Facts and myths about Uralic studies: a review article of Jazyki mira: Ural’skie jazyki. [Languages of the world: Uralic languages.] Moskva: Nauka, 1993. – 398 pp. (Otvetstvennye redaktory: Ju. S. Eliseev; K. E. Majtinskaja)

by Tapani Salminen (Helsinki)

“Ural’skie jazyki” is the first volume in an ambitious publishing project dealing with all of the world’s languages. It is a synchronic survey of a basically random collection of modern languages, not a diachronic study of a group of genetically related languages. For a reader less familiar with the Uralic languages, the status of the relationship between the languages may even remain obscure. This should not be surprising, since both the inter- and intra-branch diversity of the Uralic family equals that of Indo-European. This fact has sometimes been denied, e.g. in a recent handbook by Ruhlen (1987). Another point where Ruhlen, following his general presumptions, takes an untenable position concerns the alleged genetic relationship between Uralic and Yukagir. The lack of diachrony, however, has led to the rather positive result that so-called long range comparisons, including the notorious Nostratic hypothesis, are barely mentioned.

The lack of a historical aspect is further reflected in the absence of any lengthy treatment of the history of Uralic studies. In this context, it is apt to mention a competing volume with the same name (Sinor 1988; reviewed by Salminen 1990), which includes a historiography of Uralic linguistics by Wickman (1988). Korhonen (1986) is a larger study of the Finnish scholarly tradition. Ruhlen (1987, 64–77), in spite of its other shortcomings, can also be recommended for its excellent summary of the topic.

There is no consensus about the actual number of Uralic languages. The extreme minimum count, still supported by the most conservative viewers, would give us no more than eighteen languages, viz “Lappish” (i.e., Sámi); five Finnic languages, Livonian, Estonian, Votian, Finnish, and Vepsian; Mordvin; “Cheremis” (Mari); two Permian languages “Votyak” (Udmurt) and “Zyryan” (Komi); three Ugrian languages, Hungarian, “Vogul” (Mansi), and “Ostyak” (Khanty); and five Samoyed languages, “Tavgi” (Nganasan), “Yenisei Samoyed” (Enets), “Yurak” (Nenets), “Ostyak Samoyed” (Selkup), and Kamas, which would represent all of “Sayan Samoyed”.

In contrast, the most comprehensive list of Uralic languages, in the sense that each one of them requires a separate linguistic description, would include forty-six languages, i.e. eleven Sámi, nine Finnic, two Mordvin, two Mari, three Permian, eight Ugrian, and eleven Samoyed languages. “Ural’skie jazyki” covers twenty-four languages, which means that it adds Ingrian and Karelian to the minimum list of Finnic languages, but subsumes Olonetsian and Ludian under Karelian, splits Mordvin, Mari, and Komi into two languages, and recognizes Mator as a language distinct from Kamas. The Sámi languages, peripheral as they are from the Russian point of view, are represented by Kildin Sámi alone, and Yurats, an extinct and sparsely documented Samoyed language, is ignored. Of the so-called main dialects of Mansi, Khanty, Enets, Nenets, and Selkup, only one variant of each is described. The dialects of at least Estonian and Finnish are also divergent enough to require special treatment, but in general, the description of linguistic variation is supressed to the minimum in “Ural’skie jazyki”.


“Ural’skie jazyki” is oriented to material rather than theory. In this respect, it is as a linguistic handbook should be. Many authors may be unaware of recent versions of, say, autosegmental phonology or government-and-binding syntax, but then, they are free to exploit any grammatical means they find suitable for describing the relevant linguistic structures irrespective of fashionable but often ill-founded trends. On the other hand, many if not all of the authors are keenly aware of the existence of the so-called morphemic model, which is a pity, since this model is bound to obscure the often complex morphological and morphophonological processes found in most Uralic languages. A realisational approach to word-structure would have allowed many things to be dealt with both more iconically and more compactly. By contrast, the phonemic theory, despite its temporary neglect in recent literature, could have been exploited more correctly and effectively. Concrete instances will be mentioned later.

