The translation of the Chapter 11 is only partially edited, the extensive listings of the finds are left in a barely readable machine translation. The subject are the tamgas in the closest to us period, when the archeological finds and the extensive literature of the period should have left us with mountains of evidence indicating their absolutely non-Türkic ethnicity. However, there is none of that. Instead, in spite of the nearly-complete destruction of any Türkic-related monuments, the surviving evidence of the Alanian Türkishness unexpectedly pops up here and there, requiring extensive fantasy to rationalize the phenomena and contrive bypass scenarios, which are performed unfortunately without an artistry achieved by the best artists of science during Stalinist times.

The chapter's title ”Alans” must be stated exclusively for the attention of a censor, for there are no credible criteria so far to segregate the Alans from the other Türkic population. All attempts to pick out ”Iranism” from the Türkic alphabet inscriptions of the Middle Age population ended with failures. So, under ”Alans” the chapter systematizes Bulgarian and Khazarian states with ethnic populations called in the sources  Bulgars, Khazars, Alans, Huns, Savirs, Esegs (Esegels), Burtas, etc., who most likely also included a mass of other ethnic and tribal names like Hunogurs, Onogundurs, Horesmians, Bukharians etc. In a typical Russian tradition, we say ”Lenin” but mean ”Party”, we say one and always mean another.  If this study had ”Türkic-lingual” in its title, it would probably never see the light of the day, and we would have never learnt that the Kangars, Sogdys, Horasmians/Horezmians, Sarmats and Alans had common clans and common clan tamgas, that the royal families of the Horesmians and Sarmatians belonged to the same clans, that the dynasties of the Bospor, Olbia, Bactria were intimately connected all the way up to the Chach by their royal bloodlines, and none of the enigmatic details of these fascinating popular and royal carousels. The multiple Translator's Notes address some of the grossest  translated misrepresentations weaved in the text of the chapter.

As with the signage traditions of our days, the deviations from traditional orientation of the tamga are irrelevant, like is irrelevant the orientation of a 5-point star on the US emblems, or a 6-point star of the Mogen Dovid, their significance is relayed exclusively by their form and not by the conditions of the object where they are placed. On the other hand, the relative orientation of modifier markings is significant, allowing, for example, to segregate the animals of a ”right hand” family from the ”left hand” family and other kins of the clan, and therefore the mirror images carry discerning connotations. At the same time, an ”assignment” of the tamga would be quite an aberration, like renaming ”McDougall” clan into ”Stuarts”, because every member of the society has his own parents, and is destined to carry their tamga to his progeny. This aspect is especially pronounced in the societies where upbringing of the youngsters is traditionally a treasured right of their grandparents.

The English rendition of the extensive citation of the work is much simplified, with many details omitted and much reduced references, but with an eye to preserving the logic, facts, and evidence. The citation brings forward the author's comparisons, to allow for easy visual collation of the evidence. A reader should be aware that absolutely none of the dating was performed even with rudimentary scientific instrumentation, and with the exception of the dated coinage, all other dates are within the accuracy of educated opinions, which at times significantly differ, and time to time abruptly change. Any posting's comments notwithstanding, this work is a first major overview of the accumulated research, and we all should be limitlessly grateful for the titanic work performed by the author in researching, assembling and mapping the data, and for many of his insights that lay the ground for the future researchers. The author also notably deviates from the cleansed lexicon that came to use in the Russian academical works of the Soviet period by using Türkic terminology, integral to the Russian language, for the authentic Türkic phenomena, including the very term ”tamga”.

