Etymological Dictionary of Tocharian B Online

A Dictionary of Tocharian B

Introduction

Purpose

In this second edition of A Dictionary of Tocharian B its purpose remains twofold. Firstly, it is a dictionary, in the ordinary sense, of Tocharian B (or, more properly, Kuchean, the language of Kucha and the Kuchean Kingdom), attested from the fourth through at least the ninth centuries of our era along the northern rim of the Tarim Basin, in the Turfan basin to the northeast, and with outlying finds on the southern rim of the Tarim basin and in Dunhuang in Gansu. Thus the dictionary’s entries have been glossed with enough detail, and provided with sufficient grammatical explanation and examples, that the user can employ it profitably in reading and interpreting actual Tocharian B texts. The dictionary attempts to treat fully the currently known Tocharian B vocabulary, including for instance all proper names and a host of rare technical terms borrowed from Buddhist (Hybrid) Sanskrit. Secondly, it is an etymological dictionary of the language and thus an aid to understanding the place of Tocharian within the Indo-European language family. Defining, translating, and etymologizing, particularly in a language as fragmentarily known as Tocharian B, are not exact sciences; readers are cautioned to take the question marks and “possibly’s,” etc., at full value and should, no doubt, add many of their own.

In the fifteen years since the publication of the first edition, those interested in Tocharian have witnessed a flourishing of Tocharian studies. Available to me now, but not in the nineties, are the full inclusion of words appearing in the Karmavācana and in Broomhead’s edition of the British Library texts. Revolutionizing the field are the re-edition of the British Library texts, published on-line by the International Dunhuang project, and the on-line publication of the Berlin collection as part of TITUS. These published pictures and texts have supplied the dictionary with a not inconsiderable group of new words and filled out the paradigms further of scores of words already in the first edition. Together with texts published more conventionally, particularly by Pinault from the Paris collection, these new sources have produced an increase of almost exactly 20% in the dictionary’s entries. An even grander revolution is under way in the form of the Comprehensive Edition of Tocharian Manuscripts (CEToM) project just begun in Vienna which aims to make digitally accessible all Tocharian manuscripts, wherever their physical location may be. The first fruits thereof have been included here.

Among an increasing list of books and articles published on Tocharian topics (semantics, grammar, etymology), three major works must be singled out: Pinault’s Chrestomathie Tokharienne: Textes et Grammaire (2008), Peyrot’s Variation and Change in Tocharian B (2008), and Malzahn’s The Tocharian Verbal System (2010). Particularly the latter two are “game-changing” for Tocharianists. Peyrot, building on the earlier, and neglected, work of Stumpf, has demonstrated conclusively that Tocharian B has an internal history. We have texts dating from the fifth, perhaps even the fourth, century AD until at least the 9th century. Not surprisingly, the language shows significant linguistic develop-ment within that period. One result for the dictionary is that the examples in the various entries have been reordered so as to put the earliest attestation first and the later ones last. An incomplete search for earlier, or later, attestations has been made so as to show a certain sense of historical development, but more needs to be done in this area before we truly have a dictionary of Tocharian B “on historical principles.” Malzahn’s encyclopedic discussion of the Tocharian verb has informed every verb entry in the dictionary and made many of them better and more complete. In almost all cases where we originally differed, I have conformed the dictionary to her work. There are a few places where our under-standing of the facts differ, however (e.g., auk-), and there are a few (some fifteen) ‘late arrivals’ (e.g., kwänt-, śäp-, wey-), so the Tocharianist still has to read both works.

Other improvements have included the complete redoing of the definitions of botanical terms, both as to the provision of an English equivalent and the provision of a Linnaean name. In many instances there is more than one English name and/or more than one competing Linnaean name. Often I have given more than one competing Linnaean name where they occur. Obsolete Linnaean names are sometimes added parenthetically, preceded by “aka.” The reader should always keep in mind that botanical classification is never fixed but always evolving.

On the etymological side, scores of etymologies have been revised or reconsidered. I have been able to take into account the wonderful new series of Indo-European etymological dictionaries coming out of Leiden (the Slavic, Hittite, Iranian, Latin, Celtic, Greek, and Armenian ones having appeared and been incorporated into this work). The Carling, Pinault, Winter dictionary of Tocharian A has been fully taken into account, insofar as it has been published.

