The Cultural Assimilation of the Varangians in Eastern Europe from the Point of View of Language and Literacy

Elena A. Melnikova

Since the 1960s it has been a locus communis in the writings on Old Russian and Scandinavian relations that those Vikings who settled in Eastern Europe in the second half of the eighth to the end of the tenth century underwent rapid assimilation in the Slavic milieu. The second generation of the settlers is considered to have fully adopted the local material culture, to speak Old Russian, and to practice the customs and rituals current among the Eastern Slavs[840].

This conclusion is based on archaeological materials that indeed show that Old Norse ethnically indicative elements were forced out of the material culture of the Old Russian elite already by the end of the tenth century and that they utterly disappeared in the eleventh century. The only non-archaeological argument commonly used to prove this point is the appearance of Slavic personal names in the family of Kievan princes of Scandinavian origin – Svjatoslav, Predslava, and Volodislav, as attested in the Russian treaty with Byzantium of 944[841].

The validity of both arguments is subject to doubt.

Most archaeologists maintain that the introduction of Christianity in Rus’ speeded up the ethno-cultural assimilation of Scandinavians, as the unified religion and rites replaced different heathen cults and practices. It is true that Christianity exerted a deep influence on Old Russian culture as well as on the culture of other previously pagan peoples, first and foremost on their burial customs in which ethnic indicators are more obvious than in other cultural traditions. At the turn of the tenth and eleventh centuries the diversity of burial customs in town necropolis in Ancient Rus’ is replaced with a unified Christian rite. Therefore the disappearance of specifically Old Norse features in Old Russian burial monuments at that time can be regarded not as an indication of the end of the process of assimilation of Scandinavians, but rather as a result of the introduction of Christianity. It is worth mentioning, that in Scandinavian countries too conversion into Christianity led to a rapid disappearance of burials typical for the tenth century and to the spread of simple graves with no objects to manifest the ethnicity of the buried in any way.

Another consideration to be taken into account is the fact that clothes, ornaments and weapons served as indicators not only of ethnicity, but of a social status as well, and therefore they could have been rather easily changed when a person entered a new social and political entity.

Thus, speaking in terms of archaeology, the assimilation of Varangians by the beginning of the eleventh century arouses doubts. The disappearance of ethnically indicative objects can be easily explained by the introduction of Christianity and changes in the status fashion of the Old Russian elite that included Varangians.

However, ethnic and cultural assimilation does not consist of changes in material culture only. It is a complicated and manifold process of the replacement of one set of features characteristic of an ethno-cultural community with a different one. Together with material manifestations such as house-building techniques, implements and everyday objects, costume and ornaments, these features include cultural phenomena like language and literacy, folklore and literature, religion and common beliefs. Taken together they form the means for a community to become aware of its integrity and of its opposition to other communities, that is, they provide it with self-consciousness and self-identification. The assimilation then can be considered to have taken place only if all these factors underwent certain changes.

No systematic studies exist so far, which would describe the cultural and mental aspects of assimilation of Varangians in Eastern Europe. The reasons are both the concentration on archaeological evidence and scarcity of non-material sources. In fact, there is no way to learn what language was spoken or what alphabet was used by a noble warrior buried in the Black Mound in Chernigov, which gods were worshipped by those cremated in Gnjozdovo or how Vladimir the Saint and Jaroslav the Wise, Russian princes of Scandinavian descent, regarded their identity. The concrete questions of that kind are doomed to remain unanswered. Still there are indications to provide for more general answers, even if in a preliminary and hypothetical form at the moment.

The most important indications of ethno-cultural self-identification are language and literacy. Their retention by a community living in an ethnically and culturally alien milieu is an unequivocal testimony to isolated position of this community as perceived both by its members and its neighbours. Therefore the time when Scandinavians who had settled in Rus’ stopped to speak their native language and to write with runes is of crucial importance.

