Nomads, Northmen and Slavs. Eastern Europe in the ninth century

Imre Boba




A. The fortress of Sarkel on the Don  70

B. The Proto-Hungarians in the Pontic steppes  74



Between the time of the collapse of the Avar Khaganate (796-803) and the date of the first arrival of the Northmen in Kiev (c. 856), there was a half-century for which the sequence and details of events are not yet fully reconstructed, although they were of crucial importance in the history of Eastern Europe. In the east, by the year 800, the frequent wars between the Khazars and the Caliphate had ended or at least had become less evident. The Muslim lands of the Middle East and especially of Central Asia enjoyed at that time a period of economic prosperity. As a result, the cities of the Khazar Empire became important centers for trade carried on between Europe and the lands of the East, especially as the traditional trade routes across the Mediterranean, connecting Western Europe with the East, were still disrupted by the wars between the Arabs, Byzantium, and the Carolingian Empire. Merchants preferred to take the overland routes across the Slavic-populated areas of Central and Eastern Europe, and then the routes used by the Khazars. [1]


The opportunities offered by the prosperity of the East also attracted the Northmen of Scandinavia, who in the first decades of the ninth century became one of the most active groups of European tradesmen frequenting Oriental markets. [2] Of advantage for the seagoing Northmen were the convenient waterways of Eastern Europe, connecting the Baltic with the Caspian Sea. Although Scandinavians maintained trade contacts with Eastern Europe and the lands further to the East as early as the Gothic period, it was only in the ninth century that regular and large-scale contacts were re-established.


Eastern Europe was, furthermore, not only a crossroads for trade, but also an important source of commodities, such as furs, honey, and wax, sought by foreign merchants. Both the Slavs and the Ugro-Finns benefited from this commerce, as well as from the contacts with people of different cultures. The expansion of trade stimulated the economic growth of the lands of Eastern Europe and promoted the exchange of cultural and



1. See the description of such routes by Ibn Khurdādhbeh, quoted above on p. 28;

cf. also Lewicki, Źródła arabskie, pp. 141-52.


2. J. Arne, La Suède el l'Orient (Upsal, 1914), passim.





political ideas. As a result, the consciousness of national identity became stronger, and conditions were created for the emergence of political organizations as a means of self-preservation. Were it not for the abrupt end of the peace in the Pontic Steppes in the thirties of the ninth century, the first East European states would have emerged in a different form.





The tranquillity of Eastern Europe was disturbed by some nomadic tribes which were at that time unknown in the West. The danger posed by these nomads affected the Khazars from the West, the Crimean Greek cities from the North, and also the Byzantine possessions on the Balkans. Soon afterward, the nomads began the invasions of Western Europe and already in 862 were reported on German soil. [3] Their presence was to remain the dominant political factor of the Pontic Steppes for at least another seven centuries, until the Tartars were finally absorbed into the political structure of Russia.


The names and the ethnic identity of the tribes which first created the disturbance in the early ninth century is still the subject of conjectures. Their appearance is implied in the description made by Constantine Porphyrogenitus of defense measures undertaken by the Khazars on the Don and by the Byzantine Empire in its Crimean possessions. Constantine introduced his description with the statement that in his own times (the middle of the tenth century) there was a city of the Khazars named Sarkel with a garrison of 300 men; then he presented its history, as follows:


Sarkel among them means ‘white house’, and it was built by the spatharocandidate Petronas, surnamed Camaterus, when the Khazars requested the emperor Theophilus that this city should be built for them. For that famous Khagan and the Pekh of Khazaria sent envoys to this same emperor Theophilus and begged that the city of Sarkel might be built for them, and the emperor acceded to their request and sent them the aforesaid spatharocandidate Petronas with ships of war of the imperial navy, and sent also ships of war of the captain-general of Paphlagonia. This same Petronas arrived at Chersonnes and left the ships of war at Chersonnes, and having embarked his men on ships of burden, went off to that place on the Tanais river where he was to build the city. And since the place had no stones suitable for the building of the city, he made some ovens and baked bricks in them, and with these he carried out the building of the city, making mortar out of shingle from the river. [4]



3. Annales Bertiniani, s.a. 862: “Dani magnam regni eius [i.e., of Louis the German] partem caede et egni vastantes praedantur. Sed et hostes antes illis populis inexperti, qui Ungri vocantur, regnum eius depopulantur.”


