CC BY
24
Attila Rakos
Eotvos Lorand University, Budapest
CLEAR SCRIPT AS SOURCE FOR THE HISTORY OF OIRAT DIALECTS
1. Introduction
Studying the history of a language is highly dependent on the existence and reliability of written monuments of the given language. The value and usefulness of these written sources can be very different depending on several factors. Some of them provide a lot of reliable information, while others are almost useless for certain purposes. Naturally, type of the writing system and the content deeply influence the extent of information that can be extracted from a source, but these are not the only aspects that matter. Hereby I provide some thoughts concerning these aspects in general and also closely in reference to monuments in the Oirat Clear script and their usefulness in the study of the history of Oirat language and its dialects. We should make distinction between written and spoken languages — here I am going to focus on monuments written in Clear script as sources for the history of spoken Oirat dialects.
Why is Oirat script so important for studying the history of Oirat language? Because its monuments represent the largest corpus of sources related to the Oirat language from the creation of Oirat script in 1648 by Zaya Pandita until the beginning of the 20th century. Although correspondence between written texts and contemporary spoken dialects is often quite loose, the proper and careful analysis of sources in Oirat script can provide new and valuable data on the Oirat language's history, especially if data from sources in other scripts (e. g. Latin, Cyrillic, and Arabic) are also involved.
2. Open questions concerning the history of Oirat
No general and comprehensive work has been published on the history of spoken Oirat dialects, and evidently there are a lot of questions to be cleared up in this field. Some minor publications
dealing with narrow details have been published, and we have an imposing book of D. A. Susejeva on the letters of Kalmyk khans and their language focusing on the 18th century [Susejeva 2009]. Gerhard Doerfer also has published a really useful book with facsimiles of several European sources on Oirat and Kalmyk languages dated between 1692 and 1827, and he also drew a draft on some aspects of the Oirat language's history in his book [Doerfer 1965]. Of course, publications dealing with the history of Mongolic languages contain several useful information concerning Oirat, but we still have unsolved questions. Some additional material that has been published recently can push the researches further: the publication of Cornelius Rahmn's Kalmyk grammar and dictionary from the beginning of the 19th century by Jan-Olof Svantesson [2009, 2012], and also the Kalmyk-Khalkha grammar and Kalmyk field-work material of Gabor Balint of Szentkatolna collected between 1871-1872 and published by Agnes Birtalan [Balint 2009; Birtalan (ed.) 2011].
Thorough analysis of available sources is inescapable in order to make clear and well-grounded statements concerning the history of Oirat. Creating the full image requires analysis of several well-known and little-known changes occurring in present day dialects of Oirat including (but not limited to) the following ones:
Phonetic and phonological changes:
• palatalizing effect exercised on illabial and labial back vowels (/a/ and /u/, /o/) by /i/ of non-first syllables, and its impact on the set of phonemes;
• breaking of /i/ in the first syllables;
• formation of long vowels from VCV complexes, paying particular attention to cases where V1 ^ V2 and Oirat script uses ou and ou;
• formation of long vowels from diphthongs;
• shortening and reduction of vowels in non-first syllables;
• labializing effect of labial consonants exercised on vowels;
• possible changes in quality of /0/ and /u/ (neologism or archaism);
• changes of Common Mongolian affricates /c/ and /J/;
• spirantization of /k/ in back vowel words and in the suffix of nomen futuri;
• loss of the initial consonant in the suffix of the perfect verbal noun (nomen perfecti).
Changes in morphology and syntax:
• evolution of personal possessive markers;
• evolution of personal predicative markers;
• grammaticalization of the verb bayi- 'to be';
• evolution of negation, changing role of ugei, bisi, ese and ulu.
Lexical changes to be examined mostly concern borrowings from Turkic, Chinese (especially in Xinjiang and Qinghai), Russian (especially in Kalmykia), Mongolian Proper (Xinjiang and Western Mongolia) and Tibetan (Qinghai). Researches on the influence of these languages however should not be limited to the lexicon since they exercised influence on the Oirat sound system, morphology and syntax, too.
The two main research tasks to be done concerning these processes enlisted above are
• determining their absolute and relative chronology;
• separating original features of Oirat (heritage of Common or Middle Mongolian) from its new inventions.
Although some attempts have been made already for solving some of these questions, I think there is much to clear up in more details.