Mini-Glossary of the Türkic ethnic names of the period

The intermediate epoch between the Sarmatian epoch and the early Middle Ages epoch, the Hunnish epoch (end of the 4th - 5th centuries), with its political instability, large migrations and politically denigrated position of Alans under the Hunnish rule, almost does not produce any finds of the Alans' tamgas in all area of their dispersal (The army and the population should not be confused. When the army shows up in Italy, 4/5th of the family continue their pastoral life along their traditional routes.  For the army, there were two periods of Alano-Hunnish allegiance, in the first the Alan armies served as close allies, in the second period the Alanian armies split off and a more significant part deserted the Huns. The non-military part of the Alanian population weathered the kaleidoscope of overlords and lasted well into the 14th century as a significant and powerful entity. In the 7th c. Eastern Bulgars, and together with them Alans, fell under Khazar overlordship. In the 8th c. Alans sustained a 30-year war with the Arab armies, the Alan population was severely decimated, and Alans, along with other Türkic people, fled away from the Caucasus war theater, many of them established residence along the Severski Donets, interspersed with other refugees and local, mostly Suvar, people; a significant part saved themselves in the mountain gorges and returned to their Caucasus valleys. In that dispersed state Alans, with other Türkic people, survived Khazar domination till the 10th c, and the incipient Rus domination till the 13th c, when the Mongol-Alan war further decimated Alans in the Caucasus. The Donets Alans, integrated with other Türkic people, carried their distinction well into the 17th and 18th centuries. Agusti Alemany's ”Sources on Alans” provides a detail analysis of the sources - Translator's Note).

Until recently in the science dominated a version about a great prevalence of the Türkic language population in the early Middle Ages European steppes. It was convincingly criticized in the dissertation and monograph of O.B.Bubenok based on the all-around analysis of the written, archeological and other sources (Bubenok, 1995; 1997). Large groups of Alans on the basis of archeological and written sources are documented in the Western and Central Caucasia, in the upper course of Don and Severski Donets, in the Mountain Crimea. The analysis of written sources, craniology of one of the groups in the burial of the fortress Sarkel on the Lower Don has also allowed O.B.Bubenok to show convincingly enough a presence there of rather significant group of the Alanian population (in July, 1998 O.B.Bubenok kindly gave me a manuscript of a big article ”Alans-Yases in Sarkel/White Tower” being published in MAIET) (Anyone who would want to personally peruse the work of O.B.Bubenok would run into major obstacle: the only copy in the US is in the Library of Congress, untranslated. O.B.Bubenok forked over enough material,  predominantly of the early Rus and Russian annals, and including references in the Persian sources glossed over in the Agusti Alemany's ”Sources on Alans”, but the plethora of unsubstantiated assumptions, and the superfluity of accidental comparisons completely void a scientific value of his work. That Prof. Dr. S.A.Yatsenko could turn solely to the work of O.B.Bubenok to substantiate his main thesis, that is that the N.Pontic Alans in the post-Sarmatian time were Iranian-lingual, bode not too favorably for the author's credibility when he gets to the interpretation of the facts. Neither craniologically, nor ethnographically, nor archeologically can the Alans/Ases be unequivocally segregated from the list of the surrounding Türkic population, each consisting, like the Alans, of various multi-ethnic Türkic subdivisions, the major conglomerates of which were Huns, Bulgars, Khazars, Sabirs, Esegels, Besenyos, Kipchaks, Oguzes etc. Even the idea that there was a homogenous nomadic population that grew up in isolation is surreal. And a student of the clan tamgas should know it better than anybody else. The references to Alans in this chapter must be taken with a pile of salt. A clan can be Huns at one time, Bulgars the next time, Khazars later on, and Russian/Ukrainian/Kazak in the end. Just go to the Arbat boulevard in the heart of Moscow and think what the Arbat's tamga was, and what journey it made to end up right there where you stand - Translator's Note).

The tamgas on these regions of the 7th-10th centuries are shown by me in the table (fig. 36).

Fig. 36. Tamgas of medieval Alans 7th-10th cc.
(There were so many events in the N.Pontic area between the 7th-10th cc., and so much movement,
That to lump all movements into one pile, Alan or anything else, is preposterous.
Area I was shared by Bulgars, Alans, Nakhs, Adygs, and a multitude of other small and large people,
Area II was a domain of Khazars,
Area II was a domain of Suvars, with  Bulgar and Alan refugees from Arab wars,
Area IV was the territory of Ak-Bulgars, with leftover Scythian and probably Sarmatian population,
Area V was settled by Khazars and Sogdians.
Even this general, and therefore inaccurate listing shows that to treat all these territories as Alanian is a fallacy.
On top of that, the Badjanak invasion of ca 840 severely changed the demographic picture again,
forcing Kara Bulgars, aka Kupi-Bulgars, to flee with their allies to the Samara Bend)