4. De administrando imperio, pp. 182-85.





After more than a century of conjecture by historians and research by archeologists, the site of Sarkel has been located on the lower Don, [5] the Tanais of medieval sources. The fortress was constructed at some time around 833. It was known to the early Eastern Slavs by the name of Bela Vezha (White Tower, White Fort) (cf. the ‘aspron ospiton’ of Constantine’s Greek text).


Another fragment of Constantine Porphyrogenitus’ work describes the Greek defense measures undertaken on the Crimea:


Now this aforesaid spatharocandidate Petronas, after building the city of Sarkel, went to the emperor Theophilus and said to him: "If you wish complete mastery and dominion over the city of Chersonnes and of the places in Chersonnes, and not that they should slip out of your hand, appoint your own military governor and do not trust to their primates and nobles." For up to the time of Theophilus the emperor, there was no military governor sent from here, but all administration was in the hands of the so-called primate, with those who were called the fathers of the city. The emperor Theophilus took counsel in this matter, whether to send as military governor so-and-so or such-an-one, and at last made up his mind that the aforesaid spatharocandidate Petronas should be sent, as one who had acquired local experience and was not unskilled in affairs, and so he promoted him to be protospatharius and appointed him military governor and sent him out to Chersonnes, with orders that the then primate and everyone else were to obey him; and from that time until this day it has been the rule for military governors in Chersonnes to be appointed from here [i.e., from Constantinople]. [6]


Unfortunately, Constantine Porphyrogenitus did not specify against what enemy the fortress of Sarkel was constructed, or who was endangering the peace around the Greek cities of the Crimea. To solve this problem, various theories have been advanced, some suggesting that the disturbance was caused by the Pechenegs, some considering the Magyars, and still others proposing the Ruses (Slavs or Normans, depending on the views of the proponent).


The theory that the Pechenegs were already in the Pontic Steppes in the thirties of the ninth century is the least acceptable, because, according to available sources and prevailing opinion, they did not cross the Volga, coming from the East, until the eighties of that century. [7] In most recent scholarship, the Pechenegs have been suggested as the cause of Khazar



5. For more recent articles on Sarkel see:

M. I. Artamonov, “Khazarskaia krepost' Sarkel", Acta Archaeologica, 7 (1956), pp. 321-41;

by the same author “Sarkel-Belaia Vezha", Trudy volgo-donskoi arkheologicheskoi ekspeditsii, Vol. I (Moscow, 1958), pp. 7-84;

and his lstoriia Khazar (Leningrad, 1962), pp. 288-323.


6. De administrando imperio, pp. 184-85.


7. Byzantinoturcica, Vol. I, p. 87;

C. A. Macartney, The Magyars in the Ninth Century (Cambridge, 1930), p. 86;

V. Minorsky in his comments to Hudud al-Alam, pp. 312-15, dates the event in the early 90's of the ninth century.





and Greek anxiety, e.g., by T. Lewicki and B. A. Rybakov. [8] The theories, as formulated by them, however, appear to be self-defeating. Lewicki is of the opinion that the Pechenegs were directly and solely responsible for the fact that Sarkel was constructed, but in the same paragraph he shares the general opinion that they did not cross the Volga before the end of the ninth century. A similar inconsistency is apparent in the formulations of Rybakov, who connects the strategic importance of the fortress with the danger posed first by the Magyars and Bulgars (Uturgurs) and later by the Pechenegs, but in the same study demolishes his otherwise plausible theory by asserting that Sarkel was constructed on the northern borders of the Khaganate as a defense measure against the Pechenegs. Unless some new piece of evidence or sound reasoning is brought forward, the presence of, or danger from, the Pechenegs around the Don and the Crimea as early as the thirties of the ninth century cannot be accepted.


