Suomen kaan-liitteen alkuperä

• Alpo Räisänen

Avainsanat: fonologia, liitepartikkelit, liivi, morfologia, murteentutkimus

Abstrakti

The origin of the Finnish enclitic particle -kaan (englanti)

3/1996 (100)

Alpo Risnen (University of Joensuu)

THE ORIGIN OF THE FINNISH ENCLITIC PARTICLE -kAAn

The Finnish enclitic particle -kin (e.g. poikakin 'a/the boy too, even a/the boy') has a variant form, -kaan/-kn, which is used in negative clauses (e.g. poikakaan '(not) even a/the boy'). In the majority of the other Baltic-Finnic languages, the particle -kin is used in both positive and negative clauses. A negative variant occurs only in Izhorian (-kaa/-k) and in White Sea Karelian (metathesized as -kana/-kn); in the rest of Karelian -kin is usually used in both contexts. Geographical considerations would thus suggest that -kAAn has developed during the separate development of Finnish.

The received view assumes that -kAAn was originally a compound suffix: the kA is the same suffix that occurs in the interrogative pronouns kuka 'who?' and mik 'which (one)?' and the final part developed from the enclitic particle -hAn (a marker of given or assumed information). The particle -kAAn is assumed to have developed from -kAhan/ -gAhAn, the weak consonant occurring after a vowel in an unstressed syllable, according to suffixival consonant gradation (a feature of Late Proto-Finnic) (Hakulinen 1979: 237; Mkel 1993: 10).

This article puts forward an explanation that differs in part from the received one. This explanation rests, firstly, on the observation that northern Finnish dialects did not originally contain an intervocalic h, in spite of the fact that it generally occurs in these dialects (Mantila 1992: 64-67). Secondly, it is argued that the -kAAn particle in the eastern Finnish dialects originally contained an intervocalic h. The long vowel AA which resulted from the loss of the voiced fricatives "g" and "d" in non-initial syllables has not been preserved in the Savo dialects: e.g. *antaga- > central Savo antoo = standard Finnish antakaa 'give (2. pl. imp. )', *kertada > kertoo = standard kertaa 'time (partitive sing.)'. However, AA has been preserved in the -kAAn particle: ei kuk(k)aa 'no-one', ei miesk 'not even a man' (cf. Savo lampaat < lampahat 'sheep'). A variant "weak grade" of the particle, -An (the Savo variant is generally derived from *-Ak), is also frequently used in the Savo dialects: e.g. ei siin = standard ei siinkn 'not even there', ei tullunnaa = standard ei tullutkaan 'didnt come after all'. This "weak grade", however, is completely independent of the syllable count in a word, which is a clear indication that it is not the result of phonologically conditioned consonant gradation.

On the basis of data from dialects and old literary Finnish, it is argued that the -kAAn particle developed from the interrogative pronoun in the old western dialects (i.e. the Hme and the southwest dialects). Interrogative pronouns underwent suffixival consonant gradation, clear evidence of which is to be found only in the Savo dialects: kuka: kettee (< *ketg), mik: missee (< *missg). The function of the suffix in the weak grade became blurred with the loss of the fricative "g": the long vowel in the forms ket, mit, miss etc. was gradually understood as a means of emphasising the negation. As in other Baltic-Finnic languages, the -kin particle was originally also used in negative clauses in Finnish (this still happens to some extent in the eastern dialects), and thus a final n was attached to the long vowel AA. The long vowel formed from the weak grade spread by analogy to interrogatives in the strong grade: kukaan, mikn, which, from a historical viewpoint, are pleonastic (*ku-ka-ga-, *mi-k-g-).

Interrogatives with -kAAn also form exceptions in contemporary Finnish dialects and standard Finnish: they are unique in that they have a weak grade (ketn, mitn, missn etc.) in all Finnish dialects. Interrogative -kAAn spread and became a general enclitic particle, which could be attached to any nominal form or verb. The weak grade of the enclitic generally occurs only in the Savo dialects.

As evidenced by the quality of the vowel, the long vowel in the -kAAn particle is clearly a late development in the eastern dialects. Thus, we can assume that it was borrowed from the western dialects into the eastern dialects, apparently sometime between 1000 and 1500 A.D. As there was no long AA in non-initial syllables at this stage in the development of the eastern dialects, AA was replaced by the sequence AhA. This process also accounts for the intervocalic h in the enclitic particle kAAn; and on this basis, it can be concluded that -kAAn never contained the particle -hAn. Particularly conclusive as evidence are the following instances from the northernmost Savo dialects, the Kainuu dialects: ei kertovaa (< *kertoa < *kertaa < *kertada + hak) = standard ei kertaakaan 'not even once' and ei vhej (< *vhehk) = standard ei vhkn 'not even a bit', where the long AA created with the loss of the fricative was diphthongised prior to the attachment of the enclitic particle. These particular forms are now conceived of as idioms.