How to Read an Entry

Entries in this dictionary may consist of the following parts: (1) the lemma itself (in bold), (2) designation of word class (in parentheses), (3) gloss, [and, in a second paragraph] (4) morphological information, (5) textual examples and text references, (6) run-on entries, (7) special notes (preceded by the symbol ●), (8) etymological notes (preceded by the symbol ■), and (9) cross-references (►). Not all entries of course contain each of these parts: the shape of the lemma is dependent on the type of word in question; in the case of uninflected words for instance there will of course be no morphological information. All citations to the London and Berlin texts (save those published in Tocharische Sprachreste) have been updated so as to give and their current Berlin press marks (THT) and London press marks (IT, short for IOL Toch).

(1) Words in Tocharian B may be divided formally into two groups: those with inflection and those without. For the latter group, composed of prepositions, postpositions, adverbs, particles, most numbers, and a small number of adjectives, the shape of the lemma is a simple matter—it is the sole form of the word. Inflected words, nouns, verbs, and most adjectives, are, of course, a bit more complicated. Nouns and adjectives are normally given in their nominative singular form (nominative singular masculine in the case of adjectives). If the nominative singular (masculine) is not actually attested, its probable form is reconstructed and given with a following asterisk. Occasionally, when only the plural (less often the dual) is attested (and particularly when there is some likelihood that the noun in question is a plurale or duale tantum), the nominative plural (dual) is taken as the appropriate lemma. Verbs are given in their root form and thus end in a hyphen. (Occasionally other words are known only fragmentarily and they will also end in a hyphen.)

(2) The designation of the class of the word is fairly simple in the case of most parts of speech. For nouns and verbs it is more complex. For nouns an indication of gender is given where it is known (nouns may be masculine, feminine, or neuter—the latter being nouns with masculine concord in the singular and feminine concord in the plural). For verbs there is an indication of transitivity, transitive or intransitive (with the understanding that a transitive verb may always be made intransitive by passivization). Verbs that are marked as both are those with an underlying intransitive “Grundverb” and a derived transitive causative (which in turn may be passivized). The semantic subset to which a particular causative belongs (according to the analysis of Malzahn [2010]) is indicated by a superscript number.

(3) As suggested above, the gloss is intended to be relatively specific (and thus more likely to provoke correction). The gloss may also include on occasion particular set phrases or idioms of which the lemma forms a part. The numerous botanical terms (almost always from some medical formula, are supplied with the parenthetical notation, “a medical ingredient” or sometimes simply “MI”).

(4) For inflected words the attested inflected forms are given between square brackets. (The universe of attested forms is substantially complete for texts published in books and articles; it is not complete for texts published on-line by THT or IDP.) In the case of nouns attested forms are given in the order: nominative singular, genitive singular, accusative singular, nominative dual, genitive dual, accusative dual, nominative plural, genitive plural, accusative plural. The numbers, singular, dual, and plural, are separated by slashes. If a form is not attested, its place is taken by a hyphen. Thus for raso (n.[m.sg.]) ‘span’ we have “[raso, -, raso/rsoñc, -, -/rsonta, -, -]” indicating that in the singular both the nominative and accusative are attested but no genitive, while in the dual and plural only the nominatives are as yet found. For pat (nnt.) ‘stūpa’ we have “[pat, ptantse, pat//-, -, ptanma]” indicating that all three forms of the singular are attested but there are no dual forms and only the accusative plural is attested for that number. Where known, the vocative form is given, in paren-theses, after the corresponding accusative, thus for waṣamo (nm.) ‘friend’ we have “[waṣamo ~ wāṣmo, waṣamontse ~ wāṣmontse, waṣamoṃ (voc. waṣama)/ /waṣamoñ ~ wāṣmoñ, wāṣmoṃts, wāṣmoṃ].” This latter entry illustrates another possible complexity, the existence of alternative forms. Where alternative forms are known, they are given separated, as here, by a swung dash. The lack of an attested alternative, as in the genitive and accusative plural, of course does not mean that such did not exist. In this, and similar cases, the second (or only) alternative is one found exclusively in poetry. Underlyingly the noun is /wāṣä́mon-/ and appears always in prose, and presumably in speech, as [waṣámon-]; when metrically necessary the stress may shift, giving [wā́ṣmon-]. The inflected forms of the adjective are shown in exactly the same way, though they will have both masculine and feminine forms and thus the potential array of forms is doubled and each part preceded by a “m.” for masculine or an “f.” for feminine as needs be.