3. The underlying language
In order to put to use an Oirat script source, first we have to determine its underlying language, the language in which it was written. Although Sanskrit or Tibetan glosses are common in some sources and even Telengits have used the Oirat script for a while [Vladimircov 1929: 26, Luvsanbaldan 1975: 61], we can exclude here non-Mongolic languages since our focus is the history of Oirat. So the language is clearly Mongolic or even Mongolian, but saying that sources in Oirat script are simply written representations of the Oirat language is undoubtedly an over-simplification. There are at least two questions to be examined here more thoroughly: (1) Was the Oirat script used for writing a Mongolian dialect other than Oirat? (2) What does Oirat really mean?
Let's begin with answering the second question since it is required for answering the first one. Oirat is the language of several Western Mongolian groups known in general as Oirats and Kalmyks and living scattered in Asia from Western Mongolia through China's Xinjiang to the Eastern shore of the Caspian Sea. Their present-day
living spoken language has several non-standardized dialects (Dorbot, Torgut, Khoshut, Bayit, etc.) and a standardized one used in Kalmykia as an official language. Oirat also has newer and older written forms (some of them are obsolete) in three writing systems (Oirat, Latin and Cyrillic scripts) with additional orthographies and variants. Latin and Cyrillic based variants will not be discussed here, only the variants written in Oirat script and usually called Written Oirat collectively. Written Oirat has a standardized, but today non-official variant used in Xinjiang and based on the somewhat modified form of Zaya Pandita's Clear script. Since it is a quite recent variant, it is much less important for studying the history of Oirat and will be neglected here. What is more important for us, is the corpus of sources written in Oirat script and datable between 1648 and the beginning of the 20th century. These are sources in so-called Written Oirat, but what is Written Oirat really? Does it correspond to one or another spoken Oirat dialect? Is it something uniform, unvarying and non-changing? If it did change, what was its original form, how did it change and why?
When a language gets written down first time, its written form directly reflects one of the variants and registers of the living spoken language. This direct correspondence however usually ceases and becomes more distant. Changes of the spoken language are reflected in the written texts to some extent, but written language becomes an independent register or variant (or even a dialect) of the given language with its own life and own changes — or on the contrary it will be characterized by conservatism and lack of changes. This is the case with the adopting of the Uighur alphabet to Mongolian: the written form became separate and — due to its conservatism — quite distant from the spoken language. The ambiguous letters of the Uighur-Mongolian script and the growing distance between the spelling and the pronunciation have lead Zaya Pandita (1599-1662), a Buddhist monk-scholar of Oirat origin, to the idea of a reform. He has created new diacritics, letters and modified variants of the existing graphemes in order to write the sounds of the contemporary language unambiguously, and also simplified the system of positional allographs.
When Zaya Pandita has invented his new script, he also created a new literary language — now usually we call it (and its descendants) Written Oirat, but in Zaya Pandita's time the language was called simply Mongolian, while the script was referred to as todorxoi uzuq or
todo uzuq 'clear script'. This literary language and its orthography were closer to the contemporary spoken language than Written Mongolian in some aspects (e. g. indicating long vowels with separate letters), but had a plenty of features very far from the spoken idioms (e. g. neglecting vowel harmony in the case of certain suffixes, excessive use of verb uyiledku 'to perform, to do' not typical of Mongolian). It is clearly not true that Zaya Pandita's literary language is the direct equivalent of the "contemporary Western Mongolian" and that "he has handed down to us the exact pronunciation of Western Mongolian in the middle of the 17th century" as stated by Udo Posch [1957: 209] and some other scholars. Gyorgy Kara describes it as a mix of colloquial and bookish elements [Kara 2005: 148], which is true for its later forms, but less characteristic of the original variant appearing in Buddhist translations. The very original idea of Zaya Pandita is probably closer to B. Ja. Vladimircov's opinion, who writes that the language of Zaya Pandita's Buddhist translations is a purely artificial creation, and it is not even similar to any Oirat dialect [Vladimircov 1929: 25-26].
