Page images
PDF
EPUB

change of meaning was intended to be expressed by this change of sound. But these sounds, found ready to hand, were employed by the Greek with marvellous skill. Thus, in our present example, the original padas could be differentiated into Todós for the gen. sing., wódes for the nom. plur., and Tódas for the acc. plur. No confusion between the different cases was any longer possible. The weakening of a into a, e, o, was turned in this instance by the Greek into clear gain; as in many others which will be given in their place. At present I turn back from the results of phonetic change, to repeat its cause-the desire for ease or saving of sound; and its general effect—to substitute a weaker for a stronger sound. This is not always so, for reasons which I shall shortly point out; but the new sound will always be an easier one for the speaker to pronounce under the circumstances in which it occurs. I say, for the speaker: because there are few sounds of which it can be said that they are absolutely easier or more difficult than others. Every one knows what contradictory variations may be met with among his own acquaintance: one will pronouncer as 7, another (though very much more rarely) l as r. Similar differences occur on a large scale in different nations; the Englishman avoids in divers ways the German ch: the German finds great difficulty in our th. But as with the individual, so is it with the nation. It is the desire to avoid a sound difficult to him, which makes a man pronounce r as l. And I do not know any other reason which can be given for the loss by a whole nation of some sound which was certainly uttered by their forefathers.

CH. I.

different changes in different

It may perhaps be asked what special causes deter- Peculiar mined the different operation of this principle in different causes of languages. This question—which amounts to an enquiry into the causes of diversity of language itself cannot of course be fully answered here. No people has preserved unchanged all the letters of the original alphabet. Different peoples have modified it in different ways from

languages.

CH. I.

causes at which we can give probable guesses, but which we can never certainly know. Occasionally we may see in the altered alphabet something which seems to correspond to the genius of the people which spoke it, or to be due to the country, climate and general circumstances among which they were placed. Thus we may think that we can see in the flexibility of the Greek language the impress of the versatility of the Greek genius, and the effect of that λαμπρότατος αἰθήρ amid which at least the most brilliant section of the Greek family lived: whilst the effect of the hot enervating climate of India may be seen in the numerous weakened forms of the consonants in Sanskrit. Thus lacking energy to bring the root of the tongue firmly against the back of the palate, the Hindus produced in some cases instead of the original k a peculiar sibilant (denoted variously in philological works by s or c). In like manner, probably through the influence of an adjoining s, they weakened k into ch, and g into j, the sound of ch and j being much the same as in England. Such weakenings are especially common in Sanskrit and that they are due to some extent to the climate of India would probably be denied by few. Within the same language we may see variations arising from difference of occupation or circumstances. The different ways in which men have to exercise their voices will affect certain classes of sounds: and these differences, if found among a considerable body of people within the same. area, have a great tendency to be perpetuated'.

1 Prof. Jos. B. Mayor in a review of the first edition of this work (Camb. Journal of Philology, No. 6) holds that the causes of difference of articulation may be roughly classified as "mental, physical, and circumstantial." By the first he means "excitability, vehemence, nervousness, preciseness, artistic sensibility, the analogical disposition always seeking after resemblances, and its opposite which we may call the analytical disposition, always seeking after differences" (p. 332). I certainly should not deny that all these causes have weight; but they are personal, not peculiar to any large body of people living together: therefore they have little tendency to perpetuate themselves, and affect language: they die out with the individual. A quick excitable person may drop half his syllables, a man of "artistic sensibility" may have some

It is not indeed always easy to say where the effect of climate may be traced. In England we see much the same weakenings as in Sanskrit. In different parts of the island we find the hard k sound of the Roman castrum either retained, or weakened to ch or soft c: we have Caistor and Lancaster, but Manchester and Dorchester, and weakest of all Leicester and Gloucester. But these varieties are perhaps less likely to be due to the effects of climate in Britain, than to the mixture of different tribes, each of which had its own phonetic laws before it left its original abode1. Still (to return to our point) these last forms are the result of a weaker articulation; they are corruptions of the harder sound; it is not a strengthened form of one of them. Similar corruptions in English are our pronunciation of Ocean as Oshan, and Nature as Nachure, and a thousand other instances which will at once occur to every one. If the spelling in England were not in the main fixed by the standard of the literary dialect, these words would long ago have been written as they are pronounced. In countries where there is no literary dialect, or where there are several, but no one distinctly predominant, variation of spelling is the inevitable result. When writing, and still more when printing has become uniappropriate method of pronunciation: but in neither case is the peculiarity caught by a man's companions or commonly inherited by his descendants. The more intellectual causes referred to by Prof. Mayor, act, I think, principally on mixed languages: they will be noticed at the end of this chapter. Under the second class-physical causes-are given "dullness of hearing and defectiveness in the organs of speech." With this, I quite agree, and also with the third class, in which are mentioned the effect of cold, living in the open air as a labourer, a hunter, or a sailor, &c. Under all these circumstances modifications of speech will take place: they are all particular instances of my general principle.

