Ogur and Oguz
Alans and Ases
Russian Version needs a translation
|“Eastern Iranian” linguistic group|
THE KHWAREZMIAN LANGUAGE
Zeki Velidi Togan'a armagan, Istanbul, 1955, pp. 421-436
Reprinted in “Acta Iranica”, 1b15; Deuxieme Serie. Hommages et Opera Minora, Vols. V-VI, 2 vols, pp. -
This posting is a part of citations from the works of W.B. Henning, an outstanding philologist of Iranian languages, with bearings on the history of the Türkic peoples. It provides a context for his linguistic diagnosis, not challenged since it was first formulated, on the affinity of the Horezmian language that was heavily misrepresented by the advocacy of the Scytho-Iranian theory. The theory came up with the chimeric “Northeastern Middle Iranian language”, a category that included Ossetian as a descendent language, along with Pashto and Yaghnobi, and holds Khorezmian and Sogdian as “closely affiliated”. The category is also called East Iranian group of languages. However, Yaghnobi is also rated as a direct descendent of the Sogdian, which would exclude it from being a direct descendent of the “(Pra-)Ossetian”. The Sogdian is defined not an East Iranian language, but a “Sprachbund” Creole in a symbiosis of the pre-Indo-Iranian flexive and possibly Türkic agglutinative languages “... it does not seem possible to regard the Eastern Iranian group as a whole, even excluding Parachi and Ormuri, as a genetic grouping. Such a concept would imply the existence of an ancestral “Proto-Eastern Iranian” intermediate between the “Common Iranian” and the attested Eastern Iranian languages; but if one reconstructs the “Proto-Eastern Iranian” in such a way as to account for all the features of the group, it proves to be identical to the “Common Iranian” reconstructible as the ancestor of the whole Iranian family. It is therefore more plausible to conceive of Eastern Iranian as a “Sprachbund” or areal grouping of languages. In this case the members of the “Sprachbund” happen to be genetically related, but the special features which mark them out as a group result rather from centuries of contiguity, during which innovations will have spread from one language of the group to another and neighboring languages will have supported each other in the retention of shared features.” ( A.Dybo “Lingivistic Contacts Of Early Türks”)
In other words, the very concept of the Eastern Iranian group is as much a fiction as any Creole conjured to be a daughter of a mother tongue. This “Sprachbund” Creole East Iranian group includes documented languages Pashto, Pamir, Ormuri-Parachi, Khwarezmian, Sogdian, Khotanese, and Yaghnobi, and a number of non-documented speculative attributions for Saka, “Scytho-Sarmatian”, and Bactrian languages. The “Scytho-Sarmatian” language is a phantom category peculiar to the Scytho-Iranian theory, concocted to crate the genetic chain linking the “Ossetian blend” with the Scythian: Ossette - Alanian - Sarmatian - Scythian. And the “Ossetian” is a political chimera that cobbled together the languages and ethnoses of Nakh-based Irons (Persians), Digors (Dügers, Tokhars), and Jasses (Ases), with Iron being a Nakh/Persian amalgamation, and Digors/Jasses being known members of the ancient and well-attested As–Tokhar confederation genetically distinct and unrelated to the Irons. With that background, the observation of W.B. Henning makes perfect sense: “As regards the position of Khwarezmian within the circle of the Iranian languages, the closeness of its relation to Sogdian on the one hand and Ossetic on the other has perhaps been overstressed; some of the most striking resemblances to Sogdian may be due to loans. Some of the sound changes that so far have been neglected may be mentioned here, especially those that show the connection of Khwarezmian with the languages adjoining it towards the south and south-east” (a transparent reference to Pashto et al.) The rest of the posting is only an illustration depicting material leading to this profound linguistic conclusion.
The sad consequence of the non-sensical Indo-European-centered quasi-scientific brouhaha is the loss of the historical identity by the people with fascinating and glorious history, which so far has been lost in the Indo-European-spoofed nationalistic borsch; the famous Tokhars/Yuezhi melted away like a mountain stream in the desert sands, and their descendents in the name of the Indo-European pedigree are re-molded into faceless and illiterate “Eastern Iranians”. The Ephthalites lost their original Abdaly identity, their original true Tokharian/Yuezhi language, their glorious tamgas that once were known across Eurasia, and became obscure Turkmen Abdals and Yazyrs (Ases), and obscure Kazakh and Bashkir and Uzbek and Azeri and Turk Abdals, the obscure Eastern Turkestan (Uiguristan) Abdallar Abdals, and obscure Volga Tatar (Itil Bulgar) Abdals; in 1747 the Bactrian/Tocharistan Abdaly were renamed to Durrani in the Afganistan, and became the core of the Afgani Pashtuns that speak the “East Iranian” Creole “Sprachbund”; no wonder that W.B. Henning found the Horezmian language closest to the Turco–Bactrian–Persian–Indian Pashto; the Persian Abdaly united with the Mazanderan Lurs and were re-christened into Khojavends; the Indian Abdaly became Gujars and Rajputs, of all other Ephthalites they alone to some degree preserved their historical connection with the Abdaly and Huns. In the quasi-scientific melee became obscure the original homeland of the Tokharian tribes, the ancient and splendid Horezm, and the linguistic distinction that embraces all Tokharian slivers: morphologically agglutinative language, SOV structure, and obviously Türkic substrate lexicon, which either make the “Eastern Iranian” group conspicuously non-Indo-European, or brings into the Iranian subdivision a wealth of conspicuously non-Indo-European traits. Ethnologically, the Pashtuns, like the other descendents of the Abdaly Huns, did not lose their hallmark military spirit, they have already defeated one superpower, and are well underway in defeating the only other remaining superpower.
Page numbers are shown at the beginning of the page NNN (Istanbul, 1955) and at the end of the page [NNN] (Reprint). Detailed linguistic analysis has not been proofread, and is posted in smaller font. Page breaks in continuous text are indicated by //. Chinese hieroglyphic spelling and Arabic script are skipped, with  indicating placeholders. Most of diacritics are dropped to avoid font conflicts. The bold highlighting was added for the posting.
Many of those who come to-day to offer their congratulations to Professor Zeki Velidi Togan may be unaware of the great services our learned friend has done to the study of the Iranian languages. It is owing solely to his restless energy and searching mind that we now possess a fairly adequate knowledge of the ancient Khwarezmian language, the mother tongue of al-Beruni and al-Zamakhshari, an Eastern Iranian dialect that continued to be spoken in Khwarezm well into Mongol times.