Zaya Pandita's aim with reforming the Uighur-Mongolian script and creating a new literary language was not to provide a more precise and effective writing system for the Oirats only, but he targeted the whole Mongolian community. As X. Luvsanbaldan cites from Uzugiyin nayiralya 'Composition of letters' ascribed to Zaya Pandita, he created his script in order to make it easy for the Mongols — and no Oirats are mentioned here [Luvsanbaldan 1975: 23-24]. The political situation in the mid-17th century pointed towards a possible unity or at least closer alliance of Eastern and Western Mongols (against the Manchus), the emerging Buddhism and its strengthening positions also made a positive atmosphere for cultural innovations, and as it is obvious from his life and travels to various Eastern and Western Mongolian territories, Zaya Pandita himself had a wider perspective, so it is quite plausible that he created his script for all the Mongols. Why did it not succeed as he expected? That has probably several reasons, but surely not because his script was not suitable for writing Eastern Mongolian dialects or because his literary language contained too much features taken from the spoken Oirat dialects and differing slightly from the Eastern dialects. The main point here is that if Zaya Pandita's literary
language was created for all Mongols, it could not rely significantly on an Oirat dialect, but had to bear a common Mongolian character.
As such I suppose that Written Oirat, the literary language created by Zaya Pandita for mostly religious purposes in its original form was not either the direct equivalent of the contemporary spoken Oirat (or other Mongolian) language or the language of Oirat folklore texts and heroic epics as stated by D. A. Pavlov [1962: 113]. Most probably it recorded the way in which the Mongols and Oirats (or at least Zaya Pandita) read out loudly and pronounced formal and religious texts written in Uighur-Mongolian script. This clerkly pronunciation of Written Mongolian texts has been following the phonetic changes of the spoken language (e. g. spirantization of q) and also phonological ones to some extent (e. g. long vowels), but still greatly insisted on the written letters and morphemes, even on those that did not exist in the spoken language anymore. Insistence on the written forms lead also to such "misunderstandings" that some Written Mongolian suffixes having the same graphical representation in their front and back harmonic variants (due to the same grapheme for medial a and e) were treated as front vocalic and transferred to Written Oirat as non-harmonic suffixes or postpositions with front vowels: -ece < -aca/ece, -yer < -iyar/iyer, -ber < -bar/ber, -yen < -iyan/iyen. This is clearly an artificial invention not characteristic of any Mongolian dialect.
By representing this clerkly pronunciation, Zaya Pandita's literary language could be far enough from the different dialects of spoken Mongolian (including Oirat) to serve as a supra-dialectal literary language. So, the new script and the new orthography did not change too much the archaic style of written texts being distant from the colloquial speech, but made their reading clearer and easier. Even Written Mongolian texts could be easily transcribed to Oirat script (and some evidences prove that it was done sometimes).
Now if it is clear enough what Oirat means and what is the relation of Oirat to Zaya Pandita's original literary language, then we can answer our first question: was the Oirat script used for writing a Mongolian dialect other than Oirat? The answer is yes, it was. Specifically, in its original and first form it was not used for writing either an Oirat dialect or any variant of spoken Mongolian, but a supra-dialectal literary language.
However this situation rapidly changed. Due to various, probably mostly political reasons (Oirat-Khalkha conflicts) Zaya Pandita's invention did not reach wider (if any) popularity among Eastern Mongols and only Oirats (including Kalmyks) have started to use it. Since Clear script had a very good and precise system for indicating the sounds of the contemporary spoken Mongolian, this feature facilitated the infiltration of colloquial elements into the written language. Inasmuch as Oirats were those who used the new script, colloquial elements of their speech started to infiltrate Zaya Pandita's literary language and it became Oirat or Written Oirat — clearly distinct from Written Mongolian and typical of Oirats only. As time progressed this colloquialization became stronger and stronger, but it was not even and uniform everywhere. Religious texts and Buddhist translations kept the original bookish and clerkly character for a long time, historical texts (e. g. Sarayin gerel 'Moonlight', the biography of Zaya Pandita) have borrowed more from colloquial speech, while official and personal letters contained a lot of colloquial forms. Written Oirat became a mix of archaic, artificial and colloquial elements whose ratio depended on the period, location, context, author and some other factors. As Written Oirat is not equal with spoken Oirat, but reflects its influence, they have separate, but partly overlapping history. One can study Written Oirat's history in whole as the history of a written language, but only its colloquial elements matter as far as history of spoken Oirat is concerned.
Since Oirats living on huge territories of Asia did not have a central authority controlling the unification and standardization of their literacy (except of modern Xinjiang), just the unifying influence of some local cultural centres (monasteries, chancelleries) could exercise some effect on the variants of Written Oirat. As a result of this, Written Oirat had vivid and diverse varieties, a lot of orthographical variations — the same colloquial element could appear in many forms, but mostly with minor differences only.
