V.Stetsyuk's personal site logo
Mail to site
Print version
News (RSS)
Nostratic Languages.… / The Areas of the Uprising of the Tocharian, Albanian, Thracian, Phrygian Languages.

The Areas of the Uprising of the Tocharian, Albanian, Thracian, Phrygian Languages.


The areas of uprising the Tocharian, Albanian, Thracian, Phrygian languages can be determined by the comparison of suppositions of many scholars with the location of the Urheimats of speakers of other Indo-European languages received by the graphic-analytical method. Let’s start with Tocharian. Many scholars believed that the first place of the Tocharian habitat should be somewhere close to the Greek, Baltic, Germanic settlements (KRAUZE V. 1959. 157; GORNUNG B.V., 1963, 25, 80); GAMKRELIDZE T.V., IVANOV V.V., 1984, 424), but more accurately, though not entirely, the Tocharian area was determined by W. Porzig:


The Tocharian language, of course, is adjacent to this group of languages (German-Baltic-Slavic – VS). It is characteristically that the common phenomena combined the Tocharian language with Baltic and Slavic and simultaneously a part of these phenomena connects all three languages to the west, and another part – to the east. In addition, there are special connections of Tocharian with German and at the same time with Greek, to which Baltic and Slavic have no relation… In such way the relative and absolute coordinates of the place of the origin of Tocharian are determined so: it is located near the German-Baltic-Slavic space in the basin of rivers flowing into the Baltic Sea. (PORZIG W. 1964: 315-316).


They consider the Tocharian area to be in the region of the rivers flowing in the Baltic Sea also on the warrant of the "salmon argument". This fish spawns in the basin of the rivers of the Baltic Sea. The word related to the Germanic, Baltic and Slavic words which means “salmon” is present in Tocharian but it is absent in the other Indo-European languages. There is a vacant area on the map between the Berezina and Dnepr Rivers close to the rivers of the Baltic basin that matches to all the requirements. The Tocharian habitat can be placed only here. According to the calculation, the number of mutual words in Tocharian and Greek is 146, and in Tocharian and Germanic – 145, in Tocharian and Baltic – 121, in Tocharian and Indic – 118. The number of mutual words in Tocharian and other Indo-European languages is considerably less. These results confirm the location of the Tocharian area between Baltic, Greek and Indic areas and near to the Germanic area.


There are other evidences in favor of this location of the area of the Tokharian language.



At right: The map of the Indo-European habitats located according the graphical model of relationships



And now about the Albanian language. Authorities have different opinions about its origin. The reason of this is the significant impact of foreign languages, particularly affected Albanian vocabulary (ZHUGRA A.V., SYTOV A.P. 1990: 64). As Agnia Desnitsky indicated, some scientists believe that Albanian is a direct continuation of the Thracian language others think that it is offspring of Illyrian. She herself supported the Illyrian hypothesis (DESNITSKAYA A.V.1966: 4), although supposed that this problem will be never solved by linguists completely. (DESNITSKAJA A.V. 1984: 727). The phonetic correspondences between the individual Illyrian lexemes and the words of the modern Albanian language "or have (because of the scarcity of material) too general character.., or do can be hardly uniquely determined". (ZHUGRA A.V., SYTOV A.P., 1990: 68).

W. Porzig considered Albanian being an individual language on a par with the other Indo-European languages, including the Illyrian and Thracian (PORZIG W. 1964: 223). Such divergence of notions is caused by the extremely poor and random lexical material from the Illyrian and Thracian languages, as well as by the large number of loan-words in Albanian from Greek, Latin, Slavic and other languages. R. Troutman presented such data of G. Meyer about the composition of the Albanian language: out of 5110 Albanian words, 1420 words have Romance origin, Slavic origin have 540 words, Turkish -1180, Modern Greek – 840, Indo-European heritage – 400 words and 730 words have unknown origin. (TRAUTMAN REINHOLD. 1948.)

