Cumans in Southern Dobrudja

Some remarks on the Second Bulgarian Empire

during the twelfth and thirteenth centuries

THOMAS BRÙGGEMANN

The end of the First Empire - Byzantines, Bulgarians, Vlakhs, and Cumans
The downfall of the First Bulgarian Empire and the re-integration of its territory
between the Balkan Mountains and the Danube at the beginning of the eleventh
century achieved just a short respite for the Byzantine Empire in the Balkans.1
Soon the Byzantines were confronted with permanent uprisings of the resident
population, because they had confined their re-conquest mainly to the elimina-
tion of the Bulgarian elite and taking over its military infrastructure, which, if
they were unable to control, they destroyed.2 Constantinople had to face insur-
gents on two fronts. On the one hand there were dispersed parts of the remaining
Bulgarian population, which due to its loss of political leadership was beset by

1 The Byzantine Emperor Basileios II Bulgaroktonos had brought about the décliné of
the First Bulgarian Empire by his victory at the Kimbalongos pass in the gorge of Klei-
don, where he defeated Tsar Samuel and his troops in luly 1014; Th. Brüggemann, "Die
Staatswerdung Bulgariens zwischen Rom und Byzanz. Migration, Christianisierung
und Ethnogenese auf der Balkanhalbinsel (6.-11. lahrhundert n. Chr.)," in S. Conrad et
al., eds., Pontos Euxeinos. Beiträge zur Archäologie und Geschichte des antiken Schwarzmeer-
und Balkanraumes, Manfred Oppermann zum 65. Geburtstag, Langenweißbach 2006, 468
with notes 66-70; R.-J. Lilie, Byzanz. Das zweite Rom. Berlin 2003, 250, also H. Ditten,
Ethnische Verschiebungen zwischen der Balkanhalbinsel und Kleinasien vom Ende des 6. bis
zur zweiten Hälfte des 9. Jahrhunderts. Berlin 1993, 91ff.
2 For the "Byzantine Intermediate", between 1014 and 1185, see the description of Geor-
gios Akropolites, Chronike syngraphe, in Georgii Acropolitae opera. ed. A. Heisenberg,
Leipzig, 1903, 18 6-10; in général, see I. Vâsâry, Cumans and Tatars. Oriental Military in
the Pre-Ottoman Balkans, 1185-1365. Cambridge 2005,13-17; G. Ostrogorsky, Geschichte
des Byzantinischen Staates. Munich 1963; and F. Grabler, ed., Abenteurer auf dem Kaiserth-
ron. Die Regierungszeit der Kaiser Alexois II, Andronikos und Isaak Angelos (1180-1195) aus
dem Geschichtswerk des Niketas Choniates. Byzantinische Geschichtsschreiber 8. Graz
1958,171ff.

57
 THOMAS BRÜGGEMANN

social crisis and economic suffering. On the other hand, there was the endemic
old-Balkanic Vlakh populace,3 who, being transhumant animal-breeders, mainly
inhabited the mountainous regions. The latter had been a source of continuous
trouble for the Bulgarian tsars and especially for their sedentary rural subjects al-
ready, but now the Byzantines,4 who obviously lacked a consistent political con-
cept for the region, received the Bulgarians in addition, who were thrown from
their fairly stable life into chaos. Due to the fact that the hegemony of Constan-
tinople had been restored only superficially, one uprising was followed by the
next immediately.5 Therefore, in spite of their own weakness, the Byzantines
owed their fragile 'control' and the at best formal existence of the Byzantine 'Da-
nube-border' entirely to the poor organization and the total lack of powerful po-
litical-military leadership of the rebels.6

3 The old-Balkanic population, which mostly consisted of transhumant pastoralists, is

mentioned under this name for the first time in sources of the tenth century; see Annae
Comnenae porphyrogenitae Alexias, 2 vols. ed. A. Reifferscheidt, Leipzig 1884. 2: 8 , 1 1 - 1 3 .
The sources clearly distinguish the Vlakhs from the sedentary rural population of the
Balkans. They settled and wandered mainly in the area between Thessaly and the Bal-
kan Mountains, the Thessalian region is even known as "Megale Vlachia"; see M.
Blagojevic, "Vlachen" in Lexikon des Mittelalters 8 (1997), 1789; G. Schramm, Eroberer
und Eingesessene. Geographische Lehnnamen Sùdosteuropas im 1. Jahrtausend n. Chr. Stutt-
gart 1981, passim; and T. J. Winnifrith, The Vlachs. The History of a Balkan People. London
1995,3-7.
4 The Byzantine conquerors were either unwilling or unable to integrate the former Bul-
garian territory. At first they did not seem to have a concept for the Bulgarian lands at
their disposal which went beyond the "pure" military re-conquest: this caused the re-
futation of the new "old rules" by the resident population, because the Byzantines
failed to offer their new subjects any rewarding political, social or economic incentives.
For the subjected people, just the political-military leaders changed, and the Byzantines
remained an "alien element", a "head" without connection to its "body". The break-
down of Byzantine supremacy up to the Danube therefore was bound to happen be-
fore long.
5 For the connections between Byzantines and Cumans in the southern Dobrudja during
the "Byzantine Intermediate" throughout the eleventh and twelfth century, which
were not only of a military nature, see H. Gockenjan, "Kumanen" in Lexikon des Mitte-
lalters 5 (1991), 1568, 1068; V. Spinei, The Great Migration in the East and Southeast of Eu-
rope from the ninth to the thirteenth Century. Cluj-Napoca 2003, passim, also P. Diaconu,
Les Coumans au Bas-Danube aux IXe et XIIe siècles. Bucharest 1978,14-21.
6 Soon after the re-integration of the former Bulgarian territory, the Byzantines, due to
their neglect of governance, were confronted with uprisings of the residual Bulgarian
population, whose living conditions especially in political and economic matters wor-
sened rapidly under Byzantine rule. The first, which took place in 1040/41, was led by
Peter Odeljan, who called himself a grandson of the last Bulgarian Tsar Samuel. When
the Bulgarians rose again in 1072 they had to look for a leader firstly, because no direct
descendants of the old Bulgarian dynasty were left. Bodin, son of the Serbian prince
Zeta, was appointed as "Bulgarian" Tsar at the Prizren court, see I. Dujcev, "Bulgarien"
in Lexikon des Mittelalters 2 (1983), 914-927 at 920. These uprisings of the Bulgarians

