Chronica_2023-21

CHRONICA
Nomads and their Neighbours in the Middle Ages
Editors: Konstantin Golev – Delyan Rusev

Contents
KONSTANTIN GOLEV AND DELYAN RUSEV
Preface: Medieval Nomads (MeN) at the dawn of a Modern Global Pandemic ........ 5
Articles
RICHÁRD SZÁNTÓ
The Carpathian Basin in the Cosmography of the Unknown Ravennese .................... 9
ULI SCHAMILOGLU
Medieval Nomads and the First Pandemic: Theoretical Considerations ................... 21
KÜRŞAT YILDIRIM
Some Notes on the Missions of Pei Ju ................................................................................. 35
AH RIM PARK
Art in Mongolia Between the 6th and 8th Centuries: Cultural Diversity in the Arts of
the Turks .................................................................................................................................... 43
EDINA DALLOS
The Tree Cult and the “Piece of Wood the Size of a Phallus” ....................................... 55
IRINA KONOVALOVA
Pechenegs in the Imperial Space of Byzantium in the Middle of the 10th Century .. 67
TSVETELIN STEPANOV & IVELIN IVANOV
Late Medieval Nomads in the Bulgarian Historical Apocalyptic Literature: Images
and Realities .............................................................................................................................. 77
FRANCESCO DALL’AGLIO
Shifting Foundations: The Cumans of Paristrion and Their Influence on the Second
Bulgarian State .......................................................................................................................... 89
IVELIN IVANOV
Nomads against Knights: Cumans and Latins in the Military Campaigns in the
Balkans (1205–1225) ............................................................................................................... 105
KONSTANTIN GOLEV
Ecology of Warfare: The Seasonal Campaigns of the Cuman-Qïpchaqs in Eastern
Europe and Central Asia ...................................................................................................... 117
SIMON BERGER
Nomadic State in the Sedentary World: The Military Mode of Government of
Conquered Lands in the Mongol Empire......................................................................... 183
DELYAN RUSEV
Sarı Saltuk and the Nomads in 13th-Century Dobruja: Pre-Ottoman Sources and
Realities .................................................................................................................................... 201
LORENZO PUBBLICI
Experiencing Alterity: Italian Merchants and the Local Population in the 14th
Century Venetian Azov Sea. Changes and Continuity ................................................. 247
ISHAYAHU LANDA
Some Preliminary Notes on the Aftermath of the Eurasian Steppe’s “Mongol
Moment” .................................................................................................................................. 269
MICHAL HOLEŠČÁK
The Oriental Coin Hoard from Ružomberok, Slovakia ................................................ 291
HATIM MAHAMID & CHAIM NISSIM
Rakb al-Ḥajj al-Shāmī: Relations between Nomads and Pilgrims during the Mamluk
and Ottoman Periods ............................................................................................................ 301
HARUN YENI
Vlachs and Yörüks as Military Groups under Ottoman Rule and the Evolution of
Their Roles ............................................................................................................................... 317
TATIANA A. ANIKEEVA
The Battle of the Epic Hero with Death: The Problem of the Relationship between
Medieval Turkic and Byzantine Folklore ......................................................................... 329
BRUNO GENITO
Early Mediaeval Nomads between Iran, Central Asia and Europe: The Issues of the
Archaeological Evidence ...................................................................................................... 337
Preface
Medieval Nomads (MeN) at the dawn of a Modern Global Pandemic:
Eighth International Conference on Medieval History of the Eurasian Steppe,
“Nomads and their Neighbours in the Middle Ages”,
held in Sofia, Bulgaria, on November 20–3, 2019
In 1997, 2000, and 2002, the Department of Medieval History at the University of
Szeged organized several conferences on the history of medieval nomads of the
Eurasian steppe, the proceedings of which were subsequently published in Hun-
garian. In 2004, the Department of Medieval History and the Department of Ar-
chaeology at the same university together with the Research Group on Hungarian
Prehistory of the Regional Committee of the Hungarian Academy of Sciences in
Szeged decided to convene an International Conference on Medieval History of the
Eurasian Steppe. The first conference of this kind was held in Szeged in 2004, fol-
lowed by events in Jászberény (2007), Miskolc (2009), Cairo (2011), Moscow (2013),
Szeged (2016), and Shanghai (2018).
As a continuation of this series, Sofia University “St. Kliment Ohridski” and the
Institute for Historical Studies of the Bulgarian Academy of Sciences organized the
Eighth International Conference on Medieval History of the Eurasian Steppe,
“Nomads and their Neighbours in the Middle Ages”, held in Sofia, Bulgaria, on
November 20–3, 2019. It was only to be expected that the regular scholarly migra-
tions of the MeN series would follow in the footsteps of their medieval nomadic
forerunners and will appear south of the Lower Danube to gather on a venue in the
foothills of the Balkan Mountains. Thus, Bulgaria, which shares with Hungary a
steppe origin of its medieval statehood, also shared with the modern Magyars the
historiographical interest in the nomads of the past as well as the honour to host a
scientific forum of internationally recognized researchers in this field.
Given the fact that the medieval Eurasian nomads affected in one way or anoth-
er the history of most of the Old World, it comes as no wonder that the participants
in the 8th MeN were also globally represented. As could be observed on the website
of the event,1 more than forty scholars from Hungary, China, Slovakia, the United
Kingdom, Italy, Canada, France, Russia, Turkey, South Korea, Poland, Israel, Fin-
land, Germany, Kazakhstan, and, of course, Bulgaria, presented their research.
Two distinct key notes were offered to this diverse audience—by Prof. Lorenzo
1 www.medievalnomads2019.weebly.com
6 KONSTANTIN GOLEV – DELYAN RUSEV
Pubblici on the Italian commercial presence in the Azov Sea, and by Prof. István
Zimonyi on the history of Volga Bulgaria.
Yet, modern scholars, not unlike medieval nomads and indeed humans in gen-
eral, are not immune to the vagaries of nature and politics. Although this was nei-
ther the first time when an event with a nomadic connection foreran a global pan-
demic, nor the first MeN conference that found itself in proximity to historical
events,2 in late November 2019 nobody could have expected that very soon the
Horsemen of the Apocalypse will gallop through Eurasia—Pestilence starting from
the east, to be followed by War in the west. One could only speculate that when
their hooves finally come down, our world will never be the same—what is certain
is that it is now a bit more similar to the world of our medieval forefathers. Thus,
as mortal human beings we cede our humble offerings to the everlasting Clio—
written about those who went before us for those who come after.
Acknowledgments
This volume appears with the help of Dobromir Dobrev whose role in providing
financial and administrative support was instrumental. The editors are also espe-
cially indebted to the language editor Nikolay Dyulgerov for his devoted efforts in
the editing process.
This publication is supported by the National Research Program Cultural Her-
itage, National Memory and Society Development (https://kinnpor.uni-sofia.bg),
funded by the Bulgarian Ministry of Education and Science. The same program
financed the 8th International Conference on Medieval History of the Eurasian
Steppe, “Nomads and their Neighbours in the Middle Ages”, Sofia, Bulgaria (No-
vember 20–3, 2019).
KONSTANTIN GOLEV – DELYAN RUSEV
2ZIMONYI, I. Medieval Nomadic conference in the shadow of the Egyptian revolution.

Chronica 11 (2011), 3–5.
PREFACE 7
This volume is dedicated to the seventieth anniversary of Prof. Valery Stoyanov (b.
24.02.1951) – a historian of the medieval Eurasian nomads and their historians.
Prof. Valery Stoyanov

The Carpathian Basin
in the Cosmography of the
Unknown Ravennese
RICHÁRD SZÁNTÓ
Abstract
Anonymous of Ravenna, an unknown cleric, described the whole world in his
Cosmography. The exact time of the writing of the work is disputable. Joseph
Schnetz assumed that Anonymous wrote his work in the early 8 th century. In the
opinion of Franz Staab, the Cosmography was compiled in the late 8th or early 9th
century. The Cosmography consists of five books, of which the fourth book contains
a detailed description of Europe. It presents a description of Central Europe and
the Balkan Peninsula, including Saxony, Dacia, Pannonia, Illyria, Dalmatia, and
Moesia. Anonymous of Ravenna listed the settlements and rivers in Dacia and
Pannonia and mentioned the peoples living in these territories. The author also
uses the name Patria Albis Ungani, which does not appear in other sources. This
study presents the Carpathian Basin based on the description of Pannonia and
Dacia and determines the geographical position of this Patria Albis Ungani. Finally,
a hypothetical interpretation explains its name and origin.
Keywords: Anonymous of Ravenna, Cosmography, Pannonia, Patria Albis Ungani
Unknown Ravennese (Anonymous of Ravenna) wrote his cosmography at the
beginning of the 8th century. The anonymous author was an Italian cleric living in
Ravenna at the end of the 7 th and the beginning of the 8th century. He collected the
geographical and cosmographical knowledge of his age. The autograph manu-
script of the Cosmography did not survive, but three copies have survived.
The first critical edition of the Cosmography was issued by Moritz Pinder and
Gustav Parthey in Berlin in 1860. The second critical edition of the Cosmography
was published by Joseph Schnetz in 1940, but his work did not include an index of
Institute of History, Faculty of Humanities and Social Sciences, University of Szeged, Hun-
gary. The article is a revised version of a paper delivered at the 8 th International Conference
on Medieval History of the Eurasian Steppe Nomads and their Neighbors in the Middle Ages,
held in Sofia, November 20–3, 2019.
10 RICHÁRD SZÁNTÓ
the toponyms.1 Marianne Zumschlinge has prepared the index of toponyms for
Schnetz’s publication and issued the whole work in 1990.2 Additionally, a German
translation of the Cosmography was published by Joseph Schnetz in 1951.3
The medieval manuscripts of the Cosmography do not include any maps, but
two maps were created on the basis of its text in the 19 th century. Heinrich Kiepert
drafted the first one in 1859,4 and Konrad Miller created the second one in 1898.5
For the composition of the Cosmography, Anonymous of Ravenna used the
works of ancient Roman and Greek authors. 6 Furthermore, he could have obtained
additional information from merchants, envoys, and sailors. Thus, in the Cosmogra-
phy, the data from preceding historical periods is mixed with the information from
Anonymous’s own age.
Various researchers have examined the book. Joseph Schnetz published several
studies on the Cosmography,7 and Franz Staab wrote a study on the sources of the
European part of the text.8 Tuomo Pekkanen researched the names of the cities on
the Black Sea coast and their variations in the various chapters of the Cosmography.9
Alexander Podossinov translated most of the Cosmography into Russian and pub-
lished it with commentary and studies.10 Kurt Guckelsberger and Florian Mit-
tenhuber summarized the geographical information of the Cosmography and de-
scribed some elements of the geographical concepts that emerged in it. 11 The chap-
ter on Britannia in the Ravenna Cosmography was studied by Keith J. Fitzpatrick-
Matthews.12 Daniel T. Potts published studies on the Armenian toponyms and the
placenames of India and the Persian Gulf on the basis of the Cosmography.13
In general, the Cosmography contains the description of the countries and peo-
ples on the Earth; it consists of five books, and the description of Europe is in the
fourth. Anonymous of Ravenna listed the Roman provinces, the countries, peoples,
rivers, and cities of Europe. This study presents the Carpathian Basin on the basis
of the Cosmography. Anonymous of Ravenna depicted the rivers and cities of Dacia
and Pannonia, which represented the Roman provinces in the Carpathian Basin.
1 SCHNETZ 1940.
2 ZUMSCHLINGE 1990.
3 SCHNETZ 1951.
4 KIEPERT 1859.
5 MILLER 1898.
6 SCHNETZ 1942.
7 SCHNETZ 1925; SCHNETZ 1932; SCHNETZ 1934.
8 STAAB 1976.
9 PEKKANEN 1979.
10 PODOSSINOV 2002.
11 GUCKELSBERGER – MITTENHUBER 2011.
12 FITZPATRICK-MATTHEWS 2013.
13 POTTS 2019; POTTS 2017; POTTS 2018.
THE CARPATHIAN BASIN IN THE COSMOGRAPHY… 11
The Cosmography also included an unknown country under the name of Patria Albis
Ungani that can be localized between the Elbe and Pannonia.
I. Dacia
Anonymous of Ravenna wrote down the name as Datia (!), and he mentioned the
names Datia prima et Secunda. According to Anonymous, Dacia had a large territory
and he also noted that this province was called Gepidia in his age. The Huns and
Avars (unorum gens, avari) were the inhabitants of Gepidia. The Cosmography men-
tions that Illiria was between Gepidia and Dalmatia (provincia). The author listed
ancient cities in Dacia: Drubetis, Medilas, Pretorich, Panonin, Gazanam, Masclunis,
Tibis, Tema, Tiviscum, Gubalis, Zizis, Bersovia, Arcibada, Canonia, Potula, Bacau-
cis.14 The names of these cities are identifiable within Dacia’s toponyms depicted
on the Tabula Peutingeriana. The origin of the Tabula Peutingeriana goes back to
an old Roman map, which was copied several times in the late Antiquity and the
Middle Ages.15 Anonymous of Ravenna could have used an early copy of this Ro-
man map.16 A great part of the toponyms in the Cosmography come from the geo-
graphical names in this map, of which the Tabula Peutingeriana is the last medie-
val copy.17
Table 1. Settlements in Dacia
An.
Drubetis Medilas Pretorich Panonin Gazanam Masclunis Tibis
Rav.
Ad Me- Ad Pan- Tivisco
Drubetis Pretorio Gaganis Masclianis
TP diam nonios VII. 4.
VII. 4. o VII. 4. o VII. 4. o VII. 4. o
VII. 4. o VII 4 o o
An.
Tema Tiviscum Gubalis Zizis Bersovia Arcibada Canonia
Rav.
Caput
Tivisco Azizis Bersovia Arcidava
TP ? bubali ?
VII. 4. o VII. 3. o VII. 3. o VII. 3. o
VII. 3/4.
An.
Potula Bacaucis
Rav.
Centum
TP Putea ?
VII. 3. o
The rivers of Dacia were also listed in the Cosmography: Tisia, Tibisia, Drica, Ma-
risia, Arine, Gilpit, and Gresia. Anonymous of Ravenna also noted that the rivers
flowed into the Danube. He noted that the Flautasis River ended in Dacia. This
14 ZUMSCHLINGE 1990, 53–4.

15 RATHMAN 2018, 8–26.
16 WEBER 1976, 20–3.
17 PEKKANEN 1979, 111–112.
river may be identical to the Aluta, or Olt in modern Romania. Anonymous of Ra-
venna mentioned that he adopted information referring to Dacia from Iordanes’s
work.18 Nowadays, the rivers of Dacia can be identified as the following rivers.
Table 2. Rivers of Dacia
An. Rav. Tisia Tibisia Drica Marisia Arine Gilpit Gresia Flautasis
Flutaus/
Iordanes19 Tisia Tibisia Dricca Marisia Miliare? Gilpil Grisia
Aluta?
Identifi- Crişul Crişul Crişul
cation of Tisza/ Tamiš/ Maroş/ Alb/ Negru/ Repede/
? Olt
the rivers Tisa Timiş Maros Fehér- Fekete- Sebes-
names20 Körös Körös Körös
Romania,
Serbia, Romania, Romania, Romania, Romania,
Country Hungary, Romania
Romania Hungary Hungary Hungary Hungary
Serbia
II. Pannonia and Valeria

Anonymous of Ravenna mentioned Pannonia superior, Pannonia inferior and Valeria.
He named the two Pannonias (Pannoniae) as countries (patriae), but they were orig-
inally Roman provinces and were not independent political formations. In the
Cosmography, the geographical names related to the two Pannonias are divided into
three groups. The first list contains 31 toponyms. 21 As already mentioned, Anony-
mous of Ravenna adopted these geographical names from an earlier copy of the
Tabula Peutingeriana (TP).22 The TP was a copy of an old Roman imperial road
map that also included the stations on the roads. Many stations were settlements,
but some were named after a geographical point, for example the Drina River
(Drinum flumen). Anonymous of Ravenna copied these station names of the Roman
roads into the Cosmography.
18 ZUMSCHLINGE 1990, 54.

19 Tisia: MOMMSEN 1882, 62, 104, Flutaus: MOMMSEN 1882, 62; Aluta: MOMMSEN, 1882, 75. Tibi-
sia: MOMMSEN 1882, 104; Dricca: MOMMSEN 1882, 104; Marisia, Gilpil, Grisia, Marisia, Miliare:
MOMMSEN 1882, 87.
20 BÓNA – CSEH – NAGY – TOMKA – TÓTH 1993, 172–3.
22 TABULA PEUTINGERIANA (TP).
Table 3. The first list of the geographical names in Pannonia

An.
Confluentes Taurinum Idominio Bassianis Fossis Sirmium
Rav.
Confluentibus Tauruno Idominio Bassianis Sirmium
TP –
VI. 5. o VI. 5. o VI. 5. o VI. 4. o VI. 4. o
An. Mursa Mursa
Drinum Saldis Bassantis Marsonia
Rav. maior minor
Drinum Mursa Mursa
Saldis ad Basante Marsonie
TP fl[umen] maior minor
VI. 3. o VI. 3. o VI. 2. o
VI. 3/4. o VI. 2. o VI. 1. o
An.
Ioballios Berevis Sorenis Marimanus Balenilo Sirote
Rav.
Berebis
Seronis Marinianis
Iovallio (Berbis/ Bolentio Sirotis
TP (Serenae?) (Mariniana)
VI. 1. o Vereis?) V. 5. o V. 4/5. o
V. 5. o V. 5. o
VI. 1. o
An.
Cucconis Lentulis Sonista Botivo Populos Aquaviva
Rav.
Aqua
Cuccio Luntulis Sonista Botiuo Populos
TP Viva
VI. 3. o V. 4. o V. 4. o V. 4. o V. 3. o
V. 3. o
An.
Remista Petaviona Vincensimo Ligano Salla Aravona
Rav.
Remista Petavione Advicesimum Arrabone
TP – –
V. 3. o V. 2/3. o V. 3. o V. 2/3. o
An.
Savaria
Rav.
Sabarie
TP
V. 2. o
Table 4. The second group of the geographical names in Pannonia 23

An.
Burgenis Spaneatis Ansilena Cibalis Certisiam Lavares Cuminion
Rav.
ad La-
Burgenis Ulmospaneta Cansilena [Cibalae]? Certis
TP bores -
VI. 4. o V. 3. o VI. 3. o VI. 3. o VI. 2. o
VI. 2. o
In the Pannonian chapter of the Cosmography, the third list of toponyms con-
tains 33 place names, five of them are identifiable on the basis of the Tabula Peu-
tingeriana.24

24 ZUMSCHLINGE 1990, 57; TABULA PEUTINGERIANA (TP).
Table 5. The third list of the toponyms in Pannonia

An. Rav. Serbitium Fines Lamatis Lausaba Baloia
Seruttio ad Fines Lamatis Leusaba Baloie
TP
VI. 1. o VI. 1. o VI. 1/2. o VI. 2. o, VI. 2. o
Twenty-eight geographical names are not identifiable: Siclis, Ecclavia, Sanigion,

Persetis, Netabio, Speridium, Bedini, Necal, Brindia, Clande, Assino, Bercio, Apeva,
Sapua, Bersellum, Ibisua, Derva, Citua, Anderba, Sarminium, Charmenis, Scaladis, Sapua,
Aleba, Suberadona, Asione, Clande, Berginio. 25
Anonymous of Ravenna mentions that several rivers flow through the two
Pannonias (Pannoniae): Parsium, Ira, Bustricius, and Dravis. Schnetz identified Ira
with the Mura/Mur River between Austria and Hungary nowadays. Dravis can be
identified as the Drava River between Hungary and Croatia. 26 Parsium is an uni-
dentified river in Pannonia. In the territory of Croatia, Dillmann and Miller identi-
fied Bustricius as the Bistritza, a tributary of the Drava.27 According to the linguists,
the root of Bistritza originates from the Old Slavonic *bystr meaning fast-running
clear water/stream. There were many Bystric, Bistrica, Beszterce, Besztercény topo-
nyms in the medieval Carpathian Basin. 28
According to Anonymous of Ravenna, there was a large lake in Pannonia bear-
ing the name of Pelsois. This lake can be identified as today’s Balaton in Hungary.
Anonymous of Ravenna also depicts the province of Valeria as lying between
the two Pannonian provinces and he mentioned that Valeria was called Media Pro-
vincia.29
The cities (civitates) of Valeria are divided into two groups in the Cosmography.30
Table 6. The first group of the toponyms in the province of Valeria
An. Rav. Acunum Usum Malatis Catio Cornacum
Acunum Cusum Milatis Cuccio Cornaco
TP
VI. 3/4. o VI. 3. o VI. 3. o VI. 3. o VI. 2. o
An. Rav. Alusione Annama Clantiburgum Livori Donatianis
Lusiene Annamatia Tittoburgo (?) ad Labores Donatianis
TP
V. 5. o V. 5. o VI. 2. o VI. 2. o VI. 1. o
An. Rav. Antiana Lugione Belsalino Lumano Cardelaca
Antiana Lugione Vetusallo Lusomana Gardellaca
TP
VI. 1. o VI. 1. o V. 4/5. o V. 4. o V. 4. o

26 SCHNETZ 1951, 62.
27 DILLEMANN 1997, 99; MILLER 1898, 18.
28 KISS 1988, vol. 1, 208.
The second group of settlements in the province of Valeria consists of seven

toponyms.31 These geographical names are identifiable with the stations of the
Roman imperial road between Siscia and Emona.
Table 7. The second group of settlements in Valeria
An. Rav. Sicce Fines Romula Nomiudum Cruppi
Siscia ad Fines Romula Novioduni Crucio
TP
V. 5. o V. 4. o V. 3. o V. 2/3. o V. 2. o
An. Rav. Acerbo Atamine
ad Emona > Ata-
Aceruone
TP32 mine (?)
V. 1. o
V. 1. o
Anonymous of Ravenna additionally mentions that the border of Valeria was

the Saus River (Sava), but this is a mistake as the southern boundary of the prov-
ince of Valeria was actually the Drava River.
III. Patria Albis Ungani
In addition to Dacia and the Pannonian provinces, Anonymous also places another
country (patria) in the Carpathian Basin and west of it. According to him, there was
a country (patria) called Albis Ungani south of Saxony. The word Albis reflects the
name of the Elbe River, so this country may have been near the Elbe. The Cosmog-
raphy describes that the territory of Albis Ungani was mountainous to the east and
the extent of its territory was large in that direction. Anonymous Ravennese men-
tions that a part of this country is called Baias. According to the Cosmography, the
Pannonian provinces (Pannonia Superior and Inferior) had a long and extensive fron-
tier with Albis Ungani.33 It is also important to note that the word patria has two
meanings in the text of the Cosmography. The anonymous author uses it to describe
both the Roman provinces and the political formations of the barbarian tribes and
peoples.
The name Baias originated from the antique Boius, which was the name of a
Celtic tribe. The Pannonian Boius tribe settled in the territory of ancient Pannonia,
modern Slovakia and Bohemia.34 Therefore, based on the text of the Cosmography, it
seems that the area of Albis Ungani was located south of Saxony, east of the Elbe,
and north of the Pannonian provinces. As already mentioned, the territory of this
country was large and mountainous to the east.35 The mountain range of the North
Carpathians is identifiable with the mountainous territory of Albis Ungani. Only
Anonymous of Ravenna mentions the name Albis Ungani, it is not found in any

32 TABULA PEUTINGERIANA (TP).
34 SZABÓ 2005, 51–3, 63–8.
other medieval text, and the description of Albis Ungani represents only a short
passage in the Cosmogrpahy. The author does not include the names of any settle-
ments in Albis Ungani and he names only the Elbe River as situated near its territory.
Thus, the question arises: where does the name Ungani come from? What kind
of people could be called Ungani? In my hypothetical opinion, the ethnonym Un-
gani may have developed from Onogun(dur).
Anonymous of Ravenna had information on the migration and territories of the
different Bulgar tribes (Bulgars, Onogurs, Onogundurs). He claimed that the Bul-
gars came from Scythia and settled in Moesia, Macedonia, and Thracia. He also
lived after the age of migration of the Bulgars and knew the Bulgar people well. He
mentions that Onogoria was next to the Black Sea and the Sea of Azov (Maeotis).36
The author could get information about the people and the territory of the On-
ogurs/Onogundurs along the boundary of Pannonia, but the name of the Onogun-
durs apparently underwent a change in the time of Anonymous. Thus, he could
have recorded the name Onogundurs in the form Ungani.
The Byzantine authors Patriarch Nikephoros (ca. 758–828) and Theophanes the
Confessor (ca. 758/60–818), describe the history of the Onogundur-Bulgars. Ac-
cording to Theophanes, Kubrat’s fourth and fifth sons crossed the Danube after the
death of their father and went to the land of the Avars in Pannonia. One of them
was subjugated by the Avar Khagan. Kubrat’s fifth son did not stay in Pannonia,
but he and his people wandered to Italy and settled near Ravenna. 37
After the collapse of the Avar Empire, the Bavarian Geographer described the
regions east of the Kingdom of the Franks. In this source, Betheimare (Czechs), Mar-
harii (Moravians), Vulgarii (Bulgars) and Merehani (South Moravians) are located
north of the Danube in the territory of the Carpathian Basin in the middle of the 9 th
century.38 The Marharii (Moravians), living in the Morava valley, were the eastern
neighbours of the Czechs. The Vulgarii (Bulgars), a component of the population of
the Avar Empire, survived the collapse of the Avar Khaganate and lived in the
Great Plain in the Carpathian Basin. The region of the Bulgars was east of Panno-
nia and the Danube, and it came under the supremacy of the Danubian Bulgarian
Empire in the 9th century. The territory of Merehani (South Moravians) covered the
southern part of the territory between the Danube and the Tisza, and the Morava
valley south of Belgrade. 39
36 ZUMSCHLINGE 1990, 45, 48–9.

37 MANGO 1990, 86–9; NIEBUHR 1839, 544–7; TURTLEDOVE 1982, 55–6.
38 BAVARIAN GEOGRAPHER 1618, 321; POLGÁR 2003, 243–5.
39 SZÁNTÓ 2017, 152.
At the end of the 9th century, the Hungarians conquered the Carpathian Basin and
established their kingdom, Hungary. The name of the Onogundurs survived in the
form of Nándor [(O)no(gu)ndur] and became a toponym in medieval Hungary.40
Conclusion
Anonymous of Ravenna describes the Earth based on books by ancient authors
while also using information from his own age. The Cosmography includes a de-
scription of the provinces of Pannonia and Dacia. The author notes that Dacia was
also called Gepidia in his time. In the Roman age, these provinces were located in
the Carpathian Basin. The cities and rivers of Pannonia and Dacia were listed in the
Cosmography, and Anonymous of Ravenna also mentions the Avars (Huns) who
lived in Dacia. In addition to Pannonia and Dacia, the author mentions another
country by the name of Albis Ungani. This country was south of Saxony, its territo-
ry was large and mountainous to the east, and it had a boundary with Pannonia. A
part of the territory of Albis Ungani was called Baias. The area of Albis Ungani in-
corporated certain parts of the territory of the modern Czechia and Slovakia and
the northern part of the Carpathian Basin. Albis Ungani appears only in the Cos-
mography; it was a short-lived political formation (patria) – and not a Roman prov-
ince – that existed at the end of the 7th and the beginning of the 8th century. Around
670/80 AD, Kubrat’s fourth and fifth sons and their Onogur/Onogundur people
migrated into the Avar Empire and some of them settled in the northwestern re-
gion of the Charpathian Basin. The name of the Onogundurs changed to Ungani and
this ethnonym reached Ravenna where it was recorded by the writer of the Cos-
mography. Albis Ungani was ruled by the Avar Khaganate but Anonymus of Ra-
venna describes its geographical position and preserves the name of the Onogun-
durs in the form of Ungani.
Bibliography
BAVARIAN GEOGRAPHER (1618): Descriptio civitatum et regionum ad septentrionalem
plagam Danubii. In: Ex Bibliotheca Serenissima Utriusque Bavariae Ducum. As-
tronomische und mathematische Sammelhandschrift Geographus Bavarus.
Bayerische Staatsbiliothek BSB CLM 560 Mikrofilm, 321-322.
https://daten.digitale-sammlungen.de/~db/0001/bsb00018763/images/
BÓNA, I. – CSEH, J. – NAGY, M. – TOMKA, P. – TÓTH, Á. (1993): Hunok – Gepidák –
Langobardok, Szeged.
Dillemann, L. (1997): La Cosmographie du Ravennate, Brussels.
FITZPATRICK-MATTHEWS, K. J. (2013): Britannia in the Ravenna Cosmography: A reas-
sessment.
40GYÖRFFY 1977, 142; GYÖRFFY 1959, 64; KISS 1988, I, 190 (Belgrád, Nándorfehérvár), II, 221
(Nándor).
https://www.academia.edu/4175080/BRITANNIA_IN_THE_RAVENNA_COSM
OGRAPHY_A_REASSESSMENT
GUCKELSBERGER, K. – MITTENHUBER, F. (2011): Überlegungen zur Kosmographie
des anonymen Geographen von Ravenna.
https://www.academia.edu/12281964/Überlegungen_zur_Kosmographie_des_an
onymen_Geographen_von_Ravenna
GYÖRFFY, Gy. (1959): Tanulmányok a magyar állam eredetéről, Budapest.
GYÖRFFY, Gy. (1977): Honfoglalás, megtelepedés és kalandozások, in: Barta, A. –
Czeglédy, K. – Róna-TAS, A. (ed.:) Magyar őstörténeti tanulmányok. Budapest,
123–156.
KIEPERT, H. (1859): Descriptio orbis terrarum secundum geographum ravennatem.
[map]. Berlin. In: Mappae mundi. Die ältesten Weltkarten. Vol. 6. Hrsg. von Mil-
ler, K. Stuttgart, 1898, 53.
KISS, L. (1988): Bystric, Bistrica, Beszterce, Besztercény, Belgrád, Nándor, in Kiss, L.
(ed.): Földrajzi nevek etimológiai szótára, I‒II. Budapest.
MANGO, C. (1990): Nikephoros Patriarch of Constantinople Short History. Nicephori
Patriarchae Constantinopolitani Breviarium Historicum, Text, Translation and
Commentary Mango, C. (Corpus Fontium Historiae Byzantinae 13.) Washington.
MILLER, K. (1898): Mappae mundi. Die ältesten Weltkarten, Vol. 6. Hrsg. von Miller, K.
Stuttgart.
MOMMSEN, T. (1882): Iordanis Romana et Getica. Monumenta Germaniae Historica,
Auctorum antiquissimorum tomi v. pars prior, Berlin.
NIEBUHR, B. G. (1839): Theophanis Chronographia, Vol. I. Ed. Niebuhr, B. G. (Corpus
scriptorum historiae Byzantinae 43.) Bonn.
PEKKANEN, T. (1979): The Pontic ‟civitates” in the Periplus of the Anonymus Ra-
vennas. Acta Philologica Fennica 13/1, 111–128.
PINDER, M. E. – PARTHEY, G. (1860): Ravennatis Anonymi Cosmographia et Guidonis
Geographica, Berlin.
PODOSSINOV, A. V. (2002): Vostochnaya Evropa v Rimskoi kartograficheskoi tradicii,
Texty, perevod, kommentarii, Moskva.
POLGÁR, Sz. (2003): A »Bajor Geográfus« néhány Kelet-Európára vonatkozó ország
és népnevéről. Studia Archeologica 9, 243–258.
POTTS, D. T. (2017): Patria Persorum, India Dimirica-Evilat, and India Thermantica-
Elamitis in the Cosmographia of Ravennas Anonymous. Studia Iranica 46, 163–192.
POTTS, D. T. (2018): The Persian Gulf in the Cosmographia of the Anopnymous Ge-
ographer of Ravenna, c. 700 AD. Dabïr (Digital Archive of Brief notes and Iran Re-
view) 5, 57–70.
POTTS, D. T. (2019): Armenian Toponyms in the ‘Patria Quae Dicitur Parthia’ ac-
cording to the Cosmographia of Ravennas Anonymus, in Avetisyan, P. S. – Dan,
R. – Grekyan, Y. H. (ed.:) Over the Mountains and Far Away. Studies in Near East-
ern history and archeology presented to Mirjo Salvini on the occasion of his 80th
birthday. Oxford, 434–437.
RATHMAN, M. (2018): Tabula Peutingeriana. Die einzige Weltkarte aus der Antike,
Engeleitet und kommentiert von Rathman, M. Mainz.
SCHNETZ, J. (1925): Onogoria. Archiv für slavische Philologie 40, 157–160.
SCHNETZ, J. (1932): Neue Beiträge zur Erklärung und Kritik des Textes der Raven-
natischen Kosmographie. Philologus 87/1, 80–113.
SCHNETZ, L. (1934): Neue Beiträge zur Erklärung und Kritik des Textes der Raven-
natischen Kosmographie. Philologus 89/1–4, 85–101, 228–251.
SCHNETZ, J. (1940): Ravennatis Anonymi Cosmographia et Guidonis Geographica, Itin-
eraria Romana. Vol. II. Lipsiae.
SCHNETZ, J. (1951): Ravennas Anonymus: Cosmographia. Eine Erdbeschreibung um das
Jahr 700 [Acta Universitatis Upsaliensis. Nomina Germanica 10], Uppsala.
SCHNETZ, J. (1942): Untersuchungen über die Quellen der Kosmographie des anonymen
Geographen von Ravenna, München.
SZABÓ, M. (2005): A keleti kelták. A késő vaskor a Kárpát-medencében, Budapest.
SZÁNTÓ, R. (2017): A Bajor Geográfus és a korai magyar történelem. Acta Historica,
Acta Universitatis Scienciarum Szegediensis 139, 149–178.
STAAB, F. (1976): Ostrogothic Geographers at the court of Theodoric the Great. A
Study of some sources of the Anonymous Cosmographer of Ravenna. Viator
Medieval and Renaissance Studies 7, 27–64.
TABULA PEUTINGERIANA (TP): https://www.tabula-peutingeriana.de/
TURTLEDOVE, H. (1982): The Chronicle of Theophanes, Translation with introduction
and notes Turtledove, H. Philadelphia.
WEBER, E. (1976): Tabula Peutingeriana: Codex Vindobonensis 324. Vol. 1: Faksimile:
Originalformat. Vol. 2: Kommentar, Graz.
ZUMSCHLINGE, M. (1990): Ravennatis Anonymi Cosmographia et Guidonis Geographica.
Edidit Joseph Schnetz. Editio stereotypica editionis primae (MCMXL), Stuttgart.
Medieval Nomads and
the First Pandemic:
Theoretical Considerations
ULI SCHAMILOGLU
Abstract
This paper attempts to apply the author’s model for understanding the impact of
the Second Pandemic of bubonic plague (the “Black Death” in the mid-14th century
and its later waves) on the Golden Horde and Anatolia to the events in the steppe
and sedentary regions during the era of the First Pandemic (6th-8th centuries) be-
ginning in the time of the Byzantine Emperor Justinian. The paper tries to analyze
selected direct and indirect evidence for political disruption, depopulation, popula-
tion movements, decline of sedentary populations urban centers, cultural and
technological regression, the decline of existing literary languages and the rise of
new vernacular-based languages, and an increase in religiosity. It argues that, as in
the 14th-15th centuries, there is a close relationship between depopulation in seden-
tary centers and the in-migration of nomadic populations.
Keywords: Bubonic plague, Black Death, First Pandemic, Second Pandemic, de-
population, nomadic migrations, Byzantium, Avars, Bulgars, Türk
Introduction
A little over one month after the 8th International Conference on the Medieval His-
tory of the Eurasian Steppe took place in Sofia, Bulgaria in late November 2019, the
world began to be confronted by the growing COVID-19 pandemic caused by the
novel coronavirus known as “severe acute respiratory syndrome coronavirus 2”
 Nazarbayev University & University of Wisconsin-Madison (Emeritus). An earlier version

of this paper was presented at “Nomads and Their Neighbours in the Middle Ages. 8 th In-
ternational Conference on Medieval History of the Eurasian Steppe”, held in Sofia, Bulgaria,
on 20–3 November 2019. The publication of this work was included under Nazarbayev
University Grant Award Number 090118FD5332. I did not have physical access to a research
library or my own personal library while working remotely during the COVID-19 pandemic
of 2020–21.
22 ULI SCHAMILOGLU
(SARS-CoV-2). It is a reminder that pandemics have been with us throughout our

history and pre-history as well. Another important point is that the lives of seden-
tary and nomadic populations were intertwined in the past. In this article I would
like to draw upon my research on the Second Pandemic beginning in the 14 th cen-
tury to address the question of the First Pandemic beginning in the 6 th century and
how it might have affected the various populations in the Balkans, including both
sedentarists and nomads.
1. The Three Pandemics in History
Historians speak of three historical pandemics of bubonic plague caused by the
bacterium Yersinia pestis, namely the First Pandemic (6th-8th centuries CE), the Sec-
ond Pandemic (14th century on), and the Third Pandemic (19 th century on). There is
a substantial body of historiography for the Second Pandemic—the initial years of
which are better known as the “Black Death”—for Europe and a growing body of
literature for the Middle East as well. I have devoted numerous studies to trying to
understand the role of the Black Death in the collapse of the Golden Horde and its
aftermath in Central Eurasia. There is also a small but growing body of literature
devoted to the First Pandemic, also known as the “Plague in the Time of Justinian”.
The study of the Third Pandemic is still emerging as a field of inquiry among his-
torians.1
The cause of bubonic plague is the bacteria Yersinia pestis, which is a clone of
Yersinia pseudotuberculosis, with the date for this split given as 1,500-20,000 BP. I
should note that for the period preceding written historical sources samples of
ancient DNA (aDNA) of Y. pestis have been identified from Scandinavia (4900
years BP) and from Yakutia and the Cis-Baikal region (4400 years BP).2 These sam-
ples indicate that Y. pestis has been a Eurasia-wide phenomenon since the 3rd mil-
lennium BCE. It has been suggested that it may be responsible for the decline of
Neolithic and Bronze Age communities across Eurasia, including the disappear-
ance of large settlements such as Arkaim. These arguments are in line with the
arguments I have been making for the Second Pandemic and more recently for the
First Pandemic.
Studies of the genetic history of Y. pestis have established for the first time that
the site of the polytomy (or multifurcation) of the bacterium Y. pestis for all three
historical pandemics is the Qinghai-Tibet Plateau, along the intersection of the Silk
Road and the Horse-Tea Road.3 Recent scholarship has focused on what we can
learn from the genetic history of Y. pestis, the various rodents which form a vector
for animal to human transmission by fleas, and the role of trade networks in the
1 LITTLE 2011.
2 RASCOVAN 2018; KILINÇ 2021.
3 CUI et al. 2013.
MEDIEVAL NOMADS AND THE FIRST PANDEMIC: THEORETICAL CONSIDERATIONS 23
transportation of fleas.4 This is a rich and complex field and a review of this litera-
ture would be beyond the scope of this article. Rather, I would like to draw upon
my own studies to try to sketch out a framework for understanding what the re-
sults of the First Pandemic might have been in the Balkans.
2. The Second Pandemic as a Model for Understanding the First Pandemic
As a model for thinking about the effects of the First Pandemic I would first like to
dwell briefly upon the Second Pandemic. I shall begin with the general model
which I have constructed for understanding the collapse of the Golden Horde with
reference to Anatolia as well.5 This model considers the following phenomena:
1. Political disruption, with the collapse of the Mongol World Empire beginning
in some regions already in the 1330s. Contrary to expectation, the Golden Horde
remained a stable state until falling into anarchy following the death of Berdibek
Khan in 1359. I attribute this political disruption to the sudden sharp rise in mortal-
ity among the élite resulting in a crisis of succession in which it was no longer clear
who was closest in line to succeed the previous khan and/or the previous tribal
leader(s).
2. Sudden large-scale depopulation and later in-migration of new nomadic ele-
ments into newly depopulated territories. We see the case of relatively more severe
depopulation of the western territories of the Golden Horde (the Aq orda or “White
Horde”), with new tribal populations moving into the western territories under
Toqtamış Khan. These new tribal populations would form the basis of the later
Khanates of Kazan and the Crimea, for example. 6 We can also take the relative size
of the new khanates of the Later Golden Horde as a proxy for differences in the
size of population: whereas the Golden Horde controlled a territory from Lake
Zaysan to the west, the later khanates were much smaller states centered around
Kazan, the Crimea, Kasimov, Astrakhan, and Tümän.
3. In the case of the Golden Horde we see the decline of urban centers, includ-
ing the capital city Saray Berke, and probably sedentary agricultural populations as
well. Concurrently we see in the case of the territories of the Golden Horde the rise
in the relative power of nomadic confederations such as the Great Horde and later
the Nogai Horde. In the Middle Volga region it is clear that during the Golden
Horde there was a sizable sedentary population speaking Volga Bulgarian, but in
the aftermath of the Black Death we see that these centers have disappeared along
with the in-migration of Qipchaq Turkic speakers and the use of Old Tatar on
tombstones, replacing the earlier Volga Bulgarian language. 7
4 GREEN 2020.
5 SCHAMILOGLU 1993, SCHAMILOGLU 2004; SCHAMILOGLU 2017a.
6 SCHAMILOGLU 2018a; SCHAMILOGLU 2018b.
7 SCHAMILOGLU 2016b.
24 ULI SCHAMILOGLU
4. Cultural and technological regression is a well-known phenomenon for me-

dieval Europe. In the case of the Golden Horde a cessation in the manufacture of
tombstones in Volga Bulgarian as well as the sudden disruption in the production
of works in the literary language of the Golden Horde can be seen as signs of cul-
tural and technological regression. 8 The Islamic high culture of Saray Berke, which
included the production of scientific works in Arabic, was totally wiped out along
with the population of the city.
5. In the territories of the Golden Horde and Central Asia there appear new
vernacular-based literary languages replacing earlier languages. In Anatolia we see
the emergence of the modified orthographic system of Ottoman Turkish replacing
the earlier Central Asian orthographic tradition. 9 There is a temporary “renais-
sance” in the use of the Uighur script in the late 14 th–15th centuries as well, proba-
bly reflecting in-migration of scribes from territories further to the east which
would have been less affected by the pandemic than territories further to the west.
6. It is well known from medieval Europe that there was a shortage of labor (in-
cluding skilled labor) resulting in a rise in the price of labor leading to inflation as
well. There are some indicators for this for the territories of the Golden Horde as
well.
7. Finally, as we know well from medieval Europe, there is a heightened religi-
osity in this period. In the case of the Golden Horde there is a rise in the produc-
tion of religious literature focusing on how to get to Heaven (since, after all, the
population was being punished by God for being insufficiently devout Muslims).
There is also literature in Anatolia which reflects an increase in religiosity and
morbidity in this period. In the second half of the 14 th century, we also see a
movement to build cathedral mosques in Anatolia as well as the establishment of
new monasteries in Greek Orthodox territories, for example.
Lest one argue that the Balkans (or Central Europe for that matter) escaped the
ravages of the Black Death in this period, I would suggest that the rather limited
coverage of this phenomenon in the historiography does not mean that it escaped
its devastation. Quite the opposite. As I have argued elsewhere, the rise of the Ot-
toman Empire is to be understood as the semi-nomadic or nomadic Ottomans be-
ing less impacted by the pandemic which afflicted the coastal regions of the Medi-
terranean and the Aegean.10 I am writing these lines while riding out the pandemic
in Bodrum, Turkey and slowly collecting indirect evidence for establishing the
effects of plague upon the beylik of Menteshe.
There is extensive evidence for the ravages of the Black Death in Byzantium. 11
In Constantinople, the population would fall to around 40,000 by the time of its
8 SCHAMILOGLU 1991.
9 SCHAMILOGLU 1991; SCHAMILOGLU 2012.
10 SCHAMILOGLU 2004.
11 SCHAMILOGLU 2004; STATHAKOPOULOS 2016.
conquest by Mehmed the Conqueror in 1453. In Anatolia there is evidence for un-
harvested fields and untended herds of animals as well. The Black Death is rarely
considered as a factor in the historiography, however, see for example the article
on medieval Anatolia in the Cambridge History of Turkey by Rudi Lindner, who ig-
nores the topic completely.12 In the same volume Machel Kiel writes only that “the
long Byzantine civil wars, combined with the disastrous effects from 1348 onward
of the Black Death (to which the settled population was more vulnerable than the
mobile Turks), had well prepared the soil for Ottoman take-over”.13 Fortunately,
by now we have excellent studies of the Black Death in the Ottoman Empire.14
Already beginning in the mid-14th century, the Ottoman Empire began its ex-
pansion from Anatolia into the Balkans, bypassing Constantinople until its con-
quest later in the mid-15th century. I would argue that the rapid expansion of the
Ottomans into the Balkans can be explained in large measure by expansion into a
territory affected by severe depopulation. Yet I am not aware of sources available
to me offering direct written evidence for the depopulation or its causes. Kiel be-
lieves the territory was prepared by the Black Death for rapid expansion by the
Ottomans, but he offers no sources. Varlık points out that the Black Death spread
widely throughout Rumelia and that Bulgaria was astride one of the major trade
routes through the Balkans.15
In an article in honor of Bernt Brendemoen, who is an expert on the Turkish
dialect of modern Trabzon, I undertook to reexamine the problem of the similari-
ties which have been noted between that NE Anatolian dialect and western Bulgar-
ian dialects of Turkish which were already endangered when they were studied by
Gyula Németh in the mid-20th century.16 I began to understand, largely on the basis
of the studies of the Ottoman historian Ömer Lütfi Barkan, that there is evidence
available on forced internal migration (sürgün) in Anatolia and Rumelia over the
course of the 15th century. The Ottomans had a regular practice of resettling no-
mads throughout Anatolia as well as in Istanbul. Once the Ottomans seized control
of the region of Trebizond in 1461, they began to resettle Türkmen nomads from
that region to Western Bulgaria. In the view of Barkan, it was not only a question
of reducing overcrowding, maintaining control over the nomads, and getting more
tax revenue from a sedentarized population in western Bulgaria as opposed to the
agriculturally poor region of Trebizond, but also of establishing control over the
newly-conquered territories. I understand that the Ottomans were repopulating a
region which had apparently emptied out as a result of recurring waves of the
12 LINDNER 2009.
13 KIEL 2009, 154.
14 AYALON 2015; VARLIK 2015.
15 VARLIK 2015, 107ff., 134 (Map 2. Plague Networks, 1453-1600), 137 (Map 3. Plague Net-
works in the First Phase, 1453–1517).

16 NÉMETH 1965; SCHAMILOGLU 2018c.
26 ULI SCHAMILOGLU
Black Death beginning in the mid-14th century and continuing over the course of
the 15th century. This policy of forced internal migration resulted in the Turkifica-
tion and Islamization of the Balkans, which would last until the rise of modern
nation-states in the late 19th century and the subsequent ethnic cleansing of Turks
even late into the 20th century. In his study of the Ottoman expansion into the Bal-
kans, including a critique of Barkan’s studies on the Ottoman policy of forced re-
settlements in the Balkans, Nikolay Antov does not consider that this policy might
have been related to depopulation as a result of the Black Death, a phenomenon
which he does not mention at all.17
3. The First Pandemic: an outline
Now that we can point to the Qinghai-Tibet plateau as the site of the origin of the
outbreak of the First Pandemic, we understand that it may have arrived in Ethiopia
via the Indian Ocean and the Red Sea.18 According to Procopius (see below), after
reaching Ethiopia, the plague then reached Pelusium in Egypt, whence it spread
throughout Byzantium. There is indirect evidence suggesting that it afflicted sed-
entary Central Asia, too, but we do not have any direct evidence for this. 19 At any
rate, the First Pandemic did not reach Constantinople via the Black Sea, unlike in
the 14th century. Once the pandemic arrived in Byzantium in 541–2, the plague
revisited Constantinople in 543, 544, 558, 573, 574, 599, 608? (sometime between
602-10), 618, 640 (?), 697, 700 (?), 717, and 747. 20 Theophanes mentions the plague
under the years 541–2, 555–6, 557–8, 560-1, 683–4, 725–6, 732–3, and 745–6.21
Procopius in his History (Book II, xxii-xxiii and xxiv: 8, 12) describes the feverish
symptoms and the human toll reaching 5,000 deaths per day and later rising to
10,000 deaths per day.22 As in the 14th century, Procopius reports that the popula-
tion could not keep up with the burial of the dead. A similarly grim portrayal is
offered by Theophanes two centuries later for the outbreak of 745–6.23 There can be
no doubt that the pandemic had a profound impact on the population of Byzanti-
um and spread widely to the rest of Europe, the Middle East, and presumably Af-
rica as well, see the essays in the volume Plague and the End of Antiquity (individual
chapters cited below).24
There is no basis for assuming with confidence that the Balkans or Pannonia
were able to avoid the ravages of the disease, since it appears that the Second Pan-
17 ANTOV 2017, 37, 41–7.

18 SARRIS 2006, 121–2; SCHAMILOGLU 2016c.
19 SCHAMILOGLU 2016c.
20 BIRABEN, 1975–6, vol. 1, 27–32.
21 MANGO 1997, 322, 337, 340, 345, 389, 503, 559, 569, 573, 585–6.
22 DEWING 1954, 451–73, 475, 477.
23 MANGO 1997, 585–6.
24 SCHAMILOGLU 2004; MATANOV 2004; STATHAKOPOULOS 2006; DIMITROV 2011; DIMITROV
2015; DIMITROV 2016; KOVACHEV 2017.

demic had a devastating impact on the Balkans, for example. Of course, at the
same time we also know from the 14th century that while some towns or neighbor-
hoods could lose 90% of their population, other nearby towns or neighborhoods
might remain unaffected. For the purposes of this paper, I will assume that regions
such as the Balkans and Pannonia also suffered from massive depopulation, since
on the scale of the Old World this pandemic was a universal phenomenon. The rest
of my paper will be devoted trying to think systematically about the First Pandem-
ic to outline a framework for discussing some major historical developments in the
Balkans and Pannonia in the 6th–8th centuries.
4. Seeking direct and indirect evidence: an agenda for research
We are at a great disadvantage in the study of the First Pandemic in comparison to
the Second Pandemic because of the dearth of primary sources, though Byzantium
represents a notable exception compared to other regions. This does not mean,
however, that we should not seek direct and indirect evidence for its impact, on the
contrary.25 After all, it was a universal phenomenon in the Old World, as already
noted, and at least for Byzantium there is substantial evidence for numerous waves
of plague striking especially Constantinople. The Black Sea coastal regions of the
Balkans are adjacent to the region of Constantinople and, as we know, maritime
routes were another highly effective route of transmission during all three histori-
cal pandemics.
The first area we might consider is the Ponto-Caspian steppe region. This zone
north and northeast of the Black Sea apparently had a substantial population of
various nomadic groups speaking a Western Turkic language with ethnonyms
ending in -gur such as the Utrigur, the Kutrigur, etc. plus other groups speaking a
Turkic (?) language such as the Sabirs.26 Golden has noted that the Sabirs could
field an army of 100,000 early in the 6th century, but later he also expressed the
view that such a claim was an exaggeration. 27 Nevertheless, we may consider that
when the Avars first arrived in the Ponto-Caspian region in 557–8, so approximate-
ly 16–17 years after the arrival of the pandemic in Constantinople, it is possible that
the populations north of the Black Sea may already have been affected by the dis-
ease, in which case the result would have been a dramatic decline in population. It
is not clear whether the pandemic may have already affected the territories in the
east whence the Avars came and whether that gave them a numerical advantage
(which becomes a factor in the late 14th–early 15th centuries). Later, the Avars cap-
tured Sirmium in 582 and moved into Pannonia and the other territories which
they now controlled. They remain a force to be reckoned over a considerable peri-
od of time. It appears that the Avars also absorbed elements of the earlier local
25 LITTLE 2006, 18–21.

26 RÓNA-TAS 1999, 209–14.
27 GOLDEN 1980, vol. 1, 33–9; GOLDEN 1992, 105.
28 ULI SCHAMILOGLU
groups already mentioned, but the relative proportions cannot be known except
that the Avars were clearly the dominant force militarily. It has also been argued
that the Avars lost population as a result of the wave of pandemic in 597–9.28
The flight of the Avars represents both an out-migration out of the eastern Eur-
asian steppe region following their defeat by the founders of the Türk Khaganate
and concurrently an in-migration to a relatively depopulated area, first the area
north of the Ponto-Caspian steppe, then to Pannonia. Following the departure of
the Avars, the Türk themselves began an expansion across Central Eurasia, which
also led to the beginning of the process of the Turkification of Central Asia. (This
suggests that they had not yet been afflicted by the pandemic.) I have argued else-
where that this must be seen at the same time as a period of decline for Sogdian
population and language.29 There is also some indirect evidence for the restructur-
ing of urban centers in Central Asia. These phenomena could be the results of
waves of epidemic, even though I am not aware of direct historical evidence for
any specific waves. The Türk appear to have reached the Black Sea region very
quickly, but their rapid and dramatic expansion across a territory which must have
been undermined demographically in the meantime was followed by an equally
rapid and dramatic disintegration of the First Türk Empire. The Western Türk
Khaganate found itself in a situation of anarchy in the western regions and because
of this we may never understand exactly the emergence of the Khazar state after
630.30 A clarification of the chronology of the waves of plague during the First
Pandemic remains a critical desideratum for the study of sedentary Central Asia
and the steppe region in this period.
I have argued that in Byzantium the rise of the themes system in Anatolia (640s-
?) was a response to depopulation and the lack of manpower to govern directly. As
a part of this process there was also an influx of Slavic populations into the Byzan-
tine territories to serve as soldiers, which may have served to reduce Slavic popula-
tions in the Balkans (?) even further.31 It has also been suggested that in the 7th-8th
centuries NE Thrace may have been a zone of depopulation because of frequent
outbreaks of plague.32 It is against this backdrop that we may consider the estab-
lishment of the Bulgarian states.
Magna Bulgaria was a confederation of Onogur Turks led by Kubrat ca. 630–5
which was defeated by the Khazars several decades later. Eventually, one group
went west under Asparukh and founded the Danubian Bulgarian state ca. 681.
Another group went north to the Volga-Kama confluence and was visited by Ibn
28 STATHAKOPOULOS 2006, 102–3; KOVACHEV 2017. On the Avars see SZÁDECZKY-KARDOSS

1990; RÓNA-TAS 1999, 209–14; POHL 2018.
29 SCHAMILOGLU 2016c; SCHAMILOGLU 2017b.
30 GOLDEN 1980, i; GOLDEN 1992, 233ff.; RÓNA-TAS 1999, 228–234.
31 SCHAMILOGLU 2004; LITTLE 2006, 24.
32 KOVACHEV 2017.
Faḍlān in 922 CE. (This region was apparently largely uninhabited for a period of
time before the arrival of the Bulgars, no doubt another consequence of the First
Pandemic.) A third group stayed behind and some scholars consider them the an-
cestors of the modern Balqars.33 Given the limited number of reports of plague
years for the mid-7th century, how do we explain the in-migration of the Onogurs
into the Balkans under Asparukh? Were there additional waves about which we do
not have reports, or might it be that the retrenchment of Byzantium to its core terri-
tories in Anatolia led to a vacuum in the future territories of the Danubian Bulgari-
an state, just like in NE Thrace? This remains a matter for further consideration.
In the east there appear to have been periods of major outbreaks of plague in
636–55 and 682–707; these waves may or may not have extended to the west. 34 It is
in this period that the Second Türk Khaganate emerges (682–744). As these waves
of plague subside we see the growing predominance of the Second Türk Khaga-
nate, whose great figures are described in the Old Turkic runiform inscriptions of
the early 8th century. Is it possible that the waves of plague did not impact the Türk
adversely in the precise location where they are residing in the period 682-702(?)?35
In the 14th century the eastern territories were affected less drastically by
plague, for which reason elements of eastern populations were able to migrate to
the west. This may be indicative of a population rebound in the east already in the
late 14th century. Can we speak of a parallel phenomenon then during the rise of
the Second Türk Khaganate? If so, this raises the question of what the underlying
reason for such parallel phenomena in the 7th and 14th–15th centuries might be,
either some aspect of the climate and geography in the western region in compari-
son to the eastern steppe region which meant that the effects of plague were not as
profound in the east, 36 or perhaps the creation of foci in the west which made the
situation markedly worse compared to the eastern territories. Some population
movements between 541 and 747 or so might be groups of population escaping to a
relatively unpopulated region, as might be the case with the Avars and later the
Bulgars who migrated and founded states in the Balkans and the Middle Volga
region.37 There is little basis, however, for discussing the First Pandemic as a recur-
ring phenomenon in the western regions past the midpoint of the 8th century.
Finally, some cultural considerations as informed by an awareness of the mid-
6th to mid-8th centuries as having been a time of pandemic. As I have already noted
elsewhere, Vasiliev considered this period to have been a “dark age” in Greek let-
33 GOLDEN 1992, 244–58; RÓNA-TAS 1999, 215–28.

34 For occurrences of plague in the west in this period see MORONY 2006, 65, 73, 76, 84–5;
STATHAKOPOULOS 2006, 104; KULIKOWSKI 2006, 154; MADDICOTT 2006, 200; MCCORMICK 2006,
311n.
35 See my “Turkological Notes on Pandemics”: SCHAMILOGLU (in press).
36 SCHAMILOGLU 2016a.
37 For a general discussion of population movements in this period see HAYS 2006, 46–9.
30 ULI SCHAMILOGLU
ters.38 This would be an example of cultural and technological regression. Of

course, we can hardly speak of vernacular literatures in this period, but as in the
14th–15th centuries, we see the beginnings of the use of a vernacular language, if not
a vernacular literature. This is a very complex topic, as will be seen below, and for
the west in the 8th century it is much more speculative than for the east. 39
At some point in time, we see the beginning of Danubian Bulgarian inscriptions
using the Greek alphabet, with many of them dating from the time of Omurtag in
the first half of the 9th century (r. 814–31).40 We also see the controversial issue of
the use of the runiform alphabet for Avar as well as for Danubian Bulgarian and in
the “Treasure of Nagyszentmiklós”, which also used the Greek alphabet. The pre-
cise dating of all of these is problematic. 41 If we can say in the end that the eastern
Eurasian runiform alphabet as used for Old Turkic is introduced in the west, could
this be considered a parallel to the situation in the mid-14th to 15th centuries when
there was a “renaissance” of the Uighur script in the place of the Arabic script?
One problem with this is the wide variation in the kind of runiform alphabet used,
while another is that these inscriptions have not all been reliably deciphered. 42 The
fact they may date from a period after the end of the pandemic is not a problem,
however, since this would represent a parallel in some ways to the “renaissance”
of the Uighur script in the late 14th-15th centuries, in all likelihood following the
end of the pandemic in the east. Another question is whether we should conclude
that the territories of the Golden Horde and the Chagatay Khanate were already in
a far more advanced position with regard to the level of literacy in the 14th-15th
centuries in comparison to the Balkans and Pannonia in the 6 th–8th centuries, or did
the Black Death end relatively earlier in the east, where it must have been less de-
structive? Reversing the question, may we conclude that the devastation of the
First Pandemic really was so great (as some scholars believe)? In this regard litera-
cy and literary production can serve as proxy data for discussing the severity of the
effects of the pandemic.
Finally, I will consider one more issue here, namely the question of religious
conversion, which seems to run parallel to the question of the rise of vernacular
writing, if not literature. It is only the conversion of the Khazars to Judaism which
may date to before the mid-8th century, but is more likely to have taken place (at
once or in stages) in the 8th century.43 The Danubian Bulgarian conversion to Chris-
tianity took place in 864. As with the inscriptions produced by the ruling élite of
these peoples, it is a century or more later than the end of the waves of pandemic,
38 VASILIEV 1958, 181–2.

39 SCHAMILOGLU 2017b.
40 MENGES 1951; BESHEVLIEV 1962; RÓNA-TAS 1999, 113, 167; AGYAGÁSI 2019, 18.
41 For a critical analysis of the corpus of the Treasure of Nagyszentmiklós and the dating of
its inscriptions to the mid-7th to 8th centuries, see RÓNA-TAS 2016.

