Although numerous scholars voiced their criticism about V.I. Abaev's postulation of Iranian-linguality of the Scythians, none went as far and as methodically as G. Dremin. G. Dremin reviewed the V.I. Abaev's methods of research, assumptions, origin of his lexical material, timeframe of his lexical material, his principle in choosing his linguistic material, type of material he ignored, his reason for ignoring major evidence. At the same time, G. Dremin left out details of purely linguistic manipulations, the fundamentally absurd idea of using onomasticon to "reconstruct" a language, replacement and transposition of phonemes in order to adjust the source to the conclusion, use of random words from random languages as ostensible cognates for targeted stems, use of uncalibrated constants in fictitious chronology, and a statistical probability of false isoglosses in random languages. V.Abaev chose to ignore most of the real Scythian words that inconvenienced him, they clipped his fantasy by having a real translation from the Scythian; they did not suit his criteria of selecting only those words that could be matched with any random roots from his selection of multitude hypothetical IE relatives. In the end, G. Dremin came to the same conclusion that was suggested by the generations of critics: the V.I. Abaev's Scytho-Iranian paradigm is nothing but a hoax.

Posting notes are highlighted in blue and in blue boxes.

Introduction

The famous Russian linguist and expert on Iran Vasily Ivanovich Abaev (1899-2001) in the first volume of his book "Fundamentals of Iranian Linguistics", 1979, included an essay titled "Scythian-Sarmatian vernacular". ¹ A central place in the essay holds a "Dictionary of Scythian words" with more than two hundred lexemes, recreated by Abaev by reconstructing preserved "Scythian-Sarmatian" lexical material. Actually, the essay is as a sort of comment to the dictionary, where the author explained methods he used in reconstructing lexis of extinct language, and in addition made a number of conclusions on the phonetics and morphology of the Scythian language.

Perusing this V.I. Abaev's work, one's attention is drawn to some inconsistency in presentations of the material. So, in the essay on the "Scythian-Sarmatian vernacular", this phrase is found only in the work introduction. In the main part all author's arguments pertain to the Scythian language, and the Scythian-Sarmatian vernaculars are not mentioned. V.I. Abaev titled his Dictionary Scythian, and the main conclusions also refer exclusively to the Scythian language. So he writes: "Our essay on certain phenomena of the Scythian language is very far from what can be described as a description of the language. It gives a no more complete picture of the entire system of the Scythian speech than few surviving fragments can give an idea of an ancient statue. Yet however, in every section of our essay - in vocabulary, phonetics, word formation - we derived a certain sum of positive, solid and indisputable data... These data describe the Scythian language as an Iranian language with unique features and well-defined individuality". Such discrepancy between the title and content prompted to examine the work of the renowned Iranist scholar more thoroughly and critically apprehend the material exhibited in the essay.

Vsevolod Fedorovich. Miller (1848-1913), a prominent Russian linguist and ethnographer, was the first to draw attention to a genetic link of the Ossetian language with dialects spoken by the Sarmatian tribes that once lived in the Azov steppes in the southern Russia. Miller made several trips to Ossetia, where he studied Ossetian language, and recorded a number of Ossetian legends. With the acquired knowledge later Miller was able to explain a significant part of non-Greek names in the ancient epigraphic inscriptions of the Northern Pontic. Following publication in 1887 of a series of his works called "Ossetian etudes", many folks realized that the Ossetian language is the tool which can solve many of the problems of the historical linguistics related to the ancient Iranian tribes in the Azov steppes.

Ethnically Ossetian, V.I. Abayev, being an expert in the native language and in addition a specialist in the ancient Iranian languages, took upon this idea and apparently this idea became his guiding star throughout his laborious life. One of central themes in his work was a thorough identification of the Ossetian language linguistic connection with the language of the Alans, and via them - with the Sarmatian vernaculars, and even deeper - with the language of the Scythians. The Scythian-Ossetian parallels became an idee-fixe of the linguist and historian Abaev, his forte, his passion. The strongest Abaev's aspiration as a scientist was to prove a direct continuity between the Ossetian and Scythian languages. Not by chance Abaev concludes his essay with an energetic statement: "Of all the Iranian languages, the Scythian language is closest to the Ossetian. This proximity is so bright and marked by such features of continuity that the Scythian and Ossetian languages can be viewed as two developmental stages of the same language".

Abaev's Dictionary

Appendix A reproduces a list of "Scythian" words from the Abaev's dictionary. There, for every "Scythian" words are given stems used for reconstruction of the word. Most of stems are "Iranian" names preserved in epigraphic inscriptions found in the ancient Pontic cities. At the time, the American linguist M. Swadesh compiled a list of 207 words most resistant to change. In linguistics, this list is used to determine the degree of relatedness between languages. The number of "Scythian" lexemes reconstructed by Abaev is  very similar to the number of lexemes Swadesh included in his list. In this context, it is interesting to analyze how "basic" is the range of terms that Abaev reconstructed of the original "Scythian-Sarmatian" lexical material. It turns out that the "Dictionary of Scythian words" contains 30% of the base lexemes from the Swadesh list. This means that the Abaev's dictionary adequately reflects most stable part in the vocabulary of the studied ethnos and in principle it is possible from it to make ethnic identification of the reconstructed language's speakers.

The ethnic affiliation of the speakers' lexicon reconstructed by Abaev most unequivocally could be found in the ethnonyms used to identify themselves or to designate neighboring peoples. The Abaev's dictionary found only two such ethnonyms: aryana "Aryan" and as "Ases". The first term indicates belonging of the Abaev's lexicon speakers to a wide range of Euro-Asian Aryan tribes, and the second term connects them with the Alans and the Roksalan tribe. Indeed, according to the Ibn-Rust testimony, Roksalan were also called ruhs-as, meaning Noble (Light) Ases. The name Ases is closely connected with the early history of the Alans, and the origin of Ases is connected with the Scythian-Asian or Massagetan world.