58
 CUMANS IN SOUTHERN DOBRUDJA

In this state of affairs it was just a question of time before immigrants from the
north came to disregard the tenuous status quo. Pechenegs,7 Uzes, and especially
Cumans crossed the Danube and settled on Byzantine territory, and occupied the
land almost unhindered by the Byzantines. In particular, the Cumanian upheav-
als in 1087, 1094, 1109, and finally 1160, put serious pressure on the Empire and
could be checked only with great difficulty.8 The attempts of individual Cuman
tribes for separatism and the growth of the tribal chieftains' power at the expense
of the existing federate khans can explain why, during the first half of the twelfth
century, the Cuman federation split into western and eastern branches. The mili-
tary activities of the western Cumans during this period were of great impor-
tance, their marauding expeditions into the territories of Byzantium, Hungary,
Russia and Poland caused considerable disturbance. The eastern Cuman federa-
tion had a much larger territory for itself, and the archaeological evidence also
suggests that it was more densely populated by Cumans. The power of the
Cumanian tribal and clan aristocracies was backed by retinues of the warrior-
class.9 Its members are also documented as mercenaries in foreign lands, includ-
ing the courts of Georgia and Serbia, and later of Hungary and Bulgaria.
Both as allies and enemies, the Cumans had a great influence on the internal
development of their neighbouring states. They stood in close contact to the
Kievan Rus at an early stage, being relatives of Russian princes and even exercis-
ing border patrols for them, but they also attacked their own allies regularly.10
King David II of Georgia is reported to have settled around 40,000 Cumans in
1118, after they had supported him militarily against the Seljuks and the reluctant
Georgian aristocracy.11 Other Cuman tribes either formed the substrate during

show that the Byzantine Empire was not able to control the re-conquered territories
enduringly.
7 For the Pechenegs, who had crossed the Danube since the eleventh century, when they
also came into closer contact with the Byzantine Empire, see O. Schmitt, "Die Pet-
schenegen auf dem Balkan von 1046 bis 1072," in: S. Conrad et al., eds., Pontos Euxei-
nos. Beiträge zur Archäologie und Geschichte des antiken Schwarzmeer- und Balkanraumes,
Manfred Oppermann zum 65. Geburtstag, Langenweißbach 2006, 473-490 with further
sources and literature.
8 Anna Komn. Alex. II 8, 28-31; Väsäry, Cumans and Tatars, 21 with note 28.
9 According to the codex Cumanicus (Codex cumanicus bibliothecae ad templum Divi Marci
Venetiarum, ed. G. Kuun, Budapest 1880.), they were called nögers or nökörs by the Cu-
mans, see P. B. Golden, "The Codex Cumanicus," in H. Paksoy, ed., Monuments of Central
Asia. Istanbul 1992, 33-63 passim, as well as D. Drüll, Der Codex Cumanicus. Entstehung
und Bedeutung. Stuttgart 1980, passim.
10 The Kievan Rus called the Cumans "wild Polov'cer". One of the Russian attacks, which
Igor Svjatoslavic undertook in 1185, passed into the so-called "Song of Igor". See O.
Pritsak, "The Polovcians and the Rus," Archivum Eurasiae Medii Aevi 2 (1982), 321-380;
and P. B. Golden, "The Question of the Rus' Qaganate," Archivum Eurasiae Medii Aevi 2
(1982), 77-97.
11 He reigned from 1089 to 1125, see, in general, P. B. Golden, "Cumanica I: The Qipiaqs
in Georgia," Archivum Eurasiae Medii Aevi 4 (1984), 45-87; and A. Pälöczi-Horväth, Pe-

59
 THOMAS BRÜGGEMANN

the ethnogenesis of Turkic populations in the Northern Caucasus or invaded

Hungary, first in 1091, when King Ladislaus I was still able to defeat them.12
The Cumans, who practised agriculture as a subsistence economy, should be
called semi-nomads, because of their regular raids against the sedentary rural
population to capture luxuries and slaves.13 The assumption that the Cumans
persisted in semi-nomadic habits is also proved by the existence of settlement-
centres such as Sudaq, Asaq (Asov), and Saqpin (Itil), although one should not
call them "cities".14 Starting from these centres they maintained trade connections
to Central Asia,15 Egypt,16 and via the Byzantine Crimea even to the European
Mediterranean already before their westward migration.17 Written sources men-

chenegs, Cumans, lasians. Steppe Peoples in Medieval Hungary. Budapest 1989, 313-333,
with further sources and literature.
12 He ruled Hungary from 1077 to 1095. For the legend concerning his battles against the
Cumans, see E. Szentpétery, ed., Scriptores rerum Hungaricarum, tempore ducum et re-
gumque stirpis Arpandianae gestarum, 2 vols. Budapest 1937-1938.1: 366-427; 2: 507-527,
also Gy. Györffy, "Die Nordwestgrenze des Byzantinischen Reiches und die Ausbil-
dung des 'ducatus Sclavoniae'," in P. Brière and X. de Ghellinck-Vaernewyck, eds.,
Mélanges offerts à Szabolcs de Vajay à l'Occasion de son Cinquantième Anniversaire. Braga
1971,295-313.
13 Györffy, "Die Nordwestgrenze des Byzantinischen Reiches"; Golden, Cumanica I, 4 5 -
87; and A. Pâlôczi-Horvâth, "L'immigration et l'établissement des Comans en Hon-
grie," Acta orientalia Hungarica 29 (1975), 313-333.
14 P. B. Golden, "Cumanica II: The Öberli (Öperli). The fortunes and misfortunes of an
Inner Asian Nomadic Clan," Archivum Eurasiae Medii Aevi 6 (1986), 5-29.
15 The codex Cumanicus proves in this respect not only intense trade relations to Central
Asia and Egypt, but allows us to reconstruct the Cumanian language via the docu-
mented terms and expressions. Besides that, the codex shows via its terminology and
foreign words where the Cumans came from and the hierarchy of their trading part-
ners. The economic terminology reveals Chinese and Persian terms mostly, but no
Greek; this did not change even by way of the permanent Cuman contacts with the By-
zantines from the tenth century. Their origin, ethnogenesis, and priorities become clear
in this context. See Golden, The Codex Cumanicus, passim, Th. Brüggemann, "From
Money-Trade to Barter? Some remarks on nomads and the changing economy on the
Byzantine Chersonesos (10^-13^ century)," Academia. The Magazine of the Polish Acade-
my of Sciences 17 (2008), passim; Th. Brüggemann, "Vom Geld- zum Tauschhandel. Die
byzantinische Krim zwischen Urbanität und Nomadismus," in I. Breuer, ed., Nomaden
in unserer Welt. Nomaden und Seßhafte 3, Wiesbaden 2007, passim·, and A. Bodrogligeti,
The Persian Vocabulary in the Codex Cumanicus. Budapest 1971, passim.
16 For a general overview of the Cuman-Mamluk connections in the thirteenth century,
see P. B. Golden, "Cumanica IV: The Cumano-Qipiaqs Clans and Tribes," Archivum
Eurasiae Medii Aevi 9 (1995/97), 99-122; and U. Haarmann, "Der arabische Orient im
späten Mittelalter," in U. Haarmann, ed., Geschichte der arabischen Welt. Munich 2004,
215-263.
17 Their presence in the Crimea and Cherson is proved from the second half of the ele-
venth century. For their trade relations with the Kievan Rus and the Byzantine Empire,
see Brüggemann, "From Money-Trade to Barter?" passim·, Brüggemann, "Vom Geld-
zum Tauschhandel," passim; A. I. Romancuk, Studien zur Geschichte und Archäologie des

