Abstract:
در بسیاری از زبانهای ایرانی، از جمله زبان فارسی، ستاک گذشتة شماری از فعلهایی که ریشه مختوم به r پس از یک واکه، دارند در حالی به -štختم میشود که r از ساختمان آنها حذفمیگردد. نظر به این که پایة بنیادین ستاک گذشتهساز تکواژ t- است که طی فرایندهایی واژ-واجی، بصورت تکواژگونههای مختلفی نمود پیدامیکند، نگارنده را بر آن داشت تا روند ساختار این تکواژ خاص را که در مطالعات پیشین یا نادیده گرفته شده، یا بدرستی تبیین نشده، بررسی و تحلیل نماید. فرض مطالعه بر این بوده که در خوشة همخوانی نخستی که /rstـ/ بوده، فرایند درهمشدگی [r]ی واکرفته در [s] صورت گرفته و به پیدایی صدای واجشدة /š/ انجامیده است. در شیوة پژوهش، با بهرهگیری از الگوی آواشناختی-محور در دگرگونی واجی شامل برنامهریزی حرکتی در تولید گفتار، آیرودینامیک گفتار، سازوکار حرکت بخشهای دستگاه گفتاری و ادراک گفتار و بویژه آواشناسی تعاملی (IP)، هر دو رویکرد درزمانی و همزمانی مورد نیاز، چه برای تحلیل و چه برای استخراج شواهد بهکار گرفته شدهاست. دادهها در این بررسی همة ستاکهای گذشتة دارای این موضوعیت در زبان فارسی بودهاند. شواهد نظری و تجربی، هم در دادهها و هم در دیگر نمونههای تاریخی یا بهروز، مؤید درستی فرضیه در پیدایی واج š در ستاک گذشته فعل و حذف r از پایة آن بودند. یکی دیگر از نتایجی که این بررسی بدست میدهد این است که تکواژگونة زیرساختی این دسته از ستاکها میبایست /st/ بودهباشد. پیشنهاد میشود این یافته در پژوهشهای آتی، بصورت آزمایشگاهی نیز بررسی شود.
In many Iranian languages including Persian, the past tense stem of some verbs terminates with –st, while their present stem ends with a /r/ which in fact exists in the root; but this /r/ does not appear in the past stem. These verbs are /ænbastæn/, /engastæn/, /pendastæn/, /dastæn/, /kastæn/, /gozastæn/, /gastæn/, /gozæstæn/, /gomastæn/, /negastæn/ and their combinations. Concerning the basic morpheme of │T│ for the past tense of verbs represented in different allomorphs (t, d, st, st, xt, ft, ɑd, ud and id), this research aims to detect the process of formation the -st form of this morpheme in the mentioned verbs. The question is what phonemic processes have led to the deletion of a consonant (r) from the root (which is athwart the definition of "root”) and appearance of /s/ in the past stem. The main motivation for this investigation was lack of convincing reasoning, explanation or even analyzing for the issue. This research based on phonetic factors for sound change containing the elements of motor planning, aerodynamic constraints, gestural mechanics, and especially interactive-phonetic (IP) on one hand, and historical studies on the other hand, has applied both diachronic and synchronic evidences for analysis and explanation of the sound change, and ultimately morpheme formation in the discussed subject. The absence of a consonant in the stem of a verb, in Persian, is a token for being stem constructor, but the presence of a consonant in the root and its absence in only one of the stems, indicates there must be a reason for disappearing the consonant of the root; particularly in past stems due to their construction. In the base of all past stems, “t” or “d” (depended on their environment voicing) is seen, whilst the presence of /s/ (the only phoneme without any vestige neither in the root nor in the base form of stems, unlike other phonemes than “t/d” in the stems) in /st/, has been regarded as the sole “exceptional” past stem morpheme by the scholars, due to not finding any reason for it. The assumption of this study was that the initial allomorph of past stem constructor should have been /st/ which had make the ending consonant cluster /rst/, and then through the telescoping process of the /r/ (unvoiced [ɾ/ɹ]) with [s] has triggered to emerge /s/. The question here is about the fount of /s/ in the stem. Some investigations have pointed out that in the aorist paradigm of the Old Iranian languages including Old Persian, /s/ had been the aorist stem constructor; therefore some stems were ended with /*rst/. According to researchers, this /s/ is still seen in some verbs, despite vanishing of aorist inflection, and in some cases, it has been changed into /s/. This statement, improves our presupposition about the initial allomorph of past stem constructor /st/ instead of /st/ to make ending consonant cluster /rst/ in the past stem. Moreover, one of the most frequent past tense morpheme of many stems in Iranian languages is /-st/, and based on diachronic (e.g. in Indo-European and Indo-Iranian Languages) and synchronic (German dialects and Scandinavian Languages) studies there is a universal sound change of s→s in an especial context. Hereupon, it seems reasonable to emerge such phoneme in the discussed stems, if the condition is fulfilled. One of the triggering conditions has been reported in ruki rule (s → s / r, u, k, i -). Concerning the adjacency of /r/ (the last consonant of the root) and /s/ of the morpheme /st/, the condition is acquired, but the data in Iranian Persian language show a /t/ or /n/ is needed to occur the change. The process of this phonemic change seems to be a phasic telescoping process via saltation (conversion of sound A to C, leaping over a phonetically intermediate sound B) as follows. In order to articulate /r/ at the end of the root and the /s/ of the past tense morpheme, respectively the tip and the blade of the tongue touches the back and the front of the alveolar ridge. In this companion, /s/ gets regressive assimilation with /r/ and becomes retroflex ([ʂ]: intermediate of saltation) by pulling back the tip of the tongue into hard palate, but the alveolar /t/ in /st/ (which is articulated by blade of the tongue in Persian language) removes immediately the retroflex feature by pulling up the tip of the tongue toward the blade to form /s/. Thus based on telescoping process, contrary to the opinion of most scholars, the /r/ of the root has not been omitted but has been merged into /s/ to emerge the sole phoneme /s/: CVr+st →CVst. The results not only explain the sound change in the above mentioned verbs, but also in some other words in Persian (e.g. /gosne/ hungry, /tesne/ thirsty and etc.); and in dialects of Iranian languages (e.g. /sustan/ in Mahabadi Kurdish and /sistan/ in Davani dialect for /sostan/ to wash in Persian). As previously mentioned, in all of the samples of these change a /t/ or /n/ must be after /rs/ (Vrsn/t). Moreover, as regards the morphophonemic study pertains to the phonemic changes in “morphemes via appending”, in case of simple words (free morphemic structure) like gosne (hungry), tesne (thirsty), Pastu (Pashto), pasne (heel) and so forth, this changing does not seem to be a morphophonemic change. Therefore, this view does not consent to most of the studies in this case which remark it as a morphophonemic change; athwart, it accords with all studies and findings in IP model on phonemic change of /s/ to /s/.