Suntuubi-palvelussa käytetään evästeitä. Palvelua käyttämällä hyväksyt evästeiden käytön. Lue lisää. OK
Lexicon of Early Indo-European Loanwords Preserved in Finnish

Background to Pre-Finnic Phonology

The Baltic-Finnic languages, often called simply Finnic, comprise Finnish, Estonian and their closest relatives in the Baltic sea area, such as Karelian, Vepsä, Livonian, Võru, etc. These languages are very closely interrelated, much as the Germanic, Romance or Slavonic languages are each among themselves. The Saami language of Lapland, while being geographically close and containing many lexical borrowings, does not belong to the group itself, but constitutes in a historical sense the least remote of the distant relatives in the wider Finno-Permic group of Uralic. The relation is still relatively loose, much as the standing of Celtic or Balto-Slavic in relation to the Germanic languages.

The Finnic group kept together and distanced itself from the more remote relatives, such as Saami, through a sequence of phonological innovations during the first millennium B.C., perhaps continuing into the first  millennium A.D. The innovations revolutionized especially the inventory of consonants, eliminating for example completely the ancient opposition between palatalized and non-palatalized phonemes. A good description of these innovations and their relative sequencing is available by Petri Kallio on internet in Finnish at

The concept of “Pre-Finnic” refers to all stages of development preceding these revolutionary innovations, starting from Proto-Finno Ugrian to the final departure of Baltic-Finnic from the western Finno-Permic dialect antecedent of Baltic Finnic and Saami. The phonological (and morphological) innovations during this Pre-Finnic period were very modest. A presumed lowest reconstruction level representing the common source of Saami and Finnic (Proto-Finno-Saami) would be a language system not very different from Proto-Finno-Ugric or even Proto-Uralic itself. The description of the Proto-Uralic phonemic system, available on internet, is therefore with some exceptions valid for the whole of this period, until the splitting away of Baltic-Finnic, especially as regards consonants. All intermediary reconstruction levels in this Pre-Finnic period are utterly controversial (see The historically most well known intermediate level called Proto-Finno-Ugric, combining Hungarian and Finnic, is no exception. A methodologically serious attempt to reconstruct distinctly Proto-Finno-Permic (along Uralic and Proto-Finno-Ugric) forms for Pre-Finnic word stems has been made by Pekka Sammallahti in 1988. For a further view on this subject favoring Proto-Finno-Ugric and Proto-Finno-Saami see Juha Janhunen: 2007: The primary laryngeal in Uralic and beyond.


On the substitution of Indo-European phonemes in Pre-Finnic lexical borrowings

Pre-Finnic allowed only for two syllable lexical stems. All stems had an open second syllable ending either in allophonic /a/=/ä/ or allophonic /ï/=/i/. The words followed strict vowel harmony. If the nucleus of the first syllable was one of the back vowels /u/, /ï/, /o/ or /a/, only the back vowels /ï/ or /a/ were possible in the second syllable and in the suffixes. If the nucleus of the first syllable was one of the front vowels /ü/, /i/, /e/ or /ä/, only the front vowels /i/ or /ä/ were possible in the second syllable and in the suffixes (in Proto-Finno-Ugric /ï/ had nevertheless merged with /i/ in stem final syllables). In the oldest IE borrowings there is a strong tendency to favor -i/-ï -stems over -ä/-a –stems . The former seem to have been less marked as it alters with zero-representation. It did also not cause shortening of a PFU vowel lengthened by the Uralic “laryngeal” */χ/.

These phonotactic restrictions sometimes shortened the IE words at the time of borrowing: it does not really matter whether the PIE original for ‘ale/beer’ was *h2el-u-t=*ħæl-u-(t) like in Pre-Germanic or *h2eleuio=*ħæleuio like in Pre-Balto-Slavic, the most probable substitution would in any event be the plain stem *kalï- (in some cases, like this one, it appears that a Pre-Finnic derivational suffix may have been added already at the time of the borrowing, in some cases faintly mimicking a third syllable of the original: *kali-ja or *kali-ta).