The 32 chapters were written by 17 named authors plus a workgroup. At the time of writing, all but two were carriers of a Soviet passport. The two are the eminent Hungarian scholars Péter Hajdú (Uralic languages) and László Honti (Ob-Ugrian languages, Khanty), who, however, only contributed translated versions of earlier publications. Among the others, Arvo Laanest (Finnic languages, Votian, Ingrian), Tiit-Rein Viitso (Livonian), Ago Künnap (Kamas, “Koibal”), and the workgroup (Estonian) are Estonians. The remaining twelve scholars from the Russian Federation represent eight native linguists, i.e. V. D. Rjagoev (Karelian), M. I. Zajceva (Vepsian), A. P. Feoktistov (Moksha; also authors Mordvin languages, Erzya), E. I. Kovedjaeva (Mari; also authors “Hill Mari”), V. K. Kel’makov (Udmurt), R. M. Batalova (Permyak; also authors Permian languages, Komi), K. E. Majtinskaja (Hungarian; also authors Finno-Ugrian languages), E. I. Rombandeeva (Mansi), and four non-native scholars, i.e. Ju. S. Eliseev (Finnish), G. M. Kert (Sámi), N. M. Tereshchenko (Samoyed languages, Nenets, Enets, Nganasan), and Eugene Helimski (Selkup, Mator). Furthermore, two subtitles without a chapter, viz “Volgaic languages” and “Ugrian languages”, were added to the outline of the book. Incidentally, fuller and better edited chapters by Laanest (on Ingrian), Feoktistov, Kovedjaeva, Majtinskaja (on Finno-Ugrian), Rombandeeva, Kert, and Tereshchenko can already be found in Lytkin & Majtinskaja (1966).

The range of the chapters is wide enough to make further general statements hardly possible, except one thing: the number of typing and printing errors surpasses all books I have seen in my life. What is worse, the misprints abound in language material, which makes them fatal for a non-specialist. This means that I, with a special training mainly in Samoyed and Finnic, have failed to notice many of them, but on the basis of what I have noticed I would have rather hoped that the book had been left unpublished in this condition.

It should be emphasized that many errors have appeared during the editing process, so that the authors are not responsible for them. For instance, passages with translations of language names, apparently added to the manuscript later, contain errors of a kind that the respective authors could not have committed. For example, the Finnish equivalents for Ingrian and Permian are cited as *inkerois and *permiläi instead of the correct inkeroinen and permiläiset.

Notwithstanding the poor proof-reading, “Ural’skie jazyki” is generally readable, and while the text is mostly quite standard, a few novel contributions are also included. In what follows, I deal with all (groups of) languages in an order that I prefer myself rather than that of the book. I try to concentrate on a few selected problematic or controversial issues, and aim at dissolving certain myths that prevail in the standard literature, including the volume under review.

The existence of two introductory chapters, labelled “the Uralic languages” and “the Finno-Ugrian languages”, illustrates one af the most persistent myths in the field. In practically all textbooks, the standard claim is that the Uralic family is a union of two very distantly related groups of languages, called Finno-Ugrian and Samoyed. This claim is false. While it is true that the Samoyed branch is very independent, there is little or no evidence for the rest of Uralic as a unit in its own right. The standard view is originally based on the classification presented by Otto Donner, the founder of the Finno-Ugrian Society; he correctly dissolved the so-called “Ural-Altaic” unity established by the Finnish pioneer scholar M. Alexander Castrén, but by error, he went too far and also excluded the Samoyed languages from the family. This error was later remedied, but the Samoyed branch remained as “the first branch to have left the Uralic unity”.

The standard classification continues to split the “main” branch, i.e. Finno-Ugrian, into Finno-Permian and Ugrian, Finno-Permian further into Finno-Volgaic and Permian, Finno-Volgaic into “Early Proto-Finnic” (a most unhappy appellation) and Volgaic (Mordvin and Mari), and finally “Early Proto-Finnic” into Sámi and (“Late Proto”-)Finnic. This practice is also unfounded, and originally based on a nationalistic Finnish view which wanted to see the Finnish language literally as the highest sprig of the sacred family tree.