A number of terms in the Abaev's dictionary pertain to religious and ideological persuasions of the speakers of the reconstructed "Scythian-Sarmatian" lexicon. To these terms be attributed the words arta "deity", baga "god", od "soul", spanta "sacred", farna "heavenly grace", sugda "pure, holy". On the word sugda Abaev noted that it was likely was brought from Middle Asia, from the Sogdiana area. Vasmer derived this word from Old Pers. suxta "purified by fire". And this indicates that the people who were using this term were followers of Zoroaster, who believed in cleansing by fire. As is known, the European Scythians did not belong in the circle of such followers.

Another important group of terms allows to characterize the social structure of the society where lived the speakers of the Abaev's lexicon as transitional from tribal to the slave society. This is indicated by such words as ardar "gentleman", pathaka "leader", zantu "tribe", cagar "slave", dar "hold", zaranya "gold", aspaina "iron". The Abaev's dictionary contains considerable number of terms conveying various relations. Here are words like pita "father", mata "mother", pu ra "son", brata "brother" nar "husband", os "wife". This indicates that the family continued playing an important role in the studied community.

Abundance in the Abaev's dictionary of the military terms, such as spada "army", bala "military squad", rasma "detachment", spar "attack", xsathra "valor", "courage", skunx "feats", arsti "spear" karta "sword", druna "bow", tells that the people who were using these words very often were in a state of war, and were truly militia-people.

Terms associated with cattle breeding in the Abaev's dictionary are encountered much more than the agricultural terms, which radily attests his "Scythians-Sarmatians" as nomadic herders, but they were also familiar with the concepts such as "arable land", "millet", "seed" "vine". The Abaev's "Scythians-Sarmatians" raised animals like aspa "horse", gau "cow", pasu "sheep", kuti "dog", hara "donkey". About the Abaev's "Scythians-Sarmatians" acquaintance with donkeys is suitable to recall the Herodotus' famous story about a strong effect the bellowing of the asses inflicted on the European Scythians' horses whenever Scythians tried to attack the King Darius camp. This example shows that the reconstructed Abaev's lexicon has more to do with the world of the Asian rather than the European Scythians. And the animals which the "Scythians-Sarmatians" encountered during hunts were found both in Europe and in Asia, they are arsa "bear", saka "deer", varka "wolf", varaza "boar".

To understand the substance of the method that Abaev used for reconstruction of the "Scythian" words, consider a few examples. So, Abaev expected that to designate a deity personifying light and truth, Scythians used the word arta. In his anticipation researcher proceeded from his knowledge that exactly the same word was used in the Old Persian language, where it meant "belonging to the truth". In addition, the Ossetians word ard means an oath or a deity to swear with. As evidence that Scythians used the word arta, Abaev cites several names known from epigraphic and written sources. For example, he mentions the name Αρδαγδακος (CBI No 1279) ², carved on a marble slab found in the Tanais ruins. Abaev believes that the name consists of two words - ard and agdak. The first word is a modified word arta, and the second word in the Sogdian language means "covenant". So, ardagdak in "Scythian-Sarmatian" language can mean "one who follows custom". Another name from Tanais is Οδιαρδος (CBI No 1279), consisting of two words odi and ard, according to Abaev it denoted a person "whose soul is filled with light deity". In addition, citing Theophanes Malala, Abayev quotes a name of a Byzantine originally Alan general Αρδαβουριος, whose name with a help of the Ossetian term ard-bewræ he explains as "endowed with light and truth".

Abaev presumed that the Scythian "sky" was the word abra. In his assumption he proceeded from his knowledge that in the Avesta language was preserved the word awra - "cloud", and in the ancient Indian language the same word was pronounced abhra. In addition, Ossetians have a word arv meaning "sky". As a proof that Scythians used the word abra, Abaev cites several names known from epigraphic sources. So, he cites names Αβραγος and Αβροαγος, repeatedly found in the Olbia inscriptions (IOSPE I² No 80, 82, 91, 98, 100) ³. In addition, on a marble slab recovered from the Tanais ruins can be read a name Αβροθεος (CBI No 1279).

For third example, consider the reasons for inclusion in the dictionary for the Scythian word hvar. In Abaev's assumption, this word in Scythian meant "sun" and he believed that this word Scythians borrowed without a changed from the language known from the Avesta texts. His hunch was reinforced by a similar word xor at Ossetians. As a proof of his assumption, Abaev points to two names numerously found in the Tanais epigraphic inscriptions. The first of these names is Χοροαθος or Horoath (CBI No 1245) can be read on a marble slab dating from 220 AD. Vasmer read it as a "messenger of the Sun", based on the Avesta word hvar - sun, and the Ossetian word vak - messenger. The second name is Χοφρασμος or Hofrazm (CBI Nos 1245, 1246, 1248, 1250, 1252, 1278, 1279) is found in the Tanais inscriptions at least seven times.  Based on the Ossetian xor-fazm, Abaev read "Sun-like".

Finally, examine why Abaev believed that in the Scythian language the word "father" was pronounced as pita. His hunch was prompted by the Avestan language with the word "father" as  pitar, and the Ossetian form fidœ. His suspicion of the Scythians using a close form he rationalized by referring to the names read on the epigraphs found in different Pontic cities. So, he cites Πιδανος (IOSPE I² Nos 26, 103, IO No 85) from the inscriptions of Tiras, Πιδος and Πιδεις (IOSPE I² Nos 98, 111, 135) from Olbia, Πιτφαρνακης (CBI No 1278) from Tanais. About the origin of these names Vasmer once concluded that they are derived from the words fidœ, fidon, which in Ossetian mean "father", "father's".

The above examples trace all primary techniques Abaev used to reconstruct the extinct forms of the "Scythian" vocabulary. First of all, he explored etymology of historically attested name, ethnonym or toponym, and for his reconstructions drew only the linguistic material that with sufficient justification could be derived from Indo-Iranian stems. Having substantiated the "Iranian" origin of the word's stem, Abaev tried to find its analogue in the Ossetian language. With two forms of a lexeme, an ancient Indo-Iranian and a modern Ossetian, Abaev entered into his dictionary a form that seemed to match the most ancient Scythian form.