4. Dating and periodization
Dating a certain written monument is an essential factor when it is used as source for study of the history of language. There are some lucky cases when the monument contains a proper dating of itself (especially private and official letters), but usually determining the
date of a monument's creation requires the examination of several aspects. Furthermore there are at least two dates that are important: date of the text's authoring and date of its recording in writing. These dates are often different, mostly because a written document is a copy of an earlier one. Actually this is true for most of the larger and longer sources written in Oirat script: monuments that survived are copies or even copies of copies. A copy (and here we talk about a copy made by rewriting or transcribing the original text, and not about copies created by some kind of printing process or photocopy) is surely not valueless, but it should be treated with extreme carefulness and source criticism (see more in "Originals and copies").
At present the earliest surviving Oirat script documents are from the end of the 17th century: letters of Galdan khan and his envoys to the Russian Tsar from 1691 [Sastina 1958; Krueger 1969; Kara 1974, 2005: 141]1. Currently this date determines the time-frame in which the history of Oirat dialects can be studied on the basis of Oirat script. Monuments from the 18th century are incomparably more numerous, and from the 19th century we have even more sources.
Periodization of Written Oirat sources is a rather puzzling question. X. Luvsanbaldan tries to divide the history of Written Oirat into two roughly determined periods from 1648 to the second half of the 17th century, and from the second half of the 17th century to 1924 in Kalmykia and up to the present in Xinjiang. He states that during the first period the original form of Zaya Pandita's literary language was in use lacking any distinct features typical of spoken Oirat (this seems to be far from the truth), and the second period is characterized by strong colloquial influence and even the disappearance of Zaya Pandita's original rules (this is also an exaggeration I think) [Luvsanbaldan 1975: 62-63].
In my opinion such periodization has not too much sense, since the genre is what mostly determines the character of a certain source,
1 I have to mention here a very recent information on a manuscript of Altan gerel 'Golden light', which was donated to the Bogd Khan Palace Museum in Ulaanbaatar by J. Coloo and — according to a newspaper — was written by Zaya Pandita's own hand. If that is true, then this manuscript becomes the earliest known source of the Oirat script and opens the possibility of examining the system of Written Oirat in its original form, as it was used by its inventor. See http://www.news.mn/content/169566.shtml.
although it is clear that later sources contain colloquial elements in increasing amount. If a certain period should be set apart in the history of Written Oirat, then I would choose the period of its modern, standardized form used today in Xinjiang (starting from the middle of the 20th century).
5. Originals and copies
Sources, which are copies of an earlier original, should be treated with great carefulness. Not only unintentional mistakes, typos or slips of the pen made during the copying process can cause problems for the researchers, but intentional or unintentional editing of the text, too. A copyist often makes changes in the original text, adjusts its orthography to his contemporary usage and standard (e. g. writes uu and uu instead of ou and ou as in xuucin and xoucin, tuuji and touji), replaces some words, phrases or proper names seeming obscure and/or archaic, or corrects some places where he found a mistake (at least by his opinion — not always correctly). This all is not necessarily a result of intentional action, sometimes it is caused by his everyday routine and habits.
If multiple copies of the same original exist they can be used for reconstructing the original text, at least to some extent. It may turn out however that this original is not the real original, but an older copy being the common source of further copies.
6. Genres and topics
As it was already mentioned above, the genre and topic of Written Oirat sources highly determines the occurrence of colloquial elements useful for historical studies of the spoken dialects. Religious texts, especially Buddhist translations made from Tibetan, are the most conservative and most artificial sources. These translations often hardly could be considered as real Mongolian texts, since Zaya Pandita (and his pupils following his method) translated Tibetan to Written Oirat almost word by word, copying even the grammatical structures, word order and syntax of the original [Yakhontova 2006]. This means that these translations can hardly add anything to the history of spoken Oirat except for some conclusions that can be drawn from the sound marking system of the Oirat script.
Authored texts form a very different group of sources having much more colloquial influence. Even Sarayin gerel 'Moonlight', the biography of Zaya Pandita (or at least its available copies) written by his pupil Radnabhadra and dated to 1691, contains some colloquial forms (e. g. zarliyasa [Radnabhadra 1999: 25] instead of zarliq-ece-yen) and lexical elements typical of Oirat (e. g. yaza [Radnabhadra 1999: 17; Radnaabadraa 2009: 87], Written Mongolian yaday-a 'outside'). Later historical sources such as the Xosuud noyun batur ubasi tumeni tuurbiqsan dorbon oyiridiyin tuuke 'History of the four Oirats compiled by the Khoshut prince Batur Ubasi Tumen' or the Xalimaq xadiyin tujiyigi xuraji biciqsen tobci 'A summary written by collecting the history of the Kalmyk khans' and several others that are not connected to Zaya Pandita and the Buddhist church contain considerably more colloquial elements (phonetic, morphological, syntactical and lexical, too).