The problem is that these areas (first of all the Lower Don) partially were contact zones with the Türkic ethnoses (Bulgars, Khazars, Besenyos), and in the Crimea with the Greek-Byzantines (who also were using tamgas of simple forms and individual letters for marking objects) (Greeks using clan tamgas? Greeks living in clan society?). Therefore reliably segregate the tamgas of the Persian-lingual and Türkic-lingual, and even the Greek population is difficult today in many cases. Partly, the analogies with earlier Sarmato-Alanian tamgas will help. But only partly, as, firstly, a number of types was borrowed by the Türks in the captured ”Iranian” lands, and second, in the early Middle Ages also certainly new forms have appeared at Alans (Borrowing tamgas is like borrowing a clan. And accuse Türks in that is not too scientific).
110

A.L.Yakobson tried to ignore even a very chance of continuity of the tamgas forms from the antique Sarmato-Alans to the medieval Alans of Mountain Crimea: ”The value of this source (tamgas. - S.Ya.), chronologically very much remote, seems to be doubtful” (Yakobson, 1970, 163). However the permanent location of the Sarmato-Alanian population in the Crimean foothills from the turn of our era to the 14th century makes the reference to the Sarmatian material quite natural (to  extricate credibly Alans from the mixture of the Crimean population, which different sources call a dozen of names, including Greeks, Goths, Tauars (Türk. ”Mountain people”), Akathyrs (Türk. ”Forest people”), Scythians, Sarmatians, Alciagirs, Alcildzurs and Alpidzurs whom Kurbat's son Emnetzur (Altsek) moved from the Crimea to Italy,  etc. would be an absolute feat of natural credulity - Translator's Note).

Our compilation shows a presence of significant individuality in each region, on one hand, and, on another hand a big share of the ”Sarmatian inheritance”. One more important feature of this compilation should be noted. It has (first of all among the tamgas on the ceramics, peculiar not to a certain region, but several) 17 alphabetic signs of the local Türkic alphabets, the ”Don” and ”Kuban” type (Kyzlasov, 1994, tab. VII, IX) (fig. 36).

The alphabetic letters of the local Türkic alphabets are represented once in each among the obvious tamgas in the Alanian lands of the N.Caucasia in the local ”encyclopedias”: in the Djinal grotto near Kislovodsk (fig. 36/22) (Runich, 1983, fig. 1) and on a ”dolmen” no. 2 of the nobility of the 10-12 centuries on r. Kyafar (fig. 36/8), and as 6 different letters on the walls of the Humar fortress (Bidjiev, 1993, fig. 37, No 26, 40, 48, 51, 54, 97). This all shows that these letters were used in the conditions of rule by the Türkic Kagans as usual tamgas (it really does not, there is a ton of other baseless explanations carefully assembled from pure speculations - Translator's Note). It is only possible to agree with such an expert as I.L.Kyzlasov: ”... There are no substantiating materials for an assumption that the runes played any role in the creation of the concurrent or later tamgas (at Türkic people proper. - S.Ya.)” (Kyzlasov, 1994, 38). And in fact, the custom of transforming into tamgas the alphabetic signs of this or that script in the world practice is documented only for the ethnoses who did not invent these alphabets and used them seldom (compare the borrowing of the letters of the Russian Cyrillics in the contact zones by the Circassians, Khakases, Nenetses, etc.) (it took centuries of Russian force to make them use it, though). This once again confirms that in these cases we are dealing not with the Türks, but with the Persian-lingual Alans (with that kind of logics, you can once again prove absolutely anything you are out to prove once again  - Translator's Note). Even more frequent is a marking of one letter of the local Türkic alphabet as a tamga on the vessels (found in all investigated regions). It is also indicative that all other types of the tamgas, common in various territories, with a rare exception (fig. 36, а/185) have exact or close analogies in the Sarmato-Alanian tamgas of the Roman time. In the groups of specific tamgas in various regions, the quantity of such exact analogies always exceeds one-half (if I understand this discourse correctly, the author is saying that the use of more than half of the Roman time's Sarmatian tamgas continued uninterrupted into the Middle Ages, and that from some point in time they sometimes were complemented by the letters of Türkic alphabet  - Translator's Note).