As for the quotes in the words "Iran", "Iranian-speaking" applied to the Scythian and Sarmatian lexicon it is appropriate to state the following comment. According to the ensconced tradition, the Scythians, Sakas and Massagetae that inhabited the space of the Greater Scythia are held as Iranian-speaking. Following a strange logic, the original native speakers of the common parental language that remained in their ancestral territory were named after their new homeland, where had settled a fraction of the people. Can be cited the opinion of one of the founders and leaders of the Eurasian movement, P.N. Savitsky ⁴: "It should be noted how uncomfortable, under conventional terminoligy, to call these or that steppe peoples "Iranians", even if these people were ethnically related to the Medes and Persians. Either the geographical Iran needs to be renamed with a new name, or should be invented a special designation for these nomadic people, for example, they may be called the "steppe Aryans".

Origin and principle in the selection of linguistic material

What is the origin of linguistic material based on which was performed reconstruction of the "Scythian" lexicon? Abaev writes on this subject that "elements of the Scythian-Sarmatian lexicon are found in their proper names, toponyms, and tribal names that were preserved in many historical, geographical and epigraphic sources".

Indeed, the efforts of several generations of researchers saved a huge epigraphic volume from the reading and decoding of many tombstones and dedicatory inscriptions found on the sites of the Greek Pontic cities. A significant part of these inscriptions was deciphered, read, and systematized. Information about them can for example be found in references 2, 3, and also in the book "Inscriptions of Olbia" ⁵.

These inscriptions mention many names that can be read on the tombstones, in dedicatory inscriptions, and in the citizen lists which Greeks regularly engraved on marble and limestone stelae and slabs for a variety of reasons. Due to such their tendency a lot of names of the people that used to live in these cities have reached us. Of course, most of the surviving names belonged to the Hellenes. But among them are also found clearly non-Greek names. The origin of the non-Greek names may be most diverse. It is known that the Pontic coastal cities were established mostly by the Ionian Greeks from the coasts of Asia Minor. Therefore, along with the Hellenes, to the Pontic cities were coming people from the Asia Minor connected with them. The existence in the Greek cities of various alien diasporas can't be excluded. In addition, the developing colonies interacted with the local Pontic tribes, and their members were gradually incorporated into the urban population.

The most noticeable fraction of the total material volume plucked by Abaev was a list of 284 names from the Pontic inscriptions of the "Iranian" origin. A total of 87 names on the list come from different cities of the Bosporan Kingdom, a separate group consists of 114 names attested in the Tanais inscriptions, and 83 other names belong to the Olbia epigraphics. The names from Tanais constitute a separate group, because although Tanais was founded by the Bosporan Kingdom Greeks, it was much later than the other Bosporan cities, and it was located away from the Cimmerian Bosporus, on the opposite bank of Maeotis. In addition, the barbarian neighborhood of Tanais was different than in the area around Panticapaeum due to its remoteness from the Bosporus.

Another source that Abaev drew lexical material for his reconstructions were the works of ancient historians and writers. Among about a hundred words taken from written sources are more than fifty named of the kings, chiefs and commanders, about thirty words convey names of the tribes in the Scythian-Sarmatian world, more than twenty words preserved various geographical names, including names of the rivers and water bodies. Among written sources used by Abayev, above all are the "History" of Herodotus, from which were used 20 words, the Arrian writings (14 words), and Strabo "Geography" (12 words).

It is interesting to analyze, to which of the ancient languages "leans" the linguistic material taken by Abaev for the basis for his reconstruction of the "Scythian" dictionary. Altogether, he used about four hundred (397 to be exact) word stems. There, about three hundred words (75% of total) show cognates in the Avestan language. Approximately half of these words have analogy only in the Avestan language, and in no other ancient language. For 124 word stems (31% of the total) are cited ancient Persian and middle Persian roots, and 14 out of the 124 words have cognates solely in the Old and Middle Persian, without any roots in the ancient Indian language or in the Avesta language. The number of words with ancient Indian roots is 101, or 25% of the total number of word stems. Of these, 10 words have only ancient Indian roots and have no roots in any other ancient languages. For the majority of the word stems (for almost 80%) Abaev was able to pick cognates from the Ossetian language.

Abaev did not find cognates for a tenth of the word stems in any ancient languages - neither in Avestan nor in Old Persian, nor in the ancient Indian language. But these words have common roots with the words of the modern languages such as Ossetian, Georgian, Armenian, Kabardian, Afghan, Mongolian, Hungarian, German, Lithuanian and Russian. 26 word stems used by Abaev have cognates only in the Ossetian language, without support in any ancient or modern languages. Another 19 words have cognates in the Ossetian and other modern languages.

We see that the lexical material used by Abayev as a base for his constructions is quite heterogeneous, but mainly leans toward the Avesta language. As is known, Avesta is a collection of sacred texts in Zoroastrianism consisting of heterogeneous fragments, including liturgical hymns, various epic stories, and religious laws. These texts were composed at different times, the most ancient part of Avesta, the "songs" or gathas, according to legend were penned by the prophet Zoroaster. Linguistic analysis of the "songs" allows to attribute them to the time of composition in 12th-10th cc. BC. The Avestan language belongs to the Iranian branch of the Indo-Iranian subgroup of the Indo-European languages. Scientists note the proximity of the most ancient Avestan texts to the ancient Indian language of Rig-Veda. Some experts believe that the language of the formulated texts of the Avesta was common in Central Asia and parts of the modern Iran and Afghanistan.

The selection principle of the lexical material from the available to Abaev epigraphic and literary sources was extremely simple. He chose only those names and appellations that were easily etymologically deciphered using "Iranian" roots. The Abaev's predecessors, in an effort to find the "Iranian" traces in the northern Pontic region, actively explored non-Greek names preserved in the epigraphic inscriptions in the Pontic cities. They have shown that many of these non-Greek names are sufficiently reliably etymologized based on "Iranian" roots. That linguistic material formed the base for Abaev's reconstructions. Producing such input filtering of lexical material, Abaev from the beginning restricted himself the framework of the exclusivly "Iranian" roots of the Scythian language.

Timeframe of examined linguistic material

As is known, in the Northern Pontic the Scythian epoch lasted for only five centuries: in 7th c. BC Scythians first became prominent on the world stage, and in the 3rd c. BC the Great Scythia had ceased to exist. Abayev, while selecting lexical material he needed for reconstruction of the Scythian language, did not limit the time frame to the actual period of the Scythian history. Based on an assumption that the language of the Scythians was close or even identical to the vernaculars spoken by various tribes that received a collective name of Sarmatians, who starting at 2nd c. BC flooded in successive waves the Northern Pontic steppes, Abaev also included in the zone of his examination the vocabulary material of the Sarmatian and Alanian origin.