His first discoveries of Khwarezmian material were embodied in an article (Islamica iii, 1927, 190-213) that contained extracts from a Fiqh work studded with Khwarezmian sentences, the Yatimatu 'd-dahr fi fatawi ahli 'l-'asr, and, in addition, a few glosses from the Khwarezmian version of Zamakhshari's Muqaddimatu 'l-adab. They were overshadowed by his later find of the Qunyatu 'l-munyah, another Fiqh work written in Khwarezm, of which countless manuscript copies are extant, some of them of the first quality. As the Khwarezmian words of the Qunyah, in the best manuscripts, are not only properly pointed and vocalized but also fully explained, in Arabic or Persian, they offered a secure basis for the study of this hitherto forgotten Iranian language.Professor Zeki Velidi Togan has now crowned his earlier efforts with the brilliant discovery of a considerable portion of the Muqaddimatu 'l-adab in its Khwarezmian version. The manuscript he has discovered extends to some 190 pages: it contains about a third part of the original work (corresponding to pp. 45-132 in Wetzstein's edition). The obvious fact that this manuscript is the original copy written by the author, the great Zamakhshari, himself, nearly consoles us for the lack of any vocalization; but the scarcity of consonant points, which to the author, writing in his mother tongue, may have seemed an unnecessary adornment, creates endless difficulties for us 1.
1 [When this was written, in November 1950, the even greater extent of Professor Togan's discovery could not be fully perceived. The Ms., now happily published in facsimile (Khorezmian
Glossary of the Muqaddimal al-Adab, edited by Z. V. Togan, Istanbul 1951) contains in
fact as much as 511 pages with Khwarezmian glosses. Only the first 190 pages could be
used for this article]
In correct appreciation of the immense value of his discovery, which, providing a vast vocabulary, supplements the information derivable from the Qunyah in the happiest way, Professor Zeki Velidi Togan has decided to publish the manuscript in a full-length facsimile edition. Of his edition, which may be generally available by the time this paper is published, he kindly sent me advance proof-sheets. As a token of our appreciation of his great services to Iranian Studies, what could be more appropriate to the occasion than making some use of the rich material he has provided?
When Professor Togan had found the Qunyah, he generously invited me to participate in the exploration of the documents. After some weeks of intensive work the basic structure of the language became clear. At the 8th German Orientalistentag, held in Bonn in August, 1936, both Professor Togan and myself delivered lectures on the Qunyah and the language data contained in it; our lectures, in extract, were printed still in 1936 (ZDMG., vol. 90, pp. + 27 4 + — + 34 + ; here quoted as Report). We then announced that we were engaged in preparing a full edition of the material.Three years later, before we could complete the work, the “discovery” of the very same Qunyah was announced by S.L.Volin and A.A.Freyman in two articles (Zapiski Instituta Vostokovedeniya Akad. Nauk, vii, 1939, 79-91, 306-319), of which the former repeats Professor Togan's lecture and the latter reproduces what I had said. These two articles are unhampered by consideration of our clear priority; in Freyman's article the only acknowledgement made is in a sentence of classical brevity at the end: “the study of the Khwarezmian language proceeds also abroad” (p. 319, with reference to ZDMG., 90, but without mention of our names). One gains the impression that in the opinion of these two authors the task of analyzing the Khwarezmian language was so easy that its accomplishment deserved no credit; it is true that it is easy after it has been accomplished by someone else (A major role in the treatment of the material played the atmosphere of terror in the Soviet science, where the Academy members were systematically murdered and replaced by decoy-duck “scientists” voted in by the intimidated remnants of the Academy, who later proceeded to re-write the history and filter out noncompliant facts. Giving credit to bourgeois unscientific scientist, traitor, and Panturkist Z.V.Togan [he fled to Turkey, and accordingly was automatic Panturkist] and bourgeois unscientific scientist W.B.Henning could have been a self-confession in spying in favor of Iran, Germany, etc., and a speedy dispatch to the other world. Probably, the publication in the form as it was had already been a daring deed taken with sober assessment of the risks involved. S.L.Volin [1909-1942?] was “repressed” three times, and died in a Soviet concentration camp at an age of 33. A.A.Freyman [1879-1968] barely survived the pogrom, he must have done something that pulled him from “undesirables” into a kind of “ours”. By 1948, W.B.Henning should not have any illusions about the Soviet scientific system, and harbor petty resentment against people who daily walked on hot coals).
As A. A. Freyman had declared his wish to proceed to the publication of the Khwarezmian sentences of the Qunyah, I for my part abandoned my earlier intention to undertake this task. The interference we had experienced produced in my mind disinclination to engage in work now made distasteful; and there is a sufficiency of untouched material in our field to render double work doubly undesirable.
Ever since 1939 A. A. Freyman has published a series of articles on the various features of the Khwarezmian language, without, however, fulfilling his promise of making the text of the Qunyah accessible in a proper edition. Only in his latest paper, contained in Sovetskoe Vostokovedenie, vi, 1949, pp. 63-88, has he made a beginning with the systematic publication of the Khwarezmian sentences, although still withholding the Arabic context and, for the greater part, the Arabic and Persian glosses explaining the Khwarezmian words l.
After studying Freyman's latest article (here quoted as SV. vi) one realizes, not without surprise, that there has been no advance over the inevitably incomplete and in many points faulty statements in my part of the Report. It is clear that no proper analysis has ever been attempted. In consequence, not even the principal points of either phonology or morphology have been grasped (apart, of course, from those that had already been dealt with in the Report). The explanations given by Freyman unsystematically, sentence by sentence, are marked by poverty of invention, lack of coherence, and irritating repetitiveness. There is an abundance of blunders; a few examples may be given here.
In the sentences concerning divorce the word pandmi “my co-wife” occurs more than once (SV., vi, 86). Freyman derives it from patinama. Such a word might mean “surname”, but scarcely “co-wife”; and even the most casual glance at the Khwarezmian language assures one that it would result in picndm. In fact, the mi of pandmi is pronominal affix “my”; the noun is pana, or, more properly, apana (spelt '/w in the Muqaddimalu l-adab [here referred to as Muq.]), presumably stressed apdna\ the stress moved forward on to the final syllable when a suffix was added, and the stressed vowel was lengthened, hence pandmi from apand-mi.