This variegated vocabulary impedes the establishment of family relation of the Albanian language. To establish them, let us again turn to the lexical-statistical data on which the model of kinship of Indo-European languages was constructed. The Albanian language has the greatest number of common words with Greek (without taking into account the common Indo-European lexical stock), that are 167 words, farther follow 152 common words with German, 146 ones with Baltic, 131 with Italic, 128 with Iranian, 111 with Indic, 76 with Armenian etc. Similar calculations were made by Polish linguist Witold Mańczak. He compiled a table dictionary for comparison of the texts of the New Testament (Luke II-IV and John V-VI) in the Albanian, Polish, Italian, Greek, German and Lithuanian languages and counted word matches of Albanian and other languages. The results were as follows: Polish – 184, Italian – 167, Greek – 127, German – 97, Lithuanian and – 96 matches (MAŃCZAK WITOLD. 1987: 111-115). As you can see, there are significant inconsistencies in the data. First of all, striking is a large number of similarities between the Albanian and Polish languages, which Mańczak presented as Slavic. At the same time, Ivan Duridanov, who carried out the Thracian-Dacian study, pointed that Slavic-Thracian and Slavic-Dacian language connections don't appear (A. DURIDANOV IVAN, 1969: 100). As we will see, the Albanian language genetically traced from Thracian, so the contradiction in Duridanov's and Mańczak's data must have an explanation. Mańczak took to the study the modern Albanian language which has many loan words from Slavic, while ancient lexical fund, which we used according to etymological dictionaries, has very few matches between the Albanian and Slavic languages. Compared to this, the Thracian and Dacian material used by Duridanov applies to much earlier time. The rest of the contradictions are explained by the complex development of the Albanian.

Thus, according to lexical and statistical data, it is impossible to localize the primary area of formation of the Albanian language, since there is no one of areas on the entire Indo-European territory, the position of which could be determined taking into account all the links of the Albanian language with others. For example, if this area would be near the areas of Greek and Italic languages, then the Albanian language had to have much more common words with Armenian, but they are few. Nevertheless, these data can be used. We will refer to them more than once. First we pay attention that the Albanian language has much more common words with the Iranian language, than with Indo-Aryan. This can mean that the area of arising of the Albanian language was lying closer to the Iranian area than to area of the Indo-Aryan language. In this case, it could not be located on the west and southwest of the Indo-Aryan area, but only to the east or south of the area of the Iranian language. The territory to the east of this area was already occupied by the Western branch of the Finnish people, just by the Veps and Mordvinic speakers. The “empty” area remains to the south of the Iranian one between the Desna and Seym Rivers. If the Proto-Albanian language was arisen in this area, it would have some common words with the Modvinic languages Erzia and Moksha, which ancestors were nearest neighbours of this land. Even cursory comparative analysis of the vocabulary of the Albanian, Moksha and Erzya languages gave indeed immediately results. Here are some examples of separate Albanian-Mordvinic matches:

Alb. en¸ “a vessel” – Mok. en’a “a scoop”;

Alb. kap¸rdij “to swallow” – Mok. kapordams “to swallow”, Erz. koporks “to swallow”;

Alb. kofš¸ “thigh” – Erz. kačo “thigh”, Mok. kače "abkle";

Alb. keqe “evil” – Mok., Erz. k’až “evil”;

Alb. bizele “peas” – Mok., Erz. piz¸l “berries of mountain ash”;

Alb. rroj "to live" – Mok., Erz. er’ams "to live".

Some Albanian-Mordvinic correspondences have matches in other neighbouring languages:

Alb. tani “today” – Mok. t’ani “now” (Mari tenij “today”, Est täna “today”);

Alb. dob¸t “quiet” – Mok. topafks “satiety” , Erz topafty “sated” (Mari typ “quiet”);

Alb. turi “muzzle” – Mok, Erz trva “a lip” (Mari t’arvö “a lip”);

Alb. bretkos¸ “frog” – Mok. vatraksh “frog” (Greek βατραχοσ “frog”, Rom broasca "frog" as possible spoiled Thracian substratum) etc.