Thus, the base for his linguistic reconstruction he used almost all traces of the "Iranian" vernaculars that came down to our times from their existence in the Northern Pontic territory during the period from 8th c. BC to 5th c. AD. In such extended period of time that covered nearly fifteen hundred years, the Scythian epoch changed to Sarmatian era, the Sarmatian epoch changed to Alan era, the Gothic invasion was supplanted by even more devastating Hunnic invasion, and finally came the Great Migration of peoples! It is clear that the Scythian language and its replacement by the Sarmatian-Alan vernaculars in that time had to undergo serious changes. The desire to cover all the available material of the "Iranian" nature led to the condition where the remains of the Scythian language were buried under more significant layers of the Sarmatian and Alan speech.

Names from cities of Bosporus Kingdom

Let us analyze from what historical strata came different groups of word stems used by Abaev in his reconstruction. This can be done because in the reference manuals, references 2, 3 and 5, a vast majority of Greek inscriptions with "Iranian" names is supplied with reliable dating. Of epigraphic monuments found in the cities of Bosporus Kingdom, Abaev used a total of 87 names with the "Iranian" roots. The diagram in Figure 1 shows that the majority of the "Iranian" names used by Abaev of the Bosporus origin, is taken from the monuments dated by the first three centuries of our era. The temporal distribution of barbarian names illustrates the notable saying by V.F. Gaydukevich, ⁶ who noted that "the non-Greek names of the Bosporus inhabitants are much more numerous in the Roman period than in the preceding period". From the Bosporus history is known ⁷ that Bosporan relations with Scythians started deteriorate in 310 BC. That year died the Bosporan king Perisad I, and the throne was claimed by two princes: Eumel and Satyr.

Fig. 1 - "Iranian" names from the Bosporus cities

A fierce struggle between two contenders led to a crisis in the Bosporus and the beginning of the Civil War. Eumel established relations with Siraks - a Sarmatian tribe that has recently appeared in the Northern Pontic. And Satyr relied on his tested old allies, the Scythians. Of course, Sarmatians and Scythians fought in that war not only for the interests of the princes, but also for their own, since the Sarmatian invasion in the area previously owned by the Scythians made them bitter enemies. The war was won by Eumel, who was well aware that he gained the throne only through the help of the Sarmatians, and therefrom in his further policy he relied on the alliance with them. Beginning with the Eumel's reign , the traditional alliance with the Scythians was broken, and the Bosporan policy reoriented to the Sarmatians.

It can't be excluded that a part of the "Iranian" names on the Bosporus monuments of the 4th-3rd cc. BC belonged to the Scythians. B.D. Blavatsky ⁸, who studied ethnic composition of the Panticapaeum population at that era, noted that five gravestone inscriptions of 4th and one inscription of 3rd c. contained Iranian names. He reasonably considered that these names could belong to the people from the tribes neighboring Bosporus, including the Scythians. In his essay, Abaev mentions only two names of the six that B.D. Blavatsky unequivocally considered to be "Iranian" (Allusion that Abaev could not come up with "Iranian" etymology for the other four names).

As B.D. Blavatsky noted, some 4th c. BC  kurgan burials of the Panticapaeum necropolis suggest that there were buried noble Scythians. The most a shining example in that line is the Kul-Oba burial kurgan near Kerch with interred Scythian chief. But from the second half of the 3rd c. BC the nature of the barbarian burial in the Bosporus changes As shows the archaeological evidence from that period, kurgans with burials of barbarian nobility were only located on the Asian side of the Bosporan Kingdom, closer to the Sarmatian nomadic tribes. As noted by Yu Vinogradov ⁹, the cultural remains of these burials display vivid Sarmatian features, primarily associated with the manner of decorating battle steeds with round silver plates, called falars (cheekpiece) by the specialists (Greek word).

Indirect indicator that Scythians were displaced from the political life of the Bosporus in 3rd-2nd cc. BC is the peace treaty of 179, concluded by the Bosporan king Pharnaces I with Pergamum, Bithynia and Cappadocia. Guarantors of that treaty, along with Chersonesus, were named Sarmatians led by a king Gatal. According to the archaeological sites, the Sarmatians' penetration of the Bosporan Kingdom was peaceful. On the one hand, during this period absent are any traces of destruction, and on the other hand at that time in the Bosporus appear numerous Sarmatian inscriptions with many names, and the local population begins to widely use Sarmat utensils. As noted by D.P. Kallistov ¹⁰, the "indigenous names as a rule start appearing only in the inscriptions of the 2nd-1st cc. BC. They become prominent only in epigraphic texts dating by the centuries of our era".

Our brief excurse into history of the Scythian-Sarmatian-Bosporan relations clearly indicates the illegitimacy of the use of "Iranian" names from the Bosporan epigraphy for reconstruction of the Scythian language. The overwhelming majority of these names dates from the time when the Scythians were already almost completely ousted from the political life of the Bosporan cities by the Sarmatians. The "Iranian-speaking" element begins to actively penetrate the Bosporus cities three centuries after the crisis in the Scythian-Bosporan relations. The "Iranian" names from Panticapaeum, Gorgippia, Phanagoria have solely Sarmatian origin. They can be used for the reconstruction of the Sarmatian vernaculars, but not for the features of the Scythian language.

Names from Tanais

The following figure shows a graph reflecting the change in time of the "Iranian" names in city of Tanais. Distribution for the Tanais' "Iranians" turned out to be narrow and intense. Virtually the entire chart fits in a 150-year span, from 100 to 250 AD, and on the first half of the 3rd c. fall 130 inscriptions - more than the total number of inscriptions with the "Iranian" names found in all other Bosporus cities. The distribution for Tanais of the personalities with "Iranian" names reflects a relatively short but turbulent life of the city. As is known, Tanais, an extreme north-eastern point of the ancient world, was founded at the beginning of the 3rd c. BC. Then it was destroyed, rebuilt and reached its peak in the first half of the 3rd c. of the new era. Just at that time, according to our diagram, in the city dramatically increased the number of people with names that sounded like "Iranian".

