v of the opinion that each language is :tain linguistic tendency, that the general 1 of any given territory is a very stubborn bit, a phenomenon of long standing; hence in assuming that our Germanic and Anglos pronounced their words very much as we do t is the chief tendency of English and whence e? At a very early period in Germanic speech became fixed on the first, or root, syllable of the Iwe can observe from that time on a steady tendency oncentrating the stress more and more on this syllable, gradual neglect of following syllables. The natural dation of this tendency would result in a further conrating of the accent on the first part of the vowel, to the fect of its latter part. Is not this the actual state of affairs . modern English? The diphthongal pronunciation of long Vowels is the most characteristic feature of English speech; and diphthongal pronunciation means merely that we stress the beginning, but not the end, of the vowel. How long have we been pronouncing in this way? The history of vowel development would lead us to infer that this manner of pronouncing has obtained for some centuries, because such a speech habit would naturally bring about just exactly those sound-changes which have actually occurred since Anglo-Saxon times. This statement will be more easily understood if we first recall a few facts concerning the nature of vowel formation and of accent.

4. Differences in vowels depend upon differences in the shape of the mouth, which is determined by the elevation of the tongue, the width of the angle of the jaw, and the shape of the lips; thus, to form u the tongue is raised higher in the back of the mouth than for o; similarly i represents higher front elevation than e; that is, the passage between the tongue and the roof of the mouth is narrower for i and u than for e and o, and is widest for a. The widening and narrowing of the mouth passage is accomplished by two sets of muscles, those of the tongue and those of the jaw. Ordinarily scholars attach too little importance to the angle of the jaw in vowel formation. Of course the vowels can be formed by the tongue alone, as can be tested by holding a pencil between the teeth, thus insuring no change in the angle of the jaw; but in actual practice the

THE CAUSE OF LONG VOWEL CHANGES

1. In Modern Philology for June, 1920, Prokosch quotes two passages, one from Sc other from Vendryes, which, he says, express cal platform more clearly and forcibly than They also express my view, and I should a motto at the beginning of this paper. translation, as follows:

What meaning have all the thousands of s remain isolated, as long as they are not brought ciples? . . . In the single phenomenon we must . and consequently, the recognition of a fact wh language is of greater importance than the rec

non.

(Schuchardt, Über die Li

Every phonetic change may be consider and mysterious forces, to which may well these tendencies that constantly modify the evolution of every idiom results, in its play of tendencies. . . The idea of a pho and more fecund in practice than that of determine accurately the cause of phoi cally even those which seem the most un (Vendryes, M

2. The most important s in the development of Old Ex. of the long vowels, ā>ō, ō> ü OE stan, MnE stōne; OE; OE fet, MnE feet; OE hūs. state the changes in ter ¿ has become the mid be become the high vowe!. become the diphthongs vowels have become dipthongized. It has know why such chan changes is very stri the conviction tha tendency.

〃

muscles are

And this closed to the roof of the fan for e.

made in the prespeech exhibits a one syllable and part of the vowel, aw muscles relax and see if we cannot find anges which the long ergone during the last Eng vowel sound è, as in

we stress only the first aw muscles and opening e owel we relax the tension partially closes; this partial art of the e-sound narrows

e palate just about up to cefect of the latter part of can of a pure e. Gradually e vowel encroaches more and until its quality predominates Characteristic element of the As losing any of its traditional publy, the long vowel è passes In this way, then, it seems to me,

e the Modern English feet. In r have ō> (fōd>food), and ā>ō we say now of the diphthongizing du? I believe that these changes also lt of the same tendency, namely, the e accent more and more to the very beginto attack suddenly and to release the stress Case of the two high vowels i and ū the tongue so high above the normal low position of rest, n initial stress-attack catches the tongue while p to this high position and produces first the i and óu respectively. The further advance of y results in the still earlier attack, while the tongue arting up to the i or u position, giving the diphthongs he stress is all expended on this first part of the movealthough the tongue goes on up weakly to the position ally aimed at.

o much for the development of the long vowels since times. Let us now consider the pre-AS development of the diphthongs ai and au to ā and ea respectively. I believe that these changes also are due to this same English tendency to stress only the first part of the sound. I should say first, however, that the a of the diphthong ai is a back a, while the a of au is front or mixed; that is, if the second element of the diphthong represents high front tongue elevation, as in ai, then the first element a is produced by a slight back elevation, and vice versa, in au the first element is produced by a slight front elevation. This is in accordance with the principle of antagonistic muscular reaction. But to return to the development of ai, and au; as the first part of the diphthong is stressed more and more, to the neglect of the second part, this latter part ceases to represent even a slight antagonistic contraction, that is, ceases to be a tongue elevation opposite to that of the first element, and becomes merely the same sort of diphthongal narrowing that we have described above in connection with the long vowels; in other words, the sounds become English long vowels and continue to develop as such.

If all this be true, then we have in English a scale of vowels and diphthongs steadily running, through the centuries, a sort

of double reverse curve, now up in the direction of the front vowels, now up in the direction of the back vowels, thus:

ā ē i ai ā ō û au @, etc.

6. Just a few words, now, in conclusion, concerning the older conception of phonetic laws and the newer one of language tendency. It was formerly maintained that a phonetic law is merely a statement of a historic occurrence, a statement that such and such a sound-change occurred at a certain time and within a given territory, and then ceased. But an example or two from the history of English will make it plain that one and the same change has occurred at different times, and will render more plausible the theory that all of these major vowelchanges are the result of a steady tendency which has existed from very early times, and not the result of some mysterious force that has appeared at one time and then disappeared from the language. Thus the AS >ē (in 16th-17th centuries) and then in modern English (1750+)>i (dād>deed), whereas AS reached the stage in 16th-17th centuries; AS a plus pal. spir (eage)>ME i>MnE ai; and again, the & which arose later from an older ă (as in AS bacan, wadan, MnE bake, wade) did not get such an early start in developing toward i, and has advanced only to the e-stage. Can we not predict that some day bake may become beek; that, in fact, all the vowel changes which have occurred since AS times may be repeated? They will, unless the counteracting forces of conservatism, which increase with general education, prove to be stronger than our natural speech-tendency.

University of Cincinnati

C. M. LOTSPEICH