V.Stetsyuk's personal site logo
Mail to site
Print version
News (RSS)
Far East: The Relationship of the Altaic and Türkic Languages.

Far East: The Relationship of the Altaic and Türkic Languages.

Edited by Veronica Veron Cruz Wong

The seeming self-evidence of the Altaic ancestral home of the Türks logically led to the idea of the genetic relationship of the Mongoliń and Turkic languages, but the facts contradicting this doctrine gave rise to many years of debate by specialists:

At the present time, this problem, like many others, results in a dispute between whole generations of scientists, neither of which can convince each other. Apart from the deceased scientists, the main supporters of the theory in question are such older scientists as prof. N. Loppe (Washington University) and prof. K. Menges (Columbia University), who devoted their entire lives to the study of these languages; opponents of this theory are mostly young scientists, for example, G. Doerfer and A. M. Scherbak (Leningrad) (CLAUSON G. 1969: 22).

It took about 50 years to write about these lines, but the situation has not changed, the number of supporters of the Altai theory have not diminished. In general, in approaching the problem to solve the origin of the Turks, Asian scholars often note the prejudiced attitude of the European. In particular, Balkarian turkologists consider this attitude as follows:

The origin of the Türkic peoples and their ethno-cultural traditions is one of the least studied problems of science. Its undevelopedness is due not so much to the weakness of the scientific base, but to the biased attitude of many scientists, especially the Iranianists, towards it. The Indo-Europeanists dictated and continue to dictate to the Turkologists the ways and methods of studying ethnogenesis and other problems of the history and linguistics of the Türkic peoples. At the same time, some of them deliberately distort, and sometimes simply ignore, their historical and cultural wealth (LAYPANOV K.T., MIZIEV I.M. 2010: 4).

However, this reproach is not entirely fair. Although well-known turkologists, first of all, N.N. Poppe, Baskakov N.A., S.A. Starostin, J. Benzing, L. Ligeti did not doubt the genetic relationship of the Turkic and Mongolian languages, at the same time, others, such as K. Grǿnbech and J.R. Krueger claimed that reliable evidences of this are absent but to their opinion the whole attitude was skeptical. Sir G. Clauson stand in their defense. Based on the analysis of the ancient Türkic and Mongolian texts, he came to the conclusion that they practically have no common vocabulary, except for international words such as kagan "supreme ruler" and teŋri "heaven" (CLAUSOΝ GERARD, 1956: 182). Later, Sir Gerard dealt with this issue in more detail and came to the following conclusion:

It has been suggested that the Turkish, Mongolic, and Tungus-Manchu languages form such a family, commonly called the Altaic, and that they are all descended from a lost primeval language called Altaic or Proto-Altaic. For some years now I have been coming more and more to the opinion that this is an error and that the fact that these languages gave a good deal of vocabulary material in common is best explained, not by assuming that they have inherited it from a common ancestor, but by assuming that a prolonged and complicated process of exchanges has taken place between these languages… I am quite convinced that Turkish is not genetically related to either of them. (CLAUSON GERARD, 2000: 21-22).

One can add to the words of Sir Gerard Clauson that Japanese and Korean are considered to be belonged to the Altaic family. Earlier, using the graphical-analytical method, we found the Türkic Urheimat in the Transcaucasia (see The Nostratic languages) and then localized the habitats of Türkic tribes speaking particular dialects originated from the common parent Türkic language in East Europe (see The uprising of the Türkic Languages). In this case, the Mongolian languages can be related genetically to Turkic only if they were formed in the vicinity surrounding area. Let us try to establish whether this was so.

Sergey Starostin compiled a large database the on the Altaic languages (The Tower of Babel), and this is presented in the Internet in a form that allows one to apply the graphic-analytical method to justify kinship (see below).

At left: The graphic model of the general relationship of the Altaic languages

As you can see, the Türkic languages fit quite well into the constructed model, but, as it was determined previously, they take one of the central places in the relationship model of Nostratic languages, which corresponds to an area in Asia Minor. In this case, we have to determine if the other Altaic languages belong to Nostratic or not. To resolve this issue we need to find a terrain for a graphical model of the Altai languages on which their parent languages could have arisen. The graph can be placed in different places in Europe and Asia since it only has five knots. To facilitate the search for an exact place, we first have to build models of the relationship of the Mongolian and Tungus-Manchu languages to find more confidently place on a geographical map with available ethno-producing areas where these languages could arise.