42 RÓNA-TAS 1999, 128–32.
43 GOLDEN 2007.
to the best of our knowledge. The mission of Cyril and Methodius and the creation
of the Glagolitic script are also from the 9th century. This stands in sharp contrast to
the other state conversions or the construction of great religious temples elsewhere
in Eurasia in the mid-8th century.44 There are many other topics which researchers
should also consider, such as the rise and fall of towns and urban centers, but in
the case of the Balkans and Pannonia this is an area which I will leave to other
scholars who are more qualified to discuss this topic.
In conclusion, the lack of adequate written sources should not serve as a pretext
to avoid pushing the boundaries of our understanding of the realities on the
ground in the era of the First Pandemic. I would like to think that I have demon-
strated that it is possible to frame questions and that a certain level of analysis is
still possible, albeit less satisfying than for the 14 th–16th centuries. The fact that we
know that the First Pandemic, the Second Pandemic, and the Third Pandemic all
played a major role in world history makes it incumbent upon scholars to attempt
to explain the past through the direct and indirect evidence available to us through
the framework offered by the three pandemics.
Bibliography
AGYAGÁSI K. (2019): Chuvash Historical Phonetics: An areal linguistic study. With an
Appendix on the Role of Proto-Mari in the History of Chuvash Vocalism, Wiesbaden.
ANTOV, N. (2017): The Ottoman “Wild West”: The Balkan Frontier in the Fifteenth and
Sixteenth Centuries, New York.
AYALON, Y. (2015): Natural Disasters in the Ottoman Empire: Plague, Famine, and Other
Misfortunes, New York.
BESHEVLIEV, V. (1962): Parvo-Balgarski nadpisi, Sofia.
BIRABEN, J.-N. (1975–1976): Les hommes et la peste en France et dans les pays européens
et méditerranéens, vol. 1–2, Paris.
CUI, Y. et al. (2013): Historical variations in mutation rate in an epidemic pathogen,
Yersinia pestis. Proceedings of the National Academy of Sciences of the United States
of America 110/2 (January 8), 577–582.
DEWING, H. B., transl. (1954): Procopius. History of the Wars, vol. 1, Cambridge.
DIMITROV, T. (2011): Kam vazdeystvieto na taka narechenata Yustinianova chuma
(541–748) varkhu demografskata struktura na Vizantiyskata imperiya. Istori-
cheski pregled 1–2, 5–15.
DIMITROV, T. (2015): Mors intra urbem: za vrazkata mezhdu Yustinianovata chuma
i institutsionaliziraneto na pogrebeniyata intra muros vav Vizantiya. Bulgaria
Mediaevalis 6/1, 297–306.
DIMITROV, T. (2016): ’Majores mures, qui vulgariter rati vocantur’: cherniyat plakh
(Rattus rattus) i Yustinianovata chuma. Bulgaria Mediaevalis 7/1, 303–316.
44 SCHAMILOGLU 2016c.
32 ULI SCHAMILOGLU
GOLDEN, P. B. (1980): Khazar Studies. An Historico-Philological Inquiry into the Origins

of the Khazars, Vol. 1–2, Budapest.
GOLDEN, P. B. (1992): An Introduction to the History of the Turkic Peoples, Wiesbaden.
GOLDEN, P.B. (2007): The Conversion of the Khazars to Judaism, in Golden, P. —
Ben-Shammai, H. — Rona-Tas, A. (ed.:) The World of the Khazars. Leiden, 124–162.
GREEN, M. H. (2020): The Four Black Deaths. The American Historical Review 125/5,
1601–1631.
HAYS, J.N. (2006): Historians and Epidemics: Simple Questions, Complex Answers,
in Little, L.K. (ed.:) Plague and the End of Antiquity. The Pandemic of 541–750.
Cambridge, 33–56.
KIEL, M. (2009): The incorporation of the Balkans into the Ottoman Empire, 1353–
1453, in Fleet, K. (ed.:) The Cambridge History of Turkey. Vol. 1. Byzantium to Turkey,
1071–1453. Cambridge, 138–191.
KILINÇ, G. M. et al. (2021): Human population dynamics and Yersinia pestis in an-
cient northeast Asia. Science Advances 7/2 (06 Jan): EABC4587.
KOVACHEV, G. (2017): Za chumnite epidemii prez rannoto srednovekovi. Epokhi
25/2, 437–447.
KULIKOWSKI, M. (2006): Plague in Spanish Late Antiquity, in Little, L. K. (ed.:)
Plague and the End of Antiquity. The Pandemic of 541–750. Cambridge, 150–170.
LINDNER, R. (2009): Anatolia, 1300–1451, in Fleet, K. (ed.:) The Cambridge History of
Turkey, vol. 1, Byzantium to Turkey, 1071–1453. Cambridge, 102–137.
LITTLE, L. K. (2006): Life and Afterlife of the First Plague Pandemic, in Little, L. K.
(ed.:) Plague and the End of Antiquity. The Pandemic of 541–750. Cambridge, 3–32.
LITTLE, L. K. (2011): Plague Historians in Lab Coats. Past and Present 213 (Novem-
ber), 267–290.
MADDICOTT, J. (2006): Plague in Seventh-Century England, in Little, L. K. (ed.:)
Plague and the End of Antiquity. The Pandemic of 541–750. Cambridge, 171–214.
MANGO, C. et al., transl. (1997): Theophanes. The Chronicle of Theophanes Confessor.
Byzantine and Near Eastern History, AD 284–813. Oxford.
MATANOV, KH. (2004): Chumni pandemii i istoriya. ‘Yustinianovata Chuma’ i
‘Chernata Smart’: Dva povratni perioda v istoriyata na srednovekovite Balkani,
in Stepanov, Ts. —Vačkova V. (ed.:) Civitas divino-humana: in honorem annorum
LX Georgi Bakalov. Sofia, 339–346.
MCCORMICK, M. (2006): Toward a Molecular History of the Justinianic Pandemic,
in Little, L. K. (ed.:) Plague and the End of Antiquity. The Pandemic of 541–750.
Cambridge, 290–312.
MENGES. K. H. (1951): Altaic Elements in the Proto-Bulgarian Inscriptions. Byzan-
tion 21/1, 85–118.
MORONY, M. G. (2006): ‘For Whom Does the Writer Write?’: The First Bubonic
Plague Pandemic According to Syriac Sources, in Little, L. K. (ed.:) Plague and
the End of Antiquity. The Pandemic of 541–750. Cambridge, 59–86.
NÉMETH, Gy. (1965): Die Türken von Vidin: Sprache, Folklore, Religion, Budapest.
POHL, W. (2018): The Avars: A Steppe Empire in Central Europe, 567–822, Ithaca.
RASCOVAN, N. et al. (2018): Emergence and Spread of Basal Lineages of Yersinia
pestis during the Neolithic Decline. Cell 176/1–2, 295–305.e10.
RÓNA-TAS, A. (1999): Hungarians and Europe in the Early Middle Ages. An Introduction
to Early Hungarian History, Budapest.
RÓNA-TAS, A. (2016): The Treasure of Nagyszentmiklós: A Golden Contribution to
the Reconstruction of History, in Csató, É. Á. — Johanson, L. — Róna-Tas, A.
— Bo Utas (ed.:) Turks and Iranians. Interactions in Language and History. Wies-
baden, 107–116.
SARRIS, P. (2006): Bubonic Plague in Byzantium: The Evidence of Non-Literary
Sources, in Little, L.K. (ed.:) Plague and the End of Antiquity. The Pandemic of 541–
750. Cambridge, 119–132.
SCHAMILOGLU, U. (1991): The End of Volga Bulgarian, in Varia Eurasiatica. Festschrift
für Professor András Róna-Tas. Szeged, 157–163.
SCHAMILOGLU, U. (1993): Preliminary Remarks on the Role of Disease in the History
of the Golden Horde. Central Asian Survey 12/4, 447–457.
SCHAMILOGLU, U. (2004): The Rise of the Ottoman Empire: The Black Death in Me-
dieval Anatolia and its Impact on Turkish Civilization, in ed. Yavari, N. – Pot-
ter, L. G. – Oppenheim, J.-M. Views From the Edge: Essays in Honor of Richard W.
Bulliet. New York, 255–279.
SCHAMILOGLU, U. (2012): Ortaçağ Dillerinden Modern Dillere: Avrupa ve Türk Dü-
nyasında Yeni Edebi Dillerin Varlığa Gelmesi, in Önal, M. N. (ed.:) IV. Dünya
Dili Türkçe Sempozyumu Bildirileri. 22-24 Aralık 2011, Muğla. 2 volumes. Ankara,
vol. 1, 501–507.
SCHAMILOGLU, U. (2016a): Climate Change in Central Eurasia and the Golden
Horde. Golden Horde Review / Zolotoordynskoe obozrenie 4/1, 6–25.
SCHAMILOGLU, U. (2016b): Gravestones and the History of Medieval Tatar Civiliza-
tion: A Brief Overview, in Nigamaev, A. Z. (ed.:) Islam i tyurkskii mir: Problemy
obrazovaniya, yazyka, literatury, istorii i religii. Materialy VIII mezhdunarodnoi
tiurkologicheskoi konferentsii. Kazan, 76–80.
SCHAMILOGLU, U. (2016c): The Plague in the Time of Justinian and Central Eurasian
History: An Agenda for Research, in Karatay, O. – Zimonyi, I. (ed.:) Central Eur-
asia in the Middle Ages. Studies in Honour of Peter B. Golden. Wiesbaden, 293–311.
SCHAMILOGLU, U. (2017a): The Impact of the Black Death on the Golden Horde:
Politics, Economy, Society, and Civilization. Golden Horde Review /
Zolotoordynskoe obozrenie 5/2, 325–343.
SCHAMILOGLU, U. (2107b): The Rise of Runiform Turkic as the first Turkic vernacu-
lar literary language. Turkic Languages 21/2, 161–177.
SCHAMILOGLU, U. (2018a): Reflections on Regional Variation in the Impact of the
Black Death on the Golden Horde, in Mirgaliyev, I. M. (ed.:) Epidemii i prirod-
nye kataklizmy v Zolotoi Orde i na sopredel’nykh territoriyakh (XIII-XVI vv.).
Sbornik nauchnykh statei. Kazan, 20–22.
34 ULI SCHAMILOGLU
SCHAMILOGLU, U. (2018b): Reflections on Regional Variation in the Impact of the

Black Death on the Golden Horde. Tiurkologicheskie issledovaniya / Turkological
Studies 1/3, 19–22.
SCHAMILOGLU, U. (2018c): Revisiting ‘Randosmanisch’: A Hypothesis on the Black
Death and the Turkification of the Balkans, in Csató, É. Á. – Parslow, J. – Türk-
ler, E. – Wigen, E. (ed.:) Building Bridges to Turkish. Essays in Honour of Bernt
Brendemoen. Wiesbaden, 285–294.
SCHAMILOGLU, U. (in press): Turkological Notes on Pandemics.
STATHAKOPOULOS, D. (2006): Crime and Punishment: The Plague in the Byzantine
Empire, 541–749, in Little, L.K. (ed.:) Plague and the End of Antiquity. The Pan-
demic of 541–750. Cambridge, 99–118.
SZÁDECZKY-KARDOSS, S. (1990): The Avars, in Sinor, D. (ed.:) The Cambridge History
of Early Inner Asia. Cambridge, 206–228.
VARLIK, N. (2015): Plague and Empire in the Early Modern Mediterranean World: The
Ottoman Experience, 1347–1600, New York.
VASILIEV, A. A. (1958): History of the Byzantine Empire, 324–1453, Vo. 1–2, Madison.
Some Notes on the Missions of Pei Ju
KÜRŞAT YILDIRIM
Abstract
Pei Ju was a high official in the Sui Empire. He was sent as an emissary to the
Turks in 593 and played an active role during Sui Emperor Wen's marriage alliance
with the royal house of the Turk Khaganate as well as during their conflict. His
actual rise and role in policy-making coincided with the reign of the Sui emperor
Yang. The emperor gave Pei Ju permission to oversee the commerce in Zhangye.
He also wrote Xiyu Tuji (“Illustrated Records of the Western Regions”). Pei set
about convincing Emperor Yang to develop both a new policy for the Xiyu (West-
ern Regions) as well as a divisive policy for dealing with the Turks. Using Pei Ju’s
plan, the Sui wanted to establish tributary and trade relations with the cities of
Xiyu; however, he was faced with two obstacles: the Turkic Empire and the King-
dom of Tuyuhun. The emperor accepted Pei Ju’s plan about the Turks and the
Tuyuhun. I will examine some elements of the historical and strategical policy of
the Chinese people on the example of Pei Ju.
Keywords: Pei Ju, Turks, Turk Khaganate, Tuyuhun, Koreans, Chinese, the Sui
Dynasty, Emperor Yang, Xiyu, Western Regions, Xiongnu
Pei Ju 裵矩 (547–627) was a high official in the Sui Dynasty (581–618). His courtesy
name was Hongda 弘大, and he was from Hedong. His grandfather, Ta, had served
as an imperial secretary (duguan shanghu 都官尚書) in the Northern Wei Dynasty
(386–534/5). His father, Nezhi, was the heir apparent’s secretary (taizi sheren
太子舍人) in the Northern Qi Dynasty (550–77). While Pei Ju grew up alone and un-
der very difficult circumstances, he nevertheless received a decent education. He
was wise and loved literature and rhetoric. His uncle, Rangzi, had reportedly told
him: “I see both holiness and wisdom in you, I see complete talent. I would like to
place you in the state service.”1 Before long, we would then see him as an official
serving throughout many Chinese states.
 Istanbul University. The article is a revised version of a paper delivered at the 8 th Interna-
tional Conference on Medieval History of the Eurasian Steppe Nomads and their Neighbours in
the Middle Ages, held in Sofia, November 20–3, 2019.
1 SUI SHU 1997, 1577.
36 KÜRŞAT YILDIRIM
When the Turkic Empire threatened the Sui in 593, Pei Ju was sent as an emis-
sary to the Turks. He played an active role during Sui Emperor Wen’s (581–604)
marriage alliances2 with the Turkic Khagan Dulan (Ashina Yongyulü) and prince
Ashina Rangan as well as during their conflict. Later, Wen sent him as an emissary
to Ashina Rangan, whom the Sui recognized as Khagan.
His actual rise and role in policy-making coincided with the reign of the Sui
emperor Yang (604–18). After Emperor Yang ascended the throne, he commis-
sioned the building of a new Eastern Capital in which Pei Ju was to construct the
governmental buildings. At that time, various people from the Xiyu, or “the West-
ern Regions”, came to Zhangye in order to trade with China. The Emperor gave Pei
Ju permission to oversee that commerce, upon which Pei Ju invited Hu merchants.
It was there that he also gathered information about his observations and compiled
it into the three-volume Xiyu Tuji 西域圖記 (“Illustrated Records of the Western Re-
gions”).3
Following the death of Emperor Yang’s key advisor, Gao Jiong, in 607, Pei Ju’s
influence began to increase. He became one of the five nobles at the Sui palace that
were in the inner circle of Emperor Yang. 4
Pei set about convincing Emperor Yang to develop both a new policy for the
Xiyu (Western Regions) as well as a divisive policy for dealing with the Turks.
Using Pei Ju’s plan, Sui wanted to establish tributary and trade relations with the
cities of Xiyu; however, the Emperor faced two obstacles: the Turkic Empire and
the Kingdom of Tuyuhun. Nevertheless, once various states in Xiyu were subju-
gated, Turkic and Tuyuhun peoples could be destroyed. Lengthy discussion be-
tween him and the Emperor had led Pei Ju to underline that Xiyu was filled with
many riches that the Tuyuhun could seize. The emperor accepted Pei Ju’s argu-
ments and ordered him to go ahead with his plans. 5
Most of our information on Pei Ju and his activities comes from various records
found in the Sui Shu, the official history of the Sui Dynasty written in 636, and most
notably his biography found in volume 67. The biography contains a preface that
sheds some light upon the missing Xiyu Tuji. With this, we can take a closer look at
Pei Ju’s actions, and do so under several subchapters.
Tuyuhun
During the early years of the Emperor Yang’s reign, the Sui focused on the
Tuyuhun within the framework of Central Asian politics. Pei Ju appealed to the
Töles in order to settle the Tuyuhun affairs. The Töles themselves had previously
2 Marriage alliance (Chinese: Heqin) refers to the historical practice of the Chinese emperors
to marry princesses to rulers of the neighbouring states.
3 SUI SHU 1997, 1578.
4 XIONG 2006, 47.
5 SUI SHU 1997, 1578–81.
SOME NOTES ON THE MISSIONS OF PEI JU 37
been under the control of the Western Turks. With the fall of Turkic power in 605,
the Töles rebelled and established their own khaganate, and ruled various city-
states in the Xiyu. They attacked the Sui border in 607 and then offered to make
peace with the Sui. Pei Ju saw this as an opportunity to implement his plan: he
visited the Töles and persuaded them to attack the Tuyuhun in 608. 6 The position
of the Tuyuhun in Qinghai (Köke Nur) was an important obstacle on China’s path
to the Xiyu. Additionally, they attacked Zhangye and formed a marriage alliance
with Qimin Khagan of the Turkic Empire, a close relationship that would end up
threatening China. Even though the Tuyuhun Khagan Fuyun sent his son, Shun, as
a hostage to China, Emperor Yang still wanted to destroy the Tuyuhun Kingdom.
Eventually, Prince Shun was arrested by the Sui. When the Töles defeated the
Tuyuhun in 608, the emperor sent his army against the latter. Tuyuhun’s territory
was annexed by Sui as part of the Chinese provinces Xihai and Heyuan. Districts,
garrisons, and border posts were established there. A Chinese general in exile had
established farming in Heyuan, thus assuring the Tuyuhun that the road to Xiyu
would be open, and ultimately allowing Fuyun and his people to flee. 7
Xiyu (Western Regions)
In 608, China sent a general to take over the Yiwu (Qumul). There, he established a
garrison of a thousand soldiers, ultimately ensuring trade with the Xiyu. In 609, the
Turfan king Qu Boya, under the influence of Pei Ju, came to the Chinese court and
paid homage to the emperor. Many cities followed him and that same year Pei Ju
managed to gather envoys from 27 cities in Wuwei, where the emperor granted
them gifts. Emperor Yang then established the Xiyu Jiaowei (Western Territorial
Commandery), setting up four commanderies from the east of Qinghai to the south
of Xinjiang: Heyuan 河源, Xihai 西海, Qiemo 且末, and Shanshan 鄯善.8
The Turks
In 608, again, under the influence of Pei Ju, Emperor Yang sent an envoy to Chuluo
Khagan (619–620/21). Chuluo was convinced both to join the Chinese army to
fight against the Tuyuhun as well as to come to court to pay tribute; but in the end
he did not come. In 611, Shegui of the Western Turks sent an envoy to the Sui court
to propose a marriage alliance. Pei Ju accepted the request, and then Shegui at-
tacked Chuluo. Towards the end of that year, Chuluo Khagan came to the court to
pay his respects to Emperor Yang. Following the Koguryo war in 614, Emperor
Yang began to worry about the growing power of the Turkic Empire. Pei Ju enact-
ed a strategy that envisaged confusion and disorder among the subjects of Shibi
Khagan (609–619) of the Eastern Turks, but that wasn’t a suitable policy for China
6 SUI SHU 1997, 1844–5, 1879–80.

7 SUI SHU 1997, 1504, 1845.
8 SUI SHU 1997, 1533–4, 1580–1, 1841; XIONG 2006, 205–6.
itself. In order to cope with Koguryo, Turkic support was needed. Pei Ju therefore
proposed that the Sui palace offer a Chinese princess to Shibi’s younger brother,
giving him the title of South Khagan. He was hesitant to accept as he was not
strong enough to oppose his elder brother. Shibi was annoyed when he found out
about China’s actual plans. Before long, Pei Ju proposed that Shibi’s Sogdian assis-
tant Shi Shuxi be murdered, and thus summoned both him and his associates on
the pretext of trading, and had them ambushed by the Chinese. Consequently,
Shibi broke his ties with China and decided to retaliate. In 615, the Khagan
planned to attack the emperor whilst travelling north. The Turks besieged Emperor
Yang in Yanmen, and captured 39 of Yanmen’s 41 cities, until eventually Shibi
withdrew because of false information provided by his Chinese wife, Yicheng,
about an attack that had taken place at the northern border. 9
Koguryo
Emperor Yang claimed victory along the western and north-western borders, and
then headed for Koguryo. Pei Ju stated that Koguryo had been a part of China
from its very onset right into the Han and Jin eras. He told the emperor that he
should summon the king of Koguryo to the palace, and to attack were the king not
to obey. The emperor accepted this proposal and told the King of Koguryo that if
he did not come to the Chinese palace, the Turks would be sent to Koguryo. King
Yongyang did not obey; he never sent a tribute after 609, thus instigating China to
attack Koguryo in 614. An unsuccessful campaign against Koguryo led to China
breaking out in turmoil as the Sui collapsed. 10
Having perfected traditional Chinese policies, Pei Ju truly stood out, but what
did he manage to accomplish? Let us take a look:
1) Tributary and trading system: Pei Ju planned and re-established the tribu-
tary and trading system of the Han era. Records in the Han Shu, the official history
of Han Dynasty written in 111, reveal the economic collapse of the Han Empire
because of this system:
There came the further expenses of presents sent as gifts or to accompany
escorts; of the courtesies exchanged at a distance of ten thousand li; and of
the armed forces, too high for calculation… the strength of the people was
spent, and resources were exhausted… robbers and thieves rose up every-
where and the roads were impassable. For these reasons, in his latter days
Emperor Wu abandoned the lands of Luntai and proclaimed a decree ex-
pressing anguish and sorrow.11
9 SUI SHU 1997, 89, 1582, 1876–9.

10 SUI SHU 1997, 89, 1581–2, 1875.
11 HAN SHU 1997, 3928–9; HULSEWÉ – LOEWE 1979, 201–2.
According to this practice, delegations from Xiyu cities went to the capital and
offered to pay tribute to the emperor in order to trade. Not only did they bring him
and the upper echelon of society exotic goods, but they also elevated the Sui’s po-
litical influence. While the Chinese already had influence in Xiyu, trade under the
tribute system proved not to be particularly profitable. 12 Pei Ju’s tribute and trade
system soon caused the collapse of the Sui economy. In fact, the Sui’s collapse was
related to Pei Ju’s politics with the Xiyu.
2) Pei Ju feeds the ambitions of Emperor Yang: Emperor Yang wanted to cre-
ate a world empire on the Han model. It was not only Pei Ju who encouraged it.
Although there were ministers and advisers who opposed the expansionist policy
of Yang their voice was not heard. Pei Ju’s role in foreign policy was so important
that Sima Guang blamed him for directing Emperor Yang. According to Sima
Guang, the emperor felt that he was imitating Qin’s first emperor and Han emper-
or Wu. It was as if China was weakened and ruined by Pei Ju’s fantasies. Neverthe-
less, many researches claim that it is not fair to blame Pei Ju for the catastrophes at
the border as he knew the emperor’s inclinations and told him what he wanted to
hear.13
3) “Right Arm”: The region of Xiyu, that is Gansu and Xinjiang in its narrow
sense, was referred to by Han sources as the “right arm of the Xiongnu”. In order
to defeat the Xiongnu, the Xiyu had to be seized. In his quest to restore the do-
mains of the Han Empire, Pei Ju put Xiyu at the centre of his struggle against the
Turkic Empire. This strategy essentially took the same form as the one which the
Han Empire used in order to fight against Xiongnu, as reported by Zhang Qian.
Chanyu himself had stated the importance of Xiyu for the Xiongnu.
The lands of Cheshi are fertile and fine, and lie close to the Xiongnu. If the
Han acquire them and accumulate stocks of corn from a large number of
harvests, the Xiongnu people and state would, without doubt, suffer loss.
These lands must not be left uncontested.14
In a policy aimed at destroying the power of the Xiongnu, the Han Shu clearly
records that Xiyu was placed at the centre of the strategy: “the Yumen was opened
so as to communicate with the Xiyu, and in order to sever the right arm of the
Xiongnu”.15
Pei Ju, naturally, regarded Xiyu as the “right hand of the Turks”. His book Xiyu
Tuji gave detailed information about the political situation, rulers, geographical
conditions, traditions, food, and clothing—in other words the culture—of Xiyu. Pei
Ju also studied maps showing roads leading westward. In the introduction of his
12 PAN 1997, 119.

13 PAN 1997, 117, 124.
14 HAN SHU 1997, 3923; HULSEWÉ – LOEWE 1979, 187–8.
book (only the preface of the book has survived), he wrote that the Han Empire
had opened the road to the Xiyu and expanded its control over the region. Accord-
ing to him, China’s power had now reached the Xiyu, but the Western Turks and
the Tuyuhuns ruled over most of the peoples in the region, preventing them from
coming to China to pay tribute. He advocated that the Sui Emperor would lure all
peoples, gain their support, and thus defeat the Turks and the Tuyuhun. 16
4) “Prevent Unification!”: According to Pei Ju’s strategy, it was very important
to prevent the Turks from collaborating with the Tuyuhun of Southern Xiyu. On
the other hand, one of the goals of China’s Xiongnu policy during the Han period
was to prevent the Xiongnu from joining the Qiang and Yuezhi peoples of the
Southern Xiyu. This, too, is clearly recorded in Han Shu:
In the age of Emperor Xiao Wu, policy was directed towards controlling the
Xiongnu, in the realization of the danger that they might form a union with
the western states and alliance with the southern Qiang. The Chinese there-
upon demarcated the area west of Yellow River; a line of four commanderies
was established… in order to sever the right arm of the Xiongnu and to sep-
arate them from the southern Qiang and Yuezhi.17
The Sui Empire followed the same policy against the Turkic Empire, and ulti-
mately achieved the same results. Pei Ju’s divisive tactics required that the Turkic
peoples not merge with the Tuyuhun.
5) Aggressive expansionism: Pei Ju advocated an aggressive expansionist poli-
cy and was deeply influenced by the Confucian political ambition to restore the
Han Empire. He provided the intellectual foundation for Emperor Yang’s expansion
policy.18 The exacerbation of his aggressive expansionism perhaps marked the end
of the empire. Although Pei Ju needed the support of the Turks during the Korean-
Chinese War of 614, he continued his aggressive and divisive policy towards them
by deciding to kill Shibi’s Sogdian assistant. This perhaps drew the Turkic Em-
pire’s attention towards China, which ultimately marked the end of the Sui.
6) The northern route passing through Tujue Khagan’s yurt: According to Xiyu
Tuji’s preface in Pei Ju’s biography, there are three routes from Dunhuang to Xihai
(the Western Sea), including the northern route, starting from Yiwu and passing
through Puleihai Tiele 蒲類海鐵勒 (Barköl Töles), the Tujue Khagan’s yurt (centre), and
“rivers flowing north” arriving in Fulin and reaching the “Western Sea”.19
This route is not mentioned often in history records or itineraries. There are two
possible reconstructions:
16 PAN 1997, 117, 209.

18 PAN 1997, 117, 209.
19 SUI SHU 1997, 1578.
(i.) The “rivers flowing north” may be rivers such as the Chu and Seyhun
(Sir Derya). The road must pass north of the Aral Sea and the Caspian Sea.
The Turkic Khagan was Western in origin, whose “yurt” or centre could in
fact be the city of Tokmak (Suyab). Xuan Zang had visited this city.
(ii.) The “rivers flowing north” may refer to the Ili and Chu rivers. Accord-
ingly, if one sets off from Qumul, passes through the territories of Barköl
and Töles, one then arrives at the Western Turkic Khagan’s “yurt” in the
north of Kucha and the Mount “Sanmi” (according to Chinese sources). Af-
ter that, if one crosses the Ili and Chu rivers, passes north of the Aral and the
Caspian seas, one would ultimately reach the Black Sea.
One rare record about “passing Turkic Khagan’s yurt” is to be found in the
travelogues of Wu Gong (8th century). He sets off from Chang’an and travels to the
Uighur Khagan’s yurt in the north because the Dunhuang region was under Tibet-
an occupation, whereupon he follows the northern fringes of the Taklamakan De-
sert and the southern foothills of Tianshan, until he ultimately reaches Kucha. 20
7) “Do not tease the Turks!”: The Ancient Chinese led sedentary live. Their
steppe Turkic counterparts, on the other hand, were nomads. Despite this, we
know that, even in the earliest historical records, the Chinese had been able to keep
the mighty Turkic armies away from their borders through calm, silent diplomacy,
and that they had even managed to divide them. Turkic peoples, in contrast, were
known to be short tempered, while the Chinese maintained their tranquility, but
were also sly and two-faced. When Turkic peoples were shown respect, they quick-
ly calmed down. However, when they realized that they had been deceived, they
were quick to draw their swords on a moment’s notice. Pei Ju understood the
Turks very well, and thus attempted to trick them in order to divide them. Yet, he
could not calculate the degree of their anger after they learned that they were
tricked. Although he successfully managed to deceive and thus put the Turkic
peoples against one another, they retaliated, thus rendering the Chinese unable to
fully understand their character for centuries to come.
Bibliography
HAN SHU (1997): Han Shu, Beijing.
HULSEWÉ, A. F. P. – LOEWE, M. A. N. (1979): China in Central Asia, Leiden.
PAN, Y. (1997): Son of Heaven and Heavenly Qaghan: Sui-Tang China and its Neighbors,
Washington.
SUI SHU (1997): Sui Shu, Beijing.
XIONG, V. C. (2006): Emperor Yang of the Sui Dynasty: His Life, Times, and Legacy,
Albany.
YILDIRIM, K. (2015): Wu Gong Seyahatnâmesi. Türk Dünyası Araştırmaları 217, 21–39.
20 YILDIRIM 2015, 21–39.

Art in Mongolia Between the 6th and 8th
Centuries: Cultural Diversity in the Arts
of the Turks
AH RIM PARK
Abstract
The 6th to 8th-century steppe arts in Mongolia were based on the artistic trends
running through the active relations between the nomads, such as the Turks, and
cultural agents like the Sogdians, who covered the steppe road and the Silkroad
from China, Mongolia, East Turkestan, and West Turkestan. Two recently discov-
ered tombs, the Bayannuur Tomb and the Pugu Tomb, as well as the Shiveet Ulaan
clearly exhibit the tastes of nomadic people residing along the Silk Road. The buri-
al objects and practices shown in these tombs can shed a light on the cultural back-
ground of the buried as well as the cultural interaction along the Steppe Route. The
Shiveet Ulaan ritual complex displays extraordinary artistic cultural exchanges and
is the beginning of the combination between the Turkic and the Chinese style in
the building of ritual complexes.
Keyword: Turks, Sogdians, Bilge Khagan, Kul Tegin, Shiveet Ulaan, Bayannuur
Tomb, Pugu Yitu Tomb
1. Introduction
The arts of the Turks in the period from the 6 th to the 8th centuries in Mongolia
were not well known until recently. Originally, Turkic art and archaeology are
known for horse burials, rock carvings of armed warriors mounted on a horse,
ritual complexes, and stone sculptures.
The situation has changed thanks to new discoveries from central Mongolia in
the provinces of Töv Arkhangai, and Bulgan. The continuing excavations of the
memorial complex of Bilge Khagan, located at Khöshöö Tsaidam, Khashaat Dis-
trict, Arkhangai Province, and the ritual complex at Shiveet Ulaan, located at Ar

Professor, Sookmyung Women’s University, Seoul, South Korea. This article is a revised ver-
sion of a paper delivered at the 8th International Conference on Medieval History of the Eura-
sian Steppe Nomads and their Neighbors in the Middle Ages, held in Sofia, November 20–3, 2019.
44 AH RIM PARK
Khondii, Orkhon District, Bulgan Province, have shown the unique significance of
the Turkic ritual complexes. The Shiveet Ulaan site, especially, has been excavated
for five years and has yielded many interesting features revealing the unique na-
ture of the Turkic ritual complex.
Furthermore, recent excavations made in the Bayannuur Tomb and the Pugu
Yitu Tomb in central Mongolia have provided extraordinary perspectives on the
nomads on the Steppe Route. The murals, the tomb epitaph, as well as the burial
objects, such as Byzantine coins and textiles, and the burial customs of the Bayan-
nuur tomb and Pugu Yitu Tomb are an exceptional testimony to the complicated
cultural mixtures in 7th-century Mongolia.
Those findings from Mongolia are intricately intertwined with contemporary
finds in East and West Turkestan as well as in China. The Turks interacted with the
Sogdians, who acted as intermediate agents between the East and the West. The
Afrasiab murals in Samarkand (modern Uzbekistan) have been dated to the mid-
7th century, and according to one of the interpretations the major figures in the
western wall represent the portrait of the Samarkand King Varhuman and possibly
a depiction of the Western Turkic Khagan receiving foreign envoys, and having a
hunting and a procession ceremony. On the other hand, in China, the descendants
of the immigrant Sogdians have left unique funerary monuments embodying both
the Sogdian and the Chinese secular and funerary art traditions.
By considering all related materials that can enhance our understanding of the
artistic diversity of the Turks and the artistic contacts between the Turks and the
Sogdians, the current research aims to analyze the acculturation of the Turks and
the Sogdians. This study discusses the three major monuments approximately dat-
ed to the second half of the 7th century in Mongolia—the Bayannuur tomb, Pugu
Yitu Tomb, and the Shiveet Ulaan ritual complex—in order to understand the na-
ture of the acculturation along the Steppe Route.
2. Funerary art of Mongolia in the 6th–8th centuries: The tomb murals at Ulaan
Kherem, Bayannuur District, Bulgan Province
Two recently excavated tombs in central Mongolia, the Bayannuur Tomb and the
Pugu Yitu Tomb, present significant evidence of the intermingling of different
cultures in the second half of the 7th century in central Mongolia. They were built
during period of Tang domination in the steppe (630–87), between the First Turkic
Empire (552–630) and the Second Turkic Empire (687–745).1
From July to September 2011, the joint research team from Mongolia and Ka-
zakhstan excavated a tomb with murals at Ulaan Kherem, Bayannuur District,
Bulgan Province, Mongolia. The tomb is the first excavated mural tomb in Mongo-
lia to date (fig.1). It is dated approximately to the second half of the 7 th century.
1STEINHARDT – ERDENEBOLD – PARK 2016; PARK 2019; STEINHARDT – ERDENEBOLD – PARK 2018;
STEINHARDT – ERDENEBOLD – PARK 2017; STEINHARDT – ERDENEBOLD – PARK 2014; PARK 2020.
ART IN MONGOLIA BETWEEN THE 6TH AND 8TH CENTURIES… 45
The structure and mural subjects of the tomb as well as the style of the figure
paintings resemble those of the mural tombs in the subsidiary burials of Zhaoling
near the tomb of the Tang emperor Taizong (r. 626–49), most of which were con-
structed in the second half of the 7th century. They also follow the funerary art tra-
dition in terms of the placement of paintings of the Azure Dragon and the White
Tiger, two of the four guardian gods of cardinal points, as well as the pavilion ar-
chitecture, the monstrous mask, and the lotus flower. This tradition was estab-
lished by the nomadic people known as the Xianbei in the late Northern and
Southern Dynasties period during the 6th century, and continued under the Tang
dynasty in the 7th century.
Another recently excavated tomb in the Zaamar District, Töv Province, yielded
a stone epitaph dated to 678. The tomb’s occupant belonged to the Pugu tribe of
the Tiele. The geographical closeness and the similarities in the tomb structure and
burial objects found in this tomb and the Bayannuur Tomb suggests that the later
can be dated to the second half of the 7th century as well, and more precisely to
around the 670s. The stylistic analysis of the murals and burial objects from the
Bayannuur Tomb also suggests that the tomb can be dated to the same period, i.e.
during the period of Tang occupation after the collapse of the First Turkic Empire.
Additionally, according to the scientific analysis of the tomb figurines and the lime
plaster of the murals the Bayannuur Tomb can be dated to AD 670 ± 70.
The figurines from the two tombs resemble those found in the early Tang tombs
located in the Chinese city of Xi’an, Shaanxi Province, which provides strong evi-
dence of Tang Chinese influence on the region. However, in terms of materials,
techniques, and pigments, there are certainly different features from those in the
Central Plains region of China. Actually, they are more like those from the Astana
tomb complex in Turfan in the Chinese region of Xinjiang, which appear to have a
close relation with sculptures and murals in Central Asia. Thus, the tomb figurines
from the Bayannuur Tomb and Pugu Yitu Tomb should be considered not only in
the context of Tang Chinese tombs but also in comparison with sculptures and
paintings from Central Asia.
Female wooden figurines from the Pugu Tomb wear a rectangular shaped
fragment of a textile with a pearl roundel pattern. Several fragments of such a tex-
tile from the Pugu Yitu Tomb may provide more evidence of the Sogdian connec-
tion found in Mongolia. The presence of a number of those textiles in the arms of
the wooden figurines displays a unique acculturation of Central Asian materials
for funerary purposes.
It is possible that the Pugu ruler governing the Bayannuur and Zaamar regions
invited a Chinese artisan to construct his tomb, but the tomb occupants definitely
embody a certain nomadic taste and culture exhibited by the unique burial practice
and the rich gold and silver ornaments, a culture that was a result of the broad
interactions with various peoples ranging from the territories of Byzantium to China.
The burial objects from the Bayannuur Tomb also clearly exhibit the tastes of
46 AH RIM PARK
nomadic people residing along the Silk Road and the Stepp Route. The tomb figu-
rines and Byzantine golden coin imitations discovered both there and at the Pugu
Yitu Tomb indicate that there were active exchanges with China and the Byzantine
Empire.
The tomb occupant of the Bayannuur Tomb was cremated, and the ashes were
put inside a small wooden box that was covered with a textile with golden leaf
decoration and buried together with silk bags. They contained a Turkic-style gold-
en cup, 40 Byzantine coin imitations, and other gold ornaments. The rectangular
wooden box was placed inside a wooden coffin with a wide top and narrow bot-
tom. Additionally, in the tomb chamber one can observe a silk screen painting with
a wooden frame. Male and female figures surrounded by a landscape are recog-
nizable on the fragments of a silk painting found inside. This type of a silk screen
painting with a human figure under a tree is a well-known motif popular during
the Tang dynasty of China.
While the wall paintings of the Bayannuur Tomb are a clear sign of the adop-
tion of the Chinese funerary mural style, the burial custom—namely, the interment
of a cremated body inside a wooden coffin covered with the golden leaf textile
along with a silk bag with 40 Byzantine coin imitations and a Turkic-style golden
vessel and ornaments—is certainly different from that of the contemporary Chi-
nese tombs.
The 40 gold coins from the Bayannuur Tomb seem quite unusual compared to
examples found in China in view of their number, techniques, and general burial
practices. Only one or two, or at most four or five Byzantine coins, are usually bur-
ied in one tomb in China. The tombs in China where Byzantine coins were discov-
ered, such as those of the Sogdian descendants in the southern suburb of Guyuan,
Ningxia, contained burial objects related to Central Asia as well.
A Byzantine and a Sasanian coin found in a tomb might speak not only of cul-
tural exchanges between regions where the same kind of Byzantine coins have
been discovered, but also of the background of the tomb’s occupant, the process of
regional transmission of such Central Asian objects, and the possible location of a
workshop. For instance, as most of the coins from the Bayannuur Tomb are repli-
cas based on Byzantine prototypes, they are assumed to have been made not in
Byzantium but in Sogdiana. Among them were imitations of Tiberius Ⅱ Constan-
tine (r. 578–82), Phocas (r. 602–10), and Heraclius (r. 610–41).
Currently, it is not easy to precisely determine where those coins were made. A.
Naymark suggests that some of the coins are not found in Sogdiana (fig. 2). 2 They
depict a human head and a bust, a profile found on neither Byzantine nor Sogdian
coins. The motifs in this sort of coins can be found in the vessels of a Central Asian
type, like those found in the southern suburb burials in Datong, Shanxi, China,
during the Northern Wei dynasty (386–534). Ultimately, both a Byzantine coin and
2 NAYMARK 2019.
a Central Asian vessel, which is often interpreted as a representation of Sasanian or

Sogdian style, could have been used as a prototype. The fact that the manufacturer
decided to use such a unique model is very peculiar: golden coin imitations with a
motif of a human figure found on Sasanian or Sogdian vessels. The fact that there
was a manufacturer who knew the contexts both of a Byzantine golden coin and a
Sasanian- or Sogdian-style vessel might shed some light on the artistic mindset of
the nomads in Eurasia.
Parallels to the Turkic-style vessel found in the Bayannuur tomb were discov-
ered in China, Mongolia, Xinjiang (East Turkestan), and West Turkestan, either as
actual vessels or depicted in a mural or sculptured as a part of a stone statuary.
Among the numerous stone sculptures of Eurasia, a Turkic-style statue holding a
cup is a sign of a sculpture belonging to the Turkic period. It is often seen in the
banquet depictions in Central Asia and in the Chinese funerary arts related to the
Sogdians.
As discussed above, the burial objects and the burial practice shown in the
Bayannuur Tomb and the Pugu Yitu Tomb can deepen the understanding both of
the cultural background of the occupants of two major tombs in Mongolia as well
of the cultural interaction along the Steppe Route.
3. The ritual complexes and stone sculptures of the 6 th–8th centuries in Mongolia
The ritual complexes and stone sculptures of the 6th–8th centuries in Mongolia are
important monuments of the Turkic period that signify the acculturation in ritual
architecture and art.3
In Mongolia, more than 730 Turkic stone statues in 114 districts of 21 provinces
have been registered. Typically, Turkic funerary monuments consist of one or two
stone statues, but occasionally sacred offering sites of eminent members of the elite
are marked by more than a dozen. Some well-known Turkic ritual complexes are
those of Kul Tegin (732), Bilge Khagan (735), and Tonyukuk (720). Recently exca-
vated ritual complexes include those at Shiveet Ulaan, Dongoin Shiree, Khaya
Khudag, Gunburd, Nomgon, etc. The Khaya Khudag and the Gunburd sites were
excavated by the National Museum of Mongolia in 2018 and 2019. Two stone
sculptures of males in the style of balbal, ritual offerings of 14 horses and two
sheep, and small potteries in a canal have been excavated from the Khaya Khudag
site. At the Gunburd site, several ritual complexes have been discovered. One of
them yielded a large stone sculpture of a seated figure and three stone coffin plates
3This part of the article has been prepared on the basis of two visits to Mongolia in 2018 and
2019. The trip was made possible with the generous help of Professor Erdenebold of the
Mongolian University of Science and Technology. It was supported by the 3-year grant of
the Korea Research Foundation entitled Art of the Mongolian Steppe in the 6th–8th Centuries. For
the ritual complexes and stone statues of the 6th–8th centuries, see: BAYAR 1997; EREGZEN
2016; SAMASHEV – TSEVENDORZH – ONGGARULY – CHOTBAYEV 2016; STEINHARDT – ERDENEBOLD
– PARK 2018; PARK 2019; PARK 2020.
48 AH RIM PARK
carved with a floral motif. The seated sculpture appears to be similar to those
found in the Bilge Khagan complex.
An international joint survey by Professor Takashi Osawa from the Graduate
School of Languages and Cultures, Osaka University, and the Archaeological Re-
search Institute of the National Academy of Sciences of Mongolia made an im-
portant discovery at Dongoin Shiree in the Tüvshinshiree District of the Sükhbaa-
tar Province in 2015–16. About 14 stone steles have been found at the site. The sizes
and the numbers of Dongoin Shiree stone steles are impressive. The highest stele is
6.4 m, the shortest one is 3.32 m, and their average height is 4.43 m. Some inscrip-
tions and 30 tamga signs have been found on the statues. This was the first discov-
ery of Turkic inscriptions in eastern Mongolia. The Dongoin Shiree site located in
the southeast of Mongolia is a unique example that includes only huge steles in-
stalled in two rectangular rows inside the complex. Those in the front of the stone
coffin have more tamga signs. The shape of the steles resembles the stone statuary
at the Tonyukuk ritual site with a rather circular contour. It is certainly different
from the Chinese-style stele found at the Bilge Kagan site.
Examples of Turkic stone sculptures that show early features in the evolution of
the Turkic ritual complex can be the Xiaohongnahai sculpture in Xinjiang, China,
the Choiren sculpture in the Mongolian province of Dorno Gobi, and sculptures
from the ritual complex at Ongot in the Altanbulag District, Töv Province, as well
as the ritual complex at Shiveet Ulaan in the Bayan-Agt District, Bulgan Province.
Each of these could represent an early type of sculpture for the respective region.
The Xiaohongnahai sculpture clearly indicates that the Turks knew the icono-
graphy and human representation of Sogdian art. The Xiaohongnahai stone statue
in western Xinjiang is suggested to have been built for Niri Kagan 泥利可汗 (r. 587–
99) of the Western Turkic Empire in 599–604. On its lower body there is a Sogdian
inscription of 20 lines. The shape of the crown is probably derived from the Sogdi-
an one, which can be ultimately traced back to an Iranian-style royal headwear.
Both the hand gesture and the seated posture with a cup or ewer in the hand of the
Turkic stone statue are assumed to have been largely borrowed from those of Sog-
dian-style nobility banqueters.4
On the other hand, the rather primitive representation of the face and the body
of the Choiren stone sculpture resembles traditional steppe stone sculptures such
as the balbal. It has a round carved face without too many details marking the hair
and the costume.
The 28 balbal sculptures with a primitive carving of a head on rectangular or cir-
cular stone at Ongot have been considered as an early example of Turkic funerary
art, similarly to the ritual complex built by the Rouran or the Xueyantuo in Al-
tanbulag, Töv. The Ongot statuary is a sign of the transformation from deer stone or
balbals to a style of human stone sculpture with a round shape in central Mongolia.
4 HAYASHI 2006.
It was at Shiveet Ulaan that the Turkic-style funerary ritual complex was estab-
lished. The Shiveet Ulaan ritual complex has been excavated for five years by a
joint excavation team from Mongolia and Kazakhstan (fig. 3). Contrary to the usual
location of a Turkic ritual complex on a plain, it is located on the top of a high hill.
According to the 2015 excavation, the ritual building in the center of the platform
of piled stones has an octagonal shape (107×45 m) and was made of red sandstone.
To the east of the ritual building are placed about 11 stone human figures, 8
stone lions (some have a tamga sign on the leg), and four stone sheep. To the east of
the stone sculptures is a square stone stele base (height 1.54 m, width 1.04–0.83 m,
thickness 0.55 m). The stele itself (height 2.24 m, width 0.82 m, thickness 0.24 m) is
now placed in front of the foundation of the ritual building along with stone sculp-
tures and its surface is carved with 60 different tamga signs. No Sogdian or Chinese
inscriptions have been found.
There are some debates regarding the person to whom the Shiveet Ulaan ritual
complex was dedicated. The laboratory analysis with radiocarbon dating of the
wooden part of the log cabin from sector D8 of Shiveet Ulaan was made by the
Japanese Institute of Accelerator Analysis in Kawasaki, Kanagawa Prefecture, Ja-
pan, on 21 July 2017. According to the analysis of the Institute, the date of the
Shiveet Ulaan ritual complex corresponds to the years 665–71 or 685–91. It is ap-
proximately during the reign of Elteris Kutlu Khagan (r. 682–91), who established
the Second Turkic Empire.
Unfortunately, no seated human sculpture, which is usually presumed to depict
the occupant of the shrine, is left at the site. All human figures there are standing,
holding cups and some other tools. A similar ritual complex consisting of a mound
of piled stones located on top of a hill is the one at Shiveet Tolgoi in the Khashaat
District, Arkhangai Province, located about 70 km away from Kharhorin and 50
km from Khöshöö Tsaidam. The head and a decorative pattern on the body of the
male stone sculpture with a seated posture at Shiveet Tolgoi (height 160 cm) are
preserved (fig. 4). Unfortunately, its whereabouts are currently unknown. The
vivid floral patterns on the body of the seated sculpture resemble those on the
seated sculpture (120×75×33 cm) found at the Tarnin Gol in the Bürd District,
Övörkhangai Province, behind which is also a large mound of piled stones.
The difference between the Turkic stone statuary of the 7th century and the 8th
century can be understood by comparing the Shiveet Ulaan complex with the Bilge
Khagan complex. The Bilge Khagan and Kul Tegin ritual buildings show the remains
of Chinese-style roof tiles which could have been part of a building with a rectangu-
lar plan imitating the Chinese wooden architecture, as indicated by the reconstruc-
tion of the complex. On the other hand, the octagonal-shaped ritual building at
Shiveet Ulaan is certainly different from the Chinese prototype and might have
stemmed from the steppe tradition, which can be observed in the remains of the
octagonal- or circular-shaped Turkic ritual building at Bozok in Kazakhstan, at
50 AH RIM PARK
Saryg-Bulun in Tuva, as well as at Voznesenka on the Dnieper River in Ukraine.5

In the Bilge Khagan ritual complex, the stone statues of the king and the queen
were placed inside the ritual building. Near the complex, more than 10 other Tur-
kic ritual complexes are distributed and some of them await to be excavated. Ac-
cording to the excavator, several balbals with inscriptions have been discovered
inside the Bilge Khagan complex and it is not known why the balbals were buried
inside the moat. The placement of the balbals at equal distance from each other
inside the moat might have been intended from the beginning or it could have
been a deliberate later interment. If the overall construction of the Turkic ritual
complex of Bilge Khagan shows the adoption of a Chinese-style funerary system,
the interment of balbals inside the complex might show the deliberate manifesta-
tion of the traditional steppe funerary practices, which included deer stones and
steppe stone sculptures. If the balbals found inside the moat of a ritual complex are
of the type developed from the deer stone tradition of the steppe, it is natural that
they would have a primitive and flat representation of a human being, a clear sign
that it is derived from the flatness of the deer stone.
4. Conclusion
Assumed to have been built for Elteris Kutlu Khagan (r. 682–92), who established
the Second Turkic Empire, the Shiveet Ulaan ritual complex and the stone statuary
built at the same time display features typical both for the steppe and for China.
Some features of the Shiveet Ulaan that are not seen in the ritual complexes of To-
nyukuk, Bilge Khagan, and Kul Tegin might be a sign of the original Turkic identi-
ty, which would later be discarded with the adoption of the Chinese style.
The Bayannuur Tomb and the Pugu Tomb, also dated to the second half of the
7th century, give an important context for the study of the Shiveet Ulaan site. While
the Shiveet Ulaan site was built at the beginning of the Second Turkic Empire by
the Ashina family, the two tombs were built during the period of Tang domination
probably by the Pugu family. They indicate that the Ashina family and the Pugu
family had different burial customs and a different manner of choosing how to
accept foreign cultural elements. These three monuments from the second half of
the 7th century reveal the ways in which the Turks and the Tiele (later the Uyghurs)
accepted the Chinese funerary ritual system.
In Mongolia, the Turkic stone statuary and ritual complexes are important
monuments for this period that signify the close relations between the Turks and
the Sogdians. Stone sculptures of the 7th and 8th centuries in Mongolia reveal the
transmission to the steppes of the iconography of a banquet scene and a hand ges-
ture, typical of the Sogdian arts found in Sogdiana and China.
According to the Jiu Tang Shu and the Xin Tang Shu, the ritual complexes of
Bilge Khagan and Kul Tegin were built by Chinese artisans sent by the Tang
5 KHABDULINA 2010.
Xuanzong, which suggests that the Turkic ritual complexes with stone sculptures
must have followed the Chinese funerary ritual system. However, the number and
the composition of the Turkic sculptures of a human figure and an animal did not
exactly follow the Chinese prototype. The inclusion of a seated figure and the rep-
resentation of the most prominent figure in a cross-legged postition and holding a
banquet cup have never been attested in sculptures of the royal and noble tombs of
Tang China. The ritual complex at Shiveet Ulaan and some important sculptures of
the Turkic period, including those at the Shiveet Tolgoi and the Xiaohongnahai,
reveal that the sculptures of the Turks must have followed the iconography of the
Sogdians (or an earlier one, of the Kushans and the Hephthalites). Thus, the Turkic
ritual complex and the stone sculptures of that period exhibit both more traditional
characteristics of Eurasian steppe stone sculptures and the deep influence of Sog-
dian iconography.
It was in the second half of the 6th century in China that extraordinary Sogdian
funerary monuments were produced in several major cities like Xi’an and Taiyuan.
The funerary couch from the tomb of An Jia in Xi’an, the stone sarcophagus of the
tomb of Shi Jun in Xi’an, and the tomb of Yu Hong in Taiyuan show that the buried
Sogdians had various interactions with the Turks. They even preceded the appear-
ance of major Turkic stone statuaries in Mongolia and the famous Afrasiab murals
in Uzbekistan. The Turks and the Sogdians depicted themselves in Chinese-style
funerary couches and stone sarcophagi on which they manipulated their images
within Chinese funerary iconography.
Even before the appearance of these Sogdian funerary relief sculptures, the
presence of the Sogdians in China has been well attested by the Sogdian letters and
tombs with frescos in Loulan, Xinjiang and Gaotai, Gansu during the Wei-Jin peri-
od (220–420). Metal wares of Central Asian style found along the Silk Road also
attest to the wide presence of the Sogdians in China.
Some models of iconography not derived from Chinese precedents are certainly
inspired by the Sogdians or derived originally from the arts of the Sasanian Persian
relief sculptures and silver plates. The Sogdian murals in Samarkand and Va-
rakhsha in Uzbekistan and Penjikent in Tajikistan are dated later than the related
Sogdian funerary arts found in China. The similarities among these examples of
the so-called Sogdian iconography were probably based on the painting models
distributed ever since the Wei-Jin and the Northern and Southern Dynasties peri-
ods (3rd – 6th centuries). These models probably stemmed not only from murals but
also from Sogdian or Central Asian metalware.
The routes and mediums of transmission must have been varied. Both the Silk
Road and the Steppe Route must have played a role in the artistic transmission.
The unique profile bust of a male figure from the golden coins found in the Bayan-
nuur tomb is a good example of the depth of the interconnections between the
Sogdians and the Turks.
In conclusion, the steppe arts in 6th to 8th-century Mongolia, including the
52 AH RIM PARK
Bayannuur Tomb, the Pugu Tomb, the ritual complexes, and the stone statuaries of
the Turks, followed the artistic trends running through the active contacts between
nomads such as the Turks and cultural agents like the Sogdians, who traversed the
Steppe Route and the Silk Road in China, Mongolia, and East and West Turkestan.
Bibliography
BAYAR, D. (1997): The Turkic Stone Statues of Central Mongolia, Ulaan Baatar.
EREGZEN, G. (2016): Ancient Funeral Monuments of Mongolia. Archaeological Relics
of Mongolia, vol. 3.
HAYASHI, T. (2006): Sogdian Influences Seen on Turkic Stone Statues Focusing on
the Fingers Representations, in: Compareti, M. – Raffetta, P. – Scarcia, G. (ed.:)
Ērān ud Anērān: Studies Presented to Boris Il’ič Maršak on the Occasion of His 70th
Birthday. Venezia, 2006, 245–254.
KHABDULINA, M. (2010): Cult Center of ancient Turks on the territory of Central
Kazakhstan, Bulletin of IICAS (the International Institute for Central Asian Studies
(Samarkand)) 12, 33–49.
NAYMARK, A. (2019): Gold Imitations of Western Coins and Bracteates from Mon-
golia and Their Connection to Finds in Western Central Asia, Unpublished pa-
per from the conference Middle-Period Mongolian Archaeology (2019.10.26–
2019.10.27), University of Pennsylvania.
PARK, Ah-Rim (2019): Stone Statuaries of the Turks in the 7th and 8th centuries – on
the Shiveet Ulaan Ritual Complex. Central Asian Studies, 24/2, 33–58.
PARK, Ah-Rim – ERDENEBOLD, L. (2020): Art of Mongolia in the Context of the Eurasian
Steppe Art, Hakyon munhwasa.
Samashev, Z. – Tsevendorzh, D.– Onggaruly, A. – Chotbayev, A. (2016): Shiveet
Ulaan Ancient Turkic Cult and Memorial Complex, Astana.
STEINHARDT, N. S. – ERDENEBOLD, L. – PARK, Ah-Rim (2014): The Wall Painting
Tomb at Ulaan Kherem, Bayannuur Sum, Bulgan Province, in Mongolia. Cen-
tral Asian Studies 19, 1–25.
STEINHARDT, N. S. – ERDENEBOLD, L. – PARK, Ah-Rim (2016): A Tomb in Bayan-
nuur, Northern Mongolia. Orientations 47/8.
STEINHARDT, N. S. – ERDENEBOLD, L. – PARK, Ah-Rim (2017): Tomb Figurines and
Byzantine Golden Coins from the Bayannuur Mural Tomb and Pugu Yitu
Tomb in Mongolia. Central Asian Studies 22, 73–99.
STEINHARDT, N. S. –Erdenebold, L. – PARK, Ah-Rim (2018): The Ritual Complexes
and Stone Statues of the 6th to the 8th Centuries in Mongolian Steppe. Central
Asian Studies 23/2, 143–68.
STEINHARDT, N. S. – ERDENEBOLD, L. – PARK, Ah-Rim (2020): Shoroon Bumbagar,
Artibus Asiae, Vol. LXXX, No. 2, 167–210.
Figures
Figure 1. A horse and a groom on the western wall of the first chamber, Bayannuur
tomb
Figure 2. A Byzantine coin, Bayannuur tomb

54 AH RIM PARK
Figure 3. Shiveet Ulaan ritual complex
Figure 4. Shiveet Tolgoi sculpture

The Tree Cult and the “Piece of Wood the
Size of a Phallus”
EDINA DALLOS
Abstract
In the 10th century, during his journey to the Volga River, Aḥmad ibn Faḍlān noted
about the Bashkirs: “Each of them sculpts a piece of wood the size of a phallus and
hangs it on himself. If he is about to undertake a trip or to meet an enemy, he kiss-
es it and prostrates himself before it, saying: ‘O my Lord, do unto me such and
such.’” This statement by Ibn Faḍlān is undoubtedly based on some misunder-
standing, as has already been ascertained by Zeki Velidi Togan. The present paper
seeks to place it into a new theoretical and conceptual framework—and thus to
offer an understanding of the possible meaning of this medieval account.
Keywords: tree cult, ethnology of Turkic peoples, totemism, ancestor-worship,
group identity
In the 10th century, Aḥmad ibn Faḍlān1 in his account of his trip to the Volga Bul-
gars briefly mentions the Bashkirs. There, we can read the following passage,
which is rather enigmatic:
Each of them sculpts a piece of wood the size of a phallus and hangs it on
himself. If he is about to undertake a trip or to meet an enemy, he kisses it
and prostrates himself before it, saying: “O my Lord, do unto me such and
such.” I said to the interpreter: “Ask one of them as to their justification for
this, and as to why he believes it to be his lord.” He said: “I came out of
 Edina Dallos, PhD, is a member of the MTA ELTE SZTE Silk Road Research Group, Hun-
gary. The article is a revised version of a paper delivered at the 8th International Conference
on Medieval History of the Eurasian Steppe Nomads and their Neighbors in the Middle Ages,
held in Sofia, November 20-3, 2019.
1 In 921, Caliph al-Muqtadir (r. 908–29, 929–32) accepted the invitation of the Volga Bulgars’
ruler and sent a delegation to the land of the Bulgars. Aḥmad ibn Faḍlān was a member of
this delegation and he wrote a detailed account of the trip, which lasted for almost one year.
56 EDINA DALLOS
something similar to it, and I do not know any creator of myself other than
it.”2
The “piece of wood the size of a phallus” described here has first been men-
tioned in the ethnographic literature by Zelenin. 3 In his analysis of a shamanic rite,
he interpreted Ibn Faḍlān’s reference in the sense that the Bashkirs had some kind
of a phallus cult. However, there is no trace of such a cult neither in the Bashkirian
ethnography nor in the ethnography of Turkic peoples in general, which has al-
ready been established by Zeki Velidi Togan in his book about Aḥmad ibn Faḍlān’s
travels.4
Nevertheless, this passage has not yet been analysed from other aspects even
though it raises several interesting issues. In my paper, I am going to concentrate
on one aspect only: the material of the little sculpture. That is to say, the main re-
search question is whether the wood itself is a “random” (readily available, there-
fore practical) material or whether it carries added meaning. My choice of topic
was motivated by some studies and articles that interpret Turkic people’s cults
related to trees differently from previous analyses. 5
Before detailing this particular issue, I must clarify which concepts in Ibn
Faḍlān’s description are related to the piece of wood that is the object of our study.
The following features can be identified in his description:
1. The Bashkirs sculpt a small piece of wood which they hang on themselves.
2. It is significant when they are about to undertake a trip or to meet an enemy.
3. In these potential crisis situations:
a) they kiss this object;
b) they prostrate themselves before it;
c) they pray to it.
4. They call it their Lord because they came out of something similar to it.
5. They do not know any creator of themselves other than it.
Hereinafter, I am going to examine two elements of the text that are dominant.
These are transcendence and origin, which are present in the following segments:
I: transcendence (divine/helping power): 3; 4; 5
II: origin: 4; 5
In the text, these are somehow combined in relation to the piece of wood. Such
a relation between these two aspects is not at all untypical—ethnology calls this
2 English translation: MCKEITHEN (1979). The interpretation problems of the Arabic text are
not related to the topic of my paper, but I must indicate that even in this short passage there
are some. Information about them can be found in previous editions and translations, e.g.:
TOGAN 1939; KRACHKOVSKY 1939; KOVALEVSKY 1956; SIMON 2007—the latter with a detailed
description of the text editions and the philological literature related to them.
3 ZELENIN 1928, 83–98.
4 TOGAN 1939, 148–53.
5 E.g.: KYPCHAKOVA 1983, 130–6; SAGALAEV ‒ OKTYABR’SKAYA 1990, 43–63; IVANICS 2017, 172–7.
THE TREE CULT AND THE “PIECE OF WOOD THE SIZE OF A PHALLUS” 57
phenomenon ancestor worship, one type of which is totemism. For Turkic peoples,
the totem was basically an animal (the best-known example is the wolf in the case
of the Orkhon-Turks).6 However, some scholars also suppose the totemic nature of
trees or the descent from some tree ancestor,7 especially on the basis of the sources
that I have also endeavoured to study. In my paper, I will give examples in which
one of the two concepts (transcendence or origin8) is related to a tree. I am going to
present five such examples originating from different eras and locations.
a) Maḥmūd al-Kāshgharī
Our earliest source is Maḥmūd al-Kāshgharī’s dictionary from the 11th century. The
Muslim scholar working in the Seljuq Empire compiled a dictionary of contempo-
rary Turkic languages and dialects. Under the entry Tengri, whose first meaning is
given as “God”, he also includes the following: “The infidels – may God destroy
them! – call the sky TANKRIY (täƞri); also anything that is imposing in their eyes
they call TANKRIY (täηri), such as a great mountain or tree, and they bow down to
such things.”9
Thus, albeit succinctly, this source provides the description of a kind of a tree cult.
b) Sources referring to the Uighurs
Form the 13th–14th centuries, there are several sources referring to the Uighurs
where, in a myth about the origin of the ruler, one or two trees play a key role. This
legend was recorded in most detail by Juwainī (1226–83):
In that age two of the rivers of Qara-Qorum, one called the Tughla and the
other the Selenge, flowed together in a place called Qamlanchu; and close
together between these two rivers there stood two trees; the one they call
qusuq, which is a tree shaped like a pine, whose leaves in winter resemble
those of a cypress and whose fruit is like a pignon both in shape and taste;
the other they call toz. Between the two trees there arose a great mound, and
a light descended on it from the sky; and day by day the mound grew great-
er. On seeing this strange sight, the Uighur tribes were filled with astonish-
ment; and respectfully and humbly they approached the mound: they heard
sweet and pleasant sounds like singing. And every night a light shone to a
6 The role of the wolf is not entirely clear in the early culture of the Turkic peoples. Here are
some important studies with further literature: CLAUSON 1964; SHCHERBAK 1993; GOLDEN
1982, 42‒4; GOLDEN 2018.
7 E.g. KYPCHAKOVA 1983, 134; IVANICS 2017, 174.
8 The “family tree”, signalling origin only as a symbol in many languages and cultures, does
not constitute the topic of my paper. Nor do the beliefs existing in many cultures, which are
based on the relationship of a given person with a given tree (e.g. when a tree is planted
after the birth of a child).
9 DANKOFF ‒ KELLY 1982–1985, II, 342–3.
58 EDINA DALLOS
distance of thirty paces around that mound, until just as with pregnant
women at the time of their delivery, a door opened and inside there were
five separate tent-like cells in each of which sat a man-child: opposite the
mouth of each child hung a tube which furnished milk as required; while
above the tent was extended a net of silver. The chiefs of the tribes came to
view this marvel and in reverence bowed the knee of fealty. When the wind
blew upon the children they gathered strength and began to move about. At
length they came forth from the cells and were confided to nurses, while the
people performed all the ceremonies of service and honour. As soon as they
were weaned and were able to speak they inquired concerning their parents,
and the people pointed to those two trees. They approached the trees and
made such obeisance as dutiful children make to their parents; they also
showed respect and honour to the ground in which the trees grew. The trees
then broke into speech and said: ‘Good children, adorned with the noblest
virtues, have ever trodden this path, observing their duty to their parents.
May your lives be long, and your names endure forever!’ 10
The legend recorded in chapter 22 of Yuanshi (1370) is similar in a number of
points. It is the myth of origin of the Uighur Bögü Kaghan (r. 759–80), who was
born from a tree:
There was in that country (where the Uigurs originally lived) a mountain
called Ho-lin, from which two rivers took their rise, the T’u-hu-la and the
Sie-ling-k’o. It happened once in the night-time that a stream of light fell
from heaven upon a tree standing between the two rivers; whereupon the
tree began to swell like a pregnant woman, and after nine months and ten
days gave birth to five boys. The youngest received the name Bu-k’o han.11
Now, I have no opportunity for a detailed analysis. I only want to emphasize
that in both sources the tree plays the role of a parent, so the notion of descent from
a tree is present. Rashīd ad-Dīn also mentions that the Uighur Bögü Kaghan was
born from a tree: “[...] it is said that he [Bögü Kaghan] was born from a tree.”12
Marco Polo also gives an account of a legend known to him: “Icoguristan is a large
province [...] They say that the king, who first ruled them, was not of human
origin, but born of one of those swellings that the sap produces on the bark of
trees, and that we call escal.”13
10 BOYLE 1997, 55‒6.

11 BRETSCHNEIDER 1888, 247.
12 HETAGUROV 1952, 139.
13 BENEDETTO 1931, 73.
c) The Oghuznāme
Our next source is of internal origin, i.e. a Turkic-language one, where the tree
carries at least one of the two functions mentioned above (transcendence or origin):
the Oghuznāme is written in Uighur letters from the 16th century.14 In it, on the oc-
casion of a voluntary surrender the son of Urus Bey says the following to Oghuz
Khan: “Our qut has become your qut. Our clan has become the clan of your tree.” 15
I would like to add two comments to this brief extract. The word qut is hard to
translate. In early sources it mostly has the meaning of “the charisma of the ruler”,
and later “blessing”, “life force”, “fortune”, or some kind of “soul”. 16 In this partic-
ular source it is hard to define its exact meaning and relate it with one word only,
which is why I have not translated it. A second comment must be added to the
second sentence, to the word I translated as “clan”. In the original Turkic text,
there is the word uruγ, whose first meaning was (in the Old Turkic and later, too)
“seed”,17 which later came to metaphorically mean “descendant”, “originating
from the same ancestor”, and thus also “clan”.18 Therefore, the extract can be trans-
lated as follows: “Our seeds (descendants) have now become the seeds (descend-
ants) of your tree.”
d) The Chinggisnāme
Our next sources are the Chinggisnāmes, recorded in the 17th–18th centuries (in the
Volga Tatar literary language), which were compiled from previous fragments
based on earlier oral tradition and court historiography. According to the texts,
Chinggis Khan called together his beys at the foundation of his state and gave
them territories and peoples, and he also “[...] created people-units by giving each
and every bey a tamga, a bird, a tree and a password.”19
Here are but two brief extracts about how this happened (there are fifteen such
tamga, bird, tree, and password units listed in the text, the sixteenth one being
Chinggis’s own):
Then thus spake Chinggis Khan: ‒ Ej Kereit-bey, thy tree shall be the linden,
thy bird the goose, thy password the jackal and thy tamga shall be an eye 20
[...] Then thus spake Chinggis Khan: ‒ Ej Borkit-bey, thy tree shall be the
14 Most recently on the dating: DANKA 2019. Naturally, this is the time of its recording, not
its creation, as the text is a compilation of various written texts and oral-traditions.
15 The latest edition: DANKA 2019. Translation of this sentence: “Herewith, our regal charis-
ma (qut) has become your regal charisma. Herewith, our progeny (uruɣ lit. ‘seed’) has be-
come the progeny of your lineage (ïɣač, lit. ‘tree’).” (Danka 2019, 95).
16 For the meanings of qut: CLAUSON 1972, 594, SEVORTYAN 1974‒2003, VI, 176‒8.
17 CLAUSON 1972, 214‒215; SEVORTYAN 1974‒2003, I, 604‒6
18 Cf. CLAUSON 1972, 214.
19 IVANICS ‒ USMANOV 2002, 56; 228.
20 IVANICS‒USMANOV 2002, 57; 228.
60 EDINA DALLOS
maple, thy bird the hoopoe, thy password Buruǰ [the name of a clan] and thy
tamga shall be the letter amza21 [...]22
After this, the text provides the list of tamgas, birds, trees, and passwords.23
The fact that a community or social group has some kind of communal attach-
ment to a certain tree species is first shown here. However, it is also worth men-
tioning that the signs or markers this source talks about had not originated inter-
nally and did not characterize a given clan for a long time but were given to the
people-units ordered under the beys from the outside: from Chinggis himself.24
Nevertheless, albeit only in traces, we can find similar matches in some 19 th- and
20th-century records about Altaian and Khakas peoples. Unfortunately, the availa-
ble data are not the result of systematic collection and, what is more, it is clear even
from these scattered accounts that the system was already incomplete and incon-
sistent at the time. Nevertheless, let me present some interesting data.
e) The Khakas čula and some Altaian data
Potanin recorded a few Altaian Turkic clan names during his trip to north-western
Mongolia in 1879–80,25 as well as some fragments of legends and rhymes related to
them. Some of these provide an explanation to the name and origin of the clans.
Here, we can find some short notes about tree-origins, for example the Irqït clan is
supposed to have originated from the rgaĭ26 (Cotoneaster) tree. According to an-
other source, the father of the Irqït is the rgaĭ and their mother is the birch tree.27
21 Arabic hamza.
22 IVANICS‒USMANOV 2002, 58; 228.
23 The following tree species are mentioned: larch, Caucasian elm, field elm, birch, beech,
elm, willow, linden, rowan, oak, maple, juniper, sandalwood, ash, alder, and Chinggis’s tree
is the plane.
24 Delyan Russev drew my attention to another possibility: „Isn’t it possible that these were
actually ’internal’, traditional markers that were ideologically linked to Chinggis in the
literary tradition?” The suggestion cannot be dismissed a priori.
25 POTANIN 1883, 7.
26 POTANIN gives this form (ргай) both in Russian and Turkic environments, even though it
can be found neither in the Russian nor in the Turkic languages in this form. The Latin
equivalent (Cotoneaster) that Potanin provides is the name of a shrub. Equivalents of the
Mongolian word irghai (Cotoneaster) can be found in the Turkic languages of Southern
Siberia as well as in Russian as Mongolian loan words, even though with different mean-
ings. Cf. Altaian ïrgay ‘honeysuckle’ (Lonicera) (VERBITSKY 1884, 458), which is also a creep-
ing shrub, Russian irga ‘saskatoon’ (Amelanchier); ‘meadowsweet’ (Filipendula); cotoneaster
(EsBE 1890‒1907, XIII, 310). Further data: RAMSTEDT 1935, 216; VASMER 1950‒1958, I, 486;
RÄSÄNEN 1969, 166.
27 POTANIN 1883, 7.
The Qara (‘black”) brothers’ origin is said to be the following: “They came out of
the black kind of iron, from the inside of the black tree.” 28
Among the Khakas people, L’vov and Usmanov recorded in the 1970s that their
informants had another group identity besides their clan (sök ‘bone; clan’), which
was their čula: always the name of a tree species.29 In the Turkic languages of
Southern Siberia, čula (Altaian šula) means “name, nickname, honorary name, ti-
tle”.30 The čula tree-names and the clan names did not provide a matching or com-
plementary system—whereas 43 clan names were recorded, there were only 10
different čulas. What is more, several clans shared the same čula and sometimes
two or three different čulas occurred within one clan. However, one informant said
that a man and a woman with the same čula did not get married,31 which suggests
that the čula as an exogamous group marker was somehow related to the clan sys-
tem (at least in the given social group). The reason for the incomplete system may
be that the structure that once had fully existed lost its meaning and function by
the mid-20th century and at the time of the recording only its remnants, mostly
remembered by the elderly, could be found by the researchers.
In summary, let us see what conclusions can be drawn from the sources listed
here. We have several sources referring to the Uighurs that point to a tree ancestor.
The only longer text, Juwainī, also mentions the element of religious respect (kneel-
ing and kissing the ground) besides the miraculous birth; therefore, we can say that
origin and transcendence are strongly combined here. It is worth emphasizing
again that this is not the myth of origin of a certain clan, tribe or people, but the
Uighurs only gave this explanation about the origin of their ruler. 32 In the Oghuz-
nāme, the tree “only” refers to the origin, which is only partly biological and
strongly linked to loyalty. The “common” trees have a rather different nature in
the texts of the Chinggisnāmes, where the non-consanguineous, newly-created
28 POTANIN 1883, 8.
29 This collection has not been published; some parts are found in: SAGALEV ‒ OK-
TYABR’SKAYA 1990, 54-9.
30 Some authors incorrectly identify the Khakas word čula (Altaian šula) with the Altaian
word yula / d’ula ‘a kind of soul; the soul of the deceased’. (For the word yula: VERBITSKY
1884, 102; ANOKHIN 1929, 253-8; MALOV 1929, 330). Based on the correspondence, they try to
ascribe some sacred content to the word čula, which is totally wrong! (So did SAGALAEV ‒
OKTYABR’SKAYA 1990, 56, 194; IVANICS 2017, 174). For the Khakas word čula / Altaian šula:
VERBITSKY 1884, 449; DYRENKOVA 1940, 282, 408; RÄSÄNEN 1969, 115; RYUMINA-SYRKASHEVA ‒
KUCHIGASHEVA 1995, 100.
31 SAGALAEV ‒ OKTYABR’SKAYA 1996, 56.
32 This topic deserves a separate study, so here I am only mentioning the fact that in 762
Bögü Kaghan converted to Manichaeism, which he then made a state religion. The legends
and legend-fragments describing Bögü’s miraculous birth show an origin which is very
different from the legends of other Turkic peoples: instead of the commonly known animal
origin, here we can find a tree-parent. The two facts (Manichaeism and the markedly differ-
ent legend of tree-origin) may be related to each other.
62 EDINA DALLOS
communities are given common markers in the form of birds, trees, tamgas, and
passwords.33 In the latest and rather fragmented Altaian and Khakas data, the tree
is a kind of a marker of origin.
Table 1. Origin and transcendence

Source ORIGIN TRANSCENDENCE
individual communal
Kāshgharī ‒ ‒ +
Uighur “royal”
+ ‒ +
legend
Oghuznāme ‒ ‒ > + ‒
Chinggisnāme ‒ ‒ > + ‒
Khakas čula ‒ + ‒
In the table, the signs ‒ > + mark a movement: the non-common origin is re-
placed by a fictitious common origin or at least some kind of expression thereof. In
the case of the Oghuznāme it appears as a part of the act of surrender, and in the
Chinggisnāme it is a sign of the integration of a newly-created group into the mili-
tary and imperial order.
It is clear from the table that origin and transcendence are only combined in the
Uighur myth of origin but, as we could see before, even there it is not represented
on a communal but an individual level, in a specific relation to the ruler. Examin-
ing the relationships of tree and origin more accurately, it is also clear that a real
descent from a tree is only expressed in the Uighur legend:
33Each of these can be interpreted as the marker of a genealogical community. We may not
only think of the tamga, tree or bird, since clan names and animal names also occur among the
passwords. However, among the tamgas, traditionally used as the emblem of a clan or tribe,
we can also find ones that are not of great antiquity or culturally not embedded, such as the
Arabic hamza. The four types of markers are summarized in a table by IVANICS 2017, 265.
Table 2. Origin from a tree/the tree is a symbol of origin

“blood” origin kinship is marked community is
from a tree by a tree marked by a tree34
Uighur “royal”
+
legend
Oghuznāme +
Chinggisnāme +
Khakas čula +
My original aim was to fit Aḥmad ibn Faḍlān’s information on the topic among
the more specific cases of tree worship by Turkic peoples, which may refer to some
kind of ancestor worship. Examining the background of the issue, I have found
that in the culture of Turkic peoples the reverence for trees was not combined with
totemistic concepts; in fact, we only have a single (specifically, Uighur) example of
this. On the basis of the above, I assume that in the short extract by Aḥmad ibn
Faḍlān wood as a material had no added meaning, as the religious respect and some
kind of ancestry are specifically present together. This data can rather be related to
the forms of ancestor worship where some kind of small puppet or doll (made of
different materials according to different sources 35) could constitute a helping force
or a cult object depicting the ancestor and connecting with the transcendent.
Bibliography
ANOKHIN, A. V. (1929): Dusha i eë svoistva po predstavleniyu teleutov. Sbornik
Muzei antropologii i etnografii 8, 253–269.
BENEDETTO, L. F. (1931): The Travels of Marco Polo. Translated from the text of L. F.
Benedetto by A. Ricci, London.
BOYLE, J. A. (19972): Genghis Khan. The History of the World Conqueror by ’Ala-ad-Din
’Ata-Malik Juvaini. Translated from the text of Mirza Muhammad Qazvini by J.
A. Boyle, Manchester.
BRETSCHNEIDER, E. (1888): Medieval Researches from Eastern Asiatic Sources, vol. 1,
London.
CLAUSON, G. (1964): Turks and Wolves. Studia Orientalia XXVIII/2, 1–22.
34 In the case of the data only appearing in this box, the question may arise whether this
information belongs to the topic of my paper, as they not only do not refer to descent from a
tree, but there is neither blood lineage within the groups marked with a particular tree, nor
are the trees here connected to the transcendent aspect(?). These data are important, howev-
er, as they provide some kind of communal marker for a secondary or external, (not even
fictitiously consanguineous) group, which must have been a marker or creator of group
identity, interpretable within the given culture(s). Therefore, it is worth taking them into
consideration even if only as secondary systems.
35 TOGAN 1939, 149 (with further literature).
64 EDINA DALLOS
CLAUSON, G. (1972): An Etymological Dictionary of Pre-Thirteenth Century Turkish,

Oxford.
DANKOFF, R. ‒ KELLY, J. (1982–1985): Maḥmūd al-Kāšγarī: Compendium of the Turkic
dialects (Dīvān Luγāt at-Turk), Cuxbury.
DANKA, B. (2019): The ‘Pagan’ Oɣuz-namä. A Philological and Linguistic Analysis, Wies-
baden.
DYRENKOVA, N. P. (1940): Shorskii fol’klor, Moskva‒Leningrad.
EsBE (1890-1907): Enciklopedičeskij slovar’ Brokgauza i Efrona, vol. 1‒86, Sanktpeter-
burg.
GOLDEN, P. (1982): Imperial Ideology and the Sources of Political Unity amongst
the pre-Činggisid Nomads of Western Eurasia. Archivum Eurasiae Medii Aevi 2,
37–76.
GOLDEN, P. (2018): The Ethnogonic Tales of the Türks. The Medieval Historial Journal
21/2, 291–327.
HETAGUROV, L. A. (1952): Rašid-ad-Din: Sbornik letopisei. I. Perevod s persidskogo
L. A. Hetagurova. Moskva‒Leningrad.
IVANICS, M. (2017): Hatalomgyakorlás a steppén. A Dzsingisz-náme nomád világa, Bu-
dapest.
IVANICS, M. ‒ USMANOV, M. (2002): Das Buch der Dschingis-Legende (Däftär-i Čingiz-
nämä), I. Vorwort, Einführung, Transkription, Wörterbuch, Faksimiles, Szeged.
KOVALEVSKY, A. P. (1956): Kniga Ahmeda Ibn-Fadlana o ego puteshestvii na Volgu v
921‒922 gg. Stat’i, perevody i kommentarii, Kharkov.
KRACHKOVSKY, I. U. (1938): Puteshestvie Ibn-Fadlana na Volgu. Perevod i kommenta-
ri, Moskva‒Leningrad.
KYPCHAKOVA, L.V. (1983): K voprosu o kul’te derev’ev u altaitsev. In Modorov, N.
S. et al. (ed.): Voprosy arheologii i ėtnografii Gornogo Altaya. Gorno-Altaisk, 130–136.
MALOV, S. E. (1929): Neskol’ko zamechanii k stat’e A. V. Anokhina «Dusha i eë
svoistva po predstavleniyu teleutov». Sbornik Muzei antropologii i etnografii 8,
330–333.
MCKEITHEN, J. E. (1979): The Risalah of Ibn Fadlan: An Annotated Translation with
Introduction. Unpublished PhD dissertation, Department of Near Eastern Lan-
guages and Literatures, Indiana University.
POTANIN, G. N. (1883): Ocherki Severo‒zapadnoi Mogolii. IV. Materialy etnograficheskie,
Sankt-Peterburg.
SAGALAEV, A. M. ‒ OKTYABR’SKAYA, I. V. (1990): Traditsionnoe mirovozzrenie tyurkov
yuzhnoi Sibiri. Znak i ritual, Novosibirsk.
SEVORTYAN, E. V. (1974‒2003): Etimologicheskii slovar’ tyurkskikh yazykov, vol. 1‒8,
Moskva.
SHCHERBAK, A. M. (1993): On the Chief Totem of Ancient Turks (mainly on the
basis of linguistic material). Türk Dilleri Araştırmaları 3, 203-211.
SIMON, R. (2007): Ibn Fadlán: Beszámoló a volgai bolgárok földjén tett utazásról. Arab
eredetiből fordította, a jegyzeteket és az utószót írta Simon Róbert, Budapest.
TOGAN, Z. V. (1939): Ibn Fad̟lān’s Reisebericht, Leibzig.