No consonant clusters were allowed in the beginning of a word and normally all but the last consonant of initial IE clusters remained after substitution without reflexes. The substitution of the Pre-Germanic original *ghroH-t- was therefore Pre-Finnic *rošt-. Exceptions were made where the second consonant was a syllabic sonorant. In these cases a vowel was inserted between the first and second consonants like in Pre-Finnic *marta-s ''bound to die, dead' for Proto-Indo-Iranian *mrtá-s. Combinations containing initial labiovelar phonemes (like *gw) or corresponding clusters (*Hu-=*Hw-) would all result in Pre-Finnic *ku-.

Pre-Finnic did not allow for any three consonant clusters between the initial and final stem syllables. It was also rather restrictive on two consonant clusters. Difficult clusters were simplified at the time of borrowing by simply dropping one or two of the original consonants. Also metathesis is known to have occurred (*kasva- ‘grow’ <= PIE *h2ewg-s-)

It is not possible to conclusively determine exactly which clusters were allowed under phonotactic rules because the total number of reconstructed word stems is limited. Some new clusters seem to have entered only during the later stages of Pre-Finnic and mainly through the gradual erosion of resistance to clusters in loanwords.

Pre-Finnic contained only plain unmarked stops /p/, /t/ and /k/, which substituted all IE stops, except the palato-velar series, whatever their status as regards voicing, glottalisation or aspiration (there are nevertheless a few cases known for a substitution of */d/ or */dh/ by a Pre-Finnic spirant */ð/).

The IE palato-velar series was perceived as palatal and substituted mainly by palatals:

The reflex of PIE */ḱ/ was Pre-Finnic */ć/ or */ś/ (the two phonemes later merged in most daughter languages)

The reflex of PIE */ǵ/ was Pre-Finnic */j/

The reflex of PIE */ǵʰ/ was word-initially Pre-Finnic */ś/ but intervocalically (in some early Indo-Iranian borrowings) the Pre-Finnic laryngeal */χ/


Three Uralic phonemes turn up in positions where PIE had laryngeals. Uralic, which was rich in alveolars, had few guttural phonemes to chose from. In post-vocalic positions both the post-alveolar fricatives that ever existed in Uralic are represented, an extinct (velar? laryngeal?) one in the very oldest borrowings (I use *χ on this site instead of conventional */x/ to avoid confusion with the sequence -ks-)  and a grooved one (*/š/ becoming modern Finnic /h/) in some younger ones. The velar plosive /k/ is the third correspondence and the only one found word-initially, as is to be expected under relevant Uralic phonological limitations.

The correspondences work in a perfectly equal way for h1, h2 and h3. Thus:
1) PIE laryngeals correspond to PU fricative
in cases like:
nai-/naa- 'woman' < PFU *naχi- < PU *näχi- <= PIE *gwneh2-/ > Sanskrit gnā́ 'Godess', c.f. Greek gunē ‘woman’,
2) PIE laryngeals correspond to Pre-Finnic fricative
in cases like:
-Old Finnish
inhi-(m-inen) 'human being' < PreFi *inši- (<**jinši-) 'descendant' <= PIE *ǵnh1-(i)e/o- > Sanskrit jā́- 'born, offspring, descendant', Gmc. *kunja- 'generation, lineage, kin'
3) PIE laryngeals correspond to Pre-Finnic
*k in wordstems like:
kesä- 'summer' < PFU *kesä- <= PIE *h1es-en- (*h1os-en-/-er-) > Balto-Slavic *eseni- 'autumn', Gothic asans 'summer'
kulke- 'to go, walk, wander' ~ Hungarian halad- 'to go, walk, proceed' < PFU *kulki- <= PIE *kwelH-e/o- > Greek pelomai ’(originally) to be moving', Sanskrit cárati 'goes, walks, wanders (about)’, cognate Lat. colere 'to till, cultivate, inhabit'
teke- 'do, make' ~ Hungarian tëv-, të-, tesz- 'to do, make, put, place' < PFU *teki- <= PIE *dheh1- > Greek títhēmi, Sanskrit dádhāti 'put, place', but 'do, make' in the western IE languages, e.g. the Germanic forms do, German tun, etc., and Latin faciō.

Comments and suggestions to: jschalin[h2ed]iki[dhudo]fi
©2020 Tcoimom hmom -