Tungus people

Having received such survey, it will be easier establish the place of formation of the other Altaic languages. Let us start with the construction of the model of the Tungus-Manchurian model, the data for which can be taken fromData for the Tungus-Manchu languages were taken from Databases compiled by Anna Dybo. There are Database 2294 Proto-Tungus-Manchu roots for which for are given matches of 11 following languages: Evenki, Nanai, Manchurian, Even, Negidal, Ulchi, Orok, Oroch, Udighe, Solon and Jurchen. All these languages have 134 common roots, which have matches at least in ten languages&. They were eliminated from analysis of kinship of the Tungus-Manchu languages by the graphical-analytical method. For the rest of the roots numbers of common words in each pair of these languages were calculated. These data are shown in the table below. Also data on the distances (in centimeters) between the points of sets of individual languages are submitted for convenience. They were used to build the model of kinship using a coefficient of 3500 for proportionality 3500. In the table main the diagonal shows a total number of words for each language in the database. Quantity of Solon language words is too small to reliably insert it into the scheme of kinship, although its place would have to be in the far west (according to the number of common words with the rest of the Tungus-Manchu languages). It is impossible to say that about the place of the dead Jurchen language among the Tungus-Manchu languages the based on the table.

Quantuty of common words between the individual Tungus-Manchu languages

Evenk Nanai Manch Even Negid Ulcha Orok Oroch Udighe Solon Jurch
Evenk 1431 4,4 5,4 3,8 4,2 5,6 5,5 6,7 6,5 12,6 -
Nanai 791 1218 5,4 5,8 48 59 5,2 6,0 6,2 15.8 -
Manchur 642 649 1064 7.3 7.1 7.1 8.0 9.0 8.9 20.1 -
Even 916 601 478 1063 5.1 7.3 6.7 8.4 8.1 14.8 -
Negidal 838 726 492 684 1022 5.6 5.7 6.6 6.0 13.8 -
Ulcha 623 638 492 478 621 936 5.6 6.7 7.5 17.9 -
Oroki 630 671 435 520 610 621 857 7.4 8.1 17.7 -
Orochi 519 585 388 415 529 523 470 721 7.6 19.1 -
Udighe 542 558 393 428 512 468 432 460 713 21.9
Solon 277 221 174 237 254 195 198 183 160 320
Jurchen 97 117 158 71 84 91 75 72 65 46 170

Based on this table, the graphic model showing kinship of Tungus-Manchu languages was built (see below).

At right: The graphic model of the relationship of the Tungus-Manchu languages

The selection of one of the two possible mirror variants of the model and its orientation are determined by actual residence of the speakers of the Manchu and Evenki languages. The Manchus, the most numerous people of the family, live now in the North-East China, that is, south of the rest of kindred peoples. The second largest nation, the Evenki currently settles the vast area from the Yenisei River to the Okhotsk Sea with the southern boundary along the Amur and Angara Rivers. Their small settlements are located also in Mongolia and China, but in general Evenkis migrated westward.

The residence of other Tungus-Manchu peoples largely correspond the model. This correspondence can be found on the map showing the small groups of people of the Khabarovsk Region and Sakhalin Island (see below). The map was compiled according to different sources as of the national contingent of administrative units of Region, and settlement of the bulk of the small peoples. If you do not take into account the migration of some people from the ancestral home, the correspondence is obvious.

At left: The map of actual habitats of the small peoples in the Khabarovsk Region and Sakhalin Island. The map shows the main places of settlements by large font and the sporadic ones do by smaller font.

Let us consider more detailed. The ancestors of the Evens followed the Evenkis, the bulk of which resides in central Siberia now, but a small part of them remained near their ancestral home. The ancient speakers of the Solon language which area had to be somewhere in the west of the common territory of the Tungus-Manchu, went just to the west and now dwell in Mongolia.

The actual space of the Udeghes contradicts the model (their position on the model is defined in the far north) as now they reside in the south of Khabarovsk Region. However, the Orochi, the Ulchi, the Negidals, and the Nanai kept their settlements near the ancestral home. In this case, we must assume that the Udeghes had to move to the south, bypassing the settlements of the Negidals and Nanai. Sakhalin logically had to be settled by the Orochis, but this was done by the Orokis, which had to pass Orochi settlements.

The search for the Tungus-Manchu ancestral home by other linguistic methods leads to a territory, close to the one that was determined by the graphical-analytical method:

… the vocabulary of the Tungus-Manch languages (the names of some leafy trees, salmon, as well as the names of rivers) gives reason to include Middle Amur basin in the area of the Tungus-Manchu ancestral home (country of linguistic ancestors) (PEVNOV A.M.. 2008, 67).