Fig. 2 - "Iranian" names from the city Tanais

A renowned archaeologist and historian D.B. Shelov ¹¹, analyzing the ratio of the Greek and Iranian names in the Tanais onomastics of the 2nd-3rd cc., noted a gradual barbarization of the Tanais population at that time. D.B. Shelov noted that of 130 "Iranian" names of the Tanais inscriptions that have been documented in the first half of the 3rd c. of our era, only 13 have correspondences with the names of people in other Pontic cities. He wrote ¹²: "Those Iranian names that first appear in the Tanais inscriptions at the end of the 2nd or first half of the 3rd c. AD and are not repeated in any other epigraphic monuments of the Pontic, almost all belong to the same tight group. This phenomenon can only be explained by the fact that the new names are brought to the city by a new ethnic group who joined the Tanais population". Excavations of the Tanais necropolis showed that in the 2nd c. of our era in the funeral rite appear new features, previously not documented here, but typical of the Lower Itil (Volga) region Sarmatian culture. D.B. Shelov feels that most of the "Iranian" names of Tanais inscriptions in the first half of 3rd c. of our era belonged to "a relatively small and closed, most likely Alanian ethnic group". The consonance of one name belonging to this tribal group - Αμαρδιακος with the name of the tribe Αμαρδοι, know from Strabo, allowed D.B. Shelov to suggest that this ethnic group was previously localized somewhere between the Caspian and Aral seas.

At the turn of the 4th-3rd cc. BC, the Great Scythia disintegrated, and first Yazygs and then Roxolans began penetrating gradually the steppe lands between Don and Dnieper rivers. By the time of the city foundation at the mouth of Don, territory along lower course of Don was controlled by the Sarmatian tribe Roxolan. Apparently, members of this tribe gradually involved into the city life, by the end of the 2nd c. of our era they became its full-fledged citizens, and even held leadership positions in the city government. For example, on a limestone slab found in the ancient city Tanais, was read a following inscription, dating back to 188 BC: "Didimoksarth son of Hoden, Archon of Tanaites, having rebuilt the tower collapsing form time, restored it to the emporia (emporium ~ trading post, factoria)" (CBI, No 1242).

The "Iranian" names from Tanais make up the largest group of word stems in the material Abaev used to reconstruct lexicon of the "Scythian" language. Meanwhile, analysis of the inscriptions' dates shows that by the time when Tanais was inhabited by the people with such names, at least three centuries passed after the end of the Scythian period at the Azov Sea. This means that Didimoksarth son of Hoden, who became archon of the city founded by the Greeks, probably was not a Scythian. More likely, he was from one of the noble barbarian clans that at that time controlled adjacent to the Tanais territory. Obviously, to use the Tanais "Iranian" names for reconstruction of the lexicon of the Scythian language is not quite correctly. In this case does not matter whether Didimoksarth was a Roxolan, Amard, or Alan, it is important that he belonged to the ethnicity different from the Scythian.

Names from Olbia

Olbia had a turning point that strongly influenced life of the Greek city in the Northern Pontic, and greatly altered ethnic composition of its population. In the middle of the 1st c. BC Olbia fell under the onslaught of the powerful Gets, led by the king Burebista. The city was almost completely destroyed, its inhabitants fled, and all vital functions stopped for a few decades. However, by the end of the 1st c. BC Olbia gradually revived, and its population has rebounded, partly by returning former residents, and partly due to the inflow of the alien barbarian population.

Literature on the ethnic composition of the Olbia population in the aftermath of the Get's devastation usually refers to the testimony of Dion Chrysostomos, a Greek writer, philosopher and historian of the Roman period, who visited Olbia in 83 AD. In his Boristhen speech he spoke of the "crowds of barbarians", who flooded the city rebuilt after the Getic invasion invasion. Analysis of the names in the Olbia inscriptions of the first centuries AD indicates a presence of a large number of non-Greek names, which confirms the statement made by Dion Chrysostomos.

To somehow characterize the participation of the barbarian element in the life of Olbia during the Scythian age, turn to two publications on this topic. In the T.N. Knipovich work ¹³ the ethnic analysis of the Olbia population in the pre-Getic period (6th-1st c. BC) was carried out based on the material available at the time of writing (1956). At the disposal of the historian were 45 stone inscriptions, 7 lead plates with lists of names, and 16 inscriptions on pottery. T.N. Knipovich found that the Olbia names in the pre-Getic period ethnically were strikingly homogeneous. Of the 255 names contained in the list of Olbiapolites, she could count only four non-Greek names. On the origin of two of them T.N. Knipovich did not say anything definite, one name she considered a clearly Thracian, and another a clearly Scythian.

The base for a later study, performed by Yu.G. Vinogradov ¹⁴, were all known by 1981 inscriptions from Berezan, Olbia and its chora (countryside) on stone, ceramics, metal, and coins, dating to the historical period from 6th to 1st cc. BC. In the epigraphic material Yu.G. Vinogradov counted 19 barbarian names, with more than half of them belonging to the 5th c. BC. Six names of the all barbarian names the researcher confidently identified as Scythian names, and one as a Persian name.

Of course, the 19 barbarian names that have accumulated in the Olbia epigraphy for six long centuries pale in comparison with the hundreds of Sarmatian names that appeared in the first century of the Christian Era. The latest studies ^{15, 16,17,18} have shown that in the first century AD Aorses had Olbia as a military protectorate, and that their kings Pharzoi and Inismei in 50-80s minted its own coins in Olbia. Abaev in his work mentions 83 "Iranian" name of Olbia origin. The timing of these "Iranian" Olbian names is shown in Figure 3. In form it is similar to the distribution that we obtained for the Bosporus cities. Unfortunately, the dating of many Olbia inscriptions is hindered by the lack of reliable time markers, such as for the Bosporus inscriptions are the regnal years of local kings. Therefore, not all writings of the 83 names were dated. Therefore, the diagram Fig. 3 reflects only part of the Olbia names mentioned by Abaev. However, all of these 83 names, including the undated names and the names reflected in the chart, belong to the first centuries of the Christian era. This becomes clear on comparing the list of names used by Abaev with a list of Yu.G. Vinogradov, these lists do not contain a single common name.