The study of the linguistic assimilation of Scandinavians in Rus’, i. e. their transition to usage of Old Russian and oblivion of their mother tongue, can base on at least three kinds of sources. First, it is epigraphic material, represented by inscriptions in Runic or Cyrillic script. Second, it is direct or indirect information preserved in Old Russian and foreign sources on the linguistic situation in Rus’ in the tenth and eleventh centuries. Third, it is the usage of Old Norse personal names, though in itself it might be viewed as a family tradition not supported by other cultural features. Still, the degree of phonetic adaptation of Old Norse names in Old Russian sources as well as the context of name-giving can throw some light on the problem.

The earliest written record in Ancient Rus’ is a runic inscription on a wooden stick from Old Ladoga[842]. It is archaeologically dated to the first half of the ninth century. Together with three runic amulets coming from the same area and dated to the tenth century it forms a group of runic inscriptions typical for Eastern Sweden[843]. At least one of the inscriptions, on the Gorodishche amulet II, was produced on the site where it was unearthed: it is a copy of another and earlier amulet (Gorodishche I) found on the same site. Even if the stick and the two amulets belonged to occasional visitors to the Ladoga region and were brought there from Sweden, they must have been intelligible and meaningful enough for the settlers of Scandinavian origin to order a copy of one of the amulets. It means that in the tenth century the Varangians, at least in the North-Western part of Rus’, were in full possession of their native language and beliefs.

In the first quarter of the tenth century first traces of the usage of Old Russian by Varangians appear. In a famous «Varangian» site of Gnjozdovo near Smolensk a burial of a warrior contained fragments of an amphora with a Cyrillic inscription[844]. The warrior must have been a Scandinavian buried with his wife or his concubine, a Scandinavian or a Slav by birth. Judging from the ritual, the burial was performed by his compatriots in a traditional Central Swedish way. The inscription on the amphora deliberately broken during the burial rite has several readings and interpretations, gorouhsha or goroushcha being the most probable. According to these readings, the inscription consisted of one word designating the contents of the amphora, either mustard, or inflammable fluid. As the earliest Cyrillic inscription, it raises many questions, its presence in the grave of a Scandinavian being one of them. Whoever made the inscription, a Slavic or a Russian, i.e. Scandinavian, merchant, it was intended to inform the community consisting mostly of Scandinavians about the contents of the amphora which presupposes their ability to comprehend the message. Thus, the Gnjozdovo inscription though not directly witnessing the usage of Old Russian by the Varangians, still suggests their getting familiar both with it and with the Cyrillic script.

The linguistic practices of rus ’ (Old Russian warrior elite of Scandinavian origin) of the mid-tenth century can be convincingly reconstructed on the basis of information preserved in the treatise «De administrando imperio» written by Byzantine emperor Constantine VII Porphyrogenitus (ca. 950). Parallel Old Norse («Russian») and Slavic names of the Dnieper rapids listed in the treatise suggest that Constantine’s informant, a representative of rus was bilingual. Moreover, the Old Norse names are rendered in Greek without any significant corruption of the original words.

On the contrary, identification of Slavic names is sometimes uncertain and the correlation of meanings of both names is obscure. All this means that the informant was in full possession of his mother tongue (most probably Old Swedish as some of the names derive from Old Swedish forms) but he was not so sure about Old Russian[845]. The bilinguism of Old Russian elite and professional warrior strata is corroborated by information of a Spanish traveler Ibrahim Ibn Jakub (960ies) and Leo the Deacon, a Byzantine historian of the late tenth century[846].

Two graffiti on the walls of St. Sofia cathedral in Novgorod provide another chronologically precise reference point. Although inscribed by persons with Old Norse personal names, both texts are written in Old Russian in Cyrillic. The first graffito reads «Oh, miserable feel I, Ger(e)ben the sinner»[847]. The name Ger(e)ben is an obvious rendering of ON Herbeinn, OSw. Herben, the name not otherwise attested in Old Russian sources. There can be no doubts that Gerben came from a Varangian (originally Swedish?) family, and his mastership of Old Russian was flawless. He makes no mistakes of any kind and he uses a formula widespread among Old Russian scribes (who entered it in marginalia in manuscripts) and laymen. The grafitto is dated to the second half of the eleventh century.