RYUMINA-SYRKASHEVA, L. T. ‒ KUCHIGASHEVA, N. A., eds. (1995): Teleut-orus Sözlik,
Kemerovo.
VERBITSKY, V. I. (1884): Slovar’ altaiskogo i aladagskogo narechii tyurkskogo yazyka, Kazan.
ZELENIN, D. (1928): Ein erotischer Ritus in den Opferungen der Altaischen Türken.
Internationales Archiv für Ethnologie 29, 83–98.
Pechenegs in the Imperial Space of
Byzantium in the Middle of the
10th Century
IRINA KONOVALOVA
Abstract
The paper is devoted to the analysis of the Pecheneg “dossier” in the work of Con-
stantine VII Porphyrogenitus De administrando imperio, and, in particular, to the
place occupied by the Pechenegs in the space of the foreign policy interests of the
empire. The study of information about the Pechenegs, as well as the analysis of
the terminology used to describe the Pecheneg society, showed, on the one hand, a
deep knowledge of the administrative-political and social structure of the
Pechenegs, their historical background and their important role in the balance of
power in the steppes of Eastern Europe, and on the other, revealed the ability of
Byzantine intellectuals to explain the significance of the Pecheneg factor for Byzan-
tium in terms of their own political and cultural tradition.
Keywords: Constantine VII Porphyrogenitus, De administrando imperio, Pechenegs,
mental map, imperial policy, Greek terminology
The Pecheneg “dossier” of Constantine VII Porphyrogenitus (r. 913–59) has been
repeatedly studied by historians and linguists who analysed the information of the
Byzantine emperor in various contexts: textual, historical, linguistic, ethnocultural. 1
I would like to draw attention to a less studied geographical aspect of the infor-
mation, namely to the place occupied by the Pechenegs on the mental map of the
author of the treatise De administrando imperio (hereinafter — DAI).
Descriptions of the Pechenegs are placed in several chapters of the DAI. By their
nature, all data relating to the Pechenegs can be divided into three groups. The first
 Institute of World History at the Russian Academy of Sciences, Russian Federation, Mos-
cow: irina_konovalova@mail.ru
1 The literature devoted to the analysis of the Pecheneg “dossier” is enormous, for biblio-
graphic reviews see: LITAVRIN – NOVOSEL’TSEV 1989, 279–80, 388–91; GOLDEN 1995; MALAMUT
1995; SHCHERBAK 1997; SPINEI 2003, 155–9; KONOVALOVA 2009; KOZLOV 2012; KOZLOV 2013.
68 IRINA KONOVALOVA
of them (chapters 1–8), devoted to the place of the Pechenegs in the system of for-
eign policy interests of Byzantium, is distinguished by a high degree of systemati-
zation of the material due to the fact that it is conceptually developed and didacti-
cally designed. The second group of information about the Pechenegs is a sum-
mary of historical, ethnographic, socio-political, and geographical data (chapter
37), which, although presented in a systematic form, do not have a strongly
marked didactic character. Finally, the third group includes scattered information
about the Pechenegs cited in connection with the stories of other peoples (in chap-
ters 9, 13, 31, 38–40, 42, 53). If we leave aside separate references to the Pechenegs,
it turns out that they are described in detail twice: in chapters 1–8 and in chapter 37.
Trying to explain this fact, R. Jenkins suggested to consider chapter 37 as relat-
ing to the “antiquarian”, by his definition, part of the treatise, and as such he con-
traposed it with the story of the Pechenegs in the DAI narrative of the Northern
peoples (chapters 1–13), which he thought to be the “actual” part of the treatise
composed last. R. Jenkins believed that chapter 37 could not be mechanically trans-
ferred to the beginning of the treatise since its “antiquarian style” was incompati-
ble with the tasks of the compiler, who had to show (in chapters 1–13) what benefit
the Northern peoples could provide for Byzantium.2 It should be noted that the
definition of chapter 37 as “antiquarian” contradicts the commentary on the text of
the chapter by Gy. Moravcsik, according to which the information presented in this
chapter was collected in the time of Constantine, 3 whence it follows that the com-
pilers of the treatise subjectively regarded it as actual, and not “antiquarian”. J.
Howard-Johnston rejected such an assessment of chapter 37, referring, however, its
compilation not to the middle, but to the beginning of the 10th century.4 P. Magda-
lino recognized the disjointed character of the treatise and suggested that the key
to understanding the work of Constantine VII is “the power perspective of the
DAI”, which determined the structure of the book and gave it inner integrity.5 One
more consideration regarding the structural features of the DAI was recently ex-
pressed by A. S. Shchavelev, who supposed on the basis of formal markers that the
treatise had two authors, Constantine VII himself and an “Anonymous Collabora-
tor”. According to his hypothesis, the chapters dedicated to the Pechenegs be-
longed to different compilers: chapters 1–13 were written by Constantine, and
chapter 37—by his “Anonymous Collaborator”. 6
Thus, in historiography it is recognized that the Pecheneg “dossier” lacks integ-
rity as the information about the Pechenegs placed in chapters 1–8, on the one
hand, and that in chapter 37, on the other, have significant differences, since they
2 JENKINS
1962, 7, 12.
3 JENKINS 1962,
143.
4 HOWARD-JOHNSTON 2000, 319, 324–6.
5 MAGDALINO 2013, 36.
6 SHCHAVELEV 2019, 691–701.
PECHENEGS IN THE IMPERIAL SPACE OF BYZANTIUM… 69
were collected at different times and, possibly, processed by different authors.

Nevertheless, a statement of this fact makes it impossible to understand why the
compiler (or compilers) of the book needed such a composition of the story about
the Pechenegs and what purpose it pursued.
The chapters devoted to the Northern peoples are the most conceptually devel-
oped part of the composition. In each chapter relating to this group, firstly, the
implicit question that this chapter is intended to answer is easily read, and second-
ly, the answer to it is almost always clearly formulated—going as far as to include
step-by-step instructions to the Byzantine ambassadors and officials in chapters 7–
8. For example, chapter 1 contains the answer to the question of why the Byzantine
emperor needs peaceful relations with the Pechenegs: because Pechenegs can raid
Cherson and the so-called Climates (τὰ Κλίματα). Chapter 2, devoted to what By-
zantium can extract from the relations between the Pechenegs and the Rus’, con-
tains an answer to the question of what happens when the Rus’ are at war with the
Pechenegs, the answer being that the latter cannot come to Constantinople either
for war or for trade. The third chapter, which analyses the relations of the
Pechenegs and the Magyars (οἱ Τούρκοι), shows that the latter were afraid of the
Pechenegs. In the fourth chapter, Constantine goes from analysing bilateral rela-
tions to finding out what Byzantium can extract from the mutual relations of the
Pechenegs, the Rus’ and the Magyars, and answers that, on the one hand, the peace
with the Pechenegs guarantees the safety of the empire against attacks by the Rus’
and the Magyars as well as against racketeering from them and, on the other hand,
that it is possible to induce the Pechenegs—as Byzantine allies—to attack the Rus’
and the Hungarians. The fifth chapter, to consider the balance of forces north of
Byzantium, adds the Bulgarians and concludes that they, like the Hungarians, are
interested in peaceful relations with the Pechenegs. The sixth chapter, as if summa-
rizing everything stated in the first five chapters, contains the conclusion that the
Pechenegs can be used as intermediaries in imperial relations with all the Northern
peoples. Developing this theme, chapters 7–8 give instructions to Byzantine offi-
cials on interacting with the two main Pecheneg groups, which nomadized on the
right and left sides of the Dnieper respectively. Chapters 9–12 discuss the interests
and policies of the empire regarding the Rus’, the Khazars, and the Alans. And
chapter 13 summarizes the section on all the Northern peoples, explaining how to
handle them in the interests of the Byzantine Empire.
In chapters 1–8, which deal with the Pechenegs, the power relationship in the
region north of Byzantium is presented in an extremely general form. The space in
which the region’s main political players—the Pechenegs, the Rus’, the Hungari-
ans, and the Bulgarians—are placed is almost devoid of geographical signs. In the
first five chapters Constantine manages to write about all these peoples using only
one toponym—imperial Cherson (Χερσών: 1.26–28), which throughout the whole
story about the Northern peoples continues to be the most important geographical
marker of the space they occupy (6.3; 7.1, 3, 6, 16; 8.8; 11.1, 8, 10, 12). The Northern
70 IRINA KONOVALOVA
peoples are localized either in relation to each other or through the outside world.
So, about the Pechenegs it is reported that they are adjacent to Cherson (1.25–26;
6.2–3), the Rus’ (2.2), the Bulgarians (5.5–6), and adjoin the Hungarians from the
North (13.4–5). Only the names of the Dnieper and Dniester rivers, “towards the
area of which” the Pechenegs are located (8.5–7), can be considered as an indica-
tion of the limits of the Pecheneg territory.
In the story about the Northern peoples in chapters 1–13, the compiler’s atten-
tion was not directed to a comprehensive description of the Northern nomads, but
to an analysis of their relationship in terms of the benefits that the empire could
derive from this. Therefore, speaking about the relations between the Pechenegs,
the Rus’, the Hungarians, and the Bulgarians, Constantine omits the details, giving
only the essence. For example, in chapter 5 the Pechenegs are considered as a hos-
tile force in general with respect to the Bulgarians, although Constantine is well
aware of the cases of joint actions of these peoples, in particular, their campaign of
895 against the Hungarians (40.13–16). In characterizing the relations of the North-
ern peoples, the emphasis is placed on the contradictions that exist between them,
which in fact does not exhaust the whole real picture of their ties, but it demon-
strates well how these contradictions could be used in the interests of Byzantium.
Thus, reporting on the Pechenegs and the Rus’, Constantine underlines only one
side of their relationship—the damage that the raids of the nomads caused to the
Rus’ (2.2–4, 9–11, 19–23; 4.9–13), although relations between these peoples were
actually much more multifaceted.7 The trade of the Rus’ with the Pechenegs was
evaluated by Constantine only as a foreign policy factor, which, from his point of
view, could seriously affect the interest of the Rus’ in maintaining peaceful rela-
tions with the Pechenegs (2.5–8).
It can be concluded that in chapters 1–8 the Pechenegs are characterized in the
framework of a special kind of space—the space of the foreign policy interests of
the Byzantine Empire, and this angle of view set both the principles of data selec-
tion and the parameters of the description itself.
The information about the Pechenegs included in chapter 37 is also conceptual-
ized, but in a different way than at the beginning of the treatise. This chapter tells
of the life of the Pechenegs in the Volga-Ural interfluve and the history of their
emergence in the steppes north of the Black Sea, and it describes in detail the inter-
nal structure of the Pecheneg society and the geographical location of the Peche-
negs. Despite the abundance of specific data that, at first glance, is preparatory or
even optional in relation to chapters 1–8, chapter 37 nevertheless continues the same
line on the consideration of the Pechenegs through the prism of imperial interests.
This approach is shown in chapter 37 primarily in the subtle use of Greek ter-
minology to describe certain phenomena of the Pecheneg society, and only those
7For a general outline of the Rus’-Pecheneg relations and historiography, see: TOLOCHKO
2003, 45–66; SPINEI 2003, 114–126; MIKHAILOVA 2006, 60–76.
whose adequate understanding was necessary for the effective conduct of imperial
policy towards the Pechenegs. The names of the Pecheneg “themes” are given with-
out translation and have not been commented on in any way by Constantine. At the
same time, the most important concepts describing the hierarchy of power among
the Pechenegs are given in Greek terms familiar to the imperial administration.
According to chapter 37, the Pecheneg union consisted of eight “themes”
(θέμα)—four on each side of the Dnieper—which, in turn, included forty smaller
“parts” (μέρος). The term “theme” is used here neither in its technical meaning of a
military-administrative district,8 usual for the royal office of the 10th century, nor as
a synonym for the words “territory”, “habitat”, “region”, 9 but as a designation of a
specific form of administrative-territorial, military-political, and socio-economic
organization of the Pechenegs. Like other steppe peoples, the basis of the Pecheneg
society was a clan organization with absolute dominance of the principles of gene-
alogical and blood relationship. The Byzantine officials, apparently, were aware of
this feature of the Pecheneg society, since, describing the contemporary placement
of their eight “themes” in space, the term “kin” (γένος) is used twice as a synonym
for the word “theme” (37.34, 39). In addition, from the given list of names of the
rulers of the “themes” (37.21–24), it is quite obvious that even during the clashes
between the Pechenegs and the Oghuzes in the Volga region (to which period,
according to the DAI, this list of names refers), their “themes” wore the same de-
nominations as later, when the Pechenegs migrated to the Pontic steppe (37.16–24,
35–36, 40–43). Thus, according to chapter 37, the term “theme” in relation to the
Pechenegs means primarily a people and only then the territory occupied by it.
Modern scholars of nomadism suppose that the term “theme” refers to the
Pecheneg tribes (tribal groupings)10 organized in a kinship-based political hierar-
chy of the "chiefdom" type.11
The rulers of the Pecheneg “themes” are called “great archons” (37.16: μἐγαλοι
ἄρχοντες) to contrast them with “lower rank archons” (37.33: ἐλάττονες ἄρχοντες).
However, in the same chapter, the rulers of the “themes” are twice called simply
“archons” (37.20, 32). Similarly, these terms are used in relation to the rulers of the
Hungarians: Árpád (ca. 895 – ca. 907) is once called “the great archon of Turkey”
(40.53), and in other cases he and his direct heirs are called simply “archons”
(38.49, 51, 53, 56; 40.12–13, 48, 58). The term “great archon” also refers to the eldest
son of the Great Moravian Prince Sviatopolk (r. 871–94), who is placed over his
other two co-rulers (41.6); the title of Sviatopolk himself is simply “archon” (41.2).
8 KAZHDAN 1991.
9 MORAVCSIK – JENKINS 1962, 145: “the word means here the territory or homeland of the
various Pecheneg clans”; MALAMUT 1995, 110: “le terme de ‘thème’ donné aux provinces”;
BELKE – SOUSTAL 1995, 185–187: “acht Provinzen (Themen)”; TOLOCHKO 2003, 46: the author
explains the word “theme” as “the area of settlement of particular hordes”.
10 GOLDEN 1995, 289.
11 MAREI 2000, 338.
72 IRINA KONOVALOVA
Thus, the term “great archon”, as far as one can judge from the text of the DAI, is
not used in a strictly terminological sense, unlike, for example, the term “archon of
archons” (ἄρχων τῶν ἀρχόντων), used as the title of Bagratids (see: 43.30, 34–35, 112;
44.6, 7, 9, 13, 18, 19, 20, 21–22, 26–27, 35, 37, 38–39, 42, 45, 50, 120).
Constantine’s assertion that units included in any “theme” were headed by “ar-
chons of a lower rank” in comparison with the rulers of the “themes” themselves,
indicates that Byzantine politicians and intellectuals had an idea of the hierarchical
structure of the Pecheneg society. As it was with other nomads, the Pecheneg ad-
ministrative mechanism was inextricably linked with the hierarchy of levels of
kinship. In this sense, the term μέρος (“part”, “share”) used to denote the structural
units of the Pecheneg “theme” undoubtedly had not only administrative and terri-
torial significance, but also reflected the corresponding level of the patrimonial
system of the Pechenegs.12 Thus, in the “archons of a lower rank” one can see the
leaders of individual Pecheneg clans13—those whom the missionary archbishop
Bruno of Querfurt, who visited the Pechenegs at the beginning of the 11 th century,
called the “elders” (maiores, meliores).14
The division of eight Pecheneg “themes” into two wings (with four “themes” in
each) noted in the DAI is a reflection of the ancient Turkic tradition of the dual left-
right organization of power.15 The information about the “great archons” that
headed eight Pecheneg “themes” seems to imply their complete equality from the
point of view of the Byzantines. This is evidenced by the absence of any reflection
regarding the actual Pecheneg titles of their senior officials—Yula (γύλα, the Gre-
cized form of the Magyar personal name and title “Gyula”) and Tzour (τζούρ).
These terms are only mentioned as the names of two “themes” (37.17–18, 35, 41,
69–70), despite the fact that the author of the DAI knew that the nomads that be-
longed to the corresponding “themes” of Kouartzitzour (Κουαρτζιτζούρ) and
Chabouxingyla (Χαβουξιγγυλά) were the most notable and influential among the
Pechenegs (37.68-71).
Thus, the use of Greek terminology in the story of the Pecheneg “themes” indi-
cates a deep understanding in Byzantium of the nature of the socio-political struc-
ture of the Pecheneg society. The close attention to precisely these aspects of
Pecheneg life demonstrated in chapter 37 fully reflects the significance that was
attached to the relations with the Pechenegs by the imperial administration in the
middle of the 10th century, and this circumstance allows us to consider chapters 1–8
and chapter 37 as parts of a uniform “dossier”. The inner unity of these parts of the
treatise is also indicated by the fact that chapters 1–8 clearly show how the Byzan-
12 P. Golden, for example, translates the term in question as “clan groupings” (GOLDEN 1995, 289).
13 MAREI 2000, 338.
14 KARWASIŃSKA 1973, 99–100.
15 GOLDEN 1982, 52–3, 63–5, 68.
tines’ ideas about the Pecheneg society (described in detail in chapter 37) were
applied in the real diplomatic practice of the empire.
The composition of chapters 5–8, devoted to the relations of Byzantium with
two groups of Pechenegs (left- and right-bank) is based on the understanding of
the dual structure of the Pecheneg tribal confederation, divided into two wings, the
border between which was the Dnieper (37.34–45). In their diplomatic practice, the
Byzantines distinguished precisely two groups of Pechenegs, which is clearly seen
from chapters 7–8. These chapters contain instructions for imperial officials sent to
different groups of Pechenegs—respectively—from Cherson to the left-bank
Pechenegs (chapter 7) and from Constantinople to the Pechenegs nomadizing be-
tween the Dnieper and the Danube (chapter 8).
At the same time, knowledge of the fractional structure of each of the two
Pecheneg wings allowed Byzantine officials to distinguish a special group of
Pechenegs — “another people (λαός) from the same Pachinakites” (6.2), who lived
near Cherson, traded with it, and also performed intermediary functions between
the Chersonites and the emperor, on the one hand, and Russia, Khazaria, and Zi-
chia, on the other (6). Constantine’s reference to a certain autonomy of this group
of Pechenegs (6.11) may relate not only to assessing their status as a foreign politi-
cal partner, but also to the structure of the Pecheneg society itself, in particular to
the lack of central authority among the Pechenegs. 16 Obviously, this group was
identical to one of those Pecheneg “themes” that were discussed in chapter 37. In
this case, not only γένος (37.34, 39), but λαός (6.2) is also synonymous with the term
“theme”.
The volume and content of the information about both the Pechenegs them-
selves and their place in the system of foreign policy interests of the empire, as well
as the terminology used in the description, indicate that the most likely informants
about the Pechenegs could have been people that belonged to two cultural tradi-
tions at once—this allowed them not only to provide valuable information about
the Pecheneg society, but also to present it in terms familiar to the Byzantines. This,
as it seems, confirms the assumption made by S. A. Kozlov that ethnic Turks who
converted to Orthodoxy and transferred to the diplomatic and military service of
the empire played a significant role in collecting and processing information for
the treatise. One such informant for chapter 37 could have been the Cherson strate-
gos (στρατηγός) John Bogas, who is a vivid example of the adaptation of the
Pechenegs to the imperial space of Byzantium. 17
In my opinion, the information about the Pechenegs in chapters 1–8, on the one
hand, and in chapter 37, on the other, should be considered as organically supple-
menting each other (regardless of their authorship) as parts of the treatise, and
together constituting a uniform Pecheneg “dossier”, which was undoubtedly writ-
16 MALAMUT 1995, 114.

17 KOZLOV 2012; KOZLOV 2013.
74 IRINA KONOVALOVA
ten with specific intentions. If in chapters 1–8 the Pechenegs are characterized as
one of the many peoples of Eastern Europe that substantially influenced the bal-
ance of power in the region, then in chapter 37 the focus and the scale of the de-
scription change, and the Pechenegs themselves with their historical background
come to the fore.
Bibliography
BELKE, K. – SOUSTAL, P., ed. (1995): Die Byzantiner und ihre Nachbarn. Die De adminis-
trando imperio gennante Lehrschrift des Kaisers Konstantinos Porphyrogennetos für
seinen Sohn Romanos. Übersetzt, eingeleitet und erklärt von K. Belke und P.
Soustal, Wien.
GOLDEN, P. B. (1982): Imperial Ideology and the Sources of Political Unity Amongst
the Pre-Činggisid Nomads of Western Eurasia. Archivum Eurasiae Medii Aevi 2,
37–76.
GOLDEN P. B. (1995): Pečenegs, in Encyclopaedia of Islam. 2nd ed. Vol. 8. Leiden, 289–291.
HOWARD-JOHNSTON, J. (2000): The De Administrando Imperio: A Re-Examination of
the Text and a Re-Evaluation of its Evidence about the Rus, in Kazanski, M. –
Nercessian, A. – Zuckerman, C. (ed.:) Les centres proto-urbains russes entre
Scandinavie, Byzance et Orient : Actes du colloque international tenu au Col-
lège de France en octobre 1997. Paris, 301–336.
JENKINS, R. J. H. (ed.) (1962): Сonstantine Porphyrogenitus. De administrando imperio.
Vol. 2: Commentary. London.
KARWASIŃSKA, J. (1973): List Brunona do króla Henryka, in Monumenta Poloniae
Historica. Nova series. T. 4. Fasc. 3. Warszawa, 97–106.
KAZHDAN, A. (1991): Theme, in Kazhdan, A. (ed.:) The Oxford Dictionary of Byzantium.
New York – Oxford, 2034–2035.
KONOVALOVA, I. G. (2009): Pechenezhskoe dosʹe Konstantina Bagryanorodnogo.
Vostochnaya Evropa v drevnosti i srednevekovʹe XXI, 139–146.
KOZLOV, S. A. (2012): Konstantin Bagryanorodnyi o pechenezhskikh «femakh»
(DAI. Cap. 37) i problema ego istochnikov. Vostochnaya Evropa v drevnosti i
srednevekovʹe XXIV, 113–120.
KOZLOV, S. A. (2013): Byzantinopecenacica I: Bogas i Kegen — pechenezhskie
«yazyki» na vizantiiskoi sluzhbe, Archivum Eurasiae Medii Aevi 20, 103–127.
LITAVRIN, G. G. – NOVOSEL’TSEV, A. P., ed. (1989): Konstantin Bagryanorodnyi. Ob
upravlenii imperiei: Tekst, perevod, kommentarii, Moscow.
MAGDALINO, P. (2013): Constantine VII and the Historical Geography of the Em-
pire, in Bazzar, S. – Batsaki, Y. – Angelov, D. (ed.:) Imperial Geographies in Byz-
antine and Ottoman Space. Cambridge Mass. – London, 23–42.
MALAMUT, E. (1995): L’image byzantine des Petchénègues. Byzantinische Zeitschrift
88/1, 105–147.
MAREI, A. V. (2000): Osobennosti sotsialʹno-politicheskoi organizatsii pechenegov,

in Kradin, N. N. – Kortaev, A. V. – Bondarenko, D. M. – Lynsha, V. A. (ed.:)
Alʹternativnye puti k tsivilizatsii. Moscow, 337–343.
MIKHAILOVA, I. B. (2006): Drevnyaya Rusʹ i Stepʹ, in Krivosheev, Yu. V. (ed.:) Rossi-
ya i stepnoi mir Evrazii: Ocherki. Saint Petersburg, 9–128.
MORAVCSIK, GY. – JENKINS, R. J. H., ed. (1967): Сonstantine Porphyrogenitus. De ad-
ministrando imperio. New, Revised Edition. Greek text edited by Gy. Moravcsik.
English translation by R. J. H. Jenkins. Vol. 1, Washington.
SHCHAVELEV, A. S. (2019): Treatise De Administrando Imperio by Emperor Constan-
tine VII Porphyrogenitus: Date of the Paris. gr. 2009 Copy, Years of Compiling
of the Original Codex, and a Hypothesis about the Number of Authors. Studia
Ceranea 9, 681–704.
SHCHERBAK, A. M. (1997): Pechenezhskii yazyk, in Tenishev, E. R. (ed.:) Yazyki mira:
Tyurkskie yazyki. Moscow, 107–110.
SPINEI, V. (2003): The Great Migrations in the East and South East of Europe from the
Ninth to the Thirteenth Century, Cluj-Napoca.
TOLOCHKO, P. P. (2003): Kochevye narody stepei i Kievskaya Rusʹ. Saint Petersburg.
Late Medieval Nomads in the Bulgarian
Historical Apocalyptic Literature:
Images and Realities
TSVETELIN STEPANOV & IVELIN IVANOV
Abstract
This article aims to demonstrate some descriptive strategies of unknown Bulgarian
scribes (monks) regarding the image of late nomads (Magyars, Pechenegs, Cu-
mans). The texts under study come from the so-called historical apocalyptic litera-
ture in Bulgaria from the 11th, 12th, and 13th centuries. For the Bulgarian literati, the
peak moments in the process of perceiving the arch-enemy before the Second Com-
ing of Christ were during the 10th century (personified by the Magyars before the
year 992), and during the 11th century (personified by the Pechenegs before the
year 1092). In light of these facts, it is surprising that these Bulgarian sources do not
contain any notion of the Cumans as the arch-enemy. The authors are of the opin-
ion that the image of the Cumans in medieval Bulgaria, which possessed character-
istics of both own and other, can be classified into a specific category, which does
not fit or fall into the category of “Ishmaelites” or “Peoples of Gog and Magog”.
Keywords: medieval Bulgarian historical apocalyptic literature, imagology, Cu-
mans, Pechenegs, Uzes, Magyars/Hungarians, “Ishmaelites”
A few clarifications need to be made at the beginning of this article on how the
idea of its writing came about. It happened quite naturally and is yet another con-
firmation of the significance of forums such as the 8th International Conference
“Nomads and Their Neighbours in the Middle Ages”. They have undoubtedly
given impetus to scientific development as a whole and to the research of Pax No-
madica in its relations with the world of the Others in particular.
Tsvetelin Stepanov has posed the problem alluded to in the title of the article
numerous times before1 but has never received any kind of “feedback”. So, during
one of the sessions of the Eighth International Conference, he directly proposed to

Tsvetelin Stepanov is a Professor at St. Kliment Ohridski University, Sofia, Bulgaria; Ivelin
Ivanov is a Professor at St. Cyril and St. Methodius University, Veliko Tarnovo, Bulgaria.
1 STEPANOV 2009a, 30–42; STEPANOV 2011a, 148–63; STEPANOV 2011b, 335–51; STEPANOV 2016b.
78 TSVETELIN STEPANOV & IVELIN IVANOV
the audience to comment on the relevant parts of the presentation and to attempt
to answer the questions raised in it. Ivelin Ivanov responded with a detailed com-
mentary and it is his ideas and visions that are presented in the second part of this
short article. It should be emphasized that these suggestions and comments of Ive-
lin Ivanov are of a contextual nature and this fact should be taken into account
from the very beginning. Our hope is that future studies of and searches for apoca-
lyptic texts will shed more light on the below-mentioned discrepancies in the images
of the so-called late nomads among the Bulgarian literati during the Middle Ages.
***
The Bulgarian historical apocalyptic literature is a specific “genre” that was very
productive in medieval Bulgaria. Moreover, compared with the other Slavic-
speaking countries whose population professed Eastern Orthodox Christianity,
this “genre” had already been strongly developed in the 11 th century. Here, one
can find both historical facts (e.g., realities) and fictitious personal names, topoi/loci,
as well as peoples’ names. Some of them are mainly based on the two Testaments
and the so-called Revelation of Pseudo-Methodius of Pathara (ca. 691/2, according
to G. Reinink), the others—on real actors’ deeds dating from the late 10th to the 13th
century. By the latter we mean the real Magyars/Hungarians, Pechenegs (together
with Uzes), and Cumans, who were attacking and migrating to the lands populat-
ed by the Bulgarians in the above-mentioned period of time. In these texts some of
the intruders were labelled as “Ishmaelites”, i.e. Muslims. Naturally, it’s important
to ask whether they were actual “Ishmaelites”, i.e. real Muslims. Additionally, the
others are obviously grouped in the cliché description of “unclean peoples of Gog
and Magog” known from the Bible.
These apocalyptic texts contain a number of symbols and (proto)images often
used by the medieval Christians in order for them to establish a strictly defined
perception that it was precisely their own Empire/Tsardom that was chosen by God
and had a salvational mission for mankind before the Second Coming of Christ. As a
rule, in turbulent times some “novelties” appeared in those kind of Biblical texts—
they were usually adapted in accordance with the specific addressees (on a local
and regional level, but also on the level of the elite as a whole) as this type of litera-
ture permitted such intentional “distortions” of a given archetypal text.
Therefore, the article aims to demonstrate some of the descriptive strategies of
unknown Bulgarian scribes (in fact, monks) regarding the image of late nomads
such as the Pechenegs, Magyars/Hungarians, and Cumans.
First, some words should be said about the historical context at the time. Be-
tween the 10th and the 12th centuries the Bulgarians, firstly, had been a settled
population for a long time, and secondly, they had a Christian Tsardom starting
from 927 and up to 1018. Hence, naturally, they thought of themselves as a people
with a mission, i.e. saving the world before the Second Coming of Christ and the
End of Times. It has already been said that the lands of Bulgaria were attacked and
LATE MEDIEVAL NOMADS IN THE BULGARIAN HISTORICAL… 79
severely devastated in that same period of time by different nomadic peoples who
were crossing the Danube River from the north. It should be noted, of course, that
since the beginning of the 10th century some of the Magyars had already settled
down in the modern lands of Hungary and had started attacking the Bulgarians
from the north-west. Additionally, the Pechenegs and the Uzes in the 11th century
(starting from the late 1020s), as well as the Cumans in the last decades of that
same century entered from the north-east into the lands of Bulgaria, at that time
part of the Byzantine Empire. In the eyes of the Bulgarian scribes they were all Evil
Forces before the End of Times.
The above-said perception, namely the salvation of the world during the Last
Times and before the Second Coming, was further elaborated in Bulgaria in de-
tailed schemes. It was especially clearly seen after the unsuccessful Bulgarian up-
risings against the Byzantine Empire in 1040–1 (led by Petar Delyan) and in 1072–3
(led by Georgi Voytekh and Konstantin Bodin), i.e. on the threshold between the
11th and the 12th centuries, and, in particular, before the well-known year of 1092,
which was connected to the Anticipation of the End of Times. Generally speaking,
in the years between ca. the 1060s and 1070s and ca. 1200, a number of historical
apocalyptic texts emerged in the Bulgarian territories, mainly texts that associated
the prophecies with the names of Isaiah and Daniel of the Old Testament. As a
rule, these Bulgarian texts presented the lands of the Bulgarians as the centre of the
Christian Empire that at that time was being plundered by various “barbarian”
peoples, called at times “Vugri” (meaning Magyars/Hungarians), “Pechenegs”,
and sometimes “blonde-beards” (meaning Rus’, the Varangians in general, or the
knights of the West).
What is the situation with regard to the Evil Forces invading the Bulgarian lost
Tsardom before the End of the world that was expected to happen in 1092? It is no-
ticeable that in the Bulgarian cycle of historical apocalyptic texts the invaders be-
fore the End of Times are most often called “Ugrians”/”Vugri” or “Pechenegs”.2
The first serious clash between the Bulgarians and the Magyars was in the 890s and
more precisely in 894–6, in the early years of the reign of the Bulgarian ruler Sime-
on (893–927). During the times of his successor, Tsar Petar (927–969, †970), these
same Magyars undertook a number of further attacks, especially between the 930s
and 960s. It seems all too likely that the Magyar invasions against the Christian
people of Bulgaria were placed in the memory “mould” of the Bulgarians both in
written texts and later in folklore, since they occurred before 992 (or 1000, accord-
ing to another chronology typical for the Eastern Orthodox peoples and influenced
2TĂPKOVA-ZAIMOVA – MILTENOVA 1996, 135–6, 156, 198, 202. Also see TĂPKOVA-ZAIMOVA –
MILTENOVA 2011.
by the Byzantine culture), i.e. precisely at a time when the Second Coming of
Christ was expected.3
These events obviously had a strong impact on the Bulgarians’ way of thinking
also for another far more prosaic reason: the Magyars, invading before the End of
Times, were still pagans, which made them easily identifiable according to the
cliché "Gog and Magog attacking the Christian, i.e. Bulgarian, Tsardom" before the
expected End and the Second Coming. Moreover, the Hungarians invaded Bulgaria
from the north/north-west (!), across the Danube River (!), and were pagans (!), i.e.
they corresponded to all the indicators in the definitions of the “unclean peoples
Gog and Magog”. The Magyar invasions also continued during the years of Byzan-
tine rule over the Bulgarian lands. 4 In 1072–3, the Magyars attacked Belgrade—a
fact that is well documented in the Hungarian chronicles. It is mentioned that
Pechenegs helped the Bulgarians and Byzantines to defend the city of Belgrade,
with the following detail deserving special mention: the imperial army also includ-
ed Arab mercenaries.5
Let us now direct our attention to the aforementioned Pechenegs, the next late
nomadic tribes that invaded from the north/north-east—once more across the
Danube—into the Bulgarian lands. In the so-called Bulgarian Apocryphal Chronicle,
the first Bulgarian ruler of Danube Bulgaria, Asparukh (d. 701) (called Ispor in the
same Chronicle), is presented by the unknown scribe as a victorious ruler over the
“Ishmaelites”; later in the same text he appears as the ruler who fell in battle "on
the Danube" against those same “Ishmaelites”. 6 The specific passage goes as fol-
lows:
And then after him [Tsar Slav] another tsar was found in the Bulgarian land,
a detishte7 carried in a basket for three years; he was given the name Tsar
Ispor and [he] took over the Bulgarian Tsardom. This tsar built great cities:
on the Danube, the town of Dorostorum; a great rampart (prezid) between
the Danube and the sea; and he also built the town of Pliska. This tsar slew a
multitude of Ishmaelites. [...] Tsar Ispor ruled over the Bulgarian land for
one hundred and seventy-two years and then the Ishmaelites slew him on
the Danube.8
3 See more in, MOLLOV 1997, passim. For the purely event-related aspect of these relations,
see also MAKK 1994, 25–33.
4 See further details in DIMITROV 1998, 93–104, as well as in SHEPARD 2011, 55–83.
5 DIMITROV 1998, 95–6.
6 TĂPKOVA-ZAIMOVA – MILTENOVA 1996, 196, 199–200; PETKOV 2008, 195. For more details see
BILIARSKY 2013.
7 Bulg. detishte, denoting a child of immense size, strong and heavy for its age, from dete,
literally “child”.
8 TĂPKOVA-ZAIMOVA – MILTENOVA 1996, 199–200.
Naturally, this does not mean—as indeed is unconditionally accepted by a

number of contemporary scholars—that Ispor/Asparukh perished by the hand of a
Khazar somewhere along the Danube River,9 because the Khazars widely per-
ceived the Danube as their western border in the 670s, that is, to the point where
Asparukh was allegedly chased by them. 10 It is well known that Khazaria has nev-
er controlled the lands to the west of the Dnieper River. Neither was it possible for
Ispor to die by the hand of an actual Muslim. Perhaps another explanation is more
appropriate in this case, namely that Asparukh was presented by the anonymous
scribe of the Chronicle as the victim of imaginary invaders infringing on the Chris-
tian Empire through the well-known Danube River that served as a boundary of
the Roman and Christian civilization.
Also, it is hardly coincidental that Ispor/Asparukh is associated with the con-
struction of a “prezid”, meaning a wall or protective embankment, precisely on the
Danube river bank and reaching all the way to the Black Sea. It is true that the Bul-
gars fortified these lands with a series of ramparts and ditches, something that has
been known for a long time mainly as a result of archaeological research.11 The
allusion in this case, however, is most probably to the barrier that, according to the
legend, Alexander the Great built in the Caucasus against the “steppe barbarians”
coming from the north. Thus, it is as if the anonymous author of the Bulgarian
Apocryphal Chronicle tried to persuade his readers that Asparukh had surrounded
the lands of civilization with a wall to keep it safe from the invading “barbarians”
that inhabited the steppe north of the Black Sea. At the very end, the same text
mentions the actual Pechenegs, this time labelled as "infidels and lawless", as well
as “violators and deceivers”.12
The Pechenegs began their attacks on the Bulgarian lands in the late 1020s and
started to settle permanently in the mid-11th century.13 By then, the independent
Bulgarian Tsardom was no more since it had been conquered by the Byzantines in
1018. The Pechenegs, therefore, formally invaded Byzantine lands that were inhab-
ited by Bulgarians at that time. The direction of their invasions, as that of the Mag-
yar ones earlier on, was again from the north southwards; and again, they had to
9 On the notion of the Danube as a boundary of the Christian and, prior to that, Roman Em-
pire and civilization, see TĂPKOVA-ZAIMOVA 1976; MOLLOV 1997, 33, 35, 67–69, 102, 105, 109
and n. 9; STEPANOV 2003, 14–27; STEPANOV 2016a, 299–309; VACHKOVA 2004, 135–50. Espe-
cially on the Slavic visions and in particular those of the Rus’ on the same issue, see
PETRUKHIN 2013, 35–47.
10 On this concept, see the correspondence between Hasdai ben Shaprut and the Khazar
khagan-bek Joseph from the 950s–960s in KOKOVTSOV 1932.

11 RASHEV 1982; SQUATRITI 2005, 59–90.
12 TĂPKOVA-ZAIMOVA – MILTENOVA 1996, 198, 202; PETKOV 2008, 199.
13 On the Danube frontier during the 11 th and the 12th centuries and, in particular, on the
Pechenegs, Cumans, and Magyars along it, see MADGEARU 2013, 115–66, and also TĂPKOVA-
ZAIMOVA 1993, 95–101.
cross the Danube! The reason for this is because from a geographical point of view
during the 10th and early 11th centuries the Pechenegs inhabited the lands north of
the Black Sea and the Sea of Azov, in other words, to the north – north-west of the
Caucasus. Therefore, in the eyes of the unknown Bulgarian scribe who created the
Bulgarian Apocryphal Chronicle (according to Anissava Miltenova, it is dated to the
end of the 11th century or the beginning of the 12 th), the Pechenegs perfectly fit the
cliché of the “unclean peoples Gog and Magog” who lived in the steppe north of
the Caucasus and invaded from the north. It is obvious, then, that for the Bulgari-
ans during the decades between the 1040s, in general, and ca. 1200, the arch-evil
before the Second Coming of Christ was expected to invade the Christian Empire,
the New “Holy Land”, from the north. Thus, the Magyars and the Pechenegs ideally
suited the profile of the steppe peoples from the North, known as Gog and Magog.
The first Cuman attacks south of the Danube were carried out during the reign
of Basileus Michael VII Doukas (1071–8), i.e. in the late 1070s, and the second phase
began at the start of the reign of Alexios I Komnenos (1081–1118), i.e. again before
the well-known year 1092. During this period, the Cuman invasions were strictly
related to the large-scale Pecheneg offensive on the Balkan Peninsula that was car-
ried out during the 1080s.14 Given that situation with the apocalyptic Bulgarian
sources, it seems important to point out that the only nomadic tribes that appeared
in the Bulgarian historical apocalyptic literature under the name “Ishmaelites”
during the above-mentioned period are indeed the Ugrians/Magyars and the
Pechenegs, but not the Cumans. This means that the Byzantine and Bulgarian
sources from the time contain accounts of the Cuman invasions in South East Eu-
rope, but it also means that the unknown Bulgarian scribes of the apocalyptic texts
did not “recall” the Cumans as the invaders of the Bulgarian “Holy Land”. At the
same time, the Ugrians/Hungarians were the people that can be found in the Bul-
garian written sources and later in folklore (where they were called cherno vugre,
literally “black Vugri”, i.e. black Hungarians). Therefore, in view of this, for the
Bulgarians these same Cumans never became the arch-enemy and the apocalyptic
sign of the coming of the Antichrist.
Let us draw some preliminary conclusions from the above. For the Bulgarian
literati in general, the peak moments in the process of perceiving the arch-enemy
before the Second Coming were during the 10 th century—personified by the Ma-
gyars/Ugrians as pagans coming from the north (before the year 992), and during
the 11th century—personified by the Pechenegs (this time before the year 1092). In
view of these findings, it seems quite surprising that the Bulgarian sources prior to
the Fourth Crusade do not contain any notion of the Cumans as the arch-enemy,
although they were located north of the Bulgarians, beyond the Danube River, and
they were also pagans, or in other words, they fit perfectly into the apocalyptic
scheme of the “unclean peoples” called “Gog and Magog.“ Whether this was due
14 See STOYANOV 2005, 6–9.

to the fact that a large number of these same Cumans were assimilated rather
quickly in Bulgaria after the 1230s and even managed to lay the foundations of
some royal dynasties in the 13th and the 14th centuries, including the Terterids and
the Shishmanids, shall be discussed in the next pages.
***
No doubt, this absence of the Cumans in the Bulgarian apocalyptic sources de-
serves special attention as they are an exception to the apocalyptic images of Mag-
yars/Hungarians and Pechenegs. As for the Magyars, their image of invaders and
destroyers is far greater in scope as it is attributed to their attacks not only in the
Bulgarian lands and the Byzantine Empire but also in Central and Western Europe,
reaching even Italian territories. In this respect, the apocalyptic image of the
Pechenegs, for example, is more regional than that of the Magyars. We should also
not disregard the fact that Bulgarian-Magyar relations had a much longer history
over the centuries than the Bulgarian-Pecheneg contacts. The relations between
Bulgarians and Magyars were often hostile even in the centuries after the Chris-
tianization of the Magyars, with military conflicts in the beginning of the 13 th and
even in the 14th century.15 This also contributed to the creation of a long-lasting and
clearly evil image of the Hungarians in Bulgarian medieval literature.
Going back to the image of the Cumans, we should note that they devastated
the Bulgarian lands no less than the Pechenegs did in the 10 th and 11th centuries. It
has already been pointed out that the Cuman attacks started during the rule of
Emperor Michael IV Doukas and the second wave of raids devastated Paristrion
during the reign of Alexios I Komnenos. Raids were also recorded in the written
sources in 1095, 1114, and in 1122 (the last one probably in an alliance with the
Pechenegs). But the stability of Byzantium and the fortified borders helped the
local authorities and population build up a strong defence against the Cumans in
the second half of the 12th century. However, a historical review shows that effec-
tive Cuman attacks lasted about a century—from the second half of the 11th centu-
ry to the mid-12th century. Therefore, the answer to the above question cannot be
linked to an absence of destructive Cuman raids in Bulgarian territories.
But where should we look for a plausible and logical answer to the question of
the absence of the Cumans among the “peoples of Gog and Magog” in Bulgarian
apocalyptic literature? First of all, we would emphasize the political-dynastic rela-
tions that influenced the historical apocalyptic literature under question. The issue
of the relationship between the Cumans and the first Bulgarian dynasty ruling
after 1185, the Asenids, has been widely debated, but even the fiercest adversaries
of the hypothesis about the Cuman origin of the dynasty cannot dismiss the dynas-
15 DIMITROV 1998, 110–120, 235–9.

tic relations of Bulgarian rulers such as Kaloyan (1197–1207) and Boril (1207–18)
with the mighty Cuman clans north of the Danube.16
Even the deteriorated relations with the Cumans after 1211 under the reign of
the Bulgarian Tsar Boril did not completely interrupt the traditional ties and the
Cuman (military) influence in the Bulgarian society. 17 Last but not least, the Cuman
military power was crucial to the political establishment of the restored Bulgarian
Tsardom in the late 12th and the early 13th centuries. Although not explicitly stated
in the written sources, Cuman assistance also played an extremely important role
in the Bulgarian-Hungarian military conflicts of the abovementioned period. The
Cuman support in the clash with the knights of the newly established Latin Empire
in the period 1205–8 was also of a decisive role for the Bulgarians. 18 But some of the
written sources indicate that the Cuman influence in Bulgaria was greatly under-
mined after 1211 under Boril’s politics. Whatever the main cause of this turna-
round, its consequences were dramatic and lasting. The Mongol invasion played
the role of an additional and severe blow that marked the beginning of the decline
of Cuman influence in the European South-East, but in fact the Cuman presence in
the Bulgarian lands lasted until the very end of the Second Bulgarian Tsardom.
Vivid examples of this are the abovementioned Bulgarian dynasties of the Terter-
ids and the Shishmanids.
The resumption of Cuman activity after 1185 coincides with the uprising of the
Asenids, which turned the allied campaigns against the Byzantines south of the
Haemus Mountains (Stara Planina) into a regular, relatively secure, and extremely
lucrative occupation. At the same time, the Bulgarian-Cuman alliance protected the
population and territories between the Danube and the Haemus Mountains from
raids and devastation from the north-east and the south. What is more, the same
raids and devastation of lands occurred in Thrace as well, which benefited the aris-
16 The Cumans and their influence on the Bulgarian-Hungarian and Bulgarian-Latin rela-
tions have been the subject of research by some Bulgarian medievalists such as ZLATARSKI
1940 (1994); TSANKOVA-PETKOVA 1978; ILIEV 1984, 85–94; DANCHEVA-VASILEVA 1985; DIMI-
TROV 1998; STOYANOV 2002, 680–9; NIKOLOV 2007, 30–46; NIKOLOV 2005, 223–229; PAVLOV
1989, 9–59; PAVLOV 2002, 9–18; PAVLOV 2009, 388–409; IGNATOV 2005, 83–98; IVANOV 2017,
153–64. Also see, RASOVSKY 1939, 203–11; RASOVSKY 2016; PLETNEVA 2010; DALL’AGLIO 2008–
2009, 29–54; DALL’AGLIO 2013, 299–315. The hypothesis about the Cuman origin of the Ase-
nid dynasty could explain the close Bulgarian-Cuman alliance in the period 1185-1211, but
the events after 1211 testify to complicated and hard relations with some of the Cuman
chieftains and tribes. This indicates that the close ties of the previous period were not the
result solely of kinship relations but of a common political and military interest.
17 According to some authors, young Ivan (the future tsar of Bulgaria Ivan Asen II, 1218–41)
and his brother Alexander were taken to the Cumans in 1207 and later to the Principality of
Kiev. Whether this relocation was caused by the deteriorated relations with the Cumans
after 1211 or not is still under debate.
18 PAVLOV 1989, 9–59; NIKOLOV 2005, 223–9; DALL’AGLIO 2008–2009, 29–54.
tocracy and a part of the population north of the Haemus Mountains. The Cumans,
most probably, sold a significant portion of the booty acquired in their raids in
Thrace—this task was difficult and the prices were probably lower than the real
market value of the goods, but it was meaningless to transport them many hun-
dreds of kilometres to the Dniester and Dnieper rivers. It is also reasonable to be-
lieve that some of the captives were also left or sold in the Bulgarian territories
north of the Haemus Mountains. This practice most likely led to the accumulation
of booty and power by the local aristocracy and even by the common people in
Northern Bulgaria. That was probably yet another reason why the Cuman raids, so
violent in Thrace and other Byzantine territories, did not leave the same negative
memories in the lands north of the Haemus Mountains.
The issue of the image of the Cumans in historical sources could be analysed
from another point of view—by the methods of imagology. Here we would apply a
comparative approach to the image of the Cumans in the Chronicon Pictum (Képes
Krónika), where they occupy an important place. Chronicon Pictum depicts very
close Hungarian-Cuman relations, even within the royal Hungarian family. The
image of King Ladislaw IV the Cuman (1272–90), for example, is represented in
typical Cuman clothing, unlike other Hungarian rulers in the chronicle. 19
However, despite the close Hungarian-Cuman relations, the analysis of the tex-
tual and visual information about the Cumans in the Chronicon Pictum leads to the
conclusion that, even when they were on or near the throne the Cumans remained
part of the others. On the other hand, the image of the Cumans in medieval Bulgar-
ia can be classified into a category that possesses characteristics of both own and
other. The Cuman’s image in medieval Bulgarian apocalyptic literature does not fit
or fall into the category of “Ishmaelites”, or “Peoples of Gog and Magog”, etc. Ul-
timately, the Cumans will remain between the own and the other.
Bibliography
BILIARSKY, I. (2013): The Tale of the Prophet Isaiah: The Destiny and Meaning of an
Apocryphal Text, Leiden–Boston.
DALL’AGLIO, F. (2008–2009): The military alliance between the Cumans and Bulgar-
ia from the establishment of the Second Bulgarian kingdom to the Mongol in-
vasion, Archivum Eurasiae Medii Aevi 16, 29–54.
DALL’AGLIO, F. (2013): The interaction between nomadic and sedentary peoples on
the Lower Danube: the Cumans and the “Second Bulgarian Empire”, in Curta,
F. – Maleon, B. (ed.:) The Steppe Land and the World Beyond Them. Studies in honor
of Victor Spinei on his 70th Birthday. Iași, 299–315.
DANCHEVA-VASILEVA, A. (1985): Balgariya i Latinskata imperiya (1204–1261), Sofia.
DIMITROV, H. (1998): Balgaro-ungarski otnosheniya prez Srednovekovieto, Sofia.
IGNATOV, V. (2005): Nyakoi problemi otnosno tsar Boril. Istoriya 5–6, 83–98.
19 IVANOV 2017, 157.

ILIEV, N. (1984): Otnosno vremeto na potushavaneto na bunta sreshtu tsar Boril vav
Vidin. Izvestiya na muzeite v Severozapadna Balgariya 9, 85–94.
IVANOV, I. (2017): Svoi i chuzhdi: obrazat na ungartsi, huni, balgari, gartsi, kumani,
tatari, sarbi i vlasi v „Chronica Picta” ot XIV v., in Srednovekovnite Balkani kato
svetoven krastopat. Kontakti i obmen [Institut za balkanistika s Tsentar po trako-
logiya pri BAN; Balcanica 32], Yambol, 153–164.
KOKOVTSOV, P. (1932): Evreisko-khazarskaya perepiska v X v., Leningrad.
MADGEARU, A. (2013): Byzantine Military Organization on the Danube, 10 th–12th Cen-
tury, Leiden–Boston.
MAKK, F. (1994): Relations Hungaro-Bulgares au temps du Prince Géza et du Roi
Étienne Ier. Szegedi Bolgarisztika [Hungaro-Bulgarica, 5], Szeged, 25–33.
MOLLOV, T. (1997): Mit. Epos. Istoriya. Starobalgarskite istoriko-apokaliptichni skazaniya
(992–1092–1492), Veliko Tarnovo.
NIKOLOV, A. (2005): Cumani Bellatores in the Second Bulgarian State (1185–1396).
Annual of the Medieval Studies Department, CEU-Budapest 11, Budapest, 223–229.
NIKOLOV, A. (2007): “Gostite” na srednovekovna Ungariya: nemskata i kuman-
skata kolonizatsiya i polyusite na srednovekovnata integratsiya (XІ–XV v.), in
Kazakov, G. – Stepanov, Ts. (ed.:) Mediaevalia Christiana, Vol. 2: Srednovekoven
urbanizam: pamet–sakralnost–traditsii. Sofia, 30–46.
PAVLOV, P. (1989): Srednovekovna Balgariya i kumanite. Voennopoliticheski
otnosheniya (1186–1241 g.). Trudove na Velikotarnovskiya Universitet Sv. sv. Kiril i
Metodii, Istoriya 27, no. 3. Veliko Tarnovo, 9–59.
PAVLOV, P. (2002): Kumani i alani vav Vtoroto balgarsko tsarstvo (Niakoi paraleli v
istoricheskoto razvitie na Balgariya i Ungariya prez XІІ–XІV v.), in Balgari i un-
gartsi – 1000 godini zaedno. Budapest, 9–18.
PAVLOV, P. (2009): Polovtsi vo vnutripoliticheskoi zhizni Vizantii i Bolgarii (XІІІ–
XІV vv.). Stratum plus, 6: Prichernomorskie etiudy, 388–409.
PETKOV, K. (2008): The Voices of Medieval Bulgaria, Seventh–Fifteenth Century. The
Records of a Bygone Culture, Leiden–Boston.
PETRUKHIN, V. (2013): Rus’ v IX–X vekakh. Ot prizvaniya variagov do vybora very,
Moscow.
PLETNEVA, S. (2010): Polovtsy, Moscow.
RASHEV, R. (1982): Starobulgarski ukrepleniya na Dolniya Dunav (VII–XI v.), Varna.
RASOVSKY, D. (1939): Rol’ polovtsev v voinakh Asenei s Vizantiiskoi i Latinskoi
imperiyami (1186–1207 gg.). Spisanie na BAN 58, 203–211.
RASOVSKY, D. (2016): Polovtsi, torki, pechenegi, berendei, Moscow.
SHEPARD, J. (2011): Byzantium and the Steppe-Nomads: The Hungarian Dimen-
sion, in Shepard, J. Emergent Elites and Byzantium in the Balkans and East-Central
Europe. Farnham, UK – Burlington, USA, 55–83 (Ashgate Variorum Reprints,
no. VIII).
SQUATRITI, P. (2005): Moving Earth and Making Difference: Dikes and Frontiers in
Early Medieval Bulgaria, in F. Curta (ed.:) Borders, Barriers, and Ethnogenesis:
Frontiers in Late Antiquity and the Middle Ages. Turnhout, 59–90.
STEPANOV, Ts. (2003): Aleksandar Makedonski, Kavkaz i problemat za granitsata:
Osmislyane na prostranstvoto na tsivilizatsiyata sred volzhkite i dunavskite
balgari, in Stepanov, Ts. V sveta na srednovekovnite balgari: Mezhdu realnoto i
vaobrazhaemoto. 8 etiuda. Sofia, 14–27.
STEPANOV, Ts. (2009): Pamet i zabrava v khristiyanskiya Iztok, kr. X – kr. XII vek (Tri
apokaliptichni kazusa). Minalo no. 4, 30–42.
STEPANOV, Ts. (2011a): Memory and Oblivion in the Christian East, 990s–1200: Three
Apocalyptic Cases, in Stepanov, Ts. – Vachkova, V. – Milanova, A. (ed.:)
Memory and Oblivion in Byzantium. Sofia, 148–163.
STEPANOV, Ts. (2011b): Memory and Oblivion in the Christian East, 990s–1200: Three
Apocalyptic Cases. Archivum Eurasiae Medii Aevi 18, 335–351.
STEPANOV, Ts. (2016a): Danube and Caucasus – the Bulgar(ian)s’ Real and Imag-
ined Frontiers, in Schwarcz, A. – Soustal, P. – Tcholakova, A. (Hrsg.:) Der Do-
naulimes in der Spätantike und im Frühmittelalter [Miscellanea Bulgarica, Bd. 22].
Wien, 299–309.
STEPANOV, Ts. (2016b): V ochakvane na Kraya: Evropeiski izmereniya, ok. 950–ok. 1200 g.,
Sofia.
STOYANOV, V. (2002): Cumans in Bulgarian History (Eleventh–Fourteenth Centu-
ries), in The Turks, vol. 1: Early Ages, Part 9: Turks in East Europe. Ankara, 680–689.
STOYANOV, V. (2005): Kumanite v balgarskata istoriya (XI–XIV vek). Istoricheski
pregled 5–6, 3–25.
STOYANOV, V. (2006): Kumanologiya. Opiti za rekonstruktsiya, Sofia.
TĂPKOVA-ZAIMOVA, V. (1976): Dolni Dunav – granichna zona na vizantiiskiya Zapad (Kam
istoriyata na severnite i severoiztochnite balgarski zemi, kraya na X–XII vek), Sofia.
TĂPKOVA-ZAIMOVA, V. (1993): L’administration byzantine au Bas-Danube (fin du
Xe-XIe siècle). Byzantinoslavica 54, no. 1, 95–101.
TĂPKOVA-ZAIMOVA, V. – MILTENOVA, A. (1996): Istoriko-apokaliptichnata knizhnina
vav Vizantiya i v srednovekovna Balgariya, Sofia.
TĂPKOVA-ZAIMOVA, V. – MILTENOVA, A. (2011): Historical and Apocalyptic Literature
in Byzantium and Medieval Bulgaria, Sofia.
TSANKOVA-PETKOVA, G. (1978): Balgariya pri Asenevtsi, Sofia.
VACHKOVA, V. (2004): Traditsii na sveshtenata voina v ranna Vizantiya, Sofia.
ZLATARSKI, V. 1940 (1994): Istoriya na balgarskata darzhava prez Srednite vekove, vol. 3,
Sofia.
Shifting Foundations: The Cumans of
Paristrion and Their Influence on the
Second Bulgarian State
FRANCESCO DALL’AGLIO
Abstract
Traditionally, the historiography of medieval Bulgaria tends to emphasize the Eu-
ropean dimension of the Bulgarian polity, and its links to European political and
cultural actors. While this is by no means incorrect, this position risks to limit the
study of medieval Bulgaria to only a fraction of its extensive network of interna-
tional connections, overlooking the strong ties it maintained with the steppe world
throughout its history. This is especially true for the so-called ‟Second Bulgarian
State” and its close association with the Cumans. While the military alliance be-
tween the two polities has been explored at length and it is well documented in the
sources, less attention has been devoted to the socio-cultural interaction between
the Bulgarians and the Cumans and, in general, to the role of the mixobarbaroi in
bridging the gap between “steppe and sown” in South-Eastern Europe.
Keywords: Cuman-Qïpchaqs, medieval Bulgaria, Medieval Nomads, Byzantine
History
As a general rule (which, as every rule, comes with its fair share of exceptions) the
usual historiographical representation of medieval Bulgaria, including that of the
so-called “Second Bulgarian State” (1185–1396) originating by a rebellion of the
population of Paristrion led by the brothers Peter and Asen, is one of a stable and
organized state, perfectly aligned with contemporary models of political organiza-
tion.1 The elements of stability, such as the presence of a capital city where political
 Institute for Historical Research, Bulgarian Academy of Sciences. This paper is the revised
version of the talk given at the 8th International Conference on Medieval History of the Eur-
asian Steppe Nomads and their Neighbours in the Middle Ages, held in Sofia on 20–3 November,
2019. It is my pleasure to thank the organizers and the participants in the congress, as well
as the editors of this volume.
1 “The main protagonist of the master national narrative is the state, represented by its rulers
[…] The master national narrative shows particular interest in statehood in the sense of
90 FRANCESCO DALL’AGLIO
and religious authorities had their residence, a ruling dynasty with uninterrupted
continuity, a well-defined territorial core firmly controlled by the state, and an
ethnically homogeneous population, are always stressed and analysed at length. 2
Moreover, while the role of Bulgaria as a geographical and cultural crossroad be-
tween East and West is duly noted and appreciated, 3 the representation of medie-
val Bulgaria, limited or not to the Balkans, the Slavic world, the “Byzantine Com-
monwealth”, or Orthodox Christianity, is essentially that of a European nation. The
boundaries of the Bulgarian polity, perceived as compact, homogeneous, and
“natural”, were permanently set at the Danube (although occasional forays or
temporary expansions to the north of the river were not ignored), but undefined
and stretchable, almost at will, southward and westward.4
The elements discussed above are of no small importance for the development
of the Bulgarian medieval state, its political ideology, and its culture. It is also out
of question that the great majority of its interactions with other polities, both
peaceful and hostile, involved European actors, that its religion (leaving aside the
first two centuries of its history) was Orthodox Christianity, and that Byzantine
traditions were the main influence in both politics and culture. Yet, this discourse
is at risk of omitting, or at least neglecting, Bulgaria’s relation with the Northeast,
with the Euro-Asiatic steppe macroregion and its inhabitants. This neglect is par-
ticularly evident in the first historiographical researches devoted to the medieval
history of Bulgaria:5 once the Bulgars led by Asparukh crossed the Danube and
established their polity south of the river in 680-1, their nomadic past and their
Asiatic/Eurasian origins tend to be forgotten, as if their history really began only
when they established themselves in a world of settled and, needless to say, civi-
lized polities and populations, and as if they had left little or no traces in the Bul-
garian polity, which was represented as essentially Slavic and Christian. In his
Istoriya Slovenobolgarskaya, completed in 1762, Paisiy Hilendarski omits any men-
tion of a non-European origin of the Bulgars: according to him they were a Slavic
people whose ancestral land was located along the Volga river. They crossed the
Danube in 328 and, after a period in which they were loyal subjects of the Byzan-
institutions and state traditions, as well as in the church”: DASKALOV 2020, 270. Cf. also
DASKALOV 2018. The best and most influential example of this general attitude is Vasil
Zlatarski, who entitled his seminal work History of the Bulgarian State during the Middle Ages
(emphasis and translation mine). Cf. ZLATARSKI 1918–40.
2 See, for instance, the classical treatments of ANGELOV 1971; PETROV 1981.
3 For an analysis of medieval Bulgaria’s in-betweenness, see especially STEPANOV 2020 and
the extensive bibliography provided in the text.