1 [Meanwhile the publication has been continued in Freyman's Xorezmiyskiy Yazik, 1951, which, in addition to reprints of earlier articles, including that in S.V., (here on pp.
51-78), contains further glosses (pp. 78-118) published in the same style. This work was
not yet available when the present article was written.]
Already in my part of the Report I referred to the Beweglichkeit der Vokale unter Akzentwirkung (p. 34), which is indeed as characteristic of Khwarezmian as it is e. g. of Hebrew and Aramaic; it frequently involves the loss of an initial vowel, when the stress moves forward, cf. e. g. *w6yr = u6ir- “belly” (Avestan udara-) beside Srfind - Sdrfdnd “girth” or Srfineyk rwyr = ddrfidnjik ruwira “astringent medicine” (Av. urvara). While lack of understanding such finer points as the movement of stress and its effect on the word structure may well be excused, one would have thought that the ability of detaching the more obvious suffixes from the whole word should form one of the prerequisites of the study of unknown languages. In the present case matters are made worse by the consideration that the true etymon (Avestan hapadni“co-wife”), which in any case should be obvious to anyone acquainted with the history of the Iranian languages, was expressly stated as long as ten years ago (see my Sogdica, 1940, p. 17).
The first of the Probesatze I gave in the Report, p. 34, is reproduced by Freyman, SV., vi, p. 66, but the first words, haftamidi yd duydami “I have given you my daughter”, reappear wrongly as haf$(i)rindi yd duydami. In fact, all MSS agree in spelling Kftrnyd (not h'firynd), with nun before yd. To give the text as h'firynd involves a silent correction of an altogether illicit and misleading kind; no doubt it has been caused by inability to account for the form; for as "I gave" + hi "him" is h'flrrih, "I gave" + di should appear as h'firn'd. However, the actual form, h'firnyd, illustrates one of the principal rules of Khwarezmian syntax, a rule not understood by me in 1936, after the first brief glance at the Khwarezmian material, and not understood by Freyman in 1949, after many years of study. The rule is this : if a verb precedes its object, the object is expressed a second time by an anticipatory pronoun, which is added either to the verb (if the first word of the sentence) or to a word before the verb1. It follows that in the sentence under review the verb, h'firnyd, as it precedes its object, must contain a reference to the object, in addition to the pronoun - di "to you". Indeed, it is owing to the presence of a second pronoun that the actual form differs from the (wrongly) expected form h"ftrrid.
1 The use of anticipatory pronouns in such positions is not confined to the direct object. It also occurs with certain prepositional expressions. Thus a noun preceded by /-
"in, by" is normally anticipated by the pronominal adverb wa "in him, therein, thereat"
(not understood by Freyman, SV., vi, 76, 78), which, when affixed to a verb, appears as
wa (e.g. added to the ending of the 3rd pers. sing impf., -da, it appears as dawa, spelt
-d"w). Example: c'yt'w fy n'n'm = caytdwa ft nanama ~ Arabic daxala 'alafulanin (Muq.),
where "he entered" = cdyta.
The hidden pronoun, of course, is -hi "her", and - nyd = nidi represents -nahidi. Such contraction commonly occurs, owing to the feebleness of h in Khwarezmian, whenever the pronoun - hi is followed by another element. For example, the ending of the 3rd sing. impf. is - da, which combined with -hi becomes -ddhi; but when hI6 "with him, to him" or hie "from nim" are joined with it, we find - did and - die.
Another error in which Freyman has followed me concerns the prepositions. In dealing with the words fy cwb "in the water" (Report p. 34) I wrongly explained ji as a preposition "in". In fact, two words have coalesced in Ji: the preposition f or / "in to, at, on, etc." (presumably derived from Old Iranian obi) and the masculine (and plur. comm.)
article i. Hence, Ji = "in the". Correspondingly, in combination with the feminine
article, yd, we have fa. Freyman, of course, has not understood why this preposition
should seemingly appear sometimes as/?, sometimes as fa (see SV., vi, 76, 84). The two
forms of the article are equally combined with the other prepositions; thus we have ci
and cd from c" "with, from" (Av. haca), pad and pacd from pac "after" (Av. pasca), pan
and para form par "over" (Av. upairi], and so on. This has not been clear to Freyman (see
e.g. p. 86 on para). The fact that the feminine article is a mere -a when attached to
prepositions may, at first glance, make one doubtful of the derivation of the article
from the old relative stem ya-, which was proposed in my portion of the Report (p. 34)
and has since been advocated by Professor Bailey (Asica, 15 sqq.; Transactions of the
Philological Society, 1945). Yet the volatility of initial v- in Khwarezmian entitles us
to the assumption that the difference between yd and a was so insignificant that the
former could be replaced by the latter. Indeed, one finds 'wyc'nkyk (unpointed) =
awecandkik "eternal" resulting form ydwaitdn-) or 'yw'nk "young" (from yuvan-) beside
'_>-' yw'nyk "the young". An allusion to the reverse phenomenon, the addition of y- to
initial a-, was made already in my Report (p. 32); it is applied irregularly, cf 'dyr =
ddir- "fire", "styr- — astir "lining" (Pers. dstar), y (with tasdid) — ay- "life" from Old Iranian dyu-, against y'k = yak - "egg" (with article 'y y'k), and the like; it must
not be confused with the occasional coalescence of the article with a word that begins
with a vowel or with y~, as in the case of y'mnk "the summer" (from I + ^mnk from hamina-)
or y'sm "the sky" (from i -f- 00 asm-; why Freyman, SV., vi, p. 66, attributes the
feminine gender to this word escapes me). Clear support of the derivation of the
Khwarezmian article from the old relative pronoun is afforded by the repetition of the article before an apposition (an adjective or a genitive) that follows a determined noun.