The second inscription reads: «Lord, help your slave Far’man, Gleb’s retainer»[848]. It has a more precise date, most probably 1137, as the inscription mentions prince Gleb who then ruled in Novgorod. As in the first graffito, this one also contains an Old Norse personal name, or rather a nickname, Far'man < Farmadr in an Old Russian formulaic text. The accuracy of spelling, the usage of correct forms, and the sureness of hand show that both, Gerben and Far’man, were experienced in writing and that their mother tongue was Old Russian though they had Old Norse names – probably due to family tradition.

The same tendency is characteristic of runic literacy[849]. Besides amulets from the Volkhov region (Old Ladoga and Gorodichshe) the tenth century is rich in graffiti on Islamic coins coming from hoards buried in different parts of Eastern Europe, but mostly on the sites along the Dnieper, the Western Dvina, and the Upper Volga, the main river routes of that time. The majority of inscriptions, ca. 200, are made with runes and rune-like signs. There are only about 10 Arabic inscriptions, some 10 made with Turkish runes, up to 5 made with Armenian or Georgian letters, and one Greek graffi to. Quite recently a coin carved with two Cyrillic letters was found in Gnjozdovo. The Old Norse inscriptions are very short, they consist of one word or one or several rune-like signs. The readable graffi ti, not more than twenty in number, contain words gu? and ku?, go? «god», kiltR, gildr «of full weight, of good quality», or personal names like ubi, ?bbi. More numerous are pictures of warrior goods, namely weapons (swords, knives, arrows), ships, drinking horns, banners, and symbols like swastika, Thor’s hammers, or the emblems of the Rurikides. The runic and rune-like inscriptions and the pictures on coins are made in the same technique and must have been executed by warriors and merchants of Scandinavian origin during their stay in or voyage through Eastern Europe[850].

In the eleventh century the number of runic inscriptions found in Eastern Europe decreases and their character shows traces of degradation. Except for the Berezan’ stone that was erected by Scandinavians on their way to or from Byzantium there is only one inscription typical for Scandinavia, which represents a part of futhark carved on a bone from Novgorod[851]. Identification of other inscriptions as runic is uncertain. Though the letters are similar to runes, there are misspellings, syntactical constructions unusual for Old Norse, etc. These features can be probably regarded as a result of gradual loss of writing-in-runes habits of the descendants of the Varangians as well as their loss of native language.

As scanty as the sources for the ninth to the eleventh century are, they seem to reflect different stages in the linguistic assimilation of Scandinavians in Rus’. By the mid-tenth century the Varangians became bilingual; by the end of the eleventh century they used Old Russian as their mother tongue.

The developments in the interval between the mid-tenth and late eleventh centuries are partially attested by personal names current in the family of Kievan grand princes and among the warrior elite. The Old Russian annalist of the beginning of the twelfth century, the author of the earliest extant Russian chronicle (though preserved in manuscripts of the fourteenth century and later) was sure that Russian princes descended from Rurik (< Hrcerekr), a leader of a Norse people rus who had been invited by the Novgorodians in 862 to be their ruler[852]. His commander-in-chief Oleg together with Rurik’s son Igor’ moved to Kiev and founded the dynasty of Russian rulers. Though the traditions about Rurik, Oleg and Igor’ were most probably put together in a genealogical sequence by a late eleventh-century annalist and before that existed as separate tales or cycles of tales[853], there can be little doubt that the tales were based on historical memories and preserved some real details, names of the heroes among them.