Theories connecting Sarkel with the Ruses are based on information contained in the “Life of Stephan of Sugdea” and the “Life of Georg of Amastris”. According to these biographies, Ruses plundered the cities of the Crimea and along the Black Sea coast of Asia Minor as early as the end of the eighth or the first half of the ninth century (before 842). On the basis of this, it is concluded, e.g., most recently by B. D. Grekov and G. Vernadsky, [9] that it was the same Ruses who provoked the Khazars to defense measures, such as the construction of Sarkel. The hagiographic character of the two sources makes it difficult to accept the stories at face value, especially with respect to chronology. It has been suspected that the references to the Ruses in these works are interpolations reflecting Rus attacks against the Paphlagonian and Crimean cities in the late ninth and tenth centuries. The main argument against the credibility of the “Life of Stephan of Sugdea” is that the original Greek version does not refer to Ruses. Such a reference is included only in the Russian translation dating from the fifteenth century. [10]



8. Lewicki, Źródła arabskie, pp. 33-34;

B. A. Rybakov, “K voprosu o roli khazarskogo kaganata XX v istorii Rusi”, Sovetskaia arkheologiia, XVIII (1953), pp. 146, 149.


9. B. D. Grekov, Kievskaia Rus (= Izbrannye trudy, Vol. II (Moscow, 1959), p. 100;

Vernadsky, The Origins, pp. 180-89, where earlier opinions are cited.

See also M. V. Levchenko, Ocherki po istorii russko-vizanriiskikh otnoshenii (Moscow, 1965), pp. 45-58. Levchenko is in favor of an earlier dating of the Rus attacks, but much of his argumentation is based on a mistranslation of Ibn Hurdadbeh’s description of the routes used by the Rus merchants. Where Ibn Hurdadbeh has ‘Bahr ar-Rum’, i.e., ‘the Sea of the Romans’, or ‘the Mediterranean’, Levchenko reads ‘more russkoe’ (p. 45).


10. G. da Costa Louillet, “Y eut-il des invasions russes dans l’Empire Byzantin avant 860?”, Byzantion, XV (1940-41), pp. 231-48.

See also George Vernadsky’s article “The Rus in the Crimea and the Russo-Byzantine Treaty of 945” in Byzantina Metabyzantina, I (1946), pp. 249-59. In comments on this article (on pages following the article) Henri Gregoire restated the current opinion that “the life of St. Stephen of Surozh has absolutely no value . . .". Gregoire has pointed out that A. A. Vasiliev also doubted the value of the two Lives for the study of the Ruses (Ibid., pp. 259-60).





However, before dismissing the theories that some militant Ruses were already present in the Don region in the third decade of the ninth century or even earlier, we should explore the possibilities of a connection between the construction of Sarkel and the presence of Rus envoys in Constantinople in 838/9. This date is very close to that of the alleged Rus raids against the Black Sea cities which are supposed to have taken place sometime before 842. Between the two facts, that of the embassy and that of the raids, a direct relation is possible, especially if we recall that King Louis I of Germany had suspected that the envoys were on a spy mission. [11]


Spying envoys were commonly employed both by the nomads and the Northmen as a preparation for armed invasions. But even the possibility of a sinister reason behind the Rus mission of 838/9, and the probability of Rus attacks against the Black Sea cities even before 842 will not explain the construction of Sarkel as a defense measure directed against the Ruses in the early thirties of the ninth century. There must have been some other tribes at that time in the Steppes - namely, those who in 839 prevented the Rus envoys to Constantinople from returning home.