Now should be considered one important question about the ethnic attribution of the above  tamga collections. In the literature dominates a conviction (so far only aprioristic, not confirmed by the systematic comparison of material), that almost all tamgas in the territory of the Khazar Kaganate of the 8th - the1-st half of the 10th c. first of all to belong to the main Türkic-lingual ethnos of the Kaganate, the Bulgars. Let's try to verify this hypothesis. The matter is, in our disposal is a huge collection from the hundreds of types of the synchronous and certainly authentic Bulgarian tamgas of the 7th-10 centuries found in the capitals of the Danube Bulgaria, the Pliska and Preslav, and a multitude of other places, and in a number of the centers in the Itil Bulgaria. As a rule, the authors confine to a simple indication about a ”significant similarity” of the types of tamgas (alas, such argument does not look convincing to me). A comparison of ours ”Saltov” collection with a cumulative collection of the early tamgas from the Danube Bulgaria shows that the number of identical types is extremely meager (see, first of all: Doncheva-Petkova, 1980, examples 4, 6, 8-9, 16; Tab. II, IV-IX, XI, XIII-XIV, XVI-XVII, XXII, XXVI-XXVII; Ovcharov, 1982, tab. XXV, LXI). Of all the five regions of the Saltov tamgas' distribution, among the specific regional types we find a number of coincidences only in the Lower Don area, where nobody ever denied the activity of the Bulgarian ethnic element (fig. 36/55, 61, 63, 66-67, 72, 74-75, 78, 90, 95, 97, 99). But even there the number of the types similar with authentically Bulgarian makes only 13 out of 48, i.e. only 27 %!

In the other regions of the Saltovo-Mayak culture the number of coinciding types is so scanty that it can't be seriously discussed: in the N. Caucasia fig. 36/7, 22, 27; on the Upper Don fig. 36/105, 111, in the Crimea fig. 36/130 (Which indicates that Bulgars are not identical with Sarmatians or whatever tribes S.A.Yatsenko calls Sarmatians, but leaves out a question of Onogurs, whose continued presence in the N.Caucasus is documented from the 6th c. BC, before the arrival of Masguts/Alans to the N.Caucasus area in the 4th c. BC).

It is likewise insignificant (on a background of hundreds of types of the tamgas from Bulgaria) the similarity of the ”common-Saltov” types (fig. 36, а/170, 172, 178, 180-182, 195, 198-199) (and, certainly, of the single signs in the form of the letters of the Türkic alphabets, which in the Bulgaria were most likely used by non-Türkic subjects: fig. 36, а/179, 182, 196-197, 200-201) (Apparently, S.A.Yatsenko does not buy into the idea of some scholars that Türkic alphabet contains some tamgas adopted as letters).

Equally small is also the number of the common types tamgas for the pre-Mongolian Itil Bulgaria. Especially  is important  a most representative collection of potter's brands from the Bilyar fortress,  Bulgar (Grigorieva, at al., 1976, fig. 91; Kochkina, 1983, fig. 2). As is already easy to guess, the individual common types find analogies only in the Lower Don (fig. 36/80, 95, 97).

And if we take into account the fact that the majority of the common, for the three compared territories, tamga types are represented by the widely distributed tamgas of the simplest geometrical forms, and because of that usually used for alphabetic letters of the very various late-antique and medieval cultures of Eurasia (see first of all: Lavrov, 1978, Tab. IX), the picture becomes totally clear. Thus, with the ethnic Bulgars from the known today tamgas in the territory of the Saltovo-Mayak culture is possible to connect (with some allowances and qualifications) only a small part of the tamgas in the Lower Don.