Up to the middle of the tenth century all the names mentioned by the annalist are of Old Norse origin. These are the names of successive Kievan rulers Olg or Oleg (< Helgi) and Igor’ (< Ingvarr), of the latter’s commander-in-chief Svenel’d (< Sveinaldr), and of a group of Oleg’s emissaries who concluded a peace treaty with Byzantine emperor in 911 (or in 907 and 911) after a successful attack on Constantinople in 907. Among fi fteen representatives of prince Oleg only two have probably Finnish names, while others bear purely Scandinavian names Karly (< Karli), Inegeld (< Ingjaldr), Farlof (< Farl?fr), Veremud (< Vermundr), Rulaf (< Hr??l?fr), Gudy (< G??i), Ruald (< Hr??l?fr), Karn (< Karn), Frelav (< Fri?l?fr), Ruar (< Hr?arr), Truan (< ?r?ndr), Fost (< Fastr), and Stemid (< Steinvi?r)[854]. The set of names in the 911 treaty is homogeneous and testifi es that among the upper layer of Oleg’s retinue there are neither Slavs, no Scandinavians using Slavic personal names.

The situation changes in the mid-tenth century. The treaty of 944 includes 76 names: of representatives of the princely family (12), their emissaries (11), other agents and their masters (27), and merchants (26). It is in this list where Slavic names appear for the fi rst time. Most of them belong to members of the princely family: Svjatoslav, the son of grand prince Igor’, Volodislav, and Predslava, whose relations to Igor’ are not stated. Other princes and princesses, including Igor’s two nephews, have Old Norse names: Ol’ga (< Helga), Igor’s wife, Akun (< H?kon), Igor’s nephew, Sfanda (< Svanhildr), Uleb (< ?leifr), Turd (< ??r?r), Arfast (< Arnfastr), Sfi r’ka (< Sverkir). Princes’ emissaries bear Old Norse names too, but for three persons whose names are Finnish. There is no correspondence between princes and their emissaries in regard to the origin of their names. Ol’ga is represented by an agent with a Finnish name Iskusevi while Volodislav has an agent named Uleb (< ?leifr). The list of other agents includes a number of Finnish and no Slavic names while the list of merchants contains three Finnish and two Slavic names[855].

The predominance of Old Norse names is obvious though Slavic ones start to appear. The usage of the latter is restricted to two groups, namely the princely family and merchants. The penetration of Slavic names into princely anthroponymicon indicates the beginnings of assimilation processes. Princes of Scandinavian origin started to feel a necessity and found it possible to borrow local names for at least some of their scions thus breaking off with the ancestral tradition of naming. As to the warrior stratum, the treaties reveal no tendency for them to make use of Slavic names as yet. The «Primary chronicle», however, names a man with a Slavic name Pretich among the highest officials some twenty years after the treaty of 944[856]. Two other commanders-in-chief of Igor’ and Svjatoslav mentioned in the chronicle are Svenel’d and Asmud. The difference in the usage of Slavic names by the Rurikides and by warriors can be explained by the second group’s greater mobility. The majority of warriors came to Rus’ and went home, and only a part of them stayed forever. At the same time they must have suffered less pressure to accommodate themselves to the local population than the princes who needed support on the part of local nobility. The adherence of warrior elite to traditional names is attested by the name of Svenel’d’s son Ljut (< Lj?tr)[857].

The practice of name-giving in the princely family can be further traced since the 980ies, in the generation of Svjatoslav’s grandchildren. The «Primary chronicle» supplies information about twelve sons and a daughter of Vladimir the Saint[858]. Only one of Vladimir’s sons has an Old Norse name – Gleb (< Gu?leifr). All the rest have Slavic names, mostly compounds with – slav (< slava, «fame») as a second stem.