4 On the subject of frontiers (political, cultural, geographical) in the Middle Ages, see espe-
cially BARTLETT – MACKAY 1989; POWER – STANDEN 1999; DESPLAT 2000; ABULAFIA – BEREND
2002; MERISALO – PAHTS 2006.
5 DETCHEV, 2013.
SHIFTING FOUNDATIONS: THE CUMANS OF PARISTRION AND THEIR INFLUENCE… 91
tine empire, they created a state of their own. 6 Both Paisiy’s omission and his con-
flation of the Slavic and Bulgar migrations were justified, since the few sources he
could access offered him no insight on the matter, and he had neither the training
nor the education to be considered a historian. The belief that the Bulgars were
Slavs was also expressed in the first half of the 19 th century by Yuriy Venelin—
again an author that was only an amateur historian. 7 Marin Drinov, on the other
hand, who certainly was a historian in the technical sense of the word, and who
could profit form a much larger body of sources and secondary literature, did not
ignore the non-Slavic origins of the Bulgars, but according to his interpretation
they could not be considered proper ancestors of the modern Bulgarians, since
their contribution to the culture and the ethnicity of the local Slavs was negligible. 8
The same opinion was maintained by Konstantin Jireček a few years later.9
In the beginning of the 20th century, this approach rapidly changed and the
Eurasian roots and culture of the first Bulgarian state have since been thoroughly
analysed and detailed, with particular regard to the establishment of the state, its
relations with its neighbours (chiefly with the Byzantine Empire), its internal or-
ganization and composition, and the relations between Bulgars and Slavs. 10 How-
ever, the nomadic, and specifically Cuman, contribution to the establishment of the
Second Bulgarian State is usually studied only in its military aspects; 11 the Cumans
are considered, just as the medieval chroniclers represented them, as radically dif-
ferent from the Bulgarians. Their impact on the Bulgarian state (now truly and
fully European, in its relations with crusaders, Roman pontiffs and emperors of
Constantinople), aside from the aforementioned military assistance, either goes
unnoticed or is presented in confrontational, even destructive terms, 12 with little
attention to the interactions between sedentary people and nomads, and little con-
cern for the development of the Cumans’ own culture, and of nomadic civilization
in general. Just as in many reconstructions of Byzantine history Bulgaria is seen as
an accessory, in many reconstructions of Bulgarian history this is the role played
6 HILENDARSKI 2012, 80–5.

7 VENELIN 1829; VENELIN 1841.
8 DRINOV 1971, 81, 121–2 (first published in 1869).
9 JIREČEK 1876, 136–8.
10 See, for example, BEŠEVLIEV 1980; BEŠEVLIEV 2008; BOŽILOV 1992; BOŽILOV 1995; NIKOLOV
2005; STEPANOV 1998; STEPANOV 1999; STEPANOV 2005; STEPANOV 2010; STEPANOV 2020. On the
remembrance of the nomadic and pagan past during the First Bulgarian State, see NIKOLOV
2011; KAYMAKAMOVA 2011, 43–101.
11 Тhe classical treatment of the military alliance between Bulgarians and Cumans is RASOV-
SKY 1939. In the past, I have also approached the matter exclusively from this (limited) point
of view: DALL’AGLIO 2008–2009.
12 For the supposed existence of a “Cuman party” that hindered the development of the
Bulgarian state, see note 45 below.

by the Cumans.13 One of the main reasons for this omission is not a lack of will on
the scholars’ side, but rather the fact that there are very few literary sources or ar-
chaeological remains that can shed some light on the subject, that many of the ref-
erences found in the chronicles are indirect, ambiguous, biased, or uninformed. But
were the Cumans, and their customs, so alien and perplexing to the Bulgarians as
they were to a “poor” French knight, or to a refined Byzantine intellectual, who
considered barbaric even the periphery of Constantinople? 14
Nikethas Choniates, the main source for the establishment of the second Bulgar-
ian state, makes no reference to the Cumans in his account of the beginnings of the
rebellion led by the brothers Peter and Asen. He mentions Cuman incursions in the
imperial territories during the 11th and the 12th centuries, calling them by the anti-
quated name of “Scythians” and not differentiating between them and the other
northern nomads, to the point that at times it is difficult to understand precisely
which ethnic group he refers to. The creation of the new state (or the restoration of
the old Bulgarian state, as Choniates himself writes15) is the deed of the local
Vlachs and Bulgarians, and no Cuman presence or influence is mentioned for the
first phase of the insurrection. But when Emperor Isaac II Angelos organized a
resolute counter-attack in 1186, Peter and Asen were forced to flee across the Dan-
ube.16 Once there, with a passionate speech Peter convinced the Cumans to join his
side,17 promising a rich and easy plunder because he was certain that the emperor
would return as soon as possible to Constantinople, leaving only few troops in
Paristrion. The Cumans, characterized according to the classical Byzantine repre-
sentation of the nomadic northerners as greedy, cowardly, and not particularly
clever, accepted Peter’s offer with enthusiasm, and with their support he was able
to cross back over the river, recover the land lost to the emperor, and secure the
survival of his dominions.18 His speech was obviously made up retrospectively by
Choniates, following the canons of ancient Greek historiography, and the aim of
13 For some insightful exceptions to this trend, see TĂPKOVA-ZAIMOVA 1976; PAVLOV 1989;
PAVLOV 1990; NIKOLOV 2005; STOYANOV 2005; STOYANOV 2009; GOLEV 2013; GOLEV 2018a;
GOLEV 2018b; VLADIMIROV 2018.
14 The “pobre chevalier”, according to his own definition, is Robert De Clari, whose descrip-
tion of the Cuman customs is rightly famous: DE CLARI 2004, § 65. Nikethas Choniates is the
Byzantine writer whose chronicle is the most detailed source for the establishment of the
second Bulgarian state: CHONIATES 1975. On the medieval perception of the steppe world
and its inhabitants, see especially AHRWEILER 1998; SARDELIĆ 2019; CHEKIN 1992; on migra-
tions and mobility in the Byzantine world, see STOURAITIS 2020.
15 CHONIATES 1975, 374.
16 CHONIATES 1975, 371–2.
17 It must be pointed out (this is another usual flaw in the historiography of the Bulgaro-
Cuman relations) that it was not the Cumans as a whole but only certain clans that joined the
Bulgarian side: see PAVLOV 1989, 18-20, with extensive bibliography.
18 CHONIATES 1975, 373–4; CHONIATES 1972, 7–8.
the chronicler was not to provide an accurate description of the events but rather to
highlight Isaac’s lack of combative spirit. Nonetheless, he is correct in representing
the Cumans as willing, even eager to assist the Bulgarians in their enterprise
(which, notwithstanding what Choniates had Peter say, was not easy at all, as both
the Bulgarian ruler and the Cumans most certainly knew). This willingness to
commit their armies and their resources, after so many years of peaceful relations
with Constantinople, had different causes than the rhetorical abilities of Peter and
the supposed greed of the nomads. The Bulgarian proposal turned the attention of
some Cuman chieftains and clans towards the territories south of the Danube, an
area that had been marginal to their interest for many years. In the second half of
the 12th century their attention gravitated mostly towards the Dnieper and the rich
commercial settlements of Crimea, and further expansion in the neighbouring re-
gions had been stalled.19 Just two years before Peter’s proposal, their armies had
suffered a heavy defeat at the hands of the Rus’, 20 so the opportunity to compen-
sate for their losses was surely welcomed.
It would be interesting to know if the Cumans already stationed south of the
Danube were instrumental in brokering the alliance between the Bulgarians and
their fellow clansmen, if they facilitated or suggested it, or if they thought it was in
their best interest. It would be also interesting to know if they had retained their
old customs, or if they had adapted to a way of life more similar to that of their
settled neighbours. What we know for certain is that since the end of the 11 th cen-
tury groups of Cumans had moved in the imperial territory and that most of them
were employed as military auxiliaries by the Byzantine army. Their presence (alt-
hough we cannot be sure about the actual number of Cumans who settled in the
region) contributed to further diversification of the local population of Paristrion,
which had already experienced a constant afflux of new settlers, especially trans-
humant Vlachs,21 after the period of the great Pecheneg and Oghuz invasions that
had depopulated the region. The sources paint a vivid picture of the network of
cities, villages, and fortresses dotting the region in the 11th century, “populated by
a multitude of people speaking all kind of languages.”22 The Pechenegs and Cu-
mans in Byzantine service are called in the sources, regardless of their actual ethnic
origins, with the derogatory name of mixobarbaroi, a reference to their “barbaric”
nature only partially tamed by Byzantine customs, which placed them halfway be-
19 PAVLOV 1989, 13; SPINEI 2009, 140–1; GOLEV 2018B, 101–3, with extensive bibliography.
20 PSRL, vol. 2, 19622, 631–3.
21 STANEV 2013, 215–17.
22 ATTALEIATES 2011, 158. On Paristrion and its inhabitants in the 11 th and 12th century see
especially STEPHENSON 1996; STEPHENSON 2000; TŬPKOVA-ZAIMOVA 1976; TŬPKOVA-ZAIMOVA

2010a; TŬPKOVA-ZAIMOVA 2010b; MADGEARU 1999a; MADGEARU 1999b; MADGEARU 2013;
CURTA 2006, 293–9, 302, 314, 319; BOŽILOV – GJUZELEV 2004, 102–24; BONAREK 2007;
DALL’AGLIO 2013.
tween the sedentary and nomadic world, between civility and savagery. 23 As noted
by V. Tŭpkova-Zaimova,24 they tended to mingle with the locals, i.e. Bulgarians
and Vlachs, and to use the local language rather than Greek, which was not spoken
in the region apart from the elites and the clergy. At first rewarded for their service
with money, gifts, and honorific titles, they began to receive plots of land after the
institution of the pronoia system and became not only consumers, but also produc-
ers of agricultural goods and breeders of livestock while continuing to serve as
intermediaries, along with other local and foreign actors, for the goods exchanged
between Constantinople, Paristrion itself, and the territories to its north. They be-
came integrated in the network of Paristrian productive activities, and their num-
ber, while never particularly high, increased steadily.25 Only minor Cuman raids
are reported in Paristrion after the late 1150s, a further proof of the fact that peace-
ful relations appeared now more lucrative. Due to the lack of major incursions, the
second half of the 12th century saw a demographic increase in the region, 26 which
consequentially led to an economic expansion 27 and to the rise of a new class of
local magnates of various ethnic origins: Bulgarian, Vlach, or mixobarbaroi.
This relatively prosperous situation, however, did not last long, especially after
the death of Manuel I Komnenos in 1180. The reasons for this are manifold. The
end of the Cuman incursions and the parallel tensions with Hungary, Serbia, and
the Normans led to the gradual decommissioning of the Byzantine fortresses on
the Danube since the military personnel was needed elsewhere. 28 This in turn re-
duced the trade opportunities for the local producers, whose goods could not be
sold on the rich markets of Thessalonica or Constantinople due to the competition
with other regions of the empire closer to those markets and better served by roads
23 ATTALEIATES 2011, 158; STǍNESCU 1965; TANAȘOCA, 1973; TŬPKOVA-ZAIMOVA 1976, 126–31;
TŬPKOVA-ZAIMOVA 1979; AHRWEILER 1998, 10–15; MADGEARU 2013, 87–8; STEPHENSON 2000,
109–14; ZHEKOVA 2020.
24 TŬPKOVA-ZAIMOVA 1976, 130.
25 GOLEV 2018b, 108; RASHEV 2007. On the institution of pronoia, see especially BARTUSIS 2012,
94–7; BIRKENMEIER 2002, 148–68; LAIOU, 1977, 142–58; LEMERLE 1979, 166–88, 230–48; KA-
ZHDAN 1952, 202–3; KAZHDAN 1995, with a comprehensive bibliography. On the Cuman
pronoiars, see especially OSTROGORSKI 1977; ZHEKOVA 2020. Niketas Choniates was extremely
critical of the idea of giving military command to half-barbarians and other people that are
inadequately experienced (according to him) and of low social status, because in his opinion
this weakened the army: CHONIATES 1975, 208–9; see also AHRWEILER 1998, 2–3, for other
examples of Byzantine writers expressing their disgust at the concept of Roman soldiers
being led by barbarians.
26 HARVEY 1989, 55–66; STANEV 2013.
27 HARVEY 1989, 87–8; LAIOU 2012, 145–6; OBERLÄNDER-TÂRNOVEANU, 1979.
28 BOŽILOV – GJUZELEV 2004, 113–114; CURTA 2006, 334–5, 339, 346–7; FINE, JR. 1994, 6–9; STE-
PHENSON 2000, 279–84; MADGEARU 2017, 32–3.

and infrastructure, such as Thrace or Macedonia.29 To make matters worse, the tax
burden became increasingly oppressive. The tipping point was reached in the early
autumn of 1185, the unintended consequence of the marriage arranged between
the new emperor Isaac II and Margaret, the daughter of King Béla III of Hungary.
Isaac decided to pay the expenses of the ceremony from his estates in Thrace, but
apparently the tax collectors were overzealous and confiscated large amount of
cattle from other regions of the empire, especially from the vicinities of Anchi-
alos.30 The town was an important commercial outlet where Paristrian goods were
traded, so it seems logical to suppose that those who were taxed in excess and saw
their cattle confiscated were not the locals but the traders who brought their goods
there to be sold locally or shipped to Constantinople. This interpretation is corrob-
orated by the later testimony of Akropolites. According to his chronicle, the requi-
sition was particularly hard on the Bulgarians: ‟sheep, pigs, and oxen were collect-
ed from every province of the Roman empire. But since the land of the Bulgarians
rears more of these than do other places, more animals were also demanded from
it.”31 It may have been particularly hard on the Cumans as well if they were among
those who brought their cattle to Anchialos, and we may suppose they were, alt-
hough Akropolites does not mention them specifically.
The requisition increased the dissatisfaction of the Paristrians, and it was in this
context that Peter and Asen appeared at Kipsella, where Isaac II was preparing the
expedition against the Normans who had taken Thessalonika. 32 We ignore whether
they went to Kipsella for personal reasons (according to Choniates, they asked to
join the imperial army in exchange for a small plot of land and a small revenue) or
to voice the discontent of the Paristrians. In any case, their request was denied and
they returned home angered and ready to take vengeance. The same Choniates,
however, writes that the decision of seceding from the empire had already been
taken, and that they were just looking for a pretext. It has been speculated that,
since they were able to access the emperor, they were already at the service of the
empire, maybe even pronoiars or anyways men of great standing among the local
population,33 and that their main occupation was horse breeding; 34 and although
29 On grain production in Thrace, see HARVEY 1989, 139–40. See also LAIOU 2012, 125–46,
especially p. 130 for the 11th–12th century trade networks between Thrace and Constantino-
ple, p. 137 on Thessalonika; ASDRACHA 1976, 180–230.
30 CHONIATES 1975, 368.
31 AKROPOLITES 1978, 18 (English translation: MACRIDES 2007, 133). For an assessment of the
number of animals possessed by large estates, see HARVEY 1989, 153–7, although the exam-
ples provided date mainly from the 11th century.
32 CHONIATES 1975, 368–9.
33 MALINGOUDIS 1980, 83–5; SIMPSON 2016, 6–7; MADGEARU 2017, 41–3; TŬPKOVA-ZAIMOVA
1976, 131; ZLATARSKI, vol. 2, 435–9; FINE, JR. 1994, 10.

34 On the basis of CLARI § 64, where the initiative of the rebellion is assigned to Kaloyan (not
to his older brothers) who, according to the chronicle, was “un serjans l’empereur qui
Choniates is consistent in calling them “Vlachs” it is also possible that they were
mixobarbaroi, possibly of Cuman ancestry, 35 so despised by Choniates and other
Byzantine intellectuals.
While this is certainly possible, especially in a border area where many different
populations lived side by side with no clearly defined ethnic boundaries, with all
the obvious intermingling that this proximity caused and favoured, it must be
pointed out that all the available sources, Byzantine or Latin, unanimously consid-
er Peter, Asen, and Kaloyan to have been Vlachs or Bulgarians, not Cuman. More-
over, the Cumans were always considered a different people than the Bulgaro-
Vlachs and their allies. Akropolites, while not particularly attentive towards the
ethnic differences separating the enemies of the empire (like the vast majority of
the Byzantine writers), even writes that Kaloyan became more savage when he
married a Cuman princess. He exchanged his previous nature of Bulgarian, which,
while extremely violent, was at least more civilized, for that of the Cumans: “since
he had won over the Scythian race and was associated with them by kinship and
partook of their habits which were bestial by nature, he delighted in the murder of
Romans.”36 Choniates also differentiates between Bulgarians and Cumans, 37 but
his disdain and bias against the Asenids, the same disdain and bias he had for the
mixobarbaroi, is more evident, and he makes no effort to hide it especially when
presenting the actions of Asen and Kaloyan. In his eyes, those populations are dif-
ferent and separated, but both participate in various degrees of barbarism. The
Cumans, being nomads and pagans, are demon-worshippers: “they flogged some
of their prisoners, outstanding for their beauty, before hanging them as a sacrifice
wardoit une huiriere l’empereeur; si que quant li empereres mandoit soisante chevaus ou
chent, que chis Jehans [Kaloyan] li envoioit”. The Anatolian plain had been the most im-
portant horse breeding ground for military use, where the greater number of imperial stud
farms was located. According to HARVEY 1989, 152, “its loss in the 1070s probably caused
problems in the supply of horses to the army”. See also HENDY 2008, 54–6. We may suppose
that as a consequence of this loss the importance of Paristrian horse breeding increased
dramatically.
35 On the various theories about the ethnicity of the Asenids, see BOŽILOV 1994, 18–19.
ZLATARSKI 1933 supposes that Peter and Asen were chosen as leaders of the uprising because
of their Cuman descent since the Bulgarians needed to secure the help of their northern
neighbours. According to ANGELOV 1987, 48; DUJČEV 1945, 50; and MALINGOUDIS 1980, 85–7,
they were of Bulgaro-Cuman descent. PRITSAK 1982, 373, maintains not only that the Cu-
mans led the uprising, but also that all the subsequent Bulgarian dynasties until the Otto-
man conquest were of Cuman origin: “They [the Cumans] led the Wallachian-Bulgarian
insurrection against Byzantium which led to the formation of a second Bulgarian kingdom
with three Polovcians dynasties: Asen (1185–1280), Terter-oba (1280–1323) and Sisman
(1323–1396)”; similarly, according to VÁSÁRY 2005, 41, “the Asenids were a Cuman dynasty
whose members became Vlakhs in the twelfth century and Bulgars in the thirteenth.”
36 AKROPOLITES 1978, 24 (English translation: MACRIDES 2007, 140).
37 See, for instance, CHONIATES 1972, 3–6.
to their demons.”38 But the Bulgarians are described, at times, as being just as
fiendish: the seers who prophesize the good fortune of the insurrection are “pos-
sessed by demons”,39 Peter and Asen escape the Byzantine army throwing them-
selves in the Danube ‟like, in the Gospel, the flock of pigs in the sea”, 40 and when
Kaloyan took Varna in 1203, ‟the barbarian was not impressed by the sanctity of
the day (since it was the most holy Saturday, in which Christ slept in the Sepul-
chre) and had no respect for the Christian name, which he only honoured with his
lips, but was driven by bloodthirsty demons”.41
De Clari is the source in which the difference between Bulgarians and Cumans
is most evident. He devotes an entire chapter of his chronicle to the outlandish (for
him and for his audience) habits of the Cumans, 42 stressing the fact that they were
“dressed in animal skins”.43 At the same time, he does not feel the need to com-
ment on the habits of the Bulgarians, which evidently conformed to his standards
of normality; and when he describes the meeting between Kaloyan and Pierre de
Bracheux his representation of the Bulgarian tsar is that of a king perfectly able to
behave according to the chivalric code of conduct.44
The same diffidence towards the Cumans, and the assumption of their some-
what malignant alterity, has also at times surfaced in the Bulgarian historiography
of the past. A case in point is, again, Zlatarski, who has ascribed the dynastic trou-
bles of the first Asenids and the violent deaths of Asen and Kaloyan to the deeds of
an unspecified ‟Cuman party” whose motivations remain, however, unexplored in
his analysis.45 Needless to say, the sources do not support this interpretation: leav-
38 CHONIATES 1975, 618.

39 CHONIATES 1975, 371.
40 CHONIATES 1975, 372.
41 CHONIATES 1975, 533.
42 DE CLARI 2004, § 65.
43 DE CLARI 2004, § 112.
44 DE CLARI 2004, § 106.
45 According to ZLATARSKI, vol. 3, 94–101, no internal opposition existed in Bulgaria, since
Asen and the Bulgarian aristocracy shared the same goals. According to him, any discording
voice was external, or suborned by an external force. Ivanko, the killer of Asen in 1195, was
therefore the instrument of a “Cuman party,” expression of the Cuman nobility dissatisfied
by Asen’s apparent lack of consideration. Thus, Zlatarski pointed out at a rivalry between
Bulgarians and Cumans that, according to him, will continue in the following years and
bring also to the death of Kaloyan, killed in 1205 while besieging Thessalonika at the hand
of one of his Cuman generals, Manastras, acting on behalf of the “Cuman aristocratic party”
(“куманската болярска партия”, 260). PAVLOV 1995, 56, rightly considers that a “Cuman
party” may have well existed, but only at the time of the Terterid dynasty, whose founder
George I Terter (r. 1280–92) was of Cuman descent. He also considers (p. 36) that even if
Kaloyan was killed by the Cuman Manastras, this was not the result of a plot organized by
the Cumans, but rather because of their “Bulgarization”, i.e. their involvement in the fac-
tionalism of the Bulgarian aristocracy. They were not the maneuverers, but the maneuvered.
ing aside the fact that it makes no sense to consider all the Cumans as a single po-
litical entity with the same goal, they continued to cooperate with the Bulgarian
State even after the deaths of the sovereigns allegedly killed by them, something
which hardly seems possible if they had reasons to hold the Bulgarians in con-
tempt.
Peter came from the north in 1186 as Asparukh did in 681, both crossing the
Danube, both at the head of an army of nomads, both establishing an independent
polity with their help. The analogy, of course, ends here (even if it probably did not
escape Choniates, who was after all the first one to point out that the aim of the
rebels was to rebuild the old Bulgarian state “as it had been in the past”). 46 While
Asparukh was leading his own people, Peter was bringing a “foreign” army of
allies. The Cumans remained a distinct entity with their own customs, traditions,
and political interests; Bulgaro-Vlachs and Cumans did not create a unified polity,
and after a period of collaboration, long and fruitful as it was, they went their sep-
arate ways, even if the contacts were never interrupted. Apparently, Peter and
Asen did not include any of the Cuman traditions in their political ideology, which
emphasized the elements of continuity between their polity and the First Bulgarian
State, presenting themselves as the legitimate successors of the Bulgarian rulers
and their country as the revival of the old and glorious Bulgarian state. 47 But was
this continuity true in all its elements? Is it possible to consider the Paristrian
community, even in its predominantly Bulgarian cultural features, as the faithful
reproduction of the old state that was absorbed by the Byzantine Empire in the
beginning of the 11th century?
The answer, of course must be a negative one: and not only for the obvious rea-
son that more than 150 years had passed between the two polities. Even more im-
portant was the cultural, rather than purely ethnical, shift that the region had wit-
nessed, in which sedentary, nomadic, and transhumant elements concurred in the
development of a new identity for the Paristrian community, whose features were
partly borrowed from an imperfectly remembered past, and partly from an auton-
omous and recent creation. This mixed, intertwined, diverse community formed
the basis of the second Bulgarian State, and while its rulers were fully justified in
searching for their legitimization in the memory of the old state, and the Bulgarian
ethnic element was willing to revive its old traditions and recreate its polity “as it
had been in the past,” the Paristrian community as a whole was more than that,
and its roots were also those of the mixobarbaroi that participated in the revolt be-
The same can be said of the battle fought between the Hungarian army and some Cuman
chieftains who, according to an erroneous tradition, took control of Vidin during Boril’s
reign: see NIKOV 1912; ILIEV 1984; for a more correct assessment of the episode, see VÁSÁRY
2005, 58–60.
46 CHONIATES 1975, 374.
47 On the political use of the memory of the first Bulgarian State made by Peter and his suc-
cessors, see DALL’AGLIO 2018; DALL’AGLIO 2020.

cause they were a part of their present community rather than because they want-
ed to revive a state whose memory was alien to them. Being aware of this influence
can help us in framing Bulgaria, and the European Middle Ages as a whole, in a
network of interconnections spanning across the continent and extending outside
of the usual regions of interest and the usual distinctions between Europe and
Asia, and sedentary societies and nomads, abandoning the traditional view of mu-
tually exclusive worlds whose only interactions were violent or menacing. “A
postcolonial Middle Ages has no frontiers, only heterogenous borderlands with
multiple centres. This reconfigured geography includes Asia, Africa, and the Mid-
dle East not as secondary regions to be judged from a European standard […] but
as full participants in a world simultaneously larger and more fragmented—a
world of intersecting, mutating, incommensurable times and places.” 48
Bibliography
ABULAFIA, D. – BEREND, N. (2002): Medieval Frontiers: Concepts and Practices,
Aldershot.
AHRWEILER, H. (1998): Byzantine Concepts of the Foreigner: The case of the
Nomads, in Ahrweiler, A. – Laiou, A. (ed.:) Studies on the Internal Diaspora of the
Byzantine Empire. Washington DC, 1–15.
AKROPOLITES (1978): Georgii Acropolitae Opera, ed. Heisenberg, A. – Wirth, P.,
Stuttgart.
ANGELOV, D. (1987): Vastanieto na Asenevtsi i vazstanovyavaneto na srednovekov-
nata balgarska darzhava. Palaeobulgarica 9, 37–58.
ANGELOV, D. (1987): Obrazuvane na balgarskata narodnost, Sofia.
ASDRACHA, C. (1976): La région des Rhodopes aux 13e et 14e siècles. Étude de géographie
historique, Athens.
ATTALEIATES (2011): Michaelis Attaliatae Historia, ed. Tsolakis, E. Th., Athens.
BARTLETT, R. – MACKAY, A., ed. (1989): Medieval Frontier Societies, Oxford.
BARTUSIS, M. (2012): Land and Privilege in Byzantium: the Institution of Pronoia,
Cambridge.
BIRKENMEIER, J. (2002): The Development of the Komnenian Army: 1081-1180, Leiden.
BEŠEVLIEV, V. (1980): Die protobulgarische Periode der bulgarischen Geschichte,
Amsterdam.
BEŠEVLIEV, V. (2008): Parvobalgarite. Istoria, bit i kultura, Plovdiv.
BONAREK, J. (2007): Le Bas Danube dans la seconde moitié du XIème siècle : nou-
veaux états ou nouveaux peoples ? Byzantina et Slavica Cracoviensia 5, 193–200.
BOŽILOV, I. (1992): Razhdaneto na srednovekovna Balgariya (nova interpretatsiya).
Istoricheski pregled 1–2, 3–34.
BOŽILOV, I. (1994): Familiyata na Asenevtsi (1186-1460). Genealogiya i prosopografiya,
Sofia.
48 COHEN 2000, 7.
BOŽILOV, I. (1995): Sedem etyuda po srednovekovna istoriya, Sofia.

BOŽILOV, I. – GJUZELEV, V. (2004): Istoriya na Dobrudzha, Varna.
CHEKIN, L. S. (1992): The Godless Ishmaelites: the Image of the Steppe in the
Eleventh-Thirteenth-Century Rus. Russian History 19, 9–28.
CHONIATES, N. (1972): Nicetae Choniatae Orationes et Epistulae, ed. van Dieten, J. L.,
Berlin/New York.
CHONIATES, N. (1975): Nicetae Choniatae Historia, ed. van Dieten, J. L., Berlin/New
York.
COHEN, J. J. (2000): Midcolonial, in Cohen, J. J. (ed.:) The Postcolonial Middle Ages,
New York.
CURTA, F. (2006): Southeastern Europe in the Middle Ages, 500–1250, Cambridge.
DALL’AGLIO, F. (2008-2009): The Military Alliance Between the Cumans and
Bulgaria from the Establishment of the Second Bulgarian Kingdom to the
Mongol Invasion. Archivium Eurasiae Medii Aevi 16, 29–54.
DALL’AGLIO, F. (2013): The Interaction between Nomadic and Sedentary Peoples on
the Lower Danube: the Cumans and the ‘Second Bulgarian Empire’, in Curta, F
– Maleon. B.-P (ed.:) The Steppe Lands and the World Beyond Them. Studies in
Honor of Victor Spinei on His 70th Birthday. Iaşi, 299–312.
DALL’AGLIO, F. (2018): ”As it Had Been in the Past”: the Idea of National
Continuity in the Estаblishment of the Second Bulgarian Kingdom, in
Biliarsky, I. (ed.:), Laudator Temporis Acti: Studia in Memoriam Ioannis A. Božilov.
Sofia, vol. 1, 282–299.
DALL’AGLIO, F. (2020): The Second Bulgarian Kingdom as an «Imagined Com-
munity» and as a Community of Memory, in Balgariya i balgarite: bit,
dushevnost, natsionalna identichnost. Shumen, 117–124.
DASKALOV, R. (2018): Golemite razkazi za Balgarskato srednovekovie, Sofia.
DASKALOV, R. (2018): Historical Master Narratives and the Master Narrative of the
Bulgarian Middle Ages. Studia Ceranea 10, 259-280.
DE CLARI, R. (2004): La conquête de Constantinople, ed. Dufournet, J., Paris.
DESPLAT, C., ed. (2000): Frontières. Actes du 125e Congrès national des sociétés histo-
riques et scientifiques. Lille.
DETCHEV S. (2013): Between Slavs and Old Bulgars: “Ancestors,” “Race” and
Identity in Late Nineteenth-Century, in Gery, P. - Klaniczay, G. (ed.:)
Manufacturing Middle Ages. Entangled History of Medievalism in Nineteenth-
Century Europe. Leiden, 347–376.
DRINOV, M. (1971): Pogled varkhu proizkhozhdenieto na balgarskiya narod i
nachaloto na balgarskata istoriya, in Drinov, M., Izbrani sachineniya, vol. 1, Sofia.
DUJČEV, I. (1945): Vastanieto na Asenevtsi i kultat na sv. Dimitar Solunski. Spisanie
na Balgarskata akademia na naukite 41, 44–51.
FINE, JR., J. V. A. (1994): The Late Medieval Balkans. A Critical Survey from the Late
Twelfth Century to the Ottoman Conquest, Ann Arbor.
GOLEV, K. (2013): Desht-i Kipchak i Vneshnii mir. Tipologiya kontaktnykh zon
SHIFTING FOUNDATIONS: THE CUMANS OF PARISTRION AND THEIR INFLUENCE…101
mezhdu kumano-kipchakskoi obshchnostyu i vneshnego mira, in Kumekov, B.

E. (ed.:) Sbornik materialov mezhdunarodnoi nauchnoi konferentsii ‛Kipchaki Evrazii:
iztoriya, yazyk i pismennye pamyatniki’ posvyashchennoi 1100 letiyu Kimekskogo
gosudarstva v ramkakh Dnei tyurkskoi pisʹmennosti i kulʹtury. Astana, 142–151.
GOLEV, K. (2018a): The Bulgarophilia of the Cumans in the Times of the First
Asenides of Bulgaria. Golden Horde Review 6/3, 452–471.
GOLEV, K. (2018b): On the Edge of ‟Another World”: The Balkans and Crimea as
Contact Zones Between the Cuman-Qïpchaqs and the Outside World. Études
Balkaniques 54/1, 89–126.
HARVEY, A. (1989): Economic Expansion in the Byzantine Empire, 900–1200,
Cambridge.
HENDY, M. F. (2008): Studies in the Byzantine Monetary Economy c. 300–1450,
Cambridge.
HILENDARSKI, P. (2012): Istoriya Slavyanobalgarska. Kritichesko izdanie s prevod i
komentar, ed. Dobrev, I. – Peev, D. – Dimitrova, M. – Petkov, P. – Nikolov, A.,
Zograf.
ILIEV, N. (1984): Otnosno vremeto na potushavane bunta sreshtu tsar Boril vav
Vidin. Izvestiya na muzeite v Severozapadna Balgariya 9, 85–94.
JIREČEK C. (1876): Geschichte der Bulgaren, Prague.
KAYMAKAMOVA, M. (2011): Vlast i istoriya v srednovekovna Balgariya, VII-XIV vek,
Sofia.
KAZHDAN, A. (1952): Agrarnye otnosheniya v Vizantii, Moscow.
KAZHDAN, A. (1995): Pronoia: the History of a Scholarly Discussion. Mediterranean
Historical Review 10, 133–163.
LAIOU, A. (1977): Peasant Society in the late Byzantine Empire, Princeton.
LAIOU, A. (2012): Regional Networks in the Balkans in the Middle and Late
Byzantine periods, in Morrisson, C. (ed.:), Trade and Markets in Byzantium,
Washington DC, 125–46.
LEMERLE, P. (1979): The Agrarian History of Byzantium from the Origins to the Twelfth
Century, Galway.
MACRIDES, R. (2007): George Akropolites: The History. Introduction, Translation and
Commentary. Oxford.
MADGEARU, A. (1999a): Dunărea în epoca bizantină (secolele X-XII): o frontieră
permeabilă. Revista Istorică 10, 41–55.
MADGEARU, A. (1999b): The Military Organization of Paradunavon. Byzantino-
slavica 60/2, 421–446.
MADGEARU, A. (2013): Byzantine Military Organization on the Danube, 10-12th
Centuries, Leiden.
MADGEARU, A. (2017): The Asanids. The Political and Military History of the Second
Bulgarian Empire (1185–1280), Leiden.
MALINGOUDIS, PH. (1980): Die Nachrichten des Niketas Choniates über die
Entstehung des zweiten bulgarisches Staates. Βυζαντινά 10, 51–147.
MERISALO, O. – PAHTS, P., eds. (2006): Frontiers in the Middle Ages. Proceedings of the
Third European Congress of Medieval Studies (Jyväskylä, 10-14 June 2003), Louvain-
la-Neuve.
NIKOLOV, A. (2011): The Perception of the Bulgarian Past in the Court of Preslav
Around 900 AD, in Gjuzelev, V. – Petkov, K. (ed.:) State and Church: Studies in
Medieval Bulgaria and Byzantium. Sofia, 157–171.
NIKOLOV, Al. (2005): Cumani Bellatores in the Second Bulgarian State (1186–1396).
Annual of Medieval Studies at Central European University Budapest, 11, 223–229.
NIKOLOV, G. N. (2005): Tsentralizam i regionalizam v rannosrednovekovna Balgariya
(kraya na VII – nachaloto na XI v.), Sofia.
NIKOV, P. (1912): Tsar Boril v svetlinata na edin nov pametnik. Spisanie na
Balgarskata akademiya na naukite 3, 121–134.
OBERLÄNDER-TÂRNOVEANU, E. (1979): Quelques aspects de la circulation monétaire
dans la zone de l’embouchure du Danube au XIIe siècle. Dacia 23 (1979), 265–273.
OSTROGORSKI, G. (1977): Još jednom o pronijarima Kumanima, in Bogdanović, D. –
Jovanović-Stipčević, B. – Trifunović, Đ. (eds:) Zbornik Vladimira Mošina,
Belgrade, 63–74.
PAVLOV, P. (1989): Srednovekovna Balgariya i Kumanite. Voennopoliticheski
otnosheniya (1186–1241 g.). Trudove na Velikotarnovskiya Universitet “Sv. Kiril i
Metodii”, Istoriya 27/3, 7–61.
PAVLOV, P. (1990): Kumanite v obshtestveno-politicheskiya zhivot na sredno-
vekovna Balgariya (1186 – nachaloto na XIV vek). Istoricheski Pregled 46, 16–26.
PETROV, P. (1981): Obrazuvane na balgarskata darzhava, Sofia.
POWER, D. – STANDEN, N. (1999) (eds.) Frontiers in Question: Eurasian Borderlands
700–1700, Basingstoke–London.
PRITSAK, O. (1982): The Polovcians and Rus’. Archivium Eurasiae Medii Aevi 2 , 321–380.
PSRL 19622: Polnoe Sobranie Russkikh Letopisei vol. 2, ed. Shahmatov, A. A., Saint
Petersburg.
RASHEV, R. (2007): Kumanite na yug ot Dunav po arheologicheski danni, in
Nedelchev, N. (ed:) 800 godini ot bitkata pri Odrin, Shumen, 120–130.
RASOVSKY, D. (1939): Rolʹ Polovtsev v voinakh Asenei s Vizantiiskoi i Latinskoi
imperiyami v 1186–1207. Spisanie na Balgarskata akademiya na naukite 58, 203–211.
SARDELIĆ, M. (2019): Images of Eurasian Nomads in European Cultural Imaginary
in the Middle Ages, in Hao, C. (ed.:) Competing Narratives between Nomadic
People and their Sedentary Neighbours. Papers of the 7 th International Conference on
the Medieval History of the Eurasian Steppe, Nov. 9–12, 2018, Shangai University,
China. Szeged, 265–279.
SIMPSON, A. (2016): Byzantium’s Retreating Balkan Frontiers during the Reign of
the Angeloi (1185-1203): a Reconsideration, in Stanković, V. (ed:) The Balkans
and the Byzantine World before and after the Captures of Constantinople, 1204 and
1453, Lanham, 3–22.
SPINEI, V. (2009): The Romanians and the Turkic Nomads North of the Danube Delta from
SHIFTING FOUNDATIONS: THE CUMANS OF PARISTRION AND THEIR INFLUENCE…103
the Tenth to the Mid-Thirteenth Century, Leiden.

STǍNESCU, E. (1965): Les ”mixobarbares” du Bas-Danube au XIe siècle (quelques
problèmes de la terminologie des textes). Nouvelles études d’histoire 3, 45–53.
STANEV, K. (2013): Migratsiyata ot Trakiya kam Severna Balgariya v kraya na XII —
nachaloto na XIII vek i neynite posleditsi, in Nikolov, A. – Nikolov, G. N. (ed.:)
Srednovekovniyat balgarin i «drugite». Sbornik v chest na 60-godishninata na prof.
din Petar Angelov. Sofia, 207–226.
STEPANOV, TS. (1998): Periphery as Universe. Byzantinoslavica 59/2, 247–254.
STEPANOV, TS. (1999): Vlast i avtoritet v rannosrednovekovna Balgariya (VII–sr. IX v.),
Sofia.
STEPANOV, TS. (2005): Balgarite i stepnata imperiya prez rannoto srednovekovie, Sofia.
STEPANOV, TS. (2010): The Bulgars and the Steppe Empire in the Early Middle Ages. The
Problem of the Others, Leiden.
STEPANOV, TS. (2020): The History and Culture of Bulgaria in the 7 th – 12th
Centuries: Between Two Easts and Two Wests. Byzantina Lodzensia 39, 163–176.
STEPHENSON, P. (1996): The Byzantine Frontier at the Lower Danube in the Late Tenth
and Eleventh Centuries, in Power, D. – Standen, N. (ed.:) Frontiers in Question:
Eurasian Borderlands, 700–1700. London, 80–104.
STEPHENSON, P. (2000): Byzantium’s Balkan Frontier. A Political Study of the Northern
Balkans, 900–1204, Cambridge.
STOYANOV, V. (2005): Kumanite v Balgarskata istoriya (XI-XIV v.). Istoricheski
Pregled 61, 3–25.
STOYANOV, V. (2009), Kumanologia. Istoriografski eskizi, 2 vols., Sofia.
STOURAITIS, Y. (2020): Migrating in the Medieval East Roman World, ca 600–1204,
in Preiser-Kapeller, J. – Reinfandt, L. – Stouraitis, Y. (ed.:) Migration Histories of
the Medieval Afroeurasian Transition Zone. Aspects of Mobility between Africa, Asia
and Europe, 300-1500 C. E., Leiden, 141–185.
TANAȘOCA, N.-Ș. (1973): Les Mixobarbares et les formations politiques paristriennes
du XIe siècle. Revue Roumaine d’Histoire 12/1, 61–82.
TŬPKOVA-ZAIMOVA, V. (1976): Dolni Dunav – Granichna zona na vizantiyskiya zapad.
Kam istoriyata na severoistochnite balgarski zemi, kraya na X – XI v., Sofia.
TŬPKOVA-ZAIMOVA, V. (1979): Les Mixobarbaroi et la situation politique et
ethnique au Bas-Danube pendant la seconde moitié du XIe siècle, in Tăpkova-
Zaimova, V., Byzance et le Balkans à partir du VIe siècle, London, 615–619.
TŬPKOVA-ZAIMOVA, V. (2010a): La population du Bas-Danube et le pouvoir
byzantine (XIe-XIIe s.), in Tŭpkova-Zaimova, V., Byzance, la Bulgarie, les Balkans,
Plovdiv, 68–76.
TŬPKOVA-ZAIMOVA, V. (2010b): Les mouvements des populations en Mésie et en
Thrace entre le début du XIe et le début du XIIIe s., in Tŭpkova-Zaimova, V.,
Byzance, la Bulgarie, les Balkans, Plovdiv, 77–85.
VÁSÁRY, I. (2005): Cumans and Tatars. Oriental military in the pre-Ottoman Balkans,
1185–1365, Cambridge.
VENELIN, Y. (1829). Drevnie i nyneshnie bolgary v politicheskom, narodopisnom,

istoricheskom i religioznom ih otnoshenii k rossyanam. Istoriko-kriticheskie izyskanya.
Vol. 1, Moscow.
VENELIN, Y. (1841). Drevnie i nyneshnie bolgary v politicheskom, narodopisnom,
istoricheskom i religioznom ih otnoshenii k rossyanam. Istoriko-kriticheskie izyskanya.
Vol. 2, Moscow.
VLADIMIROV, G. (2018): Serʹgi v vide znaka voprosa iz srednevekovoi Bolgarii (XIII-XIV
vv.): o materialʹnykh sledakh kumanov i Zolotoi Ordy v kulʹture Vtorogo Bolgarskogo
tsarstva, Kazan.
ZHEKOVA, ZH. (2020): Pechati na nomadi na sluzhba vav Vizantiya (XI-XII v.), in
Acta Mediaevalia Magnae Tarnovie, vol. 1, Vladetel, darzhava i tsarkva na Balkanite
prez Srednovekovieto. Sbornik v chest na 60-godishninata na prof. d-r Plamen Pavlov.
Veliko Tarnovo, 836–845.
ZLATARSKI, V. (1918-1940, rep. 1970-72, 1994): Istoriya na Balgarskata darzhava prez
srednite vekove, 3 vols., Sofia.
ZLATARSKI, V. (1933): Potekloto na Petra i Asenya, vodachite na vastanieto v 1185 g.
Spisanie na Balgarskata akademiya na naukite 45, 8–48.
Nomads against Knights:
Cumans and Latins in the Military
Campaigns in the Balkans (1205–1225)
IVELIN IVANOV
Abstract
The paper considers the clash between Cuman and Western Latin troops in the
Balkans in the first two decades of the 13 th century by focusing on the following
questions. What was the role of the Cumans in the Bulgarian–Latin conflict of
1205–13? What was the real extent of nomadic involvement in the different stages
of the war? Can we analyse this military conflict as a clash between the nomadic
and Western, knightly warfare? Next, the study focuses on the Teutonic–Cuman
conflict of 1211–25 in the Burzenland region of Eastern Transylvania. The paper
analyses this war, drawing parallels with the Cuman–Latin clashes in the Bulgari-
an–Latin wars of 1205–13. In conclusion, the author presents a balanced assessment
of the strengths and weaknesses of Cumans and Latins in the Balkans in the first
decades of the 13th century.
Keywords: Bulgarian–Latin wars, Latin Empire of Constantinople, Cumans, Teu-
tonic order, medieval Burzenland, Medieval warfare, Nomadic warfare
The available written and archaeological sources provide seemingly sufficient but
simple and at times unreliable information about the participation and the role of
the Cumans in the Bulgarian wars with the Latin Empire in 1205–1211/13. What
was the real strength of the nomadic involvement in the different phases of the
war? To what extent were the Cuman troops controlled by the Bulgarian rulers
during the war? Can this military conflict be regarded as an example of a collision
between the nomadic and the Western, knightly warfare? Last but not least, what
was the significance of the Teutonic–Cuman war of 1211–25 in Burzenland and the
lands between the south-eastern Carpathians and the lower Danube? Usually, the
Prof. Ivelin Ivanov, St. Cyril and St. Methodius University of Veliko Tarnovo, Bulgaria. The
article is a revised version of a paper delivered at the 8th International Conference on Medi-
eval History of the Eurasian Steppe Nomads and their Neighbours in the Middle Ages, held in
Sofia, November 20–3, 2019.
106 IVELIN IVANOV
authors focus on the Cuman activity, the seasonal nature of Cuman warfare, and
other aspects of the conflict, but the issue of the clash between Latin and nomadic
warfare has not been analysed in detail. 1 Most often, historians of medieval war-
fare present this collision as a clash of two radically different military systems. A
view that Western knights did not realize the characteristics of nomadic warfare
and did not adjust their tactics to the realities in the East and the Balkans is also
popular among the medievalists. But is that correct? To what extent is the strength
or weakness of both sides exaggerated or undervalued? What does the analysis of
the Latin–Cuman clash in the Balkans of the early 13 th century show?
Chronologically, the study covers about two decades: the time between the
foundation of the Latin Empire in 1204 and the beginning of the Bulgarian–Latin
war in 1205, and the end of the Teutonic presence in Burzenland, south-eastern
Transylvania, in 1225. The paper considers the following issues: (a) composition,
numbers, mobility, and logistics of the Cumans and the Latin Crusaders in the
period 1205–25; (b) strategy and tactics of the Cumans and the Latins in the exam-
ined period; (c) the clash between the Cumans and Teutonic knights stationed in
the Burzenland area; and (d) troop numbers, composition, mobility, and logistics of
the Cumans and the Latins.
The Cumans
Most of the chroniclers exaggerate the number of Cuman troops. According to
Niketas Choniates, they were numerous (even almost innumerable and legions of
demons) and they usually operated in several directions at the same time, which
could explain this impression of multiplicity. 2 Also, all the written sources identify
Cumans as mounted bowmen, but some of the reports lead to the assumption that
these bowmen were not only Cumans but mixed detachments that also included
Vlachs and Bulgarians.
The archaeological map of the nomadic presence in the territories between the
Dniester, the Carpathians, and the Danube over three centuries—from the 10th to
the 13th century—can also be used as a source of information about the number of
Cuman troops during the period under question. Based on the archaeological data,
we are led to the conclusion that the total number of nomads in these lands hardly
exceeded 100,000, which casts doubts on the written information that the number
of Cuman armies in the military campaign in 1205 reached 14,000 or even 40,000
warriors. Most likely, Cumans from the closest territories of Cumania between the
1 Cumans and their actions in the Balkans in the late 12 th and early 13th centuries have been
the subject of study by a number of historians. See: DIACONU 1978, GOLEV 2017, IONITA 2010,
NIKOLOV 2008, PAVLOV 1989, PAVLOV 2002, PLETNEVA 2010, KNYAZʹKY 1988, RASOVSKY 1939,
RUSSEV – REDINA 2015, TSANKOVA-PETKOVA 1978, DALL’AGLIO 2008-2009, DALL’AGLIO 2013,
SPINEI 1986, SPINEI 2009, STOYANOV 2002, VASARY 2005.
2 According to some researchers, written records of high numbers of Cuman armies may be
considered credible (cf. PLETNEVA 2010, 144–5).

NOMADS AGAINST KNIGHTS. CUMANS AND LATINS… 107
Dniester and the lower Danube took part in the military campaigns under ques-
tion, but it is also possible that troops from farther Cuman clans east of the Dnie-
ster were involved as Bulgarian allies (or mercenaries).3
According to Henri de Valenciennes, there were numerous Cumans in Tsar
Boril's (1207–18) troops at the beginning of the 1208 campaign against the Latin
Empire.4 It remains unclear what part of these Cumans took part in the Battle of
Philippopolis on August 1, 1208. Many of them may have left early in the summer,
as they did during the campaigns of 1205, 1206, and 1207. The result of the battle of
Philippopolis on 1 August 1208 suggests that the Cuman cavalry was not as nu-
merous as in Kaloyan’s (r. 1197–1207) campaigns of 1205 and 1206. The next clear
account of Cuman involvement in the war dates to 1211 (or the previous 1210).5 In
April 1211 (or 1210), Bulgarians and Сumans invaded south-eastern Thrace.6 Also,
16 000
14 000
12 000
10 000
8 000 minimum number
6 000 maximum number
4 000
2 000
0
1205, 1206 1207, 1208, 1211, 1211,
May Sept. Summer Summer Autumn
Figure 1. Number of Bulgarians, Vlachs, and Cumans

in the summer of 1211, Boril and his ally Strez were defeated by allied Latin-
Epirote troops. Again, the number of Cumans in this campaign is not clear, but it
can be assumed that it was lower than in the campaigns of 1205–8. In the same year
1211, Tsar Boril undertook a new advance to Thessaloniki, but it is hard to estimate
the Cuman involvement in the campaign. Quite similar was the situation with the
next Bulgarian march against the Latins and their ally Alexios Slav in 1213. The
3 IONITA 2010; PAVLOV 1989; PAVLOV 2006.

4 VALENCIENNES 2009, 35.
5 LONGNON 1978, 144.
6 PRINZING 1973, 413 (48–51).
108 IVELIN IVANOV
failures of the last two campaigns lead to the assumption that Tsar Boril no longer
relied on large Cuman contingents. Obviously, the highest number of Cuman allies
(between 7,000 and 14,000) was reached in 1205–8. The above leads to the conclu-
sion that there was a sharp change in the number of Cuman troops in military ac-
tions compared to the period 1205–8. The chart above is an attempt to represent
and compare the numbers of Bulgarians, Cumans, and Vlachs in the campaigns of
1205–11.
The Latins
According to Geoffroy de Villehardouin, Emperor Baldwin I (1204–5) succeeded in
mobilizing about 340 knights, with whom he arrived under the walls of Adriano-
ple at the end of March 1205.7 Therefore, we can assume that the total number of
Latin troops under Adrianople varied between 1,300–1,500 armed men.8 The fact
that the siege actions were not related to a complete blockade of the town, but only
to the blockade of two of the city gates, is also quite indicative of the number of
Latin troops. If we calculate all the troops of the Latin Empire on the eve of the
battle of Adrianople in 1205 as indicated by Geoffroy de Villehardouin, we will
count about 4,000–5,000 armed men.
Apart from the Venetians, there were also Turcoples in the Latin army in the
1205–7 campaigns. Villehardouin mentions the Turcoples several times, implying
that they had acted in a coordinated way with the crossbowmen, apparently as
archers and as light cavalry for reconnaissance and other auxiliary purposes. Also,
Geoffroy de Villehardouin states that the Latin army numbered 400 knights on 23
June 1206.9 Multiplying the number of knights by four to seven times we can reach
an estimated total number of about 1,600–2,800 armed men. Based on the above, I
would summarize that except for the early spring of 1205—when the total number
of Latin troops reached 5,000—after the battle of Adrianople and by the end of the
war of 1205–13 Latin military resources did not exceed 2,500 capable men at arms.
At the same time, Latins, like their opponents, relied on allies and mercenaries.
7 VILLEHARDOUIN 1985, 104. According to Villehardouin, Emperor Baldwin left Constantino-

ple with about 140 knights. The next day, nine leagues before Adrianople, he was joined by
Marshal Geoffroy de Villehardouin and the number of Latin knights rose to about 200. They
were later joined by the Venetian Doge Enrico Dandolo, who led as many as the knights of
Emperor Baldwin and Count Louis de Blois, i.e. about 140. Therefore, we can conclude that
the Latin army under Adrianople numbered about 340 knights.
8 Calculating the Latin losses in this battle based on the written accounts of Villehardouin
and Clari I have reached a number of about 220 knights (140+300 = 440, 440/2 = 220), ex-
cluding the number of those of lower social status who lost their lives. These losses were
high not only in absolute numbers but also as a percentage of the total number: in this case,
more than 60% of the knights of the army besieging Adrianopolis.
9 VILLEHARDOUIN 1985, 120.
7000,0
6000,0
5000,0
4000,0
3000,0
minimum number
2000,0
1000,0
,0
maximum number
Figure 2. Number of Latin troops in the period 1205–13

Analysing the number of troops on each side relative to the number of success-
ful campaigns, I would underline the following. The considerable numerical supe-
riority of the Bulgarians, Cumans, and Vlachs, who had an overwhelming ad-
vantage in many clashes in the period 1205–8, does not coincide with the results of
the quantitative analysis of victorious battles and successful campaigns in the peri-
od. Despite the considerable superiority of the Bulgarian-Cuman-Vlach coalition
and the heavy losses of the Latins, the latter managed to organize their forces as
effectively as possible.
Mobility of the opposing sides
Analysing the mobility of the opposing sides, I would note that the Cuman mobili-
ty was beyond doubt very good. As all written sources testify, Cumans were high-
ly mobile both on the march, in attack, and in retreat. Their adversaries—the Latins
and their auxiliaries—could not compare with them. Unfortunately, we do not
have more specific data that allows us to determine the speed of Cumans’ daily
march, for example. However, such information on the speed of movement of Lat-
in troops can be found in some written accounts. According to Villehardouin, the
march from Fenepolis [Philippopolis] to Constantinople took about 9 (nine) days,
or an average of about 45 kilometres per day, which was a remarkable speed for
knights and sergeants.10 This means that in wartime conditions the speed of
movement of the Latins was not to be underestimated, but the Cumans were faster.
This may also explain why Tsar Kaloyan and Tsar Boril withdrew their infantry
and siege equipment as soon as the main Latin forces approached.

110 IVELIN IVANOV
Cumans’ and Latins’ strategy and tactics

After the end of the Carolingian conquests and with the beginning of the Viking
and Magyar attacks, the art of war evolved towards a defensive strategy, which
finally developed in the period from the 11 th to the 13th centuries. It brought de-
fence and siege operations to the forefront and pushed open battles to the back-
ground as they were too risky and with an unpredictable outcome. The role of light
cavalry and its effectiveness in dealing with heavy cavalry and infantry also in-
creased. This necessitated counteracting measures against the light cavalry,
through the wider use of bow and crossbow archers and spearmen. The strategy
and tactical practices of the Cumans and the Latins in the first two decades of the
13th century can be analysed in this context.
It is not an easy task to differentiate the Cuman strategy in the military cam-
paigns in question, as it was common to the allied Bulgarians, Cumans, and
Vlachs. Both in Kaloyan’s and Boril’s campaigns, the strategy of scorched and de-
populated land stands out, both in offensive and defensive actions. Niketas Choni-
ates and Geoffroi de Villehardouin testify to the destruction of Thrace by Kaloyan
and the Cumans in 1205 and 1206.11 The effect of the scorched-earth methods was
also reported by Henry de Valenciennes in his description of the summer cam-
paign and the battle of Plovdiv in 1208.12 Another characteristic of the strategy of
the Cumans, according to Choniates, was being active in several directions at a
time, which didn’t allow the enemy to concentrate forces for effective resistance.
Analysing the Latin strategy, I would emphasize the information concerning
logistics. Describing the siege of Adrianople at the end of March 1205, Villehardou-
in emphasizes the lack of food supplies for two reasons: the merchants did not
follow the Latin troops, and providing food from the surrounding area was almost
impossible because of the enemies’ attacks.13 The logistics of the Latin troops were
the main problem in the campaigns of 1205–8, which was related to the strategy of
scorched earth implemented by the Cumans, Bulgarians, and Vlachs. Also, Henry
de Valenciennes testifies that the Latin troops didn’t find food and wine for twelve
days during the march of Emperor Henry I against Tsar Boril in the summer of
1208. Even the emperor had only crumbs to eat and some wine to drink, while
many had nothing at all. It was due to the Bulgarians’ attempt at luring the enemy
into a completely depopulated and food-deprived territory.14
Another important issue in both sides' strategy was the seasonal war, usually
associated with the Cumans. However, I would like to consider the problem from a
broader perspective. Some researchers have interpreted the information about the
withdrawal of the Cumans during the summer months as directly related to their
11 VILLEHARDOUIN 1985, 114, 117, 118, 123, 128; NIKETAS CHONIATES 1983, 77, 99.
nomadic lifestyle and the relocation to summer pastures. No doubt, it is a reasona-

ble approach, but I would emphasize two accounts of Cuman actions around Pen-
tecost, i.e. in June. According to Villehardouin, just after Pentecost in 1205, the
Cumans withdrew because they could not fight due to the summer, and Tsar
Kaloyan directed his troops against the King of Thessaloniki, Boniface of Montfer-
rat (1204–1207).15 On the other hand, Choniates testifies to a Cuman assault on St
George's Day (in early May), and Valenciennes notes that the news of the invasion
of Cumans and Vlachs had reached Emperor Henry I on Pentecost 1208. 16 The
Cumans not only did not withdraw but even started a large-scale campaign in the
early summer. The report calls into question the total dependence of the Cumans
on the seasonal nomadic movement and suggests that the withdrawal of the Cu-
mans after May was not only provoked by the heat or the seasonal movement of
the herds to summer pastures. There was probably another reason that was present
in 1205–7, but not in 1208. It may be necessary to refer to Villehardouin's account
that in 1207 Kaloyan called off the siege of Adrianople because of the departure of
the Cumans, who decided to return to their homelands as they had accumulated
enough booty.17
But the Latins were also dependent on climatic conditions and seasonal chang-
es. Describing the campaign of 1206, Villehardouin reports that after the raid in the
Bulgarian lands at the end of the summer and in the autumn of 1206, the returning
Latin troops reached Adrianople and remained in its vicinity until the Feast of All
Saints (1 November): a time when they could no longer fight because of winter. 18
Additionally, the Latins weren’t fit to fight in winter and early spring not only
because of the cold but also because of the lack of fresh pasture and food supplies
in the war zones.
However, if we summarize the information on the seasonal activity of the Cu-
mans, a conclusion could be made that it was at its peak in the period January–
May. As for the Latins, they were most active in the April–September period. It is
clear that the Latin counter-raids and main success were mainly in the period
June–September. There were two reasons behind this: the absence of Cumans in
the summer and the autumn, as well as the better climatic conditions and logistical
support for people and horses.
Cumans I II III IV V VI VII VIII IX X XI XII
Latins I II III IV V VI VI VIII IX X XI XII
Figure 3. Seasonal military activity of Cumans and Latins

16 NIKETAS CHONIATES 1983, 55; VALENCIENNES 2009, 28.
112 IVELIN IVANOV
The above also raises a question of the effectiveness of the Cumans in the cam-
paigns under consideration. Setting aside the descriptions of their fierce raids and
widespread destruction, we can conclude that their role was really important, but
not the only factor for the outcome of the war. We should not underestimate the
fact that Cumans’ direct involvement in the siege warfare was insignificant, be-
cause for most of the military campaigns in question sieges were much more com-
mon and decisive than open battles.
Some chroniclers testify to the Cuman tactics in their campaigns in the Balkans,
with some specifics being highlighted. Niketas Choniates describes the Cuman
warfare as a heavy bow shooting from a safe distance, followed by false retreat and
decoying the enemy into pursuit, unexpected counterattacks and entering into
close combat only after weakening or dispersing their adversaries. Another charac-
teristic of the Cuman warfare, according to Choniates, was the ambush. A descrip-
tion of a Cuman attack on Messeni and Tzouroulos, which was taken by surprise
on St George’s Day in 1196, is also indicative of their tactics. The makeshift defence
with carts proved to be effective and the attackers were forced to withdraw. Cho-
niates himself explains this with the inability of the Scythians (Cumans) to besiege
field and city fortifications.19
Next, detailed reports of Cuman tactics can be found in Latin written sources.
Henri de Valenciennes testifies that in the battle of Beroia in June 1208, the Vlachs
and Cumans shot ruthlessly at the Latin army, which provoked a knight named
Lienard to break the battle formation and to attack the enemy. 20 This description
also testifies to the strict discipline followed by the Latin troops after the defeat of
Adrianople in April 1205. No one followed Lienard, and Emperor Henry himself
was criticized for breaking the order. Next, Valenciennes's account that due to the
perfectly flat terrain and the good weather the enemies didn't engage in battle testi-
fies clearly to the Cuman tactics of avoiding an open battle against mounted Latins
in such conditions. But despite the advantages of the terrain, the knights did not
pursue the enemy for fear of traps and ambushes on August 1, 1208. Valenciennes
also provides information on the formations of the Latins in this battle, describing
the detachments and the battle plan for the coming fight. The chronicler describes a
centre, two wings, and a split attack against the enemy. 21
Cumans against Teutonic Knights in Burzenland
The issue of the knights’ confrontation with the Cumans in military campaigns in
the Balkans can be viewed from another perspective, namely, the establishment of
the Teutonic Order in Burzenland and the war with the Cumans who lived in the
territories beyond the Carpathian range, in present-day Moldova and the north-
19 NIKETAS CHONIATES 1983, 55.