Furthermore, a fortress such as Sarkel would scarcely have been effective against the Ruses, who used the seas and rivers for their peaceful, as well as warlike, exploits. An effective defense against the Ruses would have been a garrison at the mouth of the river Don. Khazar defenses, in fact, did exist at the mouths of rivers, and they are mentioned in the so-called Jewish Correspondence exchanged between the Jews of Spain and the Khagan of the Khazars sometime around 960.


“I am defending the Mouth of the river”, writes the Khagan, “and am preventing the Ruses, who are arriving by boat, coming from the sea, from going against the Muslims. . ." [12]


Sarkel could not have served such a purpose, at it was located halfway between the mouth of the Don and the portage to the Volga, while, on the other hand, it could have served as a base for troops in control of river crossings used by the nomads.


The theories relating the upheavals of the early ninth century to the presence of the Ruses in the neigborhood of the Don and of the Crimea, therefore, cannot be sustained, unless new independent sources were to confirm the statement of the hagiographic Russian ‘Lives’ of the fifteenth century.


The third and commonly accepted theory holds that the Pax Chazarica in Eastern Europe was disrupted by the appearance of the Magyars. Most



11. See above, p. 24.


12. P. K. Kokovtsev, Evreisko-khazarskaia perepiska v X veke (Leningrad, 1932);

cf. an English translation of the fragment by D. M. Dunlop, The History of the Jewish Khazars (Princeton, N. J., 1954), p. 240.





of the modern formulations of this theory are simply rephrasings of the observation made by J. Marquart in 1903, [13] although the original idea was not necessarily his own. [14] According to Marquart, there was a connection between the construction of Sarkel and Ibn Rūsta’s remark that


“in former days the land of the Khazars was surrounded by a ditch as a defense against the ‘Majgharija’ and other nations adjacent to the [Khazar] country”.


On the face of it, the connection between the erection of Sarkel in the thirties of the ninth century and the remark of Ibn Rūsta seems to be acceptable. This theory, however, proves to be even more complex than those already eliminated concerning the Pechenegs and the Ruses. Marquart translates Ibn Rūsta’s Arabic ‘Majgharija’ by the term ‘Magyaren’, thus identifying the unruly tribes of the early ninth century with the Hungarians, whose own name today is ‘Magyar’. Upon a close confrontation of the sources, however, it appears that the theory of Marquart and those derived from his are based on false assumptions.


With this statement we enter the most entangled controversy of Hungarian protohistory. [15] Without attempting to solve the problems of the origin of the Hungarians, it must be emphatically stated that the application of the names ‘Magyar’, ‘Hungarian’ or ‘Ungar’ to any tribal formation participating in the events of the ninth century is an anachronism or is based on an unsubstantiated assumption that the Hungarians (Magyars) of today can be identified with a single group active in the steppe zone in the ninth century.





In order to comprehend the dangers arising from the proposed equation of the names ‘Majgharija’ and ‘Magyars’, the following explanation should be made. The Hungarians of today are known in most languages by a name which is derived from an old Slavic term ‘*Ǫgure’ (phonetically; ‘Ongure’). From the Slavic form the Greeks in the ninth and tenth centuries adopted the form ‘Ouggroi’ (phonetically: ‘Ungroi’). The German-Latin form is ‘Ungari’, ‘Ungri’. The same name, transmitted by the Germans or the Greeks to the West, is used, inter alia, also by the English-speaking world in the ‘Hungarians’. The source of all these forms, the



13. Marquart, Streifzüge, pp. 27-28.


14. The same concept was expressed earlier by Gyula Pauler in A magyar nemzet története Szent Istvánig (Budapest, 1900), p. 14.


15. A most recent survey of the relevant theories is offered by Tamás Bogyay, “Forschungen zur Urgeschichte der Ungarn nach dem 2. Weltkrieg”, Ural-Altaische Jahrbücher, XXIX (1957), pp. 93-114. For an earlier survey see A magyarság őstörtenete, edited by Lajos Ligeti (Budapest, 1943). For a more recent study by Bogyay cf. p. 101, note 40.