60
 CUMANS IN SOUTHERN DOBRUDJA

tion the names of a number of Cumanian chieftains, amongst whom Boniak

(Bőnek) Khan was the most significant personality. It was he who led the army
which gave the Byzantines assistance against the Pechenegs in 1091, and of which
a part, after the victory at Mount Levunion, turned towards Hungary.18

Between the Empires - Cumans, the Balkans, and the Byzantine Intermediate
As holders of imperial tenures, some Cuman tribes were put in charge of defen-
sive duties for the Empire and in this capacity were even transferred to Thrace,
Macedonia, and Asia Minor.19 However, by taking its inconvenient Cuman con-
temporaries into imperial service Byzantium just bought a short period of peace.
Soon Cuman horsemen, whose loose confederation stretched from the northern
steppe regions to the territories south of the Danube anyhow, unified with the
Bulgarian and Vlakh populaces of the southern Dobrudja with ease.20 The final
collapse of Byzantine rule in the southern Dobrudja, rudimentary at least since
the middle of the twelfth century, just needed organized military leadership to
bundle the merged but disparate Bulgaro-Vlakh population and the marauding
Cumans into a unit with mutual interests and aims.21 When in 1185 the Byzan-
tines were bound up with the invasion of the Normans,22 who took Thessaloniki

byzantinischen Cherson. Leiden 2005, 11-19; M. Strâssle, Der internationale Schwarzmeer-

handel und Konstantinopel im Spiegel 1261-1484 im Spiegel der sowjetischen Forschung.
Frankfurt 1990, 23-38; and finally G. Bràtianu, La Mer Noire: Des origins à la conquête ot-
tomane. Munich 1969,46ff.
18 A historical overview can be found in Lilie, Byzanz, 328ff.
19 Schmitt, Die Petschenegen auf dem Balkan, 473-490. The Byzantine "approach" to inte-
grating the uninvited immigrants followed the conventional course. On the one hand,
Byzantium was forced to "embrace" the Cumans by the assignation of land and the
transfer of duties and titles, quite similar to the Byzantine treatment of the Pechenegs
before. On the other hand, some Cuman tribes were moved to regions which the By-
zantines though would hinder and restrict the Cumanian nomadic way of life. This
would force the Cumans to concentrate on their own economic survival and therefore
they would be kept from raids against the Byzantine central power and its rural inha-
bitants. That such resettlements were mostly unsuccessful is shown by H. Ahrweiler,
"Byzantine Concepts of the Foreigner: The Case of Nomads." in H. Ahrweiler and A. E.
Laiou, eds., Studies on the internal diaspora of the Byzantine Empire. Dumbarton Oaks
1998, 1-16; and P. Charanis, "The Transfer of Population as Policy in the Byzantine
Empire," Comparative Studies in Society and History 3 (1961), 140-154.
20 Golden, "Cumanica I," 45-87; Golden, The Question, 5-29; and Diaconu, Les Coumans au
Bas-Danube, 14-21.
21 The process is comparable with the ethnogenesis of the First Bulgarian Empire be-
tween the sixth and seventh centuries, when one could observe the merging together of
a military and politically well organized Turkic elite with a Slavic majority that was in
this respect undeveloped. See Brüggemann, Die Staatswerdung Bulgariens, 462ff.
22 Nicetae Choniatae Historia, 1: 368, 38ff; Georg. Akr. Chron. 1: 18, 6 - 1 0 for the Norman
invasions and the reign of Isaak II Angelos; also R.-J. Lilie, Handel und Politik zwischen
dem byzantinischen Reich und den italienischen Kommunen Venedig, Pisa und Genua in der

61
 THOMAS BRÜGGEMANN

and rushed for Constantinople, and the pressure of the Seljuks in Asia Minor out-
side the Balkans, finally, two charismatic leaders, the brothers Petâr and Asen,
took the chance to start the ultimate uprising against the Byzantine Empire, in the
Tarnovo area. As the Byzartine Emperor Isaakll Angelos was not able to check
the rebellion, he had to agree to a peace-treaty with the insurgents. This rebellion
not only unified Bulgarians, Vlakhs and Cumans, but also led to the establish-
ment of the Second Bulgarian Empire.23
Although it is not the task here to trace the process of migration and settle-
ment of the Cumans north of the Danube,24 the so-called Cumania, some remarks
should nevertheless be made. At the end of the eleventh and the beginning of the
twelfth century, political leadership was in the hands of the tribes living to the
west of the Dnjeper,25 but the land of Cumania beyond the Dnjester was less well
known to the Byzantines and Westerners. Therefore literary information about
the names and dispositions of the several Cuman tribes that settled on the grassy
steppe to the north of the Black Sea is rather sparse. However, at least some of the
tribal centres can be ascertained by information provided by the Russian chroni-
cles. By comparing these with the distribution of archaeological finds, a rough
picture emerges of the areas of settlement.26 The southern frontier was the Da-