Fig. 3 - "Iranian"names from Olbia

Could at least some of the 83's names mentioned in the Abaev's essay belong to the Scythians? This can not be ruled out. However, an analysis of the military-political situation prevailing in the Northern Pontic in the early centuries of the Christian era leaves no hope for that quantity of names to be significant in any way. As noted T.N. Knipovich, the epigraphic sources, and from such sources come all 83 names in the Abaev's list, may reflect a position of only one social group, namely the ruling elite of the Olbia society. As is known, the power of the Scythian Kingdom was seriously damaged during the war with Diophantus in 109 - 107 BC. After their defeat in the war, Scythians stopped playing a leading role in the history of the Northern Pontic. Gradually, Sarmatians took their place on the center stage of history. The latest minted Olbia coins of the Scythian kings were the Skilur coins. The Olbiapolites, using the Scythian struggle with Chersoneses, freed from the Scythian protectorate. By the beginning of the 2nd century BC Sarmatians finally stabilized in the region, and then in the history of the Northern Pontic began the Sarmatian Epoch ¹⁹.

Analysis of barbarian names' dating on Northern Pontic epigraphic monuments showed that in contrast with the later Sarmatian era, in the early Scythian epoch such names are very rare. The reason for this of course rests not with the weak economic ties of the Pontic Scythians with the Greeks. One reason for the almost complete absence of Scythian names in the Greek epigraphic monuments could be a special religious prohibition, imposed by the Scythian priests on the possibility of burial a Scyth in the Greek rite with a grave slab bearing the name of the deceased. A.M. Hazanov ²⁰ wrote about a special role of the priestly class in the life of the Scythian society. Herodotus' stories about the murder of the Scythian prince Anacharsis, and later of the king Skil murder on religious grounds hint to the religious vision of the Scythians had very serious restrictions and limitations.

Scythian and Sarmatian - are they separate languages or dialects?

Above it was shown that an overwhelming number of epigraphic monuments that served as Abaev's source of the "Iranian" names appeared in the first three centuries AD. The military-political situation in the Northern Pontic at that era has developed in such a way that the bearers of the "Iranian" names are more likely to be from various Sarmatian tribes than the Scythians, who by that time were completely displaced from the historic proscenium onto back plane. That is why the use of this kind of linguistic material for reconstruction of the Scythian loom as highly flawed. Even if the Scythian language and Sarmatian vernaculars were related, reconstruction of more archaic form of the language based only on a much later related languages can not be done.

In the introductory part of his essay, Abaev uses the term "Scythian-Sarmatian vernaculars" and explains it as a common name for all Iranian dialects and vernaculars that existed in the territory of the northern Pontic region during the period of the 8th-7th cc. BC to 4th-5th cc. BC. Thus he combines the Scythian, Sarmatian, and Alanian languages, attested in the Pontic at that time, into some artificial linguistic association: "Speaking of the Iranian language of the Northern Pontic, we take for granted that that speech was fragmented into many varieties. But we are convicted that they had a number of common features that let viewing all of the Scythian-Sarmatian vernaculars as linguistically undivided".

Aware of the weakness of the evidence base, which could support such a statement, Abaev tried to reinforce his position, citing opinions of prominent linguists, who studied this problem at various times. However, the authoritative linguists do not give a clear answer to the question whether the Scythian language was a separate language in respect to the Sarmatian vernaculars or Alanian language. The fact is that all of their constructions base on a rather limited data - all in the same barbarian names found in Greek inscriptions from the North Pontic cities. In addition, they all turned to that material to justify their individual linguistic concepts.

So Vs.F. Miller ²¹ once defended the idea that the Ossetian language belongs to the Iranian group of the Indo-European language family. He studied the ancient inscriptions of the Northern Pontic in order to prove that the Ossetian ancestors belonged to that group of the Iranian tribes who once lived in the steppes between the Pont and Meotida. He discovered that most of the names with the "Iranian" roots from the northern Pontic and, first of all, Tanais inscriptions have Sarmatian-Alan affiliation. Although Miller was distinguishing Scythian and Sarmatian onomastic material, he still considered these languages as equal linguistic ancestors of the Ossetian language.

M. Vasmer ²² turned to the Scythian and Sarmatian names, studying the problem of Iranian loanwords in Old Church Slavonic and Russian. According to Janos Harmatta, Vasmer demonstrated a cautious and prudent approach, carefully differentiating Scythian and Sarmatian elements in the studied onomastic material. He definitively separated the Scythian language from Sarmatian, writing separate chapters for Scythian and Sarmatian language in his research paper. However, at the time of the publication of his work Vasmer's attempts to treat Scythian language separately from the Sarmatian language was perceived unfavorably by the scientific community.

L. Zgusta in his work ²³ analyzed all the known to him personal names of non-Greek origin, found on the Pontic epigraphic monuments. He sorted these names into the following groups: Avestan, Old Persian, Middle Persian, Ossetian, New Persian, Thracian, Asia Minor's, and names of unknown origin. As is easy to note, classification of the "Iranian" names Zgusta performed in accordance with the classification of the documented written languages, he sorted these names not by the ethnicity, but by phonetic matching with this or that ancient Iranian languages. In doing so, Zgusta was able to identify from the total body of the N.Pontic "Iranian" names two groups. In the first group he included names with phonetic nature closer to the Old Persian and Avestan language, and in the second group he included names with features of the Middle Persian phonetics. After studying geographical distribution of these names, Zgusta found that more ancient Avestan names were evenly distributed throughout the Northern Pontic, while the later Middle Persian names mostly concentrated in the east.

According to Zgusta, all examined names belonged to two dialects of a single language, and not to separate languages. His such conclusion does not seem particularly convincing, for the reason that the division of the N.Pontic names into two groups Zgusta made based on phonetic similarity of examined names with the phonetic features of two separate languages, so far no one called the Avestan and Middle Persian languages the dialects of the same language.

