The Altaic Language Family 

By Dugarsuren Munkhtsetseg


The name of the Altaic language family comes from the name of the Altai mountain which the part of Himalaya mountain district and derived along into Mongolia. (The tallest point in Mongolia is Altai Tavan Bogd peak is in Altai Mountain which very famous between hunters by its' mountain sheep and a beautiful wild nature.) There are quite few things share the name of this mountain, and one of them is Altaic language family. 

The first hypothesis with regard to the relations between the Altaic and some other languages date from the first half of the XVIII century. The first scholar who noticed similarities existing between Turkic, Mongolian, and Manchu-Tungus was Johann von Strahlenberg, a Swedish officer. (Nicholas Poppe, p 125) He spent many years in Eastern Siberia and studied about Finno-Ugric, Turkic, Mongolian, Kalmuck, and other languages. He classified “The Tatar languages” into six different groups such as 

1. the Uighurs as he calls the Finno-Ugric people, the Barraba Tatars, and the Huns; 
2. the Turco-Tatar people; 
3. the Samoyeds; 
4. the Mongols and Manchu; 
5. the Tungus; and 
6. the tribes living between the Black and Caspian sea. 

This classification cannot be accepted at the present and further researches made it clear.

A hundred years after von Strahlenberg, Altaic languages was treated again by the famous Danish linguist Rasmus Rask. He renamed the language group and named them “the Scythian languages”, and added to them languages spoken in Greenland, North America, all of Northern Asia and Europe, and in the Caucasus. He also included in the group the non-Indo-European languages spoken in Spain. And the after all, the Scythian group included Mongolian, Manchu-Tungus, Turkic, Finno-Urgic, Eskimo, the Palaeo-Asiatic languages, the Caucasian languages, and the languages of the ancient, non-Indo-European in habitants of Europe.

In the middle of the XIX century linguistics were in such an advanced stage that comparative language studies could be conducted on a relatively solid foundation, once the Indo-European comparative linguistics had been established. (Muller Max p126) One of the criteria for establishing language affinity was the morphological structure. The infective structure being characteristic of the Indo-European languages, the agglutinative grammatical structure of the Altaic languages and a vast number of other languages was elevated to the main principle for judging what languages should be regarded as related. On the basis of the agglutinative features of a large number of languages, Max Muller included in the groups postulated by von Strahlenberg and Rask also Siamese, Tibetan, Dravidian, and Malayan. This expanded group was called “the Turanian languages”. The Turanian languages were only nomadic languages, as Max Muller defined them, and never could be regarded as possessing features as clearly and distinctly defined as those of the Indo-European or Semitic languages. Anyway, the problems set by Max Muller were never solved, and his own statements with regard to the languages concerned were rather vague. 

 A new period began, as Castren, M. A, who first applied linguistic criteria to languages to distinguish the language families.(Castren, M. A 49-62 St. Petersburg 1856) He was not fully satisfied with conglometing vast numbers of languages almost unknown or very little investigated. To prove the affinity of languages he used identity of morphemes as an essential feature addition on the agglutinative grammatical structure. Castren included in the language group only the Finno-Ugric, Samoyed, Turkic, Mongolian, and Manchu-Tungus and also finally this language group called Altaic by him.  The language group what was called Altaic by Castren is still called so are two different things: in Castren's writings Altaic means Ural-Altaic. At the present time only Turkic, Mongolian, Manchu-Tungus, and, with certain reservations and even reluctance, Korean are counted among the Altaic languages. 

Since Castren the Ural-Altaic theory has been uniting, in general, four language families, namely, Uralic (comprising Samoyed and Finno-Ugric), Mongolian, Manchu-Tungus, and Turkic. However, relapses into the previous stages occurred from time for a long period. Thus, several unsuccessful attempts at including Japanese were made. On the other hand, some extinct ancient languages were declared as belonging to the Altaic group, and even the name of Turanian reappeared for brief periods.

Serious investigation of the Altaic languages and their mutual relations on the basis of comparative linguistic studies began with W. Schott's work. He based his observations on vocabulary, not neglecting, morphology. His research was limited to the Chudic (his term for Finno-Ugric) and Tatar (his term for Turkic, Mongolian, and Manchu-Tungus) languages and called this group the Altaic or Chudic-Tatar group. He was the first scholar who integrated the Chuvash language into Altaic comparative studies and he made a correct observation that Chuvash is closest to the Turkic languages. 