Since then the number of Scandinavian personal names among the Rurikides’ gets restricted to four masculine names and one feminine name. The most popular among them were Oleg, Igor’, and Gleb (due to canonization of prince Gleb who was murdered in 1015). Rurik is met for the fi rst time in the mid-eleventh century and it was used later from time to time but did not enjoy wide spread. The only feminine name that remained in Old Russian anthroponymicon was Ol’ga. Three other Old Norse names known from the treaties, namely H?kon, ?leifr and Ivarr, continued to be used, now not by the Rurikides, but by Russian nobles. Drastic decrease in number of Scandinavian personal names at the end of the tenth century can indicate that assimilation processes had intensifi ed.

There is a great diff erence in the forms of names in the treaties, between those that do not occur later and those current among the Rurikides. In the treaties all the names except for Oleg, Ol’ga and Igor’ are rendered in a form as close to the original as the Old Russian phonetic system allowed. At the same time there exist certain fl uctuations in rendering vowels /? / > o and u (?leifr > Oleb / Uleb) and /?/, /a/ > a and o (H?kon > Akun, ]akun, Arnfastr > Arfast and Fastr > Fost, the latter two occur both in the treaty of 944). The interdentals /?/ and /?/ that lack in Russian are systematically refl ected as /d/ (very seldom /z/) and /t/ respectively (??r?r > Turd, Gu?leifr > Vuzlev). The initial Fr– uncommon in Old Russian is usually substituted by Pr– (Freysteinn > Prasten, but sometimes also Frasten). It seems that there was no stable tradition of spelling Old Norse names and the scribe was free in choosing this or that variant[859].

On the contrary, the princes and princesses’ names that came into permanent usage in the eleventh century have stable forms common for the whole text of the chronicle. These forms reflect changes that originated in the course of adaptation of the names in Old Russian. Thus, the initial H– in Ol(e)g and Ol(‘)ga is omitted in all cases and the combination of consonants – lg– started to be occasionally divided by a reduced vowel. Igor’ (< Ingvarr) has a denasalized group Ig– and a compressed second stem while in Gleb (< Gu?leifr) it is the fi rst stem that turned to be compressed.

Though phonetically altered forms of these three names are used in the treaties of the first half of the tenth century, they could hardly appear then or even by the end of the tenth century. Foreign sources of the time of the treaties and of the late tenth century render these names in forms closer to their Old Norse variants than to those found in the Old Russian chronicle. Thus, Byzantine authors of the middle and the second half of the tenth century preserve nasalization in Yngvarr – ?????[860], or ?????[861] and Inger[862]. In the so-called Cambridge document written in Hebrew and telling about an attack of the Rus-people on SMKRYY (Tmutarakan’?) the name of the leader of the assailants is rendered as HLGW, i.e. Helgi with the initial H-[863]. The fact that the adaptation of the name Gu?leifr was not complete even in the middle of the eleventh century is attested by a manuscript of 1073 made for prince Svjatoslav where the name Gleb is spelled with a reduced vowel between G and l, the remains of the stem Gu?-[864].

As these sources reflect authentic pronunciation of the names, it is more probable that the process of adaptation of Old Norse personal names used by Russian princes came to its end not in the mid-tenth century, but only in the second half of the eleventh century. The annalist who wrote at the beginning of the twelfth century knew the already adapted forms and used them throughout his text. He must have had no knowledge of the earlier pronunciation of the names, as he was unable to identify the contemporaneous form Gleb and the mid-tenth century form Vuzlev (< Gudleifr) that he left unchanged.

Were Old Norse names perceived at the turn of the eleventh and twelfth centuries as alien to Eastern Slavic tradition? Two cases seem to suggest an answer to this question. First, it is the case of the name Vuzlev in the text of the treaty of 944. The Old Norse form was unfamiliar to the annalist and he did not correlate it with the Old Russian Gleb. One can suppose that Gleb sounded for the annalist as an original Old Russian name, contrary to Vuzlev and dozens of Old Norse and Finnish names in the preamble.