Verbs are of course more complex yet, but follow the same basic format, though the three divisions are the first, second, and third persons rather than nominative, genitive, and accusative. Thus for the present-stem forms of pär- ‘bear (away), carry (off); take up; wear’ we have “[A -, -, paräṃ//-, parcer, pareṃ; AImpf. //-, -, priyeṃ; nt-Part. preñca; m-Part premane; Ger. pralle]” indicating that in the present active (a combination which is unique in not being explicitly marked in the dictionary) the third person singular and the second and third persons plural, in the active imperfect the third person plural is known, while of the non-finite forms belonging to the present-stem we know both the nt- and m-participles as well as the gerund. (As an adjectival form the gerund is given in the masculine singular nominative; if the masculine singular nominative is not actually attested, it is reconstructed and given with a following asterisk.) Where securely known, the underlying place of the stress is also given for each paradigm. One should also note that in the case of verbs the inflected forms and examples are divided among the attested tenses and moods.

(5) The textual examples have been chosen to illustrate the various meanings of the word and to give some idea of the collocations into which it may enter. Unsystematically, examples have also been chosen to highlight a word’s presence in Archaic ([4th and] 5th centuries AD) or Early Tocharian B (6th century). To emphasize the time depth of the attestation of Tocharian B, all exemplifying texts have been “dated” where possible by the addition of Peyrot’s designation of linguistic periods. (Classical Tocharian B texts are attested from the beginning of the 7th century onward. To include, with greater regularity, attestations in Late or Colloquial Tocharian B [7th-9th centuries] is a desideratum of the next edition.) In poetic texts some attempt has been made to include verse punctuation (: and •) and verse numbering, which may be read as “here ends śloka/stanza such-and-such”), since those punctuation marks so often delimit syntactic units of some sort. The translations are designed to maximize the possibilities of one-to-one correspondence with the Tocharian B original, even at the sacrifice of elegance or even, at times, ordinary English word order. Words inserted for the sense in the English but which have no Tocharian B correspondent are placed in square brackets (except for articles and other “little words” which are added without comment). Particular care is taken to cite all the attested Buddhist (Hybrid) Sanskrit (B[H]S) equivalents of Tocharian B words, since so much of our knowledge of Tocharian B vocabulary comes from the fact that a majority of our Tocharian B texts are translations of B(H)S originals. (B[H]S is an admittedly unsatisfactory cover-term which takes in a wide variety of Sanskrit and somewhat Sanskritized Middle Indic used by Buddhists but no simple equivalent can be substituted. In most cases we have no idea the exact route by which a Sanskrit, Sanskritized Prakrit, Prakrit, or Pāli word has become a part of the Tocharian B vocabulary.) In the case of proper names and obvious loanwords there are usually no examples given as the meaning in the case of proper names is self-defining while in the case of the many Buddhist technical terms borrowed from B(H)S the meaning may be discovered to a much greater degree of detail by consulting dictionaries of (Buddhist Hybrid) Sanskrit than it could be inferred from the limited examples known to us in Tocharian B texts. However, even in the latter cases at least one reference to the word’s occurrence in a Tocharian B text is given.

(6) Given as run-on entries, or sub-lemmas, are regularly derived adjectives, abstract nouns, and those compounds which contain the main lemma as the first member. The run-on entries are preceded by a long hyphen (if the run-on entry is a derivative which is only attested as the second member of a compound, it is preceded by a long hyphen, a space, and then a single hyphen). Thus, appended to the main entry whose lemma is śaul (nnt.) ‘life,’ we find —śaulaṣṣe ‘prtng to life,’ —śaulanmaṣṣe ‘prtng to lives,’ —śaulassu ‘life-possessing,’ —śaula-preñca ‘life-bearing,’ —śaulu-wärñai ‘life-long,’ and —śaultsa warñai ‘id.’ Derived adjectives other than the productive ones in -ṣṣe, -ññe, -tste, -ssu, are normally given separate entries; thus ypiye (adj.) ‘prtng to barley’ is separate from yap (n.[m.sg.]) ‘barley.’