Rather than a result of recursive bifurcations with the Finnish language as its most highly developed representative, the Uralic family is a chain of seven, or more likely, nine, separate branches, which derive directly from Common Uralic. The hesitation about the exact number of primary branches is due to the fact that while Sámi, Finnic, Mordvin, Mari, Permian, and Samoyed are beyond doubt as historical linguistic entities, with a host of common innovations in each, the same is far from true about the alleged Ugrian branch, consisting of Hungarian, Mansi, and Khanty.

Consequently, the differences between the two introductory chapters by Hajdú and Majtinskaja derive from personal views of the authors, not from observed differences in the subject matter. In other words, the chapter for “the Finno-Ugrian languages” does not contain any general statements that would not encompass “the Uralic languages” in the alleged wider sense. Incidentally, most details of one chapter recur in the other. Essentially, both chapters deal with differences rather than similarities across the family, which is expected in view of the vast diversity one encounters there.

Both of the introductory chapters present the classification of the Uralic languages in the form of a family tree. On p. 8, we find one which corresponds exactly to the binary model described and criticized above, the only peculiarity being the three-fold rendering of “Proto” in Russian, which, to my knowledge, is not standard. On p. 21, the picture is almost identical, except that a dashed line leads from Proto-Volgaic to Sámi, which is supposed to reflect the obsolete view of the “non-Finno-Ugrian” origin of Sámi speakers, but because of other, more obvious errors, happens to confuse the relevant corner of the picture. Both trees are further accompanied by standard ad hoc datings of the alleged intermediate proto-languages.

Hajdú’s contribution is an abbreviated version of relevant chapters in Hajdú & Domokos (1978) and Hajdú & Domokos (1987). It seems, however, that the editors of “Ural’skie jazyki” subsequently rewrote passages of the text, apparently in an attempt to unify the style. The unfortunate result was that inconsistencies increased, and a number of errors appeared. For instance, the Nenets material is transcribed vaguely, and differently from the elaborate system of Hajdú (1968) himself, and on p. 15, most of the predicative forms of the Nenets word for ‘child’ have turned out wrong. On the other hand, as much as Hajdú is a specialist of Samoyed, his account of the predestinative declension is not correct: contrary to the scheme on p. 15, there is no 3rd person possessive suffix after the stem, but the suffix in question is not only functionally, but also materially distinct (Janhunen 1989). Further, the table of the grammatical features of the Uralic languages found on p. 19, earlier published in Hungarian (Hajdú & Domokos 1978, 127) and German (Hajdú & Domokos 1987, 396, illustrates the risks of concentrating too much data in a small space. Phonological features, in particular, are of relative nature and cannot therefore be straightforwardly isolated from the system of oppositions where they function. For example, it appears from the table that there are voiced plosives in Mari but not in Nenets, while these languages are actually quite similar in having a series of lax obstruents which can be manifested as voiced stops or fricatives, depending on a number of conditioning factors. Furthermore, the “major dialects” of most languages (read: groups of languages) in the table are phonologically very different from each other, for instance, the above statement about obstruents is valid for Tundra Nenets only. By contrast, a + sign for Umlaut/Ablaut in Nenets refers to Forest Nenets in the first place. It is also unclear what is meant by palatalization: both Finnish and Estonian are marked with a ± sign for it, while most forms of Finnish, including the standard language, are void of distinctly palatal consonants of any sort, and only dialects show a prosodic palatalization similar to the common Estonian system. Among grammatical features, a few concepts seem highly controversial. It is not clear what is the use of the notion “multifunctional nominative”, supposedly present in Ugrian languages, where it appears to be a logical consequence of the lack of the genitive case, and therefore a redundant feature, but also in Nenets, which is rather an error based on a misinterpretation of phonologically conditioned sandhi variants of normal genitive forms. Contrary to the table, ‘passive’, admittedly a concept subject to various definitions, is similar in Finnish and Estonian, while the Sámi “passive” is a derivational category, which can also be identified in languages without passive formations according to the table. Another obscure concept, viz ‘conversion’, seems to refer to the existence of nominal conjugation in Mordvin and Samoyed, which is not conversion, and the function of the ± sign in the columns of Ugrian languages remains obscure. Finally, the distribution of the infinitive in -ni is confused through a simple misprint: such an infinitive exists in Hungarian, but not in Mordvin.