112

Does it mean that Bulgars-Türks did not widely use tamgas in the territory of Khazaria or even that they did not have them? Certainly, not! What is only obvious is that these tamgas were made mainly on the organic objects (animal body in herds, wooden and leather products), which in the materials of excavation, unfortunately, are seldomly preserved, and are almost absent in our collection. Besides, we actually have a whole series of the Bulgarian (probably, sometimes Khazarian) objects with tamgas in the lands of the Kaganate. These are artifacts with Türkic runic inscriptions. By the way, such inscriptions in the Eastern Europe are known practically only in area where we noted the presumed Bulgarian tamgas,  in the basin of Don (Lower and Upper) are 18 inscriptions which locations are widely scattered in this area (In the N. Caucasia are found 10 inscriptions only in one location, in the Khumarin fortress, where justly expected a presence of the Bulgarian ethnic component: Bidjiev, 1993, 186) (see: Kyzlasov, 1994, 241-279) (I.L.Kyzlasov in his listing attributed, in a number of cases, to the ”Türkic inscriptions” the stones with one alphabetic sign or with two, located far apart among the images, and other like transgressions. Such interpretation, in view of the representation above about the probable use by the Alans and, probably, other ethnoses, of individual letters as the tamgas seems to be rather risky). The undertaken attempts to read these inscriptions ”in Alanian” (see first of all: Turchaninov, 1990, 78-154) have not find any sympathy in the scientific linguistic community.

Tamga ”Encyclopedias”

At medieval Alans tamga ”Encyclopedias” are connected, like the earlier Sarmatian, with the most different objects. A certain regional specificity is notable. In the Caucasus sometimes  they are caves and grottoes (Djinal grotto: Runich, 1983, fig. 1) where undoubtedly (like among the successors of the Alans - ethnographic Ossetes) were pagan sanctuaries, and also various objects connected with the cult of the deceased: crypts cut in cliffs (where tamga are usually carved in a horizontal raw at an entrance into the chamber, and identical tamgas are not duplicated - Tokmak-Kaya, Gnakyzy and Kreidy in the upper course of Kuban: Minaev, 1971, fig. 14, 17, 22) and dolmen-like mausoleums of the nobility (”dolmen” No 2 on r. Kyafar, where are 4 original tamga types: fig. 36/7-10), stone gravestone crosses with Greek inscriptions of the 10th-13th centuries (Baraban cliff at the r. Urup. There on one of the sides of the cross are chaotically carved not less than 9 tamgas: Lojkin/Malakhov, 1998, fig. 1, 1 and 2, 2). Assemblies of the tamgas in the Ukraine are represented on some portable objects (belt sets: a treasure from Martynovka in Lower Dnieper, where the proportions of the usual Alanian type tamgas are fancifully elongated: Prihodnuk, at al., 1991, fig. 1 (2-4), 5 (1-5, 20), 9 (2-3), 10, 12 (2); Korzuhina, 1996, tab. 164, a clay cubic model of a temple from the Ay-Foka cape in the Eastern Crimea. Djanov, 1996, 187-188, fig. 1).

Should be noted the specific iconographical similarity of a set of the ”dancing” idols and paired animals in the Martynov treasure, and in authentically Alanian monuments of Caucasia (village Pregradnaya) (Yatsenko, 1992, 71-72, fig. 1/7). Spay to These two facts do not note the supporters of the Türkic, Early Slavic, and German attribution of the treasure and its separate elements.
113

In various areas of Alan settlement, the ”encyclopedias” are carved on the walls of large fortresses, neighboring individual tamgas on separate stone blocks, which the researchers of the Saltovo-Mayak culture call ”tamgas of the builders” or ”brands of the masters” (first of all see Humara: Bidjiev, 1993, fig. 37; Mayak: Pletneva, 1984; Flerova, 1997, tab. of the 3rd-4th; Mangup (Doros): Gertsen, 1990, fig. 15-18). Were found individual blocks with even three marked sides (Nahapetyan, 1988, 102). The tamgas of some types are found several times on the walls of one site (but certainly not so frequently that they could be taken for ”brands of the masters”!) . The special analysis conducted by V.E.Nahapetyan has shown that the large Mayak fortress was built not by the professional crews, but the local residents who left their tamgas on individual blocks (Nahapetyan, 1988, 105) (see Baichorov study of Türkic inscriptions). It is speculated that tamgas in  these ”encyclopedias” were carved there during some ceremonies accompanied with a collective oath (Nahapetyan, 1990, 60) (what an ignorant nonsense!). In Humar, as was already noted, tamgas were restored after whitewashing of the walls (Bidjiev, 1993, 165). Tamgas not necessarily belonged to a clan: there could be only simple additional markings of individual large families, as was, for example, still recently observed at the construction of clan towers in the Central Caucasia (Krupnoe, 1971, 136) (the development of constituent tamgas among descendents is a separate field, and should not be based on casual suppositions. Each tamga is a representation of individuality ”I”, ”me”, ”mine”, and can't be expressed by ”only simple additional markings”).

