Page images
PDF
EPUB

fallen off in cau-os, cau-ea, caulae, &c., all apparently from the same root. From LU to wash (as in lu-ere, and λúOpov), we have lau-tus (but in Greek Nov-ew): from RU, seen in opów, we have raucus: from PU to strike (Greek тa(F)-íw) we have pau-ire and pau-imentum: from the other PU mentioned above we have pa-eni-tet, as well as po-ena: there seems to be even a third root of the same form, expressing rottenness, in pu-tris, pa-edor, and pū-s, pu-tere: it is the German faul, our "foul," in the old sense of crumbling, decayed, which it retains in Cumberland, e.g. Foulsyke: apparently also Foulmire, for foul mere, not far from Cambridge. The derivatives of KUP, to trade, are numerous, the simple form may be found in cupio, and is in caup-o and cop-a, where the change from au to o is regular: but a comparison of κáπ-ŋλos would seem to shew that the original form was KAP: and this is confirmed by the Teutonic: in O. H. G. we have chauf-an, and also kouf-ōn, the modern kaufen: in Danish we have Copenhagen: the variations in English of this root are well known; the names "Chapman," "Copeman," and probably "Cooper;" the market places, "Cheapside," "Chepstow" (contrast Stow-market), " Chipping" Norton; and the verbs to chaffer and to chop (of horses, or of the wind)1.

There is the same quantitative increase of U as of 1; that is, we often find both in Greek and Latin, without being able to say it is a contraction of au, eu, or ou. Thus we have φύσις but φῦλον, κἔτος but σκῦτος, from SKU mentioned above: from DHU we have ovos but Oûμós, and in Latin fumus and funus: we have ruber and rūpes, Rupilius and rupes, pronuba and nubo.

1 See Isaac Taylor, Words and Places, p. 394.

CH. VI.

CH. VI.

NOTE ON CHAPTER VI.

Prof. Whitney dissents from the view which I have given in this Chapter of Dynamic change. He writes: "A marked tendency in the best modern researches, if we are not mistaken, is towards the entire elimination of the symbolical element from the history of the languages of our family, and the recognition of all internal change, whether of vowel or of consonant, as at first only the accidental accompaniment of external accretion, or its remoter euphonic consequence: even though sometimes seized upon later by the language-making faculty, and turned to account in a secondary way, or inorganically." In other words, he thinks that all so-called dynamic change was originally phonetic: as some changes (afterwards used dynamically) undoubtedly were. His arguments are put very briefly, but they deserve careful consideration, and are chiefly these:

(i) "Vowel change comes along with the ordinary means of external derivation." For example, in the instance given in the text, where vid is changed to veda and vaidika, there is in each case a suffix added, as well as a vowel-change. What reason is there, he asks, to regard the vowel-change as all important, the suffix as mere machinery?

(ii) "Vowel-change is not to be brought into any definable relation with ideas or classes of ideas, as their expression." That is, if we take again an example already given, there is no intelligible gradation of idea conveyed by the modification of vowels in fides, feidus, foidus.

1 In a very courteous review of the first edition of this book, in the Journal of the American Oriental Society for 1870.

2 This indeed is not always so. Thus in Sanskrit, from Buddha is formed Bauddha, a Buddhist, with no further suffix. Prof. Whitney will not hold, I imagine (with Sanskrit grammarians), that in such cases there has been a "lopa," i.e. that a suffix has been put on, only to be knocked off again. This shews that a suffix was not absolutely necessary: though no doubt it is far more often used than not.

In reply, I have no doubt that all suffixes once had a meaning. But it certainly seems to me that such meaning was commonly (if not universally) lost before the stage of language of which we are now treating. The Indo-European language, at the earliest period when we become acquainted with it, had passed out of the stage at which each formative accretion had a separate and intelligible meaning, when it could be stripped off again at pleasure from the root, and used separately. This question has been frequently raised already, and I can only refer back to what I have said at pp. 50-52, respecting the origin of the suffixes: see especially the discussion on ma and ta. Therefore it seems to me that the change of meaning must be looked for, either not at all or very slightly, in the suffix; and, either altogether or principally, in the vowel-change.

:

The second argument is more important. In the nouns it certainly must be allowed that the motive of the change is no longer traceable with any certainty. But in the verbs the vowel-change does seem to have been consciously used for a special purpose, that of denoting different stages of action. I grant that there are many irregularities in the application of the principle, chiefly in the a-class, but also in the others. Still, I think that the principle is sufficiently traceable and its use is parallel to that of reduplication: that certainly must have been performed consciously at first; and this seems to me to remove any a priori objection to vowel-change having at first been conscious too. It may be said that changes purely phonetic in their origin have been employed afterwards dynamically. This I have myself assumed in the change of ▲ to a, e, o. But if the change from i to ai and oi be phonetic, it is one which to the best of my knowledge has no parallel in any Indo-European language. It is an increase of sound which certainly requires some explanation. Sometimes vowels are modified by contiguous consonants: thus, in Gothic, i and u become ai and au before r and h. But this change of which we are speaking is not confined to any special combination; it occurs before any consonant indifferently. I have already said that I do not think it can be traced to accent: and that I hold Corssen's view to be probable, that accent and vowel-change were employed quite independently of each other.

CH. VI.

CH. VI.

The confusion of use in Greek and Latin, where the vowelsystem had been increased by è and ŏ, is much greater than in Sanskrit, which remained nearer to the original language. This shews that we are justified in not regarding the irregularities of later times as a conclusive argument against the regular application of the principle in simpler days. The lapse of time must tend ever more and more to obscure the "symbolical element of language." The principle of decay has been at work as now from the very beginning of all speech. But there must have been a creative epoch or epochs also.

Since then I seem to myself to see in these changes a dynamic principle, however often obscured, and as I know of no rea sonable explanation which can be given of the change if merely phonetic, I still adhere to the account which I gave in my first edition. I consider, however, that the question is one of much uncertainty neither view admits of any positive proof.

CHAPTER VII.

VOWEL-CHANGE

WE have now cleared the way for the discussion of the nature and extent of phonetic variation in Greek and Latin-such change of sound as was originally caused purely by the desire for easier articulation, and was not intended to denote any modification of idea, though the new forms may in some few cases have been afterwards so employed. Two different sets of phenomena resulted from this striving for ease of sound. Either a new sound was substituted for the old more difficult sound'; in which case we have the result of Weak1 Articulation: or in consequence of a lazy, perhaps sometimes drawling pronunciation, an entirely new sound became heard in connection with an old one-a sound to which I have already applied the expressive term of Prof. Curtius, "parasitic;" such cases will be considered separately under the head of Indistinct Articulation.

Again, Weak Articulation may be viewed under four different aspects. In all the tendency is the same; but the results are different from the modifying effect of neighbouring sounds in certain cases. Sometimes there seems to be no such cause; here we have cases of pure

1 By the term weak I do not intend to imply any weakness in the character of the speaker. I mean that the organs are put into a position in which less exertion is called forth than the old sound requires: and therefore a new sound is the result. This, I think, may fairly be called weak articulation, whether the speaker be a lazy man who desires to save his trouble, or a busy man who desires to save his time, or an excitable man whose muscles are not under his own control. In each case the organs are not put into the position they ought to be in, but into an easier one..

CH. VII.

Phonetic

change due

to two causes

Weak Arand Indistinct Arti

ticulation,

culation.

Weak Articulation to be first considered under four heads.

« PreviousContinue »