Following the next text we have to keep in mind that the present-day Albanian language is a descendant of Thracian (see The Areas of the Uprising of the Tocharian, Albanian, Thracian, Phrygian Languages
As it is known, according to the principle of reflection of palatalized velars or in the form of affricates and fricatives, either in the form of pure velars, all Indo-European languages are divided into two groups Satem and Centum. The Satem group includes the Indo-Aryan (Indic), Iranian, Baltic, Slavic, Albanian, Armenian, Phrygian, Thracian languages. The Hittite-Luwian (Anatolian), Greek (Hellenic), Italic, Germanic, Celtic, Tocharian, Illyrian languages belong to the Centum group (POKORNY JULIUS. 1954: 376). However, there is little disagreement about this division, G. Province, for example, refers to a group of Phrygian language "centum", although it is questionable (KRAHE HANS, 1966: 30). Currently, the same view adheres to F. Kortlandt (KORTLANDT FREDERIK. 2016).
In principle, this separation could affect the location of individual Indo-European languages so that the two groups would have fairly clear geographical boundary, but hardly the such can be found. Indeed, one would assume that the division occurred on the principle "east-west" as it was supposed by W. Porzig (PORZIG W. 1964: 315). In such case, the boundary could lay along the Dnieper River, as most Satem language areas were located on the left bank but all the Centum ones did on the right bank. However the location of the area of the Centum Tocharian language between the Satem Indo-Aryan and Baltic languages contradicts this assumption. The complexity of the division of the Indo-European languages to the groups Satem and Centum due to involvement of other new data resulted that some scientists even refuse this division (GORNUNG B.V. 1963: 14; VINOGRADOV V.A. 1982: 259). But the contradictions can be resolved for the most part adopting and developing the explanation of this division proposed by Agnia Desnitsky:
Proto-Albanian belonged to the Satem group languages, preserved the row of palatal guttural consonants and carried out their assibilation. However satemization was "inconsistent" in this language like as in the Baltic languages (less for Slavic ones). This "inconsistency" can be considered as a feature of fairly wide transition band in the central part of the Indo-European space, which could consist apart from the Baltic and Proto-Balkanic also of the Illyrian (with Messapian) and Thracian languages. The tendency to assibilation of palatal consonants was powerful innovation of the period prior to rupture of the territorial contacts between different parts of the Indo-European community. This innovation weakened moving from east to the central zone, meeting with the approaching from the west the tendency of the neutralization of the opposition of palatals and velars. (DESNITSKAYA A.V. 1966: 11-12).
Illyrian could not enter the central part of the Indo-European space, and the wave of the assibilation of the palatal could not reach it, it remained the expressed Kentum language. This wave, according A. Desnitsky, arose under the influence of Finno-Ugric languages, which include a rich set of sibilants s, s', š, and only two the gutturals k and q', while the Indo-European languages had one fricative s and a large collection of guttural voiced, voiceless and aspirated consonants (DESNITSKAYA A.V.1968: 12). Since the Albanian language manifested “inconsistent Satemization”, A. Desnitsky places it in the middle of the Indo-European space, where the influence of Finno-Ugric languages were felt less. The central position of the Albanian language among Indo-European languages was further confirmed by two other facts which were given by A. Desnitsky:
The whole dialect space, which Proto-Albanian belonged, according the feature of the setting of the Proto-Indo-European scheme of three short vowels *e, *o, *a to the schema of two vowels *e, *a (in Slavic *o) opposed to the vast area in southwest and south of the Indo-European space (Celtic, Italic, Greek, Phrygian, and Armenian), where the three-part scheme was preserved, and Indo-Iranian region where three-term model was reduced to a one-term one *o (DESNITSKAYA A.V.1966: 10).
That is, three areas existed: the area A, which has kept all three of the ancient Indo-European short vowels, the area B included the Proto-Albanian, that is Thracian, language and characterized by displaying them in two vowels, and the area C where they were reflected to one vowel.
The second fact is very important also for determination of the order of migration of speaker of particular Indo-European languages from their Urheimat.