21 VALENCIENNES 2009, 32–3.
eastern part of the Wallachian lowland. 22 Some charters issued in the name of the
Hungarian king, Andrew II (r. 1205–35), in favour of the Teutons in Burzenland
reveal the impact of the order on the Cumans between 1211–25. Most valuable for
this research are the charters from 1211, 1212, and 1215, which describe the estab-
lishment of the order in Burzenland and the expansion of the Teutons in the Cu-
mans’ lands.23 The first charter issued after May 1211 testifies to the Cumans’ raids
in south-eastern Transylvania and the direction of the Teutonic expansion beyond
the Carpathians. This information is also supported by some of Pope Gregory IX’s
(1227–41) letters where the Cuman raids in the lands of Burzenland are mentioned
as a reason for the Teutonic counter-attack and the occupation of trans-Carpathian
territories. Moreover, King Andrew II permitted the Teutons to build wooden for-
tresses, both in Burzenland and in the newly occupied territories. The latter proved
to be effective both in the defence of Burzenland and in the subsequent expansion
and consolidation of the Order’s new conquests beyond the Carpathians.
In the context of this study, the charters and letters under consideration testify
to the characteristics of the Teutonic-Cuman war in 1211–12. It can be also assumed
that, as a result, the Cumans in the lands between the Dniester and the lower Dan-
ube were seriously engaged in the war, and that this led to limited participation of
Cuman troops in the military campaigns south of the Danube. Another charter of
1212 testifies that the Teutons had repelled Cuman attacks while subjecting them-
selves to mortal danger in defence of the Hungarian Kingdom. 24 Next, a diploma of
the Hungarian ruler from 1215 notes that at that time the Teutons completely con-
trolled Burzenland and the border territories, and also entered the lands of the
Cumans. The last of these four charters, dating from 1222, reveals important in-
formation about the territorial expansion of the Teutons beyond the Carpathians. 25
This document confirms the Teutonic dominance over the former Cuman terri-
tories to the east and south-east of the Carpathians and testifies that the Cumans
were under intense Teutonic pressure. Probably, in their further conquest after
1215, the Teutons cut off (in part or in full) the direct connection between the Cu-
man territories along the Dniester and those between the south-eastern slopes of
the Carpathians and the lower Danube. The above testifies to the military effec-
tiveness of the Teutonic Knights in their conflict with the Cumans and to results
22 The scope of research on Teutonic presence in Transylvania in the 13 th century is impres-

sive, but the focus is most often on Hungarian–Teutonic rather than Teutonic–Cumanian
relations. See: IVANOV 2018, HAUTALA 2015a, HAUTALA 2015b, HAUTALA 2015c, GLASSL 1971,
HUNYADI 2008, SHLOMO 2010, ZIMMERMANN 2000. Some sources must also be mentioned
here. First, some royal Hungarian letters and some papal letters from Gregory IX and
Honorius III testifying to the Teutonic actions against the Cumans in the period 1211–25 can
be mentioned (ZIMMERMANN 2000, 168–9).
23 ZIMMERMANN 2000, 162–193; HAUTALA 2015c, 80–9.
24 ZIMMERMANN 2000, 164–5.
25 ZIMMERMANN 2000, 169–172; HAUTALA 2015a, 23–4.
114 IVELIN IVANOV
that were achieved only within a decade. What were the reasons for the knights’
success in this conflict?
Based on the above analysis of the clash between the Cumans and the Latins in
the campaigns south of the Danube and the Haemus, I would emphasize the strat-
egy and tactics of the rival sides. Most importantly, the Teutons built an effective
defence with a network of fortifications in Burzenland, transferring this practice
beyond the Carpathians to the Cuman lands. As some chroniclers testify, the Cu-
mans were ineffective against field and city fortifications. Last but not least, unlike
the military campaigns south of the Danube, in which the Cumans fought in alli-
ance with the Vlachs and Bulgarians, they were deprived of such support in the
lands between the Danube Delta and the Carpathians. Therefore, the Cumans were
far more effective and victorious against the Latins south of the Haemus than
against the Teutons north of the Danube. Which, after all, could explain the close
relations and the strong alliance between Cumans and the first Asenids in the late
12th – early 13th centuries.
Bibliography
CLARI (1974): Robert de Clari. La conquete de Constantinople, Paris.
DALL’AGLIO, F. (2008-2009): The military alliance between the Cumans and Bulgar-
ia from the establishment of the second Bulgarian kingdom to the Mongol in-
vasion. Archivum Eurasiae Medii Aevi 16, 29–54.
DALL’AGLIO, F. (2013): The interaction between nomadic and sedentary peoples on
the Lower Danube: the Cumans and the “Second Bulgarian Empire”, in Curta,
F. – Maleon, B. (ed.:) The Steppe Lands and the World Beyond Them. Studies in hon-
or of Victor Spinei on his 70th birthday. Iași, 299–315.
DIACONU, P. (1978): Les Coumans au Bas-Danube aux XIе – XIIIe s., Bucharest.
GLASSL, H. (1971): Der deutsche Orden im Burzenland und in Kumanien (1211 –
1225). Ungarn: Jahrbuch der Presseagentur Budapress, Bd. 3. Wiesbaden, 22-49.
GOLEV, K. (2017): Za balgarofiliyata na kumanite pri upravlenieto na parvite Ase-
nevtsi. Minalo 4, 37–56.
HAUTALA, R. (2015a). Gramoty vengerskogo korolya Andreya II, predostavlennye
Tevtonskim rycaryam v Transilʹvanii: latinskie teksty, perevody i kommentarii.
Golden Horde Civilization 8, 13–30.
HAUTALA, R. (2015b): Papskie poslaniya v Vengriyu, kasayushchiesya perebyvaniya
tevtonskikh rycarei v Transilvanii (1211–1225). Peterburgskie slavyanskie i balkan-
skie issledovaniya 1 (Jan.–Jun.), 96–114.
HAUTALA, R. (2015c): The Teutonic Knights’ Military Confrontation with the Cu-
mans during Their Stay in Transylvania (1211–1225). Golden Horde Civilization 8,
80–89.
HUNYADI, Z. (2008): The Teutonic Order in Burzenland (1211–1225): New Re-
considerations, in Houben, H. – Toomaspoeg, K. (ed.:) L’Ordine Teutonico tra
Mediterraneo e Baltico. (Bari–Lecce–Brindisi, 14–16 settembre 2006). Galatina, 151–162.
IONITA, A. (2010): Structures de pouvoir et populations au Nord du Danube aux Xe-

XIIIe siècles reflétées par les découvertes funéraires. Transylvanian Review, vol.
19, Supplement no. 5/1, 115–34.
IVANOV, I. (2018): Tevtonskiyat orden, kumanite i Balgariya prez vtoroto desetiletie
na XIII v. (1211–1225). Epokhi 26/2, 383–393.
KNYAZʹKY, I. O. (1988): Polovcy v Dnestrovsko-Karpatskikh zemlyakh i Nizhnem
Podunavʹe v konce XII – pervykh desyatiletiyakh XIII v. Socialʹno-ekonomiches-
kaya i politicheskaya istoriya Moldavii perioda feodalizma. Kishinev, 22–32.
LONGNON, J. (1978): Les compagnons de Villehardouin. Recherches sur les croisés de la
quatrième croisade, Genève.
NIKETAS CHONIATES (1983): Historia. Gratski izvori za balgarskata istoriya (GIBI), vol. 11.
Sofia, 8–94.
PAVLOV, P. (1989): Srednovekovna Balgariya i Kumanite. Voennopoliticheski otno-
sheniya (1186–1241). Trudove na Velikotarnovskiya Universitet Sv. sv. Kiril i Metodiy.
Istoriya, vol. 27. Veliko Tarnovo, 9–59.
PAVLOV, P. (2002): Kumani i alani vav Vtoroto balgarsko tsarstvo (Nyakoi paraleli
v istoricheskoto razvitie na Balgariya i Ungariya prez XІІ–XІV v.), in Csíkhelyi,
L. (ed.:) Balgari i ungartsi – 1000 godini zaedno. Budapest, 9–18.
PAVLOV, P. (2006): Balgariya i neynite voenni sayuznitsi ot Severnoto Pricherno-
morie (XII–XIV v.) Balgarite v Severnoto Prichernomorie, vol. 9. Veliko Tarnovo,
49–60.
PLETNEVA, Sv. (2010): Polovtsy, Moskva.
PRINZING, G. (1973): Der Brief Kaiser Heinrichs von Konstantinopel vom 13. Januar
1212. Überlieferungsgeschichte, Neuedition und Kommentar. Byzantion 43,
395–431.
RASOVSKY, D. (1939): Rolʹ polovcev v vojnakh Asenej s Vizantiiskoi i Latinskoi im-
periyami (1186–1207 gg.). Spisanie na BAN 58, 203–211.
RUSSEV, N.D – REDINA, E.F. (2015): Pogrebenie kochevnika s vizantiiskimi mon-
etami u ozera Katlabukh i razvitie etnopoliticheskoj situacii na Nizhnem Du-
nae v XII veke. Stratum plus 6, 99–107.
SHLOMO, L. (2010): Between the Latin Kingdom of Jerusalem and Burzenland in
Medieval Hungary. The Middle Ages and the Crusades 10 (Jan.-Jun.), 184–195.
SPINEI, V. (1986): Moldavia in the 11th–14th centuries, Bucharest.
the tenth to the mid thirteenth century, Boston.
STOYANOV, V. (2002): Kumans in Bulgarian History (Eleventh – Fourteenth Centuries).
The Turks, vol. 1, Early Ages, pt. 9, Turks in East Europe. Ankara, 680–689.
TSANKOVA-PETKOVA, G. (1978): Balgariya pri Asenevtsi, Sofia.
VALENCIENNES (2009): Anri dyo Valansien. Istoriya na imperator Anri, Sofia.
VASARY, I. (2005): Cumans and Tatars. Oriental Military in the Pre-Ottoman Balkans,
116 IVELIN IVANOV
VILLEHARDOUIN (1985): Zhofroa dyo Vilarduen. Zavladyavaneto na Konstantinopol,

Sofia.
ZIMMERMANN, H. (2000): Der Deutsche Orden im Burzenland. Eine diplomatische Un-
tersuchung, Köln.
Ecology of Warfare: The Seasonal
Campaigns of the Cuman-Qïpchaqs in
Eastern Europe and Central Asia
KONSTANTIN GOLEV
Abstract
According to Dmirty Rassovsky, the Cumans never appeared south of the Danube
to support the Asenids during the hot time of the year, and if it occurred in the
course of a campaign they withdrew. Plamen Pavlov relativizes to a certain extent
the too deterministic model introduced by Rassovsky, which is still relevant for the
relations between the Cumans and the first rulers of the re-established Bulgarian
state. Yet, the question whether this model is valid only for the period of a few
decades of intensive military conflicts in the Balkans (at the end of the 12th and the
beginning of the 13th century) or had broader chronological and geographical im-
plications remains unsolved. On the basis of various cases from the Balkan, Rus’,
Khwārazmian, and Mongol history examined in the present paper, it can be con-
cluded that the asymmetric seasonal military activity of the Cuman-Qïpchaqs was by
far not a unique feature of their military cooperation with the first Asenids. In fact,
this phenomenon can be observed through the entire period of Cuman activities in
 Assist. Prof., PhD – Institute for Historical Studies of the Bulgarian Academy of Sciences.
This paper is a result of research carried out within the project “Nomadic Migrations in the
Bulgarian Lands during the Late Middle Ages – History, Memory and Usage” financed by
the Bulgarian Science Fund under contract no. KP-06-M30/4 from 17.12.2018. The article is a
revised version of a paper delivered at the 8th International Conference on Medieval History
of the Eurasian Steppe Nomads and their Neighbours in the Middle Ages, held in Sofia, Novem-
ber 20-3, 2019. It is dedicated to the 80th anniversary of the untimely and tragic death of
Dmitry Aleksandrovich Rasovsky (19.04.1902–06.04.1941), a scholar and researcher of the
Cuman-Qïpchaqs who was far ahead of his time. Wherever the original source texts are
given, the accompanying translations are made by the author and may differ to a certain
extent from already published translations of other scholars, which are also referred to for
comparison. The author is indebted to Prof. Mirena Slavova, Dr. Delyan Rusev, Dr. Frances-
co Dall’Aglio, Iliana Bozhova, Prof. Pavel Pavlovich, and Dr. Ishayahu Landa for their valu-
able comments and help with the translation of particular source passages, but bears alone
the full responsibility for any mistakes.
118 KONSTANTIN GOLEV
the Balkans and Central Asia (11th–13th centuries). Yet, the Cuman asymmetric sea-
sonal military activity was also not ubiquitous as it was not characteristic of the in-
tensive contacts of these nomads with the Rus’ principalities. By drawing synchro-
nous and diachronous parallels with the activities of other Eurasian nomads, the
author identifies the factors that stood behind these dynamics.
Keywords: Cuman-Qïpchaqs, climate, warfare, Byzantium, Second Bulgarian
Tsardom, Rus’, Khwārazm
Introduction
In the course of a nearly two-century-long domination in the steppes between the
Irtish and the Danube (mid-11th–mid-13th centuries), the Cuman-Qïpchaqs regular-
ly entered into the closer and even the more distant sedentary territories—both in
Europe and in Asia. As a rule, the appearance of these stockbreeders outside of
their steppe habitat (the so-called Dasht-i Qipchāq)1 was caused by an invariable
desire for participation in military conflicts either in the usual form of nomadic
raids or as sedentarists’ allies/mercenaries, but always to their own advantage.
Modern scholars explain this eagerness with the nomads’ desire to supplement the
products of the extensive steppe economy, as well as with the existence of a specif-
ic military know-how, which could be widely implemented in the territories of the
so-called outside world.2 The numerous campaigns of the Cuman-Qïpchaqs in the
domains of their settled neighbours exercised significant influence over the latter’s
political history and therefore have been an object of research in multiple historical
studies. Yet, despite the extant voluminous literature, only few scholars have paid
the due attention to one peculiar feature of the Cuman military undertakings in the
outside world—their seasonal rhythm.
Dmirty Rasovsky, a Russian scholar and white émigré, was the first to do that
more than eighty years ago— in a study about the Cumans as a factor in the wars
of the early Asenids with Byzantium and the Latin Empire. According to his ob-
servations, these nomads never appeared south of the Danube during the hot time
of the year and if it occurred in the course of a campaign, they would leave the
1 I.e., “Qïpchaq Steppe”, the Persian name of the Eurasian habitat of the Cuman-Qïpchaqs,
referred to in the Rus’ sources as “Pole Polovetskoe” and in the western sources as Cuma-
nia. Rasovsky was right to note that initially these terms did not overlap. Due to the enor-
mous extent of the steppes controlled by the Cuman-Qïpchaq tribes, their sedentary con-
temporaries were initially unable to fully realize their size, and thus the various toponyms
appeared as a designation to the steppes in the immediate proximity to the respective medi-
eval authors, RASOVSKY 1937, 71–3. In the present text Dasht-i Qipchāq and its synonyms are
used as a terminus technicus referring to the territories controlled by the Cuman-Qïpchaq
tribes in the period of their domination in Western Eurasia.
2 GOLDEN 1987–91, 68–73; see also GOLDEN 1991, 149. The term outside world is introduced by
Anatoly Khazanov as a general designation of the non-nomadic societies (mostly agricultur-

al and urban) with whom the nomads cohabitated and interacted, KHAZANOV 1994, 3.
ECOLOGY OF WARFARE: THE SEASONAL CAMPAIGNS… 119
Bulgarian army and withdraw to the steppes. 3 Rasovsky supports his conclusions
with a number of unequivocal source quotations and defines the period of Cuman
activity in the Balkans as the late autumn, winter, and spring.4 Quite naturally, the
Russian scholar explains this behaviour with the seasonal migrations of the no-
mads and the withering of the grass in the summer months. This concept is still
relevant, but Rasovsky went too far in determining that the absence of the Cumans
from the Balkans during the summer was mandatory. By pointing out some excep-
tional cases of Cuman presence in the summer months, the Bulgarian scholar Pla-
men Pavlov relativizes to a certain extent the model introduced by Rasovsky, as-
suming that only the allied contingents from the more distant steppe regions with-
drew during the hot season.5 Indeed, one can add to the cases cited by Pavlov fur-
ther evidence referring to Cuman activities south of the Danube during the sum-
mer.6 Thus, to summarize the present state of the field: Rasovsky’s concept, relativ-
ized by Pavlov, is still absolutely relevant—the Cumans avoided appearing on the
Balkan military theatre during the summer, but this does not mean that their pres-
ence during the hot season should be considered a priori impossible.
These important observations on the Cuman military activity south of the Dan-
ube raise interesting questions: is it possible to identify similar rhythms in the Cu-
mans’ military endeavours in other contact zones of Dasht-i Qipchāq with the out-
side world? Furthermore, is the model of Rasovsky applicable to the Cuman incur-
sions in Byzantium’s Balkan provinces before the restoration of the Bulgarian state
in 1185/86? In other words, what was the geographic and chronological span of
the phenomenon, which could be defined as asymmetrical seasonal military activity of
the Cuman-Qïpchaq tribes? The answers to these questions can only be found
through comparative research of the Cuman presence in various contact zones in
Europe and Asia. Such an approach could often shed additional light upon the
3 RASOVSKY 1939, 203–11.

4 In this regard, it is curious to note the fragmentary preserved text of an undated Bulgarian
stone inscription from Veliki Preslav. It mentions in an unclear context 10,800 Cumans and
apparently lists six months, each of which fits within the referred seasons, in the following
sequence: December, January, March, April, May, October, TOTEV 2007, 37-40. In an earlier
publication, Totev reads the number of the Cumans as 10,300, TOTEV 1985, 158–69.
5 PAVLOV 1992, 34–5; PAVLOV 1990a, 19–20. Especially indicative is the participation of Cu-
mans in the war with the Latin Empire in July–August 1208, pointed out by Pavlov, see
HENRI DE VALENCIENNES/BUROV 2009, 28–39.
6 Thus, the regent and future emperor of the Latin Empire Henry I (r. 1206–16) personally
states that not all Cumans had left the Bulgarian army in the summer of 1205. He writes to
his brother Geoffrey that Kaloyan’s troops, which attacked and captured Philippopolis after
the termination of the campaign in the region of Macedonia, were composed of “an infinite
multitude of Vlachs and Cumans”: “Ihoannicius cum infinita Blacorum et Commannorum
multitudine ad prelibatam civitatem veniens, eam obsedit.” (Ihoannicius [i.e. Kaloyan] arriv-
ing with an infinite multitude of Vlachs and Cumans at the aforementioned city [of Philip-
popolis], besieged it.), HENRICUS IMPERATOR/PRIMOV 1981, 14.
behaviour of these medieval nomads by pointing to parallels or contrasts that

would otherwise remain unnoticed in the research of Cuman actions in one partic-
ular geographical or geopolitical area.7 Thomas Allsen has introduced such a “ho-
listic approach” in the studies of the much larger Mongol Empire, 8 and there could
be no doubt that a similar methodology would offer promising perspectives for
those interested in the Mongols’ far less known predecessors in Western Eurasia—
the Cuman-Qïpchaq. Regarding the present subject, only after one has established
whether the seasonal rhythms of the Cuman campaigns were characteristic of the
Balkans, or on the contrary, they can be observed in other regions as well, one
would be able to seek with greater confidence the reasons behind this phenomenon.
Byzantium
Despite the relatively scarce data, an examination of the sources for the Cuman
military activity south of the Danube in the period of Byzantine domination over
the Balkans (11th–12th centuries) allows for certain conclusions regarding its sea-
sonal intensity. It is noteworthy that the first documented appearance of the Cu-
mans on the Balkans, as allies of the Pechenegs during an incursion in the sur-
roundings of Adrianople, apparently took place in the autumn or winter of 1078.
According to Michael Attaleiates and the Continuation of the Chronicle of John Sky-
litzes, the attack occurred while the imperial troops under the command of the
future emperor Alexios I Komnenos (r. 1081–1118) were engaged in a military
campaign against the rebel Basilakes in the area of Thessaloniki. 9 Since Alexios was
sent by Nikephoros III Botaneiates (r. 1078–81) against the pretender in the autumn
of 1078, the advent of the Cumans and the Pechenegs in the vicinities of Adriano-
ple must have taken place during the cold months that followed his departure. 10
Less than a decade later, another nomadic incursion into the Balkans in which
the Cumans may have played a role followed. This was the campaign of the de-
7 GOLEV 2013a, 142–51; GOLEV 2018a, 89–126; see also two separate publications that examine
the relations of the Cuman-Qïpchaqs with the urban centres in the contact zone with the
outside world in two completely different and remote regions GOLEV 2018b, 23–107; GOLEV
2021a, 11–52.
8 ALLSEN 1987, 9–11.
9 ATTALEIATES/KALDELLIS – KRALLIS 2012, 548–9; THE CONTINUATION/MCGEER 2020, 184–5.
10 SKOULATOS 1980, 37. There is another mention of the Pechenegs and the Cumans during
the reign of Botaneiates preserved only in the Continuation of the Chronicle of John Skylitzes.
According to that source, a certain Leo Diabatenos—an imperial official—was dispatched to
govern Mesembria and managed to arrange “a reconciliation and treaty” with them, THE
CONTINUATION/MCGEER 2020, 188–9). However, this seems to be a reference to a settlement
of the aforementioned conflict in the vicinity of Adrianople, and not to another joint incur-
sion of the nomads in the cold seasons, as I supposed earlier on the basis of a wrong dating
and partial publication of the source text in SKYLITZES – KEDRENOS/TAPKOVA-ZAIMOVA 1965,
340; see GOLEV 2021b, 9.
throned Hungarian king Salomon (1063–74), who attacked the European domains
of Byzantium together with unidentified nomadic allies, but was defeated and
perhaps killed.11 A juxtaposition of Anna Komnene’s Alexiad, where it is stated that
the incursion started “after the arrival of spring” (Ἔαρος δὲ ἐπιφανέντος) and the
account of The Hungarian Chronicle of the 14th century, according to which the in-
vaders were defeated during the winter and Salomon’s retreat was hindered by a
heavy snowfall, suggests that the events perhaps took place in the first cold
months of 1087.12 Therefore, this campaign, too, apparently matched the seasonal
cycles of the steppe nomads’ military activity, regardless of the particular tribal
affiliations of the protagonists.
In the spring of 1091, Cumans appeared in Thrace, where on April 29 they
helped the Byzantine emperor, Alexios I Komnenos, to defeat the Pechenegs who
roamed in his realm in the famous battle of Lebunion. 13 The Cuman campaign in
support of the so-called Pseudo-Diogenes ca. 1095 apparently also took place in the
cold seasons. This is suggested by Ana Komnene, who states that after the final
battle with the nomads near an unidentified passage of the Balkan Mountains, her
father was forced to spend the night in the mountain “since there was a violent
storm”.14
Rumors about a Cuman incursion south of the Danube compelled the emperor
to head with his troops towards his Balkan provinces in November 1114, but as far
as one can judge from the unclear chronology of Anna Komnena, the assault itself
took place in the next summer.15
11 The Hungarian Chronicle from the 14th century refers to his allies as “Cuni”, a generic term
for eastern nomads, while Anna Komnene labels them in the archaic style typical for the
Byzantine historiography: “a mixed […] army […] of both Sauromatians and Scythians”
(σύμμικτον […] στράτευμα […] ἔκ τε Σαυροματῶν καὶ Σκυθῶν), for these accounts, see n. 12.
Thus, it seems probable that at least some of the nomads who raided the Balkans with Salo-
mon were affiliated to the Cuman-Qïpchaq tribal community although this cannot be estab-
lished with certainty.
12 CHRONICI HUNGARICI/DOMANOVSZKY 1999, 409–10; THE HUNGARIAN ILLUMINATED CHRONI-
CLE/WEST 1969, 128; ANNA KOMNENE/REINSCH – KAMBYLIS 2001, 203; ANNA COMNENA/
DAWES 2000, 119; for the death of Salomon, see BERNOLDI CHRONICON/PERTZ 1844, 446.
13 ANNA KOMNENE/REINSCH – KAMBYLIS 2001, 243–51; ANNA COMNENA/DAWES 2000, 142–7.
14 “ὁ δὲ βασιλεὺς περὶ τὴν ἀκρολοφίαν τῆς Σιδηρᾶς Κλεισούρας παννύχιος διατελέσας,
χειμῶνος ὅντος σφοδροῦ, αὐγαζούσης ἤδη ἡμέρας τὴν Γολόην κατείληφεν.” (The basileus
spent the whole night on the mountain ridge of the Sidera gorge, since there was a violent
storm, and when the day dawned he reached Goloe.), ANNA KOMNENE/REINSCH – KAMBYLIS
2001, 295. All accessible translations render this passage in a different way, in the sense of a
storm, bad or cold weather, ANNA COMNENA/DAWES 2000, 175; ANA KOMNINA/LYUBARSKY
1996, 274; ANA KOMNINA/VOYNOV 1972, 112. Regarding the dating of this Cuman campaign,
see the various opinions cited by Lyubarsky, ANA KOMNINA/LYUBARSKY 1996, 556, n. 959.
15 ANNA KOMNENE/REINSCH – KAMBYLIS 2001, 454, 457, 458; ANNA COMNENA/DAWES 2000,
271, 273–4.
Another large-scale nomadic invasion, which apparently begun in the autumn,

led to a war between the invaders and Emperor John II Komnenos (r. 1118–43) in
the winter and spring of 1121/22 or 1122/23.16 In this case, the ethnic affiliations of
the attackers are not clear—some scholars identify them as Pechenegs, 17 other as
Cumans.18 In all likelihood, this was a coalition of Berendeis, Uzes, and Peche-
negs,19 but what is more important for the present topic is that a significant inva-
sion like this also followed the seasonal pattern of nomadic military activity in the
Balkans, which is by far not an exclusive feature of the Cumans.
New Cuman raids followed during the reign of the next emperor of the Kom-
nenos dynasty—Manuel I (r. 1143–80). The first of them took place in 1148 and
demanded the personal intervention by the ruler, who, according to John Kin-
namos, even dared to personally chase the nomads beyond the Danube. 20 Modern
scholars date the assault to the end of the winter or the beginning of the spring,21
and this hypothesis is fully supported by the narrative of the Byzantine chronicler.
Indeed, Kinnamos reports that Manuel I set off towards the island of Kerkira, re-
cently captured by the Normans, via a land route through the Balkans. But when
the emperor arrived at Philipopolis, he was informed about the Cuman incursion
and headed towards the Danube. While awaiting the arrival of ships from Con-
stantinople, Manuel spent the time hunting through the Danubian Plain. But be-
fore the appearance of the vessels, the Cumans withdrew with their booty beyond
the river and the emperor followed them with a detachment of five hundred men,
with whom he crossed two more rivers before catching the nomads and defeating
them. Yet, despite this significant detour, the emperor managed to reach the Adri-
atic coast “in the most suitable time” for sailing, and his plans for the siege of
Kerkira in the same year were diverted only by the delay of the Byzantine fleet.
The latter arrived in late autumn, when the sea was already dangerous.22 The fact
16 KINNAMOS/MEINEKE 1836, 7–9; KINNAMOS/BRAND 1976, 16–17.

17 See ZLATARSKI 1994, 369; MLADENOV 1931, 115–136; OSTROGORSKI 1998, 485; TAPKOVA-
ZAIMOVA 1976, 104; SPINEI 2003, 150–1; SPINEI 2009, 125–6. In the latter book Spinei allows for
the possibility that Cuman groups on the Lower Danube joined the invaders, who apparent-
ly were a combination of several tribes, but still affirms that the Pechenegs must have had
the leading role in the incursion.
18 DIACONU 1978, 62–77; FINE 1989, 225, 234; STEPHENSON 2000, 106; CURTA 2006, 312, 319.
19 See the argumentation in GOLEV 2013b, 358–64.
20 KINNAMOS/MEINEKE 1836, 93–5; KINNAMOS/BRAND 1976, 76–8; CHONIATES/VAN DIETEN
1975, 78; CHONIATES/MAGOULIAS 1984, 46; THEODOROS SKOUTARIOTES/VOYNOV 1972, 226–7;

JOHANNES TSETSES/YONCHEV 1980, 108; BIBIKOV 1981, 117–118.
21 ZLATARSKI 1994, 384–385; followed by BIBIKOV 1976, 18; MAGDALINO 2002, 53.
22 “ὁ μὲν γὰρ καίτοι τῶν Σκυθικῶν μεταξὺ περιεσπακότων ὅμως ἔφθη ἐπικαιρότατα τῷ χώρῳ
ἐπιστάς, ὅθεν κὰι τὴν ἀπόπλοιαν ποιεῖσθαι ἐχρῆν· τὸν δὲ στόλον ξυνέβαινεν, εἴτε πνεύμασιν
ἐναντίοις τῆς ἐπὶ τὰ πρόσω ἀνακοπέντα, εἴτε καὶ τῇ τοῦ δουκὸς περὶ ταῦτα ἀμαθίᾳ, ὀψὲ καὶ
κατόπιν ἐλθεῖν τοῦ καιροῦ. ἔαρος γὰρ τῶν Βυζαντίων ἀναχθεὶς λιμένων, φθινούσης ἤδη τῆς
ὥρας ἐς βασιλέα κατῆρεν ἐντεῦθέν τε τὰ Ῥωμαίων ἔσφηλε πράγματα. [...] πελάγη τε γὰρ
that Manuel I reached the Adriatic coast before the arrival of his fleet in the au-
tumn as well as the reasonable assumption that he set off from Constantinople as
early as possible suggest that the Cuman raid indeed must have taken place at the
end of the winter or in the spring of 1148.
Two more Cuman incursions followed during the reign of the same emperor:
ca. 1152–54 and less than a decade later. The latter was also the last one for which it
can be assumed in what time of the year it has been conducted. The attack took
place at the very end of the sixth or the beginning of the seventh decade of the 12 th
century, when Manuel I led a campaign against the Seljuq sultan Kilij Arslan II (r.
1156–92). Kinnamos reports that the basileus concluded peace with the sultan and
set off towards his own realm, but on the way he learned “that the Scythians [i.e.
the Cumans] had crossed the Danube intending to ravage the land of the Ro-
mans”.23 In order to face the raiders, Manuel I took a detour from the route to Con-
stantinople and crossed the Dardanelles near Gallipoli heading directly towards
them. When the nomads were informed about his approach, however, they retreat-
ed before the emperor reached the Danube. 24 Claude Cahen dates the beginning of
the Seljuq campaign in the winter of 1160/125, and according to Özaydın the fights
ended in the autumn of 1161.26 These rather general chronological assessments
allow for the assumption that the Cuman incursion in question also took place in
the cold time of the year, but due to the lack of more solid evidence one cannot
reach beyond the sphere of conjectures.
Naturally, there are a number of other episodes during the Byzantine domina-
tion over the Balkans (11th–12th century) when the Cumans appeared south of the
Danube, but in all of them the sources do not provide any hints about the season of
their activities. Yet, it is indicative that, as a rule, whenever the precise time of the
year is mentioned or can be deduced from the context, it almost always falls within
ἐνταῦθα δεινῶς ἀχανῆ τέταται, καὶ ἔστιν ἐπιεικῶς κινδυνώδης ὁ πλοῦς καὶ μάλιστα
χειμῶνος.” (For, indeed, although in the meantime he [the basileus] had been diverted by
the Scythians, nevertheless he arrived in the most suitable time [for sailing] establishing
himself on the place, whence the navigation had to be made. Yet, it so happened that the
fleet – either prevented to move forward by unfavourable winds, or due to the ignorance of
the duke [i.e. the admiral] in the [maritime] affairs – arrived too late and after the favourable
time [for departure]. For after [the fleet] had sailed from the Byzantine ports in the spring, it
arrived to the basileus when the autumn was already coming to an end, and thus thwarted
the affairs of the Romans. […] Because the sea there stretched terribly wide and sailing is
rather dangerous, especially in the winter.), KINNAMOS/MEINEKE 1836, 96; KINNAMOS/
BRAND 1976, 78–9.
23 “πυθόμενος δὲ Σκύθας τὸν Ἴστρον διαβῆναι ἐφ᾽ ᾧ Ῥωμαίων καταδραμεῖσθαι”, KINNAMOS
1836, 201; KINNAMOS/BRAND 1976, 153.

24 KINNAMOS/MEINEKE 1836, 201–2; KINNAMOS/BRAND 1976, 153.
25 CAHEN 2001, 25–6.
26 ÖZAYDIN 2002, 399.
the cold months of the year. The only exception is the summer incursion by the
Cumans in 1115, mentioned by Anna Kommnena. Yet, even in this case the ap-
pearance of the nomads was initially expected during the cold months, which
forced Alexios I Komnenos to gather his troops and to spend the winter of 1114/15
in Thrace. In fact, Anna’s account of these events is too vague and obscure, which
is not unusual for her writing style. But even if such a summer “precedent” exist-
ed, the disproportion with the significant number of documented incursions dur-
ing the cold seasons strongly suggests that most of the undated Cuman raids dur-
ing the period of Byzantine domination over the Balkans must have also taken
place in the cold time of the year. Of course, such a tendency is not a deterministic
one and does not mean that the nomads were absolutely absent from the peninsula
during the summer. Such instances were simply much rarer as compared to the
peaks of nomadic military activity in the Balkans during the cold seasons—just as
it was during the later period of intensive cooperation between the Cumans and
the Asenids (end of 12th and early 13th centuries) studied by Rasovsky and Pavolv.
Khwārazm
It is only logical that the present comparative research would include the most
distant from the Balkans contact zone of the Cuman-Qïpchaqs with the sedentary
societies—the Central Asian region that for thousands of years has been known as
Khwārazm. Luckily, the medieval sources for the history of these lands provide
some evidence that is directly related to the researched topic.
In the first place, the geographical features of Central Asia should be taken into
consideration—the oases of agriculture in the region are dispersed amidst steppes
and deserts, which makes the frontier between the habitats of nomads and seden-
tarists much more fluid as compared to Eastern Europe.27 Due to this, the climate
turned to be a decisive factor defining the access to the sedentary oases, which
according to the medieval sources were inaccessible in some periods of the year.
Thus, Bayhaqī (11th century) narrates that the Khwārazmshāh Hārun b. Altuntash
(r. 1032–5), after his meeting with Shāh-Malik—the ruler of Jand near the Syr Dar-
ya—commented that the latter was only able to come in Khwārazm during the
winter, when the desert is covered with snow.28 This statement may be too deter-
ministic, but there is no doubt that the winter was the most suitable season for
military campaigns in the region. Due to the harsh natural conditions, this observa-
tion proved to be valid not only for the northern nomads, but also for the invading
armies from the south, who customarily entered Khwārazm during the cold part of
27 KHAZANOV 1992, 69–74; see also BAYPAKOV 1986, 7–12; BREGEL 2003, 2–3 and map 1.
28 »‫( «و جز زمستان که این بیابان برف گیرد از جند اینجا نتوان آمد‬and except in the winter, when this desert
is covered with snow, it is not possible to arrive here from Jand), BAYHAQĪ/KHAṬĪB RAHBAR
1393/2014, vol. 3, 1116–1117; BEYHAQI/BOSWORTH 2011, vol. 2, 391-392; vol. 3, 395, n. 113;
BAYHAQĪ/ARENDS 1969, 827.
the year, as evidenced by the campaigns of the Seljuq sultans Alp-Arslan (r. 1063–
72)29 and Sanjar (r. 1118–57).30
Furthermore, in the case of the nomads their military activity was defined not
only by the climatic conditions in the vicinities of Khwārazm, but also by the sea-
sonal cycles of their own migrations. Owing to the aridity of the steppes, these
migratory routes stretched over much longer distances in Central Asia as com-
pared to Europe.31 As a result of these migrations, the steppes along the lower and
middle course of the Syr Darya sheltered the winter camps of prominent nomadic
groupings throughout the centuries. 32 Thus, it is hardly surprising that the combi-
nation of the two aforementioned tendencies defined the peak of nomadic military
activity in Khwārazm during the cold months of the year. In fact, the appearance of
the steppe dwellers in the region was so rhythmical that the locals sedentarists also
organized their campaigns against them in a strictly defined time of the year. This
is revealed by al-Bīrūnī’s account about annual early winter campaigns of the in-
habitants of Khwārazm aimed at driving the Oghuz away from their borders. 33
The question of how the activities of the Cuman-Qïpchaqs in Khwārazm and its
surroundings fit within this general model of nomadic behaviour arises. It is note-
worthy that as early as 1133 the Khwārazmshāh ʿAlāʾ al-Dīn Atsïz (r. 1127/8–1156)
of the Anushteginid Dynasty (1097–1231) led a campaign from Jand in “the depth
of Turkistān”, where he “encountered a ruler and chief who was considered the
most powerful among the infidels”, managed to defeat him, and slew many of his
followers.34 Apparently this attack occurred in the winter or early spring of 1133,
29 MĪRKHWĀND 1339/1960, 274–6; MITT 1941, 466–7; İBNÜʾL CEVZÎ/SEVIM 2011, 149–50; see
also TURAN 1993, 159; BREGEL 2003, 28–9.
30 IBN AL-ATHĪR/RICHARDS 2005, 348, 370; JUVAYNĪ/QAZVĪNĪ 1916/1334, 5, 7–8, 8–10; JU-
VAINI/BOYLE 1958, 280, 281–2, 282–4; BARTOL’D 1898, 44–7; BARTOL’D 1900, 347–8, 350–1; PAUL
2013, 81–129 (especially p. 103 for Sanjar’s second campaign in Khwārazm).
31 See in general KRADER 1955, 301–26; see also GOLDEN 1987–1991, 78.
32 Thus, according to the anonymous author of Ḥudūd al-ʿĀlam, as early as the late 10 th centu-
ry the malik of the Ghūz (i.e the Oghuz) wintered in Dih-i Naw, apparently identical with
Yangikent in the river’s delta, ḤUDŪD AL-ʿĀLAM/SOTOODEH 1340/1962, § 26, p. 123; ḤUDŪD
AL-ʿĀLAM/MINORSKY 1970, § 26, pp. 122, 371; for the connection between the two toponyms
see Golden, who draws attention to this evidence, GOLDEN 1992, 209.
33 GOLDEN 1992, 211. Likewise, in the cold part of the year Alp-Arslan launched an expedi-
tion against Khwārazm (1065–6) and campaigned against the nomadic tribes from the
steppes between the Caspian and the Aral Seas, some of whom apparently were Qïpchaqs,
see the sources in n. 29.
34 ‫«از جانب خوارزم چند ماه بود تا خوارزمشاه با لشکری عظیم از جانب ثغری که معروف و مشهور است و آن را جند خوانند‬
‫بقعر ترکستان فرو رفته بوده است و خطرها تحمل کرده و با ملکی و مقدمی که او را در میان کفّار بزرگتر دانند او را مالقات‬
‫افتاد و ایزد تعالی بلطف و فطل خویش نصرت و تایید ارزانی داشته تا آن کافر را بشکسته است و هزیمت کرده و خلق بسیار از‬
»‫مقر حویش رسیده‬ّ ‫( ایشان بکشته و غنایم و سبی و مال بی قیاس او را روزی بوده و در ضمان سالمت با‬from Khwārazm:
several months ago the Khwārazmshāh, with an immense army, entered the depth of
Turkistān from a frontier, which is famous and glorious and is called Jand. [He] endured
since it is reflected in a letter from the summer of the same year, in which it is said
to have taken place “several months ago”. 35 This must have been a successful raid
against the nomadic camps beyond the Syr Darya and the prominent leader of the
unbelievers was most probably one of the numerous Cuman-Qïpchaq chiefs. Dur-
ing his reign, Atsïz organized a number of campaigns against the steppes inhabited
by the Cuman-Qïpchaq tribes,36 and although in the rest of the cases there is no
preserved information about the season of fighting, one can assume that most of
them must have taken place during the cold period of the year.
Yet, the relations of the nomads with the Khwārazmshāh were not limited ex-
clusively to the sphere of mutual conflicts. The Cuman-Qïpchaqs apparently pro-
vided contingents for the troops of Atsïz, again in the cold seasons. This is revealed
by the fathnāma37 circulated by Sultan Sanjar after an autumn-winter campaign
against the Khwārazmshāh, who also happened to be Sanjar’s unruly vassal.38 Ac-
cording to this document, in the defeat that he suffered under the walls of the
Hazārasp fortress, Atsïz lost “nearly ten thousand men Turks, some [of them] from
the infidels,39 who were among the auxiliaries and allies of the governor who had
gone astray [i.e. the Khwārazmshāh]”.40 Most probably, these infidels were mem-
bers of the Cuman-Qïpchaq tribal community41 most of whom remained faithful to
their traditional pagan religion until the very end of their domination in Dasht-i
Qipchāq.42
perils and encountered a ruler and chief who was considered the most powerful among the
infidels. And the Most High God with His grace and justice rendered [His] assistance and
support so that infidel was broken and defeated and many people from them [the infidels]
were killed. He [the Khwārazmshāh] won an unprecedented [quantity of] booty, captives,
and wealth and safely arrived at his residence.), BARTOL’D 1898, 37; see also BARTOL’D 1900, 347.
35 See the source text in the above note and the arguments of Paul, who believes that the
campaign probably took place during the winter of 1132/3, PAUL 2013, 93, n. 54.
36 JŪZJĀNĪ/HABĪBĪ 1343/1964, vol. 1, 299; AL-JUZJANI/RAVERTY 1970, 236–7; JUVAYNĪ/QAZVĪNĪ
1916/1334, 10, 12; JUVAINI/BOYLE 1958, 284, 286; see also RASHĪD AL-DĪN/RAWSHAN 2010, 3.
37 I.e., victory proclamation in Buniyatov’s words, BUTINYATOV 1986, 10.
38 This is the first of several campaigns that Sanjar led against Khwārazm trying to affirm his
power in this isolated region. The Seljuq ruler started his expedition in the autumn of 1138
and came back to his capital Nīshāpūr in February 1139, see n. 30.
39 In this regard, see Paul’s thoughts: PAUL 2015, 145.
ّ ‫ «قرب ده هزار مرد ترک بعضی از کفّار که از انصار و اعوان مدیر ضا‬, BARTOL’D 1898, 45.
40 »‫ل بودند‬
41 PAUL 2015, 145.
42 GOLDEN 1998b, 223–6. Even as late as the 1180s, the Qïpchaq allies of the Khwārazmshāhs
were still pagans, although they were already in-laws of the Anushteginids, PAUL 2015, 147;
KAFESOĞLU 2000, 94, n. 76; see also: BARTOL’D 1900, 365; BIRAN 2005, 61. By the late 12th and
early 13th century the contacts with the Anushteginid Dynasty led to the Islamization of
some Cuman-Qïpchaq tribes and especially their elite. Yet, on the whole, the inhabitants of
Dasht-i Qipchāq continued to follow their traditional beliefs until the arrival of the Mongols
and even after that.
A letter by Atsïz’s court poet, Rashīd al-Dīn Vaṭvāṭ, written soon after the death
of the ruler, however, suggests that the appearance of the steppe nomads on
Khwārazmian territory was by far not always connected with the duties of merce-
nary or allied troops. It seems “the infidels” did not always remain mere passive
spectators to the Khwārazmshāh’s numerous steppe campaigns. According to this
source, the inhabitants of Khwārazm also had to bear the incursions of the said
“unbelievers”, which, unsurprisingly, occurred in the winter. In his letter, Vaṭvāṭ
gives the following explanation for the Khwārazmian troops’ failure to support
Sanjar, who had just saved himself from Oghuz captivity: 43 “in the winter, this
frontier is fearful from the blows of the infidels (may God abandon them!), espe-
cially now when the lord, the late malik (may he rest in peace!) has passed in the
proximity of the Almighty God, and the infidels became more confident due to his
death.”44 Vaṭvāṭ also notes that the Khwārazmian troops, who were still occupied
with a campaign against the unidentified settlement of Sāq, will arrive in Khurāsān
“as soon as the winter passes away and the vanguard of spring becomes evident
and secures the frontier of Khwārazm against the evil[doings] of the infidels (may
God abandon them!)”.45
Naturally, the court poet tried to excuse the Khwārazmian leadership’s failure
to provide help for their overlord in Khurāsān by using all kinds of arguments, and
it is entirely possible that he overstated the problems caused by the nomads. Yet,
despite this possibility, Vaṭvāṭ’s letter undoubtedly refers to real or at least realistic
political circumstances from the mid-12th century. Furthermore, the quoted passag-
es leave the impression of regular nomadic incursions against Khwārazm during
the winter and of a relief with the advance of the spring, which provided protec-
tion against “the evil[doings] of the infidels”. Thus, the steppe dwellers’ pressure
had strictly defined (and hence predictable) seasonal intensity and the warming of
the weather could limit it. Therefore, it was precisely the seasonal cycle that de-
fined the dynamics of the conflicts between the Cuman-Qïpchaqs and their settled
neighbours in Central Asia—both within the steppe and in the territories of the
sedentarists.
The rule of the last Khwārazmshāhs was characterized by the growing im-
portance of the Qïpchaq contingents for the successful expansion of the dynasty.
The sources for this period continue to mark the seasonal intensity of the nomadic
activities and become more specific as regards the tribal identity of the steppe
43 By the end of his long reign Sanjar was defeated by his Oghuz subjects and spent three
and a half years as their captive, TOR 2000/2010; ÖZAYDIN 2009, 510–11.
44 ُ ‫ردَ هللا‬
َّ َ‫ ب‬،‫ ملک ماضی‬،‫خداوند‬
ّ ُّ ‫ این‬،‫«بوقت زمستان‬
‫ خاضّه در ین وقت کی‬،‫ خوف باشذ‬،‫ خَذ َلَ ُهم ُهللا‬،‫خطۀ را از صدمات کفّار‬
».‫ و کفّاربسبب وفات او پر و بال کرده اند‬،‫ انتقال کرده است‬،‫ تَعالی‬،‫ بجوار ایزذ‬،ُ‫ض َجعه‬ ْ ‫ َم‬, VAṬVĀṬ/TŪYSIRKĀNĪ
1338/1960, 128. Jürgen Paul brings attention to this fragment of Vaṭvāṭ’s correspondence,
PAUL 2013, 108.
45 »‫ ایمن گردذ‬،ُ‫ خَذ َلَهم هللا‬،‫خطۀ خوارزم از غوائل کفار‬
ُْ ُّ ‫ و‬،‫ و طالئع بهار پیذا آیذ‬،‫ «چندانک زمستان بگذرذ‬,
VAṬVĀṬ/TŪYSIRKĀNĪ 1338/1960, 128.
dwellers. First, several letters of Khwārazmshāh ʿAlāʾ al-Dīn Tekesh (r. 1172–1200)
to other Islamic rulers should be pointed out. They have been preserved in a collec-
tion of documents of the Khwārazmian senior administrative official al-Baghdādī.
In these letters, it becomes evident that for at least two subsequent winters in the
early 1180s the Qïpchaq in-laws of Tekesh—Alp Qara Uran46 and his son Qiran—
appeared with their troops in the surroundings of Jand. Their aim was to join the
Khwārazmian contingents in this region and to head together with them against
the territories of the Qara Khitai Empire (1124–1218) in the region of Ṭarāz.47 Fur-
thermore, in a letter to the Atabeg of Azerbaijan, Jаhān Pahlvān (r. 1175–86), from
November 1182 (written about a month after the evidence for the second appear-
ance of Alp Qara Uran in the vicinities of Jand) the Khwārazmshāh boasts that he
continues to enjoy the support of a numerous Qïpchaq army from the furthermost
parts of Turkistān.48
Although evidence for only two consecutive years is preserved, perhaps it re-
flects the existence of a long-lasting practice of seasonal military campaigns. The
advent of the Qïpchaqs in the region of Jand in two subsequent winters coincides
entirely with the seasonal migratory rhythm of the Eurasian nomads. In fact, the
rhythmic appearance of their troops raises the questions whether Tekesh could
afford to abort the seasonal campaign and what the alternative would have been
for the large nomadic forces already situated on the border of his realm? Or, in
other words, was Tekesh the initiator of the allied military activity or was he just
trying to channel it in a politically favourable direction? Regrettably, the answers
to these questions remain elusive.
The establishment of an alliance with the Cuman-Qïpchaq tribes and the suc-
cessful Khwārazmian expansion in the basin of the Syr Darya did not guarantee a
smooth coexistence with the nomads. The fear of the imminent nomadic menace
that Rashīd al-Dīn Vaṭvāṭ mentioned decades earlier was apparently still to be felt
in the Central Asian oases. It is hardly by chance that in the early 1180s Tekesh
planned a campaign in Khurāsān “with great armies […] from the remotest lands
of Islam and the frontiers of the land of Khifjāq” precisely at the end of the win-
46 Apparently, this was the leader of the powerful Qïpchaq grouping Uran, which has been
mentioned as allied to the Khwārazmshāhs also by other sources, and from which—
according to some medieval accounts—originated Terken Khatun, the wife of Tekesh, JU-
VAYNĪ/QAZVĪNĪ 1916/1334, 109, 198; JUVAINI/BOYLE 1958, 378, 465; see also: RASHĪD AL-
DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol. 1, 505–6; RASHĪD AL-DĪN/KARĪMĪ 1338/1959, vol. 1,
366; RASHĪD AL-DĪN/THACKSTON 1998–9, part 2, 250; RASHĪD AL-DĪN/SMIRNOVA 1952, 209–10;
RASHĪD AL-DĪN/RAWSHAN 2010, 33; MĪRKHWĀND 1339/1960, 407. For this tribal grouping see
GOLDEN 1995-1997, 117–18; KÖPRÜLÜ 1943, 227–43.
47 BAGHDĀDĪ/BAHMANYĀR 1384/2005–6, 158, 174–5.
48 BAGHDĀDĪ/BAHMANYĀR 1384/2005–6, 180; see also BARTOL’D 1900, 366.
ter.49 Indeed, early spring was the usual time of the year when the Khwārazmshāh
departed with his troops towards this region. 50 And yet, there can barely be any
doubt that the geographical conditions in Khurāsān were not the only reason for
Tekesh’s unwillingness to leave with his troops the core region of his realm before
the end of the winter, which was also the peak season of nomadic military activity.
The letters in al-Baghdādī’s collection reflect a period when the Qïpchaqs still
arrived for seasonal campaigns and then retreated in the steppes, much like the
Cuman allies of the first Asenids in Bulgaria. In the following decades, the nomads
migrated permanently within the Khwārazmian territories and their participation
in the imperial campaigns did not follow such a strict seasonal rhythm. But this
was not the case with the usual conflicts between the Anushteginids and the Cu-
man-Qïpchaq tribes that—despite the strengthened partnership and the nomads’
integration within the imperial military-political system—did not cease at all. As it
seems, the campaigns of the Khwārazmshāhs in Dasht-i Qipchāq followed to a
large extent the same natural cycle, perhaps driven by the necessity to deliver a
blow upon the nomadic encampments in the time of the year when they were most
vulnerable –both in terms of location and cattle condition.51
Events that took place in the late 12 th century are indicative in this regard. After
the winter of 1194/5 passed, Tekesh personally led an expedition towards “Sïgh-
naq and its surroundings” with “the intention to wage ghazā against Qadïr Buqu
Khan“.52 After the Khwārazmian troops crossed the Qïzïl Qum and reached Jand,
the nomadic chief started to withdraw while Tekesh chased him. The Khwārazm-
shāh, however, ignored the dangerous fact that among his troops were many war-
riors of the Qïpchaq tribe Uran, some of them in his own retinue. The latter con-
tacted Qadïr Buqu Khan and promised him that if he faces Tekesh they will desert
him on the battlefield. During the battle that followed on May 18, 1195 the Urani-
ans kept their promise, thus causing the Khwārazmshāh’s severe defeat.53
Apparently, Qadïr Buqu Khan was a leader of a steppe grouping whose winter
pasturages were situated in the surroundings of Sïghnaq, while Tekesh’s move-
49 »‫( «با لشکرهای گران که از اقاصی بالد اسالم و تخوم دیار خفجاق در سلک خدمت منتظم شده باشند‬with great armies,
which are ordered in the line of service from the remotest lands of Islam and the frontiers of
the land of Khifjāq), BAGHDĀDĪ/BAHMANYĀR 148. According to Bartold, the letter was writ-
ten in January 1182, whereas Kafesoğlu dates it to January 1181 on the grounds of its histori-
cal context, BARTOL’D 1900, 181; KAFESOĞLU 2000, 93, n. 69.
50 As can be inferred from another letter of Tekesh, BAGHDĀDĪ/BAHMANYĀR 1384/2005–6,
175–6.
51 For the vulnerability of the animals in certain periods of the year, see below in the text.
52 JUVAYNĪ/QAZVĪNĪ 1916/1334, 34; JUVAINI/BOYLE 1958, 304–5; see also RASHĪD AL-
DĪN/RAWSHAN 2010, 13. For the various forms of this chief’s name and its meaning, see
JUVAYNĪ 1916/1334, 34, n. 2; JUVAINI/BOYLE 1958, 305, n. 67.
53 JUVAYNĪ/QAZVĪNĪ 1916/1334, 34–5; JUVAINI/BOYLE 1958, 304–5; see also RASHĪD AL-
DĪN/RAWSHAN 2010, 13; MĪRKHWĀND 1339/1960, 375–6.

ments fit entirely within the classical seasonal model of sedentary campaigns
against the Cuman-Qïpchaqs. Atsïz’s campaign of 1133 apparently followed the
same model of attacking the nomadic encampments during the winter or early
spring, which is well attested also in the European parts of Dasht-i Qipchāq (see
below). The fact that the decisive battle between Tekesh and his nomadic adver-
saries took place as late as May 18, 1195 perhaps should be explained with the pro-
longed period of time the preceding manoeuvres must have required—i.e. the re-
treat of the steppe troops and the chasing that followed.
Be that as it may, Tekesh apparently planned to avenge himself in the early
spring of 1197, but according to Juvaynī, after hearing the news about the death of
his heir Malik-Shāh, the broken monarch gave up his plans for ghazā’.54 Most prob-
ably, the unbelievers against which the aborted campaign was targeted were the
Cuman-Qïpchaq tribes from the Syr Darya basin whose recent victory had shaken
the positions of the Anushteginids in the region. Yet, the possibility that the expe-
dition was aimed at the Qara Khitay cannot be completely ruled out. The military
actions that took place in the next year represent a classical instance of the seasonal
pattern of the Khwārazmian steppe campaigns. During Rabīʿ al-Awwal (January–
February) and Rabīʿ al-Ākhir (February–March) AH 594 (AD 1198), Tekesh and his
new successor—the future Khwārazmshāh ʿAlāʾ al-Dīn Muḥammad II (r. 1200–
20)—took advantage of an internecine strife within the steppe. They intervened
through a relatively brief—yet large-scale and successful—expedition against the
nomads to the north of the Syr Darya.55 During his own reign, Muḥammad II also
continued to lead campaigns against the Cuman-Qïpchaqs, but regrettably the
sources do not give any information regarding the particular time of the year when
this happened. On the other hand, the Mongol invasion in the Khwārazmian Em-
pire followed the century-old seasonal pattern of the nomadic incursions in the
Central Asian sedentary areas: the settlements along the Syr Darya were attacked
during the cold months of 1219/20,56 while the siege operations against the
54 JUVAYNĪ/QAZVĪNĪ 1916/1334, 39; JUVAINI/BOYLE 1958, 309.

55 JUVAYNĪ/QAZVĪNĪ 1916/1334, 40–1; JUVAINI/BOYLE 1958, 309–10; see also RASHĪD AL-
DĪN/RAWSHAN 2010, 14–15; MĪRKHWĀND 1339/1960, 378–9.
56 RASHĪD AL-DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol. 1, 496; RASHĪD AL-DĪN/KARĪMĪ
1338/1959, vol. 1, 359; RASHĪD AL-DĪN/THACKSTON 1998–9, part 2, 245; RASHĪD AL-DĪN/
SMIRNOVA 1952, 203; Rashīd al-Dīn mistakenly dated the events a year later (BARTOL’D 1900,
399; see also the editorial comment in RATCHNEVSKY 1991, 257, n. 151; for the dating of the
beginning of the Mongol invasion, see BARTOL’D 1900, 438), but what matters for the present
topic are the seasons. In fact, Jand was attacked as late as April, but the original Mongol
military plan (‫حساب‬, ḥisāb) had to be modified, JUVAYNĪ/QAZVĪNĪ 1912/1329, 68–9; JU-
VAINI/BOYLE 1958, 88–9.
Khwārazmian capital Gurgānj were commenced in the next autumn and continued
until the city was conquered in the spring of 1221.57
At any rate, the evidence examined so far reveals that the coexistence of the
Cuman-Qïpchaqs and Khwārazm, and the military conflicts between them in par-
ticular, were defined to a significant extent by a seasonal cycle resembling the one
noticed by Rasovsky with regards to the Cuman presence in the Balkans. In this
case, too, the nomadic military activity was intensive during the cold part of the year
(though, due to the moderate climate, the war season in Southeast Europe seems to
have been longer, including the late spring and the early autumn). Again, it was
the same period that witnessed the Khwarazmian expeditions against the nomadic
encampments in the steppe, which, of course, had no parallel in the Balkans. 58
Naturally, exceptions to this clear pattern of warfare in the cold months were
documented in Central Asia, too. In the mid-12th century, Atsïz started a rather late
expedition officially aimed at Sïghnaq on the left bank of the Syr Darya, in whose
surroundings were the winter campgrounds of some Cuman-Qïpchaq tribes. Alt-
hough these plans were nothing more than a pretext for the surprising conquest of
the city of Jand, which was much closer to Khwārazm, it is still noteworthy that
according to Juvaynī the Khwārazmshāh set off on the campaign as late as Muhar-
ram AH 547 (April–May AD 1152).59 Even more indicative is an episode from the
reign of Atsïz’s grandson Tekesh. As already mentioned, in 1198 the latter ruler
interfered in a Cuman-Qïpchaq internecine strife and won a victory, which af-
firmed the position of his protégé within the steppes, while the chief who used to
dominate before the conflict was brought to Khwārazm enchained. Yet, it soon
turned out that the surviving followers of the defeated leader gathered around his
rival, who had just been supported by Tekesh, and continued to challenge the
Khwārazmian power in the Syr Darya basin. In the end, the Khwārazmshāh was
forced to instigate the steppe inner strife again, but this time he backed the oppo-
site site. Tekesh freed his noble captive and sent him to the steppes at the head of
“a large army” while he himself set off towards Khurāsān and ʿIrāq-i ʿAjam. What
is important for the present topic is that according to Juvaynī the ruler came back
to Khwārazm with the enchained chief in Rabīʿ al-Ākhir AH 594 (February–March
AD 1198), and the latter was freed before the Khwārazmshāh arrived in Khurāsān
on 2 Dhū l-Ḥijjah AH 594 (October 5, AD 1198).60 Therefore, the twist in the behav-
iour of the Khwārazmian protégé must have taken place within this timespan and

1338/1959, vol. 1, 371; RASHĪD AL-DĪN/THACKSTON 1998–9, part 2, 253; RASHĪD AL-DĪN/
SMIRNOVA 1952, 215; see also AL-NASAVĪ/BUNIYATOV 1996, 131–3; SĪRAT-I DJALĀLUD-
DĪN/MĪNUVĪ 1986, 123–5.
58 For further details, see below.
59 JUVAYNĪ/QAZVĪNĪ 1916/1334, 10; JUVAINI/BOYLE 1958, 284.
60 JUVAYNĪ/QAZVĪNĪ 1916/1334, 41; JUVAINI/BOYLE 1958, 310; see also RASHĪD AL-DĪN/RAW-
SHAN 2010, 15; MĪRKHWĀND 1339/1960, 379.

his Qïpchaq followers must have menaced the Khwārazmian domains during the
hot months of the year.
Exceptions could also be observed as regards the arrival of steppe nomadic al-
lies in the Khwārazmian army, as indicated by other events during the reign of the
same monarch. In his letter to the ruler of Ghūr from 1183, Tekesh mentions
“50,000 sword-bearing Turks” in his troops during a campaign in Khurāsān. There
can hardly be any doubt that such a massive number of warriors (even if exagger-
ated) could only be provided by the tribes of Dasht-i Qipchāq, and it is noteworthy
that the expedition started in the first decade of Rabīʿ al-Ākhir AH 579—i.e. at the
very end of July or the beginning of August AD 1183. 61 Thus, despite the unwill-
ingness of the Central Asian nomads to fight during the hot months, the circum-
stances could demand military operations that saw the participation of steppe con-
tingents during the climax of the heat in this harsh region.
All of these instances clearly indicate that in Central Asia, too, the climatic
frame that defined the seasonal migrations of the Cuman-Qïpchaqs and the rhythm
of their military activity should not be viewed as an absolutely deterministic factor
for their behaviour.
Kievan Rus’
Prima facie, it appears that the similar seasonal peaks of the Cuman-Qïpchaq mili-
tary activity in Central Asia and the Balkans must have been caused by similar
reasons. Before proceeding with the analysis of the specific factors, however, one
should examine the most thoroughly documented relations of these nomads with a
sedentary power—their contacts with Rus’. It is hardly surprising that the Rus’
chronicles have preserved the most detailed information regarding the seasons in
which the Cumans conducted their campaigns, as is also the case with a number of
other aspects of their history. The data is so abundant that in order to be properly
illustrated it is presented in a table (see the Appendix). Its content is by no means
exhaustive and further contextual investigation of every account of the Cumans in
the Rus’ sources would perhaps extend the source base. Yet, even at the present
state of research the data is more than enough to allow firm conclusions regarding
the seasons in which the Cumans waged war on Rus’ territory—both as independ-
ent nomadic raiders and as allies of the Rus’ princes.
The data in the table demonstrates that in certain periods of the nearly two-
century-long Rus’-Cuman cohabitation, nomadic activity was more intensive in
either the hot or the cold period of the year, but altogether it is impossible to out-
line a clear seasonal model. Contacts, mostly of a military nature, are documented
in every period of the year and even in different seasons of the same year. 62 Fur-
61BAGHDĀDĪ/BAHMANYĀR 1384/2005–6, 200; see also BARTOL’D 1900, 367.

62For example, s.a. 1095 (PSRL, vol. 1, cols. 227–9; PSRL, vol. 2, cols. 217–19); 1096 (PSRL,
vol. 1, cols. 231–4, 239–40; PSRL, vol. 2, cols. 221–4, 229–30); 1107 (PSRL, vol. 1, cols. 250,
thermore, in a number of cases the chronicles mention several different nomadic

incursions in the Rus’ lands under a given year, but only for one of them the pre-
cise time of the year can be identified. Hence, the Cumans were active also out of
the season, which is specifically mentioned in connection with their appearance.
Thus, their activities in the Rus’ principalities did not follow a strict seasonal
rhythm and—unlike the Balkans and Khwārazm—in this vast contact zone there
was no clear correlation between their arrival and the time of the year.
Such a conclusion is supported by the fact that throughout their long domina-
tion in Dasht-i Qipchāq the nomads were ready to react ad hoc to the political re-
versals in the Rus’ principalities. Thus, in 1093, 1113, and 1125, on learning about
the death of the great princes ruling in Kiev (Vsevolod I Yaroslavich, Svyatopolk II
Izyaslavich, and Vladimir II Monomakh, respectively), Cuman troops quickly set
off to the Rus’ lands.63 In the same way, the nomads took advantage of a sudden
political crisis in 1096,64 and when nearly a century later the notorious expedition
of prince Igor Svyatoslavich ended in disaster (1185), the Cumans immediately
organized a large-scale counteroffensive.65 Particularly indicative for the absence of
seasonal limitations on the Cuman activities are the events that, according to the
compiler of the Kievan Chronicle, took place in the winter of 1193. The chronicler
reports that the Rus’ princes guarded their lands on the steppe frontier for a long
time, but as soon as the word for their withdrawal spread, a nomadic attack fol-
lowed. Therefore, the steppe dwellers patiently awaited the retreat of the princes’
troops in order to take advantage of their absence. 66 Undoubtedly, if there was a
strict correlation between the nomadic activity and the time of the year, the Ru-
rikids would have awaited on the border the arrival of the safety season, like the
Khwārazmians were awaiting it according to Rashīd al-Dīn Vaṭvāṭ. Indeed, the
conclusion that such a clear correlation between the season and the risk of attack
did not exist is supported by the evidence in the Kievan Chronicle for the preceding
281–3; PSRL, vol. 2, cols. 258, 259); 1136 (PSRL, vol. 1, cols. 303–4; PSRL, vol. 2, cols. 297–300;
PSRL, vol. 3, 23–4, 208–9); 1149 (PSRL, vol. 1, cols. 321–6; PSRL, vol. 2, cols. 374–83, 386–93);
1196 (PSRL, vol. 2, cols. 695–6, 698–700; PSRL, vol. 3, 42–3, 235–6). It should be pointed out,
however, that the chronology of the Rus’ chronicles follows both the March-style and the
ultra-March style, which is why not all events given under the respective years coincide with
the modern years starting in January, and thus some of them may have in fact taken place in
different years according to the Gregorian calendar. Be this as it may, there is no doubt that
in the articles for the cited years the Rus’ chroniclers registered Cuman activities in different
seasons, consecutive or closely following each other.
63 PSRL, vol. 1, cols. 218, 250, 295–6; PSRL, vol. 2, cols. 209, 276, 289–90.
64 PSRL, vol. 1, cols. 229–34; PSRL, vol. 2, cols. 220–4.
65 PSRL, vol. 1, col. 399; PSRL, vol. 2, cols. 645–9.
66 PSRL, vol. 2, cols. 675–9 especially col. 679. Judging by the version of the Laurentian Chron-
icle for the events in 1185, the Cumans may have applied a similar manoeuvre in this case,
too (PSRL, vol. 1, col. 399).
year, 1192. Under that year the anonymous chronicler reports that the princes
gathered with their troops to guard “the Rus’ Land”, but this time they spent “the
whole summer” on the steppe frontier. 67
In fact, if there is certain visible influence of the seasonal cycle upon the intensi-
ty of the military conflicts between the Rus’ and the Cumans, it is related not to the
nomadic campaigns in the sedentary territories, but on the contrary—to the expe-
ditions of the Rus’ princes inside the steppe. As modern scholars have already
pointed out, the attacks of the Rurikids against the nomadic encampments were
strictly related to the time of the year.68 An examination of the chronicles reveals
that these expeditions as a rule were organized during two periods of the year: the
first was winter and early spring in particular; 69 the second was the beginning of
April, but this month was chosen more rarely and it is noteworthy that most of the
campaigns that started at that time were unsuccessful.70 These two traditional sea-
sons for expeditions in the Cuman steppe were separated by the spring muds of
the so-called rasputitsa (ca. mid-March–April). It rendered military operations in
the grasslands practically impossible 71 and turned into a serious trouble for the
67 PSRL, vol. 2, col. 673.

68 RASOVSKY 1939, 210-11; see also the instances, pointed out by Rasovsky, RASOVSKY 1938,
168–1, 173–4; PLETNEVA 1990, 114; as well as the considerations for the Rus’ seasonal cam-
paigns discussed by Lev Gumilyov, who, however, wrongly attributed to the Cumans the
practice of haying, GUMILEV 1992, 81.
69 For example, s.a. 1109 (PSRL, vol. 1, cols. 283–4; PSRL, vol. 2, col. 260); 1110 (PSRL, vol. 1,
cols. 284, 289; PSRL, vol. 2, col. 260; the expedition was aborted on reaching the steppe fron-
tier and the exact time in which it took place is not specified, it is only mentioned that it was
“during the winter”); 1111 (PSRL, vol. 1, col. 289; PSRL, vol. 2, cols. 264–8, 273); 1152 (PSRL,
vol. 1, cols. 339–40; PSRL, vol. 2, cols. 459–61); 1168 (PSRL, vol. 2, col. 532); 1170 (PSRL, vol.
2, cols. 538–41; PSRL, vol. 3, 32–3, 220); 1183 (PSRL, vol. 2, cols. 628–9); s.a. 1187 there are
three reported Rus’ campaigns in the steppe: one in spring and two in winter (PSRL, vol. 2,
cols. 652–3, 653–4, 659); 1190 (PSRL, Vol. 2, cols. 670–2); s.a. 1191 two expeditions took place:
one in an unspecified time and one in winter, (PSRL, vol. 2, col. 673); s.a. 1193 one futile
attempt to organize a winter raid and another successfully launched incursion in the same
season are reported (PSRL, vol. 2, cols. 676–8); 1202 (PSRL, vol. 1, col. 418); 1205 (PSRL, vol.
1, col. 420; PSRL, vol. 3, 240).
70 During that time apparently started the successful raid by Roman Nezdilovich, noted by the
Kievan Chronicle s.a. 1185 (PSRL, vol. 2, col. 637); the ill-fated campaign of Igor Svyatoslavich
began on April 23, 1185 (PSRL, vol. 2, col. 637); Vsevolod the Big Nest started his expedition
on April 30, 1199. He, however, was not able to catch the nomads (PSRL, vol. 1, cols. 414–15,
PSRL, vol. 7, 106–7). The expedition of the Rus’ princes and the Cuman chiefs against the
Mongols, which ended with the defeat near the Kalka River, also started in April 1223
(PSRL, vol. 2, col. 741).
71 As indicated by the events in the late winter and early spring of 1185. Then, the Rus’ sent
Kuntuvdey (one of the leaders of their steppe foederati known as the Black Hats) after the
defeated Cuman chieftain Konchak, but the chasing was completely hampered by the raspu-
Rus’ troops that were unwise enough to enter the plains during that time of the
year.72 However, the aforementioned distinction between the starting seasons is
purely conditional and perhaps depended on the climatic specifics of each year—
there were Rus’ expeditions that started at the apex of the rasputitsa in late March
and early April.73 The spring snow-melt floods of the rivers (the so-called polovod’e)
could also hamper a Rus’ raid in the steppe. 74 Yet, if the nomads turned to be on
the wrong river bank, it could also cause their thorough defeat. 75
In fact, in the case of the Rus’ steppe campaigns, too, the seasonal cycle was not
entirely deterministic. There are some rather rare exceptions in the hotter part of
the year,76 but the tendency for campaigns against the Cumans during winter and
spring is clearly visible in the sources.
The nomads migrated along the banks of the great Eastern European rivers—
the Dnieper, Don, Volga, and Ural—looking for fresh pasture to the north in the
summer and for sheltered camping grounds near the sea to the south during the
winter. Such was Carpini’s evidence for the early Mongol period, corroborated also
by Rubruck.77 It can be assumed with confidence that the same practice was fol-
titsa (PSRL, vol. 2, col. 636). It was again the rasputitsa that prevented Igor Svyatoslavich
from joining the coalition of Rus’ princes that defeated Konchak, PSRL, vol. 2, col. 637.
72 The eloquent description of Igor Svyatoslavich’s unsuccessful attempt to join the rest of
the Rurikids in the spring of 1185 is indicative in this regard (see the previous note).
73 For instance, s.a. 1103 (PSRL, vol. 1, col. 277–9; PSRL, vol. 2, cols. 253–5)
74 For instance, s.a 1187 for the Dnieper (PSRL, vol. 2, col. 652). The period of the breaking of
the river ice (the so-called Ledokhod) could also be an obstacle, for instance s.a. 1190 for the
Dnieper (PSRL, vol. 2, cols. 670–2). It is noteworthy that in both cases the Cumans did not
have troubles crossing the river under these harsh conditions.
75 For instance, s.a 1183, PSRL, vol. 2, cols. 628–9.
76 For instance, s.a 1183/1185 the Rus’ chronicles report a large Rus’ steppe campaign in the
middle of the summer (PSRL, vol. 1, cols. 394–6; PSRL, vol. 2, cols. 630–2), and by the same
time the Olgovichi launched their own raid (PSRL, vol. 2, col. 633). The ill-fated expedition
of prince Igor Svyatoslavich, which ended with the defeat near Kayala in 1185, took place in
late spring (PSRL, vol. 1, cols. 396–8; PSRL, vol. 2, cols. 637–44). The unfavourable for the
Rus’ outcome of the battle in its turn thwarted other princes’ plans for a summer campaign
in the steppe (PSRL, vol. 2, cols. 644–6). The aforementioned expedition of Vsevolod the Big
Nest started in end of April and the prince returned to Vladimir as late as the beginning of
June (PSRL, vol. 1, cols. 414–15; PSRL, vol. 7, 106–7). About the same time of the year, the
Rus’ princes and the Cumans launched their allied campaign against the Mongols, which
ended with the disastrous defeat near the Kalka River on May 31, 1223 (PSRL, vol. 2, col.
741; PSRL, vol. 3, 63, 267). There is even some evidence for a futile attempt to organize an
expedition in the steppes during the autumn (s.a. 1192, PSRL, vol. 2, cols. 673–4), but on the
whole this season seems to have been the least suitable for such undertakings due to the fact
that by that time the nomadic horses fattened up and the steppe dwellers were traditionally
most prepared for warfare precisely during that part of the year.
77 DI CARPINE/DAFFINÀ ET AL. 1989, 309; THE MONGOL MISSION/DAWSON 1966, 55; DI
RUBRUK/CHIESA 2011, 66, 90, 292; DE RUBRUC/VAN DEN WYNGAERT 1929, 197–8, 212, 315–16;
lowed during the period of Cuman-Qïpchaq domination in the Western Eurasian

Steppes. This is suggested by the account of Ibn al-Athīr for the existence of sea-
sonal pasture grounds in Dasht-i Qipchāq at the time of the first Mongol appear-
ance in Europe (1222–3), and it is noteworthy that according to him the winter
pastures were near the seacoast.78 Another Arabic chronicler, al-Nuwayrī, reports
that the Qïpchaq tribes lived in tents and “spent the summer in one place and win-
ter in another”.79 Furthermore, the Eurasian nomads were very conservative in
their approach to natural resources, 80 so there are no grounds to suspect that the
Mongol invasion brought any lasting changes in the seasonal cycle of the nomadic
migrations. Hypothetically, this would imply that by the time of the Rus’ attacks in
winter and spring, the Cuman camps were not in the nearest possible positions to
the frontiers of the Rus’ Land. Yet, in this regard, too, one should not adopt a de-
terministic approach: under 1187 the Kievan Chronicle reports that the Cumans
were close to the Rus’ frontiers during the spring and this led to the organization of
a Rus’ campaign,81 whereas the account of the steppe expedition of Vsevolod the
Big Nest makes it clear that not all Cuman winter camping grounds were close to
the sea.82
Be that as it may, the main factor according to which the moment for the attack
was chosen was not the distance, but the vulnerability of the nomads in the winter
and early spring. On the one hand, the relatively strict dependence of their migra-
tory routes upon the seasonal cycle made their location rather predictable. On the
other hand, the mounts and the cattle, which were not fed with fodder and sur-
vived entirely on pasture, were weakest at this particular time of the year, when
THE THE MISSION/JACKSON 1990, 109, 130, 258. Carpini explicitly states with respect to the
Mongol chiefs, who migrated along the four great rivers: “Omnes isti in hyeme ad mare
descendunt, et in estate super ripam eorumdem fluminum ascendunt ad montes.” (All these
[men] descend towards the sea in the winter and in the summer ascend towards the moun-
tains along the bank[s] of the same rivers.), DI CARPINE/DAFFINÀ ET AL. 1989, 309; THE MON-
GOL MISSION/DAWSON 1966, 55; Rubruck is even more specific regarding the seasonal cycle:
“A ianuario enim usque ad augustum ascendit ipse et omnes alii versus frigidas regiones, et
in augusto incipiunt redire.” (Indeed, from January until August he [Batu] and all the others
ascend to the cold regions, and in August they begin to turn back.), DI RUBRUK/CHIESA 2011,
90, see also 66; DE RUBRUC/VAN DEN WYNGAERT 1929, 212, see also 197–8; THE MISSION/
JACKSON 1990, 130, see also 109. Rasovsky was the first to note the seasonal nature of the
Cuman migrations and to bring scholarly attention to the source passages referred to here
and in the next note, RASOVSKY 1938, 166.
78 IBN AL-ATHĪR 1884, 26; IBN AL-ATHĪR/RICHARDS 2008, 223; see the entire passage below, n. 156.
79 AL-NUWAYRĪ 1884, 539–40.
80 KHAZANOV 1974, 213–19.
81 PSRL, vol. 2, col. 652.
82 PSRL, vol. 1, cols. 414–15; PSRL, vol. 7, 106–7; see also: RASOVSKY 1938, 171.
their energy reserves lowered to a minimum.83 This unfavourable combination

reduced both the military potential of the nomadic groupings and their capabilities
to manoeuvre with their families and herds, and given the harsh winter weather
sometimes it could have devastating consequences for the inhabitants of the at-
tacked camps.84 Thus, it appears that the seasonal cycle could have been decisive
when the nomads were the target and not the initiators of an assault, and it is not
by chance that on the other edge of the vast Dasht-i Qipchāq the Khwārazmians
launched their raids into the steppes during the same time of the year. 85
The opportunity provided by the winter-spring period was of such importance
for the success of the undertaking that the Rus’ princes insisted to campaign pre-
cisely in that time despite the objections of their retainers that the season is unsuit-
able for their own subjects due to its coincidence with the ploughing time.86 On the
other hand, missing the right moment could prove fatal for the raiders themselves.
Due to this very reason the scouts sent to the steppes by the Rus’ princes in May
1185 reported that they found the nomads with the following words: “either come
quickly, or we should go home, since [now] it is not our time”. 87 Indeed, this cam-
paign ended with a disaster. Hot weather and blocked access to fresh water also
played a role in the following defeat of Igor Svyatoslavich’s army near Kayala, but
it seems that the main factor was the better organization of the Cumans during the
warmer season.88 Perhaps this was what the scouts meant when they warned the
princes that early May was not “their time”.89 Apparently, in the late 12th century
83 For the influence of the seasonal cycle upon the herds of the Eurasian nomads, see in gen-
eral BARFIELD 1989, 22–3; GOLDEN 1998a, 8.
84 See, for instance, s.a. 1168 (PSRL, vol. 2, col. 532); s.a. 1205 (PSRL, vol. 1, col. 420; see also
PSRL, vol. 3, 24).