Epoche der Komnenen und der Angeloi (1081-1204). Amsterdam 1984, I 9-120, esp. 99ff;
and Grabler, Abenteurer aufdem Kaiserthron, 171f.
23 The sources show clearly that the motives for this insurgence, both from the brothers
(appreciation by the Byzantine Emperor and assignation of land) and from the popula-
tion (taking back the increase of taxes and fees by the Emperor), had a short-sighted,
individual nature only. See Georgii Acropolitae Opera, ed. A. Heisenberg, Leipzig 1903,
Vol. 1 : 1 8 , 1 2 - 1 3 . See also Nicetae Choniatae Historia, 2 vols. ed. H. van Dieten, Berlin -
New York 1975,1: 368, who says that "because of his pettiness he (Isaak II) did not no-
tice that he caused trouble [...], and incited to war the barbarians living in the Balkan
Mountains against himself and the Rhomaioi. [These barbarians] were formerly called
Mysians, and now they are named Vlakhs". See below note 66 also. On this back-
ground it remains unintelligible why Vasary, Cumans and Tatars, passim, calls this rebel-
lion a "liberation movement", especially since he indicates that the upheaval had no
"nationalistic" notions, as for example at page 21 with note 28. Because of the several
"trivial" reasons for the uprising one cannot truly speak of a planned military, politi-
cal or social "national movement" against the rule of the Byzantines. The result of this
revolt was the "liberation" of the southern Dobrudja and the "foundation" of the
Second Bulgarian Empire, but it has to be emphasized that this happened rather acci-
dentally; therefore the "liberation movement" of Vasary has to be rejected.
24 For a general overview, see Golden, "Cumanica II," 5-29; Diaconu, Les Coumans au Bas-
Danube, passim.
2s See Golden, "Cumanica II," 5-29; Golden, "Cumanica I," 45-87.

2 6 See, therefore, P. B. Golden, "Aspects of the nomadic factor in the economic develop-

ment of Kievan Rus," in I. S. Koropeckyi, ed., Ukrainian Economic History: Interpretive

Essays, Cambridge 1991, 36-73; Golden, "The Question of the Rus'," 77-97; and Pritsak,
"The Polovcians and the Rus'," 321-380, with further sources.

62
 CUMANS IN SOUTHERN DOBRUDJA

nube, and it comprised what was later to become Wallachia and Moldavia.27 To
the north the Russian principalities formed the frontier, and to the east it
stretched as far as the Volga. Groups of settlements have been identified by the
lower reaches of the Dnjeper, on the northern shore of the Sea of Azov, in the
Donets Basin, by the lower Don, and in the region of the river Kuban. In the
south, the steppes of the Crimean peninsula and the Italian trading ports of the
Black Sea which had formerly been under Byzantine suzerainty fell to Cuman
domination.28 From the end of the eleventh century the Cumans sought to in-
crease their wealth not just by plundering raids but also by systematic collection
of tributes from the cities and sedentary peoples that had come under their sway.
They also supervised the trade that passed along the steppe routes, with guaran-
tees of freedom of movement for merchants even in times of war, and a large
number of craftsmen employed at their headquarters.29
The climatic and ecological conditions in the southern Dobrudja were less
suitable to the Cuman nomadic way of life than those north of the Danube, as be-
comes evident not only from the limitation of the winter-pastures in Bulgaria, but
also from the distribution and density of Cuman burials in the region. Namely,
although the Cumans obviously used the lands which had been abandoned by
the "former" Bulgarians after the downfall of the First Bulgarian Empire, and
which were under the, at best, nominal rule of the Byzantine Empire, for seasonal
pastures during the eleventh century, there are few archaeological traces to show
their presence in that area. While the precise ethnic classification of burials of
non-sedentary groups becomes more problematic for the twelfth and thirteenth
centuries, in contrast to the tenth and eleventh,, there are still some invariable
characteristics, especially in the case of the Cumans.30 For example, burials of
horses next to human remains are common in Cuman cavalry graves, and embel-

27 The approximate extent of the lands of Cuman "settlement" between the Kievan Rus
and the Black Sea is revealed not only by archaeological traces, but by information de-
rived from Western chronicles from the fourteenth century, like the English annalist
Iohannes (A. F. Gombos, Catalogas fontium históriáé Hungaricae aevo ducum et regum ex
stirpe Arpad descendentium ab anno Christi DCCC usque ad annum MCCCI, 4 vols. Buda-
pest 1937-1943,1: 1330), who describes utraque Cumania, or Cornelius Zantfliet (Gom-
bos, Catalogas fontium, 1: 796), who also mentions the expansion of utramque Cumaniam.
28 For the military, political, and economic processes and disturbances in the Byzantine
Crimea throughout the tenth to the thirteenth centuries, see Brüggemann, "From Mon-
ey-Trade to Barter?" passim; and Romaniuk, Studien zur Geschichte, passim.
29 Ibid, passim; Golden, "Aspects of the nomadic factor," 23-31; Golden, "The Question of
the Rus'," 77-97; Strássle, Der Internationale Schwarzmeerhandel, 23-38; and Brátianu, La
Mer Noire, 46ff.
30 ,For recent work on the material Cuman traces in the southern Dobrudja, i.e., modern
Bulgaria, see especially R. RaSev, "Kumanite na jug ot dunav no archeologizeski Dan-
ni," in Sb. 800 g. ot pobedata na zar Kalojan pri Odrin 1205 g. Sumen 2006, 21-28.

63
 THOMAS BRÜGGEMANN

lishments of grave-statues for deceased warriors were habitual also.31 A putting

together of all the occasional finds from the southern Dobrudja which could be
attributed to the Cumans with certainty or at least high probability makes it pos-
sible to modify the "classical" view of an isolated and occasional Cuman presence
in the Balkans. Although this approach cannot confirm whether the Cumans
regularly settled south of the Danube, switching from a nomadic to a sedentary
form of life, their presence since the eleventh century can no longer be called "in-
discriminate" or "accidental". The nomadic Cumans, known for their large herds,
seem to have found favourable conditions for their way of life in the southern
Dobrudja only during the winter months. If information concerning their pres-
ence in the Balkans gives particular dates, however, it could be shown that these
are limited to this period of the year.32 That becomes evident both in the distribu-
tion of their material assets and in the text of a recently discovered Bulgarian in-
scription from Preslav.33 The Cuman cavalry, which was crucial for all rulers of
the Second Bulgarian Empire as it was previously for the Byzantines against the
Pechenegs, could apparently be employed only at times when larger Cuman
tribal groups stayed in the region for their winter pasture.34