Examples: adjective + noun, undetermined znc fmvrn “insects, lit. small animals”,
determined y' wdzy'k-mync 'we “the place of strangulation)) (wadazydk verbal noun, from
awa-darzaya-); noun + adjective, undetermined r"c wdncy “old veins” (pi. of r'£ = Sogd.
r'/e; wdnyk, fern, wdnc “old”, pi. fem. wsdanciya, cf. Sogd. wtcnyy etc.), determined 'y
st'rc 'y nhsy “the unlucky stars” (pi. of st'ryk; nahsiya from Arabic nahs), y mrc y
xw'rzm “the Khwarezmian”. Determined genitives, preceding y' jfty pcx*s “the wife's (juftiya)
clothes”, cf. 's $nck name of a parasite plant, clearly “camel's tooth” (viz. ussidancik;
Olr. ustra and dant-)\ chiefly following, 'y cyy 'y xr'x'n “the blade of the sword” (taiya-\
xr'x'n gen. of xr'x, masc., cf. Sogd. xyr etc.); 'y nwk y' nyzc “the tip of the lance” (nyzc
gen. of nyzk, fern.), 'y y'k 'y zmwrk'n “the ant's egg” (gen. masc. of zmwryk, cf. Sogd,
zm'wr'k P 2, 390, zm'wrc SCE. 366), *y hwny y ricy “the blood of the nose” (huni, Av.
vohuni-', ndciya gen. fem. of ndca), 'y 'nw/t y' qlmy 'wd 'y dfir'n “the creaking of the qalam and of the door” (qal(a)miya gen. fem., dfiaran gen. masc. of <5/far-), / S'rk y'
ywc 'wd y'psy “the udder of cow and ewe” (Sarik from udar-7 yoca gen. of ybk “cow”,
Oss. yog; Olr. pasu- here capable of being either masculine, gen. *ps*n = apsdn, or
feminine, gen. 'psy = apsiya). In these genitive constructions the gender of the first
article always conforms to the gender of the first word; but the gender of the second
article conforms to the gender of the second word only if the first word is masculine: is
the first word a feminine, both articles are feminine. This circumstance is not
necessarily due to attraction; for if the article is indeed derived from the relative
pronoun, the gender of the second article should agree with that of the first word. For
example, wdc “thing, matter” (Av. vac-) is a masculine noun in Khwarezmian, and as such
has the gen. sing, wdcdn; one finds both 'y hqyqt y w'cn “the truth of the matter” (the
Arabic fem. being treated as a masc. in Khw.) and y pc y' w*cn “the end of the matter” (paca
“end” is fem. throughout); cf. also y'yt 'y w'cn “the beginning of the matter”, where y*yt
= i + ay at- “beginning” (from ayast-). The repetition of the article has a parallel in
the repetition of prepositions in equivalent situations. Thus we have without 'dtticlefydk'wkfn'n'm'n
“in the presence of so-and-so”, f- preposition, dyadkduk “presence” (abstr. of ayadik “one who has come”), f- preposition, nanaman gen. masc. of ndndm- ~ fuldn. With the
article, fy zfrkfy trk'wyk “in Turkish)), literally “in the language in the Turkish)); cy b'dys cy 'llh'n
“from the command of God)> (from the command from the of-God).
No repetition occurs with an antecedent apposition, y>> ririm w'c “in such-and-such a matter)),./)' n/?V“Srwk'nyk “in the illness of death)) (nfi'y, masc., “death)>; Srwk-“sick)) from adruva-, cf. Morgenstierne, IIFL, ii, 224).
As I have mentioned the Probesdtze in the Report (p. 34), I will not let the opportunity pass without correcting a few further points. In the second sentence, ciyy- “to enter)) does not belong to Olr. dyaw-, but derives from all + /'-; its imperfect stem is cdy- (atiydi-\ its participle is cydyk(atigata~); cf. Parthian 'dyh-: 'dgd, Sogdian tys-: tyt-. In the third sentence, the first word should be written kimi, -ml here being contracted from -mi-hi; and the name of the small coin is not ys but *ps = ipsi, developed from pis/- with the prothetic vowel common in Kwarezmian; this ipsi belongs to Persian pisi (beside pisiz etc.), which originally means “scale)) but is used of small coins (= Arabic fals;on the history of the meaning see G. Hoffmann apudS. Flemming, Akt. Ephes. Syn., Abh. G. G. W. 1917, p. 174a).
It should not be thought that the errors found in Freyman's latest article are confined to the cases mentioned above; these are merely representative cases, showing the absence
of progress beyond the stage reached in our Report. It would be tedious to fill many
pages with the list of his blunders, but we will mention a few. P. 64, the first word,
aciwa is not a + interrog. pronoun, but 'cy interrog. pron. + wa “therein)) (= wd with
preceding stress). P. 65, mis “also)> compared to Sogd. m'yd “so”, without regard to
phonology; in fact = Sogd. ms “also”; still stranger remarks on mis on p. 80. P. 64, a
peculiar derivation of the ending of 2nd sing. subj. from the 3rd sing. opt. of “to be”.
P. 67, pdrwuzda “he became)) derived from par + vart-\ but the present stem is parwuz-
(e.g. ibid. p. 68 where wrongly with z), hence pari + waz-. P. 68, ndndm- is declared to
be a pronoun; in fact, a bahuvrlhi compound of nd(n) “that” and ndm- “name”. P. 69, hyd
and nd completely misunderstood; in fact, hy + da and n + da, hy and ri being the
enclitic pronouns of the 3rd person, hy singular and n plural. P. 71 hwpyr and hwfi'r,
from hwff- “to fall)), wrongly divided into hw (which does not exist) and forms of “to
be)). P. 71, kicydmihi “I throw it away” absurdly referred to kaft- “fall”, with the
usual confusion of present stem and past participle and the accustomed disregard of
phonology; in fact, key-, impf. k'cy-, is from kartaya-cf. Wakhi kart- “to throw)). P. 72 muxxdst is said to be from xwah-, but on the next page we are told that muxxds- comes
from vi + hrz-\ the latter statement stands on the same page as the assertion that ixxi-
is the result of grz-: truly an extraordinary language; we note that in fact rz- always
becomes z in Khwarezmian, and that (i)xxiy- belongs to Avestan xsi- (with -x- or -xx-
from -xs-, see Report 32; mxy-, not mcy-, as in Sogdica 37, is the correct reading).