The most useful and colloquial genre is letter. There are numerous letters available starting from the end of the 17 th century. A considerable corpus of letters of the Kalmyk khans from the 18th century has been published by D. A. Susejeva [2009] and letters from the beginning of the 19th century and sent to I. J. Schmidt during his service in Sarepta, Russia, at the mission of the Moravian Church (United Brethren) have been published by J. R. Krueger and R. G. Service [Krueger, Service (eds.) 2002]. These two collections are the most easily accessible sources, although there are some individual letters published elsewhere and various archives keep for the most part still unpublished.
There are several other genres of course (X. Luvsanbaldan makes 15 distinct categories [Luvsanbaldan 1975: 81-82]), but these three are the most important and hopeful ones.
7. Author of sources
The author of a source is an important factor during researches for several reasons. The more we know about the author, the more accurate conclusions can be drawn from the source. An author's person makes easier to date and localize a source and connect its content to a particular dialect or language variant. Comparing sources written by the same author helps to make investigations concerning the penetration of certain phenomena. An author's ethnic, social,
educational and professional background provides a lot of information about the context of a written text.
As we made distinction between dates of a text's authoring and its recording, we should make the same distinction concerning the authors, too. So, two types of authors exist: author of the text itself and a scribe writing it down. These two can be one and the same person or two (or more) different persons as in the case of copies.
Author of a text mostly determines its morphological, syntactical, lexical and stylistic character, but phonetic features and spelling may depend on the scribe. Even an accurate scribe can add the reflection of his own dialect (mostly pronunciation) to a copy and leave traces of his orthographical habits.
8. What do sources in Oirat script offer?
As it was mentioned before, studying the history of Oirat dialects requires the analysis of colloquial elements occurring in Written Oirat. These elements cover a wide range of linguistic phenomena including phonetics and phonology, morphology, syntax and lexicon.
Due to the peculiarities of Oirat script and its orthographical traditions, its usefulness for solving open questions of Oirat's history enlisted above is not equal. There are cases where little new can be expected from Written Oirat sources, and others where they are promising. In general, phonetic and phonological changes are reflected in Written Oirat partially, while morphological, syntactical and lexical changes can be traced to far more extent.
Let's provide some examples of phonetic changes. Question of suspected changes in quality of Kalmyk /0/ and /u/ ([e] > [0] and [u] > [y]) discussed and supposed by Doerfer [1965: 21-24] probably cannot be solved on the basis of Written Oirat since this change (if present) does not affect the phonological system and could not be reflected by this script. On the contrary, strong palatalization of vowels, lack of labial harmony, changing of diphthongs into long vowels are phenomena that can leave clear traces in Written Oirat. Even some phonetic changes that do not interfere with the phonological system can be traced in the spelling of glosses of foreign words or samples of non-Oirat proper names.
9. Conclusion
Written Oirat is undoubtedly an important and unavoidable source for carrying on researches on the past of Oirat dialects, as well as Mongolic languages. Although Oirat script and Written Oirat is an increasingly studied area in Mongolian studies, we are far from answering all questions concerning it. Even among its fundamental characteristics we found unsolved problems (such as the marking of long labial vowels or diphthongs by Zaya Pandita), and also its usage in Dzungaria is still lesser studied than would be desirable. Fortunately, new publications are appearing continuously (especially in Mongolia and Kalmykia) and new sources from the hidden treasures of the libraries and archives become available to the wider scientific community.