The Second Empire - Cumans, Bulgarians and Vlakhs

The debate concerning the ethnic origins of the brothers Petar and Asen, and
their immediate successor, Kaloyan, who were the founders and first representa-
tives of the new ruling dynasty in the Second Bulgarian Empire, is still a contro-
versial one.35 The interpretation of the name Asen as Cuman seems to be com-
monly accepted, but the historical conclusions that could be drawn from this
have been very different. Some have made the Asenids Bulgars, or at best Cu-
mano-Bulgars,36 others have made them Russians or Russo-Cumans, who were

31 These are similar to the Russian "Kamennaja baba". See, in this respect, R. RaSev,
"Prab'lgarski li sa 'Kamennite Babi' ot endusche," Musei i Pametnizi na Kulturata 1
(1972), 17-20.
32 The investigations of D. Raskovskij, "Rol' polovcev v vojnah Asenej s Vizantijskoj i La-
tinskoj imperijami v 1186-1207 godah," Spisanie na B'lgarskata Akademija na Naukite
5 8 / 2 9 (1939), 203-211 have proved their presence for winter pastures during the
months November to April.
33 Although this inscription is not datable precisely, it derives most probable from the
twelfth or thirteenth century. It proves not only that the Cumans retained their nomad-
ic skills at least in the southern Dobrudja, but also that they did so even in the time of
the Second Bulgarian Empire, see Rasev, "Kumanite na j u g " 21.
34 For the inscription and its text, see Totev (2006), cited in Rasev, "Kumanite na jug," 21.
35 A comprehensive overview of the scientific debate is given in Vasary, Cumans and Ta-
tars, 33-42, esp. 34-38 with notes 80 to 99.
36 Bulgarian scholars especially have tried to prevent the ruling dynasty of the Second
Bulgarian Empire from being of Cuman origin. See, for example, V. N. Zlatarski, Istori-
ja na b'lgarskata d'rzava prez srednite vekove, 3 vols. Sofia 1918-1940, 2: 424; and idem,
"Potekloto na Petra i Asenja, vodacite na v'zstanieto v 1185 god'," Spisanie na B'garskata
Akademija na Naukite 45 (1933), 7-48, whose "kumano-b'lgarski znaten rod" is hard to

64
 CUMANS IN SOUTHERN DOBRUDJA

practically Bulgars.37 Finally the fact that Asen was a Turkic name was sufficient
to make him and his descendants Cumans.38
The name Esen was widespread among the Turkic peoples. 39 All the Greek
forms of the name (Asan, Asanas) and the Russian ones (Osen, Asin) can be satis-
factorily explained by reference to Turkic "Esen". 40 Thus, while the Turkic origin
of the name Asen can be taken for granted, this fact cannot prove that its bearer
was undoubtedly Cuman. Asen's Turkic name must be reconciled with the fact
that the sources unanimously testify to his being a Vlakh. Since the Vlakhs lived
in the Balkans before 1185 and settled on the left bank of the Danube only from
time to time, only Turkic groupings of the Balkans can be considered as having
lent Turkic personal names to the Vlakhs.41 Given that the Cumans were the most
frequent visitors to the Balkans, and that Cuman nobles of the eleventh and
twelfth century had the same name as Asen, the most probable explanation for
Asen's Turkic name is that it came from the Cumans. Some Cuman tribes must
have remained in the Balkans even during the summer months and merged with
the Vlakhs. The fact that the nomadic way of life of both groups displayed nu-
merous common features may have facilitated their fusion.42

believe. Similarly, P. Mutafcief, "Proizod't na Asenevci," Makedonski Pregled 4 / 4 (1928),

1 - 4 2 (French summary 149-152). This nationalistic interpretation of the sources has
nowadays become less dominant even among Bulgarian scholars, see, for example,
V. Stojanov, Kumanen und Kumanologie. Über die kumanische ethnische Komponente auf
dem Balkan, in http://www.ihist.bas.bg/sekcii/CV/_private/lValery_Stojanow, passim.
37 P. Mutafcief, Istorija na b'lgarskija narod, 2 vols. Sofia 1943-1944, 33, attempted this,
wrongly creating closer ethnic ties between the Asenids and the Kievan Rus, in this re-
spect unconvincingly as Zlatarski.
38 This assumption, which can be regarded as generally accepted nowadays, has become
more and more influential since F. I. Uspenskij, Obrazovanie vtorogo bolgarskogo carstva.
Odessa 1879, 108; and C. Jirecek, Geschichte der Serben. 2 vols. Gotha 1911-1918,1: 269.
See, for more detail, Vásáry, Cumans and Tatars, 38ff.
39 See L. Rásonyi, "Les anthroponymes Comans de Hongrie," Acta orientalia Hungarica 20
(1967), 135-149; and L. Rásonyi, "Kuman özei adlari," Türk Kültürü Ara§tirmalari 3 - 6
(1966/69), 71-144.
40 Rásonyi, "Kuman özei adlari"; G. Clauson, An Etymological Dictionary of Pre-Thirteenth-
Century Turkish. Oxford 1972, 248.
41 The relevant sources cannot be dealt with here for reasons of space; due to the fact that
they are not supposed to be in the foreground, see L. Rásonyi-Nagy, "Valacho-Tur-
cica." In: Aus den Forschungsarbeiten der Mitglieder des Ungarischen Instituts und des Colle-
gium Hungaricum in Berlin, dem Andenken Robert Graggers gewidmet. Berlin and Leipzig
1927, 68-96; Vásáry, Cumans and Tatars, 40 with note 116; and Mutafíiev, "Proizod't na
Asenevci," 1-42.
42 "The prostagma of Emperor Andronikos Komnenos from 1184," in Actes de Lavra 341-
345, No. 66, ed. P. Lemerle, Paris 1970, in which mention is made of the Cumans, Vlakhs
and Bulgars in the province of Moglena in western Macedonia, makes this assumption
plausible.
 THOMAS BRÜGGEMANN