85 Nether the Rus’ nor the Khwārazmians were the first sedentary peoples to recognize this
weak point in the living cycle of the nomads and their herds as the most suitable moment to
attack them. As early as the turn of the 6th century Maurice's Strategikon recommends: “[If]
they are Scythians or Huns, they are to be attacked in the month of February or March,
when [their] horses endure the hardship of the winter...” “Σκυθικὸν ἢ Οὑννικόν ἐστι, περὶ
τὸν Φεβρουάριον μῆνα ἢ Μάρτιον ἐπιτίθεσθαι, ὅταν οἱ ἵπποι ἐκ τῆς τοῦ χειμῶνος
κακοπαθείας ταλαιπωροῦσι“, DAS STRATEGIKON/DENNIS 1981, 230; MAURICE'S STRATE-
GIKON/DENNIS 1984, 65.
86 S.a. 1103, PSRL, vol. 1, col. 277; PSRL, vol. 2, cols. 252–3; see also the article of the Hypatian
Chronicle for the year 1111, which perhaps resulted from some contamination with the
events described s.a. 1103, PSRL, vol. 2, cols. 264–5.
87 “или поѣдете борзо . или возвороти[м]сѧ домовь . ӕко не наше єсть веремѧ“, PSRL,
vol. 2, cols. 638–9; PLDR 1980, 352–3.

89 In contrast to the words of the scouts, another passage in the Kievan Chronicle (s.a. 1193)
referring to the winter as the right time for campaigning in the steppes could be pointed out:
“Тоие же зимъı сдоумавше лѣпьшии моужи в Чернъıхъ Клобоуцехъ . и приѣхаша к
Ростиславоу к Рюриковичю . и почаша емоу молвити . поеди кнѧже с нами на вежи
the Great Prince of Vladimir Vsevolod the Big Nest also missed the right time as he
attacked the Cuman winter camping grounds between late April and early June
1199: the nomads withdrew towards the sea and he was unable to catch them. 90 It
is indicative, by the way, that in this instance the expedition was aimed precisely at
the Cuman winter camps.
There may also have been other factors—unknown to the modern researcher—
that led to the organization of steppe raids in April. It should be taken into consid-
eration that the contact zone between the Rus’ principalities and Dasht-i Qipchāq
was a vast one, stretching for endless kilometres throughout Eastern Europe. Alt-
hough the main conflict area was in the steppes along the riverbanks of the Dnie-
per and Donets, there were also expeditions starting in other regions to the north-
east, as indicated by the incursions led by Vsevolod in 1199 and the princes of Rya-
zan in 1205.91 Therefore, the north-eastern location of the starting points could also
have influenced the departure date of the Rus’ troops. Yet, it is noteworthy that all
along the steppe frontier the Rus’ expeditions were generally organized during the
winter and spring, and, as already mentioned, most of those that started in April
were unsuccessful, including the one of the Rus’-Cuman coalition on the Kalka
River in 1223.92
Thus, the best documented contacts of the Cumans with the outside world—
those along the Rus’ steppe frontier—reveal an absence of a correlation between
the intensity of the numerous nomadic campaigns in the Rus’ principalities and the
seasons. On the other hand, there was a clear tendency in organizing the Rus’
steppe expeditions at a precise time of the year—winter and spring. The state of
the Rus’-Cuman contacts is in sharp contrast with the seasonal peaks of Cuman-
Qïpchaq activity in the Balkans and Central Asia and this somewhat surprising
observation undoubtedly demands an explanation.
Comparative analysis
Interpretation of the divergent data for the seasonal cycle’s influence upon the
intensity of the Cuman presence in the outside world is only possible if one con-
siders the typology of the different contact zones along the periphery of Dasht-i
Половѣцкъıӕ веремѧ ти есть […] а такого ти веремени иногда не боудеть . Ростиславъ

Рюриковичь . оулюбивъ рѣчь ихъ . сдоумавъ с ними […] а по дроужиноу свою посла .
рекъ дроужинѣ своеи . веремѧ нъı есть поѣдѣмь . на Половци“ (“The same winter the
notables of the Black Hats [i.e. the tribal nobility] held counsel and came to Rostislav Ryu-
rikovich, and they started talking to him: ‘Come, prince, with us against the Cuman camps,
it is your time […] and there will be no other such time for you.’ Rostislav Ryurikovich was
pleased by their words and held counsel with them […] and sent to his retainers telling
them: ‘It is time for us to march against the Cumans’”), PSRL, vol. 2, cols. 676–7.
90 PSRL, vol. 1, cols. 414–15; PSRL, vol. 7, 106–7.
91 PSRL, vol. 1, col. 425.
92 Although in this particular case the defeat can hardly be related to climatic factors.
Qipchāq. In this typology, Rus’ and Khwārazm occupy central positions, since
these polities both shared a vast contact zone with the steppe and intensively inter-
acted with various Cuman-Qïpchaq groupings. The nomads regularly appeared on
their territory either as raiders or as partners and allies, and this proximity forced
the Rus’ and Khwārazmians to regularly launch their own campaigns into Dasht-i
Qipchāq. This distinguished them from all other sedentary neighbours of the Cu-
man-Qïpchaqs. The Balkans, on the other hand, were a peripheral contact zone
with a relatively limited direct access to the habitat of the Cumans.93 Under partic-
ular circumstances—such as the partnership with the Asenids—the nomads could
often appear on the peninsula, but the intensity of this partnership could be com-
pared with their presence in the Rus’ principalities only for relatively short periods
of time. It is not by chance that there are no records of expeditions of any Balkan
power inside the steppes inhabited by the Cumans.94
Thus, the typological similarities between Rus’ and Khwārazm raise no doubts.
If so, then what is the explanation for the fact that the seasonal peaks of the Cu-
man-Qïpchaq activity in Central Asia find parallels in the peripheral Balkans,
whereas the Cumans intensively appeared in the Rus’ lands all year round? Could
this impression of asymmetric seasonal activity be simply a result of the dispropor-
tion in the source material, which is far more detailed as regards the Rus’-Cuman
relations and not so abundant when it comes to the nomadic presence in
Khwārazm and the Balkans? The answer to these questions is negative, since the
contemporaries of the events both in the Balkans 95 and in Khwārazm96 explicitly
mention the existence of such asymmetric seasonal nomadic activity. Therefore, the
reason for the contrast between Khwārazm and Rus’ should be sought elsewhere.
As is usually the case with premodern societies and nomads in particular—the
answer is related to geography. As previously mentioned, due to the arid nature of
the Central Asian steppes the nomadic migratory routes in the region were far
longer than those in Eastern Europe. This means that the winter and summer pas-
93 GOLEV 2013a, 142–51.

94 Except for one rather unclear evidence in the panegyric of Michael Rethor—written to
celebrate the achievements of Manuel I Komnenos—which is viewed by some authors as an
indication that the emperor initiated an expedition in the Azov area ca. 1152–3. For this
source and its modern interpretations, see GOLEV 2018b, 64–5. When the Byzantines were
forced to seek leverage over the Cumans, they rather looked for partners with а better loca-
tion and more experience in fighting with the steppe tribes than they had. This was the case
with the expeditions of the Rus’ princes against the nomads in the early 13th century (see the
sources in n. 104). The counterattacks of the Byzantines themselves were usually limited to
chasing the retreating Cumans for some distance beyond the Danube, ANNA KOMNE-
NE/REINSCH – KAMBYLIS 2001, 458; ANNA COMNENA/DAWES 2000, 273–4; KINNAMOS/MEIN-
EKE 1836, 93–5; KINNAMOS/BRAND 1976, 76–8.
95 See the accounts of Robert de Clari and Geoffrey de Villehardouin cited below.
96 See the evidence of Rashīd al-Dīn Vaṭvāṭ cited above.
ture grounds of the Cuman-Qïpchaqs in Central Asia were separated by a much

larger space, in comparison to those of their counterparts in the European parts of
Dasht-i Qipchāq. That is why for the Qïpchaqs in the east it was much harder (alt-
hough not impossible) to reach the borders of Khwārazm in the warmer months of
the year, when their camps were at a much longer distance from the sedentary
areas. This specific feature of Asian nomadism coincides with the aforementioned
fact that during the warm seasons the Central Asian steppes turn into almost im-
passable deserts,97 and undoubtedly would explain the much rarer appearance of
the Cuman-Qïpchaqs during that part of the year. For comparison, the steppe zone
of Eastern Europe is much narrower and much more humid than the one in Cen-
tral Asia, and thus the migratory routes were far shorter, while the Rus’ principali-
ties were close at hand, regardless of the particular locations of the camps in any
given season. This would explain the ability of the Cumans to react ad hoc to the
political crises and the royal deaths in the Rus’ principalities, which were men-
tioned above.
On the other hand, the growing partnership between the Cuman-Qïpchaqs and
the Khwārazmshāhs in the last decades of their empire’s existence led the nomads,
who wanted to participate in the lucrative campaigns in Iran and the adjacent terri-
tories, to resettle within the imperial lands.98 Such a mass exodus had no analogue
in the Rus’ principalities, although the Cumans maintained traditional allied rela-
tions with many princely families. This migration could be interpreted as a corrob-
oration of the observation that the extent of the nomadic routes in Central Asia
forced the local Qïpchaqs to spend most of the year away from the zone of their
interests in the Khwārazmian state. At the same time, the Cumans that neigh-
boured Kievan Rus’ as well as the Central Asian Qïpchaqs remained most vulnera-
ble (and in Khwārazm also at their closest location to the sedentary territories) in
the winter and early spring, which is why the expeditions against them were
launched at that time of the year in both contact zones.
What was the position of the Balkans? As already mentioned, they were a rela-
tively peripheral contact zone that was situated rather far from the main concentra-
tions of Cuman camps in Europe along the course of the Dnieper and in the Don
97 Thus, according to Carpini many of Prince Yaroslav Vsevolodovich’s men died of thirst
when travelling through the Central Asian steppes inhabited by Qanglï on their way to the
Mongol imperial court, DI CARPINE/DAFFINÀ ET AL. 1989, 313–14; THE MONGOL MIS-
SION/DAWSON 1966, 58. Juvaynī reports that as a result of a battle with the Qïpchaqs on May
18, 1195, Khwārazmshāh Tekesh’s “army of Islam was routed, many perished under the
sword, and yet more were buried in the dust of the desert on account of heat and thirst”
(‫)لشکر اسالم در انهزام افتادند بسیاری در زیر شمشیر هالك شدند و بیشتری در بیابان از سبب گرما و تشنگی دفین خاك گشتند‬,
JUVAYNĪ/QAZVĪNĪ 1916/1334, 35; JUVAINI/BOYLE 1958, 305. For the harsh natural conditions
in Central Asia, see also Bayhaqī’s evidence cited above, n. 28.
98 ABŪ АL-GHĀZĪ/SABLUKOV 1906, 34; AL-NASAVĪ/BUNIYATOV 1996, 82; SĪRAT-I
DJALĀLUDDĪN/MĪNUVĪ 1986, 62.
basin. Although in this case, too, the dimensions are not comparable to those in
Central Asia, in the late 11th century the nomads still had to cover significant dis-
tances. According to Anna Komnena, the Cumans—who arrived in the Northern
Balkans to support the Pechenegs against the Byzantines but were late for the bat-
tle—pointed out that they had travelled “such a long way”. 99 A century later, at
least part of the Cuman allies of the Asenids were coming from the steppes east of
the Dnieper.100 Yet, despite the distances, even in the times of the successful no-
madic partnership with that dynasty there is no evidence of a mass Cuman migra-
tion into the Balkans which could be compared to the processes in Khwārazm. This
indicates that for the steppe dwellers the abundant booty of the fertile Thracian
fields remained rather close, although not as close as the Rus’ principalities which
immediately neighboured the Cuman steppe.
Two other questions still remain to be answered: what were the reasons for the
pronounced seasonal intensity of Cuman activity in Khwārazm and the Balkans,
and were they identical in both contact zones? Rasovsky associates the rhythmic
appearance of the Cumans in support of the Asenids with the seasonal migration
and the northward movement of their camps in the summer, which took them
away from “the Danube region” during that part of the year.101 By the way, he also
noticed that the Rus’ took advantage of the Cuman expeditions to the Balkans in
order to launch their traditional attacks against the nomads’ unprotected encamp-
ments.102 Indeed, sometimes the chroniclers specifically report that the Rus’ and
their nomadic foederati from the Black Hats seized the opportunity provided by
the Cuman absence “on the Danube” in order to attack their camps, naturally dur-
ing the cold time of the year.103 In this regard, it is noteworthy that despite the Rus’
custom to raid the Cuman encampments during winter and spring, the nomads
continued to be active in the Balkans precisely in that time of the year and obvious-
ly their withdrawal in the summer was not related to these conflicts. When the
Rus’ raids were particularly devastating, the inhabitants of Dasht-i Qipchāq simply
could not arrive to help the Asenids, as indicated by the events in the first years of
99 “οἳ […] τοῖς ἡγεμόσι τῶν Σκυθῶν ἔφασαν ὡς «Ἡμεῖς μὲν τὰ οἴκοι καταλιπόντες εἰς
ὑμετέραν ἤλθομεν βοήθειαν τὴν τοσαύτην ὁδὸν διηνυκότες ἐφ' ᾧ καὶ τοῦ κινδύνου καὶ τῆς
νίκης συγκοινωνοὶ γενήσεσθαι.” (They [the Cumans] […] said to the chiefs of the Scythians
[i.e. the Pechenegs] the following: “Indeed, leaving our deeds at home, we came to your aid,
travelling such a long way, on the [understanding] that we will be partners both in danger
and in victory.), ANNA KOMNENE/REINSCH – KAMBYLIS 2001, 216; ANNA COMNENA/DAWES
2000, 126.
100 PSRL, vol. 2, col. 659, see also cols. 673–4.
101 RASOVSKY 1939, 205.
102 RASOVSKY 1939, 210–11.
103 A successful winter campaign s.a. 1187 and a futile attempt to organize a raid in the au-
tumn s.a. 1192, PSRL, vol. 2, cols. 659, 673–4.

the 13th century.104 There is no evidence, however, that such attacks have caused
mass retreat of the steppe contingents among the troops of the Asenids.
As regards Rasovsky’s hypothesis that the withdrawal of the Cumans during
the summer was a result of the movements of their camps to the north towards the
summer pastures, it must be pointed out that, unlike Khwārazm, in the case of the
Balkans the seasonal migrations could hardly influence nomadic activity south of
the Danube. The peninsula is located far to the southwest of the main migratory
routes along the courses of the great Eastern European rivers, and the variations
caused by the seasonal movements must have not been of significance given that
the distance between the Pontic steppes and Eastern Thrace is rather long anyway.
Therefore, the reasons for the seasonal intensity are related to natural factors,
but these were of a different kind. It is noteworthy that the medieval authors them-
selves also gave “naturalistic” explanations for the seasonal activity of the Cumans.
According to Robert de Clari: “During the summer in their land there are so many
flies and midgets that they do not dare to get out of their tents until the winter
draws near.”105 The French knight adds that the Cumans launched their raids dur-
ing the winter.106 Such a naïve explanation suggests a discomfort during the hot
months. However, there is no logic in the idea that such a discomfort should be
connected to Dasht-i Qipchāq, since this would not hamper Cuman presence in the
Balkans.107 On the other hand, the account of Villehardouin—who was perhaps
better informed of the real reasons behind the seasonal appearance of the no-
mads—leaves the impression that the Cumans twice left the army of the Bulgarian
tsar Kaloyan (r. 1197–1207) under the walls of Adrianople due to the arrival of the
summer at the theatre of operations.108 And, indeed, everything seems to point that
the effects of the summer heat stood behind the Cuman withdrawal both in South-
eastern Europe and Khwārazm. As previously mentioned, the nomadic war season
104 CHONIATES/VAN DIETEN 1975, 522–3; CHONIATES/MAGOULIAS 1984, 287; THEODOROS

SKOUTARIOTES/VOYNOV 1972, 260; the expedition described s.a. 1202 in PSRL, vol. 1, col. 418;
and maybe the expedition, mentioned s.a. 1205, PSRL, vol. 1, col. 420.
105 ROBERT DE CLARI 2007, §LXV, 104–5; the English translation follows the Bulgarian one.
106 ROBERT DE CLARI 2007, §LXV, 104–5.
107 Yet, there is a grain of truth in de Clari’s report about the presence of annoying insects in
the steppes inhabited by the Cumans. Among the advantages of the Mongol seasonal migra-
tions Marco Polo points out that in the “cooler regions” (i.e. mountains and valleys) in the
summer “there are no horse-flies or gad-flies or similar pests to annoy them and their
beasts”, MARCO POLO/LATHAM 1958, 97.
108 “When Pentecost [May 29, 1205] had come, Johannizza, the King of Wallachia and Bul-
garia, had pretty well had his will of the land; and he could no longer hold his Comans
together, because they were unable to keep the field during the summer; so the Comans
departed to their own country”, MEMOIRS OF THE CRUSADES/MARZIALS 1908, 102; see also the
description of similar events in late April or early May 1207: MEMOIRS OF THE CRU-
SADES/MARZIALS 1908, 126.
in the Balkans was longer and included late spring and early autumn since the
peninsula is located in the temperate climatic zone. On the other hand, in South-
west Asia the periods of high temperatures arrive earlier and last longer, making
the harsh landscapes of Iran and Afghanistan particularly inhospitable during that
part of the year. It was already noted that in the Khwārazmian case the influence of
the longer migratory routes should be added to the temperature amplitudes.
Stretched through the arid Central Asian steppes, seasonal migration caused
movement of the nomadic summer encampments far away from the frontiers of
the sedentary world, unlike the situation in the Balkans. Due to the effects of these
factors, the nature of the Cuman-Qïpchaq presence in Central Asia and the Bal-
kans, although similar, was not completely identical.
As regards the unwillingness of the nomadic armies to operate during the hot
seasons, it was far from unique to the Cumans. The Mongol campaigns offer an-
other typical example from the same historical period, although they usually were
of a much larger scale compared to the Cuman incursions into the outside world. It
was as early as the phase of strategic planning that the Mongol commanders took
into consideration the seasonal cycle and strove to avoid the summer heat by
scheduling the beginning of the operations in the early autumn—when the nomad-
ic horses were fattened up and prepared for the incoming physical efforts.109 The
high temperatures in the summer also influenced the course of the ongoing cam-
paigns. The entire Western campaign of Chinggis Khan (1219–23) was regularly
interrupted by pauses in the summer seasons. It was as early as the stage of rede-
ployment and movement toward the Khwārazmian borders that the Mongol army
stopped along the Irtysh for the summer of 1219. 110 After he conquered Utrār, Bu-
khārā, and Samarqand, Chinggis Khan (r. 1206–27) spent the hot months of the
following 1220 encamped in the meadows near Nakhshab, and Juvaynī explicitly
notes that as a result of their stay there “the mounts were fattened, and the troops
rested”.111 At the end of the spring of 1221, while he was besieging Ṭāliqān, the
109 “…on the fifth of the fifth moon, they hold an assembly with feasting [celebrating the
beginning of the summer] and plan together against whom the campaign will be that fall.
Then everyone returns to his domain to avoid the summer heat and raise their livestock [i.e.
the horses]. When the eighth moon [roughly September] comes, they all gather at Yanjing
and then set out.”, [the translation of this passage has been slightly revised by Dr. Landa]
FIVE CHINESE SOURCES 2021, 83; MENG-DA BEILU/MUNKUEV 1975, 68; see also the sources
regarding the debate, which preceded Chinggis Khan’s campaign against the Naimans in
1204, n. 155.
110 BARTOL’D 1900, 434; THE SECRET HISTORY/DE RACHEWILTZ 2006, vol. 2, 939; see also the
information of the Chinese source Sheng wu qing zheng lu: BRETSCHNEIDER 1910, 291.
111 ‫«بهار آن سال در کنار سمرقند بگذرانید و از آنجا بمرغزارهای نخشب آمد تابستان بآخر رسید و چهار پایان فربه و لشکر‬
»‫[( مرفّه شدند‬Chinggis Khan] spent the spring of that year in the vicinity of Samarqand and
from there arrived at the meadows of Nakhshab. The summer ended [there], the mounts
were fattened, and the troops rested), JUVAYNĪ/QAZVĪNĪ 1916/1334, 102; JUVAINI/BOYLE 1958,
Mongol ruler sent a message to his youngest son Tului that he should cease the
operations in Khurāsān and join him in the mountains of Afghanistan “before the
weather turns hot”.112 And, indeed, Chinggis Khan and the main Mongol forces
did spend the summer of this year on the hills113 around the already captured
Ṭāliqān, together with Tolui, as well as Cha’adai and Ögedei, who had just re-
turned from the successful campaign against Gurgānj.114 Rashīd al-Dīn reports that
during this pause the troops “rested and the mounts were fattened”.115 In the
spring of 1222, following the defeat of Jalāl al-Dīn Mingburnu’s army on the right
bank of the Indus (1221), Chinggis Khan commanded the operations upstream,
whereas Ögedei was sent to capture the regions downstream and in the interior.
After he conquered Ghazna, the prince asked his father to continue against Sis-
tān116, but “Chinggis Khan ordered: ‘The weather has turned hot, you come back so
that we will send other troops in order to besiege it’”.117 The great conqueror him-
128–9; see also Sheng wu qing zheng lu: BRETSCHNEIDER 1910, 291; THE SECRET HISTORY/DE
RACHEWILTZ 2006, vol. 2, 948; and the Secret History of the Mongols, where apparently there is
a contamination with the events of the next year, THE SECRET HISTORY/DE RACHEWILTZ 2006,
vol. 1, § 259, pp. 191–2; vol. 2, 947–50. Rashīd al-Dīn dates the stay in Nakhshab in the au-
tumn of the following year, 1221, but his account of this period is full of chronological inac-
curacies (for them, see n. 56 above), RASHĪD AL-DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol. 1,
517; RASHĪD AL-DĪN/KARĪMĪ 1338/1959, vol. 1, 374; RASHĪD AL-DĪN/THACKSTON 1998–9, part
2, 255; RASHĪD AL-DĪN/SMIRNOVA 1952, 217.
112 ».‫( «و چینگگیز خان از طالقان ایلچی فرستاده بود که فرزند تولوی خان پیش از آنکه هوا گرم شود باز گردد‬Chinggis
Khan had sent a messenger from Ṭāliqān [ordering] that his son Tului Khan should come
back before the weather turns hot.), RASHĪD AL-DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol. 1,
519; RASHĪD AL-DĪN/SMIRNOVA 1952, 219; see also THE SECRET HISTORY/DE RACHEWILTZ 2006,
vol. 1, § 259, pp. 191–2; Sheng wu qing zheng lu, BRETSCHNEIDER 1910, 292; THE SECRET HISTO-
RY/DE RACHEWILTZ 2006, vol. 2, 949.
113 Or the foothills, according to Smirnova’s interpretation, see her translation, referred to in
the next note.

114 RASHĪD AL-DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol. 1, 521, 525; RASHĪD AL-DĪN/KARĪMĪ
1338/1959, vol. 1, 375; RASHĪD AL-DĪN/THACKSTON 1998–9, part 2, 256; RASHĪD AL-
DĪN/SMIRNOVA 1952, 220, 223; YUAN SHI/KHRAPACHEVSKY 2009, 158; Sheng wu qing zheng lu,
THE SECRET HISTORY/DE RACHEWILTZ 2006, vol. 2, 949; see also de Rachewiltz’s commentary
on p. 950.
115 ».‫[( «و باتّفاق لشکرها بهم در پشته های طالقان تابستان کرده و آسوده گشته و چهارپایان فربه شده‬Chinggis Khan and
his sons] together, along with their troops, spent the summer in the hills of Ṭāliqān, rested,
and the mounts were fattened.), according to the fuller version from Rawshan and Mūsavī’s
edition and the translation by Smirnova, RASHĪD AL-DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol.
1, 525; RASHĪD AL-DĪN/SMIRNOVA 1952, 223.
116 I.e. Zarang, the capital of that province.
117 »‫«چینگگیز خان فرمود که هوا گرم شد تو باز گرد تا دیگر لشکرها را به جهت محاصره آن بفرستیم‬, RASHĪD AL-
DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol. 1, 529; RASHĪD AL-DĪN/KARĪMĪ 1338/1959, vol. 1,

377; RASHĪD AL-DĪN/THACKSTON 1998–9, part 2, 257; RASHĪD AL-DĪN/SMIRNOVA 1952, 225; see
self spent the summer in the steppes around Parvān.118 Since the Eurasian nomads
tried to avoid large operations during the hot period of the year, the troops of the
great conqueror broke their march during the summers even at the time of their
withdrawal from the destroyed Khwārazmian empire towards Mongolia.119
The comparative material about the Khwārazmian campaign is of particular
importance, as in the pre-Mongol period and in the age of the Mongol invasions
many of the eastern Cuman-Qïpchaq tribes were active on the territory of this
ephemeral empire, operating in the same or a very similar natural environment.
This is illustrated by the actions of one of the most prominent Khwārazmian amirs
and a leading member of the Cuman-Qïpchaq tribal elite—Amin Malik, the gover-
nor of Harāt. He planned to seek asylum from the troops of Chinggis Khan in Sis-
tān, but the arrival of the heat forced him to give up his intentions. 120
It must be emphasized, however, that the seasonal rhythm of Eurasian nomadic
warfare was not a result of particular conditions in Central Asia. Therefore, simi-
larities in the behaviour of the Mongols and the Cuman-Qïpchaqs can be observed
far beyond this region. This is revealed by the course of other campaigns of Ching-
gis Khan and his successors in various geographical areas. Thus, according to the
Chinese dynastic chronicle Yuanshi, in two consecutive years during his last expe-
dition against the Tanguts the founder of the Mongol Empire retreated in moun-
tainous regions for the summer in order to “hide from the heat”.121 The ruler of the
Iranian Mongols Ilkhan Ghazan (r. 1295–1304) aborted his campaign in Mamlūk
Syria in February 1300 due to the warming weather. In the firman issued by him
immediately before his withdrawal from Damascus, he states: “We will return to
also: Sheng wu qing zheng lu, BRETSCHNEIDER 1910, 293; THE SECRET HISTORY/DE RACHEWILTZ
2006, vol. 2, 950.
1338/1959, vol. 1, 377–8; RASHĪD AL-DĪN/THACKSTON 1998–9, part 2, p. 257; RASHĪD AL-
DĪN/SMIRNOVA 1952, 225; YUAN SHI/KHRAPACHEVSKY 2009, 159; see also THE SECRET HISTO-
RY/DE RACHEWILTZ 2006, vol. 1, § 257, 258, p. 191; Sheng wu qing zheng lu, BRETSCHNEIDER
1910, 293; THE SECRET HISTORY/DE RACHEWILTZ 2006, vol. 2, 944–5, 947, 950; see also the ac-
count of the visit of the Daoist master Changchun in the camp of Chinggis Khan,
BRETSCHNEIDER 1910, 86.
119 JUVAYNĪ/QAZVĪNĪ 1916/1334, 110, 111; JUVAINI/BOYLE 1958, 139, 140; THE SECRET HISTO-
RY/DE RACHEWILTZ 2006, vol. 1, § 264, p. 196; THE SECRET HISTORY/DE RACHEWILTZ 2006, vol.
2, 965.
120 »‫ روی بطرف غزنین نهاد‬،‫ چون هوا گرم شد‬،‫ در وقت فرار بطرف سیستان رفته‬،‫( «و ملک خان هرات‬Malik Khan of
Harāt, at the time of the flight [from the Mongols] had proceeded towards Sīstān, [but]
when the weather became hot, he turned towards Ghazna), JŪZJĀNĪ/HABĪBĪ 1343/1964, vol.
2, 116, see also 117; AL-JUZJANI/RAVERTY 1970, vol. 2, 1013, see also 1015–16.
121 YUAN SHI/KHRAPACHEVSKY 2009, 160, 161; see also THE SECRET HISTORY/DE RACHEWILTZ
2006, vol. 1, § 266, p. 198; vol. 2, 971;

[this] country in the autumn.”122 And, indeed, next winter Ghazan launched a new
campaign against Syria.123
Even in the moderate climate zone the intensity of warfare during the long
Mongol campaigns varied with the change of seasons. The Mongol invasion in the
Kingdom of Hungary (1241–2), led by Batu and his fellow Chinggisids, offers a
particularly useful parallel to the researchers of the Cuman behaviour in South-
eastern Europe, as it developed in a region with similar climatic conditions to those
in the Balkans. According to Rashīd al-Dīn, in this case, too, the Mongol troops
chose a suitable area where they spent the summer—the riverbanks of the Tisza
and the Danube.124 Thomas of Spalato also reports that after they conquered the
Hungarian territories east of the Danube, the invaders “settled in those places with
the intention to spend the entire summer and winter there.”125 The situation was
similar with regards to the Mongol military activities during their first attack
against the Rus’ principalities in 1237–8, at an earlier phase of the so-called Great
Western Campaign of 1236–42 (the same long expedition which led them in Hun-
gary). After they ravaged Northeastern Rus’ in the winter of 1237–8, the invaders
managed to capture the town of Kozelsk, located in the principality of Chernigov.
The Mongols had to besiege the town for two months before it finally fell—
perhaps by the end of May 1238. The Moscow-Academic Chronicle reports that “Batu
took Kozelsk and went into the Land of the Cumans.” 126 Undoubtedly, the Mongol
troops withdrew to Dasht-i Qipchāq with the intention to spend there the ap-
proaching summer,127 following a pattern of many other pauses in the summer
season.
122 AMITAI 2002, 259–60.

123 BOYLE 1968, 389.
124 ،‫ گریزانیده‬،‫ کلر‬،‫ تمامت والیات باشغرد و ماجار و ساسان گرفته و پادشاه ایشان را‬،‫«و شهزادگان بدین پنج راه مذکور رفته‬
»‫( تابستان در رودخانۀ تیسا و تُنا کردند‬The princes, proceeding along the aforementioned five routes,
seized all the regions of the Bashkirs, Magyars, and Saxons, and putting their ruler, k-l-r
[kral (Slav.), király (Hung.), i.e. king] to flight, they spent the summer on the rivers Tisza
and Danube), RASHĪD AL-DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol. 1, 678; RASHĪD AL-
DĪN/KARĪMĪ 1338/1959, vol. 1, 483; RASHĪD AL-DĪN/BOYLE 1971, 70; RASHĪD AL-
DĪN/THACKSTON 1998–9, part 2, 332; RASHĪD AL-DĪN/VERKHOVSKY 1960, 45.
125 “Interea Tartarorum exercitus depopulata omni regione Transilvana, cesis ac fugatis
Hungaris ex Transdanubialibus horis, composuerunt se in locis illis totam ibi estatem et

yemem peracturi.” (Meanwhile, the Tatar army, having ravaged the entire region of Tran-
sylvania, and having slayed and chased away the Hungarians from the areas of Transdanu-
bia, settled in those places with the intention to spend the entire summer and winter there.),
THOMAS ARCHIDIACONUS/PERIĆ ET AL. 2006, 278–9.
126 “Батыиж взем Козелескъ и поиде в земълю Половетцкоую“, PSRL, vol. 1, col. 522. For
the operations of the Chinggisids in the Rus’ principalities in 1237–8, see KHRAPACHEVSKY
2005, 354–73; for the siege of Kozelsk, see also KARGALOV 1967, 110–11.
127 In that particular case this does not exclude campaigning against the Cuman-Qïpchaqs
during that period.

Naturally, the summer was not the only season that the Mongols took into con-
sideration while planning their campaigns. Thus, they often sought to spend the
winter in regions with favourable conditions. And yet, unlike the summer pauses,
arriving at the winter quarters did not necessarily diminish the intensity of Mongol
warfare, which at certain moments quite surprised their enemies.128 Furthermore,
sometimes these nomads intentionally planned to launch their campaigns in the
cold time of the year in order to take advantage of the low temperatures. 129 As a
rule, neither the Cumans nor the Mongols demonstrated such preferences regard-
ing the summer season. On the contrary, the instances referred to above illustrate
these nomads’ unwillingness to fight during the hot period of the year.
Reasons for the asymmetric seasonal military activity
What were, after all, the reasons for this unwillingness to campaign during the
summer? It is noteworthy that this was not a matter of some personal discomfort of
the steppe warriors. Nomadic armies, and those of the Cumans and Mongols in
particular, were accompanied by a large number of horses, since every warrior
brought at least several spare mounts which he rode in turns. Robert de Clari ex-
plicitly states regarding the Cumans: “Each one of them has ten or twelve horses
and they are so well trained that they follow them everywhere where they want to
lead them.”130 Mongol warriors, too, brought with them spare mounts whose
number varied according to the circumstances. 131 Thus, even relatively small no-
128 JUVAYNĪ/QAZVĪNĪ 1916/1334, 116; JUVAINI/BOYLE 1958, 147; IBN AL-ATHĪR/RICHARDS 2008,
214–15.
129 The Hungarian Dominican Julian reports: “Hoc tamen expectantes quod sicut et ipsi
Ruteni, Ungari et Bulgari qui ante eos fugerant, viva voce nobis referebant: quod, terra,
fluviis et palidibus in proxima hieme congelatis totam Rusciam toti multitudini sic facile est
eis depredari, sicut totam terram Rutenorum.” (Yet, [the Mongols], expect—as those Rus’,
[Eastern] Hungarians, and [Volga] Bulgars, who had run before them, were telling us with
their own mouth—that after the land, rivers, and swamps freeze in the coming winter, it
will be easy for the entire [Mongol] multitude to devastate all Rus’ as well as the entire land
of the Rus’ people.), DÖRRIE 1956, 174; ANNINSKY 1940, 86; see also Khrapachevsky, who
brings attention to this evidence, KHRAPACHEVSKY 2005, 355.
130 ROBERT DE CLARI/MARKOV 2007, §LXV, pp. 104–5; the English translation follows the
Bulgarian one.
131 “Equos ita bene habent edomitos, ut quotcumque unus habeat homo, omnes ipsum
tamquam canes secuntur.” (They have so well-trained horses that however many a man has
they all follow him like dogs.), THOMAS ARCHIDIACONUS/PERIĆ ET AL. 2006, 284–5. Similar
evidence, perhaps containing inflated numbers, could be found in a letter from an anony-
mous Hungarian prelate to the Bishop of Paris ca. 1239–40: “et equi multi sequuntur ipsos
absque ductore, ita quod quando dominus equitat, xx. vel xxx. equi sequuntur ipsum” (and
many horses follow them without a leader, so that when [their] master rides twenty or thirty
horses follow him), ANNALES MONASTICI/LUARD 1865, 324. Regarding the horse training
Meng-da beilu reports: “After the horses are dismounted, it is not necessary to hobble or
madic detachments of about 1,000 men would bring with them a large number of
horses, 6,000–10,000, sometimes maybe even more. And when there were larger
Cuman armies, such as the one at Lardeas (more than 7,000 horsemen according to
Choniates); the nomadic allies of Kaloyan in the battle of Adrianople in April 1205
(some 14,000 according to Villehardouin); or the aforementioned 10,800 Cumans
from the fragmentarily preserved Preslav inscription, the number of the spare
horses must have been considerable, even if one assumes certain overstatements of
the numbers reported by the medieval authors.132
The fact that the Cumans and their Mongol contemporaries had at their dispos-
al such a large number of spare mounts gave them a number of advantages. The
Mongol practices in this regard are far better documented. Since both nomadic
communities had a similar lifestyle and military traditions, the medieval sources
for the Mongols provide rich contextual material for better understanding the Cu-
tether them, as they still do not run away, their nature being so fine and good.”, FIVE CHI-
NESE SOURCES 2021, 83; MENG-DA BEILU/MUNKUEV 1975, 69. Regarding the number of horses,
Marco Polo reports that when the Mongols “are on military service, there is not one of them
who does not lead with him six, eight, or more horses for his own use”, MARCO PO-
LO/LATHAM 1958, 152. In one version of the Venetian’s account it is stated that there was an
average of eighteen beasts to every man, but this evidence apparently resulted from an
exaggeration or mistake, MARCO POLO 1903, 264, n. 3; MARCO POLO/LATHAM 1958, 100;
Smith brings attention towards these passages, SMITH 1984, 314, 334). Marco Polo also re-
ports that the Mongol horses are very well trained, MARCO POLO 1903, 260, 262; MARCO PO-
LO/LATHAM 1958, 99, 101. In Meng-da beilu it is reported that every time the Mongols set off
on campaign each of them has several horses, MENG-DA BEILU/MUNKUEV 1975, 69; FIVE CHI-
NESE SOURCES 2021, 83. In the 1230s, the Southern Song envoy Peng Daya reported that each
warrior has “either two-three mounts or six-seven mounts”, HEI-DA SHILUE/KHRAPACHEVSKY
2009, 60–1; FIVE CHINESE SOURCES 2021, 117, n. 113; see also the evidence of his colleague Xu
Ting: “When the troops set out, the leader rides one horse with five or six to three or four
other horses following along. They regularly prepare thus for an emergency, and even with-
out [so many backups] a man must have at least one or two extra mounts.”, FIVE CHINESE
SOURCES 2021, 117; see also HEI-DA SHILUE/KHRAPACHEVSKY 2009, 61. The following order
was issued for Ghazan’s Syrian campaign in 1299–1300: ‫«و یاسا شد که چریک از هر ده نفر پنج بر نشینند‬
»‫( و هر نفری از لشکریان پنج اسب با ساز و عدت تمام و آزوق شش ماه مرتب کردانند‬It was ordered that from
every ten men [subject to military duty] five should mount [their horses and present them-
selves in] the army and each person from the warriors should prepare five horses, fully
ready and harnessed, and supplies for six months.), VAṢṢĀF 1269/1852–53, 373;
VAṢṢĀF/ĀYATĪ 1346/1967, 223; John Mison Smith and Rouven Amitai bring attention to-
wards this evidence, SMITH 1984, 329; AMITAI 2002, 229–30, see also 253–4.
132 CHONIATES/VAN DIETEN 1972, 8-12; CHONIATES/VOYNOV 1983, 95–9; MEMOIRS OF THE CRU-
SADES/MARZIALS 1908, 92; for the inscription of Veliki Preslav, see n. 4. Regarding the num-
bers of the Cuman allies of the Asenids, see also PAVLOV 1990b, 14; DALL’АGLIO 2008–9, 34–
5, where other figures are also given.
man behaviour.133 The large number of horses in these nomadic armies did not
hamper their fast marches during campaigns, since usually the animals were not
shod and were not fed with fodder,134 which means that the classical nomadic mili-
tary columns travelled without a baggage train. Furthermore, the possibilities of
changing horses often during the march 135 and leaving the animals enough time to
rest before they have to carry their rider again 136 actually allowed the nomads to
133 Indeed, the lifestyle and customs of Cumans and Mongols were so akin that their seden-
tary contemporaries could not clearly distinguish them when the Mongols first appeared in
Central Europe, ROGERIUS/JUHÁSZ 1999, 566; MASTER ROGER/BAK – RADY 2010, 172–3, n. 1;
CONTINUATIO SANCRUCENSIS SECUNDA/WATTENBACH 1851, 640–1. The proximity between the
two communities could be traced also in the material culture, especially that of the Eastern
Qïpchaq tribes. According to the Russian archaeologist Yuly Khudyakov, the Eastern Qïp-
chaqs’ arrow set is most comparable with the Mongol one, KHUDYAKOV 1997, 112.
134 Contrary to de Clari’s statement regarding the Cumans, ROBERT DE CLARI/MARKOV 2007,
§LXV, pp. 104–5. Carpini explicitly states that “cum Tartari nec stramina, nec fenum nec
pabulum habeant” (since the Tatars have neither straw nor hay nor fodder), DI CAR-
PINE/DAFFINÀ ET AL. 1989, 305; THE MONGOL MISSION/DAWSON 1966, 52. Marco Polo, too,
reports that the Mongol horses subsist entirely on grazing and there is no need to transport
fodder, MARCO POLO 1903, 260; MARCO POLO/LATHAM 1958, 99. In Meng-da beilu (1221) it is
stated that the Mongols graze their horses and never give them fodder, FIVE CHINESE
SOURCES 2021, 83; MENG-DA BEILU/MUNKUEV 1975, 69. Peng Daya also reports that the Mon-
gol “horses are pastured in the wild, without any hay or grains”, FIVE CHINESE SOURCES 2021,
115; HEI-DA SHILUE/KHRAPACHEVSKY 2009, 58; see also the observations of Ibn Baṭṭūṭa for the
Golden Horde in the 14th century, IBN BAṬṬŪṬA/GIBB 1995, 473; IBN BAṬṬŪṬA/TIZENGAUZEN
1884, 282. Thomas of Spalato also gives a valuable description of the Mongol horses: “Equos
breves sed fortes, patientes inedie ac laboris more equitant rusticano, per rupes vero et
lapides absque ferramentis ita discurrunt, ac si capre forent silvestres, tribus enim continuis
diebus labore quassati parvo stipularum pabulo sunt contenti.” (They ride in the rustic
manner short but strong horses, able to endure hunger and toil. [The horses] run over rocks
and stones without iron shoes in such a way as if they were wild goats. Indeed, exhausted
by three days of continuous riding they are satisfied with a small meal of chaff.), THOMAS
ARCHIDIACONUS/PERIĆ ET AL. 2006, 284–5. In the Georgian Chronicle of the Hundred Years it is
reported with astonishment that the Mongols of Jebe and Sübe’etei Ba’atur rode mounts
without horseshoes, KARTLIS TSKHOVREBA/MATREVELI ET AL. 2013, 337; KARTLIS TSKHOV-
REBA/MET’REVELI – JONES 2014, 321–2. This practice, however, seems to not have been ubiq-
uitous, since, according to Peng Daya, the horse hooves were vulnerable to the stones, and
that is why the Mongols used some kind of iron or wooden horseshoes, FIVE CHINESE
SOURCES 2021, 115; HEI-DA SHILUE/KHRAPACHEVSKY 2009, 58.
135 According to de Clari, the Cumans also rode “first one and then another” of their horses
ROBERT DE CLARI/MARKOV 2007, §LXV, с. 104–5, the English translation follows the Bulgari-
an one.
136 “et propter hoc debent vitare nimium cursum post eos, ne forte fatigentur equi eorum;
quoniam nostri multitudinem equorum non habent. Sed Tartari illum equum quem equitant
uno die, illum non ascendunt in tribus vel in quator diebus post hoc; unde non curant si
fatigentur equi, propter multitudinem equorum quam habent.” (and due to this [our men]
cover far longer distances in a shorter period of time as compared to most medie-
val armies.137 This impressive mobility could be achieved mainly due to the ability
of nomadic horses to sustain themselves entirely by grazing. The grass diet and the
harsh natural conditions in the steppes turned the nomadic mounts into relatively
small but very hardy ponies, which were inferior in terms of physical strength (yet
not in terms of endurance) to the fodder-fed horses of the sedentary armies. 138
Thus, the large number of spare mounts allowed the nomadic armies to move fast-
er. Furthermore, the possibility of changing horses during battle provided the
steppe warriors with a significant advantage on the battlefield.139 It was precisely
this practice that allowed the Cumans (and the other Eurasian stock breeders) to
wage their traditional mobile steppe warfare, “fighting in their ancestral and cus-
tomary manner”,140 so eloquently described by Choniates:
should avoid too long a pursuit after them, lest their horses be exhausted; since our men do
not have a large number of horses. Yet, the Tatars do not mount the horse they have ridden
for one day in the next three of four days; hence they do not care if the horses are exhaust-
ed—due to the large number of horses they possess.), DI CARPINE/DAFFINÀ ET AL. 1989, 298–
9; THE MONGOL MISSION/DAWSON 1966, 47. In Meng-da beilu it is reported that every Mongol
set on campaign with several horses, changing them every day, and that is why the animals
were not exhausted, MENG-DA BEILU/MUNKUEV 1975, 69; FIVE CHINESE SOURCES 2021, 83.
137 According to de Clari’s report, which was undoubtedly stereotyped to a certain extent,
the Cumans travelled “incessantly during the night and during the day. They move so fast
that in one night and one day they cover the distance for six days, or seven, or eight.” ROBERT
DE CLARI/MARKOV 2007, §LXV, 104–5; the English translation follows the Bulgarian one.
138 See, for instance, John Mason Smith’s comments about the horses of the Egyptian Mam-
lūks and the Iranian Mongols, SMITH 1984, 331–2.

139 As can be seen in the description of the Cuman victory in the battle of the Kayala River,
given by the Laurentian Chronicle: „изнемогли бо сѧ бѧху безводьємь . и кони и сами в

знои . и в тузѣ . и поступиша мало к водѣ . по . г҃ . дни бо не пустили бѧху ихъ к водѣ .
Видѣвше ратнии үстремишасѧ на нь . и притиснуша и к водѣ . и бишасѧ с ними
крѣпко . и бъıс̑ сѣча зла велми . друзии конѣ пустиша к ним̑ съсѣдше . и кони бо бѧху
под̑ ними изнемогли . и побѣжени бъıша наши гнѣвом̑ Бж҃ьимъ .“ (Since due to the lack
of water both the horses and [the Rus’] themselves were exhausted by heat and suffer, and
they advanced a little bit towards the water, because they were prevented from reaching the
water for three days. When the enemies [i.e. the Cumans] saw that, they rushed upon them
and pushed them to the water, and fought with them fiercely, and there was a great slaugh-
ter. The others [i.e. the Cumans] galloped [their] horses towards them [the Rus’] after they
changed them [with fresh ones], while the horses under [the Rus’] were exhausted and our
men were defeated by the wrath of God.), PSRL, vol. 1, col. 398 see also the modern Russian
translation, PLDR 1980, 369. Smith believes that the Mongols also changed their horses in
the course of battle, without referring to a specific source, SMITH 1984, 319.
140 “τὸν πάτριον αὐτοῖς καὶ ἠθάδα τρόπον μαχόμενοι”, CHONIATES/VAN DIETEN 1975, 397;
CHONIATES/MAGOULIAS 1984, 218.

…[attacking, the Cumans] were discharging javelins and entering into fight
with spears, and in short time again turning their onslaught into flight and
provoking the opposite side to pursue them closely behind as [if they were]
running away. Again, turning [their] backs like birds cleaving the air, they
appeared before the eyes, [and] facing the attacking [enemies] they began to
fight and were engaging into hand-to-hand combat with much greater brav-
ery. And doing this many times, so that they were already prevailing over
the Romans, they were no longer thinking about turning around, but draw-
ing the swords and shouting some terrifying battle cry, they fell upon the
Romans faster than a short thought; overtaking both the fighting one and the
coward alike, they mowed [them] down. 141
Therefore, the large masses of spare mounts were a key military resource that
gave the nomadic armies, and those of the Cumans in particular, their main strate-
gical (speed and manoeuvrability of the marching columns) and tactical (fresh
horses in the climax of the fighting) advantages. But these advantages came at a
cost—the presence of these resources imposed certain limitations on the nomadic
armies, which influenced their behaviour on the theatre of war. 142
141 „ἐπιόντες ἠφίεσαν βέλεμνα καὶ δόρασι προσεπλέκοντο, μετὰ βραχὺ δ' αὖ μεταβαλόντες
τὴν ὁρμὴν εἰς φυγὴν καὶ καταδιώκειν ὀπίσω αὐτῶν ὡς φυγάδων τὸ ἀνθιστάμενον ἐρεθίζοντες,
αὖθις ὡς οὐδὲ πτηνὰ τὸν ἀέρα τέμνοντα τὰ νῶτα εἰς τὴν τῶν ὄψεων μεταθέμενοι χώραν
ἐνώπιοι τοῖς ἐπιοῦσιν ἐμάχοντο καὶ πολλῷ γενναιότερον συνεπλέκοντο. καὶ τοῦτο πολλάκις
πεποιηκότες, ὡς ἤδη τῶν Ῥωμαίων ἦσαν ἐπικρατέστεροι, οὐκέτ’ ἐμέμνηντο τῶν ἑλίξεων, ἀλλὰ
τὰ ξίφη γυμνώσαντες καί τινα βοὴν καταπληκτικωτάτην ἐκρήξαντες μικροῦ καὶ ἐννοήματος
τάχιον Ῥωμαίοις ἐνέπεσον καὶ τὸν μαχόμενον ὁμοίως καὶ τὸν δειλαινόμενον καταλαμβάνοντες
ἐξεθέριζον.“, CHONIATES/VAN DIETEN 1975, 397; CHONIATES/MAGOULIAS 1984, 218. Describ-
ing the same events in his orations Choniates notes: “those [Cumans] that remained in the
front[lines] started hand to hand combat with our [men], retreating and at the same time
attacking, as indeed [is] their custom to fight” (ἐν τοῖς ἔμπροσθεν οἱ λοιποὶ τοῖς ἡμετέροις
συμπλέκεσθαί ἤρξαντο ὑποφεύγοντες ἅμα καὶ ἐπιτιθέμενοι, ὥσπερ δὴ τρόπος αὐτοῖς ἐν τῷ
μάχεσθαι), CHONIATES/VAN DIETEN 1972, 11; CHONIATES/VOYNOV 1983, 98.
142 Thus, s.a. 1190 The Kievan Chronicle reports that the Cumans rested their horses in the
course of a campaign: “Половци […] ѣхавше и легоша по Висемь и тоу перепочивше

конемь своим̑ и ѣхаша ко Боровомоу” (The Cumans […] rode [away] and lied down [in
camps along the rivers Velyka and Mala] Vis’ and here they gave rest to their horses, and
rode to [the town of] Borovy), PSRL, vol. 2, col. 669. The words of the Armenian chronicler
Hayton of Corycus are perhaps the most eloquent description of the care that the Eurasian
nomads took of their mounts: “Majorem enim curam habebant de equis quam de seipsis,
quia, dum parvo et vili cibo sciant esse contenti, de seipsis quodammodo non curabant”
(Indeed, they cared more for the horses than for themselves, because as long as they knew
that they are satisfied with a small and modest ration, they were not taking particular care
for themselves), HAYTONUS/KOHLER 1906, 321; Denis Sinor brings attention towards this
passage, SINOR 1972, 178.
Perhaps the most important of these limitations derives from the aforemen-
tioned fact that nomadic horses depended mostly on the available pasture. It is not
by chance that according to Meng-da beilu during the night breaks: “[The Mongols]
graze them [their horses] in the steppe according the greening and withering of the
grass.”143 Even more eloquent is Maurice’s Strategikon, which was written centuries
earlier (late 6th/early 7th century): “The shortage of pasture hinders them [i.e. the
nomads] due to the multitude of horses, which they bring with them…” 144 It was
this dependency that stood behind the nomadic unwillingness to wage war during
the summer heats, and it was caused by a rather trivial circumstance—during the
summer, the pasture grasses dry up under the scorching sun and their nutritional
value for the animals considerably diminishes. In his letter to the French king Saint
Lewis IX (r. 1226–70), the founder of the Ilkhanate, Hülegü (r. 1259–65), pointed to
the lack of pasture among the reasons that caused the withdrawal of his main forc-
es from Syria in 1260.145 This is why during the hot months the steppe armies were
bound to particular areas where the natural conditions support the necessary
amount of grazing, just like the troops of Chinggis Khan during the summers of his
Khwārazmian Campaign. Thus, in summertime, especially when operating outside
143 MENG-DA BEILU/MUNKUEV 1975, 69; FIVE CHINESE SOURCES 2021, 83.
144 „Ἐναντιοῦται δὲ αὐτοῖς ἔνδεια βοσκῆς διὰ τὸ πλῆθος ὧν ἐπιφέρονται ἀλόγων…“, DAS
STRATEGIKON/DENNIS 1981, 364; MAURICE'S STRATEGIKON/DENNIS 1984, 117. Slightly above
the text states: “And a multitude of horses follows them, both male and female, [serving] at
the same time, for nourishment and to make them look numerous […] until the day of the
battle, spread about according to clans and tribes, they continuously graze their horses in
summer and winter.”, „Ἀκολουθεῖ δὲ αὐτοῖς καὶ πλῆθος ἀλόγων, ἀρρένων τε καὶ θηλειῶν,
ἅμα μὲν πρὸς ἀποτροφήν, ἅμα δὲ καὶ διὰ πλήθους θεωρίαν […] μέχρι μὲν τῆς τοῦ πολέμου
ἡμέρας διεσπασμένως κατὰ γένη καὶ φυλάς, τοὺς ἵππους βόσκοντες διηνεκῶς ἔν τε θέρει καὶ
χειμῶνι.“, DAS STRATEGIKON/DENNIS 1981, 362; MAURICE'S STRATEGIKON/DENNIS 1984, 116–117.
145 “Quia tamen nostri moris est caumatibus estiuis locis frigidis montanis niueis libencius
adherere, supra dictis Halapensibus et Damaszenis deuastatis nec non uictualibus et herbar-
iis pro maiore parte consumptis, placuit nobis uersus montana maioris Armenie paulisper
redire, in dictis quidem locis paucos de nostris pro quorumdam residuorum castrorum
Hassassenorum destructione dimittendo” (Yet, since in our custom it is more pleasant dur-
ing the summer heats to settle in cool places in the snowy mountains—after we have rav-
aged the aforementioned inhabitants of Aleppo and Damascus, and also after we consumed
for the greater part the provisions and the pastures—it pleased us briefly to return to the
mountains of Greater Armenia, while leaving in the mentioned places a few of our men for
the destruction of certain remaining fortresses of the Assassins), MEYVAERT 1980, 258; LET-
TERS FROM THE EAST/BARVER – BATE 2010, 159; David Morgan draws attention to this evi-
dence: MORGAN 1985, 231–5. Regarding this custom, see also the account of Marco Polo,
MARCO POLO 1903, 251–2; MARCO POLO/LATHAM 1958, 97. Naturally, the seasonal cycle and
the summer heats were hardly the only reason for the Mongol withdrawal, but there is no
doubt that Hülegü chose to justify his actions with a real practice, notwithstanding the polit-
ical rhetoric in the letter.
of the steppes, the nomads lost their main advantage over the sedentary adver-
saries—their mobility. Furthermore, they became easy to detect and vulnerable. 146
In comparison, the winter did not cause such problems, even during intensive
snowfalls, since as specifically noted by Plano Carpini, the hardy Mongol horses
knew how to dig up the grass from under the snow in the steppes, 147 and the Cu-
man mounts undoubtedly had the same skills. Naturally, the limited amount of
available pasture, especially in regions whose geographical conditions in principle
do not suggest the presence of abundant grazing, may influence the course of no-
madic campaigns in other seasons, too. 148 But for no other time of the year have the
sources documented such a clear correlation as that between the arrival of the
summer and the suspension of military activity by Mongols and Cumans in vari-
ous regions located throughout the entire sedentary periphery of the Eurasian
steppe belt. Thus, there can hardly be any doubt that the main factor behind the
asymmetric seasonal military activity of the Cuman-Qïpchaqs in Khwārazm and the
146 In fact, due to the large number of horses, even in these cases nomadic armies could not
stay at the same place for a long time since the grass would be consumed. This is indicated
by Marco Polo’s observations regarding the Mongols (MARCO POLO/LATHAM 1958, 97), and
as it seems such was also the case with the Cuman refugees of Kuten in Hungary in 1239,
ROGERIUS/JUHÁSZ 1999, 554; MASTER ROGER/BAK – RADY 2010, 138–41. It should be taken
into consideration that both cases are related to nomads who migrated with all of their
flocks. Yet, there could hardly be any doubt that while on a campaign, the steppe warriors
faced similar problems, although perhaps to a lesser degree, if one keeps in mind that they
were accompanied only by the horses necessary for the military operations (see, for instance,
SMITH 1984, 338–9). The need for a constant search for fresh grass made the manoeuvres of
the steppe armies through regions with limited pastures during the summer rather predict-
able and hence they became vulnerable.
147 “Qui responderunt nobis quod si diceremus in Tartariam equos illos quos habebamus,
cum nives essent magne et nescirent fodere erbam sub nive, sicut equi Tartarorum, nec
inveniri posset aliquid aliud ad manducandum pro ipsis, cum Tartari nec stramina, nec
fenum nec pabulum habeant, morerentur omnes.” (Who [the notables of Kiev] answered us
that if we were to bring in Tartaria those horses which we had, they would all die, for the
snows were great and they would not know how to dig up the grass from under the snow
like the Tatar horses, neither could be found anything else to feed them, since the Tatars
have neither straw nor hay nor fodder.), DI CARPINE/DAFFINÀ ET AL. 1989, 304–5; THE MON-
GOL MISSION/DAWSON 1966, 52.
148 Such was the case with the arrival of the Mongol troops in Dalmatia in March 1242, de-
scribed by Thomas of Spalato: “Ecce autem paucis diebus elapsis venit Caydanus cum ali-
quota parte sui exercitus, quia non erant herbe pro toto equitatu sufficientes, erat enim prin-
cipium Martii asperis frigoribus inhorrescens.” (And now, moreover, after few days had
passed, Kadan arrived with a small part of his army, since there was not enough grass for
the entire cavalry. Indeed, it was the beginning of March, bristled up with sharp cold.),
THOMAS ARCHIDIACONUS/PERIĆ ET AL. 2006, 298–9. Denis Sinor reasonably assumes that in
this particular case the shortage of grass was not only due to the harsh climatic conditions in
March, “but also to the general bareness of the Karst mountains” in the region, SINOR 1972, 178.
Balkans was the insufficient amount of grazing for their horses in the hot part of
the year, which in the former case coincided with the growing distance between
the Central Asian operational theatre and the nomadic summer encampments
within the steppes.
The adversaries of the nomadic peoples quickly realized the strategical im-
portance of grazing for the movement of their armies and sought to hamper it by
destroying this resource. The Egyptian Mamlūks developed the practice of burning
the fields and pastures along the frontier regions in the course of a war with the
Mongol Ilkhanate, which lasted for six decades in the second half of the 13 th and
the early 14th century.149 Sometimes the Mamlūks applied this strategy in an even
more radical form, by setting on fire their own territories so that the horses of the
invaders could not find grazing.150 Far away to the east, on the other edge of the
Eurasian steppe belt, the officials of the Ming dynasty (1368–1644) also set on fire
the pastures along the imperial frontiers during autumn and winter in order to
deter the Mongols from approaching Chinese territories. 151 In Europe, Plano
Carpini recommended the hay and straw to be burned or hidden before the ad-
149 AMITAI-PREISS 1995, 205–6; AMITAI-PREISS 1999, 134; 138–9. The fields were burned in
order to prevent the crops being used as fodder for the Mongol horses. Amitai notes that
both known cases of setting fire in the frontier regions were in the autumn—i.e. the usual
season for the Mongol advance in Syria.
150 Hayton of Corycus reports that when the troops of Ghazan reached the Euphrates in 1303
with the intention of invading Syria: “Sarraceni quidem timuerunt de adventu Cassani, et
dum in bello sibi diffiderent esse pares, totam terram eorum combusserunt ante faciem
Tatarorum, ita videlicet quod, collectis frugibus et aliis in terra crescentibus ac animalibus, et
omnibus in castris et munitionibus congregatis, residuum totum ignibus devastarunt, ut,
cum Tatari venirent, victualia non invenirent neque pabula pro jumentis. Quando Cassanus
intellexit ea, que fecerant Agareni, et qualiter igne dissipaverant totam terram, videns quod
equi per terras illas taliter devastatas non possent modo aliquo sustentari, accepit consilium
remanendi super flumine Eufrates illa hyeme, et, veris tempore veniente, dum herbe inci-
perent pululare, perficere iter suum.” (The Saracens, indeed, were afraid of the advent of
Ghazan, and since they did not consider themselves equal to him in war, they burned their
entire land before the face of the Tatars—in such a way, indeed, that gathering the crops and
[everything] else which grows from the land, as well as the animals, and concentrating them
in fortresses and fortifications, they wasted everything else with fires, so that when the Ta-
tars arrive they would find no provisions [for themselves] nor pasturages for the beasts.
When Ghazan realized what the Hagarenes had done and how they had utterly destroyed
with fire the entire country, seeing that the horses could not be sustained in any other way
through these so devastated lands, he decided to stay on the Euphrates River this winter,
and after the arrival of springtime, as soon as the grass begins to grow, to accomplish his
march.), HAYTONUS/KOHLER 1906, 321; NIKOLOV 2017, 78. Sinor brings attention towards this
evidence, SINOR 1972, 177–8. It seems that in this case, too, the Mongol offensive started in
the autumn.
151 JAGCHID – SYMONS 1989, 205, n. 93.
vancing Mongols.152 Thus, the sedentarists practically tried to simulate to a certain

extent the limitations which the summer heats imposed upon the otherwise ma-
noeuvrable nomadic armies.
The scarcity of pasture in the sedentary territories during the summer coincides
with another circumstance that perhaps also contributed to the sudden Cuman
withdrawals towards the steppes in this season. Quite often in the sources for the
summer pauses of the nomadic armies (the Mongol in particular) it is reported that
during this season the steppe dwellers strove to fatten up their horses which had
grown lean during the harsh winter months (or during the intensive campaigns in
the cold seasons), and due to this reason, the beginning of the military operations
for them traditionally fell within early autumn. It should be pointed out that the
most suitable conditions for such summer grazing are to be found precisely in the
Eurasian steppes.153 Furthermore, in this case it was not a simple fattening of the
animals, but rather a special diet aimed at bringing the horses in optimal form for
the forthcoming campaigns.154 Thus, the summer was a key period both in the liv-
ing cycle of the animals and in the military calendar of their nomadic masters. 155
152 “Eis fugientibus de terra, debent fenum et stramina comburere vel fortiter occultare, ut
equi Tartarorum minus inveniant ad comedendum.” (Those [adversaries of the Mongols]
who flee from their land must burn the hay and straw, or to hide them well, so that the Tatar
horses would find less to eat.), DI CARPINE/DAFFINÀ ET AL. 1989, 300; THE MONGOL MIS-
SION/DAWSON 1966, 48. It should be pointed out that the nomads did not practice haying,
but used such reserves when they were found gathered in the lands of their sedentary ad-
versaries.
153 Peng Daya notes about Mongolia: “Their country’s ‘crop’ is wild grass, which begins to
green in moon IV, becomes lush starting in moon VI, and in moon VIII withers again. Apart
from grass, there is nothing else at all.” FIVE CHINESE SOURCES 2021, 96; HEI-DA SHILUE/
KHRAPACHEVSKY 2009, 32. Since the Chinese months roughly correspond to the Roman ones,
according to Peng Daya the steppe grass was most nutritional precisely during the summer.
For the quality of the steppe herbage in general see the observations of Ibn Baṭṭūṭa, IBN
BAṬṬŪṬA/GIBB 1995, 473; IBN BAṬṬŪṬA/TIZENGAUZEN 1884, 282.
154 Peng Daya reports: “Their horses are pastured in the wild, without any hay or grains. In
moon VI [roughly July], they eat their fill of green grass and begin to fatten.” Xu Ting gives
more detailed information: “From early spring, after they call off the campaigns, all good
horses that have been on campaign are released to enjoy water and pasture without being
ridden or driven, until the west wind is about to come. The horses then are caught, bridled,
and tied up near the tents. They are given only a little water and grass until a month has
gone by and their fat has dropped. When ridden fully several hundred lǐ, they naturally do
not sweat, and thus they can endure the long distances of going on campaign.”, FIVE CHI-
NESE SOURCES 2021, 115–16; HEI-DA SHILUE/KHRAPACHEVSKY 2009, 58–9.
155 This fact should not be viewed as over-deterministic. The campaign of Chinggis Khan
against the Naimans in the summer of 1204, conducted against the opinion of most of his
notables, demonstrates that under pressing circumstances the nomadic armies could carry
out large-scale operations with insufficiently fattened horses, RASHĪD AL-DĪN/RAWSHAN –
MŪSAVĪ 1373/1994, vol. 1, 415–16; RASHĪD AL-DĪN/KARĪMĪ 1338/1959, vol. 1, 303–4; RASHĪD
Having in mind the conservative nature of the traditional steppe military culture
and the fact that the seasonal living cycle of the animals in the various parts of the
Eurasian steppe belt did not differ significantly, it could be assumed that this con-
clusion is absolutely valid with respect to the Cuman-Qïpchaq tribes. Therefore, it
seems very likely that the Cuman intention to prepare in the most suitable time
their animals for the forthcoming cold months of prolonged military activity
played a role in the withdrawal of their armies to Dasht-i Qïpchaq during the
summer along with the withering of the grass in the sedentary regions under the
scorching sun.
But why did the heat not prevent the Cuman activity in the Rus’ principalities,
where their presence was attested multiple times during the summer? It is hard to
answer this question, but everything leads to the conclusions that it was a matter of
geography. Unlike the Balkans and Khwārazm, the Rus’ principalities were located
to the north of Dasht-i Qipchāq, i.e. to the north of the Cumans’ own summer pas-
ture grounds where the grass remained nutritious during the hot season.156 Appar-
ently, the sparsely populated Rus’ territories, watered by dense river systems
which had no analogue neither in the Balkans nor in Khwārazm, could provide
pastures with sufficient nutrition even in the hot time of the year. This would ex-
plain the numerous incursions of the Cumans in the Rus’ lands during the sum-
mer, especially in the areas near the steppe, which were most often a target of no-
madic attacks.157 And yet, it is noteworthy that in 1238 the Mongols withdrew from
AL-DĪN/THACKSTON 1998–9, part 1, 201–2; RASHĪD AL-DĪN/SMIRNOVA 1952, 146–7; THE SECRET
HISTORY/DE RACHEWILTZ 2006, vol. I, § 190, 193, 194, pp. 112–113, 115–117; YUAN
SHI/KHRAPACHEVSKY 2009, 144–5. It should be taken into consideration, however, that in this
case the fighting took place within the Eurasian steppe belt, where the animals could be
compensated for the insufficient grazing, at least partially, in the course of the campaign.
156 In the 13th century Ibn al Athīr reports: “The Tatars remained in the Qipjaq lands, which
are plentiful in pasturages summer and winter. There are places that are cool in summer
with plentiful pasturage and others warm in winter, also with plentiful pasturage, the latter
being wooded areas on the sea coast.”, IBN AL-ATHĪR/RICHARDS 2008, 223; IBN AL-ATHĪR 1884,
26; see also AL-NUWAYRĪ 1884, 540. Jūzjānī also refers to the natural advantages of the
steppes inhabited by the Cuman-Qïpchaqs without relating them to a particular season:
‫ و‬،‫ دانست که در همه جهان زمین از ان نزه تر‬،‫ هوا و آب و زمین قفچاق را بدید‬،‫« چون توشی که پسر مهتر چنگیز خان بود‬
»‫ و مرغزارها و چراگاهها از ان وسیع تر نتواند بود‬،‫ و آبی از ان لطیف تر‬،‫( هوائی از ان خوشتر‬When Tushi [i.e.
Jochi], who was the eldest son of Chingīz Khān, saw the climate of the Qifchāq land, he
considered that in the whole world there could not be a healthier land than that, a climate
better than that, a water pleasanter than that, meadows and pasture-lands more extensive
than that.), JŪZJĀNĪ/HABĪBĪ 1343/1964, vol. 2, 150; AL-JUZJANI/RAVERTY 1970, 1101;
JŪZJĀNĪ/TIZENGAUZEN 1941, 14.
157 Already Rasovsky pointed out the fact that the frontiers between the territories populated
by the Cuman-Qïpchaqs and the habitat of the sedentary societies were rather fluid and not
always clearly discernible. During their seasonal migration, the nomadic groups often ad-
Rus’ in order to spend the summer in Dasht-i Qipchāq.158 Apparently, strategical