On walls of Mayak were repeatedly found 12 types of tamgas (frequently, the tamgas are grouped together) ⁵ . Of them five are the types spread in different parts of the Alanian world (i.e. Bulgarian) (fig. 36, а/171, 178, 187, 198-199), 5 more types are specific to the Upper and Lower Don (fig. 36/101, 105, 116-117, 122) ⁶, and, at last, were found letter signs of the Türkic ”Don” and ”Kuban” alphabets (fig. 36, а/197, 201) which, as was already noted, in the Alanian context of that time could be tamgas also. Of the singular tamgas, 5 types are specifically local (fig. 36/102, 111, 115, 123-124) and only one was found in different regions (fig. 36, а/180). For deep carving of tamgas on the fragile chalk blocks was used a special techniques (Nahapetyan, 1988, 91). The tamgas are frequently located at two different levels, as they were drawn on the walls while sitting or standing (Pletneva, 1984, 57, 59).

In the Khumar fortress on one of the towers of the excavations A, on the wall, gate, staircase, and also in a sanctuary (Bidjiev, 1993, 164-165) were found 9 types of tamgas, known only to Caucasia (fig. 36/11-19, of them were repeatedly found No 12, 17), 8 types found in various parts of the Alanian world (i.e. Bulgarian) (fig. 36, а/14-16, 17, 22-24, 30, of them are repeatedly found No 14, 16-17, 23-24), and 3 letters of local Türkic alphabets (fig. 36, а/184, 193, 197 of which first two letters were carved repeatedly).

On the end face of the blocks (not visible after construction) were carved only 8 % of tamgas, which in this case realy belong to ”tamgas of the builders” (Nahapetyan, 1990, 57).

6 Some of them could be initially connected with the certain scene drawings on the walls. So, the type No 25 in one case is associated with an image of a horse, and type No 82 in one case is associated with a five-pointed star.
114

On the walls of the Mangup (Doros) northern fortifications, in accordance with published materials, were found 21 types of local tamgas (fig. 36/130-141, 147-153, 156-157, of them were repeatedly represented only No 140-141), and 8 types known in different parts of the Alanian world (i.e. Bulgarian) (fig. 36/1, 2, 7, 10, 15, 24, 36, 39, of them repeatedly were depicted No 1, 2, 15), and also 3 tamgas with the letters of the local Türkic alphabets (fig. 36, а/179, 197, 200) ⁷. The population of Mangup (Doros) in the 6-10 cc., in the opinion of its leading researcher A.G.Gertsen, mainly consisted of Alans (see, for example: Gertsen, 1998, 748) (The Gothic fortress Doros, present Mangup, may have had some Alans there, but to call it Alanian settlement is too much, especially in view of not only Türkic, but clearly Greek brands. The last Türkic speaking Crimean Goths were deported in 1778 by Catherine I of Rissia to Azov area, where they established a city Mariupol. Since tamgas are the only positive way to identify the ethnicity of the population without ambiguities of the archeological and written evidence, appealing to the studies that do not trace tamgas directly to make ethnic definition of the tamgas is a clear case of a circular logic. But aside from frivolous speculations, the facts are more then valuable).

In addition, tamgas are known on the stone walls of such fortresses as Saltov, Olshan and Right-bank Tsimlyan on the Don and Donets (Lyapushkin, 1952, 135), and also in Bakla in Crimea (Rudakov, 1979, fig. 1/4) (In most cases the real names and belonging of the fortresses are well known and substantiated. The use of conditional names is not helpful in the line of discourse that addresses attribution of the tamgas).