reason for the political and social instability of the Second Empire: many of its
"new" nobles, the Bojars, were of nomadic Cuman origin and hence subordinated
under a "central" ruler with reluctance. The Cumans, with their nomadic back-
ground, were obviously not willing or able to accommodate to the conventions of
feudal Bulgaria from one day to the next. Inevitably conflicts between sedentary
inhabitants and rural dwellers of the Second Bulgarian Empire and Cumans were
commonplace.43 After the assassination of Tsar Petar in Tarnovo in 1195, and of
his brother Asen in the following year, their younger brother Kaloyan succeeded
as tsar.44 His marriage to a Cuman noblewoman shows strikingly that the Bulgar-
ian house of tsars obviously needed a dynastic connection with the Cumans.
There are two further possibilities still under discussion besides the 'Cuman the-
ory' when looking for the ethnic origins of the ruling Asenid dynasty of the Sec-
ond Bulgarian Empire. These soon turn out to be at best hypothetical, however,
becausee trying to establish a Bulgarian or Vlakh descent of Asen, Petar, and Ka-
lojan does not succeed convincingly. That is because one usually cannot avoid
basing both assumptions mainly on the testimonies of Pope Innocent III, which
are conflicting in this respect. There are two groups of statements in the pope's
correspondence, each seemingly contradicting and excluding the other. One
group seems to support the Bulgarian descent of Asen's family, the other the
Vlakh descent.45 Innocent III wrote to the Hungarian King Imre, in 1204, saying
that "Peter and Joannica, who descended from the family of the former kings, be-
gan rather to regain than to occupy the land of their fathers".46 By contrast, the
pope wrote to Kajolan in 1199, saying that he had heard of Kalojan's Roman de-
scent. In his reply, Kajolan expressed his satisfaction that God "made us remem-
ber our blood and fatherland from which we descended".47 In another place, In-
nocent remarks on "the people of your land who assert that they descended from
Roman blood".48 As far as the first statement is concerned, it is a medieval con-
vention: the ruling house is always seen as the legitimate successor of the previ-
ous one. If the pope said that Asen, Petar and Kalojan were descendants of the
earlier Bulgarian kings, he simply wanted to express that they were the legitimate
rulers of Bulgaria. That is why they do not occupy the land, but reoccupy it as

43 In this respect, cf. Rogerius' description in the Carmen Miserabile, §§ 2-12; Szenpétery,
Scriptores, 2: 553-559, for Hungary in 1244: "But when the king of the Cumans, with his
nobles and commoners, began to roam about Hungary, since they had innumerable
herds of cattle, caused serious damage to the pastures, sown lands, gardens, orchards,
vineyards, and other property of the Hungarians".
44 He was also called loanitza and reigned from 1197 to 1207, see, for example, Dujcev,
"Bulgarien," 921.
45 See Uspenskij, Obrazovanie vtorogo bolgarskogo carstva, 153; Vâsâry, Cumans and Tatars,
34; and N. Bànescu, Un problème d'histoire medieval: Création et caractère du second empire
bulgare (1185). Bucharest 1943,13-21.
46 A. Theiner, éd., Vetera monumenta Slavorum meridionalium historiam illustrantia, 2 vols.
Rome and Zagreb 1863,1875 (henceforth Mon. Slav, merid.) 1: 36.
47 Mon. Slav, merid. 1:15.
48 Mon. Slav, merid. 1:16.

66
 CUMANS IN SOUTHERN DOBRUDJA

their heritage, which in the meantime had been usurped by the Byzantines. As
for the pope's second statement, according to which the Asenid dynasty was of
Roman descent,49 similar caution is necessary. The pope must have known about
the Latin origin of the Vlakh language, and consequently identified its speakers
as descendants of Rome. Although Kalojan and his Vlakh subjects must have
spoken a neo-Latin language, the precursor of modern Romanian, it can almost
be taken for granted that the Vlakhs of the Balkans had no historical awareness of
their Roman descent. Therefore the pope's assertion that the Asenids were of
Roman descent was a mere expression of the fact that the Asenids were Vlakhs,
and has nothing to do with the Vlakhs' alleged Roman consciousness. If they
were Romans, this was true only in the sense that they were subjects of Byzan-
tium and thus called "Romaioi", because Byzantium regarded itself as the true
heir of Rome.50

The archaeological perspective - Cumans in the Southern Dobrudja

Since at least some of the Cumans did not gave up their nomadic habits when
they settled in the Balkan Peninsula in winter, the material remains of their pres-
ence are normally found outside built up settlements and necropoleis. Although a
more intense approach to the habits of the sedentary population, at least on the
part of the Cuman aristocracy in the vicinity of the Bulgarian court at Tarnovo,
seems predictable, archaeological traces in this respect are as yet missing.51
The presence of the Cumans south of the Danube and their role within the
Second Bulgarian Empire has been designated mainly via literary references.
Though the Cumans settled sporadically on the right bank of the Danube before
the twelfth century, the fact remains that there are only a few archaeological
traces to show their presence in the southern Dobrudja which can be securely
dated later than the eleventh century, i.e., in the period of the Second Bulgarian
Empire. The fact that until recent times no archaeological traces of the Cumans in
the southern Dobrudja could be dated after the "foundation" of the Second Em-
pire neither denies their continuous presence in the region nor their connection to
the "princes" of the rising state. The absence of Cuman finds outside the two cen-
tres which could identified for the eleventh century moreover underlines that, af-
ter the end of Byzantine supremacy and its substitution by the Second Bulgarian
Empire, whenever they occurred here in larger formations, they continued cross-
ing the Danube to the south for seasonal purposes only. The Cumans went on
with their nomadic way of life even in the probably Cuman-led 'Bulgarian' Em-

49 In this respect, the term 'Roman descent' was not just used by the pope synonymously
with 'Vlakh descent', as Vasary, Cumans and Tatars, 36, points out convincingly.
50 For example, see Lilie, Byzanz, passim.
51 Therefore no remarks can be made about the number of Cuman courtiers, their possi-
ble residence in particular quarters and their relationship to the city's inhabitants or
even to Christendom; see RaSev, "Kumanite na jug," 22.