considerations for the further stages of the campaign stood behind this decision
(the Chinggisids opened the next military season, i.e. the autumn of 1238, with
operations against Caucasia, Crimea, and the Qïpchaq tribes),159 and perhaps it
was also influenced by the intention of the conquerors to fatten up their horses
within the steppe during the summer. 160 Maybe the Cuman tribes never operated
in the Rus’ principalities in such large numbers and for such a long time as to be
forced to conduct a mass exodus during the hot part of the year. In any event, the
Cumans faced no difficulties in campaigning in Rus’ during the summer in the
previous two hundred years.161
Conclusions
Thus, it turns out that something as natural and trivial as the quality of the availa-
ble grazing stood behind the asymmetric seasonal military activity of the Cuman-
Qïpchaqs in Southeastern Europe and Central Asia, and influenced the political
history of these regions. For it was the seasonal rhythm of Cuman warfare that
forced the Bulgarian tsar Kaloyan twice to cede his dream of capturing Adrianople
(1205 and 1207) while he was on the verge of conquering it. More than eight dec-
ades ago, Rasovsky apparently noticed the meaning of the sun-burnt grass around
vanced beyond the closest arable fields, whereas in the steppe there were sedentary en-
claves, see RASOVSKY 1937, 71–85; RASOVSKY 1938, 155–78.
158 See above in the text and n. 126.
1338/1959, vol. 1, 476–7; RASHĪD AL-DĪN/BOYLE 1971, 60; RASHĪD AL-DĪN/THACKSTON 1998–9,
part 2, 327; RASHĪD AL-DĪN/VERKHOVSKY 1960, 39.
160 Perhaps this was not the first time the Mongols spent the summer within the steppes of
the Cuman-Qïpchaqs in order to fatten up their horses, as indicated by events that took
place in the course of the same campaign. A year earlier, after the Qïpchaq chief Bachman
was finally defeated, at least part of the Mongol troops headed by the future Great Khan
Möngke (r. 1251–9) spent the summer of 1237 in Dasht-i Qipchāq, in the Volga region,
RASHĪD AL-DĪN/RAWSHAN – MŪSAVĪ 1373/1994, vol. 1, 668; RASHĪD AL-DĪN/KARĪMĪ 1338/1959,
vol. 1, 476; RASHĪD AL-DĪN/BOYLE 1971, 59; RASHĪD AL-DĪN/THACKSTON 1998–9, part 2, 326;
RASHĪD AL-DĪN/VERKHOVSKY 1960, 38.
161 Apparently, such was also the situation in the Pontic steppes in later centuries. The Cri-
mean and Nogai Tatars raided the Polish-Lithuanian and Muscovite territories in the Early
Modern Age all year round. In a similar way to the Cumans, the Tatars were able to conduct
a whole series of raids in one year. In contrast to such regular incursions, L. Collins consid-
ers that the campaign season for large Crimean forces lasted from early February to October
(the period between February and August being the most suitable time, according to him).
But even in this case there were exceptions of large-scale campaigns “in the depths of win-
ter”, COLLINS 1975, 264, 267, 268; DAVIES 2007, 20, 22; KHODARKOVSKY 2002, 17.
Adrianople for the withdrawal of the Cumans.162 But in his research the Russian
scholar put the accent upon the seasonal migrations of the Cuman-Qïpchaqs inside
the steppes. However, the problem of the availability of grazing on the battlefield
should be examined and placed in the broader context of the traditional nomadic
warfare, which was shaped by the seasonal asymmetry of the climatic factors.
There is no doubt that climatic conditions varied each year, while political and
social factors could force some Cuman groupings to endure extreme stress cam-
paigning during an unfavourable season. This explains the exceptions to the rule of
the Cuman-Qïpchaq seasonal activity in the Balkans and Khwārazm. Viewed in a
broader perspective, the climate is nothing more than a fluid frame within which
the human societies, and these of the nomads in particular, developed. Besides the
moments of particularly harsh climatic extremes, the limitations of this frame could
often be overcome and should not be viewed as an absolute determinant. Yet, in
the long term the seasonal cycle continued to define the curve of Cuman-Qïpchaq
activity in these zones and nature prevailed over politics. Even such a cataclysm as
the Mongol invasion in Dasht-i Qipchāq and the resulting establishment of the
centralized Golden Horde did not alter the asymmetric seasonal models of nomadic
warfare in the Balkans. Those Cumans who survived under the shadows of the
Chinggisids continued to support the Bulgarian tsar Michael II Asen (r. 1246–56)
during the cold months,163 whereas Pachymeres, describing events of 1282, explic-
itly reports that the Tatars “are accustomed to wage war in the winter”. 164 Indeed,
according to the Mamlūk chroniclers the joint incursion by the Bulgarian tsar Kon-
stantin Tikh Asen (r. 1257–77) and the troops of the Golden Horde in Byzantine
Thrace (which could be dated to 1263–4 and led to the liberation of the former Sel-
juq Sultan ʿIzz al-Dīn Kaykāvus II (r. 1246–61, d. 1279/80)) took place in “the win-
ter time”.165 The only nomads that mаnaged to break from the seasonal cycle of
162 In the beginning of his article, Rasovsky comments on the sun-burnt grass in the steppes,
but further in the text he seems to touch on the problem of the dried grass and the lack of
pasture for the Cuman horses in the area of Adrianople, RASOVSKY 1939, 203, 208.
163 The described events took place at the end of 1255 and the beginning of 1256, ACROPOLI-
TA/HEISENBERG 1903, 123, 124–6; AKROPOLITES/MACRIDES 2007, 297, 300–1; see also: THE-
ODOROS SKOUTARIOTES/VOYNOV 1972, 292–4; Chronographia of EPHRAIM/GYUZELEV – LYU-
BENOV 1981, 93–4; ZLATARSKI 2007, 457–8, 462–3; DPI 1993, 59; PAVLOV 1996, 193, 197, n. 23.
164 This information is in the context of the campaign which Emperor Michael VIII Pal-
aiologos (r. 1261–82) organized against the ruler of Thessaly Sebastokrator John I (ca. 1271–
89): “Εὔκαιρον δ' εἶχε καὶ τὸν ἐφεστῶτα χειμῶνα – χειμῶνος γὰρ καὶ σύνηθες ἐκείνοις
στρατεύειν –, ὃν καὶ προκαταλαβεῖν ἠπείγετο, ὡς ἔξω πόλεως συμβαλεῖν Τοχάροις.” (And he
also considered well timed the coming winter—because they are accustomed to wage war in
the winter—and he was in a hurry to get ahead of it [the winter] in order to join with the
Tocharians [i.e. the Tatars] outside the City [i.e. Constantinople].), PACHYMERES 1984, 658–9;
PACHYMERES/KARPOV 1862, 484–5; for these events, see also VÁSÁRY 2005, 84; KRASTEV 2011, 95.
165 AL-ʿAYNĪ/TIZENGAUZEN 1884, 511. For the dating of the campaign, see MUTAFCHIEV 1973,
615–6. Ibn Bībī also reports that during this campaign the winter was so harsh that the Dan-
their activities in these contact zones were those who permanently left the steppes
and settled within the outside world. Such were the Qïpchaq migrants who were
pushed by the Rus’ pressure to the Georgian Kingdom in the early 12 th century; the
Qïpchaqs who migrated to the empire of the Khwārazmshāhs-Anushteginids in
the late 12th century;166 and the Cuman refugees who ended their odyssey in Nicaea
during the 1240s.
Appendix: Seasonal activity of the Cumans in their contacts with the Rus’
The present table contains only cases when the season of the campaign is relatively
clear. The events are dated according to the chronology of the chronicles, which
often diverge by one or several years from the real dates. The year is given accord-
ing to the leading chronicle, without referring to possible variants in the chronolo-
gy of additional sources. Since the real dating of the listed events is not of primary
importance for the present topic, the voluminous commentaries related to this
issue are omitted with the exception of the cases when it is directly related to the
season of the military activities. Years that, apart from the listed campaigns, also
include reports of other military operations without specific information on the
time of the year are marked with an asterisk (*) in order to indicate Cuman activi-
ties outside of the explicitly mentioned season. For the sake of simplicity, the reign-
ing years of the Rus’ princes in the various principalities are omitted.
ube froze and the Tatar army crossed over without difficulties, IBN BĪBĪ/MOTAḤEDDĪN
1390/2011, 553; İBN BIBI/ÖZTÜRK 2014, 590 (I am indebted to Dr. Delyan Rusev, who drew
my attention to this evidence); see also PACHYMERES 1984, 294–5; PACHYMERES/KARPOV 1862, 205.
166 As regards the latter, however, it should be taken into consideration that the frontiers
between the nomadic and the sedentary habitat in Iran and Central Asia were much more
fluid than in Eastern Europe due to the geographical specifics of these regions. Thus, the
two modes of life often coexisted in the same zones located thousands of kilometres away
from the Eurasian steppe belt.
Cold Months Hot Months

Year
(October–April) (May–September)
The first Cuman incursion into the Rus’
1061
lands took place in the winter.167
Attack against the Rus’ principalities in
the autumn:
1068
- The Battle of the Snov River (No-
vember 1); 168
The Rus’ princes Oleg Svyatoslavich
and Boris Vyacheslavich attacked the
Rus’ lands with an allied Cuman
army in the summer:
1078
- The Battle of the Sozhitsa River
(Аugust 25)170
The Battle of Nezhatina Niva (October
3);169
The winter of 1078–9(?): Cumans pil-
laged the area of Starodub and were
subsequently defeated twice near Des-
na by Vladimir Monomakh and his
own Cuman allies.171
The Rus’ prince Roman Svyatoslavich
1079
led an allied Cuman army against the
Rus’ lands in the summer and negoti-
ated with his uncle Vsevolod Yaro-
slavich near Voin’. He was killed by
the Cumans during the retreat
through the steppes on August 2.172
In the summer, Vladimir Monomakh
persecuted the Cumans who had
captured Goroshin before that.175
1080–6173
The following autumn Monomakh
captured Minsk with allied Cuman
troops.174
167 PSRL, vol. 1, col. 163; PSRL, vol. 2, col. 152.

168 PSRL, vol. 1, cols. 171–2; PSRL, vol. 2, col. 161.
169 PSRL, vol. 1, cols. 201–2; PSRL, vol. 2, cols. 192–3. The participation of Cumans in this
battle is not certain.

171 PSRL, vol. 1, col. 248.
172 PSRL, vol. 1, col. 204; PSRL, vol. 2, col. 195–6.
173 The events during this period cannot be accurately dated by years.
174 PSRL, vol. 1, col. 248.
175 PSRL, vol. 1, col. 248.
Monomakh fought against significant

Cuman forces in the environs of
Priluk and Belaya Vezha (August 14–
15).176
A Cuman incursion after intense
1092
heat.177
A large Cuman offensive in the spring
and summer, started after Apri 14:
- The Battle of the Stugna River
1093
(May 26);
- The Battle near Zhelan’ (July
23);178
Oleg Svyatoslavich attacked Vladimir
Monomakh in Chernigov with an
1094 allied Cuman army in the height of
summer. Monomakh surrendered
Chernigov to Oleg on July 24.179
In late autumn, the Cuman chiefs Itlar
and Kitan arrived at Pereyaslavl’ for
peace talks with Vladimir Monomakh.
After they were treacherously mur-
1095
dered there, a spring Rus’ raid in the
steppes followed.180
“The entire summer” the Cumans
besieged Yur’ev.181
Intensive military activity of the Cu-
mans against the Rus’ principalities in
spring and summer:
- The Cuman Chief Bonyak at-
tacked the environs of Kiev;
- The Cuman Chief Kurya attacked
the environs of Pereyaslavl’ (May
1096
24);
- Tugorkan besieged Pereyaslavl’
(from May 30 until July 19) and
was killed in the Battle of the
Trubezh River (July 19);
- Bonyak attacked the outlying dis-
tricts of Kiev (July 20).183
176 PSRL, vol. 1, cols. 248–9.

179 PSRL, vol. 1, col. 249.
In late winter (end of February), a

Cuman named Kunuy with Cuman
forces supported two of Monomakh’s
sons against Oleg Svyatoslavich in
Northeast Rus’.182
Late April: Cumans headed by
Bonyak took part in the Battle of the
Vigor River (on an unidentified day
between April 9 and June 12) on the
side of David Igorevich against Yaro-
1099
slav Svyatopolchich and his Hungari-
an allies;
A second appearance of Bonyak’s
Cumans in support of David Igo-
revich after August 5.184
Late summer: Large meeting for peace
talks between many Rus’ princes and
1101
Cuman chiefs in Sakov (September
15).185
Rus’ incursion into the steppes in late
1103
winter.186
Bonyak attacked Zarub during the
1105
winter.187
Cumans attacked Zarechesk in the
1106
spring (before 24 June).188
- Bonyak attacked Pereyaslavl’ in
May;
- Cumans besieged Lubno during
the summer;
1107
- The Battle of the Sula River (Au-
gust 12).190
Peace negotiations between the Rus’
princes and the Cuman chiefs (January

182 PSRL, vol. 1, cols. 239–40; PSRL, vol. 2, cols. 229–30; PSRL, vol. 3, 19, 202. The chronology
of the Rus’ chronicles follows the March-style or the ultra-March style, and in both cases the
New Year starts on March 1. Thus, the events in February described s.a. 1096 actually took
place in early 1097 according to the Gregorian calendar. This is why in this and subsequent
similar instances the events are displayed in the table after the description of the summer
months of the given year.
187 PSRL, vol. 2, col. 257.
12).189
A Rus’ winter incursion into the
steppes. On December 2, the Rus’ at-
1109
tacked the Cuman camps “on the
Don”.191
An unfulfilled Rus’ campaign against
the Cumans in the spring.192
An unsuccessful Cuman incursion.
1110* Two more undated Cuman incursions
followed, at least one of which per-
haps took place in the hot part of the
year.193
A spring Rus’ campaign in the steppes,
1111 the Battle of the Salnitsa River
(March).194
Cuman forces arrived after the death
of Svytopolk on April 16.195 This un-
1113
successful incursion perhaps took
place in late spring.
After the death of Vladimir Mono-
makh (May 19), the Cumans attacked
1125
Pereyaslavl’ in late spring or in the
summer.196
Cumans supported Vsevolod
1135
Ol’govich during the winter.197
Cumans supported the Ol’govichi in
the summer (August).199
1136
Cumans supported the Ol’govichi in
the winter (December and January).198
Cumans joined the campaign of
Svyatoslav Ol’govich (ruling in Novgo-
1137
rod) against Pskov in the cold part of
the year (?).200
190 PSRL, vol. 1, cols. 250, 281–2; PSRL, vol. 2, col. 258.
189 PSRL, vol. 1, cols. 250, 282–3; PSRL, vol. 2, col. 259.
193 PSRL, vol. 2, col. 260.
194 PSRL, vol. 1, col. 289; PSRL, vol. 2, cols. 264–8, 273.
195 PSRL, vol. 2, col. 276.
197 PSRL, vol. 1, сol. 303; PSRL, vol. 2, cols. 295–6.
198 PSRL, vol. 2, cols. 299–300; PSRL, vol. 3, 23–24, 208–9.
199 PSRL, vol. 1, cols. 303–4.; PSRL, vol. 2, cols. 297–9; PSRL, vol. 3, 208.
200 PSRL, vol. 3, 210.
“Wild” Cumans supported Vsevolod

Ol’govich in late winter or early spring.
“Wild” Cumans supported Svyatoslav
1146* Ol’govich next January. In February,
Svyatoslav Ol’govich allowed his Cu-
man allies to come back to the
steppes.201
- Cumans supported Svyatoslav
Ol’govich several times (between
Palm Sunday and September).203
- Izyaslav Mstislavich concluded peace
with the Cumans (after July 27).204
- Many Cumans swore an oath of
1147 alliance with Svyatoslav Ol’govich
and Gleb Yur’evich (after Igor
Ol’govich was killed in September,
but before the rivers froze).
- Cumans supported the Ol’govichi
and the Davidovichi after the rivers
froze.202
Cumans flocked to the army of Yury
Dolgoruky in July. After he met the
Ol’govichi, they waited other Cuman
troops for a month. The entire coali-
tion participated in the Battle of
1149 Pereyaslavl’ on August 23.206
Yury Dolgoruky attacked with allied
Cuman troops Izyaslav Mstislavich in
Luchesk at the end of winter. After the
arrival of spring, both parties conclud-
ed peace.205
Cumans arrived to support Yury
Dolgoruky without being called and
1150/1 created troubles in the surroundings
of Pereyaslavl’ before the end of
summer.207
Cumans arrived to support Yury
1151/2
Dolgoruky after Saint George’s Day,
201 PSRL, vol. 2, cols. 319, 329, 333–5, 339.

202 PSRL, vol. 2, cols. 353, 356–8, 359;
203 PSRL, vol. 2, cols. 340, 341–2, 353.
204 PSRL, vol. 1, col. 315.
suffered defeat, and withdrew to the

steppes before July 24.208
Cumans arrived to support Yury Dol-
goruky in the autumn, and after that
the rivers started to freeze.
1152 Around the end of winter, Mstislav
Izyaslavich organized a campaign against
the Cumans with “all the Black
Hats”.209
A small Cuman party joined an un-
successful campaign by Yury Dol-
goruky in the summer(?).211
At the end of autumn or the beginning
of winter, Gleb Yur’evich attacked
Mstislav Izyaslavich in Pereyaslavl’
1154 with a multitude of Cumans.
In the winter, Gleb Yur’evich with a
multitude of Cumans supported Izyaslav
Davidovich against Rostislav Mstislav-
ich near Chernigov.
In the same winter, the Cumans plun-
dered the surroundings of Pereyaslavl’.210
In the spring, Cumans attacked the
territories around the Ros’ river and
1154/5*
were defeated by Vaslilko Yur’evich
and the Berendei.212
Cumans supported Ivan Rostislavich
Berladnik in an unspecified moment.
1158/9* Cumans joined a campaign by Izyaslav
Davidovich that ended in defeat in late
autumn or early winter.213
- Oleg Svyatoslavich defeated a
Cuman force and killed a Cuman
chief(?).
- Vladimir Andreevich, Yaroslav
1159/60*
Izyaslavich, and troops from the
Pricipality of Galicia defeated
Cumans on Rus’ territory and lib-
erated many captives(?).
208 PSRL, vol. 1, cols. 330–4; PSRL, vol. 2, cols. 422–40, 443.
209 PSRL, vol. 1, cols. 338–40; PSRL, vol. 2, cols. 455–61;
210 PSRL, vol. 1, cols. 341–344; PSRL, vol. 2, cols. 468–9, 471–6;
213 PSRL, vol. 1, col. 348; PSRL, vol. 2, cols. 497, 500–2.
- Campaign by Izyaslav Da-

vidovich with a multitude of
Cumans against Chernigov(?).215
- Izyaslav Davidovich made an incur-
sion into the Rus’ lands with a mul-
titude of Cumans (probably in the
autumn?).
- A winter campaign of Izyaslav Da-
vidovich with Cumans against Smo-
lensk.214
A multitude of Cumans arrived to
support Izyaslav Davidovich (maybe
in August).217
1161
Many Cumans took part in the offen-
sive by Izyaslav Davidovich against
Kiev in February.216
The Cumans served as a vanguard for
the troops of Izyaslav Davidovich in
1162* the winter. Izyaslav Davidovich was
killed in a battle (March 6) and many
Cumans were slain.218
A winter campaign of the Ol’govichi in
1168
the steppes against the Cumans.219
In early March, Mstislav Izyaslavich
headed a campaign against the Cu-
1170
mans aimed at protecting the trade
routes in the steppes.220
In the winter, Cumans attacked the
environs of Kiev. Mikhalko and Vsevo-
1171
lod Yur’evichi counterattacked with
Berendei and Torki.221

215 PSRL, vol. 1, cols. 349–50; PSRL, vol. 2, cols. 504–8. These three conflicts are described by
the chronicler after the meeting of Rostislav Mstislavich and Svyatoslav Ol’govich in May
and before the fights in the winter. Therefore, perhaps at least some of them took place in
the hot season.
216 PSRL, vol. 2, cols. 515–16.
217 PSRL, vol. 2, cols. 512, 514–15.
218 PSRL, vol. 2, cols. 517–18; PSRL, vol. 3, 31, 218. These events were part of a campaign that
started in the previous year.

219 PSRL, vol. 2, col. 532.
220 PSRL, vol. 2, cols. 538–41; PSRL, vol. 3, 32–3, 220.
221 PSRL, vol. 1, cols. 362–3.
Cumans supported Gleb Yur’evich and

his allies around Easter.222
Gleb Yur’evich tried to lead parallel
parleys with two Cuman groupings.
1172
He reached an agreement with one of
them; the other pillaged the environs
of Kiev but suffered defeat afterwards
(between May and August?).223
In the winter, Cumans supported the
Ryazan Prince Gleb Rostislavich
against Vsevolod the Big Nest. Togeth-
er with Gleb, the Cumans suffered
defeat at the end of winter.224
At the end of spring or the beginning
of summer (the Green Week), Cumans
attacked the Principality of Kiev,
pillaged some settlements, and de-
1177
feated the Rus’ troops.
After Svyatoslav Vsevolodovich en-
tered Kiev on the Feast of the Prophet
Elijah (July 20), he called the Cumans
to support him. When the nomads
learned that he had withdrawn from
the city, they attacked the area of
Tortsk (in late summer or early au-
tumn?).225
Svyatoslav Vsevolodovich attacked
Vsevolod the Big Nest with Cuman
1181
allies but retreated before the river
levels rise in spring.226
- In February, the Cumans, led by
Konchak and Gleb Tirievich, at-
tacked Rus’.
- The leading Rus’ princes ruling the
1183
lands along the Dnieper marched on
a retaliatory campaign but retreated
and decided that it would be better
to attack again in the summer.
222 PSRL, vol. 2, cols. 548–50.

223 PSRL, vol. 1, cols. 355, 357–61; PSRL, vol. 2, cols. 552, 555–9. The negotiations and the
incursion are described in the chronicle between events dated in May and August, respec-
tively. The same Cuman activities are reported in the Laurentian Chronicle s.a. 1169.
224 PSRL, vol. 1, cols. 383–5.
225 PSRL, vol. 2, cols. 603–5.
226 PSRL, vol. 1, col. 388.
- Igor Svyatoslavich launched a retal-

iatory campaign at the end of winter
and ravaged the Cuman camps on
the Khorol River.227
A large Rus’ steppe campaign that
ended with a battle at the end of July
and a victory over the nomads. A
parallel campaign of the Ol’govichi
about the same time, during which
they confronted and defeated a Cu-
man host that was marching on a raid
against the Rus’ territories.228
Konchak marched against the Rus’
territories bringing with him a Muslim
specialist in siege warfare. A surprising
Rus’ counterattack and battle on March
1, 1185. The rasputitsa hampered the
chase of the defeated nomads.
In April, the Voivode Roman Nezdilo-
vich with the Berendei attacked the
Cuman camps.229
1184/5*
Igor Svyatoslavich and other princes
marched on a steppe campaign at the
end of April and suffered a defeat in
the Battle of the Kayala River in May.
Svyatoslav Vsevolodovich had
planned a campaign in the area of
Donets for the entire summer, but his
plans were thwarted by the defeat
near the Kayala River.230
A campaign of the Rus’ princes and the
Black Hats against the Cumans in
spring. The Rus’ were unable to chase
the Cumans beyond the Dnieper be-
cause the water level rose quickly.
1187*
An unsuccessful winter campaign of
the Rus’ princes against the Cumans.
The same winter, Roman Nezdilovich
and the Black Hats attacked the Cuman
camps beyond the Dnieper since the
227 PSRL, vol. 2, cols. 628–9.

229 PSRL, vol. 2, cols. 634–7.
nomads were on the Danube to sup-

port the Asenids.231
During the winter, the Cumans and the
renegade Kontuvdey from the Black
Hats often raided the frontier areas of
the Principality of Kiev.
A retaliatory expedition of two Rus’
princes and the Black Hats against the
1190
Cumans at the end of winter.
An unsuccessful Cuman raid at the end
of winter.
Another joint attack by the Cumans
and Kontuvdey at the end of winter,
which ended in defeat.232
A campaign of the Ol’govichi against
the Cumans in an unspecified time.
1191*
Another steppe incursion by the
Ol’govichi during the winter.233
During the summer, the princes
guarded the Rus’ Land from the Cu-
mans.235
In the autumn, the Black Hats wanted
to attack the camps of the Cumans,
who were on the Danube, but Ryurik
1192 Rostislavich did not allow them.
An unsuccessful campaign of the Rus’
princes and the Black Hats against the
Cumans.
Negotiations by Ryurik Rostislavich
with Kontuvdey and the Cumans in
the winter.234
Parallel unsuccessful peace talks with
the Cumans in the autumn.
The princes wanted to organize a win-
ter campaign against the Cumans, but
1193 were unable to do so due to the poor
harvest.
A successful incursion by Rostislav
Ryurikovich and Mstislav Mstislavich
with the Black hats against the Cu-
231 PSRL, vol. 2, cols. 652–4, 659.

232 PSRL, vol. 2, cols. 668–73.
233 PSRL, vol. 2, col. 673.
234 PSRL, vol. 2, cols. 673–4
235 PSRL, vol. 2, col. 673.
mans. They came back on Christmas.

Svyatoslav Vsevolodovich and Ryurik
Rostislavich guarded for a long time
the frontiers of the Principality of Kiev,
but as soon as they withdrew the Cu-
mans launched a winter incursion.236
The “Wild” Cumans arrived to support
Ryurik Rostislavich in the winter, but
the campaign was aborted.237
1195
At the end of winter or in spring,
Vsevolod the Big Nest summoned
allied Cuman troops.238
The “Wild” Cumans supported Ryu-
rik Rostislavich in the summer, until
autumn arrived.240
Cumans took part in a campaign by
Vsevolod the Big Nest against the
1196 Ol’govichi, which perhaps started in
the summer and continued until the
In the autumn, the “Wild” Cumans autumn. 241
supported the Ol’govichi against
Vsevolod the Big Nest, whose troops
also included Cumans.239
From the last day of April until the
early June, Vsevolod the Big Nest led
a steppe campaign against the Cu-
1199 mans. The nomads withdrew towards
the sea and the prince traversed their
winter pasture grounds, but was
unable to catch them.242
In the winter, Roman Mstislavich at-
1202
tacked the Cuman camps.243
In the winter, a coalition of Rus’ princ-
1203 es, supported by Cuman allies, pil-
laged Kiev.244
1205 In the winter, Ryurik Rostislavich,
236 PSRL, vol. 2, cols. 675–9.

237 PSRL, vol. 2, cols. 689–90; and retrospectively s.a. 1196, PSRL, vol. 2, col. 694.
238 PSRL, vol. 3, 42, 235; PSRL, vol. 1, col. 413.
239 PSRL, vol. 2, cols. 698–700.
240 PSRL, vol. 2, cols. 695–6.
241 PSRL, vol. 3, 42–3, 235–6; PSRL, vol. 1, col. 413; PSRL, vol. 2, cols. 698–700.
242 PSRL, vol. 1, cols. 414–15.
243 PSRL, vol. 1, col. 418.
244 PSRL, vol. 1, cols. 418–19; PSRL, vol. 3, 45, 240.
Roman Mstislavich, Yaroslav Vsevo-

lodovich, and other princes attacked
the Cumans.245
In the winter, the Cumans supported
1206* Vsevolod The Red and the Ol’govichi
against Ryurik Rostislavich in Kiev.246
On July 20, Cumans participated in
the treacherous massacre of six princ-
1217
es in Ryazan organized by Gleb Vla-
dimirovich.247
A winter campaign of Mstislav Msti-
slavich, Mstislav Romanovich, and
1220 other princes with Cuman allies
against the Hungarians in the Princi-
pality of Galicia.248
Cumans supported Mstislav Mstislav-
[1221] ich in the Principality of Galicia (end of
March).249
Around the end of summer, Cumans
supported Daniil Romanovich in one
[1233]
of his conflicts with the Hungarians
over Galich.250
Izyaslav Vladimirovich attacked with
his Cuman allies the environs of Kiev.
They plundered the area and won a
[1235]
battle against Daniil Romanovich and
Vladimir Ryurikovich soon after the
Ascension of Christ (May 17, 1235).251
In the summer, Mikhail Vsevolodo-
vich and Izyaslav Vladimirovich
[1236] summoned allied Cuman contingents
against Daniil Romanovich. Later on,
however, the Cumans refused to fight
245 PSRL, vol. 1, col. 420; PSRL, vol. 3, 240.

246 PSRL, vol. 1, col. 428.
247 PSRL, vol. 1, cols. 440–1; s.a. 1218: PSRL, vol. 3, 58.
248 PSRL, vol. 7, 128.
249 PSRL, vol. 2, col. 737; KRONIKA HALICKO-WOŁYŃSKA/DĄBROWSKI – JUSUPOVIĆ 2017, 79, for
the dating see n. 267. The dates of the events described in the Galician-Volhynian Chronicle are
given according to the real course of events and do not follow the chronology of the chroni-
cle, which has been inserted much later in the original text in a completely random manner.
250 PSRL, vol. 2, cols. 770–1; KRONIKA HALICKO-WOŁYŃSKA/DĄBROWSKI – JUSUPOVIĆ 2017,
178–84; HRUSHEVSKY 1901, 24.

188–92.
against Daniil and retreated after they

had pillaged “the entire land of Ga-
lich”.252
The Cumans took part in the Battle of
Yaroslavl’ on August 17 on the side of
[1245] Daniil and Vasil’ko Romanovichi, and
they were the first to cross a ford over
the San River.253
In the winter, the Cumans took part in
[1251- a campaign of Daniil and Vasil’ko
1252] Romanovichi against the
Lithuanians.254
Bibliography
ABŪ АL-GHĀZĪ/SABLUKOV (1906): Rodoslovnoe drevo tyurkovʺ, transl. Sablukov,
G. S. Izvestiya obshchestva arheologii, istorii i etnografii pri Imperatorskomʺ Ka-
zanskom universitete, 22/5–6, i–xvi, 1–224.
ACROPOLITA/HEISENBERG (1903): Opera, ed. Heisenberg, A. Vol. 1, Lipsia.
AKROPOLITES/MACRIDES (2007): The History, transl. Macrides, R. Oxford.
ALLSEN, T. T. (1987): Mongol Imperialism. The Policies of the Grand Qan Möngke on
China, Russia and the Islamic Lands, 1251–1259, Berkeley–Los Angeles–London.
AMITAI-PREISS, R. (1995): Mongols and Mamluks. The Mamluk-Ilkhanid War, 1260–
1281, Cambridge.
AMITAI-PREISS, R. (1999): Northern Syria between the Mongols and Mamluks: polit-
ical boundary, military frontier, and ethnic affinities, in Power, D. – Standen,
N. (ed.:) Frontiers in Question: Eurasian Borderlands, 700–1700. Basingstoke, 128–152.
AMITAI, R. (2002): Whither the Ilkhanid army? Ghazan’s first campaign into Syria
(1299–1300), in Di Cosmo, N. (ed.:) Warfare in Inner Asian History (500–1800).
Leiden, 221-264; reprinted in: AMITAI, R. (2007): The Mongols in the Islamic Lands.
Studies in the History of the Ilkhanate, Ashgate, article XV.
ANA KOMNINA/VOYNOV (1972): Aleksiada, in: Gratski izvori za balgarskata istoriya,
vol. 8, ed. Voynov, M. Sofia, 7–149.
ANA KOMNINA/LYUBARSKY (1996): Aleksiada, transl. Lyubarsky, Ya. N. Saint Peter-
sburg.
ANNA COMNENA/DAWES (2000): The Alexiad, transl. Dawes, E. A. S. Cambridge,
Ontario.
ANNA KOMNENE/REINSCH – KAMBYLIS (2001): Alexias, CFHB, vol. XL/1, eds. Reinsch,
D. R. – Kambylis, A. Berlin–New York.
252 PSRL, vol. 2, col. 775; KRONIKA HALICKO-WOŁYŃSKA/DĄBROWSKI – JUSUPOVIĆ 2017, 196–8.
253 PSRL, vol. 2, col. 802; KRONIKA HALICKO-WOŁYŃSKA/DĄBROWSKI – JUSUPOVIĆ 2017, 276–7.
326–9.
ANNALES MONASTICI/LUARD (1865): Annales monastici. Vol. 2. Annales monasterii de

Wintonia (A.D. 519–1277). Annales monasterii de Waverleia (A.D. 1–1291), ed.
Luard, H. R. London.
ANNINSKY, S. A. (1940): Izvestiya vengerskikh missionerov XIII-XIV vv. o tatarkh v
Vostochnoi Evrope, Istoricheskii Arkhiv 3, 71–112.
ATTALEIATES/KALDELLIS – KRALLIS (2012): The History, Transl. Kaldellis, A. – Kral-
lis, D. Cambridge MA–London.
AL-ʿAYNĪ/TIZENGAUZEN (1884): in Sbornikʹ materialovʹ, otsnosyashchihsya kʺ istorii
Zolotoj ordy, vol. 1, transl. Tizengauzen, V. Saint Petersburg, 475–534.
BAGHDĀDĪ/BAHMANYĀR (1384/2005-2006): Al-Tawassul ilā l-tarassul, ed. Bah-
manyār, A. Tehran.
BARFIELD, T. J. (1989): The Perilous Frontier. Nomadic Empires and China, Cambridge,
M. A.–Oxford UK.
BARTOL’D, V. (1898): Turkestanʹ v” epokhu mongolʹskago nashestviya. Part. 1. Teksty,
Saint Petersburg.
BARTOL’D, V. (1900): Turkestanʹ v” epokhu mongolʹskago nashestviya. Part. 2, izsledo-
vanie, Saint Petersburg.
BAYHAQĪ/KHAṬĪB RAHBAR (1393/2014): Tārīkh-i Bayhaqī, ed. Khaṭīb Rahbar, Kh.
Vol. 1–3. Tehran.
BAYPAKOV, K. M. (1986): Srednevekovaya gorodskaya kulʹtura Yuzhnogo Kazakhstana i
Semirechiya (VI – nachalo XIII v.), Alma-ata.
BERNOLDI CHRONICON/PERTZ (1844): Bernoldi Chronicon, in Monumenta Germaniae
Historica, ed. Pertz, G. H. Scroptorum, vol. 5, Hannover, 385–467.
BAYHAQĪ/ARENDS (1969): Istoriya Mas‘uda, (1030-1041), transl. Arends, A. K. Moscow.
BEYHAQI/BOSWORTH (2011): The History of Beyhaqi. (The History of Sultan Masʻud of
Ghazna, 1030-1041), transl. Bosworth, C. E. Vol. 1-3. Cambridge, MA–London.
BIBIKOV, M. V. (1976): Vizantiiskie khroniki i lokalizatsiya polovetsko-vizantiiskoi voiny
1148 g., in Letopisi i khroniki. 1976 g. M. N. Tikhomirov i letopisovedenie. Moscow,
117–122.
BIBIKOV, M. V. (1981): Vizantiiskie istochniki po istorii Rusi, narodov Severnogo
Prichernomorʹya i Severnogo Kavkaza (XII–XIII vv.), in Drevneishie gosudarstva
na territorii SSSR. (1980), Moscow, 5–151.
BIRAN, M. (2005): The Empire of Qara Khitai in Eurasian History, Cambridge.
BOYLE, J. A. (1968): Dynastic and Political History of the Īl-khāns, in Boyle, J. A.
(ed.:) The Cambridge History of Iran. Vol. 5. The Saljuq and Mongol Periods. Cam-
bridge.
BREGEL, Y. (2003): An Historical Atlas of Central Asia, Leiden–Boston.
BRETSCHNEIDER, E. (1910): Mediaeval Researches from Eastern Asiatic Sources, vol. 1,
London.
BUTINYATOV, Z. M. (1986): Gosudarstvo khorezmshakhov Anushteginidov, 1097–1231,
Moscow.
CAHEN, C. (2001): The Formation of Turkey. The Seljukid Sultanate of Rūm: Eleventh to
Fourteenth Century, Harlow.
CHONIATES/VAN DIETEN (1972): Nicitae Choniatae Orationes et Epistulae, ed. van Di-
eten, J-V. Berlin–New York, 1972.
CHONIATES/VAN DIETEN (1975): Nicitae Choniatae Historia, ed. van Dieten, J-V. Ber-
lin–New York.
CHONIATES/VOYNOV (1983): Rechi, in: Gratski izvori za balgarskata istoriya, vol. 11,
ed. Voynov, M., Tapkova-Zaimova, V., Yonchev, L. Sofia, 94–117.
CHONIATES/MAGOULIAS (1984): O City of Byzantium, Annals of Niketas Choniatēs,
transl. Magoulias, H. J. Detroit.
CHRONICI HUNGARICI/DOMANOVSZKY (1999): Chronici Hungarici compositio
saeculi XIV, ed. Domanovszky, A. in Szentpétery, E. (ed.:) Scriptores rerum
Hungaricarum. Vol. 1, 217–505.
COLLINS, L. J. D. (1975): The Military Organization and Tactics of the Crimean Tatars
16th – 17th centuries, in Parry, V. J. – Yapp, M. E. (ed.:) War, Technology and Soci-
ety in the Middle East. Oxford, 257–276.
CONTINUATIO SANCRUCENSIS SECUNDA/Wattenbach (1851): Continuatio Sancrucen-
sis Secunda – ed. Wattenbach, W. in Monumenta Germaniae Historica. Scripto-
rum, vol. 9, ed. Pertz G. H. Hannover, 637–646.
THE CONTINUATION/MCGEER (2020) Byzantium in the Times of Troubles. The Continu-
ation of the Chronicle of John Skylitzes (1057–1079), transl. McGeer, E. Boston–
Leiden.
CURTA, F. (2006): Southeastern Europe in the Middle Ages 500–1250, Cambridge, 2006.
DAVIES, B. L. (2007): Warfare, State and Society on the Black Sea Steppe, 1500–1700,
London and New York.
DI CARPINE/DAFFINÀ ET AL. (1989): Storia dei Mongoli, ed. Daffinà, P. – Leonardi, C.
– Lungarotti, M C. – Menestò, E. – Petech L. Spoleto.
DALL’АGLIO, F. (2008–9): The Military Alliance Between the Cumans and Bulgaria
from the Establishment of the Second Bulgarian Kingdom to the Mongol Inva-
sion. Archivum Eurasiae Medii Aevi 16, 25–54.
DIACONU, P. (1978): Les Coumans au Bas-Danube aux XIe et XIIe siècles, Bucarest.
DÖRRIE H. (1956): Drei Texte zur Geschichte der Ungarn und Mongolen. Die Mis-
sionsreisen des fr. Iulianus O.P. ins Ural-Gebeit (1234/5) und nach Ruß-land
(1237) und der Bericht des Erzbischofs Peter über die Tartaren. Nach-richten der
Akademie der Wissenschaften in Göttingen, Philologisch-Historische Klasse, 125–202.
DPI (1993): Lazarov, I. – Tyutyundzhiev, I. – Pavlov, P. Dokumenti za politicheskata
istoriya na srednovekovna Balgariya XII-XIV vek, Veliko Tarnovo.
EPHRAIM/GYUZELEV – LYUBENOV (1981): Stikhotvorna khronika, in Gyuzelev, V.
(ed.:) Samiyat Tarnovgrad shte raztrabi pobedite. Srednovekovni poeti za Balgariya.
Sofia, 57–96.
FINE, J. V. A. (1989): The Early Medieval Balkans. A Critical Survey from the Sixth to the
Late Twelfth Century, Ann Arbor.
FIVE CHINESE SOURCES 2021, The Rise of the Mongols. Five Chinese Sources, ed., transl.
Atwood, Ch. P. – Struve, L. Indianapolis–Cambridge.
GOLDEN, P. B. (1987–1991): Nomads and Their Sedentary Neighbors in Pre-
Činggisid Eurasia. Archivum Eurasiae Medii Aevi 7, 41–81.
GOLDEN, P. B. (1991): The Qipčaqs of Medieval Eurasia: An Example of Stateless
Adaptation in the Steppes, in Seaman, G. – Marks, D. (ed.:) Rulers from the
Steppe: State Formation on the Eurasian Periphery. Los Angeles, 132–157.
GOLDEN, P. B. (1992): An Introduction to the History of the Turkic Peoples. Ethnogenesis
and State-Formation in Medieval and Early Modern Eurasia and the Middle East,
Wiesbaden.
GOLDEN, P. B. (1995-1997): Cumanica IV: The Cumano-Qıpčaq Clans and Tribes.
Archivum Eurasiae Medii Aevi 9, 99–122.
GOLDEN, P. B. (1998a): Nomads and Sedentary Societies in Medieval Eurasia, Washing-
ton, D. C.
GOLDEN, P. B. (1998b): Religion among the Qipcaqs of Medieval Eurasia. Central
Asiatic Journal 42, 2, 180–237.
GOLEV, K. (2013а): Desht-i Kipchak i Vneshnii mir. Tipologiya kontaktnykh zon
mezhdu kumano-kipchakskoi obshchnostyu i vneshnego mira, in: Kumekov,
B. E. (ed.:) Sbornik materialov mezhdunarodnoi nauchnoi konferentsii „Kipchaki Ev-
razii: istoriya, yazyk i pismennye pamyatniki“ posvyashchennoi 1100 letiyu Kimeks-
kogo gosudarstva v ramkakh Dnei tyurkskoi pisʹmennosti i kulʹtury. Astana, 142–151.
GOLEV, K. (2013b): Prichini za kumanskite migratsii na Balkanite. Unpublished Doc-
toral Dissertation, Sofia University “Saint Kliment Ohridski”.
GOLEV, K. (2018a): On the Edge of “Another World”: The Balkans and Crimea as
Contact Zones Between the Cuman-Qïpchaqs and the Outside World. Études
Balkaniques 54, 1, 89–126.
GOLEV, K. (2018b): The Cuman-Qipčaks and Crimea: the Role of the Peninsula in
the Nomads’ Relations with the Outside World. Archivum Eurasiae Medii Aevi
24, 23–107.
GOLEV, K. (2021a): The Cities along the Syr Darya in 11–13 cc.: Jand and Sïghnaq
between the Cuman-Qïpchaqs and the Khwārazmshāhs Anushteginids. Acta
Orientalia Academiae Scientiorum Hungaricae 74, 1, 11–52.
GOLEV, K. (2021b): „Te ne mozhekha da voyuvat poveche poradi lyatoto“: asimetrichnata
sezonna voenna aktivnost na kumano-kipchakite na Balkanite, Istoricheski
pregled, 77/4, 5–23.
GUMILEV, L. (1992): Ot Rusi k Rossii. Ochrki etnicheskoi istorii, Moscow.
HAYTONUS/KOHLER (1906): Flos Historiarum Terre Orientis, ed. Kohler, Ch. in Recueil
des historiens des croisades. Documents Arméniens, vol. 2, Paris, 255–363.
HEI-DA SHILUE/KHRAPACHEVSKY (2009): Zolotaya Orda v istochnikakh. Vol. 3. Kitais-
kie i mogolʹskie istochniki. Sobranie svedenii, otnosyashchikhsya k istorii Zolotoi Ordy
(ulus Dzhuchi), iz kitaiskikh i mongolʹskikh istochnikov, transl. Khrapa-chevsky, R.
P. Moscow, 27–120.
HENRI DE VALENCIENNES/BUROV (2009): Istoriya na imperator Anri, transl. Burov, I.

Sofia
HENRICUS IMPERATOR/PRIMOV (1981): Pisma, in: Latinski izvori za balgarskata istoriya,
vol. 4, ed. Voynov, M., Gyuzelev, V., Lishev, S., Petrova, М., Primov, B. Sofia,
11–23.
HRUSHEVSKY, M. (1901): Khronol′ohiya podii Halyts′ko-volyns′koï lïtopysy, Zapysky
tovarystva imeni Shevchenka 41/3, 1–72.
ḤUDŪD AL-ʿĀLAM/SOTOODEH (1340/1962): Ḥudūd al-ʿĀlam: min al-Mashriq Il al-
Maghrib. Compiled in 982-3 A.D. = 372 A.H., ed. Sotoodeh, M. Tehran.
ḤUDŪD AL-ʿĀLAM/MINORSKY (1970): Ḥudūd al-ʿĀlam. ‘The Regions of the World’. A
Persian Geography, 372 A. H. – 982 A. D., transl. Minorsky, V. Cambridge, 1970.
THE HUNGARIAN ILLUMINATED CHRONICLE/WEST (1969): The Hungarian Illuminated
Chronicle. Chronica de Gestis Hungarorum, transl. West, A. Budapest.
IBN AL-ATHĪR/RICHARDS (2005): The Chronicle of Ibn al-Athīr for the Crusading Period
from al-Kāmil fi’l-ta’rīkh. Part 1. The Years 491–541/1097–1146: The Coming of the
Franks and the Muslim Response, transl. Richards, D. S. Aldershot–Burlington.
IBN AL-ATHĪR/RICHARDS (2008): The Chronicle of Ibn al Athīr for the Crusading Period
from al-Kāmil fi’l-ta’rīkh. Part 3. The Years 589–629/1193–1231: The Ayyūbids after
Saladin and the Mongol Menace, transl. Richards, D. S. Aldershot–Burlington.
IBN AL-ATHĪR/TIZENGAUZEN (1884): in Sbornik’ materialovʹ, otsnosyashchihsya kʺ isto-
rii Zolotoi ordy, vol. 1, transl. Tizengauzen, V. Saint Petersburg, 1–45.
IBN BAṬṬŪṬA/TIZENGAUZEN (1884): in Sbornikʹ materialovʹ, otsnosyashchihsya kʺ istorii
Zolotoi ordy, vol. 1, transl. Tizengauzen, V. Saint Petersburg, 278–314.
IBN BAṬṬŪṬA/GIBB (1995): The Travels of Ibn Baṭṭūṭa, A. D. 1325–1354, transl. Gibb,
H. A. R. Vol. 2. Chippenham.
IBN BĪBĪ/MOTAḤEDDĪN (1390/2011): Tārīkh-i Ibn Bībī, ed. Motaḥeddīn, Zh. Tehrān.
İBN BIBI/ÖZTÜRK (2014): El-Evâmirü’l-Alâ’iyye fi’l-Umûri’l-Alâiyye. Selçuknâme, vol. 2,
transl. Öztürk, M. Ankara.
İBNÜʾL CEVZÎ/SEVIM (2011): Mirʾâtüʾz-zaman fî târîhiʾl-âyânʾda Selçuklular, transl.
Sevim A. Ankara.
JAGCHID, S. – SYMONS, V. J. (1989): Peace, War and Trade along the Great Wall. Nomad-
ic-Chinese Interaction trough Two Millenia, Bloomington and Indianapolis.
JUVAYNĪ/QAZVĪNĪ (1912/1329): Tārīkh-i Jahān-Gushāy, ed. Qazvīnī, M. Vol. 1, Ley-
den–London.
JUVAYNĪ/QAZVĪNĪ (1916/1334): Tārīkh-i Jahān-Gushāy, ed. Qazvīnī, M. Vol. 2, Ley-
den–London.
JUVAINI/BOYLE (1958): The History of the World-Conqueror, transl. Boyle, J. A. Vol. 1-
2, Manchester.
JŪZJĀNĪ/TIZENGAUZEN (1941): Sbornik materialov otsnosyashchikhsya k istorii Zolotoi
ordy. Vol. 2, ed., transl. Tizengauzen, V. – Romaskevich, A. A. – Volin, S. L.
Moscow–Leningrad, 13–19.
JŪZJĀNĪ/HABĪBĪ (1343/1964): Tabaqāt-i Nāsirī, ed. Habībī, ʿA. Vol. 1-2, Kābul.
AL-JUZJANI/RAVERTY (1970): Tabaqat-i-Nasiri: A General History of the Muhammadan

Dynasties of Asia, including Hindustan; from A.H. 194 (800 A.D.) to A.H. 658 (A.D.
1260) and the Irruption of the Infidel Mughals into Islam, transl. Raverty, H. G. Vol.
1-2, New Delhi.
KAFESOĞLU, İ. (2000): Harezmşahlar Devleti Tarihi (485–618/1092–1221), Ankara.
KARGALOV, V. V. (1967): Vneshne-politicheskie faktory razvitiya feodalʹnoi Rusi
(Feodalʹnaya Rusʹ i kochevniki), Moscow.
KARTLIS TSKHOVREBA/METREVELI ET AL. (2013): Kartlis tskhovreba, ed. Metreveli, R.
et al. Tbilisi.
KARTLIS TSKHOVREBA/MET’REVELI – JONES (2014): A history of Georgia, ed., transl.
Met’reveli, R. – Jones, S. Tbilisi.
KHAZANOV, A. M. (1974): Sarmato-kalmitskie paralleli (k voprosu o zakonomernos-
tyakh kochevogo khozyaistva v odnotipnoi ekologicheskoi obstanovke), in:
Problemy Altaistiki i mongolovedeniya. Materialy Vsesoyuznoi konferentsii. Elista.
17-19 maya, 1972 g. Vol. 1, Elista, 213–219.
KHAZANOV, A. M. (1992): Nomads and oases in Central Asia, in Hall, J. A. – Jarvie,
I. C. (ed.:) Transition to modernity. Essays on power, wealth and belief. Cambridge,
69–89.
KHAZANOV, A. M. (1994): Nomads and the Оutside World, Cambridge.
KHODARKOVSKY, M. (2002): Russia’s Steppe Frontier. The Making of a Colonial Empire,
1500–1800, Bloomington and Indianapolis.
KHRAPACHEVSKY, R. P. (2005): Voennaya derzhava Chingiz khana, Moscow.
KHUDYAKOV, YU. S. (1997): Vooruzhenie kochevnikov Yuzhnoi Sibiri i Tsentralʹnoi Azii v
epokhu razvitogo srednevekoviya, Novosibirsk.
KINNAMOS/MEINEKE (1836): Epitome rerum ab Ioanne et Alexio Comnenis Gestarum,
ed. Meineke, A. Bonn.
KINNAMOS/BRAND (1976): The Deeds of John and Manuel Comnenus, transl. Brand,
Ch. M. New York.
KÖPRÜLÜ, М. F. (1943): Uran Kabilesi. Belleten 7, 26, 227–243.
KRADER, L. (1955): Ecology of Central Asian Pastoralism. Southwestern Journal of
Anthropology 11, 4, 301–326.
KRONIKA HALICKO-WOŁYŃSKA/DĄBROWSKI – JUSUPOVIĆ (2017): Kronika Halicko-
Wołyńska, Dąbrowski, D. –Jusupović, A (eds.), Kraków–Warszawa.
KRASTEV, К. (2011): Balgarskoto tsarstvo pri dinastiyata na Terterevtsi, Plovdiv.
LETTERS FROM THE EAST/BARVER – BATE (2010): Letters from the East. Crusaders,
Piligrims and Settlers in 12th – 13th Centuries, transl. Barver., M. – Bate, K. Ash-
gate.
MAGDALINO, P. (2002): The empire of Manuel I Komnenos, 1143–1180, Cambridge.
MARCO POLO/YULE (1903): The Book of Ser Marco Polo the Venetian concerning the
Kingdoms and Marvels of the East, vol. 1, ed., transl. Yule, H. – Cordier, H. Lon-
don.
MARCO POLO/LATHAM (1958): The Travels of Marco Polo, transl. Latham, R. Har-
mondsworth.
MASTER ROGER/BAK – RADY (2010): Anonymus and Master Roger. Gesta Hungarorum.
The Deeds of the Hungarians, Rady, M. – Veszpremy, L. Epistola in miserabile
Carmen super destructione Regni Hungarie per tartaros facta. Epistile to the Sorrow-
ful Lament upon the Destruction of the Kingdom of Hungary by the Tatars, transl.
Bak, J. M. – Rady, M. Budapest–New York.
MAURICE'S STRATEGIKON/DENNIS (1984): Maurice’s Strategikon. Handbook of Byzan-
tine Military Strategy, transl. Dennis, G. T. Philadelphia.
MEMOIRS OF THE CRUSADES/MARZIALS (1908): Memoirs of the Crusades by Villehardou-
in & de Joinville, transl. Marzials, F. London–New York.
MENG-DA BEILU/MUNKUEV (1975): Men-da bei-lu («Polnoe opisanie mongolo-tatar»),
transl. Munkuev, N. Ts. Moscow.
MEYVAERT, P. (1980): An Unknown Letter of Hulagu, Ilkhan of Persia, to King Lou-
is IX of France. Viator 11, 245–259.
MĪRKHWĀND (1339/1960): Tārīkh-i rauẓat al-Ṣafā, vol. 4, Tehran.
MLADENOV, S. (1931): Pechenezi i uzi-kumani v balgarskata istoriya, Balgarska isto-
richeska biblioteka 4, 1, 115–136.
THE MISSION/JACKSON (1990): The Mission of Friar William of Rubruck. His journey to
the court of the Great Khan Möngke 1253-1255, transl. Jackson, P. London.
MITT (1941): Materialy po istorii turkmen i Turkmenii. Vol. 1. VII–XV vv. Arabskie i
persidskie istochniki. ed., transl. Volin, S. L. – Romaskevich, A. A. – Yakubovsky,
A. Yu. Moscow–Leningrad.
THE MONGOL MISSION/DAWSON (1966): The Mongol Mission: Narratives and Letters of
the Franciscan Missionaries in Mongolia and China in the Thirteenth and Fourteenth
Centuries, ed. Dawson, Ch. New York, 1966.
MORGAN, D. (1985): The Mongols in Syria, 1260–1300, in Edbury, P. W. (ed.:) Cru-
sade and Settlement. Cardiff, 231–235.
MUTAFCHIEV, P. (1973): Mnimoto preselenie na seldzhushki turtsi v Dobrudzha
prez XIII v., in: Mutafchiev, P. Izbrani proizvedeniya, ed. Angelov, D. Vol. 2, So-
fia, 607–745.
AL-NASAVĪ/BUNIYATOV (1996): Sirat as-sultan Dzhalal ad-Din Mankburny (Zhizneopi-
sanie sultana Dzhalal ad-Dina Mankburny), ed., transl. Buniyatov, Z. M. Moscow.
NIKOLOV, A. (2008–2017): Khaiton ot Korikos i negoviya traktat „Tsvete na istoriite
na Iztochnata zemya“ (Istoriya na tatarite). Godishnik na Sofiiskiya Universitet
„Sv. Kliment Okhridski“ 102, 7–117.
AL-NUWAYRĪ/TIZENGAUZEN (1884): in Sbornikʹ materialovʹ, otsnosyashchihsya kʺ istorii
Zolotoj ordy, vol. 1, transl. Tizengauzen, V. Saint Petersburg, 128–171.
OSTROGORSKI, G. (1998): Istoriya na vizantiiskata darzhava, Sofia.
ÖZAYDIN, A. (2002): Kılıcarslan II, in TDV İslâm Ansiklopedisi, vol. 25, Ankara, 399–403.
ÖZAYDIN, A. (2009): Sencer, TDV İslâm Ansiklopedisi, vol. 36, Ankara, 507–517.
PACHYMERES/FAILLER (1984): Relationes Historicas, CFHB, vol. 24/2, ed. Failler, A.,
transl. Laurent, V. Paris.
PACHYMERES/KARPOV (1862): Istoriya o Mikhaile i Andronike Paleologakh. Trinadtsatʹ
knig. Vol. I. Tsarstvovanie Mikhaila Paleologa (1255–1282), ed., transl. Karpov.
Saint Petersburg.
PAUL, J. (2013): Sanjar and Atsız: Independence, Lordship, and Literature, in: Paul,
J. (ed.:) Nomad Aristocrats in a World of Empires. Wiesbaden, 81–129.
PAUL, J. (2015): Who Makes Use of Whom? Some Remarks on the Nomad Policy of
the Kwarāzmshāh, 1150–1200, in Frantz, K. – Holzwarth, W. (ed.:) Nomad Mili-
tary Power in Iran and Adjacent Areas in the Islamic Period. Wiesbaden, 137–153.
PAVLOV, P. (1990а): Kumanite v obshtestveno-politicheskiya zhivot na srednove-
kovna Balgariya (1186 g. – nachaloto na XIV v.). Istoricheski pregled, (7), 16–26.
PAVLOV, P. (1990b): Za rolyata na kumanite v balgarskata voenna istoriya (1186–
1241 g.). Voennoistoricheski sbornik, (6), 14–23.
PAVLOV, P. (1992): Srednovekovna Balgariya i kumanite: voennopoliticheski ot-
nosheniya (1186-1241). Trudove na Velikotarnovskiya Universitet „Sv.Sv. Kiril i
Metodii” 27/3, 8–61.
PAVLOV, P. (1996) Kam interpretatsiyata na nyakoi arkheologicheski i numizma-
tichni nakhodki ot Severnoto Chernomorie (XIII-XIV v.). Balgarite v Severnoto
Prichernomorie 5, 189–198.
PLETNEVA, S. A. (1990): Polovtsy, Moscow.
PLDR (1980): Pamyatniki literatury Drevnei Rusi: XII vek, ed. Dmitriev, L. A. – Likha-
chev, D. S. Moscow.
PSRL, vol. 1 (1928): Polnoe sobranie russkikh letopisei. Vol. 1. Lavrentʹevskaya letopisʹ,
Leningrad.
PSRL, vol. 2 (1908): Polnoe sobranie russkikh letopisei. Vol 2. Ipatʹevskaya letopis, Saint
Petersburg.
PSRL, vol. 3 (1950): Novgorodskaya pervaya letopisʹ starshego i mladshego izvodov, ed.
Nasonov, A. N. Moscow–Leningrad.
RASHĪD AL-DĪN/SMIRNOVA (1952): Sbornik letopisei. Vol. 1, pt. 2, transl. Smirnova, O.
I. Moscow–Leningrad.
RASHĪD AL-DĪN/VERKHOVSKY (1960): Sbornik letopisei. Vol. 2, transl. Verkhovsky Yu.
P. Moscow–Leningrad.
RASHĪD AL-DĪN/KARĪMĪ (1338/1959): Jāmiʾ al-tawārīkh, vol. 1–2, ed. Karīmī, B. Teh-
rān.
RASHĪD AL-DĪN/RAWSHAN – MŪSAVĪ (1373/1994): Jāmiʾ al-tawārīkh, vol. 1–4, ed.
Rawshan, M. – Mūsavī, M. Tehrān.
RASHĪD AL-DĪN/BOYLE (1971): The Successors of Genghis Khan, transl. Boyle. J. A.
New York–London.
RASHĪD AL-DĪN/RAWSHAN (2010): Jāmi’ al-tawārīkh: tārikh-i sаlāṭīn-i Khwārazm, ed.
Rawshan, M. Tehrān.
RASHĪD AL-DĪN/THACKSTON (1998–9): Rashiduddin Fazlullah’s Compendium of Chroni-

cles. A History of the Mongols, Parts 1–3, transl. Thackston, W. M. Cambridge, MS.
RASOVSKY, D. A. (1937): Polovtsy, III. Seminarium Condacovianum 9, 71–85.
RASOVSKY, D. A. (1938): Polovtsy, III. Annales del’ Institut Kondakov (Seminarium
Condacovianum) 10, 155–178.
RASOVSKY, D. A. (1939): Rolʹ polovtsev v voinakh Asenei s Vizantiiskoi i Latinskoi
imperii v 1186–1207 g. Spisanie na Balgarskata akademiya na naukite 58, 203–211.
RATCHNEVSKY, P. (1991) Genghis Khan: His Life and Legacy, Oxford.
ROBERT DE CLARI/MARKOV (2007): Zavoyuvaneto na Konstantinopol, 1204 г., transl.
Markov, N. A. Sofia.
ROGERIUS/JUHÁSZ (1999): Rogerii Carmen Miserabile, ed. Juhász, L. in Szentpétery,
E. (ed.:) Scriptores rerum Hungaricarum, vol. 2, Budapest, 543–588.
DE RUBRUC/VAN DEN WYNGAERT (1929): Sinica Franciscana, vol. I, Itinera et Rela-
tiones Fratrum Minorum saeculi XIII et XIV, ed. van den Wyngaert, A. Quarac-
chi–Florence, 145–332.
DI RUBRUK/CHIESA (2011): Viaggio in Mongolia, ed. Chiesa, P. Fondazione Lorenzo
Valla.
THE SECRET HISTORY/DE RACHEWILTZ (2006): The Secret History of the Mongols. A
Mongolian Epic Chronicle of the Thirteenth Century, vol. 1–2, transl. de Rachew-
iltz, I. Leiden.
SINOR, D. (1972): Horse and Pasture in Inner Asian History. Oriens Exstremus 19,
171–184; reprinted in Sinor, D. Inner Asia and its Contacts with Medieval Europe,
London, 1977, article II.
SĪRAT-I DJALĀLUDDĪN/MĪNUVĪ (1986): Sīrat-i Djalāluddīn. Persian Translation, made in
the 7th Cent. H. from the Arabic Text of Shihabuddin Muhammad Nasawi Zeydari
Khurandizī, ed. Mīnuvī, М. Tehran.
SKYLITZES – KEDRENOS/TAPKOVA-ZAIMOVA (1965): Kratka istoriya, in: Gratski izvori
za balgarskata istoriya, vol. 6, ed. Duychev, I. – Yonchev, L. – Tivchev, P. – Tap-
kova-Zaimova, V. – Tsankova-Petkova, G. Sofia, 198–340.
SKOULATOS, B. (1980): Les personnages byzantins de l’Alexiade : Analyse prosopogra-
phique et synthèse, Louvain-la-Neuve–Louvain.
SMITH, J. M. (1984): ʿAyn Jalut: Mamluk success or Mongol failure? Harvard Journal
of Asiatic Studies 44/2, 307–345.
SPINEI, V. (2003): The Great Migrations in the East and South East of Europe from the
Ninth to the Thirteenth Century, Cluj-Napoca.
Tenth to the Mid-Thirteenth Century, Leiden–Boston.
STEPHENSON, P. (2000): Byzantium’s Balkan Frontier. A Political Study of the Northern
Balkans, 900-1204, Cambridge.
DAS STRATEGIKON/DENNIS (1981): Das Strategikon des Maurikios, ed. Dennis, G.,
transl. Gamillscheg, E. CFHB, vol. 17, Vienna.
THEODOROS SKOUTARIOTES/VOYNOV (1972): Obzorna khronika, in: Gratski izvori za