In the Djinal grotto were found 12 almost exclusively original local N.Caucasian type tamgas (fig. 36/3, 4-6, 9, 21-25, 27-28, of them No 27 was  found 2 times, No 9 was  found 4 times), and only 2 types belong to the prevailing tamgas (fig. 36/15, 21), and the tamga No 158 is known in Crimea (in the grotto it was depicted three times); letter tamgas were completely absent there. All that allows to view it as a sanctuary of specially local significance, and probably not intended for strangers (Applied to the Vietnam Memorial or the cemetery in the Kremlin wall, where most of the names are different, and only a few are of prevailing type, those monuments are also ”of specially local significance, and probably not intended for strangers”. Wow!).

A original group of tamgas was depicted in a raw with equal intervals at the butt ends of a clay model of a temple, ornamented with strips of red paint, found on the settlement 8th-9th centuries at Ay-Foka cape. The product has Only a half of the object was preserved, and from 8 (?) initial tamgas only four reached us. Three of them are specific (fig. 36/142-144), the fourth is a five-point star (which was used as tamga in other regions of Saltovo-Mayak culture also).

On smart zone set from a treasure of 7th century in Martynovke on the Lower Dnieper the majority of tamgas is specifically local (fig. 36/162-168) and only two Alanian worlds distributed in different parts (fig. 36, а/173, 181), two more letters of local Türkic alphabets (fig. 36, а/186, 197) (the 7th century Lower Dnieper area was a center of Ak-Bulgars, interspersed with Akathyrs, Scythians, and probably Sarmatians. The period and location are very significant for the demographical picture at the end of the Great Bulgaria, and may include tamgas of the tribes that migrated westward with Asparukh).

Significant interest also present a set of solitary tamgas on the bricks from a series of Khazar 9th century fortresses constructed along a small strip on the banks of the Lower Don: Sarkel, Right-bank Tsimlyan and Semikarakor (Scherbak, 1959, 363-365; Flerova, 1997, tab. VI-VII). The ethnic identification of the people producing bricks there is not clear; however, a participation of qualified Alanian masters, because the Alans obviously lived in that district, can not be excluded. The greatest variety present the collection from Sarkel of the 830s, probably built with participation of Byzantines, but directed by local builders (Artamonov / Pletneva, 1997, pp. 549, 622). In the collection sharply prevail 20 local types (fig. 36/54-70, 76, 99, 101, of which repeatedly are found No 55, 57, 62). Another 8 types are much less prevailing (fig. 36, а/170, 174, 185, 194-195, 199, 202-203) and 6 are letters of local Türkic alphabets (fig. 36, а/184, 186, 196-197, 200-201).

7 In Mangup (Doros) was found a a series of new tamgas, which are being prepared by A.G.Gertsen for publication (oral message, November, 1998).
115

The collection of tamgas from the Right-bank Tsimlyan fortress has a small number of types, from which a half is the original local tamgas (fig. 36/93-94, 98), another half is prevailing tamgas (fig. 36, а/172, 178, 198). At the Semikarakor fortress the picture is about same: 4 of the Don types (fig. 36/93-96, of them No 93-94 are also known on the Right-bank and nowhere else), and 3 prevailing types  (fig. 36, а/178, 197, 203).

A mass kind of objects with tamgas was potter's ceramics, where such tamgas were frequently carved on the surface of a potter's wheel, thus being the original form of a clan symbol (or a small complementary symbol of his patronym) of the potter. It is well noticeable on an example of such Upper Don burial as Dmitriev necropolis (Flerov, 1979, fig. p. 98-99). There, a widely spread tamga No 194а is found on the ceramics 14 times, a local tamga No118 - 6 times, and another local No 120 - 3 times. In the necropolis Wet Beam in the N.Caucasia (Afanasiev, 1976, fig. 2) the tamgas also in two cases out of three are specifically local (fig. 36/12, 31). In the N.Caucasia is distinguishable a group of potter's disk imprints on the bottoms of the jugs and pots in a catacomb burial at the Zmei village. There, only during a season of 1959 were found 19 types of potter tamgas (4 of poor quality). Five of them are talisman crosses of various forms. The others are fairly various (fig. 36/45-53) (Kuznetsov, 1995).