67
 THOMAS BRÜGGEMANN

pire.52 Accordingly, most probably just their elites seem to have been assimilated
and integrated in the court society of the Second Empire, which then became ab-
sorbed in the sedentary urban population and therefore left no specifically
Cuman archaeological trace, as for example in the capital Tarnovo.
The presence of such nomadic parts of the Cuman tribes, especially the
horsemen-warriors with their families and clans, whose services hence were
available for the princes of the Bulgarian Empire on a temporary basis only, can
be restricted to two zones in the southern Dobrudja. The archaeological material
at and close to the centre of the First Bulgarian Empire implies a Cuman inci-
dence even in the twelfth century, whereas the burials situated more to the south,
in the Thracian basin between the rivers Tundscha and Mariza,53 seem to prove a
Cuman presence as allies of Constantinople against the Pechenegs and similar
groups, which endangered its rule during the "Byzantine Intermediate" through-
out the eleventh and twelfth century.54 Therefore the region of Pliska has to be
dealt with first of all.55 In 1927, on the plain of Pliska56 in the vicinity of Zarev
Brod, two larger-than-life stone statues were found, put in the ground on top of a
Bronze Age tumulus.57 These sculptures are regarded as being an aristocratic
couple who were worshipped here as ancestors of an important Cuman noble
clan.58 Considering many analogies in the northern Black Sea area, i.e., Cumania,
one could postulate in this case a Cuman sanctuary, in which seasonal rituals of a
cult for the ancestors were performed.59 The existence of this material in the
Pliska basin allows us to assume the presence of a large and important Cuman
group. Moreover, the sanctuary may be evidence that the area was visited abun-

52 At best, the seasonal presence of larger Cuman groups in the southern Dobrudja
proves this assumption; see Raskovskij, "Rol' polovcev v vojnah," 203-211, and above
note 34.
53 In this respect, see, convincingly Rasev, "Kumanite na j u g " 24, nos. 6 and 7.<
54 For the Cumans as allies of the Byzantines, especially against the Pechenegs, see Lilie,
Byzanz, 328ff; and Schmitt, Die Petschenegen auf dem Balkan, 473-490.
55 For the geopolitical significance and for the development of Pliska from a nomadic
campus to the residence and capital of the First Bulgarian Empire, see Th. Brüggemann,
"Campus - Sedes Principalis - Civitas Regni. Pliska und das frühe Bulgarien im Spiegel li-
terarischer und archäologischer Zeugnisse (7.-9. Jahrhundert nach Christus)." in K.
Boshnakov, ed. Jubilaeus VI. Das antike Erbe des westlichen Pontosraumes, Sofia 2008, pas-
sim.
56 See Rasev, "Kumanite na jug," 22, nos. 1-3.
57 The site is located close to modern Sumen, see Raäev, "Kumanite na j u g " 22, no. 1 with
figure 1 no. 1.
58 Ra§ev, "Kumanite na jug," 23 with figure 2; and "Prab'lgarski," 17-20.
59 In kind and size these two statues were for a long time singular on the Balkan Peninsu-
la. However, recent excavations at the citadel of Pliska have brought to light a head of
50 cm, which undoubtedly must have belonged to a larger-than-life statue; besides
that, the new piece shows generally significant similarities to Cuman "round-plastic"
(Valeri Grigorov/Stanislav Stanilov). Despite the significance of this example for the
Cuman presence in the Pliska region, it has to be emphasized that its attribution cannot
be called certain pending detailed publication by the excavators.

68
 CUMANS IN SOUTHERN DOBRUDJA

dantly and that eventually even a permanent Cuman base may have existed here.
This then would have been situated between the Bulgarian court in Tarnovo and
the areas where the Cuman khans resided, north of the Danube. From these ar-
chaeological data, the plain of Pliska furthermore is of outstanding importance
concerning Cuman presence in the southern Dobrudja. In this respect a yet un-
published secondary burial from tumulus 25 in Pliska seems of significance also,
as it is situated approximately 250 m outside the western earthen wall of the
outer city.60 Because of the total lack of embellishments, a group of four nomad-
burials from Dewtaschlari near Pliska, which obviously belong together, remain
uncertain in this respect. But on the ground of bearing and funeral habits they are
readily interpreted as being Cuman also.61 The same can be said for the late no-
madic burials from the eleventh-twelfth century from Madara.62 If the Pliska ba-
sin was predestined to become the gathering place of the Cumans in the Second
Bulgarian Empire, both because of its proximity to the court in Tarnovo and also
its symbolic importance as the "cradle" of the first Bulgarians, the findings from
southern Bulgaria show undoubtedly that Cuman horsemen covered the whole
of the imperial territories at least during Byzantine rule.63
In addition to the tombs, many small and scattered finds from the southern
Dobrudja dating from the eleventh to the thirteenth century, which have been
summed up in older research as belonging to "late nomads", have more recently
been assigned to the Cumans, as for example stirrups, arrowheads and other

60 Although the author has to thank A. Evglevszkij (Donetzk) for the hint that this Cu-
man burial may be dated earlier than late twelfth century due to its embellishments,
and therefore does not belong to the Second Bulgarian Empire but to the period of By-
zantine supremacy, the date assigned by Rasev, "Kumanite na jug," 22 no. 1 with fig. 1
(no. 2) and fig. 5 (excavated 1998 by Rasev/Stanilov, but still unpublished) has to re-
main for the moment, because further archaeological data, which would allow more
precise statements in this respect are not yet accessible. This tomb, which is surely
Cuman, consists of a hollowed trunk as sarcophagus, which was covered with thin,
smooth boards. To the left of the coffin, the skull, flayed skin, bones of the back-legs,
and bridles of a horse were found in situ, which were put there correlating to the anat-
omy and posture of the deceased man. In the narrow interstice between the grave-
trough and sarcophagus to the right of the dead man, his sabre and lance, which are
clearly typologically of Cuman origin, were found.
RaSev, "Kumanite na jug," 23 no. 5 with fig. 1 (also no. 2) and no. 6.
62 RaSev, "Kumanite na jug," 23 no. 2 (with fig. 1 no. 3; 3) and no. 4 with (fig. 1 nos. 3; 4).
The burials were discovered in 1934 in the Bronze Age tumuli 1 and 3 from Madara.
Although no embellishments were found here either, Cuman identification at least of
tumulus 1 seems to be correct, because of a horse's upper jaw-bone, which was found 6
m west of the tomb. It apparently belongs to this funeral since the custom of burying
horses next to men is known in that region and period only from the Cumans.
63 Ibid. 24 no. 6 (with fig. 1 no. 4). The secondary burial in a Roman-era tumulus from
Mednikarovo with its Cuman sabre and stirrup gives satisfactory information about
the deceased being a horse-soldier; next to the handle of the sabre a horse-skull was
found as well; 24 no. 7 (with fig. 1 no. 5). The secondary burial from Kovazevo offered
an iron sabre, which probably has to be defined as being Cuman also.