Names with more archaic, Avestan phonetics Zgusta connected with the Scythians, while the names of the second group, more common in the Tanais and Cimmerian Bosporus. with the Sarmatians. Such ethnic identification of the isolated by Zgusta two groups of the "Iranian" names contradicts the obvious historical facts. First, as is well known, the spread of the Sarmatian tribes absolutely was not limited to the eastern parts of the Northern Pontic region; various written sources testify of repeated Sarmatian attempts penetrate beyond Danube already in the first centuries of our era. Second, the analysis of the dated epigraphic monuments shows that a vast number of the names attributed by Zgusta to the Scythians appeared in the Sarmatian era, that is at a time when into the ruling elite of society easier could penetrate members of the Sarmatian tribes than of the Scythians. It would be more correctly to consider two groups of the "Iranian" names identified by Zgusta as belonging to two groups of the Sarmatian tribes, independently settling in the N.Pontic territory.

Janos Harmatta ²⁴ in his work criticized concept of building language classification in the Indo-European family of languages following a family tree principle. He noted that a vast majority of researchers involved in the study of ancient Iranian languages were captives of idea that that linguistic community developed following a principle of the tree structure. It was mainly manifested in the abnormality that the earlier stage of the language was studied, more reluctantly they recognized even a small degree of linguistic differences. In particular, one of the results of that approach was that these researchers saw the Scythian and Sarmatian languages as a single language.

Analyzing the non-Greek names from the Northern Pontic cities distinguished by "Iranian" articulation, J. Harmatta showed that Sarmatian language was not a single language, but was divided into separate vernaculars. He wrote: "The study of the Pontic Greeks inscriptions and Iranian names preserved in the northern Pontic has clearly shown that the language of the Iranian tribes in the first centuries of the new era in the steppe zone of Eastern Europe was not undifferentiated and homogeneous language. Phonetic differences displayed in the names indicate that these tribes spoke several different vernaculars, apparently in accordance with the tribal divisions". Janos Harmatta confidently distinguished at least four languages or vernaculars. He showed that from the fact that the "Iranian tribes in southern Russia in the first century AD spoke several languages or dialects, clearly distinguishable one from another, follow important implications for explaining the relationship between the Sarmatian, Alanian, and Ossetian languages". According to Harmatta, the idea of direct historical continuity of the Scythian, Sarmatian, Alan, and modern Ossetian language is incorrect. Because the names attributed to the Alans and other Sarmatian tribes indicate an existence of several dialects, it is obvious that the language of Sarmats or Alans on the whole can not be regarded as an ancestor of the Ossetian language. The "Iranian" names of the Northern Pontic clearly indicate that Ossetian language emerged as a result of development of only one dialectal group. Equally, this statement applies to the degree of kinship between the Scythian and Sarmatian languages. Already divided into dialects by the beginning of interethnic contact between Sarmatians and Scythians, the Sarmatian language as a whole can not be regarded as a direct heir of the Scythian language.

And how the current Russian linguistics answers the question of individuality of the Scythian and Sarmatian languages? To find out, turn to the A.Yu. Musorin ²⁵ and D.I. Edelman ²⁶ articles posted on the Internet at http://www.philology.ru. A.Yu. Musorin in his work noted that vernacular or dialects are always some common varieties of a single language. With reference to the term "Scythian-Sarmatian vernaculars" introduced by V.I. Abaev the above statement implies for the Scythian and Sarmatian languages a common linguistic ancestor. Indeed, such a common language ancestor of these languages existed. Methods of linguistic glottochronology established that approximately a thousand years before Scythian-Sarmatian contacts took place in the 3rd c. BC, the role of common proto-language for the Scythian and Sarmatian language could play the language of the early texts of the Avesta. But if closer in time relatives at the Scythian and Sarmatian languages are not found, it simply means that for such a long period of time the "Scythian-Sarmatian vernaculars" had to evolve into separate languages.

According to A.Yu. Musorin, a strong criteria of language individuality is "the presence of linguanym - a name of the language used by its speakers and different from all designations used by the linguistic collective for the languages of their neighbors". History did not have such cases that an ethnicity, conscious of its language as something separate and distinct, used to denote it with the same name as the language of neighboring people. It is obvious that regardless of theoretical constructs of our time, the Scythian language as an individual linguanym was attested in the written tradition by many ancient writers.

As noted by A.Yu. Musorin, linguistic affinity among related ethnic groups emerges as a consequence of extended proximity and close economic cooperation. Until recently, a common was an idea that the ancestral home of the Scythians and Sarmatians was located somewhere in the Middle Asia, and the emergence of the Scythians in southern Eastern Europe was connected with the famous Herodotus' story of about their invasion from Asia in the 8th c. BC. Even within the framework of that concept, it turns out that the clash of the Sarmatians and Scythians in the Northern Pontic in the 3rd c. BC occurred after 5-6 centuries after a loss of contact between the languages of their ancestors. It is obvious that such historical situation does not fit the definition of a "extended proximity and close economic cooperation of related ethnic groups".

Nowadays, the concept of Asiatic origin of the Scythians undergoes strong criticism and is being revised. So the work of  Yu.A. Vinogradov and K. Marchenko ²⁷ connects the formation of Scythia with two migrations of nomads from the east. The first migration wave occurred in the second half of the 8th c. BC. Moreover, the totality of archaeological materials in the Northern Pontic indicates that the autochthonous population was peacefully incorporated in the course of this migration into political and socio-economic structure of the newcomers. In the first half of the 5th c BC the power in the Northern Pontic seized the tribes from the North Caucasus. B.B. Piotrovsky ²⁸ believed that at all times "the term "Scythians" denoted many disparate tribes, different in their ethnic origin." MI Artamonov noted that the most important prerequisite for the emergence of the Scythian civilization was a formation of two profoundly different types of economy: pastoral and agricultural ²⁹. The pastoral nomadic economy could not do without constant economic ties with farming. So emerged the Scythian confederation, which should not be depicted as a mono-ethnic state. Rather, it was a fluid in its composition alliance of tribes with many dynasts and leaders. L.A. Elnitsky ³⁰ noted that for the Scythian culture are typical wide inter-tribal confederations under a rule of the royal tribe. He believed that "it is hardly possible to talk seriously about any ethnic homogeneity of the Scythians, even not going in the definition of the ethnos beyond a linguistic community".

