ИСТИНА

In this talk I discuss the difficulties in part-of-speech attribution of Old Russian криво. My research is based primarily on the Old Russian subcorpus of the Russian National Corpus (RNC). In addition, I consider the following sources: historical dictionaries, tagged texts from the collection of Old Russian manuscripts («Рукописные памятники Древней Руси», http://www.lrc-lib.ru/), and also a number of offline sources. I analyze the following contexts: (1) да аще кто ѿ руси или ѿ грекъ створи криво да оправлѧеть то (Povest’ Vremennykh Let, Laurentian Chronicle, 12) ‘If any of the Russians or Greeks did a bad (“awry”) job, let them correct it’. In this context the dictionaries, as well as the RNC and lrc-lib.ru, consider криво an adverb. It is quite likely, however, that it is a noun (compare to добро ‘good’ and зъло ‘evil’). Such an example is found in Ptchela, 287: вѣрных͡ мни не тѣхъ иже по твоемоу словоу молвѧть но иже противѧть сѧ гл͠емымъ тобою по кривоу ‘Consider faithful not those who repeat your words, but those who disagree with your untrue sayings (“saying by wrongfulness”)’. Consider also the following example: (2) поутѧта пьсалъ . даче криво да исправите а не кльните (Menanion XI c., 135) ‘Putyata wrote (this). If (it is) crooked, correct (it), but do not curse’. It appears that in cases like (1) and (2) криво can either be classified as an adverb meaning “making a mistake in action” or as a noun meaning “a wrongful deed”. Note that the first sense is retained in modern Russian: Они как-то криво договорились и не смогли встретиться ‘They made an awry arrangement and failed to meet’. In contexts like (3), however, we may not classify криво as an adverb: (3) дажь въ нѧ поѧ обрѧще криво а вьсе <...> (Novgorod Menanion 1095-1097, September, 176 – a margin note by scribe Domka) ‘Should (anyone) conducting a sermon by them (these books) find a mistake, …’ (fragment ends). Обрѧще криво does not signify making a mistake in the process of acquiring (обрѣсти). Here криво is an object, and therefore a noun, which votes for the noun interpretation in (2) and, evidently, in (1). In a number of contexts the homonymy is impossible to resolve, and this should be reflected in the morphological markup in corpora and other electronic resources. The solution may be to introduce double tags for these cases. The same attribution problem is observed in other Old Russian words ending in –o, e.g. in право. According to the corpus, however, ambiguity is much less common for these lexemes.