people distinguish
Ba, bà, ba, bá, “their significations only by means of “ different accents in pronouncing them. would mean, if properly pronounced, “The same syllable, for instance, dai, “ Three ladies gave a box on the ear to "signifies twenty-three entirely different the favourite of the prince.” How much "things, according to the difference of these accents must be exposed to fluc“accent, so that people never speak with- tuation in different dialects is easy to “out singing.” 1 This description, though perceive. Though they are fixed by somewhat exaggerated, is correct in the grammatical rules, and though their nemain, there being six or eight musical glect causes the most absurd mistakes, accents or modulations in this as in they were clearly in the beginning the other monosyllabic tongues, by which mere expression of individual feeling, the different meanings of one and the and therefore liable to much greater diasame monosyllabic root are kept dis- lectic variation than grammatical forms, tinct. These accents form an element properly so called. But let us take what of language which we have lost, but we might call grammatical forms in which was most important during the Chinese, in order to see how differently primitive stages of human speech. The they fare in dialectic dispersion, as comChinese larguage commands no more than pared with the terminations of inflec450 distinct sounds, and with them it tional languages. Though the gramexpresses between 40,000 and 50,000 matical organization of Latin has been words or meanings. These meanings well-nigh used up in French, we still are now kept distinct by means of com- see in the s of the plural a remnant of position, or in other languages by deri- the Latin paradigm. We can trace the vation, but on the radical stage they one back to the other. But in Chinese, would have bewildered the hearer en- when the plural is formed by the adtirely, without some hints to indicate dition of some word meaning “ multitheir real intention. We have some- tude, heap, flock, class,” what trace of thing left of this faculty in the tone of original relationship remains when one our sentences. We distinguish an in- dialect uses one, another another word ? terrogative from a positive sentence by The plural in Cochin-Chinese is formed the raising of our voice. (Gone? Gone.) by placing fo before the substantive. We pronounce Yes very differently when This fo means many, or a certain we mean perhaps (Yes, this may be number. It may exist in Chinese, but true), or of course (Yes, I know it), or it is certainly not used there to form really (Yes ? is it true ?) or truly (Yes, the plural. Another word employed I will). But in Chinese, in Annamitic for forming plurals is ñung, several, and (and likewise in Siamese and Burmese), this again is wanting in Chinese. It these modulations have a much greater fortunately happens, however, that a importance. Thus in Annamitic ba pro- few words expressive of plurality have nounced with the grave accent means a been preserved both in Chinese and lady, an ancestor ; pronounced with the Cochin-Chinese; as, for instance, choung, sharp accent, it means the favourite of a clearly the Chinese tchoung,' meaning prince ; pronounced with the semi-grave conflux, vulgus, all, and used as an 1 Léon de Rosny, 1.c. p. 301.
1 Endlicher, S 152.