However, location of the areas that formed the individual Tugnus-Manchu languages by purely linguistic methods is impossible.

Now we turn to the construction of a graphic model of Mongolian languages. It was built by means of using the Database, compiled by Oleg Mudrak. He gave data on the following languages: Mongolian writing, Middle-Mongolian, Khalkha, Buriat, Kalmuk, Ordos, Dungxian, Baoan, Dagur, Shary-Youghur, Mongour, and Mughal. The database is constantly updated, so it became necessary to build a new graphical model showing the kinship Mongolian languages instead of what was published previously. Data for Mongolian writing, Middle-Mongolian, and Mughal were not taken into account, for obvious reasons. The data were summarized in the Table (see below).

Quantuty of common words between the individual Mongolic languages

Khalkha Buriat Kalmuk Ordos Dagur Mongour Shary-Yg Dungxian Baoan
Kalkha 1540 - - - - - - - -
Buriat 1297 1319 - - - - - - -
Kalmuk 1287 1297 1308 - - - - - -
Ordos 1051 942 964 1062 - - - - -
Dagur 561 538 531 486 585 - - - -
Mongour 472 435 430 402 301 492 - - -
Shary-Ygoughur 329 317 318 306 253 202 342 - -
Dungx 178 167 168 164 113 140 83 179 -
Baoan 119 112 114 90 82 92 61 67 121

Due to the insufficient number of words in the database for the Dungxian and Baoan languages, they cannot be included in the model (see it at right).

At right: The graphic models showing the relationship of the Mongolic languages

The experience of the research of different language families and groups using the graphical method allows us to assert that a high density clusters on the graphical model often displays not only proximity of language areas but also a greater quantity of documented words in these languages compared to others. This phenomenon was taken into consideration when searching for a suitable place for placing the obtained model on a geographic map. Having the graphic model of kinship of the Altaic languages and the found territory of the Tungus-Manchurian, we can determine the place for the Mongolic languages in the basin of the right tributaries of the Amur River. The supposed areas of the Dungxian and Baoan languages were located from general considerations (from modern settlements and the proximity to other Mongolic language).

Thus, we come to the obvious conclusion that the entire territory on which the formation of the Altai languages took place should be located in the Amur basin and in the adjacent territory (see the map below).

At left: The areas of origin for the Mongolic, Tungus-Manchu, Korean, and Japanese languages.

According to the model of relationship, the Korean language (the most remote from the rest of the Altaic) was formed on the peripheral area, i.e., the Korean peninsula. That the Koreans populate this territory at presentm it is an additional fact in favor of the correctness of the location indicted by the graph. The ancestors of the modern Japanese dwelled in the territory of Primorye, well limited by the Ussuri River, Amur, and the shore of the Sea of Japan. From there they moved to the Japanese island trough Sakhalin, either directly on the frozen sea.

The peoples of the Far East, being on the outskirts of the civilized world, for a long time maintained their existence by hunting, fishing and gathering, i.e. were still at a rather low level of social development. Common culture and languages of the local population corresponded to this level. Structural changes in the economy of the Amur region begin with a gradual transition to productive management. The data of archeology display the appearance of "agricultural features" in this region roughly since the end of the III and beginning of the II millennium B.C. (BROMLEY Yu.V., 1986: 257). There are evidence that the development of agriculture is associated with the penetration here of tribes from Central Asia and Southern Siberia.

In accordance with the graphical model of kinship of the Altai languages, the territory inhabited by the Turks was supposed to be in Transbaikalia being limited in the east by the Argun River, beyond which the region of the Mongolian tribes was already beginning. Thus, we can assume that when in the beginning of the III millennium BC. some Turkic tribes left their ancestral homeland in the interfluve of the Dnieper and the Don Rivers and moved eastward, some part of them reached Altai and further (See The First Great Migration). Here they met a population of Mongoloid anthropological type that spoke Mongolian dialects as a whole less developed and poorer in comparison with the languages ​​of the newcomers-Turks, who stood at a higher level of cultural development. The consequence was a massive infiltration of the Türkic lexical and grammatical forms in the language closest neighbours – the Mongols and Tungus-Manchus, and from these languages in languages of more eastern ethnic groups.