P. 74 y-atti “lip” wrongly compared with Sogd. ydi- “meat”; in fact, from Av. aosta-“lip” with -t-j-tt- from -st- as commonly. P. 75 xurnana explained as xurin + ndna, while in fact it is xurina “I ate” + na, end. pron. 3rd pers. pi. (see above my remark on p. 69); xurind-na results in xurnana quite regularly; the pronoun is extremely common, especially in combination with the 3rd sing. impf. -da, when -da-na produces -dana. P. 75 ws- “to say”, from Av. vasa-, is wrongly referred to Sogd, wys-, which in fact would have -xs- in Khw. (cf. e.g. 'mxs- “to learn”). P. 76 ndz Si “l (am) a woman” has been “corrected)) into ndz wude (in itself an impossible form, “wife” being u'“<5); but Si1, also dyn — din, “woman” (not “wife”), which belongs to Av. daenu-, is the correct word. Even stranger is the importation of the word for “wife” into \ydywO (in our MSS. 'cydwO, vocalized aciduff), p. 77, which means “what ('cy) to you (-di) there (-u'0)”, (*)w& “there” — Olr. awada, Sogd. 'wS. P. 11 jfty “of the wife” is wrongly transcribed jifte; in fact, the oblique case of feminine stems (excluding those ending in -k) has the ending -iya (from Olr. ayah etc.); it is vocalized -iya throughout; when a suffix is added, the ending is spelled -y'-, for example y' 'we “the place”, c' °wcy “from the place”, but c' 'wcy'h “from his place”. P. 11 pw^rcymn completely misunderstood; in fact two words, buwaracci from buwdradci from buwdrad-ci, pi. of buwdrddik “separated” (cf. BSOAS., xii, 310), and ymn “we are”, P. 78 manba, explained as “to me”, means in fact “as soon as”. Ibid., yd “they are” is either a bad misreading or a “silent correction)). The correct form is yl, mostly vocalized yalli (it occurs often enough). It may be difficult to account for it; but such consideration does not entitle one to suppress the true form and replace it by a purely speculative word. The pres. indie, of “to be” is spelled (without vowels) sing, ym, y'h, yt, plur. ymn, yf, yl\ forms with vowels are rare and uncertain, sing. 1st yimmi, 2nd ydhi, 3rd yatti, yitti (probably yetti)\ pi. 1st yimmini, yimminni, yuminni, 2nd not found, 3rd yalli, yalla, yilli (probably yelle). P. 80 shydk “damaging action, obstruction” is queerly referred to Arabic Sahid “witness”. Since sd- normally represents Olr. frd-, we should pose frdhiti- or a similar form (from hay-)', cf. Gathic dhoiOoi. P. 81 fynd (fiynd) “husband” (whence fyndk “master, God”) is derived by Freyman from band-', in my view it belongs to jsuyant-, Saka ksundaa-, Pers. suy, etc., with/- from/5- (cf. x- from xs-).
1 With nasalized j.
P. 81, ma, -fd, the enclitic pronouns, are here and elsewhere referred to Olr. mam and Owdrn, while in fact they belong to the enclitic Olr. -ma and -0\vd (see Report 32). P. 82 zywk — ziwak "anyone" someone))1 appears in the transcription as ev ka-, distributed over two words and deprived of its initial consonant; and hfly, the personal pronoun of the 2nd pers. plural, is said to be a "prefix" (it forms a pair with mfiy "we" = ma$yi, genitive mpy'n = mafiydn; hence = hafiyi). P. 82-3, Bbrkk has been arbitrarily changed into tbrkk\ in fact, it is glossed du barakat'" and therefore is an adjective, while tabarruk (which in Khw. is spelled tbrwk) is a substantive. Adjectives indicating possession are ordinarily formed with prefixed 6a- in Khwarezmian; they are very numerous and have their prototype in the Avestan compounds of the type of haoa-hunara, haSa-barasman- etc. Hence, Oabarakak (vulgarism for the expected da-barakat).
Before we leave this appreciation of the value of the contribution Freyman has made to the study of Khwarezmian, a few words must be said on a major error he has disseminated, namely on his assertion that Khwarezmian shared the Sogdiano-Ossetic plural ending -t (Zap. Inst. Vost,, vii, 314, and elsewhere). The presence or absence of this ending is of some importance for the classification of the Iranian languages; it figures prominently among the material Professor H. W. Bailey collected for the demonstration of close relations between Khotanese, Sogdian, Khwarezmian, and Ossetic (Asica, 24 sqq.). However, Freyman's assertion is misleading; no such ending, in fact, existed in Khwarezmian. It is true that plurals frequently end in -c; but this -c derives not from -t but from -k. One only has to look at a list of the plural forms found in Khwarezmian to discover that the plural in -c is restricted to words that in the singular end in -k, and, contrariwise, that all words in -k form their plural by changing -k into -c. For example z'dyk "son", pi. zdyc\ 'stryk "drachm" pi. 'stryc (*ester-from safer- = stater);p'cm'nyk "a weight (= mann)", pi. p'cmnc. (Pers. paimdne etc.); 'ywck "single", pi. 'ywcyc\ snk "stone", pi. snc: swk "nail", = suwik, pi. swyc (from sruva-, cf. also sw "horn", and, for s from sr, 'swc "tears", xs "mother-in-law)2; xdk "self, same; he", pi. xd'c (Av. A''a/o); r'k "vein", pi. rY; and so on.
1 From "living, existing".
That the ending was -a, not a mere -c, is shown by the spelling when a suffix is added, e.g. ~*nwk "knee", pi. with suffix 'y fnwcyh "his knees"; hence, sing. zanuk, pi. zdnuci, with suffix zdnuclhi. Cf. also the passage printed by Volin, Zap. lust. Vost., vii, 91 where xw'jcyh and r'ycycyh, the plurals of xw'j(y)k and r'ycyk, face msryfn'h and riybrih, the plurals of msryf and n'yb. The two latter forms indicate the normal plural of nouns other than those ending in -k\ it is in -ina (in final position -ina, before a suffix ~(i)na-). Certain such words, however, possessed a plural1 that remained unmarked in unvocalized text, e.g. 'my = "bird" and "birds"; the rare suffixed forms show that in these cases we have to presume the presence of a final vowel, namely -/, cf. y pnyh (nom.) "its feathers" (parna-, with regular -n- from rn-, cf. w'nync "sheep-wool", Av. cardna, cn-'ngbyn = Pers. tar(r)- angubin, znwk "crane", Pashto zdna, h'ny-"shake", Sogd. >“-, -knynk "making", etc.); they differ from the corresponding singulars (nominative), which are spelled masc. z'dkh "his son", $wmh "its tail", nsh "his lancet" (note ywh "its colour" = yuhi with nasalized -“-, from ywn), fem. n'c'h "his nose", dyd'h "his daughter", kfwk'h "its foam", y* n'xk'h "its claws".
