印欧祖語のアップラウトと文法構造の発達

The origin of the Proto-Indo-European (hereafter PIE) innate vowel gradation or so-called Ablaut is still troublesome in Indo-European comparalive linguistics despite valuable contributions made particularly by Brugmann, Saussure, Kurylowicz, etc. We must be aware, in the first place, that our predecessors failed to clarify the origin, because 1) the majority of them wrongly took it for granted that PIE is a completely stable system which remained unchanged for thousands of years; 2) they have tried to approach the phenomenon exclusively from the phonological point of view, even though it often plays a significant role in the field of morphology and/or syntax. As long as PIE was a living language, there must have been certain concrete processes that have brought about the Ablaut. We would like to propose the following consecutive phonological and/or grammatical developments, which we believe could clarify the origin of the phenomenon: 0. The earliest stage imaginable of PIE comprised only monosyllables (s)Ce(R)C or (s)C(R)eC (C: consonant, R: resonant), which always included the only original vowel ^*e. There was no concept of part of speech: any mnnosyllabic "word" could be used either as a noun or a verb depending on the context. 1. Particle ^*-(H_1)es (assumed to be from an ex-independent word ^*H_1es "exist(ence)") gradually came to be employed in a transitive construction so as to ensure that the word ending in ^*-(H_1)es functioned as the agent of the given action (creative). This opened up various derivational processes: more and more plurisyllables were created. 2. Stress emerged. The vowel in the stressed syllable was preserved, while that in the unstressed syllable reduced to ^*[〓], which is a common phenomenon. 3. The reduced vowel ^*[〓] reduced further and disappeared when the syllable comprised the resonant, which, again a common phenomenon seen in English mountain [mauntn](including a syllabic n), etc., now produced new syllabic nuclei: ^*i, ^*u, ^*r, ^*l, ^*m, ^*n. 4. The reduced vowel ^*[〓]flanked by obstruents, which, theoretically speaking, allows no further reduction process, was inevitably preserved as the second vowel and finally became ^*o, probably gaining the roundedness that marked the secondary character of its origin. On the contrary, ^*[〓]immediately followed or preceded by so-called laryngeals ^*H_1, ^*H_2, ^*H_3 remained unchanged. 5. Morphological development. There arose the formal distinction of nouns and verbs. ^*O tended to be employed to mark the new derivative forms as secondary. 6. ^*H_1, ^*H_2, ^*H_3 (phonetically assumed to be [h, X. X^W]respectively) coalesced to ^*H and the adjacent vowel took over the articulatory characteristics of the laryngeals: i.e. ^*eH_1, ^*eH_2, ^*eH_3, > ^*eH, ^*aH, ^*oH ; ^*H_1e, ^*H_2e, ^*H_3e > ^*He,^*Ha, ^*Ho. Thus the third PIE distinctive vowel ^a came into being. 7. The residue of the laryngeal element ^*H was completely lost, since it no longer fulfilled any distinctive function, now that the distinctive vowels already ensured the contrastiveness between the syllables which used to comprise the laryngeals. The loss of ^*H after the vowel caused compensatory lengthening of the vowel, while the process did not have any influence when ^*H preceded. Thus ^*eH, ^*aH, ^*pH > ^*e, ^*a, ^*o ; ^*He, ^*Ha > ^*e, ^*a, ^*o. This process produced distinctive length in PIE and there also arose long syllable resonants ^*i, ^*u, ^*r, ^*i ^*m, ^*n. Further compensatory lengthening gave rise to the quantitative alternation seen in ^*P[〓]ter-s > ^*p〓ter (Gk. nom. πατηρ) vs. ^p〓ter m (ace. πατερα) and the like. 8. The reduced vowel ^*[〓], which was adjacent to ^*H_1, ^*H_2, ^*H_3, survived till these laryngeals were completely lost. It was at this stage that it obtained a full phonemic status and here arose the fourth PIE vowel phoneme ^*∂(schwa indogermanicum). 9. Break-up of the PIE linguistic unity. The initial ^*∂ was lost in all the branches of the family with the exception of Greek. It merged with ^*a elsewhere. Further changes arose in each independent branch. These consecutive processes could successfully explain the gradual growth of the PIE vowel system, the isolated status of ^a, and such mysterious alternations as -*e/^*o/zero, ^*e/^*∂, ^*a/^*∂, *^0/^*∂, ^*e/^*e, ^*o/^*o.