‘*Ǫgъre’ is the old Slavic adaptation of the Altaic tribal name of the ‘Onogurs’. [16] The latter were a numerous group of nomads, active from the Hunnic invasion until the tenth century in various areas of Eastern Europe.


Although the name ‘Onogur’ is used today for the Hungarians by all of their neighbors, there is no derivative of that name known to the Hungarians themselves. The name which the Hungarians use to denote themselves is ‘Magyar’. This name is directly related to the name ‘Majghari’, used by Ibn Rūsta. The two forms, ‘Magyar’ and ‘Majghari’ are derived from a ‘*Mogyeri’ or ‘*Megyeri’.


After these linguistic preliminaries, it should be noted that, whereas the names ‘Hungarian’ (derived from ‘*Ǫgъry’) and ‘Magyar’ (derived from ‘Mogyeri’) are today applied to only one nation, in the ninth century the derivatives from ‘Ǫgъre’ and from ‘Mogyeri’ were applied to two distinct, unrelated tribal formations. A source of primary importance, the anonymous Persian geographic work known as udud al-'Ālam (“The Regions of the World”), knows the Majghari and the Onogurs as two groups acting independently, but simultaneously, in the ninth century somewhere north of the Black Sea. [17] There is also the Persian author Gardīzī, who makes the same distinction. [18]


Both the ‘Mogyeri’ and the ‘*Ǫgъry’ participated in the final formation of the modern Hungarian (Magyar) nation, but the merger of these two ethnic groups did not take place until the end of the ninth century and, even at that time, not all the ‘Ǫgъre’ or all the ‘Mogyeri’ took part in the merger. A disassociation between the ‘Majgharija’ of the Oriental sources and the modern name ‘Magyar’ is, therefore, as important as a distinction between the modern ‘French’ and early medieval ‘Frank’. Consequently, neither the name ‘Magyar’ nor the name ‘Hungarian’ (Ungarn, etc.) should be used for any tribal formation of the ninth century.


For the sake of precision, one should speak of ‘Hungarians’ only from the time of the conquest of the Carpathian Basin, i.e., c. 900, and even then in the political rather than ethnic sense. The “Covenant of Blood,” preceding the conquest, was the formal union of the heterogeneous tribes into one political entity, but this was not yet the formation of an ethnic unity. Constantine Porphyrogenitus was correct when he observed that in his times, i.e., in the middle of the tenth century, the tribes forming the federation were learning each other’s languages. [19] The Onogur element spoke an Altaic; the Majghari, a Finno-Ugric dialect.


On the basis of the observations above, the suggestion may be made



16. János Melich, “Über den Ursprung des Namens Ungar”, Archiv für slawische Philologie, XXXVIII (1906), pp. 244-50.


17. Hudud al-Alam, text, pp. 101, 160, 162-63; V. Minorsky’s comments, pp. 465-71.


18. Gardizi, quoted i.a. by C. A. Macartney, The Magyars in the Ninth Century (Cambridge, 1930), p. 206.


19. De administrando imperio, pp. 174-75.





that the theories linking the construction of Sarkel to the appearance of the ‘Magyars‘ are to be accepted only as tentatively valid, with the condition that instead of ‘Magyars’ (‘Hungarians’, ‘Ungarn’), we will understand ‘Proto-Magyars’, i.e., some tribes which eventually, entirely or in part, combined to produce the modern Hungarian nation. With this statement we actually have posed again the question: who were the people who disturbed the peace in Eastern Europe in the early ninth century? In our opinion the disturbances were caused both by the Onogurs and by the ‘Majghari’, two independent groups active in the Pontic Steppes simultaneously.


[Previous] [Next]

[Back to Index]