Indo-European voiced aspirated stops *bh, *dh, *gh transformed in Proto-Albanian to simple voiced stops b, d, g, and coincided with the voiced stops which were preserved from the Indo-European state. In this respect, the Proto-Albanian language has evolved the same way as a lot of languages, including Baltic, Slavic, Germanic, Illyrian and Thracian, Celtic, and Iran. On this basis, these languages were objected to Italic, Venetian and Greek lost of voiced aspirated row, but have retained the distinction of three rows. An important innovation of the vast area, stretching from the Iran language area on the east to the Celtic area on the west, was an uniting of two rows of the Indo-European stops (simple voiced and aspirated voiced) into one row. Only the German dialect area in the centre of this area did not carry out this innovation, having carried out Sound Shift (Grimm's law") and keeping the original distant relationship between the three rows of the Indo-European stops (Ibid: 11-12.)
However we need to do some clarification here. There were in PIE language not three but four rows of stops – the voiced aspirated, voiced plain, unvoiced aspirated, and unvoiced plain ones. However unvoiced aspirated stops were very rare, so Desnitsky said only about the three series, although included in consideration also unvoiced aspirated stops in a later paper (DESNITSKAYA A.V.1968). But the most important is the unity of voiced aspirated and voiced plain stops in one row of voiced plain stops in the Celtic, Germanic, Slavic, Baltic, Iranian, Armenian, Thracian, Albanian, and Illyrian languages, i.e. voiced aspirated stops did not survive in these languages. Voiced aspirated stops were preserved in Greek, Italic, Indo-Aryan (one can not say sure about Tocharian), another thing is that they reflected in each language in different ways later. It implies of this that the ancestors of Italics, Greeks, and Indo-Aryans would have been among the first to leave their Urheimat and so saved the old Indo-European sound composition. Having lost contact with each other and got into the neighbourhood with native speakers of another sound structure, they could fall under the different language influences and therefore, e.g., Greek voiced aspirated bh, dh, gh transformed in φ, θ, χ, and Latin bh, gh did in the f, h accordingly. Unvoiced aspirated ph and th reflected in Greek to φ and θ but coincided with p і t In Latin. The languages of other Indo-European peoples, who remained on their previous places, were developed by more or less common phonetic laws, and so all they have lost aspirate bh, dh, gh, ph, th, kh (including the Germanic languages, although there we have Sound Shift).
Reflection of stops in individual Indo-European languages
Taking into consideration the location of the primary areas of individual Indo-European peoples and subsequent places and time of population them, we can assume that the first who left their Urheimat were Italiсs, Greeks, and Indo-Aryans. It is possible that the first migration wave reached Asia Minor at the end of the III millennium BC, when it came to the Balkans:
The archaeological cultures of the last quarter of the 3rd mill. BC in Asia Minor came the really dramatic changes. These changes suggest the emergence of new ethnic elements that can be identified with the ancient Anatolians – but rather about their appearance from the west than from the east. (DIAKONOV I.M. 1968: 26-27).
More confident we can talk about the migration of the Greeks, who in their motion to the Balkans could use the waterways. In contrast, Italians traveled through the land. The movement of the Indo-Aryans had to be prevented by the Phrygians and Armenians, therefore, the latter migrated so that did not prevent the movement of the Indo-Aryans and at the same time remained near the Indo-European territory. The Tocharians should still stay in their Urheimat for some time, when it Italians, Greeks and Indo-Aryans already went away (PORZIG V., 1964, 319). The Thracians (Proto-Albanians) were involved in this movement of Indo-European populations and settled somewhere in the center of the Indo-European territory, what can explain the localization of their Urheimat by Agnia Desnitsky:
The outpouring of the relation of Albanian with the North-Indo-European languages (Baltic, Slavic, Germanic) give reason for searching For-Balkan homeland of the group of Indo-European tribes to which belonged also the ancestors of the Albanians, somewhere in the neighbourhood of the settlement space of the North-Indo-European tribes (DESNITSKAYA A.V. 1984: 220).