Ural-Altaic studies were continued by the German scholar Winkler, and he proposed his own division of the languages into two groups, one of them comprising Finno-Ugric, Samoyed, and Tungus, and the other one including Mongolian and Turkic. 

One of the most serious works in the field of Ural-Altaic comparative studies is the book of the French linguist Sauvageot. He pointed out main weakness of most of the previous works in this field. He tried to do comparison of establishing regular sound correspondences. Such words were very few and the results of his work was unconvincing. 

The Ural-Altaic theory is accepted by very few scholars. On the other hand, a number of linguists believe that the Uralic languages are somehow related to the Indo-European languages. The Altaic theory, the theory about the affinity of Turkic, Mongolian, and Manchu-Tungus, including or excluding Korean, is not accepted by all scholars. The scholars who do not accept the Altaic theory can be divided roughly into two groups. The first group includes scholars who believe that the Altaic theory is premature and the affinity of the languages in question still needs further proof. The scholars concerned do not reject the Altaic theory but believe the evidence presented so far is insufficient. (Poppe 254) 

At the present, approximately 40 Altaic languages are spoken by about 100 million people. This language family is traditionally divided into three main subgroups: Turkic, Mongolian, and Tungusic (Crystal, 309). Although there is doubt as to whether these subgroups are related genetically, many topological similarities exist, such as SOV order, agglutination, and vowel harmony (Dalby, 18). In direct reputation, many scholars claim that the linguistic similarities displayed in these main groups can be explained in other ways, such as widespread borrowing as a result of socio-linguistic contact and mutual interaction (Bright, 50).  Some scholars further the controversy by asserting Korean, and even Japanese, as fellow family members raising the number of Altaic speakers to about 200 million (Campbell, 50).   However, I believe that in order to produce solid evidence to connect Korean and Japanese with the Altaic family, in addition to proving genetic relations by reconstructing protoTurkic, proto Mongolian, and protoTungusic, scholars will need to do a great deal of research.  As part of the discussion, therefore, I will discuss the subgroups of the Altaic family and some of the problems associated with them.

Over half of the Altaic languages belong to the Turkic subgroup, whose best-known member is Turkish (Crystal, 309). Today over 50 million people speak Turkish. Outside of Turkish, the other main languages are Azerbaijani with 14 million speakers, Turkmen with around 3 million speakers, Uzbek with over 15 million, Tatar with over 5 million, Kazakh with over 7 million, Kirghiz with over 2 million, and Bashkir with around 1 million speakers. Additionally, there are over 300,000 speakers of the geographically isolated Yakut and 190,000 speakers of Tuva (309). 

The main Mongolian language is called Mongol or Khalkh, spoken in two basic varieties by over 4 million people in the Mongolian People’s Republic and China. Related languages include Buryat (about 300,000), Dongziang (about 280,000), and Dagur and Tu (both fewer than 100,000). In addition, the Oirat, Kalmyk, and Mogol all total around 320,000 native speakers (309).

The Tungusic subgroup is spoken in many different dialects over a vast area. Evenki possibly reaches 20,000 speakers, although some other languages have fewer than 10,000 speakers, such as Lamut, Nanai, and Manchu. Interestingly enough, even though the Manchu people now number more than 4 million, less than only 1,000 people now speak it natively while the others have become tainted with Chinese (309). 

Because of the tireless hard work of the scholars this language family already distinguished from others and set it's own direction. But there is still uncertain matters that cause confusion and doubt. And it shows that further research and investigation is necessary of the Altaic language family. 


Shirokogoroff, S. M “Etnological and Linguistical aspects of the Ural-Altaic hypothesis” Anthropological Publications. Oosterhout N.B.-The Netherlands 1970.

Miller, R. A “Japanese and the Other Altaic Languages”The University of Chicago Press” London. 1971

Pittman, R. S “Languages of the World” Ethnologue: Language Family Index.

“A brief Exploration of the Altaic Hypothesis”

Corff, O “Sound Comparisons between Turkish and Mongolian” Hugjiltu, Inner Mongolian University. Dec 1995.

Poppe, N “Introduction to Altaic Linguistics” Otto Harrassowitz-Wiesbaden. 1965


Instructor | Textbook & Materials | Course Objectives | Major Learning Activities | Course Requirements & Grading Scheme | Resources |Language Reports | Home

1998-1999 © Dr. Cynthia L. Hallen
Department of Linguistics
Brigham Young University
Last Updated: Monday, September 6, 1999