Another prevailing myth concerns the essence of the Sámi linguistic unity. In the spirit of the old school, with its colonialist perspective, the headline of the chapter (p. 134) refers to a single “Sámi language”. By all criteria, however, the Sámi branch consists of several separate languages, and according to the most consistent count, they number no less than eleven. For instance, within the territory of the Republic of Finland, there exist absolute language boundaries between all three extant Sámi languages, and the fourth extinct one is known to have been equally distinct. It is true that the languages from Pite to North Sámi on the one hand, and those from Skolt to Kildin Sámi on the other, form two chains where there is some mutual intellibility between the neighbouring languages. Nonetheless, all Sámi languages are quite clearly demarcated, and each of them exhibits wide internal variation. North Sámi, for example, is divided into three markedly different dialect groups, viz Torne Sámi, Finnmark Sámi, and Sea Sámi. All in all, the linguistic differentiation of the Sámi branch equals that of Finnic, or any other branch of Uralic, for that matter. The four extant Sámi languages of the Russian Federation are those of the eastern language chain, i.e. Skolt Sámi (Notozero Sámi as its main dialect), Akkala (Babino) Sámi, Kildin Sámi, and Ter Sámi; the Sea Sámi dialect of North Sámi was formerly spoken within the present Russian boundaries, too. The chapter presents material on Kildin Sámi only, closely following the lines of an earlier monograph by the same author (Kert 1971). However, because of the paucity of data generally available, the chapter would be extremely welcome, if it did not suffer from certain shortcomings in the phonological section, further reflected in the treatment of morphology. Kert’s transcription system is very confusing, so that it does not reveal the phonological and morphophonological properties of the language too well. Furthermore, only a few scattered examples serve to clarify the morphophonology of Kildin Sámi, one of the most complex languages on earth in this respect.

Seven Finnic languages are enumerated in their collective chapter by Arvo Laanest, with a further note that some Finnish scholars view Ludian, officially a dialect of Karelian, as a dialect group transitional from Karelian to Vepsian. A more ample characterization of the position of many non-Soviet scholars would be that Ludian is a language that has developed from Old Vepsian under heavy Old Karelian influence. Olonetsian, by contrast, can be seen as a descendant of Old Karelian, but with a notable Old Vepsian substrate. Also for modern sociolinguistic reasons, Ludian and Olonetsian are better treated as languages separate from Karelian proper. It is no doubt true that there is more diversity within the two major Finnic languages, Estonian and Finnish, but in a synchronic framework, their single-language status is not in question. The seven individual chapters for the Finnic languages follow in an inexplicably random order. A more well-founded order from west to east is used here.

Livonian is in most respects the most aberrant of the Finnic languages. Recent studies of it are not widely available, so that the chapter by Tiit-Rein Viitso is most welcome. Viitso uses standard orthography, which is not strictly phonemic, but seems to capture the essential phonological contrasts better than the traditional impractical, pseudo-phonetic transcription, which is also briefly exemplified and contrasted to the orthography in the chapter. Livonian can be characterized as the most prosodic of all Uralic languages, in that it uses quantitative and tonal oppositions which are not easily dealt with by segmental phonemics. Viitso’s summary is solid and compact, but a fuller analysis of the topic would be much needed. Its reflections on morphology, in the form of complex morphophonological alternations, would also require a more comprehensive treatment.

