É. Apor (ed.): Codex Cumanicus. Ed. by Géza Kuun with a Prolegomena to the Codex Cumanicus by Lajos Ligeti. (Budapest Oriental Reprints, Ser. B 1.)

L. Ligeti: Prolegomena to the Codex Cumanicus

PROLEGOMENA TO THE CODEX CUMANICl'S 31 The original manuscript itself must have contained mistakes arising from mishearing, and misunderstanding. The simplification of the consonant cluster in kale for kark «dowing» might belong to this category. The Italian Latin orthography of the Codex is generally a good source of reconstruction. However, there are cases 'w hen the reconstructed pronounced form is problematic or clearly wrong. Such is the si sign for S: already Salemann (p. 953) pointed out this feature of orthography. Such are: tasina for taSna «thirsty» (M tiSna, B tafana), siac for Sax «twig». Similar is the case of ys, is: teysa for tèSa «a carpenter's axe» (M, B id.), harais for hatais read dtaS «fire» (M àtaS, B hataS)*

0

,piuskseu for puiskseu read puSkSeu «froast» (MpaS{ak)-i Sab, B puSkSaw), aysdaha read aSdaha «dragon» (M azdaha, B azdalian). In modern publications, the misidentification of the Persian words of the Codex is rare. E.g.: jawan [gaouá] «millet» (105:3), the presumed equivalent of P jau-an «rice» (B p. 151); gaouá «pistù» should he correctly read as hdvan, meaning «mortai». It is an earlier mistake later corrected (without the cancellation of the erroneous reading).

4

1 There are other points of contention between the two modern editors regarding details of transcription, phonemic and semantic interpretation of the Persian items in the Codex. Differences of opinion are possible in future as well. Salemann's position is valid to date: the Codex reflects a 13th century vulgar dialectal Persian, which has nothing to do with the contemporary Literary language (pp. 951 952). Consequently, greater attention should he devoted to the Persian linguistic records of the 13th 14th centuries, and on the findings of Persian dialectology. The excellent work of Gilbert Lazard

4

2 is highly instructive with regard to classical texts containing dialectal vulgarisms. To improve our knowledge on 4

0 B (p. 29) emended P [harais] to [harmis] «fire» and compaied it with 01. garmiS (meaning «heat; a fever»). Obviously there was a mistake here, realized by B himself who left it out of the glossary (p. 139), replacing it with the reading hàtaS [harmis for hatais], compared with P atiS «fire». Personally I would think the initial A-is a characteristic of Italian orthography, and would disregard when dealing with Persian. M (p. 20) started from an *átaié form. He considered the initial h a prostetic element, and tried to derive the ai diphtong from Middle Persian antecedents. This is completely unwarranted: is is a special sign to denote the phoneme cf. supra. 4

1 The incorrect «millet» javán also slipped into the English index (p. 223). Interestingly enough, the word in question occurs twice in the Codex, parallel with another word. Thus, e.g. on p. 105, lines 2 — 3: mortale — P haouá — C touguf and pistú — P gabouá C cheli; the two words are listed here among tlie names of kitchen utensils. On page 82, lines 17 — 18 they can be read in the section on spices among relevant utensils: mortalis — P aouan dasta (the second word is crossed out) — C touguf and pistú - P haouan dasta — C touguf id cheli. M (p. 164) would change gaoua to g. dasta on p. 105. He is right, too. Kachikian's list of wares also includes this article (p. 249) in the form Khavandasta, meaning «mortier et pilon». 4

2 Gilbert í.a,7.axd,Lalanguedes plus anciens monuments de la prose persane, Paris 1963.