Given the ethnic diversity of the Scythian world, it is logical to accept that the Scythian language was produced in the course of successive relatively heterogeneous lexical strata. Most likely, the Scythian language was a result of assimilation by local pastoral and agricultural population of the cultural elements of newcomer nomadic tribes, and under influence of the peripheral sedentary civilizations. It is appropriate here to recall the works ^{31, 32, 33} of the Acad. O.N. Trubachev, where he convincingly has shown that the language of the nearest neighbors of the Scythians: Sinds, Meotians, Tavrs was close to the Indo-Aryan or pra-ancient Indian language.

Same V.I. Abaev in one of his works ³⁴ showed that some of the Scythian-European, in particular Scythian-Latin isoglosses could emerge no later than 15 c. BC. In this regard, he wrote that "not the Scythians came from Asia, breaking away from the rest of the Iranians, but on the contrary, the other Iranian tribes moved into the territory of India, Persia and Middle Asia from the South N. Pontic. And Scythians were those people who stayed in their homeland in the N. Pontic, and continued to maintain contact with other peoples of Central and Eastern Europe for many centuries". Obviously, the facts found by Abaev on sustained, supported for nearly two millennia interaction of the European Scythians' language with a number of other European languages poorly agree with his concept that the Scythian and Sarmatian vernaculars were identical. This concept looks less and less convincing with the discovery of more and more evidence that the Sarmatian tribes originated from the depths of Asia.

D.I. Edelman in the work reference 25 lists a number of criteria used in modern linguistics to address a question whether two language varieties are individual closely-related languages or the dialects of a single language. He writes that the two language varieties are likely to be recognized as dialects, if the answers to the following questions would be positive:

a) did the ethnic groups that use for communication the examined language varieties belong to the same state?

b) did the speakers of the  examined language varieties belong to a single ethnic group?

c) did the speakers of the  examined language varieties use some common communication language between themselves and outside their areas?

d) were the speakers of the  examined language varieties guided to a single common cultural center?

With our knowledge of the military-political and cultural life of the Scythians and Sarmatians, to positively answer a single the question above is difficult.

Lack of understanding at the daily level between speakers of different forms of speech is a key point for recognition of two language varieties as separate languages. In the absence of mutual understanding direct communication with the use of these forms of speech is no longer possible, and the speakers have to resort to a third form. Here is a question, how well were Scythians able to understand their Sarmatian "brothers" during the mentioned above Battle at Thatis in 310 BC? Then Sarmatians, headed by Ariafarn, fought on the side of the Bosporus Prince Eumel. and the Scythians fought for Satyr, a rightful heir to the throne. Because of incompleteness of our knowledge of the of the Scythians' and Sarmatians' history, the earliest period of the loss of the linguistic contact between the two ethnic groups can be assumed to be 13th-15th cc. BC, and the latest - 9th-8th cc. BC.

The American linguist M. Swadesh believed that a 500-year limit is a natural boundary that determines when two different dialects of a common parent language become different languages. Russian scientist S.Ya. Yahontov allows the existence of some mutual understanding between the speakers of the former common language even 500 years after their separation, as is the case with the modern Russian and Ukrainian. And after 1000 years of separate life, in a foreign language remain familiar only a individual words and phrases, like in modern Polish and Russian.

It comes out that in the heat of battle on the Taman Peninsula the Scythian warriors could hardly have a better understanding of the Sarmatians' warlike cries than we now understand the "Ukraine state language". More likely that the Scythians sould understand Sarmatians not better than we are now able to "understand Polish". Criteria of Swadesh and Yakhontov unequivocally testify that Sarmatians brought their own language to the northern Pontic. Apparently, it was akin to the language of the European Scythians, but still this was a separate language.

In a variety of ancient written sources, Scythian language is desribed as a separate and independent language, which was developing independently of the Sarmatian and Alan vernaculars. Creation of a special term "Scythian-Sarmatian vernaculars" was needed to Abaev only to justify the substitution of the Scythian linguistic remains with more numerous Sarmatian and Alanian linguistic traces, where the "Iranian" nature is coming more pronounced.

Scythian lexicon according to written sources

Although V.I. Abaev repeatedly expressed regrets about the paucity of the used by him lexical material, and that he had to judge the language of the Scythians only from their proper names, but even the material that can be found in the written sources, in the historical works of Herodotus, Strabo, Pliny, and other ancient authors, he largely ignored. The Appendix below "Scythian lexicon according to written sources" provides a list containing about two hundred words with the origin unequivocally attributed by the ancient writers to the Scythians. The base of this list consits of the lexicon of the Pontic Scythia, drawn from the Herodotus' Scythian story. This list, however, does not include the words or Alanian or Sarmatian origin. This list is not claimed to be complete, and of course can easily be expanded with more in-depth study of the ancient authors' works.

Abaev used a little more than 30% of the total number of the words contained in the list (these words are underlined). As was mentioned above, Abaev for its reconstruction was choosing only the lexical material with obvious "Iranian" origin. We see that this approach leaves out about 70% of the lexicon, which has no "Iranian" roots, but which undoubtedly is Scythian. So, of the seven names of the Scythian pantheon, Abaev cites only the names of the god Papai and goddess Tabiti. Of the 60 names of the Scythian kings and Scythian nobles in the list he mentions only 18 names. Thus, trying to reinvent the Scythian vocabulary exclusively on the "Iranian" basis, Abayev on one hand ignored much of the real Scythian words, and on the other hand intensively used alien-lingual onomastic material of the Sarmatian and Alan origin.

What conclusions can be made as a result of a critical review of the material presented in the V.I. Abaev's article? Obviously, the Scythian lexicon, even in the form that survived to this day, does not entirely fit into the Procrustean bed of notions about its exclusively "Iranian" origin. The situation with the Scythian language is more complex than it appears to the supporters of "Iranian-linguality" of the Scythians. Manipulating the "Iranian" names from the epigraphic monuments of the Northern Pontic, Abaev managed to reinvent a part of the Sarmatian and Alanian lexicon. However, there are no reasons to call the dictionary compiled from the lexemes obtained in  such a way a "Dictionary of Scythian words".

Illustrarions: Reconstructions of Ptolemaic maps