Ural-Altaic Languages


The many Ural-Altaic languages--constituting the Uralic and the Altaic languages--extend from Scandinavia, Hungary, and the Balkans in the west, to the easternmost reaches of the Amur and the island of Sakhalin, and from the Arctic Ocean to central Asia. According to some investigators, Japanese and Korean should also be considered Altaic languages (see JAPANESE LANGUAGE; KOREAN LANGUAGE).

Linguistic Features

All the Ural-Altaic languages share certain characteristics of syntax, morphology, and phonology. The languages use constructions of the type the-by-me-hunted bear rather than "the bear that I hunted," and a-singing I went rather than "I sang as I went." There are few if any conjunctions. Suffixation is the typical grammatical process--that is, meaningful elements are appended to stems, as in house-my, "my house," go-(past)-I, "I went," house-from, "from the house," go-in-while, "while (in the act of) going," and house-(plural)-my-from, "from my houses."

A great many Ural-Altaic languages require vowel harmony; the vowels that occur together in a given word must be of the same type. Thus poly, "dust," is a possible word in Finnish because o any y are both mid vowels and hence belong to the same phonetic class; likewise polku, "path," is possible because o and u are both vowels. Words such as polu or poly are not possible, because o and u, or o and y, are too dissimilar. Stress generally falls on the first or last syllable; it does not move about, as in the English series family, familiar, familiarity.

Typically, the Ural-Altaic languages have no verb for "to have." Possession is expressed by constructions such as the Hungarian nekem van, "to-me there-is." Most of the languages do not express gender, do not have agreement between parts of speech (as in French les bonnes filles, "the good girls"), and do not permit consonant clusters, such as pr-, spr-, -st, or -rst, at the beginning or end of words.

Family Status

According to the standards set by linguists, languages that make up a family must show productive-predictive correspondences. The shape of a given word in one language should be predictable from the shape of the corresponding word, or cognate, in another language. Thus Hungarian -d at the end of stems, as in ad, "he gives," is known to correspond to the Finnish consonant sequence -nt- in the interior of words, as in Finnish anta-, "give."

All of the Uralic languages have been shown to be related--the vocabulary and grammar of each member language can be examined in the light of correspondences such as that which obtains between Hungarian -d and Finnish -nt-. But Altaic is not a language family in the same sense that Uralic is, for laws of correspondence such as those available for Uralic have yet to be discovered in Altaic.

Altaic does have three branches, however--Turkic, Mongolian, and Manchu-Tungus--each of which forms a subfamily. Turkic and Mongolian on the one hand, and, to a lesser extent, Mongolian and Manchu-Tungus on the other, exhibit many striking resemblances. But the shared features may reflect only borrowing, and not a common origin.

Uralic Languages

The Uralic languages are traditionally divided into two major branches, Finno-Ugric and Samoyed. Finno-Ugric in turn contains two subgroups: Finnic and Ugric. The former is divided into the Baltic-Finnic, Volga-Finnic, and Permian languages; the latter comprises Hungarian and the Ob-Ugric languages.


Finnish, with 5 million speakers, and Estonian, with 1 million, are the best known of the Baltic-Finnic languages. Others are Karelian, spoken by 175,000 people in northwestern Russia and eastern Finland; Veps, spoken by 8,000 people between the Dnepr and the Volga; Votian, spoken by 700 people of the Udmurt Autonomous Republic of the former USSR; and Livonian, spoken by 500 people in the Livonia district of Latvia. Lapp is similar in structure to Finnish, but the various Lapp dialects--spoken by 40,000 people spread over Norway, Sweden, Finland, and the Kola Peninsula of Russia--diverge greatly from each other in phonology and even to some extent in grammar.

Finnish is famous above all for its many cases, 12 of which are productive--that is, any Finnish noun can be followed by one of the 12 case suffixes. Another pervasive feature of the language is consonant gradation, such as the t/d alternation found in the declination of the Finnish word for "hundred": nominative sata, genitive sadan, ablative sadalta, partitive sataa, and so on. Finnish is also distinctive in having a verb that, translated roughly, means "not to." Compare ulvon, "I howl," with en ulvo, "I do not howl," and ulvo, "you howl," with et ulvo, "you do not howl," where en and et mean, respectively, "(I) do not" and "(you) do not."

Volga-Finnic and Permian

Mordvinian, spoken by 1,262,000 people along the middle Volga, and Cheremis, spoken by 600,000 people in the district where the Kama joins the Volga, constitute the Volga-Finnic language group. Both of them, but especially Mordvinian, are close to Finnish in grammar and vocabulary. Less like Finnish are the Permian languages--Zyrien with its 628,000 speakers, and Votyak with its 704,000 in northeastern European Russia. All of the Volga-Finnic and Permian languages have a negative verb and a large number of cases.