Even more demonstrative is the second case. Wide spread of the name Igor ’ in the princely family did not prevent the borrowing of its original (Ingvarr) for the second time in the twelfth century in an unaltered form Ingvar’. This form is attested in the Hypatian chronicle as the name of a son of great prince Jaroslav Izjaslavich. Ingvar’ Jaroslavich must have been bom in the mid-twelfth century and died in 1212[865]. At the beginning of the thirteenth century two princes of Rjazan’ had the same name. The second of them mentioned in 1207–1219 was a son of prince Igor’ and the annalist called him Ingvar’ with the patronymic Igorevich[866]. The etymological relationship of the two names was not apparent and they were regarded as different ones.

Thus, cardinal changes in the cultural traditions of former Varangians seem to have been completed in the second half of the eleventh century. By the end of this century both runic script and the Varangians’ mother tongue fell into disuse and most probably became forgotten. They were replaced by Cyrillic alphabet and Old Russian language. Personal names of Old Norse origin changed phonetically and stopped being identified with their prototypes. They were no longer viewed as foreign but probably were equaled to Old Russian pagan names.

It should be noted that this conclusion bases on materials coming from larger sites or towns, Gnjozdovo, Kiev, Novgorod. The evidence concerns the princely family, the warrior elite and the descendants of the Varangians of high status. There is no doubt that the social elite and town citizens adopted cultural innovations much easier and quicker than the peasantry. Therefore the processes of assimilation could develop faster in towns than in smaller, especially rural communities.

There are not so many traces of Scandinavian rural colonization in Ancient Rus’, especially in its Southern part. Therefore it is all the more surprising to find vestiges of Scandinavian cultural traditions in rural areas as late as the fourteenth and early fifteenth century. The evidence is again supplied by runic or rune-like inscriptions and by personal names.

The anthroponymicon of ca. 1000 birch-bark letters found by now in Novgorod numbers several hundred personal names. Most of them are Slavic or Christian, but seven birch-bark letters contain Old Norse names[867].

The oldest among these birch-bark letters is dated to the second half of the eleventh century (stratigraphic date is 1080ies) and mentions As gut who lived in a village near lake Seliger and owed a Novgorodian moneylender or tax collector several grivnas[868]. The context of the letter suggests that Asgut was a resident of the village. There is no way of finding out what brought Asgut (or his ancestors) to this village, but it is significant that the settlement was located on the Seliger route from Novgorod to the central part of Rus’.

Other birch-bark letters with Old Norse names were written in the second half of the fourteenth century and came to Novgorod from different parts of the Novgorod land. They name Vigar’ (< V?geirr or Vigeirr)[869], a «man of Mikula» Sten’ (< Steinn)[870], Jakun (< H?kon)[871] and a widow of another, Jakun[872]. The most interesting is the birch-bark letter No.[873] that mentions a place-name Gugmor-navolok deriving from ON Gu?marr, and two persons living nearby named Vozemut (< Gu?mundr) and Vel’jut (< V?lj?tr). The combination of these names suggests that a certain Gudmarr once settled on the site near a portage (navolok) on the way to the lands north of lake Onega and that the tradition of using Old Norse names was preserved in the family (or in the community?) into the fourteenth century. It is highly improbable that Gudmarr was a newcomer, as there are no traces of fourteenth-century immigration in this area, as well as of earlier Scandinavian antiquities in the vicinity. It seems that the descendants of Gupmarr adopted the material culture of Slavic and Finnic neighbours, but retained their own name-giving traditions.

The content of all birch-bark letters but No. 249 (Sten the «man of Mikula», could be either a resident or a migrant from Sweden or Swedish Finland) points to the fact that the bearers of Old Norse names were residents fully incorporated in the local social and economic life. They are mentioned among persons with Slavic names, they pay taxes in firs, become debtors of Novgorodians, receive money, they deliver homemade products together with the Slavs and the Finns. According to the topography of later birch-bark letters, persons with Old Norse names lived in villages dispersed in the northeastern periphery of the Novgorod land. In the eleventh century the area was occasionally visited by Novgorod tribute collectors, but in the twelfth and thirteenth centuries it became the territory of Novgorod colonization. As the Varangians constituted a large part of the administrative machinery in early Rus’, they could easily penetrate into this region and sometimes settle there. The usage of specific family names still in the fourteenth century attests to the conservation of some cultural traditions among the distant descendants of those Varangians.