Bibliography
Balint 2009 — G. Balint of Szentkatolna. A Romanized Grammar of the East-and West-Mongolian Languages, With Popular Chrestomathies of Both Dialects [Budapest Oriental Reprints. Ser. B 3]. Budapest: Library of the Hungarian Academy of Sciences: Csoma de Körös Society, 2009. Birtalan (ed.) 2011 — A. Birtalan (ed.). Kalmyk Folklore and Folk Culture in the mid-19th Century: Philological Studies on the Basis of Gabor Balint of Szentkatolna's Kalmyk Texts [Keleti Tanulmanyok / Oriental Studies 15]. Budapest: Library of the Hungarian Academy of Sciences — Elista: Kalmyk Institute of Humanitarian Studies of the Russian Academy of Sciences, 2011. Doerfer 1965 — G. Doerfer. Ältere westeuropäische Quellen zur kalmückischen Sprachgeschichte (Witsen 1692 bis Zwick 1827) [Asiatische Forschungen 18]. Wiesbaden: Harrassowitz, 1965. Kara 1974 — G. Kara. Popravki k cteniju ojratskix gramot 1691 g. // V. M. Solncev (ed.). Issledovanija po vostocnoj filologii: K semide-sjatiletiju professora G. D. Sanzeeva. Moskva: Nauka, 1974. P. 111-118. Kara 2005 — G. Kara. Books of the Mongolian Nomads: More than Eight Centuries of Writing Mongolian [Indiana University Uralic and Altaic Series 171]. Bloomington, IN: Indiana University, Research Institute for Inner Asian Studies, 2005. Krueger 1969 — J. R. Krueger. Three Oirat-Mongolian Diplomatic Documents
of 1691 // Central Asiatic Journal 12, 4, 1957. P. 286-295. Krueger, Service (eds.) 2002 — J. R. Krueger, R. G. Service (eds.). Kalmyk Old-Script Documents of Isaac Jacob Schmidt 1800-1810: Todo Biciq Texts, Transcription, Translation from the Moravian Archives at Herrnhut [Asiatische Forschungen 143]. Wiesbaden: Harrassowitz, 2002.
Luvsanbaldan 1975 — X. Luvsanbaldan. Tod üseg, tüünij dursgaluud [Clear Script and its Monuments]. Ulaanbaatar: SUA-ijn xevlel, 1975.
Pavlov 1962 — D. A. Pavlov. K voprosu o sozdanii «Todo bicig» // Zapiski Kalmyckogo naucno-issledovatel'skogo instituta jazyka, literatury i istorii pri Sovete ministrov Kalmyckoj ASSR 2, 1962. P. 109-132.
Posch 1957 — U. Posch. The Syllabic Groups and Their Further Development in the Written Oirat Language. Pt. 1 // Central Asiatic Journal 3, 3, 1957. P. 206-219.
Radnaabadraa 2009 — Radnaabadraa. Ravjam Zaja bandidyn tuuj Sarny gerel xemeex orsiv [Moonlight: The Biography of rabjam Zaya Pandita] [Bibliotheca Oiratica 12]. Ulaanbaatar: Sojombo printing, 2009.
Radnabhadra 1999 — Radnabhadra. «Lunnyj svet»: Istorija rabjam Zaja-pandity: Faksimile rukopisi [Pamjatniki kul'tury Vostoka: Sankt-Peterburgskaja naucnaja serija 7]. Sankt-Peterburg: Peterburgskoe Vostokovedenie, 1999.
Sastina 1958—N. P. Sastina. Russko-mongol'skie posol'skie otnosenija XVII veka. Moskva: Izdatel'stvo vostocnoj literatury, 1958.
Susejeva 2009 — D. A. Susejeva. Pis'ma kalmyckix xanov XVIII veka i ix sovremennikov (1713-1771 gg.): Izbrannoe. Elista: Jangar, 2009.
Svantesson 2009 — J.-O. Svantesson. Cornelius Rahmn's Kalmuck grammar // Turkic Languages 13, 2009. P. 97-140.
Svantesson 2012 — J.-O. Svantesson. Cornelius Rahmn's Kalmuck Dictionary [Turcologica 93]. Wiesbaden: Harrassowitz, 2012.
Vladimircov 1929 — B. Ja. Vladimircov. Sravnitel'naja grammatika mongol'skogo pis'mennogo jazyka i xalxaskogo narecija: Vvedenie i fonetika [Leningradskij Vostocnyj Institut imeni A. S. Enukidze 33]. Leningrad: Leningradskij Vostocnyj Institut imeni A. S. Enukidze, 1929.
Yakhontova 2006 — N. Yakhontova. The Mongolian and Oirat translations of the Sutra of Golden Light. Paper read at the conference "Silk Road documents: the transmission of the Suvarnaprabhasottamasutra", Beijing, November 22, 2006. (http://www.nlc.gov.cn/newhxjy/wjls/wjqcsy/wjd17q/ 201011/P020101123700814793877.pdf)