Alb. piče “pine-tree” – Mok, Erz piči “pine-tree” (the words of this root are present in other Finno-Ugric and Indo-European languages but they mainly “galipot”, for example, Lat picis, Gr πισσα, but Lat picea “pine-tree” remains something far;

Alb zhavor “gravel” – Mok šuvar “sand” (this word in several different forms was spread in the Baltic and Slavic languages).

Above examples are enough to recognize that the ancient Albanian-Mordvinic connections were real and therefore allow us to locate with a high probability the area of formation of the Albanian language just between the Desna and Seym Rivers. But then how do we explain the fact that the Albanian language has the most common words with the Greek, Germanic and Baltic languages? As we shall see, this is the result of later borrowing after the first stage of resettlement of Indo-European tribes.

Very few lexical data are available for the localization of the Urheimat of the Illyrians, but, according to the other language facts, Walter Porzig found arguments to claim that Illyrian and Celtic areas were adjacent in early history (PORZIG W. 1964: 159). He also indicated that “Illyrian and Greek have remarkably few mutual connections though the both peoples lived in permanent proximity since the time of Illyrian migration.” (Ibid. 224). If we conjecture the Illyrian area to be near the Celtic and far from the Greek ones, we can find some additional data to prove the location of the Illyrian area somewhere on the West Indo-European territory. Place names can be very helpful in this case. The Illyrian toponymy was studied by the linguist Oleg Trubachiov and the archaeologist Dmitriy Telegin. Pointing out at the relative vicinity of the Celtic and Illyrian onomastics in general, Trubachiov wrote: "…hydronyms with West-Balkan connections are concentrated on the Dnestr narrow space and are sporadic present to the north in the Goryn’ river's basin and to the north-east in the catchment of the river Teterev river". (TRUBACHIOV O.N. 1968: 279). D. Telegin confirms Trubachiov's idea and specifies that the Illyrian (Celtic-Illyrian) hydronyms make three concentrations in the Ukraine: the Kiev, Zhytomir and Upper Dnestr accumulations. When the first two of them have only ten names, the last one has almost 30 names (TElEGIN D.Ya. 1990-1). Thus we have the reason to believe that Illyrian people resided in the Upper Dnestr river's basin and populated the territory to the north at certain period. Their Urheimat was just in the region between the rivers Sluch, West Bug and Pripyat’.

The Urheimat of the Phrygians was to be close to the Greek and Armenian areas because the vicinity of the Phrygians to the Greeks and the Armenians is confirmed by numerous linguistic data. For example, Gamkrelidze and Ivanov wrote: “Phrygian language… has structural features that bring it close to the dialects of Greek-Armenian area”. (GAMKRELIDZE T.V., IVANOV V.V. 1984: 910). Armenian scholar Gr. Kapantsyan indicated that Greek annalists (Herodotus, Eudoksus and other) were writing about the vicinity of Phrygians and Armenians. ”Phrygians and Armenians were together under the same banners in the army of Xerxes and they were dressed and armed identically” (KAPANTSIAN GR. 1956: 164). Russian scholar T. Moisejeva also wrote about closeness of Phrygian to Greek and Armenian (MOISEYEVA T.A. 1986: 13). At the same time, the area of the formation of the Phrygian language should have been close to the Thracian language, since this two languages according to the schemes of Hirt and Ler-Splavinsky are also closely related.


Under these conditions, the place of the formation of the Phrygian is most suitable for the area between the Desna and Iput Rivers, south of the area of the Indo-Aryan language (se the map at right). If such an arrangement corresponds to reality, then the Phrygian language should also have close relations with the Indo-Aryan language, however, checking this is prevented by the scarcity of the available Phrygian lexical material.

