Fifty years after Rahimi & Uspenskaia’s inescapable Tadzhiksko-russkii slovar’ (Moscow, 1954), this new and somehow official Tajik-Russian dictionary, encompassing 65,000 words and expressions according to the colophon (a rough calculation on the amount of entries per pages gives a total of approx. 40,000 separate entries) reflects many trends in the written language of the past two decades.  Forgotten and unwanted literary words, the use of which has been recurring since the mid 1980s, have found their formal place here.  This includes classical and dismissed lexical items (e.g., ozin, mavodi mukhaddir), but also many disused Arabic plurals (e.g., ibod, ruaso) or rare derivatives (e.g., mutanakkir, taahhul) and obsolete words (e.g., be'r, dizh), most of which are to be found in Shukurov et alii’s Farhangi zaboni tojiki and in S. Ayni’s Lughati nimtafsilii tojiki but some even not (suds, ghurri).  A few Persianisms, who have found their way in media and high level language these last years, have been also included (e.g., vozhanoma, siyosatmador).  In some cases, the possible indecision on majhul vowels (e.g., vezha, spelled also vizha, alongside with niru and neru) has been reflected by cross-references.  Lexical and semantic innovations (e.g., shahrvandi) have also found their place.  Some few colloquial forms have been included, but no regional or dialectal elements at all; some accepted dialectal forms have even been removed.   As customary for Soviet lexicography, no shameful body parts are to be found.  Unsurprisingly, the dictionary follows the new alphabetical order adopted in 1998 and the new spelling conventions but one may regret that it does not provide any Arabic spelling.  One may also note that present stems are not shown any more.  At first sight at least, definitions seem pretty much accurate and precise.  Time will tell whether this dictionary is to be the reference for the next fifty years—even if it may become outpassed by the long-awaited Tajik monolingual dictionary.  In any case, its low print run (1,000 copies) will certainly go against it. 

François Ömer Akakça, Humboldt University, Berlin
CER: I-6.2.B-537