1 | Introduction |
| Daniel Abondolo |
The term âUralicâ refers to the largest language family of northern Eurasia. This family consists of at least thirty languages, spoken in communities scattered over a vast area with western limits in Norway and Hungary and with eastern limits on the Taimyr peninsula and along the Yenisei and Obâ rivers of western Siberia. Their large number, and the considerable typological diversity of the phonology, morphology and even syntax of the Uralic languages make it impossible for this chapter to provide anything more than a brief survey of some of the more salient synchronic and historical features. The newer and older names for the languages and the names for their reconstructed ancestors are introduced below; the following sections treat phonology and morphology, from both a descriptive and a historical-comparative perspective; pp. 30â1 look briefly at some of the more common kinds of syncretism and suppletion in Uralic languages; syntax is broached on pp. 31â3; and the final section in this chapter gives a glimpse into Uralic vocabulary by presenting eight selected synonyms.
Good short general introductions to the Uralic languages are Janhunen 1992, Comrie 1988, and Austerlitz 1968. For a better understanding of the greater linguistic context, Comrie 1981b should be consulted; this book has good chapters not only on the Uralic languages but on âAltaicâ (TurkicâMongolicâTungusic) and Paleosiberian (Chukotko-Kamchatkan, Eskimo-Aleut, Yukagir, Ket, and Nivkh) as well.
Internal Subdivisions and Nomenclature; Possible External Connections
The locus from which these languages emanated, in other words: the proto-homeland of the speakers of the language from which all Uralic languages come, is unknown, but a relatively large and sparsely populated region at or near the southern end of the Ural mountains is likely. Some of the internal divisions of the Uralic language family are not entirely clear, but there is close to universal agreement within the profession that the primary chronological break was between Samoyedic on the one hand and Finno-Ugric on the other. The Samoyedic languages, spoken today chiefly to the east of the Urals, are thought to descend from a form of the Uralic protolanguage which spread eastward, partly through the migration of its speakers, partly by pure linguistic expansion, and in the course of this separate existence developed its own traits and distinctive vocabulary. The population speaking proto-Samoyedic must have been quite small at first, and there is no reason to assume that it underwent any significant expansion before the dissolution of its linguistic unity, probably in the centuries immediately BCE. The bulk of the vocabulary which can be safely called common Samoyedic is surveyed in Janhunen 1977, which contains 650 root morphemes, of which only about 150 go back to proto-Uralic. This relatively small proportion makes it likely that the primary split of Uralic occurred at least six millennia ago, and possibly â given the sociological and demographic features of the early proto-Samoyedic population and the rate of change which these imply â much earlier (Janhunen 1992).
Many of the languages which resulted from the breakup of proto-Samoyedic have doubtless been lost without trace, but reports by European explorers from as early as the seventeenth century, and then by linguists beginning with Castrén 1854, have provided us with information about six distinct Samoyedic languages. Listed roughly from north to south, these are (older designations given in parentheses): Nganasan (Tavgy), Enets (Yenisei-Samoyed), Nenets (Yurak), Selkup (Ostyak-Samoyed), Kamass(ian), and Mator (Motor). The southernmost languages, Kamass and Mator, are now no longer spoken: Mator was replaced by Turkic idioms during the first half of the nineteenth century, and the fact that it is known at all today is because of intensive philological work done with word lists; the last Kamass speaker died in 1989. Of the more northerly languages, only Nenets is spoken by a relatively large number of people (some 27,000); Selkup, which has sharp dialectal divisions, has fewer than 2,000 speakers; Nganasan, some 600; and Enets, perhaps 100.
Compared with Samoyedic, the Finno-Ugric branch is and probably always was the larger, in terms of both absolute speaker numbers and internal subdivisions. At least one of these subdivisions must be quite old, dating back at least as far as the third millennium BCE. There is no universal agreement about the precise membership of the two groups which resulted from this first break within Finno-Ugric. As the name Finno-Ugric itself suggests, the traditional view since Donner 1879 has been that the two main subdivisions of this branch are:
1 a âFinno-â sub-branch, which consists of Saamic (Lapp), Fennic (more commonly termed Baltic-Finnic), Mordva (Mordvinian), Mari (Cheremis), and the Permian languages Udmurt (Votyak), and Komi (Zyrian, Zyryene);
2 a âUgricâ sub-branch, consisting of Hungarian and the ObUgrian (Ob-Ugric) languages, Mansi (Vogul), and Khanty (Ostyak).
Although basically correct, Donnerâs work was a false step, taken in haste by one whose forte was not the methodical side of historical-comparative linguistics (it was meant as a response to another, but intriguingly, mistaken publication of the same year by his rival, Budenz). The problem with Donnerâs subdivision lies not so much in its postulation of a primary divide between âFinno-â and âUgricâ, as in the order of the subsequent subdivisions of the non-Ugric branch. Whereas Donner assumed that first Permian, then Mari-and-Mordva, and only then Saamic and Fennic had broken away, work since the 1970s comparing the Samoyedic branch with the Uralic family as a whole, but more particularly with the reconstruction of Fennic and Saamic, has led to a revision of this view. This revision sees Fennic and Saamic as forming a node of western peripheral languages, one which broke away from the rest of Uralic quite early, perhaps not long after the breakaway of Samoyedic; the SaamicâFennic protolanguage itself had already begun to break up in the second half of the third millennium BCE (Sammallahti Saamic chapter 2, and 1984). Working with developments of the consonants in the Finno-Ugric languages, Viitso (1996) has now proposed a refinement to this revision, according to which the breakup of Finno-Ugric began in the west, with first SaamicâFennic, then Mordva, then Mari, and finally Permian leaving the core. Expressed in terms of the conventional downward-growing ancestral tree, we would then have:
in which the âcoreâ is the linguistic cauldron from which âUgricâ was to emerge.
As an alternative to the tree type of diagram, the meiotic amoeba model is useful for rendering transparent certain areal/typological zones (Sc = Saamic, Fe = Fennic, Mr = Mari, Md = Mordva, Pn = Permian, Hu = Hungarian, Kh = Khanty, Mn = Mansi, Sa = Samoyedic):
Whatever the cause of the split in the SaamicâFennic node, there was a predictably uneven demographic result, with speaker populations living in climatically more clement areas expanding more quickly than, and eventually at the expense of, their northern linguistic relatives. The number of people speaking a Saamic language today is no more than about thirty-five thousand, while Fennic languages are spoken by more than six million. Within each branch the figures are no more evenly distributed: while Northern Saami has some thirty-thousand speakers, Akkala, Pite, and Ume Saami have next to no speakers left; while Finnish is spoken by about five million and Estonian by about one million, Veps, an eastern Fennic language with strong dialectal subdivisions, is spoken by some six thousand people in two non-contiguous regions of Russia, and Livonian, a western Fennic language, is spoken by fewer than forty. Treatments of the subdivisions and designations of Saamic and Fennic may be found in Chapters 2 and 3.
The next node to break free from the centre was the ancestor of Mordva, which survives today as two varieties called âErzyaâ and âMokshaâ, with a combined number of speakers in excess of one million. The similarities and differences between Erzya and Moksha have frequently been exaggerated or minimized for extralinguistic reasons; it is in any case difficult to quantify divergence. What deserves to be emphasized is the fact that ther...