In this way, the Türkic and Mongolic languages acquired some formal similarities that deceive to-day linguists that Altaic and Türkic language families have the same genetic origin. Meanwhile, commingling between two populations resulted in the acquisition of Mongolic physical traits by the Türkic peoples. Later, during the Mongolic expansion (around the start of the 13th century) the reverse process of borrowing from the Mongolic to Türkic languages started quite naturally. Mongolic loan words in the Türkic languages hid the question about relation the Mongolic and Türkic languages still more. Sir Gerard Clauson noted that “the existence of this massive volume of Mongolic intruders seems somehow to have escaped notice, or, if noticed, to have been regarded by those who accept Altaic theory as evidence of a common “Altaic” heritage in both language groups” (CLAUSON GERARD, 2000: XI). Such circumstance gave us exactly the chance to construct the graphical model of typological relation of Altaic languages.

View The Mongolic and Tungus-Manchu languages in a larger map

We have strongly reject the genetic relationship of the Turkic languages with Mongolian and Tungus-Manchu and further evidence of this statement are presented separately in the section Discussion. However for the reader's convenience, we will briefly repeat them here.

The main proof of the absence of a relationship between Türkic languages with the rest of Altaic is the formation of a individual proto-languages, which are considered belonging to Altaic, occurred in different places. The Urheimat of the Türks (determined by using the graph-analytical method) was in Asia Minor, and not on the Altai, as it is commonly believed.

The kinship of the Turkic and Mongolic languages is contradicted by absence of sufficient counterparts between the words that can be considered as the most ancient, most of which are usually the most commonly used in speech. They constitute the main lexical core of a language. At one time, the American scientist Morris Swadesh tried to compile a list of lexical nucleus that should be present in all languages and that, in his opinion, decomposes with the loss of individual words at a constant speed. Swadesh included in the list a hundred words, but later expanded it to two hundred (SWADESH M. 1960-1, 1960-2). If we compare these lists of restorated Mongolic and Turkic patern languages, it turns out that they have almost nothing in common, except for some pronouns that may arise even in a human proto-language, the names for men and heart, and three colors (black, yellow and green/blue), clearly borrowed from Turkic. Experts can add to these words two or three more, but for the hundred words of the lexical nucleus this is very few, because it didn't decay so quickly in any language.

Many other linguistic facts are also contrary to the Asian origin of the Türks, but one looks far-fetched explanation for them. For example, Sir Gerard Clauson, considered the common Türkic name of cannabis kendir, wrote: "Unlikely to have been an indigenous plant in the area originally occupied by the Türks and probably an Indo-European (?Tokharian) l.-w." (CLAUSON SIR GERARD, 1972). Strange thought as the word is wide spread in the Türkic languages and is completely absent in Indo-European.

Sir Gerard Clauson brings up another interesting fact. He believed that the Latin. virga "a branch, twig" has no matche in other Indo-European languages, but gave Türkic ones – OT. bergä "a rod, whip", Uighur. berge "a whip, stick". He wrote:

It is suggested.., that it is a loan-word from Latin virga ‘a rod, a stock’ obtained through Middle Persian but there does not seen to be any trace of the word in Persian, and the theory is importable. Ibid.

In favor of the Asiatic origin of the Türks, many borrowings from the Chinese language could say, but the phonetics, morphology and syntax of these languages are so different that there is no reason to speak of their kinship. Moreover, the Chinese and other Altai languages also have no kinship. This is because their formation occurred in different places, and their speakrs migrated to Asia at different times (see the map below)

Left: Spreading homo sapiens.
1. Homo sapiens.
2. Neanderthals.
3. Early Hominids.

The map on the left shows that the Amur River basin was inhabited by modern humans about 20 thousand years ago, ie, before the speakers of the Sino-Tibetan languages have penetrated into China (see the sec The Relationship of the Sino-Tibetan languages). This explains the fact that the Sino-Tibetan languages, on the one hand, and the Mongolian and Tungus-Manchu, on the other hand, have no genetic relationship.

Searching of connection of the last should be carried out, where they came from, that is, from the Mediterranean region. Minoan and Japanese have certain similarities in sound structure that are considered strange and therefore the assumption that there is a relationship between the Japanese and ancient Cretans seems impossible due to geographical remoteness and so this possibility is excluded (SHEVEROSHKIN V.V., 1968 310). Perhaps a more careful examination of the structure of the Tungus-Manchu and Minoan languages will help to find the truth.

Please this page? Help us to develop this site!

© 1978 – 2018 V.Stetsyuk

Reprinting of articles from site are encouraged while
reference (hyperlink) to my site is provided

Site powered by

Load count : 6908

Modified : 8.04.2018

If you look up the type error
on this page, please select it
by mouse and press Ctrl+Enter.