ᠮᠤᠩᠭᠤᠯ ᠬᠡᠯᠡ ᠬᠢᠭᠡᠳ ᠲᠦᠷᠦᠭ ᠬᠡᠯᠡᠨ ᠤ ᠰᠤᠳᠤᠯᠭ᠎ᠠ （ᠬᠡᠯᠡ ᠰᠢᠨᠵᠢᠯᠡᠯ ᠤᠨ ᠪᠤᠲᠢ）

STUDIES IN TURKIC AND MONGOLIC LINGUISTICS This book, now back in print after having been unavailable for many years, is one of the most important contributions to Turkic and Mongolic linguistics, and to the contentious ‘Altaic theory’. Proponents of the theory hold that Turkic is part of the Altaic family, and that Turkic accordingly exists in parallel with Mongolic and Tungusic-Manchu. Whatever the truth of this theory, Sir Gerard Clauson’s erudite and vigorously expressed views, based as they were on a remarkable knowledge of the lexical of the Altaic languages and his outstanding work in the field of Turkish lexicography, continues to command respect and deserve attention. Royal Asiatic Society Books The Royal Asiatic Society was founded in 1823 ‘for the investigation of subjects connected with, and for the encouragement of science, literature and the arts in relation to, Asia’. Informed by these goals, the policy of the Society’s Editorial Board is to make available in appropriate formats the results of original research in the humanities and social sciences having to do with Asia, defined in the broadest geographical and cultural sense and up to the present day. The Man in the Panther’s Skin Shota Rustaveli Translated from the Georgian by M.S.Wardrop Women, Religion and Culture in Iran Edited by Sarah Ansari and Vanessa Martin Society, Politics and Economics in Mazandaran, Iran 1848–1914 Mohammad Ali Kazembeyki The Zen Arts Rupert Cox Studies in Turkic and Mongolic Linguistics Gerard Clauson New Introduction by C.Edmund Bosworth The History of the Mohammedan Dynasties in Spain Ahmed ibn Mohammed al-Makkari Translated from the Arabic by Pascual de Gayangos New Introduction by Michael Brett The Courts of Pre-Colonial South India Jennifer Howes Persian Literature: A Bio-bibliographical Survey Volume V: Poetry of the Pre-Mongol Period François de Blois Studies in Turkic and Mongolic Linguistics Sir Gerard Clauson LONDON AND NEW YORK First published 1962 by The Royal Asiatic Society of Great Britain and Ireland Second edition first published 2002 by RoutledgeCurzon 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by RoutledgeCurzon 29 West 35th Street, New York, NY 10001 RoutledgeCurzon is an imprint of the Taylor and though men be so strong that they come to fourscore years; yet is their strength then but labour and sorrow, so soon passeth it away and we are gone. Psalm 90, verse 10 (Prayer Book version). There is a gracious practice in the learned world, whereby the friends and pupils of a distinguished scholar celebrate his seventieth birthday by clubbing together and producing a volume of articles dedicated to him and called alternatively Charisteria, Symbolae, Festschrift, Arman or, in Eng lish, more prosaically Anniversary Volume. It would be uncharitable, and indeed unjust, to suggest that one of the motives at the back of the contributors’ minds is a vague hope that the old gentleman will take this as a discreet hint that he is past work and ought to retire and make way for others. Nevertheless the fact remains that a seventieth birthday is something in the nature of a milestone indicating that from then onwards the road can only go downhill. Not being myself an established denizen in the learned world—at the best I might be described, in the language of the National Insurance Acts, as a late entrant, at the worst as a mere gate- crasher—it would be wildly inappropriate for me to be given the compliment of an Anniversary Volume; and so, being immune from professional competition, I felt that I might suitably reverse the process and celebrate my seventieth birthday, a little in arrear, by presenting to my many friends—I never had any pupils—a report on some of the conclusions which I have reached as a result of nearly ten years’ intensive study of the early history of the Turkish language. There is less vanity and more good sense in this than might at first sight appear. We are all too familiar with the phenomenon of scholars who absorb knowledge like a sponge all their lives and in the end take it with them to the grave, leaving little or nothing to show for it. My purpose in presenting what is, I hope, only an interim report is to put together and present in logical order some of the facts about the Turkish and Mongolian languages which I have learnt during these past ten years, and to indicate certain directions in which I suggest that further research can profitably be directed either by myself or by others. Since writing the first draft of this book I have given some thought, but perhaps not enough, to my possible audience, and endeavoured by judicious revision to make it serve three rather different purposes. I hope that in doing so I have not fallen between three stools. There is a real need for an introduction to Turkish studies, suitable for elementary students of any of the Turkish languages, which will give them some idea of the