The chapter on Estonian is written by a group of Estonian specialists in a more modern spirit than other chapters. For instance, long vowels and geminate consonants are given a biphonemic interpretation, the so-called over-long (or third) degree of quantity is called a “syllabic accent”, and the complex nature of word-stress is highlighted. The authors leave open the question whether the vowel symbolized by ő is to be regarded as a mid or high vowel: from the systematic point of view, the latter solution seems superior, and to a non-native observer, it seems unnecessary to emphasize the great difference between Estonian ő and Russian y. The standard orthography, which is only partly phonemic, is used thoughout the chapter. However, its compatibility with the phonological structure is described quite exhaustively. The traditional phonetic transcription, awkward and uninviting as it is, appears in the section of phonology as well as that of dialects. Notably, the Finno-Ugrian transcription system itself is explained nowhere in the book.

Votian, while a moribund language which never had a literary use, is historically a very central language in the Finnic branch. The author of the chapter, Arvo Laanest, repeats the common view of Votian as part of the southern Finnic subgroup, but makes a proper reservation that Votian also possesses common features with northern Finnic. In a non-binary but contact-sensitive taxonomical scheme, Votian, as a conglomeration of both original and shared features, is best regarded as a descendant of one of the basic Finnic dialects. Despite the potential interest in Votian grammar, the chapter is very short and thus fails to provide anything but the sketchiest description of the language. For instance, a much fuller statement on the role of recent Russian influence would have been required in the phonological section to make Votian phonology at all clear to the reader.

The Russian specialist Ju. S. Eliseev provides a refreshingly non-clichéd treatment of Finnish. As a start, we find that contrary to the dogma of the Finnish school of historical linguists, Mordvin rather than Sámi is the branch most closely related to Finnic, a statement which in its controversiality lends support to the notion of non-binary classification of the Uralic language family. That there are dogmas in Russian scholarship as well is exemplified by a claim that Kalevala is a “popular epic”, while in fact, it belongs to belles-lettres. As a reflection of the narrow views of the Russian phonological school, Finnish double vowels and consonants, best interpreted as sequences of identical segments, are denounced as monophonemic long vowels and geminate consonants. Occasionally Finnish and Russian dogmas seem to coincide, e.g. in the case of the basic paradigm of consonant phonemes: f should be regarded as the fourteenth member of this paradigm since it is pronouncable and also pronounced by Finnish speakers while b, g, and z are restricted to the orthographic representation rather than the pronunciation of Finnish. Another inaccurate statement concerns the phonological interpretation of the letter d, since it doesn’t correspond to a voiced stop either phonetically or by virtue of its status in the phonological system: it should rather be called a tap or a flap sound. Further, a concept such as “final aspiration”, long since obsolete in most literature, is used to refer to the well-known process of consonant doubling in certain lexical and morphological environments; of these environments, a very superficial description is given in the chapter. Within morphology, it is correctly stated that a special accusative case form appears in the declension of a few pronouns only: in the basic nominal declension, no accusative case exists, even though many grammars obscure the terminology by introducing a concept of “syntactic” accusative; Finnish object rules are quite clearly presented later in the chapter without this confusing notion. As in most if not all Finnish grammars, four tenses are listed, while there are no more than two morphological tenses, present and past. Furthermore, calling the latter “imperfect”, instead of the more correct ‘preterite’, is simply a standardized error in Finnish grammatical tradition. The two-page introduction to Finnish dialects at the very end of the chapter suffers, like the whole chapter and the whole volume, from numerous misprints; as usual in Soviet linguistics, the pre-war distribution of the south-eastern dialects of Finnish is ignored.

It is a pity that only eight pages have been assigned to the Ingrian language, which is not to be confused with Finnish dialects spoken in the historical province of Ingria around St. Petersburg. Because of the lack of space, the author Arvo Laanest often refers to Finnish, which is not very instructive for readers less acquainted with these languages. The most important features that highlight the independent position of Ingrian are nevertheless well-presented, but the chapter in the predecessor of “Ural’skie jazyki” by the same author (Laanest 1966) was much fuller.

The chapter for Karelian is, in line with the specialization of its author V. D. Rjagoev, based on the Tikhvin dialect, a diaspora dialect of Karelian proper. Not much is learnt about Olonetsian and Ludian in a brief characterization of the Karelian “dialects”. The scope of consonant gradation in Ludian is said to be restricted to stops, which is a bit unorthodox, since the statement only refers to the weak grade voiceless stops which alternate with strong grade geminates; voiced stops exhibit no gradation in Ludian.