69
 THOMAS BRÜGGEMANN

military equipment, which are unfortunately mostly without context.64 The alto-
gether promising archaeological data concerning the Cumans in the southern
Danube region reveal a much more regular and more dense presence of this con-
federation of nomadic horsemen-soldiers within the Second Bulgarian Empire
than was hitherto assumed. That Cuman warriors were recruited under Kalojan
and Boril during the struggles against the Latin Empire of Constantinople also
seems probable, but this remains to be confirmed by archaeological data.65

Closing Remarks
Taking into consideration everything that has been said so far, the most plausible
supposition seems to be that Asen and his dynasty were of Cuman origin. They
stood at the head of the ultimate uprising in 1185, which caused the foundation of
the Second Bulgarian Empire, and their main support derives from their Vlakh
environment.66 They must have spoken the language of their "fellow-insurgents",
but preserved the knowledge of their Cuman predecessors' nomadic skills.
Moreover, they must been in close contact with their near "relatives" in Cumania.
That is why they turned to their kinsfolk to help them in their fight against the
Byzantine Empire.67 During the first years of the Second Bulgarian Empire, an

64 See in general Pâlôczi-Horvâth, Pechenegs, Cumans, Iasians, passim.

65 Especially in the battles at Adrianople in 1205 and Thessaloniki in 1207; see Dujcev,
"Bulgarien," 921; I. Dujiev, "Tzar Kalojan, bitkata pri Odrin pzez april 1205 g. i nejnite
setnini," Voenno-istoritzeski sbronik 4 8 / 4 (1979), 107-123; and R. L. Wolff, "The Second
Bulgarian Empire: Origin and History to 1204," Speculum 24 (1949), 167-206.
66 See above note 23. Nicetae Choniatae Historia, 1: 369, 60-64 suggests that the motives
which led the brothers Asen and Petâr to initiate the Bulgaro-Vlakh uprising against
the Byzantines were in their individual nature comparable to those of the participating
population. This again prohibits us to make to the incident a "liberation movement" as
does Vâsâry, Cumans and Tatars, passim: "Once, they [Asen and Petâr] went to the em-
peror [Isaak II Angelos], who had set up his tent in Kypsella. They asked him to regis-
ter them in the Roman army and to that effect to grant them, by his imperial edict,
a small piece of land in the Haimos which would yield them a modest income. But their
request was rejected." (transi. Vâsâry, Cumans and Tatars, 16). Only as a result of the re-
fusal by the Byzantine Emperor did Asen and Petâr begin to agitate the Bulgaro-Vlakh
population: this was vengeance, nothing more. To justify calling this a "liberation
movement", supplementary political thoughts and a concept would have to be reason-
able for the incident: these are missing here.
67 Nicetae Choniatae Historia, 1: 371, 15-372, 49, emphasizes that especially the allegedly
unwilling Vlakhs had to be tricked into participating in the rebellion and to subordi-
nate themselves under its both leaders: "The brothers had a church built in honour of
St Demetrios, and a crowd of men and women 'possessed by the devil' were assembled
inside. They were instructed to speak in a state of ecstasy about God's intention to re-
move the yoke from the shoulders of the Bulgarians and the Vlakhs [...] Convinced by
the impressive scene, the whole people took up arms against Byzantium, and one of
the brothers, Petâr, was crowned with a golden crown. The rebellion began [...]",
Vâsâry, Cumans and Tatars, 16. In the same way as the fact that the rising is wrongly
called a movement of (self-chosen) national liberation, this "Demetrios-legend" has of-

70
 CUMANS IN SOUTHERN DOBRUDJA

ethnogenesis obviously similar to that seen already in the First Empire can be ob-
served, when the Turks after their immigration merged with the Slavic majority
during the seventh to ninth centuries.68 In the southern Dobrudja, the Cumans
remained even in the Second Bulgarian Empire in large part clearly nomads, who
crossed the Danube to the south only for seasonal pastures. Unlike their compan-
ions in Hungary, whose economic and social habits adjusted to those of their sed-
entary environment, for the southern Dobrudja archaeological traces neither for
permanent Cuman settlements nor for a sedentarization of lager tribes have yet
been found. That is why the Cumans, with their nomadic way of life, were not
able to accommodate to the conventions of feudal Bulgaria from one day to the
next: inevitably, conflicts between sedentary rural dwellers of the Second Bulgar-
ian Empire and Cumans were probably common. Thus the Second Bulgarian
Empire politically and socially remained unbalanced, the more so as many affili-
ates of the new Bulgarian nobles, the boyars, were of Cuman descent and there-
fore habitually were not willing to be subordinate to a centralized rule.
At the end of 1217, the son of Asen I, Ivan Asen II,69 who had fled after his uncle
Kalojan's death to southern Russia, returned to Bulgaria and became tsar in 1218.
He made Bulgaria the most important political and military power in southeast
Europe: it stretched south to the Aegean and southwest to the Adriatic Sea. This
status was not enduring, however. The death of Ivan Asen II in 1241 and the loss
of the Bulgarian territory in Thrace and Macedonia caused the decline of Bulgaria
as the dominant power in southeast Europe - the princedom was lacking ethnic
and social coherence. Finally, in the middle of the thirteenth century, the Mongols
overwhelmed Cumania between the Volga and the Carpathians and became not
only the immediate northern neighbours of Bulgaria, but also its hegemóniái
power.70

ten been over-interpreted in its importance for the achievement of the insurgence. It
should be seen more as a piece of temporary Christian propaganda then as sufficient
for the mobilization of lager parts of the Bulgarian or Vlakh population; the erection of
the church for St Demetrios may be historical reality, its influence on the beginning of
the insurrection is not.
68 See Briiggemann, Die Staatswerdung Bulgariens, 462ff.

6 9 He reigned from 1218 to 1241, see G. Prinzing, "Ivan II. Asen" in Lexikon des Mittelalters

5 (1991), 833; and J. S. Langdon, "The forgotten Byzantine-Bulgarian assault and siege
of Constantinople 1235-36," in Sp. Vryonis, ed., Byzantine Studies in Honor of Milton V.
Anastos. Byzantina kai Metabyzantina 4. Malibu 1985,16-135.
7 0 See, in general, Dujcev, "Bulgarien," 922.

71