Haisee provosoidulta - aineiston arvioinnin ongelmia Dmitri Tsvetkovin vatjan sanaston valossa

• Johanna Laakso

Avainsanat: kieli-intuitio, lähdekritiikki, provosointi, sanaluettelot, vatja

Abstrakti

Suspicions of provocation: problems of data assessment in the light of Dmitri Tsvetkov's Votic dictionary (englanti)

2/1996 (100)

Johanna Laakso

Suspicions of provocation: problems of data assessment in the light of Dmitri Tsvetkov's Votic dictionary

The subject of the article is a dictionary of Votic (Tsvetkov 1995), in which the invented example sentences seem sometimes quite natural, but sometimes rather artificial, even provocative. The evaluation of "grammaticality" or "genuineness" of data also concerns note-taking and interview techniques and conventions, and matters of ideology and previous expectations. The basis of the evaluation is the scholar's intuition.

The genuineness of linguistic research material has traditionally been assessed in terms of "purity" (freedom from foreign-language influence) and "spontaneity" or "unconscious production". Both these criteria are problematic. Filtering out foreign-based data on closed-system criteria (such as phonotactic incongruence) restricts the material too severely, but other criteria (e.g. extralinguistic ones) are too flexible.

Clear examples of "non-spontaneous" material include metalinguistic expressions, and those that unduly stretch abstract formation paradigms, whether original or foreign-based. The situation is particularly complicated when the research concerns a dying language form which is subject to powerful influences: one typical feature of dying languages is abundant metalinguistic expression, but also typical are the same features that are considered characteristic of spontaneous, unplanned discourse. Genuine expressions would be those that are more extensively anchored to a particular time (idiomacity, lexicalization) or language community (rituals, folklore). Some criteria of genuineness are thus language-internal, others are extralinguistic. In the final analysis, the problem lies precisely on the border of the language system itself, in the treatment of its interfaces, and this also reflects the scholars own theoretical and ontological concept of language.