evolution of the Turkish language group as a whole, help them to see what position the language they are proposing to study, perhaps Republican Turkish or one of the Turkish languages spoken in the Soviet Union, occupies in that group, and encourage them to broaden and deepen their interest in the group as a whole. One of my objects has been to meet this need to a modest degree. There is also a real need to re-examine the transcriptions of the early Turkish texts, down to, say, the eleventh century inclusive, to correct them in certain respects, and so to establish the phonetic structure of the languages in which they were written, and from these to deduce the phonetic structure of the earliest form of Turkish which we can visualize. This need too I have endeavoured to meet, but the subject is a difficult and technical one and I am afraid that the elementary student, if he gets that far in the book, will find it very heavy going. Finally it would be nice to bring the question of the “Altaic theory,” which has now bedevilled Turkish studies for many years, to a head, and so, if possible, to settle once and for all whether the Turkish and Mongolian languages are genetically related to one another; in Chapter XI I have stated the reasons for which I believe that they are not. This and the two preceding chapters are designed for yet another audience, perhaps not so much the Turcologists as the Mongolists, who will find little of interest for their own discipline in the rest of the book. It may perhaps be of interest to explain how I came to write a book combining these three rather disparate themes. I first acquired an affection for the Turkish language at the age of fifteen, but the pressure of official and other duties prevented me from devoting much time to its study for the next forty-five years. However the old allegiance never wavered, and when I retired from public service at the end of 1951, I resolved to devote my declining years to the history of the Turkish language. In fact, since that date the whole of my time that has not been occupied by domestic, social or business matters has been devoted to that study. The result of this work has been a trickle of occasional papers which are listed at the end of this Foreword for convenience of reference and a larger volume of work which is not yet in a form suitable for publication. The advantages of pursuing a line of study without a salary and consequently without a master are incalculably great. Undistracted by the demands of pupils and the exigencies of a programme laid down from above, I have been able to follow my studies in whatever direction they led me, starting and discontinuing particular lines of research as circumstances seemed to demand. When I retired at the end of 1951 my first resolve was to compile a new and better “Radloff,” a historical dictionary of all the Turkish languages from the earliest times to the present day, excluding loan words from other languages. This project petered out in the sands at the end of about six years. It became quite clear that it was impracticable for me, and indeed probably impracticable for any single scholar, even if he started in his youth and not, as I did, at the age of sixty, to compile such a work single handed, and I began increasingly to doubt whether it would be of real utility even if it were compiled. My reasons are stated in Chapter II of this book. Even as I wrote, new authorities came pouring in, so that each article became more and more incomplete directly it was written. By that time I had finished, as it had already become clear quite imperfectly, the work on words beginning with vowels, and was nearly half way through the work on words beginning with ç- and c-, a group which I had chosen because I thought, in my innocence, that it might be a fairly easy one after the matted jungle of words beginning with vowels. And so I started all over again on a much more modest scale to compile a historical dictionary of the early Turkish languages including, with a few exceptions, only words which are known to have existed before the Mongolian expansion in the early thirteenth century. This work is still proceeding and I reckon that over half the first draft is now done. This does not mean that I regard the labours of those first six years as wasted; quite the contrary. It was an invaluable experience to take word after word, to establish its first appearance and meaning, to trace its history forward from that point and find out what later forms it assumed and what finally became of it. This is in fact the only way in which to get a real feel for the history of the language. It taught me at least four important lessons. The first in chronological order was that Turkish dictionaries contain many supposedly Turkish words which never really existed at all. My conclusions on this subject were embodied in Turkish ghost words. This paper contains some minor inaccuracies, but I see nothing of substance to alter in it, although, as