In accordance with the programme laid down in the opening paragraph of Book III, we shall now deal in detail with those linguistic changes which are not due to transference to new individuals. The chapter on woman’s language has served as a kind of bridge between the two main divisions, in so far as the first sections treated of those women’s dialects which were, or were supposed to be, due to the influence of foreigners.
Many theories have been advanced to explain the indubitable fact that languages change in course of time. Some scholars have thought that there ought to be one fundamental cause working in all instances, while others, more sensibly, have maintained that a variety of causes have been and are at work, and that it is not easy to determine which of them has been decisive in each observed case of change. The greatest attention has been given to phonetic change, and in reading some theorists one might almost fancy that sounds were the only thing changeable, or at any rate that phonetic changes were the only ones in language which had to be accounted for. Let us now examine some of the theories advanced.
Sometimes it is asserted that sound changes must have their cause in changes in the anatomical structure of the articulating organs. This theory, however, need not detain us long (see the able discussion in Oertel, p. 194 ff.), for no facts have been alleged to support it, and one does not see why small anatomical variations should cause changes so long as any teacher of languages on the phonetic method is able to teach his pupils practically every speech sound, even those that their own native language has been without for centuries. Besides, many phonetic changes do not at all lead to new sounds being developed or old ones lost, but simply to the old sounds being used in new places or disused in some of the places where they were formerly found. Some tribes have a custom of mutilating their lips or teeth, and that of course must have caused changes in their pronunciation, which are said to have persisted even after the custom was given up. Thus, according to Meinhof (MSA 60) the Yao women insert a big wooden disk within the upper lip, which makes it impossible for them to pronounce [f], and as it is the women that teach their children to speak, the sound of [f] has disappeared from the language, though now it is beginning to reappear in loan-words. It is clear, however, that such customs can have exercised only the very slightest influence on language in general.
XIV.—§ 2. Geography.
Some scholars have believed in an influence exercised by climatic or geographical conditions on the character of the sound system, instancing as evidence the harsh consonants found in the languages of the Caucasus as contrasted with the pleasanter sounds heard in regions more favoured by nature. But this influence cannot be established as a general rule. “The aboriginal inhabitants of the north-west coast of America found subsistence relatively easy in a country abounding in many forms of edible marine life; nor can they be said to have been subjected to rigorous climatic conditions; yet in phonetic harshness their languages rival those of the Caucasus. On the other hand, perhaps no people has ever been subjected to a more forbidding physical environment than the Eskimos, yet the Eskimo language not only impresses one as possessed of a relatively agreeable phonetic system when compared with the languages of the north-west coast, but may even be thought to compare favourably with American Indian languages generally” (Sapir, American Anthropologist, XIV (1912), 234). It would also on this theory be difficult to account for the very considerable linguistic changes which have taken place in historical times in many countries whose climate, etc., cannot during the same period have changed correspondingly.
A geographical theory of sound-shifting was advanced by Heinrich Meyer-Benfey in Zeitschr. f. deutsches Altert. 45 (1901), and has recently been taken up by H. Collitz in Amer. Journal of Philol. 39 (1918), p. 413. Consonant shifting is chiefly found in mountain regions; this is most obvious in the High German shift, which started from the Alpine district of Southern Germany. After leaving the region of the high mountains it gradually decreases in strength; yet it keeps on extending, with steadily diminishing energy, over part of the area of the Franconian dialects. But having reached the plains of Northern Germany, the movement stops. The same theory applies to languages in which a similar shifting is found, e.g. Old and Modern Armenian, the Soho language in South Africa, etc. “However strange it may appear at the first glance,” says Collitz, “that certain consonant changes should depend on geographical surroundings, the connexion is easily understood. The change of media to tenuis and that of tenuis to affricate or aspirate are linked together by a common feature, viz. an increase in the intensity of expiration. As the common cause of both these shiftings we may therefore regard a change in the manner in which breath is used for pronunciation. The habitual use of a larger volume of breath means an increased activity of the lungs. Here we have reached the point where the connexion with geographical or climatic conditions is clear, because nobody will deny that residence in the mountains, especially in the high mountains, stimulates the lungs.”
When this theory was first brought to my notice, I wrote a short footnote on it (PhG 176), in which I treated it with perhaps too little respect, merely mentioning the fact that my countrymen, the Danes, in their flat country were developing exactly the same shift as the High Germans (making p, t, k into strongly aspirated or affricated sounds and unvoicing b, d, g); I then asked ironically whether that might be a consequence of the indubitable fact that an increasing number of Danes every summer go to Switzerland and Norway for their holidays. And even now, after the theory has been endorsed by so able an advocate as Collitz, I fail to see how it can hold water. The induction seems faulty on both sides, for the shift is found among peoples living in plains, and on the other hand it is not shared by all mountain peoples—for example, not by the Italian and Ladin speaking neighbours of the High Germans in the Alps. Besides, the physiological explanation is not impeccable, for walking in the mountains affects the way in which we breathe, that is, it primarily affects the lungs, but the change in the consonants is primarily one not in the lungs, but in the glottis; as the connexion between these two things is not necessary, the whole reasoning is far from being cogent. At any rate, the theory can only with great difficulty be applied to the first Gothonic shift, for how do we know that that started in mountainous regions? and who knows whether the sounds actually found as f, þ and h for original p, t, k, had first been aspirated and affricated stops? It seems much more probable that the transition was a direct one, through slackening and opening of the stoppage, but in that case it has nothing to do with the lungs or way of breathing.
XIV.—§ 3. National Psychology.
We are much more likely to ‘burn,’ as the children say, when, instead of looking for the cause in such outward circumstances, we try to find it in the psychology of those who initiate the change. But this does not amount to endorsing all the explanations of this kind which have found favour with linguists. Thus, since the times of Grimm it has been usual to ascribe the well-known consonant shift to psychological traits believed to be characteristic of the Germans. Grimm says that the sound shift is a consequence of the progressive tendency and desire of liberty found in the Germans (GDS 292); it is due to their courage and pride in the period of the great migration of tribes (ib. 306): “When quiet and morality returned, the sounds remained, and it may be reckoned as evidence of the superior gentleness and moderation of the Gothic, Saxon and Scandinavian tribes that they contented themselves with the first shift, while the wilder force of the High Germans was impelled to the second shift.” (Thus also Westphal.) Curtius finds energy and juvenile vigour in the Germanic sound shift (KZ 2. 331, 1852). Müllenhof saw in the transition from p, t, k to f, þ, h a sign of weakening, the Germans having apparently lost the power of pronouncing the hard stops; while further, the giving up of the aspirated ph, th, kh, bh, dh, gh was due to enervation or indolence. But the succeeding transition from the old b, d, g to p, t, k showed that they had afterwards pulled themselves together to new exertions, and the regularity with which all these changes were carried through evidenced a great steadiness and persevering force (Deutsche Altertumsk. 2. 197). His disciple Wilhelm Scherer saw in the whole history of the German language alternating periods of rise and decline in popular taste; he looked upon sound changes from the æsthetic point of view and ascribed the (second) consonant shift to a feminine period in which consonants were neglected because the nation took pleasure in vocalic sounds.