We can also assume that the Phrygians and Proto-Armenians went by this way before, departing later somewhere near southward because they still stayed in the space of Indo-European linguistic influences. Obviously, these effects were smaller than on Albanian, as Armenian remained as well as Satem language. At the same time, with such an assumption, the Phrygian centumism should be considered problematic. The Thracians settled in the triangle between the Teterev, Ros, and Dnieper rivers. As a result the satemization of Thracian was stopped because the powerful Dnieper border prevented contacts of the Albanian speakers with the Iranian tribe speaking languages of the Satem group. Thus Western influences took over in the Thracian language. In contrast, the Balts and Slavs moving eastward after the departure of Tocharian ancestors came into direct contact with Iranian speakers and later with the Finno-Ugrians therefore palatalized velar k', g' of their language transformed in front sibilants. That is, these languages, primarily belonging to the Centum group, came under the process of satemization later than others Satem languages. V. Abayev thought so, referring to V. Georgiev, who suggested that the assibilation of palatals occurred in Indo-Iranian language no later than the 3rd mill BC, while this process occurred in Slavic "in an era not so distant from the before written records (ABAYEV V.I. 1965: 141). Obviously, the satemization of the Slavic languages occurred even later than in the Baltic when settling the territory of the Balts under the influence of the local substrate.
According to the lexical and statistical data on which the model of the Indo-European languages was built, the Albanian language has the greatest number of common words (excluding the common Indo-European lexical fund) with Greek. On their ancestral homeland, the Proto-Albanians did not have direct contacts with the Greeks; therefore, the total number of Albanian-Greek correspondences included Greek borrowing in Albanian from Greek already during the time of the Thracians in the Balkans. If we exclude these borrowings, it turns out that Albanian has the most lexical correspondences with Germanic languages. The Thracians also did not have contacts with the Germanic tribes during the formation of their language, but having crossed the Dnieper, they found themselves very close to the habitats of the settlements of the Proto-Germans (see the section Germanic Tribes in the Eastern Europe at the Bronze Age). This neighborhood has identified Germanic-Albanian connections that are found in the phonetics, vocabulary and grammar of the Albanian language.
Below are some typical cases of Albanian-Germanic lexical correspondences given by A. Desnitsky (DESNITSKAYA A.V. 1965: 33-38). Only a part of them is taken into account in the etymological work of Yu. Pokorny. Other languages of the North Indo-European area are also involved in some of these correspondences:
1. Alb. barrё (IE *bhorna) «burden» — Goth., O.H.G, O.Icl. barn «child».
2. Alb. Gheg. bri (stem brin-) «horn» (IE. *bhr-no-) — Sw. dial. brind(e) (*bhrento), Norw. bringe «elk». Also cf. Let. briedis, Lith. briedis «elk».
3. Alb. bun "shepherd's hut in the mountains", originally "housing", buj/bunj “sleep” out of IE *bheu, *bhu-. Close by meaning: Goth. bauan, O.Icl. būа, O.H.G., O.Saxbūаn "live, dwell, cultivate (land)"; O.Icl. būа, O.Eng bū "housing".
4. Alb.dhi, «goat» (Proto-Alb. *diga) — O.H.G ziga, IE *digh "goat".
5. Alb. gjalm, gjalmё «lace, string» — O.H.G sell, O.Saxsel, O.Eng. sal «rope», Goth insailjan «to rope».
6. Alb kale «fish bone, awn of the ear» (IE. *skel- «cut») — Goth. skalja «shingles», O.Icl. skel «flake», O.H.G scāla "a shell of the grass." Formations from this root are also represented in other Indo-European languages. But only in the Albanian and Germanic have a special similarity in the structure of the stem and in the meaning.
7. Alb helm "poison, sadness, grief" — O.H.G. scalmo «plague», skelmo «criminal». Out of IE. *skel «cut».
8. Alb. hedh «throw out» — O.Sax. skiotan, O.H.G. skiogan, O.Eng. sceotan «throw, shoot».
9. Alb. Gheg. lâ, Tosk. le (*lədnō) «leave», participle lane (*lədno-) «left» — Goth., O.Eng. lētan, O.Sax. lātan, O.Icl. lāta «leave»; adjective Goth lats, O.Icl. latr «sluggish, lazy».
10. Alb. (i)lehtё «light» (IE. *legik-, *length-) — Goth. leihts, O.H.G, līhti), O.Eng leaht «light». Adjectives with a similar meaning, derived from the same root, are represented in a number of Indo-European languages, but only in Germanic and Albanese the stem of the adjective has the suffix -t-.
11. Alb. lesh «wool, fleece» — Dt. vlies, M.H.G. vlius, O.Eng fleos «sheep skin, fleece». IE *pleus- «pluck wool, feathers».
12. Alb miell, «meal» — O.H.G. melo, melaues, O.Eng. melu"meal". This is a case of complete identity in the structure of the stem and meaning.
