CC BY-NC-ND
34
Book Reviews / Рецензии
Anastasiya Krylova
Russian State University for the Humanities (Moscow); [email protected]
А. И. Коган [Anton Kogan].
Проблемы сравнительно-исторического изучения языка кашмири [The problems of comparative-historic study of Kashmiri language].
Moscow: Fond razvitiya fundamental'nykh lingvisticheskikh issledovaniy, 2016. 208 pp. In Russian.
The monograph of the Russian linguist Anton I. Kogan "The problems of comparative-historic study of Kashmiri language" has become a long awaited treat for Indo-Iranian studies. Indeed, the question of attributing Kashmiri to either the Indic or the Dardic sub-branch has occupied the minds of researchers for many decades. Linguists of the past, such as G. Grier-son and G. Morgenstierne, gave ambivalent answers to this question. Nor was the Kashmiri language well-studied from a diachronic perspective, and so the work of Anton Kogan fills in this important gap.
The first chapter of the book is dedicated to the problem of using the philological method for studying the history of Kashmiri. South Asia is a region with a variety of written and literary traditions, and the philological method is widely used in studying the history of Indo-Aryan languages. However, Kashmiri written tradition is only three hundred years old, and there are no reliable written sources for earlier periods. Some researchers considered the poetic inclusions in the Sanskrit philosophical work Mahanayaprakasa and the phrase "Rangassa Helu dinna" from a Sanskrit chronicle of 12th century Rajatarangini as the earliest medieval examples of the Kashmiri language. However, the phrase from Rajatarangini does not reveal any specific features which would distinguish its language from a literary Prakrit. Poetic inclusions in Mahanayaprakasa offer more extensive and interesting material; nonetheless, detailed analysis reveals that it is not possible to establish a system of regular phonetic correspondences between this language and Old Indo-Aryan. The language of the inclusions looks like an artificial literary lect, created with the aid of traditional grammars and dictionaries of Prakrits and Apabhramsha. Such practice was widespread in Indian literary tradition — and it is quite probable that the scribes were speakers of Kashmiri; it is also possible to trace certain elements of Kashmiri influence in the language of Mahanayaprakasa that can be explained as
scribes' mistakes. However, Indo-Aryan historical phonetic development is much more characteristic of these texts in general. Therefore, we do not really know any texts in medieval Kashmiri, and the use of philological method for studying the language is rather limited.
In the second chapter the author studies a number of phonetic changes in Kashmiri and their dating. According to the data of internal reconstruction, regressive assimilation of vowels took place before the fall of final short i- and u-matra, but already after the period of massive Persian lexical influence. Umlaut in Kashmiri has developed as a result of regressive assimilation of vowels and deletion of final short vowels. In addition, this deletion of final short vowels has led to a new phonological opposition of palatalized and non-palatalized consonants. This opposition, distinguishing Kashmiri from most Dardic languages, is, therefore, a relatively late innovation.
Hesitation in attributing Kashmiri to Dardic languages is due to copious borrowings from Indo-Aryan languages, as well as a number of characteristic typological features that distinguish Kashmiri from most languages in the Dardic group. Thus, in Kashmiri, the opposition of affricates by place of articulation is twofold (dentals and palatals), rather than three-fold (dentals, palatals, and retroflexives). This brings the system of Kashmiri consonantism closer to certain dialects of Western Pahari.
Based on external comparison, the author convincingly shows that this situation is the result of transition of retroflex consonants into palatals, and then, at a relatively later stage, of palatals into dentals. Mor-phophonological alternations and comparison with Shina and Phalura languages indicate that dental affricates existed in Kashmiri during the period preceding the transition. Therefore, until relatively recently Kashmiri must have had three rows of affricates.
Finally, the last section of the second chapter is dedicated to the shift of sibilants. In most cases,
Journal of Language Relationship • Вопросы языкового родства • 14/2 (2016) • Pp. 153—155
Book Reviews / Рецензии
Common Aryan *s corresponds to Kashmiri h. The palatal sibilant s of Kashmiri corresponds to s in most Dardic languages. The contemporary state of the systems of sibilants and affricates in Kashmiri can possibly be explained by the influence of Modern Indian languages, where retroflex consonants are neither affricates nor sibilants. It is typologically unlikely that the shift of sibilants preceded the shift of affricates, although we do not have a firm basis for relative dating.
Therefore, the main differences of Kashmiri phonology, compared to other Dardic languages, turn out to be a result of late changes.