(7) Special notes (marked by a ●) are relatively rare. They are intended to draw the reader’s attention to difficulties of morphological or semantic analysis or may draw his or her attention to places where the analysis presented in this work differs from that of its predecessors or contemporaries.

(8) The etymological notes (preceded by ■) are intended to sketch the history of the etymological discussion concerning the lemma and to present what I think to be its most probable history. The history of the discussion is usually admirably taken up through 1976 by A. J. Van Windekens’ Le Tokharien confronté avec les autres langues indo-européennes, Vol. I: La phonétique et le vocabulaire. I have not found it necessary to repeat everything in this work, especially early speculations that are dead-ends in both Van Windekens’ opinion and mine. Nor have I given in full Van Windekens’ own conclusions when subsequent research has clearly made them impossible (such cases are noted by “otherwise VW”). I have contented myself with summarizing the major lines of thought through 1976 and giving relatively full coverage of the discussion since 1976. It is my hope that I have not missed anything of note in that time period. However, given the episodic nature of this work’s gestation and the relative isolation of my work environment makes it more or less certain that something that should be included has been missed. The absence of a citation in the etymological discussion should be taken as a lamentable omission, not an implicit rejection or denigration of the work overlooked. If the word is of inherited Indo-European origins, a sample (in square brackets) of Indo-European cognates is given with a reference to general Proto-Indo-European etymological works, Pokorny (1959) and, where applicable, Mallory and Adams (1997), but the cognates given are not to be considered the totality of related forms.

(9) Cross-references (either italicized and followed by “q.v.” in the text or in bold at the end of the entry and marked with an arrow (►) are intended to guide the reader to words certainly or probably related to the entry in question.

Alphabetic Order

The alphabetic use in this work is that customary in Tocharian studies and that used by Tocharian writers themselves (cf. Couvreur, 1965): a, ā, ä, i/ī, u/ū, r̥, e, ai, o, au, k, kh, g, gh, ṅ, c, ch, j, jh, ñ, ṭ, ṭh, ḍ, ḍh, ṇ, t, th, d, dh, n, p, ph, b, bh, m, y, r, l, ly, v, w, ś, ṣ, s, h, ts. The symbol ṃ is given its own alphabetical order when it precedes a sibilant (ś, ṣ, s) or h, but otherwise is taken as an allograph of n and alphabetized accordingly. (If, as rarely, it precedes a k or p, it is taken as an allograph of ṅ or m respectively.) The extremely rare ḷ is alphabetized as plain l.

Acknowledgments

I acknowledge with deep gratitude the financial and moral support, including the award of two sabbaticals that the University of Idaho and its Department of English have provided for this much longer-term project than anyone ever foresaw. For technical support and assistance at various stages in this project I am very much grateful to William Pyle, Wolfgang Schubert, Glenna Tibbetts, Kim Sarff, Arla Marousek, Cathy Myers, Keri Moore, Anna Thompson, David A. McArtor, James M. Reece, Alexander Lubotsky, Marijn van Putten, and particularly H. Allen Adams, Michael D. Adams, and Gordon Thomas. For review and comment, particularly on the etymologies in the first edition, I am indebted to Eric P. Hamp, Jörundur Hilmarsson, H. Craig Melchert, and John W. H. Penney. On-going conversations with Melanie Malzahn have informed, and improved, many entries in the second edition, as has an extensive written commentary compiled by Werner Winter before his death. Had I availed myself more often of their suggestions, no doubt the book would have been a better one. In any case, the faults of the work must be laid squarely at the feet of the author.

It remains to acknowledge with deep gratitude the patience and support of my wife, for whom the dictionary has been a constant almost the entirety of our married life, expanding repeatedly from office to dining room table, to living room and then back again, and of my sons who have known no life without the dictionary in the background and who themselves have contributed greatly to the solution of its formatting problems. And, finally, I owe a profound debt to my grandfather, Howard A. Adams, whose enthusiasm for language, and particularly the classical languages, turned out to be quite infectious and set off the train of events which has resulted in this work, and to Eric P. Hamp, whose own enthusiasm for, and apparently limitless knowledge of, all things Indo-European, has ultimately informed this undertaking at all levels.

All Saints’, 2012

A Dictionary of Tocharian B

ABBREVIATIONS and SYMBOLS

A Dictionary of Tocharian B

ENGLISH – TOCHARIAN B REVERSE INDEX