Vepsian seems to be regarded as a relatively important language because it is allotted double space compared to the briefest chapters, on Votian and Ingrian. The chapter by M. I. Zajceva is relatively thorough, with inflectional paradigms of verbs though not of nouns, and with a lengthy albeit far from complete analysis of morphophonology. The transcription system is mostly phonemic, except that the influence of Russian grammatical tradition shows in distinguishing a class of “central” vowel, which are allophonic rather than phonemic, and occasional marking of non-syllabicity of latter parts of diphthongs.

“Volgaic languages”, i.e. a conglomeration of Mordvin and Mari, is only present in the form of a subtitle. In the competing handbook mentioned above, there is a short chapter devoted to the topic, with the inevitable conclusion that there is no justification for such a subgroup (Bereczki 1988, 314–315).

The Mordvin languages have a short common treatment by A. P. Feoktistov, who authors both individual chapters on Erzya and Moksha. Since they are rather thoroughly studied languages, the chapters provide but highly abbreviated versions of major grammars. It appears that there are still open questions in the phonology of Moksha vowels: the status of the reduced vowel as well as the position of the allophones included in the table on p. 180 is not made clear. In the examples that are supposed to be written phonemically, labial and velar consonants are also marked as palatalized, though palatality is only an allophonic feature for them. There are examples of possessive forms in various case forms for Moksha, but not for Erzya, which is a pity because important differences between the languages appear there, and the complexities of the issue may well have wider significance.

The question of the status of Mari either as a single language or two has received a solution that might be called schizophrenic. There is supposed to be one language, but there are two chapters, the first of which is simply about Mari, and the second, curiously, about “Hill Mari variant of Literary Mari”, both authored by E. I. Kovedjaeva. More accurate headings would undoubtedly be “Literary (variant of) Meadow Mari” and “Literary (variant of) Hill Mari”, since it is the two existing literary languages that are dealt with separately. Literary Meadow Mari serves for all speakers of Eastern Mari, i.e. Meadow Mari and the more easterly dialects, and Literary Hill Mari is confined to the respective dialect, while the other Western Mari dialect, North-western Mari, has no functioning literary norm. The main objection about the contents of the chapters is their silence about one of the most striking and theoretically important peculiarities of Mari, its variable order of case, possessive, and number markers. Thus, in the grammatical oblique cases (accusative, genitive, and dative), possessive markers precede both number and case markers, while in the local cases, possessive markers appear last. Incidentally, in the chapters for the Permian languages, a similar situation is mentioned.

Despite the relative closeness of the Permian languages, they are dealt with quite briefly in their introductory chapter by R. M. Batalova. The individual chapters on Udmurt (by V. K. Kel’makov), Permyak, and Komi (both by Batalova) are, however, written consistently enough to make comparisons unproblematic. For the Komi languages, the unhappy coinages of the Soviet era, “Komi-Permyak” and “Komi-Zyryan”, are used in the book, as they are the official terms. Nevertheless, they are not recommended for further use; if Komi is adopted as the common name of the two languages, then “Komi-Zyryan” can be referred to as Komi proper, while “Komi-Permyak” is simply Permyak.

While there are six obvious and unquestionable branches in the Uralic family, the alleged seventh, Ugrian, is more problematic. Notably, the book does not contain a chapter called “the Ugrian languages”, only such a subtitle. This seems to be the right solution in view of the evidence, or rather lack of it, that there is to support the Ugrian hypothesis. Basically, the same arguments that appear in connection with the Ob-Ugrian hypothesis, below, are valid here.