The Finnic languages are more or less geographically contiguous, but the Ugric languages lie at opposite ends of the Finno-Ugric area--Hungarian occupying the extreme west, and the Ob-Ugric languages, Vogul and Ostyak, occupying the extreme east. Hungarian has 13 million speakers--the largest number of any Uralic language--who live in the Danube Basin and adjacent areas. Vogul's 8,000 speakers and Ostyak's 21,000 live east of the Urals, in the Ob Valley.

One of the most striking Ugric linguistic features is the so-called objective conjugation. In Hungarian, for instance, adok means "I give," and adom means "I give it" or "I give them." Thus the object of the verb--"it" or "them"--is incorporated in the verb form and does not need to be expressed separately. Vogul and Ostyak are still more precise. In these languages the objective conjugation has three distinct forms, to indicate whether the object is "it," "them" (plural), or "the two things" (dual). Furthermore, Vogul and Ostyak can also express the subject in the singular, plural, or dual.

Hungarian has more productive cases--upward of 20--than even Finnish has. Vogul and Ostyak, however, have only from four to seven cases, depending on dialect. The Ugric languages have no consonant gradation.


The Samoyed languages are the easternmost representatives of Uralic. Presumably they were the first to separate, as a group, from the original, proto-Uralic language. They are spoken in the northeastern corner of Europe, near Zyrian, and in north-central Siberia.

Yurak, with 28,000 speakers, Tavgi, with 1,000, and Yenisei, with 500, form a North Samoyed group, and they can be distinguished from the South Samoyed language, Selkup, with 4,000 speakers. Other Samoyed languages, now extinct, are known only from 18th- and 19th-century records.

Loan Words and Early Records

In the course of their histories, the individual Uralic languages have come into contact with a great many languages from other families--Turkic, Germanic, Baltic (an earlier form of Latvian and Lithuanian), and Slavic. Finnish kuningas, "king," is an early loan from a Germanic language, hence its resemblance to English king and German Konig. Finnish vapaa, "free," was borrowed from a Slavic language--compare the Slavic root svobod-. The same Slavic root found its way, independently, into Hungarian, as evidenced by the word szabad.

The oldest significant text written in a Uralic language is a funeral sermon in Hungarian from about 1195. Finnish and Estonian texts survive from the Protestant Reformation, which swept over Scandinavia and much of the Baltic in the 16th century; the reformer of the Finns, Michael Agricola (1512-57), also translated the Bible into Finnish. Zyrien was recorded in the 15th century by Saint Stephen of Perm, apostle of the Zyriens, who fashioned a special alphabet for the language.

Altaic Languages

The Altaic languages are spread over an area that is even larger than that covered by Uralic. Of the three branches of Altaic, Turkic ranges from Anatolia to the Volga basin and central Asia; Mongolian extends from China and Mongolia as far west as the lower Volga and Afghanistan; and Manchu-Tungus occupies the northern coast of northeastern Siberia, and runs as far south as the Amur and as far west as the Yenisei, which divides Siberia into its eastern and western halves.


Written evidence of the Turkic languages begins with the Orkhon inscriptions of the 8th century AD, found near the river Selenga in Mongolia, and continues wherever and whenever a Turkic population came into contact with one of the higher religions, such as Islam, Judaism, Christianity, or Manichaeism.

Linguistically, the Turkic languages form a tightly knit group. Knowledge of one Turkic language usually enables an investigator to analyze words and simple sentences in any other Turkic language except Chuvash. To explain this, it is hypothesized that an original, proto-Turkic language split into two branches: West Turkic and East Turkic. West Turkic went its own way, both phonetically and in terms of contact with other languages, and eventually became Chuvash, now spoken by 1,700,000 people living in the Volga Basin in the Chuvash Autonomous Republic of the former USSR.

The early speakers of East Turkic must have remained together for a longer time and split up only comparatively recently into the many present-day languages. Still, the East Turkic languages are usually classed into five subdivisions: Oghuz, mainly represented by Turkish, the language of Turkey; Kipchak, which has over a dozen representative languages, including Kazan Tatar, Kazakh, Kirghiz, and Bashkir; Sayan Turkic, represented by Tuvan, Altai, Shor, and several other languages; Turki, represented primarily by Uigur and Uzbek; and Yakut, which comprises Yakut Proper, Khakas, and Dolgan.