In the third chapter, Kogan analyzes Indo-Aryan loanwords in Kashmiri. Their percentage in Kashmiri vocabulary is quite high, but definitively identifying these words in the absence of formal criteria is a difficult task. Easily identifiable strata are loanwords from Urdu, which became an official language in Kashmir in 1889, and Sanskritisms in Indian Kashmiri. Other borrowings from Indo-Aryan languages require more complicated analysis. The author proposes the following criteria to distinguish them:
1) a front vowel corresponding to Old Indo-Aryan e and common Iranian *ai (the regular Kashmiri reflex is a);
2) laryngeal h corresponding to Old Indo-Aryan h and common Iranian *j (the regular Kashmiri reflex is z);
3) h corresponding to Old Indo-Aryan *s (for borrowings that took place before the shift of sibilants);
4) s corresponding to Old Indo-Aryan *s (for borrowings that happened after the shift of sibilants);
5) kh corresponding to Old Indo-Aryan *ks;
6) etymological parallels existing in Indo-Aryan, but absent in Dardic languages;
7) d and t corresponding to *rd and *rt (regular reflexes are d and i);
8) sequences of a + voiceless consonants corresponding to sequences a + nasal + voiceless (the regular etymological reflex should be voiced).
Combination of features (4) and (8) in the same words allows us to suppose that the source language may have belonged to the Pahari group, where some languages have preserved the distinction of sibilants s and s, and have also undergone fronting of *a before consonant clusters of the "nasal + voiceless" type. Additionally, a number of semantic and morphological isoglosses that unite Kashmiri with Indo-Aryan languages could be the result of Indo-Aryan influence. The author supposes that Kashmiri was likely influenced by an Indo-Aryan substrate language that was
common in the Kashmir valley before becoming assimilated by the Dardic-speaking population.
The fourth chapter establishes the genealogy of Kashmiri dialects. The Siraji and Rambani dialects, which Grierson considered as mixed, can be attributed to Indo-Aryan based on a number of features. Thus, Proto-Aryan short ai is reflected as i and e in these dialects. Besides, they have voiced aspirates which usually correspond to voiced aspirates in Indo-Aryan languages. The distinction of dental and palatal affricates in Siraji and Rambani is not an exclusively Dardic feature; it is also characteristic of certain Pahari languages. As to the morphological and lexical features that Siraji and Rambani have in common with Kashmiri (pronominal suffixes, the stem of the copula etc.), this also does not seem a sufficiently solid basis for classification. The author makes his final decision upon conducting lexicostatistical analysis based on Swadesh's 100-item wordlists. The mean percentage of matches between Siraji and Indo-Aryan languages is 68,6%, between Siraji and Dardic languages — 50,6%, which allows to classify Siraji (and the closely related Rambani) as an Indo-Aryan language.
On the other hand, the Poguli and Kashtavari dialects should, most probably, be attributed to the Dardic group. Thus, voiced aspirated consonants are found mostly in Indo-Aryan borrowings. Some cases of development *s > h are explained individually by the author. It is surprising that the author does not apply lexicostatistic analysis to this pair of dialects as well, but dives instead into the explanation of examples that contradict his hypothesis by means of analogies, metatheses, contaminations, etc. This leaves an impression of asymmetric composition and somewhat inarticulate evidence. Another strange peculiarity is the urge to prove the originality of basic vocabulary even in those cases where the phonetic form of the word clearly indicates a borrowing (pp. 67, 127). Apparently, though, these details do not affect the author's final conclusions.
In the fifth chapter the question of Eastern Dardic linguistic unity, as identified by G. Grierson, is researched. The author criticizes the historical-phonetic innovations, proposed by G. Buddruss, which supposedly unite the Eastern Dardic languages (the shifts *w > b, *st > t(h) and *st > t(h)), noting that the first two cannot be considered common for Kashmiri, Phalura, and Shina languages. Lexical isoglosses identifying Eastern Dardic among other Dardic languages unite them with Indo-Aryan languages. This certainly raises suspicions that the areal cohesion of different Dardic languages could be a consequence of common Indo-Aryan influence.
Anton KOGAN. The problems of comparative-historic study of Kashmiri language
The author's own criteria are as follows: (a) front shift sr > s; (b) voicing of voiceless consonants; (c) deletion of voiced consonants after a nasal; (d) the fate of the Proto-Indo-European cluster *ks, which is reflected in Eastern Dardic languages as the palatal affricate ch, and its later developments in the original intervocalic position before a short vowel in the last syllable.
In the concluding section of the fifth chapter a lexi-costatistical research is conducted, utilizing such methods as "nearest neighbors" and "least mean deviation». In both cases lexicostatistics confirms the fact of close genetic affinity between Eastern Dardic languages (Kashmiri, Shina, and the languages of Kohis-tan). Languages of Kohistan share a high percentage of common vocabulary with Kashmiri and Shina. At
the same time, the fraction of correspondences between the Kashmiri and Shina lists is rather small. These facts allow us to suppose migration of the speakers of medieval Kashmiri from the Swat river valley to their modern habitat.
Overall, the monograph is a fascinating piece of historical research on the Kashmiri language, distinguished by the variety of methods employed by the author. Several shortcomings in the book's design slightly hinder the ease of comprehension: for example, only in the fourth chapter do etymological examples begin to be regularly separated from each other by paragraph marks. However, this technical glitch should not detract the reader from the substantial merits of the book.