The Hungarian chapter by the late K. E. Majtinskaja is a concise summary of standard grammar books. A strange instance of apparent Russian influence appears in the table of consonants on p. 258, where cs is regarded as palatal while its voiced counterpart dzs as well as s and zs are thought to be postalveolar. As usual in standard morphological descriptions of Hungarian, no explicit mention is made of the fact that for consonant stems, two separate suffixes appear in plural possessive forms, e.g. csónak ‘boat’ : poss.sg3sg csónakja : poss.pl3sg csónakjai, while only one suffix is attached to vowel stems, e.g. hajó ‘ship’ : poss.sg3sg hajója : poss.pl3sg hajói). Such cases, which run counter the traditional morphemic model, are simply forced into the mould, for instance, by regarding the combination -jai as a single suffix. Another shortcoming of the morphemic model manifests itself in the complex system of connecting vowels in Hungarian: the tables of verbal suffixes on pp. 268–270 are practically useless because it is not shown how the suffix variants actually combine with the stems.

There was no chapter for the Ugrian languages, but there is one for Ob-Ugrian. The author of this chapter, László Honti, uses more space to illustrate the differences rather than the common features of Mansi and Khanty, and correctly emphasizes that the bulk of those common features are either retentions from Proto-Uralic or relatively late areal innovations. However, he fails to draw the apparently inevitable conclusion that no Ob-Ugrian unity needs to be assumed at all.

The chapter on Northern Mansi is essentially a summary of the earlier grammar by the author (Rombandeeva 1973). In phonology, it is remarkable that a long mid central vowel, generally ignored by other scholars, is consistently recognized, a fact that points to the need for a fresh look into the Mansi vowel system in general. The synthetic forms of future and conditional in the tables of verbal inflection on pp. 294–295 derive from imitating of the traditional format of Russian grammar, and are thus unnecessary in a description of the Mansi grammar.

The chapter for Eastern, more specifically Surgut, Khanty by László Honti is a revised translation of his earlier account in German (Honti 1988). Some changes were made to adapt the article to the general format of the book, but many features of the original article also remained. The Russian version offers no new insights, rather it is slightly abbreviated from the original. In particular, important paradigm tables were removed, which limits the usefulness of the chapter. An interesting detail is that before the publication of “Ural’skie jazyki”, it had already been discovered by the Hungarian scholar Márta Csepregi that two local dialects dealt with by Honti, of Tremyugan and Tromagan, actually refer to variant names of one and the same river, and two stages of development of the same dialect (Csepregi 1993), but no mention of this discovery is made in the chapter.

Four of the chapters on the Samoyed languages, including the introduction and those on northern Samoyed languages, i.e. Nganasan, Forest Enets, and Tundra Nenets, are written by the late N. M. Tereshchenko, who played a key role in the development of Samoyed studies for decades. Since these chapters were written a long time ago, they were partly outdated even before they were published. In particular, the descriptions of phonology cannot be regarded as entirely reliable, since they draw heavily on certain unsuitable traditions of Russian grammar, e.g. giving mid vowels a phonemic status in Enets and Nenets.

Eugene Helimski’s chapter on Northern Selkup is a synopsis of the excellent grammar book by Kuznecova & Xelimskij & Grushkina (1980). The main problems in the synchronic analysis of Northern Selkup appear in phonology, especially in the alleged tense/lax (naprjazhennyj/nenaprjazhennyj) opposition of vowels, which in the first place differentiates two series of illabial front vowels, but is also used as an ad hoc device to distinguish the labialized long low vowel. In both cases, it seems that no true phonological contrast is present, which leaves us with a highly symmetrical vowel system with ten qualitative units, viz five cardinal and five non-cardinal vowels, all of which appear both as short and long.

Ago Künnap, the only modern and the last linguist who did field-work on Kamas, has written chapters on Kamas and “the Koibal language”. The former is a larger one, while the latter is just a brief mention pointing out that Koibal is very close to Kamas. As acknowledged by the author, there are good reasons for treating Koibal as a a co-dialect of a language, which can simply be named after Kamas. The main difficulty with Kamas studies lies in the fixation of true phonological oppositions: the last speakers of Kamas were, expectedly, multilingual, and the phonological systems of neighbouring languages seem to be responsible for the striking variability of Kamas pronunciation. It is likely that the basic Kamas system can nevertheless be discerned, but Künnap decided not to strive for it.