Prof. Mayor objects strongly to what he calls the libel that "man as a speaking animal is actuated only by laziness." I never said or thought he was. The term "laziness" hardly occurred half-a-dozen times in the book; but I fully allow that some of those passages were calculated to mislead, and I have altered them. I constantly spoke of the desire for an easier sound. But a man is not necessarily lazy because he goes by an easier road instead of a hard one, or because he takes a short cut.

66

1 We find "ceaster" in A.-S. to which the ch is often attributed. But this is not always the case. Thus our calf" is A.-S. "cealf," "cold" is "ceald." Mr Skeat thinks that the softening to ch may be due to Norman influence.

CH. I.

CH. I.

Application of this principle.

versal, the progress of phonetic change is considerably checked; but how much still goes on will be evident to any one who will consider the difference between the English of Chaucer and that of the present generation1.

Now what is the importance of the principle of phonetic change which I have stated? Its importance is this-it is our best guide in etymology. We learn from it that we must hold it a rule, never to derive a harder from an easier sound; that a word which has retained a strong letter can only under exceptional circumstances be derived from another word which has a corresponding weaker letter. I have said above that few sounds are universally easier than others. There is no standard to fix the relative strength of all sounds available for all languages. Still there are some general rules which can be obtained by two kinds of evidence, physiological and historical. I shall describe in the fourth chapter the methods by which the different sounds are produced and shew from their character what interchange of them is a priori to be expected in any given language. It will there be shewn that, for example, k is a stronger sound than p; that is, that k demands a larger amount of muscular exertion to produce it with the same intensity as p; the check is applied to the current of air issuing from the lungs at an earlier point in its course; and for this reason (with others less obvious) the sound requires more effort to pronounce. In harmony with this is the historical fact that in Sanskrit, Greek, Latin, and Gothic, the gutturals are found less frequently than the dentals or labials; and we should naturally expect those letters to be more sparingly used which required the largest amount of labour in production; they would either be not employed at all, or would pass into easier sounds, or be altogether dropped, in words in much use, like pronouns, or in suffixes where neatness

1 See the tables in A. J. Ellis' Early English Pronunciation, Vol. I. p. 28.

2 Curtius, Griechische Etymologie, p. 407.

and convenience were essential. Again, in many languages we find by-forms, weaker gutturals existing beside and sometimes superseding the full gutturals k and g: while we do not find similar by-forms of the labials to anything like the same extent. Accordingly from these two distinct lines of reasoning-the a priori road of physiology, and the more positive arguments supplied by observed facts in different languages which are not operating the one upon the other-we infer that k is always a stronger sound than p for our group of languages, and we are justified in applying that result to any language of the group. For example, in Greek we shall conclude that κoîos is an older form than Toĉos; that Tolos must be derived from Kolos, not vice versa. So in Latin where we find side by side the words coquina and popina, we shall conclude that popina is a later, probably dialectical, variation of coquina which at an early period fell out of use at Rome, but was originally, as Varro tells us, used for a kitchen; and we shall see a possible reason for the change in the parasitic labial sound u which forms no part of either root or suffix, which had power to assimilate the final c of the √ coc (whence cocus, &c.), and so turn the guttural to a labial : which in turn assimilated also the initial c.

Thus then the general principle of phonetic change, and the general directions which such change will take, are given by comparative philology assisted by physiology. But different peoples varied much in the extent to which they proceeded along these different paths of change. Thus the Greeks made but one variation in dealing with the original aspirates; then they stopped, and the Greek aspirates are used with as much regularity as those of the original language. The Italians on the contrary, feeling the aspirates too difficult sounds, allowed them to degrade so completely, that the single Italian spirant frepresents not only the labial aspirate bh, but dh not unfrequently, and occasionally even gh. On the other hand the Greeks. have thoroughly weakened the spirants y, s, v; the Italians

CH. I.

« PreviousContinue »