The collection graffiti on amphorae from Sarkel is also of interest (Flerova, 1997, tab. XV). There are sharply prevailing individual original local tamgas (fig. 36/55, 63, 73, 75, 77-79, 81-90, 101), 7 letters of local Türkic alphabets (including the fig. 36, а/179, 196-197, 200-201) and a few tamgas of the prevailing types (fig. 36/178, 180, 187-188, 205). On astragal game mutton bones was usually found one tamga, and only on one exemplar from a burial in Upper Saltov were two tamgas (fig. 36, а/199-200) (Flerova, 1997, tab. VIII, No 9, 24-26, 27-28, 31, 33, 55-57, 59, 62). In an Alanian burial Rocky in the Southwestern Crimea are known individual tamgas on a shoulder of the jugs, and on wooden dishes, where the letters of the Türkic runiform alphabet served as tamga (crypts 153 , 760) (fig. 36/146, 204) (Veimarn/Aibabin, 1993, fig. 12/17, 118/20). Similar tamga runiform letters were located on glazed vessels from Five Mountains area in the Northern Caucasus (Kovalev, 1984, 173). Separate tamgas, identical with those from Saltov, were on ceramics of the later Kherson (Yakobson, 1950, fig. 81).
116

In respect to the early Middle Age roof tiles from Southwestern Crimea (Potter's, Bobrovka: Yakobson, 1970; Eski-Kermen and Mangup (Doros): Yakobson, 1950, 20-22), in contrast, usually depict Greek and less frequently Türkic letters, and also tamgas of widely spread types (fig. 36, а/172, 177, 189, 196-197, 201) and unique types only in rare cases (fig. 36/129-130, 154-155, 159-160). A part of craftsmen there could belong to the Alanian and Türkic ethnoses (Why not to mention Goths? Why not to mention Akathyrs? Akathyrs lived in Crimea until its conquest by the Russia in 1771. In the 6th-10th cc. AD, the Akathyrs and Scythians were main Türkic ethnoses in the mountain Crimea). These tamgas as a whole differ from the tamgas on the ceramide roof tiles of the 8th-9th centuries from such a Byzantian city on the Southern Coast as Partenit (where the letters of the Greek alphabet prevail: Parshin, 1991, fig. 4). It is interesting to also compare them with the brands on the tiles of the later Khersoness, where along, with prevailing Greek letters are also found dozens of the types identical with the tamgas of the Saltovo-Mayak culture (Yakobson, 1950, fig. 85, tab. 7-9, 11, 16-18). A majority of them are of fairly simple geometrical forms, and they can be accepted both as tamgas of the actually ethnic Greeks, and tamgas of the descendants of ”Saltovians”.

On the gravestone stelae from the Rocky (Southwest Crimea) is repeatedly found a standalone tamga (fig. 36/127-128, 177) (Veimarn/Aibabin, 1993, fig. 11/1-2, 16, 19).

A placement of standalone tamgas at the N.Caucasia Alans in some cases also raises a great interest. Round bronze plaques, sown on a frontal part of a headdress, with the tamga image as triquetrum, oriented clockwise or counter-clockwise are curious (fig. 36/20, 26, 32). They were found in a Martan-Chu necropolis in the Chechen Republic (Ichkeria) (Vinogradov/Mamayev, 1984, fig. 8, 33; 12, 7-8; 13, 16) and in Starokorsun in Kuban (Kaminsky, 1984, fig. 4/1).

Triquetrum (triquetra) on a Funbo runestone U 937 with runic inscription in Uppsala

One more tamga, repeatedly encountered and specific to the N.Caucasia (fig. 36/17), is known from a dolmen-like crypt of an aristocrat of the 10th-13th centuries in Upper Teberda (Kuznetsov, 1961, fig. 44, 4-6). It is depicted above an image of standing to the right horse archer (deceased?), toward whom are running 5 deers and fawn; near him are depicted 7 men with hands raised upwards in prayer. On ”deer slabs” No 2 and 8, placed at the entrance to the Kyafar fortress, among the images of deers and Christian crosses was noted one tamga on each (fig. 36/44а-í) (Arjantseva/Albegova, 1999, fig. 4 and 6). The first of them represents an ancient (Türkic?) alphabetic character; the second is a distortion of the Greek letter ”A” (a similar tamga still was recently depicted on the tower of Tolhanov's clan in the village Tsimiti in Ossetia: Tmenov, 1996, fig. 78/8).