Turkish, like Finnish, has vowel harmony. It also uses cases and possessive suffixes, which can combine as in ev-ler-im-in, "of my houses," made up of the word elements found in ev-ler, "houses," ev-im, "my house," and ev-in, "of the house." Such agglutination is also characteristic of Turkish verbs: compare gel-mek, "to come," gel-ir-im, "I come," gel-iyor-um, "I am coming," gel-di-m, "I came," gel-me-mek, "not to come," and gel-me-d-in, "I did not come."


Despite their considerable geographical dispersion, the present-day Mongolian languages or dialects are all closely related and all descend from a common proto-Mongolian parent language. The vigorous but short-lived military conquests of Genghis Khan in the 13th century brought the Mongols well into Europe, and to this day traces of Mongolian may be discovered in a few provinces of Afghanistan, and over 100,000 Kalmyk-Mongols live in the Kalmyk republic of the former USSR.

Khalkha is the language of the Mongols of Mongolia, with its capital at Ulan Bator. Buryat (Buriat) is spoken in the Buryat Autonomous Republic of Russia. Other Mongolian languages include Dagur, with 24,000 speakers in northwestern Manchuria and the Chinese province of Xinjiang (Sinkiang); Monguor, in Qinghai (Tsinghai) province; Kalmyk; Oirat; Moghol; Santa; Paongan; and Yellow Uigur.

The grammatical processes encountered in the Mongolian languages are similar to those of Turkic. The Mongolian languages have many cases and in that respect they resemble some of the Uralic representatives, notably Finnish and Hungarian.


Just as the Turkic languages can be thought of as the western wing of Altaic, the Manchu-Tungus--also known simply as the Tungus--languages constitute the eastern wing. Most of these languages have been known only since the 19th century, but two of them, Manchu and Jurchen, are preserved in historical records that go back much further. Manchu, now spoken by only a few thousand people, was the original language of the tribe of horsemen that became the Qing (Ching) dynasty and occupied the Chinese throne from 1644 to 1912. Similarly, Jurchen, now extinct, was the language of the tribes that became the Jin (Chin) dynasty, ruling from 1115 to 1234.

The Manchu-Tungus languages fall into two groups. South Tungus includes Manchu, Goldi, Olcha, Orok, Udihe, and Orochon. The North Tungus languages are Eveneki, or Tungus Proper, and Even, also known as Lamut.


The grammatical structures of Uralic and Altaic are quite similar, and about 70 words in each group--such as the Finnish kaly, "sister-in-law," and Uigur kalin, "bride" and "daughter-in-law"--appear to be cognates. But the correspondences between the two groups of languages are unsystematic; they could be the result of borrowing or chance. No precise predictive-productive sound laws, for instance, have been established. Alternatively, it is argued that the parallels between Uralic and Altaic are slight because the two groups split apart a long time ago.

In addition to the Ural-Altaic hypothesis, which is that Uralic and Altaic form a superfamily of languages, there is also an Indo-Uralic hypothesis, in which Uralic is linked with the INDO-EUROPEAN LANGUAGES; a Uralic-Yukagir hypothesis, according to which Uralic and Yukagir, a Paleosiberian language, are related; a Uralic-Chukotko-Kamchatkan (another Paleosiberian language or language family) hypothesis; a Uralic-Eskaleut (Eskimo and Aleut) hypothesis (see INDIAN LANGUAGES, AMERICAN); an Altaic-Korean hypothesis; an Altaic-Japanese hypothesis; and an Altaic-Ainu hypothesis--Ainu being the language of the prehistoric inhabitants of the northern islands of Japan.

Robert Austerlitz

Abondolo, Daniel, Hungarian Inflectional Morphology (1989);
Benke, Lorand, and Imre, Samu, eds., The Hungarian Language (1972);
Collinder, Bjorn, An Introduction to the Uralic Languages (1965);
Comrie, Bernard, Languages of the Soviet Union (1981);
Hakulinen, Lauri, The Structure and Development of the Finnish Language (1953-55; Eng. trans., 1961);
Menges, Karl H., The Turkic Languages and People: An Introduction to Turkic Studies (1968);
Miller, Roy Andrew, Japanese and Other Altaic Languages (1971);
Poppe, Nicholas, Introduction to Altaic Linguistics (1965) and Mongolian Language Handbook (1970);
Raun, Alo, Essays in Finno-Ugric and Finnic Linguistics (1971);
Shirokogoroff, S. M., Ethnological and Linguistical Aspects of the Ural-Altaic Hypothesis (1970);
Vago, R., The Sound Pattern of Hungarian (1980).