13. Alb. mund 1) "be able, to be able to" 2) "to win, to overcome" mund "effort, hard work." These words, which in Albanska give a variety of derived formations, are usually associated with O.H.G. muntar "cheerful, alive", munt(a)rī «zeal, diligence», Goth mundrei «goal» and are assotiated with IE. *mendh "to direct thoughts, to be alive." The above Albanian words that convey the meaning of “physical strength, physical effort, victory in a fight” clearly stand out among the tribes of Indo-European formations associated with the root *men-. The development of both Germanic and Albanian meanings is well explained from the primary meaning “hand” (Gmc. mundō "arm, protection", Albanian verb mund "to be able to win" is an ancient derivative).
14. Alb. Gheg rū, rūni, Tosk. rёndёs «abomasum» — M.H.Grenne «abomasum».
15. Alb rē «cloud» – O.H.G. rouh, O.Saxrōк, O.Icl. reykr «smoke», Gmc. *гauki.
16. Alb. shparr (*sparno-) — a kind of oak (Quercus conferta) — O.H.G., O.Sax, sparro, M.H.G. sparre «log, beam, rafter», O.Icl. spari, sparri «log, beam» (common Gmc. *spar(r)an), O.H.G. sper, O.Eng. spere, O.Icl. sparr «a spear».
17. Alb. shpreh «speak, express» (*spreg-sk-) – O.H.G. sprehhan, O.Sax., OE. sprēcan «speak».
18. Alb lapё, lapēr "hanging, flabby piece of skin; belt, inedible piece of meat; skin hanging on the neck of an ox; flap" — O.H.G. lappo, lappa «hanging piece of leather, fabric», O.Sax., lapp «skirt, flap of klothing», Ger. Lappen «flap».
19. Alb flakё «flame», flakёroj "flicker, blaze", flakoj «flash» – M.H.G. vlackern, Ger. flackern "to tremble (about a flame), to flicker, to blaze", O.Eng. flacor «flying», cf. Eng. flakeren «flit». Here we have a vivid case of analogy in expressive word creation, which has given such a similar formation in all respects that it can hardly be considered an accidental analogy. Probably, these similar formations arose in an atmosphere of territorial contacts of the ancient Germanic and Thracian tribes. This also applies to the following group of words:
20. Alb. flater, fletё, -a «wing», flatroj «flit», flutur «butterfly», fluturoj, fluroj «fly, flit». Cf. Ger. flattern «flit», Eng. flutter, flitter "flit, flutter their wings". There are options with voiced intervocalblastic: Erly Ger. vladern, cf. Fledermaus (Eng. flittermouse) "microbat". Expressive onomatopoeia, violating the laws of sound correspondences, led to the creation of strikingly similar lexical units in the now territorially distant from each other, but had been contacting at some time (in the prehistoric era), the Germanic and Albanian languages . The corresponding formations are widespread in both the Germanic and Albanian dialects, which indicates their antiquity.
In the field of grammar, the connection of the Albanese language with the Germanic are manifested in the preservation of some specific features of the Indo-European inflections, in the types of morphological structures, in the development of the phonological structure:
In the development of alternations of vowels and their morphological functioning, the Germanic and Albanian languages are very similar. The similarity is not limited to the remnants of the Indo-European quantitative ablaut in some verb paradigms… This similarity is especially striking in the chronologically later sound alterations of assimilative nature, which are very common in the Albanese language and represent, in their morphological use, a complete analogy to similar phenomena of Germanic languages. It can be noted that it looks brighter and more convincing than the similarity of the processes of assimilative variation in the Germanic and Celtic languages, noted by the authors of the Comparative Grammar of the Germanic Languages as one of the isogloss of the Celto-Germanic area. Comparison of morphological structures of modern Albanian and Germanic, for example German, languages show close analogies in the distribution and functioning of the phenomena of internal inflection as a means of expressing grammatical meanings. In the nominal inflection of alternating vowels (more precisely, umlaut) in both language types are used in the formation of the foundations of the plural nouns, for example Alb. dash «ram»— pl. deske, natё «night» — pl. netё, cf. Ger. Gast «guest» — pl. Gäste, Nacht «night» — pl. Nächte (DESNITSKAYA A.V. 1965: 40)
The study of Baltic place names outside ethnic territories gave grounds for hypothesizing the migration of a part of the Balts to the Balkans supposedly in I millennium BC. Examples of toponymic correspondences on the territory of Romania, where the Proto-Albanians could be, are given below.