The birch-bark letters provide the latest date of Old Norse cultural relics. Nothing more than personal names survived up till that time. The latest remains of the runic script belong to the twelfth century. But for the birch-bark letter from Smolensk written by a Scandinavian, there are no objects with runic inscriptions from that time found in Old Russian towns. In the remote areas of Rus’ the runic script seems not to be utterly forgotten however and two finds prove it.

The first find was made in Zvenigorod in the southwestern part of Rus’. It was a slate spindle-whorl with an inscription sigri? on its flattened top, and two crosses and two runes f on its side[874]. The layer in which the spindle-whorl was found is dated to 1115–1130, the time when the settlement started to grow into a town. No other objects of Scandinavian origin were excavated there except for two other spindle-whorls with rune-like inscriptions dating approximately to the same time. One more spindle-whorl with a rune-like inscription was found on an Old Russian fortified site Plesnesk several kilometers from Zvenigorod[875]. It was a strategically important point on the borders of the Old Russian state and it is in Plesnesk where several warrior burials of the late tenth century were unearthed. These burials are believed to belong to warriors of a rather high standing of a Kievan grand prince and some of them could be Scandinavians by origin.

It is tempting to suppose, that the spindle-whorls were inscribed by the descendants of the Varangians who had settled in the region in the late tenth century to defend western borders of the Russian state. The archaic features of the sigri? inscription with rune g of the older futhark could be the result of copying the inscription for several generations. In this case the name Sigri?r must be a constantly occurring name in one of the families. The combination of crosses and f-runes seems however to speak against this surmise. It could not be meaningless for the carver as well as for the owner of the spindle-whorl. Both the cross and f-rune had rather similar symbolic values, although in different religious systems. It was possible to combine both only for a Christian convert who had earlier been an adherent of Old Norse paganism. One also needed an understanding of symbols’ significance, which presupposes survival of old cultural traditions. It is feasible that this group of the descendants of the late tenth-century Varangians living in a remote area of Ancient Rus’ and having no contacts with their homeland managed to preserve family names, remembrances of the runic script in an archaic form, ancestor’s beliefs, and probably a little of their mother tongue as the usage of runes suggests.

Another complex of rune-like inscriptions comes from a fortified site Masko-vichi on the Western Dvina route. This fort on the border with Latvian lands was located several kilometers from the mainstream but was still able to control it. The fort functioned in the twelfth and thirteenth centuries and later became a petty castle[876]. About 110 fragments of bones with graffiti found on the site include pictures of warriors and weapons, and inscriptions. All letter-like graffiti are short and do not form any readable text of several words. These are rather groups of three to six letters, some of which can be interpreted as a word. About 30 inscriptions are obviously Cyrillic while 48 are supposed to be made with runes[877]. Among rune-like graffiti there are several unreadable inscriptions made with ‘mirror’ runes (amulets?), several inscriptions that can be interpreted as personal names, isolated words, and separate letters. Though the reading and the interpretation of the Maskovichi inscriptions are uncertain, there can be little doubt that they were made by persons who had some knowledge or remembrances of the runic script.

Thus in the remote areas of the Old Russian state the descendants of the Varangians who had settled there in the late tenth and eleventh centuries preserved some of their cultural traditions for several centuries while being included in the local political and economic life. The assimilation processes in rural communities seem to have developed much slower than in towns.

(Впервые опубликовано: Runica – Germanica – Medievalia / W. Heizmann und A. van Nahl (Erganzungsbande zum Reallexikon der Germanischen Altertumskunde / H. Beck, D. Geuenich, H. Steuer. Bd. 37.). B.; N.Y., 2003. P. 454–465)