Evidently nouns in -&, as the words mentioned at the end of the preceding paragraph, formed their plurals by adding -/ to the singular stem, and -ki became -ci (by -e/). Similarly, the feminine nouns that in the nom. sing, end in -k(a)t have the genitive and ablative ending -c(a), while other feminine nouns have -iya in the genitive and ablative; clearly, this -ca is the outcome of an earlier -kiya (by -kya and -ca). Several fem. genitives have been mentioned in this paper, such as ywc from ywk “cow”, nyzc from nyzk, nyzyk “lance”; cf. further y° dnc from dnk, dnyk “bow” (with tasdid over -“-; prob. from druna- by dm-); y' fi'rcyc from fi'rcyk “riding animal”; y' zycwk “life”; and contrast y* dwyty from dwyt“ink-pot”, y" *sy (ussiyd)“of the she-camel”, y' *kty “of the bitch”, y' xry “of the she-ass”, y' *wdry “of the belly”, etc. (an exception to the rule is y' psky from psk “back”). There are not many ablatives; a good example is c" rknbwrc “from the she-hyena” (rknbwryk “hyena”, masc. 2 or fem. ace. to natural gender), in contrast with c' fiwmy “from the land” (flumiya, nom. j$wm\ c-xby “at night” (xsap-), and the above-mentioned c' 'wcy\ c* 'wcy'h “from \ place”.
1 A form of this type may have been in Freyman's mind when he attributed a third plural ending -e to Khwarezmian (and moreover to Sogdian). No such ending exists in Khwarezmian
(nor in Sogdian).
As regards the position of Khwarezmian within the circle of the Iranian languages, the closeness of its relation to Sogdian on the one hand and Ossetic on the other has perhaps been overstressed; some of the most striking resemblances to Sogdian may be due to loans. Some of the sound changes that so far have been neglected may be mentioned here, especially those that show the connection of Khwarezmian with the languages adjoining it towards the south and south-east (Essentially, a reference to Pashto, without indication which Pashto is meant: modern Pashto, Middle Age Pashto, or the Pashto contemporaneous or preceding the ancient Horezmian language).
Not only wi- becomes “- (cf. also 'wsyc “twenty”, 'wsd “wide”), but also MT-, e.g. 'wr(y)k “wolf” (fem. 'wrk'n), °wsyk “hunger”, "w&ynd “hungry”, 'wsnycyk “male”, 'wznyk “neighbour” (Av. \3rdzdnya-). -h- is mostly lost (e.g. s'k “hare”, nc “nose”), but sometimes maintained (w'h “price”, pxw'h-nyc ptcple. from pxy- “cut”, Sogd. pxw'y-; hwny “blood”), sometimes added (t'h “thief” from tayu-, hrs “bear”, h'ks “mountain goat”, Wakhi yuks). -y- is lost after -s-and -d- (s'w “black”, ksb “tortoise”, md'n “middle”). Initial xw-becomes ux- or ax~, but sometimes x-, cf. 'x “sister” = uxa (from xwaha), 'xyS “§weat”, mxyd “he beat” (Oss. xvayun), but xyr “sun” beside rxr, xsr “father-in-law”; xwd- remains or is axwd-, e.g. 'xw*dk “weak” (Sogd. xw'/), beside which we find mx'st — mux(x)ds-t “grew tired” (Sogd. xw's-, Parth. wx's-). While -$w~ appears as -/-,I cfr “four”, cfrys- “forty”, as in Parthian, -dw- becomes -6jt- as in Sogdian, e.g. dficy, dficym “second”, Sfizk “thick”, 6fir “door”, 'rdpk “erect” (Av. araSwa-), Sfi'ny- “winnow” (see IIFL., ii, 222), b'dflzyd'h “he unfolded, spreat out”, intrans. impf. b'dfixsyd, participle b'dfiycyk (cf. Sogd. wySfiys- etc.). Similarly, zw is -zft- in z$*k [so, not zfi'k) “tongue”. Irregular^ for -w- perhaps in fine (?) “rice”, finyk “mark, characteristic” (wor/ia-?), and bz'fir “lean”, cf. Pers. nizdr (if indeed from nizawar).
While old z(j has become -z- throughout (yyz “snake” from azi-, zyw- “live”, zywy- “revive”, z*my- “bring”), both rz and rz appear as 2, cf, kz “difficult”, (Sogd. krj), wz “thread” (Yidgha wirz etc.), "znd “worthy” (Chr. Sogd. 'yznd-), sz “wild beast” (Pers. sarze), *ndz “(knee-) fetters” (Av. handaraza), xz, xzk “good, sound” (Oss. xorz), perhaps 'zd'nk “sheath (of a stallion)” if from Av. zrszi- + ddna-. Correspondingly, s often becomes -s~ (see below), and both rs and rs turn into s, cf. ps and '&s “mane”, ksyk “strap”, mwsyd “he was hungry”, ysyk “glad”, fsyd “he was g!ad” (Sogd. wys- etc.), w'smyk “a wrapper or veil for the head” (Arabic ximdr) — Pers. basame, Arm. varsamak. Further, 5s “goat-hair” (Wakhi dirs, Yidgha lirs) and, with sp for original -rsw-, psp- and pspyk “side, rib” (Av. parasu-).
1 Initially perhaps as Of- in mQfnc<Th
“he collected it” (Pers. alfanj-), see BSOS., x,
105; Sogdica, 17, 32.
