Note on Transcription and Style

In: Sufism in Central Asia
Free access

Note on Transcription and Style

To the usual set of complications surrounding the consistent transcription of Central Asian names and technical terms rooted in multiple languages clothed in the Arabic script, the aim of this volume to bridge the divide between historical and contemporary perspectives adds still more complexity, largely because of the 20th-century crystallization of multiple ‘national’ languages in the region, written in varieties of the Cyrillic script, and the more recent adaptations of the Latin script as well. In specific terms, the present volume involves the use of sources in Persian, Darī, Chaghatay Turkic, Arabic, Pashto, and Russian, as well as in the Tajik, Uzbek (in Cyrillic and Latin scripts), Kyrgyz, and Kazakh national languages that took shape in Soviet times. It may help to clarify both the transcription principles adopted and some of the exceptions and inconsistencies allowed in our attempt to properly represent these many languages and alphabets.

For the transcription of Arabic and Persian, we have adopted the system used in the International Journal of Middle East Studies (IJMES); this includes the privileging of the tripartite Arabic vowel system for Persian contexts (on the assumption that contemporary Persian vowel quality in Tehran may be misleading for medieval and modern Tajik or Darī), but we have dispensed with strict consistency in rendering the same letters in different ways depending on the language context in which they appear: thus, in Persian environments, ث is “s̱”, ذ is “ẕ”, and ض is “ż”, but in fully Arabic environments they are “th”, “dh”, and “ḍ”, respectively. For Chaghatay Turkic, or other Arabic-script Turkic, consonants are rendered as in Persian, but vowels are assumed to reflect vowel-harmony, informed in particular cases by both medieval and modern languages, and distinguishing, on that basis, initial or medial “é” from “i” (both represented by ای - or -ی-), e.g., “bék” instead of “bik.”

For Russian, a modified Library of Congress system is employed (but without the glides or brêves used for ю, я, or й).

The contemporary and Soviet-era national languages naturally present their own specific problems of transcription. Multiple systems have been proposed during the past two decades for rendering Cyrillic-script Turkic languages and Tajik, without any gaining widespread acceptance or significantly improving upon the pioneering transliteration tables assembled by Allworth in 1967. These tables provide the basis for the rendering of Cyrillic-script Kazakh in Ashirbek Muminov’s contribution, and of Cyrillic-script Tajik in Jo-Ann Gross’s study (alongside transcriptions from Arabic, Persian, and Russian). Occasional transcriptions from the Cyrillic Uzbek alphabet that was used through most of the 20th century appear alongside contemporary, post-independence Latin-script Uzbek in the contributions of Yayoi Kawahara and Eren Tasar (transcriptions from Cyrillic-script Kyrgyz also appear in Tasar’s study, together with Cyrillic transcriptions, in Russian or Kyrgyz or Uzbek ‘style,’ of Arabic-script material that still circulated in Soviet-era Central Asia). At the same time, we have mostly opted for the standard forms of ethnonyms currently in use: thus “Uzbek” instead of “Oʻzbek,” “Kyrgyz” instead of the older “Kirghiz” or Russian-based “Kirgiz,” and “Tajik” instead of “Tojik.” In the case of “Qazaq” and “Kazakh,” however, both are employed, the former for contexts prior to the 1920s (in Allen Frank’s contribution, for instance), and the latter reflecting the prevailing international standard at present (in Muminov’s study).

The extensive presence, in these national languages, of names and terminology drawn from Arabic-script Islamic literary traditions on the one hand serves as a constant reminder of the pervasive indigenization of Muslim culture and religion throughout Central Asia, in the past and at present, but on the other hand entails the dual risk of obscuring significant distinctions (we may recall, for instance, the Sovietological conflation of khojas and ḥājjīs) and masking important conceptual and terminological continuities across the supposedly firm barriers of ‘modernity’ and ‘secularism.’ With this in mind, the editors have encouraged contributors, where relevant, to use the forms of particular titles and terms found in the national languages, but to indicate their ‘classical’ Arabic-script equivalents as well.

For one term in particular we have purposefully rejected strict consistency in order to distinguish different connotations of the term: the technical transcription “khwāja” is used when the term is used as a title or honorific for particular individuals, while a form more closely indicative of actual pronunciation, “khoja,” is used to refer to the familial groups claiming descent from specific saints of the past (often alongside contemporary ‘national’ forms, e.g., xoʻja, or qozha). On the other hand, for a small set of commonly occurring ‘religious’ or Sufi terms, we have tended to privilege a more rigorous transcription (Qurʾān, ḥadīth, imām, qāżī, muftī, mullā, ʿulamā or ʿulamāʾ, murīd), unless contributors have included explicit discussion of their usage.

A less rigorous rendering, on the other hand, has been adopted for most geographical names. Relatively well-known place-names, that is, are given without diacritics, but in forms that often do not reflect those used in the official or national languages of the countries in which the sites they designate are presently located, e.g., Bukhara (not Bukhārā or Bukhoro), Tashkent (not Tāshkand or Toshkent), Kashghar (not Kāshghar or Kashgar or Qäshqär or Kashi), Herat, Kabul, Qandahar, Peshawar, Khurasan, Kashmir, Farghana, Syr Darya, Amu Darya; Kazan is used instead of Qāzān, Qazān, or Kazan’. Diacritics are retained, however, for adjectival formations (Khurāsānī, Qandahārī, Kashmīrī, etc.), and in direct transcriptions from Arabic-script sources; diacritics are also retained for less familiar place names (e.g., Badakhshān, Ghaznī, Namangān, Marghīnān, Khwārazm). More broadly, we have not sought to eliminate or obscure what may be regarded as ‘natural’ inconsistencies (along the lines of “Qazaq” vs. “Kazakh”) resulting from shifts in the prevailing ‘official’ archival languages (transcribing from Arabic-script Persian or Turkic in some cases but from Russian in others, e.g., “Namangān” vs. “Namangan,” “Astarkhān” vs. “Astrakhan”), on the assumption that these will pose few problems for most readers.

It is hoped that the resulting mixture of rigor, accessibility, and contextual consistency will help familiarize the general reader with some of the complexities of Sufi religiosity in Central Asia, while also allowing those familiar with Central Asian languages to recognize indigenous terminology as encountered in multiple venues and forms.

  • Collapse
  • Expand