Balta Albă, a commune and village in the county (județ) Buzău – the most convincing Baltic toponym in Romania, since the double name consists of the Baltic and Romanian words having the same meaning "white" (Lith. baltas, Let. balts, Rom. alb). What was the original name, remains to be seen.
Suveica, a village in Mureş County – the village of Suviekas in Zarasai District, Lithuania. Cf. Lith. suvaikyti "drive together, round up".
Țuțora, a village in the județ Jassy – good compliance with the name can be seen in the Baltic place-names: Lake Čičirys in the northeast of Lithuania, near the village of Suviekas in the Zarasai district, the Ciecere River, rt of the Venta River in Latvia. In Ukraine, there is a village Tsitori (Ternopil Region in Ukraine).
Vârleni, a village in the județ Vâlcea – Lith. varlė "frog".
The Bulgarian scientist Duridanov Ivan, having studied the Thracian-Baltic and Dacian-Baltic language connections, found 60 convincing separate lexical correspondences between the Dacian and the Baltic languages and 16 more possible, and between Thracian and Baltic – 52 and 19 respectively. At the same time, there were only 14 common Thracian-Dacian-Baltic correspondences.(DURIDANOW IWAN. 1968: 100). These numbers may seem small, but it should be borne in mind that the vocabulary of the Thracian and Dacian languages has been preserved in very small quantity, so it is not the absolute numbers that are important, but their comparison with the data on the connections of the Thracian and Dacian languages to those close to the Baltic. First of all, we mean the Slavic languages, but no special Dacian-Slavic or Thracian-Slavic ties were found ibid. The data of Duridanov are confirmed by Proto-Albanian and Pro-Baltic connections. Especially impressive, according to Desnitskaya, are lexical correspondences that stand out for their specificity.
Alb. ligё «illness» (< Proto-Alb. *ligā), i lige «ill; bad» (< Proto-Alb. *ligas), i ligfhte «weak, powerless» (< Proto-Alb. *ligustas) – cf. Lithligà, Let. liga «illness», Lithligustas «ill»;
Alb. mal «mountain» (< Proto-Alb. *malas) – Let. mala «bank, shore»;
Alb. mot «year; weather» (< Proto-Alb. *metas) – Lith*metas «time», pl.: metai «year»;
Alb. i thjermё «grey» (< Proto-Alb. *sirmnas) – Lithsirmas, siřvas «greyй» (DESNITSKAYA A.V. 1990: 10).
Продолжительные миграции фракийцев со своей прародины к адриатическому побережью и их длительное пребывание в зоне контактов с носителями языков разных групп имели следствием редкие типологические черты албанской грамматики. Как утверждает А.Ю. Русаков, албанский язык обладает восьмью-девятью из двенадцати признаков ареала SAE (Standard Average European) и, кроме того, характеризуется также определенными типологическими схождениями с иранскими, балтийскими и тюркскими языками. (Русаков А.Ю. 2004, 259-274). Все это затрудняло установление генетических связей албанского языка с другими индоевропейскими.
Now we can find explanation for the fact why the Albanian language has so many common words with the Germanic and Baltic languages (see data on the number of common words in the Albanian and other Indo-European languages above) in the assumption that ancient speakers of Proto-Albanian language settled in close proximity to the settlements of the Germanic and Baltic tribes.
Such complex migration route of the Proto-Albanians and their long stay in the zone of contact with speakers of different language groups explains the rare typological features of the Albanian grammar. According to A. Rusakov, Albanian language has eight or nine of the twelve features of the space SAE (Standard Average European) and, in addition, is also characterized by certain typological correspondences with the Iranian, Baltic, and Turkic languages. (RUSAKOV A., 2004, 259-274). It will be shown further where and when the Proto-Albanians (Thracians) had contacts to explain to a certain extent, these correspondences.
Now, knowing the origin of the Albanian language from Thracian, we can find an explanation for the fact that Albanian has so many common words in German and Baltic languages (see Data on the number of common words in Albanian with other Indo-European languages above) – Thracians have long been in close proximity to the settlements of the Proto-Germanic and Proro-Baltic peoplea. Their difficult migration path and long stay in the area of contact with native speakers of other groups explain the rare typological features of Albanian grammar.
More detailed about migration of Indo-European peoples in the section
The First Great Migration