I shall mention below, I have added some elaborations in another place. A by-product of this discovery was the realization that there was a large concentration of these “ghost words” in the dictionaries of Çatay, owing chiefly to the fact that scholars who worked in this field had never had the advantage of basing their studies on the only reliable dictionary of Çatay compiled by an eastern scholar, the Sanglax of Mirza X. My early studies of this manuscript were pursued with the help of manuscript Or. 2,892 in the British Museum, which is a sort of calligraphic nightmare, but I was so fortunate as to find an incomparably better manuscript in the possession of the E.J.W.Gibb Memorial Trustees, and even more fortunate in being able to persuade them to publish a facsimile of it with an introduction and indices by myself. This meant taking a good deal of time off from my regular lexicographic work, but I do not regret it. The publication of this book has, I hope, laid a more solid foundation for Çatay studies in future. I hope too that the further elaboration of my remarks on the genesis of ghost words and the explanation of the way in which the compilers of Çatay dictionaries worked, which I included in my introduction, will stimulate other scholars in the pursuit and elimination of more ghost words. The second lesson which I learnt, more particularly from the Sanglax, was that there is a much greater proportion of Mongolian loan words in most mediaeval and modern Turkish languages, but more. particularly those in the north-eastern group (the languages of southern Siberia, including Tuva), the north central group (Kazax and Kırz) and in Çatay than had hitherto been suspected. It ha s of course long been realized that all the languages spoken by Moslem Turks are full of Arabic loan words, and some of them, more particularly in their literary forms, full of Persian loan words as well, but the existence of this massive volume of Mongolian intruders seems somehow to have escaped notice, or, if noticed, to have been regarded by those who accept the Altaic theory as evidence of a common “Altaic” heritage in both language groups. So far as Çatay is concerned, I devoted a good deal of attention to the subject in the introduction and relevant indices to the Sanglax. The third lesson which I learnt was that the phonetic structure of the earliest kind of Turkish that we can reconstruct, that spoken appreciably earlier than the eighth century, the date of the earliest substantial remains of Turkish which have come down to us, was a good deal more elaborate than had hitherto been realized, both as regards the general repertoire of sounds, and also more particularly as regards the sounds which could be used at the beginning of a word. This first came actively to my attention when I was working on words beginning with ç- and c-, and I embodied some preliminary observations on the subject in The Turkish Y and related sounds. I returned to this subject recently in The initial labial sounds in the Turkish languages. Another by-product of my lexicographic labour at this period was The Turkish numerals, which inter alia established some unexpected facts about the phonetic structure of some numerals, and showed that one class of collective numerals was a Turco-Mongolian hybrid and not pure Turkish. The central point of the Studies which follow is Chapter VIII, in which I have embodied what I have so far learnt about the phonetic structure of pre-eighth century Turkish. It is not intended to replace any existing work; for the most part it deals with an earlier stage of the language than any hitherto studied, but it is intended to provide a solid basis for the re-examination, and where necessary revision, of the conclusions set out in such classical works on the phonetic structure of the modern Turkish languages as:— W.Radloff, Phonetik der nördlichen Türksprachen, Leipzig, 1882; N.F.Katanov, Opyt izsledovaniya Uryankhayskago Yaz yka, Kazan, 1903; M.Räsänen, Materialen zur Lautgeschichte der Türkischen Sprachen, Helsinki, 1949; L.Bazin, Structure et tendences communes des langues turques (Sprachbau) in Philologiae Turcicae Fundamenta, Wiesbaden, 1959; N.A.Baskakov, Tyurkskie Yaz yki, Moscow, 1960. The fourth lesson which I learnt was that the early Turkish vocabulary had an extraordinarily tenacious hold on life. There are a good many words which occur in the earliest languages, especially Uyr and , but do not seem to be noted as occurring in any later literary text or in any dictionary of the literary languages, and yet have survived in the spoken languages and so can still be found in V.V.Radloff Opyt slovarya tyurkskikh narechiy, St. Petersburg, 1888–1911 (cited as Opyt), as existing in languages of the north-eastern group, or in the Türkiyede halk aından söz derleme dergisi, Istanbul, 1939 ff. (cited as SDD) as existing in the contemporary spoken language of the Turkish Republic. It is very probable that further search in other mod