Most other -r- groups offer little of interest; sr and rn (becoming s and n respectively) were mentioned before, -dr- remains (drwk “sick”, drycfh “he reaped it”), and so do -gr- (yryw “sejf”, cyr “sharp” from tigra-, yr'cy “awake”, yr'm “weight” cf. Oss. dryom, cyr “Falco sacer” — Pers. cary, carx, Arab. saqr\ -rg- (mryy* “wild (animal)” from Av. marayd with adj. suff. -yan or -yd with nasalized -a; sry “lion”; dryyc “long”), and -br- (fir'd “brother”, many forms from braig “fry”, e.g. firyy “kabab”, mfiryzd'h “he fried it”, firyck “fried”). Difficult groups are Or and jr. The former becomes -r in postvocalic position, pur “son”, purandir “son-in-law”, cwrSys “fourteen)), and so also in the often divergent Khwarezmian dialect of al-Beruni, e.g. Yiv from aOro, 'xsrywry from xsaOrahe vairyehe, 'ryyn from vdraQraynahe, etc. In initial position we have s-, as in Sogdian, sy “three)), sys “thirty”, s'sy- “strew)> (Sogd. dr"s- beside s's~), but once hr~, in hrdys “thirteen)) (cf. Parthian hry\ once arc-, in "rcy^dyk “third part)) (with Av. ydta-) beside 'rcy'my (reading doubtful) “third part)) (the ordinal number is sym), and once 6- in Oyd (as pointed out by Freyman); the latter form is due rather to loss of ~r- in a cluster of consonants (from Brxta- instead of Braxta-) than to any regular change from dr- to 9-1. Similarly divergent treatment, due no doubt to the influence of a number of dialects, occurs in the case of fr-. It may remain (wfrk “snow, zfrk “deep”, mfrysdyO “he sent to him”) or become /: (fyk “oar”), and sometimes appears both as fr- and f- in one and the same stem. E.g. fratara- is 'ftyr “before)> and frdr “better”, fratama-is j'rdmy frdym “chief, best)) and ftmyck “first)) (~t- remaining after reduction2),fri- gives b*fnd“he created)), b'fneCh “he created it)>, beside y" bfryc “the creature(s)” (unpointed and doubtful except for -fr-), friya- supplies fy*ny (unpointed) “love affair(s))), 9fy"n “friend” (Ba- + fy'n-),fyywnyk “friendly)) (pa.r\,h.fryhgwn\fyywny*d “kindne$$, friendliness))3).
1 Cf. dnyk "bow” from druna-.
However, the normal development of the preposition fra-is sa- (cf. Balochi and Ormuri); I have noted more than a dozen examples, among them smctTh “he took it off (clothes))) (fra- + muc)\ smsyd- “he rubbed)) (Bal. musag etc.), s'k'ry- “to paint” (cf. Pers. nigar); s'w'c “voice)) (cf. Av.jravaka-), snsyd “he went astray”, sntyk “astray”, caus. impf. Frisyd (Jranas-, franasta-), s'wzd “he jumped”, caus. impf. sw'zy- “throw” (fra-waz-) sbw'k “puro> (fra-pawaka-1), cf, also s'S “nine” if from frdd- “increase” (the only aberrant numeral of Khwar.; “90)> is nwyc, “19” nw'&ys). In addition, there are a few cases where one suspects that/r- has become r-, e.g. rxnd winsultw rxyz- (?) “arise, occur)), cf. *rs'wyk “pumice” (fra-saw-1) and p'cfwyd'h “he recalled it” (Sogd. pt$r*w-)\ that this change existed in a Khw. dialect is assured by Beruni's rwcn (rawacina) = Av. j'ravasinqm. — xr remains in the middle of a word, txryk “bitter”, cxyr “wheel)), but loses -r- at the beginning of roots, xn- “buy”, sxn- “ransom)) (Sogd. syr'yn-), p'cx'wd-“he scratched)) (BSOS., X, 509); doubtful is the occurrence of -rx in p'rxwdd “he became lost, giddy, dazzled”, possibly from apa + Av. xraoda- (cf. the metathesis of gr in Beruni's *wnry).
As confusing as the treatment of Or and fr is that of postvocalic -s. It remains in a few words, perhaps throughout loanwords, pws “cat”. pwpsyk “hoopoe” (Pers. pitpas), 'swk “target” (cf. Arm. nsavakt), but normally becomes -s, e.g. *ws “attention” &ws “intelligent” (Ba- + 'ws). 'nyws- “listen)), Sws- “to milk)>, etc.; cf. also sry “lion” for initial s-(of doubtful origin). Besides, we have /i-sounds, namely x in ywx “ear); and h in 'mh “ewe”, obviously from Av. maesi (cf. also mhynk “ram”; formed as 'fizynk “he-goat” from '&z “goat”, Pers. buz), and in sp'k “louse)>; in nhst- “sitting” (cf. nyO- “to become))), -h- is probabl> original. And in addition, there are, surprisingly, clear cases of -/- a^ the result of-,?-, namely mw/“mouse” and/w/ (or firwf) “flea”; this change occurred also in the dialect spoken by al-Beruni, who has VHJ for sraosahe (with sr maintained in contrast with the other forms ol Khw.) and wwfyk “Virgo” from (h)ausaka- “ear of corn” [/from S onl> after labial vowels].
A few groups with sibilants: sn (wsn “because of)>), zn (pznyk] “cushion”), zg (mzy “brain”, zyyk “horn)) from azga-, zyryk “coat-of-maib) Oss. zyar, Psht. zyara), zd ('zcTk, unpointed, “clever”, frorr azda? yzdk “rich)> Oss. qazdig, on which differently Bailey, Asica 15).1. zn from rzn.