The first grammatical treatment of Mator, by Eugene Helimski, hopefully ends the tradition of referring to five or six “Sayan Samoyed languages”, usually presented as an alphabetic list. Like Kamas and Koibal on the one hand, Mator, Taigi, and Karagas (and the practically non-attested “Soyot”) on the other, are dialects of a single language, which Helimski calls Mator-Taigi-Karagas to underline their unity rather than their separate status. Furthermore, the two languages are not particularly close to each other, and there is little basis for assuming a “Sayan Samoyed” proto-language. Mator being a language known through old word-lists, Helimski has clearly decided to resort to the attested material only, which may be wise for the present purpose, though a more comprehensive treatment would undoubtedly benefit from drawing upon comparisons with other Samoyed languages, too.

Largely a collection of standard pieces of knowledge, “Ural’skie jazyki” contains a few extremely valuable contributions, in particular the chapters on Livonian, Kamas, and Mator. It is a pity that the merits of the book are overshadowed by the disastrous proof-reading. A more general objection could be that the size of the volume is in disproportion with the diversity of the material. There simply is no way to organize even the most basic data of the Uralic languages in four hundred pages. What we have, then, is a book from which a lot can be learnt, but many more things are left out, and every detail there must be treated with strictest caution.


Bereczki, Gábor 1988. Geschichte der wolgafinnischen Sprachen. In: Sinor 1988, 314–350.

Csepregi, Márta 1993. Lintunen ja hänen sisarensa. Hantilainen satu Tromagan-joen varrelta. In: Kulonen 1993, 49–52.

Kulonen, Ulla-Maija (hrsg.) 1993. Festschrift für Raija Bartens. Mémoires de la Société Finno-Ougrienne 215; Helsinki.

Hajdú Péter 1968. Chrestomathia Samoiedica. Budapest: Tankönyvkiadó.

Hajdú Péter & Domokos Péter 1978. Uráli nyelvrokonaink. Budapest: Tankönyvkiadó.

Hajdú, Péter & Domokos, Péter 1987. Die uralischen Sprachen und Literaturen. Budapest: Akadémiai Kiadó.

Honti, László 1988. Die ob-ugrischen Sprachen. II. Die ostjakische Sprache. In: Sinor 1988, 172–196.

Janhunen, Juha 1989. Samojedin predestinatiivisen deklinaation alkuperästä. Journal de la Société Finno-Ougrienne 82, 298–301.

Kert, G. M. 1971. Saamskij jazyk. Leningrad: Nauka.

Korhonen, Mikko 1986. Finno-Ugrian language studies in Finland 1828–1918. The history of learning and science in Finland 1828–1918; Helsinki: Societas scientiarum Fennica.

Kuznecova, A. I. & Xelimskij, E. A. & Grushkina, E. V. 1980. Ocherki po sel’kupskomu jazyku. Tazovskij dialekt. Tom I. Moskva: Izdatel’stvo Moskovskogo universiteta.

Laanest, A. 1966. Izhorskij jazyk. In: Lytkin & Majtinskaja (ed.) 1966, 102–117.

Lytkin, V. I. & Majtinskaja, K. M. (ed.) 1966. Jazyki narodov SSSR. Tom III: Finno-ugorskie i samodijskie jazyki. Moskva: Nauka.

Rombandeeva, E. I. 1973. Mansijskij (vogul’skij) jazyk. Moskva: Nauka.

Ruhlen, Merritt 1987. A guide to the world’s languages. Vol. 1: Classification. London: Edward Arnold / Stanford: University Press.

Salminen, Tapani 1990. Uralilaisten kielten käsikirja vihdoin ilmestynyt. [Review of Sinor 1988.] Virittäjä 94, 493–502.

Sinor, Denis (ed.) 1988. The Uralic languages: description, history and foreign influences. Handbuch der Orientalistik 8: Handbook of Uralic studies 1; Leiden: Brill.

Wickman, Bo 1988. The history of Uralic linguistics. In: Sinor 1988, 793–818.