The number of common Hittite-Luwian words with other Indo-European languages suggests that the area of its formation should be located somewhere between the areas of Greek, Armenian, and Italic languages. Common matches for Hittite, excluding common Indo-European roots, found in the Greek – 106. Slightly less in the Armenian – 102, a significant portion of which was included in the table according to Kapantsyan (KAPANTSYAN Gr., 1956), which gives a lot of Armenian-Hittite parallels that are not present in the etymological dictionary of Pokorny . 86 matches were found as in the Italic and in Germanic languages. According these data, the area of formation of Hittite was located in the triangle between the Dnieper, Teteriv, and Ros Rivers, that is on the part of Tripilla culture. But the main argument for this decision was the common reasoning. The boundaries of the territory of the settlement of any ethnic group should not look like a broken line, that is, the area should not have distinct bulges and concave, otherwise the inhabitants of these places should be assimilated or pushed out by non-native neighbors.

However, there are reasons to believe that the ancestors of the Hittites never were present in Eastern Europe and were a part of the Indo-Europeans which remained at the site of the original settlement, and later they spread almost all over Asia Minor. This opinion was suggested to me by Peter de Reijk (the Netherlands ) with reference to Alvin Kloekhorst which stated:


The Anatolian branch can be shown to have been the first to split off from Proto-Indo-European because several instances can be identified in which Hittite shows an original situation where all other Indo-European languages have undergone a common innovation. This means… the ancestors of the speakers of the non-Anatolian Indo-European languages shared a period of common innovations that no longer reached the ancestors of the speakers of Proto-Anatolian (KLOEKHRORST ALWIN, 2008, 88).

To go along with this idea, there are good reasons and they are given by Alvin Kloekhorst, but this difference of the Hittite language from the rest of Indo-European has long been noted, especially by Gamkrelidze and Ivanov (GAMKRELIDZE T.V., IVANOV V.V. 1984, 395). First a small number of Hittite words, which have matches in the other Indo-European languages are striking. The Indo-European origin of Hittites is most evidenced by grammatical forms:


… Hittite root-words have in most non-Indo-European origin, only ten percent of them originated from PIE, while the forms of declension and conjugation, as well as forming new word are distinctly Indo-European, although there are also representatives and morphological forms of Asia Minor (Asiatic) (KAPANTSYAN Gr., 1956, 79).


Obviously the words of Indo-European origin are the oldest in the Hittite language therefore they are important for historical-linguistic analysis. But the grammatical forms, which are different from the Indo-European, can be both archaic and arisen at the time when the bulk of the Indo-Europeans migrated to Eastern Europe. The timing of their occurrence may be a good material for a general history of the development of language grammatical structure.

Therefore, under such circumstances, it must be assumed that the area between the Dnieper, Teteriv, and Ros' Rivers was populated by some other Indo-European tribe, for example, by the Thracians. Some researchers, in particular, D. Telegin argue that the greatest concentration of Thracian (Daco-Thracian) hydronyms can be found in the basins of the Southern Bug, Ros', Teteriv, (Ibr, Yantra, Alta, etc.) (TELEGIN D. Y, 1990-1). I. Zhelezniak had the same reflection (ZHELEZNIAK I. M, 1987). However by the time when Thracian language was arisen, the area near these rivers could not be hold by the Thracians, because, in this case, they should have been the first of the Indo-Europeans who start moving southward, but this could not be. Usually people in their migrations adhere to the order, which is determined by their places of previous settlements. If the Thracians later occupied the territory north of the Greeks, their ancestral home could not be in south of Greek one. Obviously mentioned area was populate by some Indo-European tribe, whose descendants are lost in history without a trace.

After such placement of areas of Indo-European languages it is appeared that the Thracian language has no free area at all. Since it is close to Phrygian and Albanian and the successor or Illyrian or Thracian, then there is no choice but to place the Thracian area where we placed the ancestors of the Albanians, ie to consider Albanian as a continuation of Thracian. This area is bordered by the Desna River and its left tributary of Nerussa in the west and north, and in the south and east by the Seym River and its right tributary Svapa.

Having the areas of uprising of the Tocharian, Albanian, Thracian, Phrygian languages we can construct the whole map of the Indo-Europeans territory.




The common territory of the settlement of Indoeuropeans in the Dnepr basin
Below the same in GoogleMap.



View Indo-European in a larger map





Free counter and web stats

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 : 5891

Modified : 12.12.2018

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