zm (zm “fuel”), sm (y'sm and y'sym “the sky))), ms (6ms- “be tamed”) sm (cm “eye”, dmn “enemy”, Smn'wy'd “enmity)>) xsm (y'xmyk “the moon” from MJCSWJ-?), xsn (pxnwr'w “ungratefulness)) from apa-xsnauOra-), xs (normally x, \v"xd “he grew)) xwfcyk “sweet)) frorr xwfc- “milk”, 'xfi'cyk1 “kingdom”, etc., but xsw6y-, unpointed, “to wash” from xsaudaya-), fs (mostly f,fynd “husband”, cfk, cjyc “sour, vmegar” from trjs-,ft'n “breast” 'rdf-ffn or 'rdf-frin corresp. to Av. arddva-jsni, but bsprm'c “shamelessness” with sp from sp from /?; mdjsydh or mdfi- “he envied him” from Av. dvafsa-1 both with -s-, perhaps s'fs'nytfh or &f$- “he smoothed it with a file” from fra-fsanaya-1), fs (xfst “he was eclipsed” /rom grfsa-), s£ (pc “after”, pcs'r “backwards”, pcrmk “arms (tied) on back” cf. Sogd. 'ps'rm'y), sxw ('nx'r ~ Pers. nisxvar), rst ('sc “lance”, but psk “back” from prst-), sir and str (both -s- or -ss~, e.g. ws “grass, pasture”, fi$ “dam, dyke” from bastra-, ys “tooth”, 's- “camel”, cf. 'smy “ostrich”; also = -.s- in 'scyk “female” from Sogd. 'stryc + -k). St either remains, 'stnb “last”, b'stwd “he denied”, 'styk “bone”, mstkyk “fresh butter” (from mand-; Pers. masge from mastke), etc., or appears shortened to -s-, as in/sW “master”, 'swr2 “large animal” (but once, disconcertingly, 'scwr3, as pi.), w'syd “he placed” from awastayata (beside wst'dyk from awastdtaka-), cf. also sxyr “osprey” (from asti-xwara-); possibly -c-, by -sc-t in pck “egg-shell”, if from pustaka- (butpwst also exists, as “fur”, perhaps loan-word). St similarly either remains, 'st, 'st-Sys, 'st'c, 'stzd “8, 18, 80, 800”, 'stye “brick”, 'syw, pause form of 'sw from sravah-, probably also in Sstw “poor”, dstw'wyk “poverty” (Sogd, Sstw'n), if, as is likely, ~st- is due merely to the common habit of incomplete pointing; or becomes -/-, 'yt “beginning)), sntyk “astray)), mtyk “broken)), mt-zyyk “with broken horns” (m'z-d“he broke)), Parth. 'mst, Psht. mat, Ormuri maz- mast-), nk and rtynk “true)) (beside rst and rst'wnd), perhaps y't “the lip” and wtr “twirl)) (wastar-l), cf. perhaps also yt “is” from asti by as//? The resulting -/- has further developed into -c- in mck, 'mck “fist(ful)” from musti- (cf. the word for “egg-shell” above).
A remarkable feature of Khwarezmian, which it shares with Younger Avestan, is the occasional appearance of -0- in the place of -S-, as in 0- “with” (~hiO “with him”, &xr “sunny”, Owz “threaded (needle))), 6'wfy “faithful”), mOx “locust)), nyd- “become)), 00 pQk “house” (probably from pada-\ k^lbyQ “form)); presumably in nmQk “salt)) (see Sogdica 8, BSOAS., xii, 55). It is true, though, that devoicing occurs sometimes in K_hw. (e.g. in crs “bustard)) — Pers. tarz).
1 '\h- "to rule".
Ordinarily, Khw. -6- represents Olr. -9- (e.g. ywO “excrements”); it may render foreign t + h (m&'ryk ~ Arabic matharah); and in one case Khw. -s- responds to Olr. -Oy- (perhaps by -s-, cf. OPers. -siy~ MPers. -s), namely in xbsk — xubisk “own” (MPers. xwybs), supported by xbsk'wnd “master” (cf. Sogd. xypS'wnd).
A brief list of some interest may conclude this survey, which, it is hoped, will afford some guidance to those who may wish to study the Khwarezmian material contained in the Muqaddimatu -/- Adab. “Paradise” is yrdrrin, “hell” tm- (as in Sogd.), 'sbnd'rmd “earth)), fiyk, 'fiyyk “doll” (from “god)); cf. Pers.ywy), 'rd “feast” (Av. ratu-\ Beruni has ryd). K's “pig)) (Sogd.), p"Q “arrow” (Asica 11), pd"ryk “slate” (Sogd. pytfr, S. T., i), mand- negative prefix as in Sogd. (mndtrinynd “dissimilar)), mndcyr “blunt”, mnd'ktrm'n1 “disobedient))); rwbs “fox”, 'bwd I) “silk”, 2) “woof” (Sogdica 19), p'rS-“selh), tsyckyk “axe)> (Sogd. tsycq), wyryk “saw” with wyryd'h “he sawed it” (Sogd. cnn wyr'kh “with a saw” P 21, iii, 3; en wyr\ S. T., ii)2; w'rynyk “royal falcon” (Av. and Sogd.), mrk “monkey” (Sogd. mkr'), drmcyk “scorpion)) (from drmad- cf. Sogd. nyrd^k from drdab-and, from drdam-, Pasthto laram, Pers. dilamak), yrSk “neck” (Sogd. yrd'kh). byrfik “cloud” (Sogd. pr'y/J'k), 'ks “lean” (Sogd.). Knbynk “linen” (Sogd. kynp\ cf. BSOAS, xi 724), wyn “eye-sight”, ywr “wild ass)> (also Sogd., Pers. gdr). JfCm “mouth)) (Oss. kom\ 'rn'ny- “guide, show)) with intrans. m'm'nsyd “he was guided)) (Oss. amonin), n'rk “narrow (Oss. nareg, Psht. narai), 'lx “top of the spindle” (Oss. alxui\ excludes etym. Asica 36), pdyk “large axe)) (from p(d)rt-, cf. Asica 13), rsy “barley” (Saka rrusa- etc.). Spdyr “mule)) (Pers as tar, but Sogd. yrtr-), mr'w “date )> (Parth. 'mr'w), ywndyk “sin” (Parth. gwyndg), fi'r “cup”, 'fiyw-, trans, 'fiy'w- “increase”, 'pywnk “additional” (Parth. 'bgw-, *bg'w-), nbyk “Qur'an” (Pers. nubi, MPers. nbyg, etc.), nx'w'z “a he-goat that leads a flock” (naxaw- from naxu-, and waz-*1?
1 'ktrm'n shortened from *'kt-frm'n.
Cf. Arm. noxaz, Pers. nuhdz). Krbwn “lizard” (Av. kahrpuna-), nkdyk = Arabic Ibn 'irs “weasel, ichneumon)) (Skt. nakula, hence Indo-Iranian nakuda\ not loan-word), *zyd or 'zyd “silver” (rzata), pyxk “node” (Av. pixa-\ prw “grey” (cf. Pers. pir), twy “ji/yah” (Pers. tuxtan, Arm. toiz), fywd “beestings” (obscurely related to Skt. piyusa or MPers. frusag, BSOAS., xi, 719?), rxt “red” (Skt. rakta-) l.1 [For technical reasons, a few diacritical marks, especially points etc. under letters etc. have been omitted in this article.].
Besenyos, Ogur and Oguz
Alans and Ases
θδğŋγČčşŠšāáäēə ï öōüūû“” ’ Türkic ’Ακατζιροι