balgarskata istoriya, vol. 8, ed. Voynov, M. Sofia, 214–303.
THOMAS ARCHIDIACONUS/PERIĆ ET AL. (2006): Historia Salonitanorum atque Spalatino-
rum pontificum. History of the Bishops of Salona and Split, Latin text by O. Perić.
Edited, translated and annotated by D. Karbić, M. Matijević Sokol and J.
Sweeney, Budapest–New York.
TOR, D. G. (2000/2010): Sanjar, Aḥmad b. Malekšāh, in: Encyclopaedia Iranica,
(https://www.iranicaonline.org/articles/sanjar)
TOTEV, T. (1985): Za kumani v edin nadpis ot Preslav, in: Kulturata na srednovekov-
niya Tarnov. Sofia, 158–169.
TOTEV, T. (2007): Koga sa pomagali kumanite na balgarite pri Asenevtsi? In: Nedel-
chev, N. (ed.:) 800 godini ot bitkata pri Odrin. Shumen, 37–40.
TAPKOVA-ZAIMOVA, V. (1976): Dolni Dunav – granichna zona na vizantiyskiya zapad,
Sofia.
TURAN, O. (1993): Selçuklular Tarihi ve Türk İslâm Medeniyeti, İstanbul.
VÁSÁRY, I. (2005): Cumans and Tatars. Oriental Military in the Pre-Ottoman Balkans,
1185-1365, Cambridge.
VAṢṢĀF (1269/1852-1853): Tārīkh-i Vaṣṣāf (Tajziyat al-amṣār va-tazjiyat al-aʿṣār), Bombay.
VAṢṢĀF/ĀYATĪ (1346/1967): Taḥrīr-i Tārīkh-i Vaṣṣāf, ed. Āyatī, ʿA. M. Tehran.
VAṬVĀṬ/TŪYSIRKĀNĪ (1338/1960): Nāmahā-yi Rashīd al-Dīn Vaṭvāṭ, ed. Tūysirkānī,
Q. Tehran.
JOHANNES TSETSES/YONCHEV (1980): Otkasi, in: Gratski izvori za balgarskata istoriya,
vol. 10, ed. Voynov, M., Tapkova-Zaimova, V., Yonchev, L. Sofia, 104–108.
YUAN SHI/KHRAPACHEVSKY (2009): Zolotaya Orda v istochnikakh. Vol. 3. Kitaiskie i
mogolʹskie istochniki. Sobranie svedenii, otnosyashchikhsya k istorii Zolotoi Ordy
(ulus Dzhuchi), iz kitaiskikh i mongolʹskikh istochnikov, transl. Khrapachevsky, R.
P. Moscow, 121–291.
ZLATARSKI, V. (1994): Istoriya na balgarskata darzhava prez srednite vekove, vol. 2, Sofia.
ZLATARSKI, V. (2007): Istoriya na balgarskata darzhava prez srednite vekove, vol. 3, Sofia.
Nomadic State in the Sedentary World:
The Military Mode of Government of
Conquered Lands in the Mongol Empire
SIMON BERGER
Abstract
The Mongols built their state around the essential institution of the army, which
was the main instrument of control over the population. The Mongol Empire was
thus characterized by a military-administrative system in which the administration
was carried out by military commanders. This paper aims to show, with a special
emphasis on the case of pre-Ilkhanid Iran and the Caucasus, that the Mongols first
transposed this model to the conquered sedentary territories instead of using na-
tive administrators. In particular, it attempts to highlight the military character of
the officials called darughachin, or basqaqs, sometimes hastily defined as civilian
governors.
Keywords: Mongol Empire; Mongol army; Mongol administration; darughachi;
basqaq
The dominant interpretative model for the relations between Eurasian nomads and
their sedentary neighbours is that of a centre-periphery approach involving eco-
nomic and political dependency of the former towards the latter, and where turbu-
lent and anarchic nomadic tribes and clans subsist on the margins of central, bu-
reaucratic, and territorialized states of sedentary societies. To face the establish-
ment of powerful sedentary states, nomads would be forced to unite into large
confederations, militarily powerful but internally decentralized, maintaining the
tribal structure intact and relying on the consensus of tribal leaders towards the
paramount leader, and thus not being states to the full extent of the term. 1
 EHESS Paris, CETOBaC. The article is a revised version of a paper delivered at the 8th In-
ternational Conference on Medieval History of the Eurasian Steppe Nomads and their Neigh-
bours in the Middle Ages, held in Sofia, November 20–3, 2019. I would like to give special
thanks to Martyna Kosik who drew the map that accompanies this article.
1 See especially BARFIELD 1989.
184 SIMON BERGER
According to some, such would have initially been the case of the Mongol Em-
pire. The Mongols, once they had made their conquests, would not have really
known how to administer the numerous populations and immense sedentary terri-
tories that fell under their dominion. They would thus have been forced to rely on
a sedentary administrative staff, especially Chinese or Persian. 2 This would corre-
spond to the famous adage repeated by Yelü Chucai to Ögödei: “The empire has
been conquered on horseback, but it cannot be ruled on horseback.” 3
This kind of interpretation is, however, grounded on a conception of the state
based on typologies that have the modern (and mainly European) State as their
model. Many recent studies have highlighted alternative approaches to the state
that allow the expression of a greater variety of forms of political complexity and
integration. In particular, the State should no longer be conceived as a centralized,
territorialized, and bureaucratic extra-social super-structure, but rather as a system
of codified social relations between a ruling stratum and subordinate populations. 4
In my view, the framework for these relations and for the large-scale integration of
subject populations is precisely the army. The army would therefore not only be an
external manifestation of a rather loose confederal political form, but an organiza-
tional tool at the heart of what could be called the nomadic state. The case of the
Mongol Empire is particularly telling in this regard. If a form of nomadic state
existed, then it may be necessary to reconsider the idea that the Mongols needed
the help of sedentary advisors recruited from among the defeated sedentary peo-
ples to impose a coherent administrative structure on their empire.
The Yuanshi (the Mongol Yuan dynasty’s official annals compiled in 1370–1
during the Ming dynasty) asserts quite bluntly and with much contempt that dur-
ing the time of Chinggis Khan the administration consisted “only of commanders
of ten thousand to command the army, and judges to enforce laws and sentences”. 5
It was only after the conquest of some Chinese territories that Chinggis Khan
would have set up a proper administration, recruiting local Chinese and Jurchen
officials. One should not take this tendentious account too literally and infer from
this absence of bureaucratic apparatus, which raised the eyebrows of the Confu-
cians, an absence of state organisation at all. Quite the contrary, the mention of the
army commanders, the noyad (sing. noyan), as the sole agents of the early Mongol
administration—along with the judges, called jarghuchin—points, in fact, to a mili-
2 See for instance RACHEWILTZ 1966, 135–6; BUELL 1979, 131–2; MORGAN 1982, 134–6; MORGAN
2007, 94–6; ENDICOTT-WEST 1989, 17; KRADIN – SKRYNNIKOVA 2006, 115; AIGLE 2008, 67,
among others. David Morgan revised his opinion in MORGAN 1996.
3 以馬上取天下，不可以馬上治 ; SONG LIAN 1976, 157.3688.
4 For a broad theoretical approach to the issue, see CHANDHOKE 1995, 49; JESSOP 2016, 53–90.
For its application to nomadic contexts, see SNEATH 2007, 185–9, and, in a way, STANDEN 2018.
5 惟以萬戶統軍旅，以斷事官治政刑; SONG LIAN 1976, 85.2119.
NOMADIC STATE IN THE SEDENTARY WORLD… 185
tary mode of government of populations, indeed alien to the Chinese and other
sedentary state traditions.6
It is obvious from the many available sources that the famous decimally orga-
nized Mongol army was a much larger structure than just all the fighting men and
that it included the entire population. Chinese sources indicate that all adult males
from the age of fifteen were conscripted. 7 Juvaynī (d. 1283), who devotes a long
passage to the military organization of the Mongols in his Tārīkh-i jahāngushā, strik-
ingly equates the army and the civilian population:
It is an army under the guise of a people, being liable in all manner of con-
tributions, and rendering without complaint whatever is enjoined to them,
whether qupchur, occasional taxes, the maintenance of travellers, or the up-
keep of the yam with the provision of mounts and food therefor. It is also a
people under the appearance of an army, all of them, great or small, noble,
and base, in time of battle becoming swordsmen, archers and lancers and
advancing in whatever manner the occasion requires. 8
The Yuanshi expresses the very same thing in a much more lapidary way when
it says that “mounted, they were ready for fighting; dismounted, they settled to-
gether to pasture animals”.9 The purpose, however, is not to point out that the
Mongol army was a people in arms, in accordance with the stereotype of the no-
mad as a natural warrior.10 The idea expressed here is that there was no distinction
among the Mongols between the armed force and the people subject to authority
and taxes. It refers to the incorporation of the entire population into a military-
administrative system: not everyone was in arms, but everybody was conscripted.
Not only men of fighting age, but also their families, the a’uruq, or encampments
for women, children, and dependants who followed the army. The entire popula-
tion was divided into the decimal units that composed the army, as Juvaynī writes:
“They have divided all the people into companies of ten, appointing one of the ten
to be the commander of the nine others.”11
6 The Qubilai era Zhiyuan yiyu thus translates noyan as guanren 官人, “official”: ZHIYUAN YIYU
2006, 12; tr. LIGETI – KARA 1990, 263; KARA 1990, 313. We will see that in fact even the judges
were not distinct from the military officers, but that on the contrary the jarghuchin were
generally also army commanders.
7 PENG DAYA 1962, 501 [19r]; tr. OLBRICHT – PINKS et al. 1980, 172. SONG LIAN 1976, 98.2508.
8 JUVAYNĪ 1912–1937, vol. 1, 22; tr. BOYLE 1958, vol. 1, 30, slightly modified. Persian text:
‫لشكرى اندر شيوۀ رعيّت كه احتمال صنوف مؤن كنند و بر اداى آنچ بريشان حكم كنند از قوبجور و عوارضات و اخراجات‬
‫ى لشكر كه وقت كار از خرد تا بزرگ شريف تا‬ّ ‫رعيّتى اندر ز‬،‫صادر و وارد و ترتيب يام و اوالغ و علوفات ضجرت نكنند‬
‫وضيع همه شمشيرزن و تيرانداز و نيزهگذار باشند بهر نوع كه وقت اقتضاى آن كند‬
9 上馬則備戰鬭，下馬則屯聚牧養; SONG LIAN 1976, 98.2508.
10 SINOR 1981.
‫و ّنواب آن را معزول تمامت خاليق را دهده كرده و از هر ده يک نفس را امير نه ديگر كرده‬
186 SIMON BERGER
Using the expression “all the people” (tamāmat-i khalāʿīq), he makes it clear that
it was not only men that were concerned. In the same way, the Secret History clear-
ly shows how the symbolic birth of the Mongol State took place during the quriltai
of 1206, when Chinggis Khan “sets the Mongol people in order” by creating the
ninety-five mingghad (“thousands”) of his army and assigning them areas of pas-
tureland.12
The army was consubstantial with the State. As already mentioned, the state,
especially in pre-modern times, was first and foremost the relation of power from a
ruling class over individuals and social groups rather than a territorialized struc-
ture. This is what the Mongol notion of ulus, “state”, or more precisely “people
subject to the State”, refers directly to. 13 The Yeke Mongghol Cherig, the “Great Mon-
gol Army”, is in a way the practical application of the theoretical concept of Yeke
Mongghol Ulus, the “Great Mongol State”. The ulus is the state as the authority ex-
ercised over subjects; these subjects are the cherig, the army.
Juvaynī thus describes the apportionment of appanages among members of the
Chinggisid family as a distribution of military units:
[Chinggis Khan] divided the houses [qabā’īl]14 and peoples [aqvām] of the
Mongols and the Naimans and all the armies between the aforesaid sons;
and to each of his younger sons and to his brothers and kinsmen he allotted
their share of the army.15
All members of the imperial family had their shares, not of lands, but of the
people and army altogether. The very same idea is expressed in the Jāmiʿ al-tavārīkh
of Rashīd al-Dīn (d. 1318) when he describes the repartition of the various hazāras
(thousands) after the death of Chinggis Khan. 16
The decimal units of the army, and especially the mingghad (sing. mingghan),
thus constituted administrative divisions. They formed the framework within
12 SECRET HISTORY 1972, § 202, 114; tr. RACHEWILTZ 2004, vol. 1, 133.
13 MUNKH-ERDENE 2011, 218, 221; ATWOOD 2015, 25ff.
14 As I have argued in a paper entitled “The Mongol military system of government and its
imposition in Iran as seen through the work of ʿAlāʾ ad-Dīn ʿAṭā Malik Juvaynī”, given on
September 10, 2019, at the 9th European Conference of Iranian Studies (Freie Universität, Berlin),
a thorough examination of the use of the word qabīla (pl. qabā’īl) by Juvaynī shows that he
does not mean it in the sense of “tribe”, but in the much narrower meaning of a family, a
household, and in particular of a noble house. See also remarks in SNEATH 2013, 178–80.
Whether Juvaynī means in this instance all the nuclear families of the Mongol and Naiman
peoples, as a kind of metonymic synonym of the following word (aqvām), or more specifical-
ly the noble lineages of the Mongols and the Naimans (to which aqvām were attached?) is
ambiguous. Therefore, I translated qabāʿīl here as the neutral word “houses”.
‫قبايل و اقوام مغول و نايمان و تمامت لشکرها بر پسران مذکور بخش کرد و ديگر پسران خردتر و برادران و خويشان هركس‬
‫را از لشكرها نصيب تعيين كرد‬
16 RASHĪD AD-DĪN 1395/2016, vol. 1, 530–51; tr. THACKSTON 2012, 207–13.
which Mongol power was exercised, whether it was the mobilization of men for
war or the collection of taxes or the corvée labour, as illustrated, for example, by
the testimony of Xu Ting recorded in the Heida shilue (which collects the reports of
the two Song ambassadors, Peng Daya and Xu Ting, who travelled to the Mongol
court in 1233 and 1235–6 respectively).17
In addition, both Juvaynī and the Song ambassador Zhao Gong (who visited the
Mongol army’s headquarters in Northern China in 1221) write that the decimal
system, because of its rationality and efficiency, made it possible to do without
written documents and civil servants:
The reviewing and mustering of the army has been so arranged that they
have abolished the registry of inspection [daftar-i ʿarẓ] and dismissed the of-
ficials and clerks.18
To mobilize several hundred thousand soldiers, they hardly need any regis-
ters. March orders are transmitted [orally] from the commander in chief to
the commanders of thousands, hundreds, and tens. 19
This is probably an exaggeration since there were most likely idiosyncratic
Mongol equivalents of such registers.20 Thus, for example, the Secret History shows
how Chinggis Khan gave to the Yeke Jarghuchi (“Great Judge”) Shigi Qutuqu the
task to write “in a blue register [köke debter] all decisions about distribution and
about judicial matters of the entire population”.21 But what emerges from this de-
piction is that the Mongol system made it possible to do without civil servants and
registers, that is, without the traditional administration of the sedentary state.
The Mongols had a rational and coherent state organization, the backbone of
which was the army. It existed even before the conquests, and in fact predates the
empire itself.22 It can therefore be assumed that it was first and foremost their own
system of government that the Mongols tried to impose on the conquered lands. A
most striking example of that is the infamous way the Mongols conscripted local
17 XU TING 1962, 489–90 [13rv]; tr. OLBRICHT – PINKS et al. 1980, 142. On decimal military
sections as social and tax-paying units, see most notably ATWOOD 2012, 25ff.
18 JUVAYNĪ 1912–1937, vol. 1, 22; tr. BOYLE 1958, vol. 1, 31. Persian text:
‫و عرضگه و شمار لشكر را وضعى ساختهاند كه دفترعرض را بدان منسوخ كردهاند و اصحاب و ن ّواب آن را معزول‬
19 起兵數十萬，略無文書，自元帥至千戶、百戶、牌子頭，傳令而行; ZHAO GONG 1962, 445 [8r]; tr. OL-
BRICHT – PINKS et al. 1980, 53.

20 Professor Christopher Atwood addressed this point in his lecture “Population Mobilization
in Early Mongol Conquests”, given at the EHESS Paris on May 17, 2019, as part of Professor
Étienne de la Vaissière’s seminar: Migrations et démographie en Asie centrale (IVe-XIe siècle).
21 SECRET HISTORY 1972, § 203, 116; tr. RACHEWILTZ 2004, vol. 1, 135.
22 I intend to demonstrate this in a work currently in progress.
188 SIMON BERGER
vanquished populations into levies, which were decimally organised just as the
regular nomadic army.23
Another illuminating case to illustrate this point is that of pre-Ilkhanid Mongol
Iran and Caucasus during the 1230s, which allows us to reconsider the establish-
ment of a so-called sedentary and civilian administration after the conquest of
these regions.
It was under Ögödei (r. 1229–41) that a large part of Iran came under the effec-
tive control of the Mongols, when at the end of 1230 he sent the general Chorma-
qan24 at the head of four tümed (sing. tümen, i.e., a unit of ten thousand men).25 In
his wake, a first Mongol administration was established in Northern Iran, from
Khurāsān to Āzarbāyjān and the Caucasus through Rayy and Hamadān. The bulk
of this huge nomadic army, which was established as a tamma (frontier garrison)
and included women, children, and herds, had its winter quarters in the large and
fertile Mughān plain, from which it controlled the whole region. 26 Juvaynī gives a
retrospective depiction of the essentially military nature of this Mongol administra-
tion in Iran, precisely when it tended to be replaced by a civilian administration
including Persian officials under the action initiative of the Uyghur governor
Körgüz after 1239:
When they arrived in these areas, they had many conflicts with the com-
manders of Chormaqan, until they freed these provinces from their control
and set taxes. For every province had a noyan, and every city a commander
[amīrī], and they were content to give little to the Dīvān and used to seize the
remainder for themselves. It was all taken from them and [considerable]
sums of money were drawn upon them. 27
23 See ZHAO GONG 1962, 445 [8r]; tr. OLBRICHT – PINKS et al. 1980, 53. JUVAYNĪ 1912–1937, vol.
1, 70–1; tr. BOYLE 1958, vol. 1, 92. IBN AL-ATHĪR 1965–1967, vol. 12, 367; tr. RICHARDS 2006-
2008, vol. 3, 209. KIRAKOS GANJAKECI 1986, 209–10. During the same lecture quoted above,
Christopher Atwood showed that these levies included not only able-bodied men that
served as soldiers, but also all their families, including women, children, and the elderly,
who followed them and provided a workforce for corvées.
24 Or Chormaghun according to Persian sources. His original name was Chorman though,
Chormaqan being a diminutive form of the latter, as stated by GRIGOR AKNERCI 1949, 300–1.
25 Some sources, like Juvaynī, attribute to him only three tümed, but others, like Rashīd al-
Dīn, indicate that he commanded four, and I think this is indeed the correct number. See
discussion in BERGER 2021.
26 The anonymous Georgian Hundred Years Chronicle, included in the compendium entitled
K‘art‘lis C‘xovreba, states that the four tamach noins (Mong. tammachi noyan), that is, the com-
manders of each of the four tümed of Chormaqan’s tamma, ruled from Khorāsān to Anatolia:
K‘ART‘LIS C‘XOVREBA 2014, 345. On the tamma institution, and particularly on Chormaqan’s
army, see MAY 2012; BERGER 2021.
27 JUVAYNĪ 1912–1937, vol. 2, 237–8; tr. BOYLE 1958, vol. 2, 501, slightly modified. Persian text:
Juvaynī describes this first military administration as dysfunctional, but it

should not be forgotten that he was himself a representative of the Iranian admin-
istrative elite, and that his own father, Bahāʾ al-Dīn Muḥammad, was among
Körgüz’s envoys.28 A real administrative network appears nonetheless in the back-
ground of this depiction, since every province had a noyan, and every city a com-
mander. This is consistent with what the Armenian author Grigor Aknerci (d. ca.
1335) describes for the Caucasus, saying that the Mongols “divided the countries
among the one hundred and ten chieftains” under the command of Chormaqan.29
Other Armenian and Georgian sources have similar accounts of apportionments of
conquered lands into administrative districts among military commanders. 30
Moreover, a discourse that the same Grigor Aknerci attributes to Chinggis Khan—
conflating him with his son Ögödei—clearly describes the administrative and eco-
nomic tasks that were assigned to the tamma of Chormaqan:
It is the will of God that we take the earth and maintain order, and impose the
(y)asax [jasaq, i.e., the law], that they abide by our command and give us tzγu
[tuzghu, i.e, food offering], mal [land tax], t‘aγar [taghar, i.e., food requisition
for the army] and γp‘č‘ur [qubchur, i.e., tax on livestock or crops].31
In a remarkably similar way, the Secret History shows that tax collection in Iran
was indeed part of the responsibilities of Chormaqan’s army:
[Ögödei] ordered as follows: “Chormaqan Qorchi shall reside at that very
place [as a commander of] the tamma troops. Every year he shall make deliv-
er yellow gold, naq-fabrics, brocades, and damasks with gilded [thread],
small and big pearls, fine Western horses with long necks and tall legs, dark
brown [Bactrian] camels and one-humped [Arabian] dromedaries, pack-
mules and [riding] mules, and he shall send them [to Us].” 32
A further example is that of Molar Noyan, one of the Mongol commanders un-
der Chormaqan, who promised to build villages and to plant fields for those who
‫ايشان چون بدان ممالک رسيدند با امراى جورماغون بسيار مخاصمتها كردند تا وقتى كه واليات را از دست ايشان مستخلص‬
‫صۀ ديوان قناعت كرده بودند و‬
ّ ‫كردند و مالها قرار نهاد چه هر واليتى نوينى داشت و هر شهرى را اميرى و به اندک چيزى ح‬
‫تصرف می نمودند تمامت ازيشان بازگرفتند و مبالغ بريشان متوجّه گردانيد‬
ّ ‫باقى بجهت خويش‬
28 JUVAYNĪ 1912–1937, vol. 2, 237–8; tr. BOYLE 1958, vol. 2, 501.
29 GRIGOR AKNERCI 1949, 302–3. See BAYARSAIKHAN 2016, 218–9.
30 KIRAKOS GANJAKECI 1986, 229; VARDAN AREVELCI 1989, 214; K‘ART‘LIS C‘XOVREBA 2014, 339.
31 GRIGOR AKNERCI 1949, 300–1, my emphasis. The fact that this speech is ascribed to Ching-
gis Khan might, however, be a hint that the dispatch of Chormaqan’s army to the West was
already decided at the end of his reign. The Shengwu Qinzheng lu shows that the decision to
send Chormaqan (Shuoliman 搠力蠻, Chorman) to the West was discussed between Ögödei
and Tolui as early as 1228, that is, before Ögödei’s enthronement: SHENGWU QINZHENG LU
1962, 201 [98r].
32 SECRET HISTORY 1972, § 274, 165; tr. RACHEWILTZ 2004, vol. 1, 205. On Mongol taxation, see
ALLSEN 1987, 144ff.

190 SIMON BERGER
surrendered.33 All of this points to the fact that the tammachi troops based in Iran
and the Caucasus were not a mere army of conquest but were also supposed to
administratively rule and economically exploit the conquered lands. 34
Juvaynī depicts how Chormaqan’s officers were drafted to represent the vari-
ous Chinggisid princes, in accordance with the system of appanages that prevailed
in the Mongol Empire:
And when the ruler of the world Qā’ān [Ögödei] appointed Chormaqan to
the Fourth Climate and issued a yāsā stipulating that leaders and basqaqs on
all sides come in person with levies and that they assist Chormaqan, Chin
Temür came from Khwārazm via Shahristāna, and placed as companions [of
Chormaqan] other commanders representing the princes. And Chormaqan
also, in the same way, set up commanders for each sovereign and prince
alongside Chin Temür: Kül Bolad for Qā’ān, Nosal for Batu, Qïzïl Buqa rep-
resenting Chaghadai, and [Yeke] for Beki [Sorqotani]. 35
One of them, Chin Temür, was Chormaqan’s lieutenant in Khurāsān when the
latter moved Westward in 1230. Either a Qara Khitai or an Önggüd, he was first
appointed by Jochi in the city of Ürgench as a basqaq—which is the Turkic equiva-
lent of the Mongolian darughachi, “overseer”, often translated in Persian as
shaḥna—before he became “basqaq of Iran”.36 This figure is often portrayed as the
first Mongol civilian administrator in Iran as opposed to the soldier Chormaqan,
and his rise described as part of the natural development that occurred during
Ögödei’s reign, in which the early Mongol military and “tribal” administration had
33 KIRAKOS GANJAKECI 1986, 208.

34 Grigor Aknerci even goes on to say that Chormaqan was a “sage” leader, although his
depiction of other Mongol commanders constantly extorting the Georgian princes is in line
with that of Juvaynī: GRIGOR AKNERCI 1949, 319, 321. Kirakos has a similar statement about
Chormaqan, saying he was “a judicious and just man” (according to Bedrosian’s transla-
tion). Dashdondog Bayarsaikhan may be right to interpret this as Chormaqan being a judge,
or jarghuchi, which would again reveal the administrative role of the military commanders:
KIRAKOS GANJAKECI 1986, 237; BAYARSAIKHAN 2016, 220.
35 JUVAYNĪ 1912–1937, vol. 2, 218–9; tr. BOYLE 1958, vol. 2, 482–3, slightly modified. Persian
text:
‫و چون پادشاه جهان قاآن جورماغون را بإقليم رابع نامزد گردانيد و ياسا رسانيد كه سروران و باسقاقان هر طرفى بنفس خويش‬
‫بحشر روند و معاون جورماغون باشند از خوارزم ج نتمور بر راه شهرستانه روان شد و از جوانب پادشاهزادگان امراى ديگر‬
‫در صحبت او بگذاشت و جورماغون نيز هم بر آن موجب از قبل هر پادشاه و پادشاهزاده اميرى را با جنتمور نصب كرد و‬
‫كلبالت از قبل قاآن و نوسال از قبل باتو و قزل بوقا از جانب جغتاى و ىىكه از طرف بکی سرقوقيتى‬
36 JUVAYNĪ 1912–1937, vol. 2, 218ff.; tr. BOYLE 1958, vol. 2, 482ff. RASHĪD AD-DĪN 1395/2016,
vol. 1, 123, 591–5; tr. THACKSTON 2012, 50, 229–30. For darughachi and basqaq, see DOERFER
1963–1975, vol. 1, 319–23, vol. 2, 241–3; ALLSEN 1994, 373–4; ATWOOD 2004, 134; JACKSON
2017, 107ff.; MAY 2018, 88ff.
to give way to a more sophisticated civilian one. 37 While such an interpretation is

based on some reality, I believe it ought to be tempered.
Chormaqan himself had appointed basqaqs in the wake of his campaign in Iran,
among them Chin Temür, whom he dispatched to Khurāsān to deal with the threat
caused by two officers of Jalāl al-Dīn Khwārazmshāh, Qaracha and Yaghan
Sonqur.38 These basqaqs were therefore, at least initially, part of the military admin-
istration of the region. Unaware that Chin Temür had been successful in quelling
the rebellions which agitated Khurāsān, Ögödei had resolved to put an end to it by
sending a new military governor, the tammachi noyan Dayir Ba’adur,39 who was
charged with massacring the entire population of the province. 40 However, Chin
Temür refused to step down and for the first time, in 1232–3, sent submissive local
potentates to the court of Ögödei: the malik of Ṣuʿlūk, Bahāʾ al-Dīn, and the iṣfahbad
of Kabūd Jāma, Nuṣrat al-Dīn. Delighted, the qa’an then confirmed Chin Temür as
governor of Khurāsān and Māzandarān—and probably conferred to him the supe-
rior dignity of yeke darughachi—and officially endorsed the position of the two in-
digenous potentates. Most notably, Juvaynī writes that “the malik Bahāʾ al-Dīn was
established as malik of Khurāsān, [that is] Isfarāyin, Juvayn, Jājarm, Jūrbad and
Arghiyān, for at that time this was Khurāsān”.41
One could see this passage as the devolution of the province of Khurāsān and
Māzandarān to an indigenous administration, supervised by Chin Temür at the
head of the dīvān of Iranian civil administrators he had composed and which in-
cluded Bahāʾ al-Dīn Muḥammad, among others.42 The strange definition of
Khurāsān given by Juvaynī is thus explained by Boyle as the result of the Mongol
devastation of the province, so that only the places mentioned here were not deso-
late at the time.43 But he thereby interprets it as still referring to the whole of histor-
ical Khurāsān, over the entirety of which Bahāʾ al-Dīn’s governorate would have
extended. This seems highly unlikely. There were probably other small towns in
Khurāsān spared by the conquest, but Juvaynī does not mention them here, and
37 BOYLE 1968, 336; BUELL 1979, 131–3, 141ff.; ALLSEN 1994, 374; JACKSON 1991, 567; JACKSON
2017, 107–10, 120–1; MAY 2018, 81–4, 104; WINK 2001, 295, among others.
38 JUVAYNĪ 1912–1937, vol. 2, 219; tr. BOYLE 1958, vol. 2, 483. RASHĪD AD-DĪN 1395/2016, vol. 1,
591–2; tr. THACKSTON 2012, 229. SAYFĪ HARAVĪ 1383/2004, 128. On these events, see GOLEV
2018.
39 Cf. AUBIN 1969, 70, 75.
40 JUVAYNĪ 1912–1937, vol. 2, 221–2; tr. BOYLE 1958, vol. 2, 485–6. SAYFĪ HARAVĪ 1383/2004,
128–9.
41 JUVAYNĪ 1912–1937, vol. 2, 222–3; tr. BOYLE 1958, vol. 2, 486–7, slightly modified; my em-
phasis. Persian text:

‫مقرر فرمود و در آن وقت خراسان آن بود‬
ّ ‫ملكى خراسان و اسفراين و جوين و جاجرم و جوربد و ارغيان بر ملک بهاء الدّين‬
Rashīd al-Dīn adds Bayhaq to the list, but this seems to be a later interpolation not shared by
all manuscripts: RASHĪD AD-DĪN 1395/2016, vol. 1, 593, vol. 3, 1593; tr. THACKSTON 2012, 229.
42 JUVAYNĪ 1912–1937, vol. 2, 218–9; tr. BOYLE 1958, vol. 2, 483.
43 BOYLE 1958, vol. 2, 487, n. 19.
192 SIMON BERGER
refers only to places situated in a specific circle located in the northwest corner of
Khurāsān.44 It is also wrong to affirm that this specific region had been spared,
since in fact Isfarāyin had been massacred by Sübe’edei’s troops in 1221. 45
Interestingly, in his Nuzhat al-Qūlūb Mustowfī Qazvīnī (d. ca. 1344) writes about
Jājarm that at a distance of a one- or two-day’s journey the city is surrounded by
poisonous grass that prevents any army from passing through the area. 46 Consider-
ing that a full day of caravan travel represents up to a little less than forty kilome-
tres,47 it can safely be assumed that this dangerous vegetation extended to a radius
of at least fifty kilometres, encompassing all the western part of the Khurāsān de-
fined by Juvaynī. No other source from the same time refers to these poisoned
pastures, but they still seem to be known in the 1960s. 48 European travellers of the
19th century do not mention them, but Napier describes the salinity of the soils, the
briny saltness of the few springs, and the scanty vegetation both north and west of
Jājarm, while Houtun-Schindler explains that the place is so infested by sheep ticks
(ixodes ricinus) that it is impossible to herd sheep there.49 Although Mustowfī
Qazvīnī might be exaggerating and even if this poisonous grass does not seem to
have prevented circulations of nomads, 50 all in all the area was certainly not the
best place to garrison nomadic troops with their families and herds. While on the
contrary, immediately northeast of it lay the huge and excellent pastures of
Rādkān:
When [Sübe’edei] came to Rādkān, the greenness of the meadows and the
copiousness of the springs so pleased him that he did that people no harm
and left a shaḥna there.51
Round Ṭūs lies a meadow known as the Rāykān [Rādkān] Meadow, twelve
farsakhs in length by five across, and it is among the most celebrated in the
world.52
44 On the cities of Isfarāyin, Juvayn, Jājarm, and Jūrbad, and the district of Arghiyān, see
AUBIN 2018 [1971].
45 JUVAYNĪ 1912–1937, vol. 1, 115; tr. BOYLE 1958, vol. 1, 146. In his parallel passage, Rashīd al-
Dīn omits any reference to this massacre (RASHĪD AD-DĪN 1395/2016, vol. 1, 454; tr. THACK-
STON 2012, 177), but the fact that Juvaynī had detailed and first-hand information about his
home region—as evidenced by his mention of the devastation of the small village of Adkān,
west of Isfarāyin and Jūrbad—leaves no doubt about the authenticity of this event.
46 MUSTOWFĪ QAZVĪNĪ 1381/2002, 184; tr. LESTRANGE 1919, 149.
47 Cf. MINORSKY 1948, 293.
48 SPOONER 1965, 101, with no references.
49 NAPIER 1876, 108–9; HOUTUM-SCHINDLER 1877, 223.
50 Cf. SPOONER 1965, 101; AUBIN 2018 [1971], 109.
51 JUVAYNĪ 1912–1937, vol. 1, 115; tr. BOYLE 1958, vol. 1, 146. Persian text:
‫چون برادكان رسيد خضرت مرغزار و انفجار عيون سبتاى را خوش آمد آن جماعت را آسيبى نرسانيد و شحنۀ آنجا بگذاشت‬
52 MUSTOWFĪ QAZVĪNĪ 1381/2002, 185; tr. LESTRANGE 1919, 149–50. Persian text:
This would require further investigation, but what Juvaynī means by Khurāsān
is certainly only a small portion of the region that the Mongols were abandoning to
an indirect civilian Iranian government, as it was not suitable for the nomadic way
of life. The two malikates of Kabūd Jāma and Khurāsān were probably separated
from each other by the road from Radkān to the öleng of Kharaqān, through the
grazing areas of Charmaqān and Kālpūsh, forming a corridor of pastures that al-
lowed nomadic troops to move back and forth between eastern and western Iran. 53
The two autonomous Iranian enclaves would therefore have been easily controlled
from these grasslands that allowed for the stationing and movement of large num-
bers of nomadic soldiers.
Figure 1. Khurāsān in 1232-3

The plain of Radkān undoubtedly served as the headquarters of Chin Temür,
which will also be the case with his successor, Arghun Aqa, 54 and this is where he
stationed his own troops, which, according to Sayfī Haravī, amounted to at least
one tümen.55 Several accounts indeed depict Chin Temür more as a military leader
than as a civilian administrator, and it should be remembered that he was initially
sent in Khurāsān to militarily quell the remnants of the Khwārazmian army. Ju-
vaynī thus says that he “attacked and reduced with troops and brutality” territo-
ries that rebelled.56 Sayfī Haravī (writing in 1322) even titles him amīr instead of
shaḥna, and the Shuʿab-i panjgāna lists him among the commanders (umarā) of Ögö-
‫و در حوالی طوس مرغزاری است که آنرا مرغزار رايکان گويند طولش دوازده فرسنگ و عرضش پنج فرسنگ از مشاهير‬
‫جهان است‬
53 AUBIN 2018 [1971], 102–5.
54 JUVAYNĪ 1912–1937, vol. 2, 247; tr. BOYLE 1958, vol. 2, 510. RASHĪD AD-DĪN 1395/2016, vol. 2,
832, 873, 975; tr. THACKSTON 2012, 325, 343, 381.

55 SAYFĪ HARAVĪ 1383/2004, 129.
56 JUVAYNĪ 1912–1937, vol. 2, 219; tr. BOYLE 1958, vol. 2, 483, slightly modified.
194 SIMON BERGER
dei.57 While Chin Temür was supposed to give way to Dayir Ba’adur, it is said that
“Chormaqan also had sent elchin to summon him and the aforementioned amīrs to
join him with their armies.”58
As we have already mentioned, he oversaw several commanders, such as Kül
Bolad, whom we see essentially acting as a military leader.59 In many ways, Chin
Temür’s administration in Khurāsān mirrored that of Chormaqan in Azerbaijan,
although the former was headed by a basqaq or darughachi, and the latter by a
noyan. Both were centred on vast plains well-suited to the encampment of numer-
ous nomadic troops; both relied, at least partly, on a hierarchy of military com-
manders; and both would allow submissive local aristocrats, whether Iranian or
Armenian and Georgian, to enjoy a degree of autonomy. 60 Both, finally, were ex-
amples of what might be called “ruling from the outside” 61 and retained a signifi-
cant military and nomadic component.
That Chin Temür was supposed to be replaced in Khurāsān by the noyan Dayir
Ba’adur tells indeed a lot about the very nature of his government. As does the
succession of Chin Temür that eventually took place after his death in 1235-6, by
the aged Kereyid soldier Nosal: Juvaynī notably writes that he succeeded him not
as basqaq but as amīr.62 Rashīd al-Dīn even tells that it was Nosal (whom he calls
Noysal) who was initially mandated to rule the province of Iran as amīr-i lashkar,
“commander of the army”, and that Chin Temür came to Iran as his nökör.63 Obvi-
ously, Rashīd al-Dīn relies here on a source which contradicts Juvaynī’s account.
Although the latter was a direct witness, I already pointed out his “anti-militarist”
bias, which might have led him to obscure any form of Mongol administration
before its partial Iranization, and to insist on the role of the bureaucratic class to
which he belonged.
Chin Temür’s prominent military duties were not exceptional as compared to
the role of other darughachin before and during his time: many of them performed
57 SAYFĪ HARAVĪ 1383/2004, 128; I would like to thank Konstantin Golev for drawing my
attention to this passage. RASHĪD AD-DĪN SP, f. 123v, where read Jin-Tämūr ‫ جنتمور‬recte
JTMUR ‫جتمور‬.
58 JUVAYNĪ 1912–1937, vol. 2, 221; tr. BOYLE 1958, vol. 2, 485, slightly modified; emphasis
mine. Persian text:

‫ او و امراى مذكور ايلچى فرستاده بود تا با لشكرها بدو پيوندد‬،‫و جورماغون نيز باستحضار‬
59 JUVAYNĪ 1912–1937, vol. 2, 220; tr. BOYLE 1958, vol. 2, 484. SAYFĪ HARAVĪ 1383/2004, 129.
Rashīd al-Dīn says he came along Chin Temür from Khwārazm as a jarghuchi: RASHĪD AD-
DĪN 1395/2016, vol. 1, 123; tr. THACKSTON 2012, 50.
60 Cf. GRIGOR AKNERCI 1949, 303. See also references to darughachin as part of the military
administration of the Caucasus in BAYARSAIKHAN 2016, 221.

61 DURAND-GUÉDY 2013.
62 JUVAYNĪ 1912–1937, vol. 2, 224; tr. BOYLE 1958, VOL. 2, 488. JACKSON 2017, 109.
63 RASHĪD AD-DĪN 1395/2016, vol. 1 123; tr. THACKSTON 2012, 50, amended.
active military tasks.64 A most telling example is the case of the Salji’ud *Chunjiqai
in North China: a contemporary of Chin Temür, he was appointed “darughachi of
the army and the people” [軍民達魯花赤] of the Yidu province by Ögödei in 1233, and
as such, followed the general Taichu in taking Xuzhou. In 1237, he was transferred
as head darughachi of Jingzhao but was soon replaced by Chaghan “as commander
of the army of Henan” [代察罕總軍河南 ].65
In Iran, it was only when Nosal and then Chormaqan were stripped of civil
power in favour of Körgüz and retained only the command of the army, thus effec-
tively separating the two spheres, that one can observe a real change in the Mongol
administration.66 Before that, the mode of government in the conquered lands cor-
responded essentially to the nomadic administrative tradition, the cornerstone of
which was the army. The institution of the darughachin was specific to the govern-
ment of sedentary populations since it is never found in a steppe context. Never-
theless, it remained an integral part of the Mongolian military system, as evi-
denced primarily by the role of army commanders that the darughachin would al-
ways retain in addition to their role of supervising the census and taxation. This is
also demonstrated by the importance of grazing areas in Mongol control in the
conquered territories: it was from these pastures that the officers, whether noyad or
darughachin, governed with their troops, which clearly illustrates the nomadic
character of this administration.
Even after the 1240s, many features of the early military administration of the
Mongol Empire remained, such as the territorial division of sedentary lands into
tümed, which were not simply territorial units, but were still supposed to contain
ten thousand households capable of contributing to the tax and the auxiliary
troops: this nomenclature indicates how the Mongol imperial state primarily exer-
cised power over individuals and social groups rather than territories, which is one
of its original features. As the late Thomas Allsen has shown, changes occurred
and native personnel were integrated into the state due to the Mongols’ ability to
pick and adapt, rather than imitate, their neighbours’ technologies of governance.67
But they did not wait for a more learned sedentary elite to come and teach them
how to administer the territories and people they had consciously come to exploit.
Bibliography
AIGLE, D. (2008): Iran under Mongol domination: The effectiveness and failings of a
dual administrative system. Bulletin d’études orientales Supplément 72, 65–78.
64 ENDICOTT-WEST 1989, 26–7, 35ff. See also Peter Jackson’s prudent commentaries on the
supposed civilian nature of darughachin: JACKSON 2017, 109–10.
65 SONG LIAN 1976, 123.3030; ENDICOTT-WEST 1989, 33–4.
66 JUVAYNĪ 1912–1937, vol. 2, 229; tr. BOYLE 1958, vol. 2, 493.
67 ALLSEN 1997.
196 SIMON BERGER
ALLSEN, T. T. (1987): Mongol Imperialism. The Policies of the Great Qan Möngke in Chi-
na, Russia and the Islamic Lands, 1251–1259, Berkeley–Los Angeles.
ALLSEN, T. T. (1994): The Rise of the Mongolian Empire and Mongolian Rule in
North China, in Twitchett, D. – Fairbank, J. K. (ed.:) The Cambridge History of
China, vol. 6: Alien Regimes and Borders States, 907-1368. Cambridge, 321–413.
ALLSEN, T. T. (1997): Ever Closer Encounters: The Appropriation of Culture and the
Apportionment of Peoples in the Mongol Empire. Journal of Early Modern Histo-
ry 1, 1, 2–23.
ATWOOD, C. P. (2004): Encyclopedia of Mongolia and the Mongol Empire, New York.
ATWOOD, C. P. (2012): Banner, Otog, Thousand: Apanage Communities as the Basic
Unit of Traditional Mongolian Society. Mongolian Studies 34, 1–76.
ATWOOD, C. P. (2015): The Administrative Origins of Mongolia’s “Tribal” Vocabu-
lary. Eurasia: Statum et Legem 1, 4, 7–45.
AUBIN, J. (1969): L’ethnogénèse des Qaraunas. Turcica 1, 65–94.
AUBIN, J. (2018 [1971]): Réseau pastoral et réseau caravanier. Les grand’routes du
Khorassan à l’époque mongole. Le monde iranien et l’Islam, 1, 105–130; repr. in
Aubin, J. (ed. Aigle, D.), Études sur l’Iran médiéval. Géographie historique et société.
Paris, 91–114.
BARFIELD, T. J. (1989), The Perilous Frontier. Nomadic Empires and China, 221 BC to AD
1757, Cambridge (Mass.).
BAYARSAIKHAN, D. (2016): Darughachi in Armenia, in de Nicola, B. – Melville, C.
(ed.:) The Mongols’ Middle East: Continuity and Transformation in Ilkhanid Iran,
Leiden, 216–236.
BERGER, S. (2021): Mingghan et tamma. L’administration nomade des frontières dans
l’Empire mongol au XIIIe siècle, in Société des historiens médiévistes de l’en-
seignement supérieur public (ed.:) Frontières spatiales, frontières sociales. Paris,
177–193.
BOYLE, J. A. (1958): History of the World Conqueror, 2 vols., Cambridge (Mass.).
BOYLE, J. A. (1968): Dynastic and Political History of the Il-K̲h̲āns, in Boyle, J. A.
(ed.:) Cambridge History of Iran, vol. 5: The Saljuq and Mongol Periods. Cam-
bridge, 303–421.
BUELL, P. D. (1979): Sino-Khitan Administration in Mongol Bukhara. Journal of Asian
History 13, 1, 121–151.
CHANDHOKE, N. (1995), State and Society: Explorations in Political Theory, London.
DOERFER, G. (1963-1975): Türkische und Mongolische Elemente im Neupersischen, 4 vol.,
Wiesbaden.
DURAND-GUÉDY, D. (2013): Ruling from the Outside: A New Perspective on Early
Turkish Kingship in Iran, in Mitchell, L. – Melville, C. (ed.:) Every Inch a King.
Comparative Studies on Kings and Kingship in the Ancient and Medieval Worlds.
Leiden, 325–342.
ENDICOTT-WEST, E. (1989): Mongolian Rule in China. Local Administration in the Yüan
Dynasty, Cambridge (Mass.).
GOLEV, K. (2018): Kangly iz Khorasana: neobychainaya sud’ba odnogo kumano-

kipchakskogo ob’edineniya v vodovorote Mongol’skogo nashestviya, in Mate-
rialy II-i nauchnoi konferencii srednevekovoi istorii Desht-i Kypchaka. Pavlodar, 103–109.
GRIGOR AKNERCI (1949): ed. and tr. Blake, R. P. – Frye, R. N., History of the Nation
of the Archers (The Mongols) by Grigor of Akanc. Harvard Journal of Asiatic
Studies 12, 3–4, 269–399.
HOUTUM-SCHINDLER, A. (1877): Beschreibung einiger wenig bekannten Routen in
Chorassân. Zeitschrift der Gesellschaft für Erdkunde zu Berlin 215–229.
IBN AL-ATHĪR (1965-1967): ed. Thornberg, C. J., Al-Kāmil fī at-taʾrīkh, 13 vols., Beirut.
JACKSON, P. (1991): Čin Tīmūr, in Yarshater, E. (ed.:) Encyclopædia Iranica, vol. 16, 597.
JACKSON, P. (2017): The Mongols and the Islamic World: From Conquest to Conversion,
New Haven–London.
JESSOP, B. (2016), The State. Past, Present, Future, Cambridge.
JUVAYNĪ (1912-1937): ed. Qazvīnī, M. M., The Ta’rīkh-i-Jahān-gushā of ‘Alā’u’d-Dīn
‘Atā Malik-i-Juwaynī, 3 vols., Leiden.
KARA, G. (1990): Zhiyuan yiyu. Index alphabétique des mots mongols. Acta Orien-
talia Academiae Scientiarum Hungaricae 44, 3, 279–344.
K‘ART‘LIS C‘XOVREBA (2014): The Hundred Years’ Chronicle, tr. Gamq’relidze, D., in
Met’reveli, R. – Jones, S. (ed.:) Kartlis Tskhovreba. A History of Georgia. Tbilissi,
315–401.
KIRAKOS GANJAKECI (1986): tr. Bedrosian, R., History of the Armenians, New York.
KRADIN, N. N. – SKRYNNIKOVA, T.D. (2006): Why Do We Call Chinggis Khan’s Poli-
ty “an Empire”? Ab Imperio 1, 89–118.
LESTRANGE, G. (1919): The Geographical Part of the Nuzhat-al-Qulūb, Leiden, London.
LIGETI, L. – KARA, G. (1990): Un vocabulaire mongol des Yuan. Le Tche-Yuan yi-yu.
Acta Orientalia Academiae Scientiarum Hungaricae 44, 3, 259–277.
MAY, T. (2012): The Conquest and Rule of Transcaucasia: The Era of Chormaqan, in
Tubash, J. – Vashalomidze, S. G. – Zimmer M. (ed.:) Caucasus during the Mongol
Period/Der Kaukasus in der Mongolenzeit. Wiesbaden, 129–151.
MAY, T. (2018): The Mongol Empire, Edinburgh.
MINORSKY, V. (1948): Tamīm ibn Baḥr’s Journey to the Uyghurs. Bulletin of the
School of Oriental and African Studies 12, 2, 276–305.
MORGAN, D. O. (1982): Who ran the Mongol Empire? Journal of the Royal Asiatic
Society of Great Britain and Ireland, 1, 1–2, 124–136.
MORGAN, D. O. (1996): Mongol or Persian: The Government of Īlkhānid Iran. Har-
vard Middle Eastern and Islamic Review 3, 1–2, 62–76.
MORGAN, D. O. (2007): The Mongols, Oxford, 2nd ed.
MUNKH-ERDENE, L. (2011): Where did the Mongol Empire Come From? Medieval
Mongol Ideas of People, State and Empire. Inner Asia 13, 2, 211–237.
MUSTOWFĪ QAZVĪNĪ (1381/2002): ed. Dabīr Syāqī, M., Kitāb nuzhat al-qulūb, Tehran.
198 SIMON BERGER
NAPIER, G. C. (1876): Extracts from a Diary of a Tour in Khorassan, and Notes on

the Eastern Alburz Tract; with Notes on the Yomut Tribe. The Journal of the
Royal Geographical Society of London 46, 62–171.
OLBRICHT, P. – PINKS, E. et al. (1980), Meng-ta Pei-lu und Hei-ta shih-lüeh. Chinesische
Gesandtenberichte über die frühen Mongolen, 1221 und 1237, Wiesbaden.
PENG DAYA (1962): ed. Wang Guowei, Heida shilue jianzheng 黑韃事略箋證, in Wang
Guowei, Menggu shiliao sizhong 蒙古史料四種. Taipei, 465–529.
RACHEWILTZ I. de (1966): Personnel and Personnalities in North China during the
Early Mongol Period. Journal of the Economic and Social History of Orient 9, 1–2,
88–144.
RACHEWILTZ, I. de (2004): The Secret History of the Mongols: A Mongolian Epic Chroni-
cle of the Thirteenth Century, 3 vols., Leiden.
RASHĪD AD-DĪN (1395/2016): ed. Rowshan, M. – Mūsavī, M., Jāmiʿ at-tavārīkh:
Tārīkh-i mubārak-i Ghāzānī, 4 vols., Tehran.
RASHĪD AD-DĪN (SP): Shuʿab-i panjgāna, ms. Ahmet III 2932, Topkapı Sarayı Müzesi
Kütüphanesi, Istanbul.
RICHARDS, D. S (2005–2008): The Chronicle of Ibn al-Athīr for the Crusading Period from
al-Kāmil fī’l-taʾrīkh, 3 vols., Burlington.
SAYFĪ HARAVĪ (1383/2004): ed. Ṭabāṭāba’ī Majd, G. R., Tārīkh-nāma-yi Harāt, Tehran.
SECRET HISTORY (1972): ed. Rachewiltz, I. de, Index to the Secret History of the Mon-
gols, Bloomington.
SHENGWU QINZHENG LU (1962): Wang Guowei (ed.:), Shengwu Qinzheng lu jiaozhu
聖武親征録校注, in Wang Guowei (ed.:), Menggu shiliao sizhong 蒙古史料四種. Taipei, 7–219.
SNEATH, D. (2007), The Headless State: Aristocratic Orders, Kinship Society, and Misrep-
resentations of Nomadic Inner Asia. New York.
SNEATH, D. (2013), Ayimag, uymaq and bayliq: Re-examining Notions of Nomadic
Tribe and State, in Paul, J. (ed.:), Nomad Aristocrats in a World of Empires. Wies-
baden, 161–185.
SONG LIAN (1976): Yuanshi 元史, 15 vols., Beijing.
SPOONER, B. (1965): Arghiyān. The Area of Jājarm in Western Khurāsān. Iran, 3, 97–107.
STANDEN, N. (2018): Followers and Leaders in Northeastern Eurasia, ca. Seventh to
Tenth Centuries, in di Cosmo, N. – Mass, M. (ed.:) Empires and Exchanges in
Eurasian Late Antiquity. Rome, China, Iran and the Steppe, ca. 250-750. Cambridge,
400–418.
THACKSTON, W. M. (2012): Classical Writings of the Medieval Islamic World. Persian
Histories of the Mongol Dynasties, vol. 3: Rashiduddin Fazlullah, London.
VARDAN AREVELCI (1989): tr. Thompson, R. W., The Historical Compilation of Var-
dan Arewelc‘i. Dumbarton Oaks Papers, 43, 125–226.
WINK, A. (2001): Conclusion, in Khazanov, A. – Wink, A. (ed.:) Nomads in the Seden-
tary World, Richmond, 285–295.
XU TING (1962): Wang Guowei (ed.:), Heida shilue jianzheng 黑韃事略箋證, in Wang
Guowei (ed.:), Menggu shiliao sizhong 蒙古史料四種. Taipei, 465–529.
ZHAO GONG (1962): Wang Guowei (ed.:), Mengda beilu jianzheng 蒙韃備錄箋證, in
Wang Guowei (ed.:), Menggu shiliao sizhong 蒙古史料四種. Taipei, 431–464.
ZHIYUAN YIYU (2006): ed. Takekoshi, T., “Zhiyuan yiyu” kango goi sakuin
「至元譯語」漢語語彙索引. Kodai moji no sekai e 古代文字資料館 “Kotonoha”, 46, 8–23.
Sarı Saltuk and the Nomads in 13th-
Century Dobruja: Pre-Ottoman Sources
and Realities
DELYAN RUSEV
Abstract
In his Oğuznāme (“Book of the Oghuz”), the 15th-century Ottoman historian
Yazıcıoğlu ʿAlī wove into the narrative of his main source, Ibn Bībī’s Persian histo-
ry of the Rūm Seljuks, a set of interpolations tracing the fate of a large group of
Anatolian Turkmens. They reportedly followed the Seljuk Sultan ʿIzz al-Dīn
Kaykāʾūs II (r. 1246–61) after his flight to Byzantium in 1262 and settled in the re-
gion of Dobruja in the Northeastern Balkans under the leadership of the famous
mystic Sarı Saltuk. This story has become the cornerstone of a longstanding histo-
riographic discussion since the early 20th century, not least because of its supposed
relation to processes of greater historical significance such as the formation of the
14th-century Principality of Dobruja and the genesis of the Gagauz, a population of
Turkish-speaking Christians, in the area. As a number of questions remain unan-
swered in the light of the increasing acceptance of the truthfulness of Yazıcıoğlu
ʿAlī’s account in recent research, the present study aims to reignite the discussion
and challenge some unfounded yet widespread conclusions by revisiting the rele-
vant pre-Ottoman evidence, both written and material. It highlights the Tatar
(Golden Horde) context of Sarı Saltuk’s presence in late 13 th-century Northern Do-
bruja and in the town of Sakçı (Isaccea) in particular, but casts serious doubt on his
connection with the alleged Anatolian migrants as well as with Babadag, a town
that was (re)established and associated with him in Ottoman times.
 Sofia University “St. Kliment Ohridski”. This paper is a result of research carried out with-
in the project “Nomadic Migrations in the Bulgarian Lands during the Late Middle Ages –
History, Memory and Usage” funded by the Bulgarian Science Fund under contract no. KP-
06-M30/4 from 17.12.2018. An earlier version of it was presented at the 8 th International
Conference on Medieval History of the Eurasian Steppe Nomads and their Neighbours in the
Middle Ages, held in Sofia on November 20–3, 2019. I owe a debt of gratitude to Dr. Konstan-
tin Golev and Prof. Pavel Pavlovitch for their assistance towards different aspects of this
study.
202 DELYAN RUSEV
Keywords: Sarı Saltuk, Turkmens, migration, Dobruja, Tatars, Golden Horde, By-
zantium, Bulgaria, Gagauz, Sakçı (Isaccea), Babadag
After the death of their great leader [Sarı Saltuk] (in 1297-8, according to the
contemporary Saradj) some of the Turks returned to Anatolia and settled in
what then had just become the emirate of Karası, while others remained [in
Dobruja], and entered converted to Orthodox Christianity Byzantine service.
Under their leaders, the Despot Balık, Culpan, Balık’s son Dobrotić and
grandson Ivanko, they maintained their political independence from Bulga-
ria as the principality of Dobrudja, which existed from the 1330s till the ﬁnal
annexation by the Ottomans under Mehmed I in 1417. As a people they be-
came known as Gagauz, a corruption of Kaykavus. 1
This passage shows the textbook reconstruction of the events that followed the
supposed resettlement of a large nomadic group from Seljuk Anatolia to the region
of Dobruja (in today’s Northeastern Bulgaria and Southeastern Romania) in the
early 1260s as summarized by Machiel Kiel in the Cambridge History of Turkey. The
assumed unconditionality of the migration as a historical fact as well as of its cau-
sality with respect to later phenomena—such as the emergence of the Principality
of Dobruja and the presence of Turkic-speaking Christians, the Gagauz, in the ar-
ea—is representative for much of the abundant research on this century-old topic
in historiography. Unfortunately, the same can be said about the attenuated atten-
tion to historical detail evident in the above quotation. 2 Besides the methodological
imperfections and the political biases that have plagued the debate, it has also been
hampered by the fact that the relevant written sources are both inconclusive and
very diverse in terms of language, genre, and availability. Moreover, the possible
relation of this historical (or historiographic) episode to several larger themes like
the ethnic and political setup of the late medieval Balkans, the advent of Islam in
the region and the famous Sarı Saltuk as the arguable originator of this process, the
ethnogenesis of the Gagauz, etc., has led to its treatment in a boundless number of
studies in a large variety of languages. 3 All this has made it increasingly difficult
for a single scholar to comprehend the accumulated body of information, let alone
to offer a convincing narrative of the events without preconceived conclusions.
1 KIEL 2009, 141.

2 Balık/Balik did not hold the title of despot, which was only granted to Dobrotitsa, who
was not Balik’s son but his brother; accordingly, Ivanko was a nephew of Balik and not his
grandson. As for the chronology of the principality’s existence, it is only attested between
the 1340s and the early 1390s, see, e.g., ATANASOV 2009.
3 Key studies concerning the separate aspects are cited in the relevant sections of the text
below. The major works focusing on the supposed migration itself are DIMITROV 1915;
BALASCHEV 1930; DUDA 1943; MUTAFČIEV 1943; MUTAFČIEV 1973 (a Bulgarian version of the
former with some minor differences); WITTEK 1952; DECEI 1968; KIEL 2000; MATANOV 2013;
OCAK 2016, 39–59 (a revised and expanded version of OCAK 2002).
SARI SALTUK AND THE NOMADS IN 13TH-CENTURY DOBRUJA… 203
While the present paper does not pursue such an ambitious goal, it aims to offer a
critical analysis of the relevant contemporary and near-contemporary sources,
including the available archaeological evidence, and thus to judge the credibility of
the narrative of the events as it appeared in the early Ottoman setting. According-
ly, later Ottoman sources will be used only marginally to the extent that they can
corroborate the pre-Ottoman evidence. In view of the modern historiography, the
aim here is to sift out the proven facts from the well-founded hypotheses and the
pure speculations, which have often been reproduced on equal grounds.
Yazıcıoğlu ʿAlī’s account
The main evidence regarding the appearance of Anatolian nomads in mid-13th-
century Dobruja is provided by the secretary of the Ottoman Sultan Murād II (r.
1421–44, 1446–51), Yazıcıoğlu ʿAlī, in his History of the Seljuk Dynasty, alternatively
titled Oğuznāme and composed in the 1420s or the 1430s.4 Following his main
source, the Persian history of the Seljuks of Rūm by Ibn Bībī, Yazıcıoğlu recounts
the fate of the Seljuk Sultan ʿIzz al-Dīn Kaykāʾūs II (r. 1246–61, d. 1279/80) after his
flight from Anatolia under Mongol pressure and his refuge at the Byzantine court
in 1262 together with his family and retinue. 5 However, the Ottoman historian has
made a number of interpolations to the original narrative including the following
key passage:
One day, Sultan ʿIzz al-Dīn and [his general] ʿAlī Bahādur told the Basileus
[Michael VIII]: “We are a Turkish people, we cannot bear staying in a big
city. If there were a place and a home for us outside, we could bring from
Anatolia the Turkish families bound to us, and we could set up our summer
and winter camps there.” The Basileus gave them as a place and a home the
land of Dobrūca, which is a hunting field with nice water and air, and
healthy for breeding good cattle. They secretly notified the Turkish clans in
Anatolia who were related to them. Many Turkish families descended [from
the mountains] to Izniḳ under the pretext that they were going to their win-
ter quarters and, in a short time, they migrated [to the European coast]
through Üsküdār. Ṣaru Ṣaltuḳ of blessed memory also crossed [the Bospho-
rus] with them. For a long time, there were two-three Muslim towns and
4 The most widely accepted dating of the work is 827/1423–4 (DUDA 1943, 138; WITTEK 1952,
646), but Paul Wittek has later proposed an alternative reading of the enigmatic chronogram
as equal to 840/1436–7 (WITTEK 1964). For the interchangeable use of the titles Tārīkh-i āl-i
Selçuḳ and (Tevārīkh-i) Oğuznāme in the available MSS, see their description in YAZICIZĀDE
ʿALĪ/BAKIR 2017, LIX–LXVIII.
5 For the dating and a reconstruction of the events, see SHUKUROV 2016, 99–105. For Ibn
Bībī’s account, see IBN BĪBĪ/DUDA 1959, 282–5.

204 DELYAN RUSEV
thirty-forty camps of Turkish clans in the land of Dobrūca. They would con-
front and crush the enemies of the Basileus.6
After this addition, Yazıcıoğlu paraphrases with a few minor yet significant
changes Ibn Bībī’s account of ʿIzz al-Dīn’s involvement in a conspiracy to over-
throw the Byzantine emperor (in which, according to Yazıcıoğlu’s interpolation,
the plotters could rely on “ten to twelve thousand men”7), the resulting detainment
of the sultan, his liberation by a military force sent from the Golden Horde, and his
transfer to Crimea, where Berke Khan (r. 1257–66) grants him the cities of Ṣogdāḳ
(Sudak) and Ṣūlgād (Solkhat; present-day Staryi Krym). Most of the changes are
meant to introduce two further interrelated plot lines developed later in the Oğuz-
nāme and missing in the Persian original. 8 The first one traces the subsequent fate
of those members of ʿIzz al-Dīn’s family who reportedly remained in Byzantium
and were sent to Karaferya (Byz. Berroia, modern Veroia in Northern Greece)
where one of his sons would become a governor and die as a Muslim, and so
would his son. Their descendants would later accept Christianity, enter Ottoman
service, and remain in the area until the time of Yazıcıoğlu, when two of them—
Dīmitrī Sulṭān and Mīho Sulṭān—renewed their privileges at the Ottoman court. 9
The other line of narration concerns the nomads who had settled in Dobruja
and their leader Sarı Saltuk. On its way back, the Tatar army that liberated ʿIzz al-
Dīn reportedly drags them along, too, and brings them to Crimea. 10 It is only after
the former sultan’s death eighteen years later that they are allowed to return to
6 For a facsimile and a Latin transliteration of the main MS (Topkapı Sarayı Müzesi Kütüphane-
si, Revan 1391), see respectively YAZICIZĀDE ʿALĪ/BAKIR 2014, 375a-b and YAZICIZĀDE
ʿALĪ/BAKIR 2017, 631. For a printed reproduction and a German translation of the text in the
Berlin MS (Staatsbibliothek, Ms. Or. Quart 1823), see DUDA 1943, 143–4. For a Latin transcrip-
tion of the passage in another Topkapı MS (Revan 1390), see DECEI 1968, 88. The translation
of the phrase “healthy for breeding good cattle” offered above is approximate, for the un-
derlying Ottoman Turkish wording is both unclear and dissimilar in the different MSS and
editions. The readings of Duda and Decei are rather unconvincing. Yet, the words “good
cattle” (eyü davār/ṭavār), and “healthy” (tendürüst) are clearly identifiable in both MSS Revan
1391 and Berlin.
7 YAZICIZĀDE ʿALĪ/BAKIR 2014, 375b; YAZICIZĀDE ʿALĪ/BAKIR 2017, 631.
8 All relevant passages from Yazıcıoğlu’s work are summarized in more detail by WITTEK
1952, 648–50, who has also highlighted the differences with Ibn Bībī’s narrative and identi-
fied the separate plot lines.
9 YAZICIZĀDE ʿALĪ /BAKIR 2014, 376a, 415b–416a; YAZICIZĀDE ʿALĪ/BAKIR 2017, 632, 701; WIT-
TEK 1952, 648–50. This subject has been studied extensively and has a rather marginal rele-
vance to the migration story, so it will not be treated here: see WITTEK 1934; LAURENT 1956;
ZACHARIADOU 1965; ZHAVORONKOV 2006, 168–74; SHUKUROV 2008; SHUKUROV 2016, 105–19.
10 DUDA 1943, 144 (text and German translation); DECEI 1968, 88 (transliteration). This ac-
count is omitted in MS Revan 1391 and A. Bakır’s edition, apparently due to an error of the
copyist.
Dobruja.11 Sometime later, they are joined by another one of ʿIzz al-Dīn’s sons, who
had remained in Constantinople, had been baptized and had entered the service of
the Patriarch, but was released by him at the request of Sarı Saltuk. The latter con-
verts the prince back to Islam, becomes his spiritual tutor, and gives him the name
Barak (“the Dog”) after he eats the sheikh’s vomit. Yazıcıoğlu’s account that Barak
was later sent to the Ilkhanid capital city of Sulṭāniyya and was the spiritual pro-
genitor of the Barakis makes it clear that the well-known early 13th-century mystic
Barak Baba is meant here.12 As for the Turks in Dobruja, pushed away by the Bul-
garian reconquest of the region, they join a certain Khalīl Ece and cross the Straits
to the land of Karasi in Anatolia. After the death of Sarı Saltuk, the descendants of
those who remain in the Balkans become apostates and unbelievers (akhriyān).13
ʿIzz al-Dīn’s supposed followers in Byzantine and Catalan service (1260s–1310s):
Persian, Byzantine, Latin, and Slavic sources
ʿIzz al-Dīn Kaykāʾūs II’s flight from Anatolia and his refuge in Byzantium were
recorded in several contemporary and near-contemporary sources in Persian (Ibn
Bībī and Aqsarāyī) and Greek (Pachymeres and Gregoras), but none of them men-
tions a largescale resettlement of Anatolian nomads to Dobruja or the involvement
of the later famous mystic Sarı Saltuk in the events. Paul Wittek has explained the
silence of Ibn Bībī (d. after 1285) and Aqsarāyī (d. ca. 1323–33) with the argument
that the “chancery at Konya”, where both authors were employed at different
points of time, would and could not “care about nomad movements on a distant
frontier”.14 More recently, Andrew Peacock has challenged that proposition by
demonstrating the considerable integration of the mostly Turkmen-inhabited bor-
der regions in the political, economic, and cultural structures of the Rūm Seljuk
state.15 In another study, he has also shown that the Seljuk court maintained both a
strong symbolic attachment to nomadic traditions and a close, albeit complex, rela-
tionship with its Turkmen subjects whom the sultans sought to appease, control,
and exploit due to their vast demographic and military potential. 16
11 YAZICIZĀDE ʿALĪ/BAKIR 2014, 411b, 414a; YAZICIZĀDE ʿALĪ/BAKIR 2017, 695, 698–9; WITTEK
1952, 649; DECEI 1968, 89.
12 YAZICIZĀDE ʿALĪ/BAKIR 2014, 415a-b; YAZICIZĀDE ʿALĪ/BAKIR 2017, 700–1; WITTEK 1952,
649–50; DECEI 1968, 89–90. The identity of Barak and his possible relation to the Dobruja
story are discussed below.
13 YAZICIZĀDE ʿALĪ/BAKIR 2014, 444a-b; YAZICIZĀDE ʿALĪ/BAKIR 2017, 741; WITTEK 1952, 651;
DECEI 1968, 90.

14 WITTEK 1952, 654.
15 PEACOCK 2014, esp. 286: “If the frontier and its Turkmen rarely feature in our Persian his-
torical sources, this has everything to do with the genre and intentions of the historian, and
nothing to do with the limits of his knowledge.”
16 PEACOCK 2013, esp. 214–15 with a focus on ʿIzz al-Dīn Kaykāʾūs II.
206 DELYAN RUSEV
This ambivalent relationship was most obvious in the reign of ʿIzz al-Dīn
Kaykāʾūs II himself and it is no surprise that the Turkmens do in fact feature prom-
inently in Aqsarāyī’s narrative tracing the sultan’s struggle for the throne and his
ultimate flight to Byzantium. The historian notes that the men of ʿIzz al-Dīn’s
brother and rival Rukn al-Dīn accused the former, in front of a Mongol envoy, of
having allied himself with the “Turks of the frontier zone” (bā atrāk-i ūj muttafaq
shuda) with the intention to rebel against his Mongol overlords. According to
Aqsarāyī, this was a false allegation (qāẕūrāt, lit. “impurities”) and ʿIzz al-Dīn had
actually left Konya in order to fight the rebellious “emir of the Turks of the ūj”
Muḥammad Beg, who ultimately defeated the sultan’s forces “between Anṭāliya
and ʿAlāʾiya” (mod. Alanya).17 The damage had been done, however, and a joint
campaign of Rukn al-Dīn and the Mongols compelled ʿIzz al-Dīn to seek refuge in
Byzantium together with his family and some of his emirs. 18 Noteworthy is
Aqsarāyī’s account that after the sultan’s flight and the enthronement of Rukn al-
Dīn in Konya, the forces of the latter and the Mongol troops “entered the frontier
areas (valāyat-i ūj) and mopped up the provinces up to the borders of the lands of
Istanbūl of contending Turks and rebellious outlaws”; Muḥammad Beg and some
other begs were captured and the subjugation (taskhīr) of the Turkmens was
achieved, at least temporarily.19 This episode did not necessarily ensue from the
links between the dethroned sultan and the nomads. The earlier struggle between
ʿIzz al-Dīn and Muḥammad Beg and the evidence provided by the Mamluk histo-
rian Baybars al-Manṣūrī (d. 725/1325) both indicate that the Turkmen leader was
acting on his own behalf and the Mongols had other reasons to detain him. 20
Nevertheless, Aqsarāyī further notes that in the same year (i.e. 660/1261–2) “the
fearless emirs of the Turks [called] Qarāmān, Zayn al-Ḥāj, and Būnsūz” rose from
Ermenāk and attacked Konya with 20,000 men out of fidelity (havādārī, lit. “affec-
tion”) to ʿIzz al-Dīn Kaykāʾūs II, before they were also crushed by Rukn al-Dīn’s
army.21 Coupled with the reported purge of the frontier zone with Byzantium, this
account adds some weight to Yazıcıoğlu ʿAlī’s story but, on the other hand, it am-
plifies the question of why the Seljuk historians would ignore the mass nomadic
migration that allegedly followed the sultan in exile. Moreover, the indicated areas
of activity of the begs affiliated with ʿIzz al-Dīn, namely the southern Anatolian
coastlands, are far removed from the region of Nicaea (Iznik), where the Ottoman
historian locates the departure of the nomadic migrants to Dobruja. It is worth
noting in this regard that some scholars have attempted to identify the latter with a
17 AQSARĀYĪ/TURAN, 66; AQSARĀYĪ/IŞILTAN, 53.

18 AQSARĀYĪ/TURAN, 69–70; AQSARĀYĪ/IŞILTAN, 54–5.
19 AQSARĀYĪ/TURAN, 71; AQSARĀYĪ/IŞILTAN, 55.
20 For an English translation of al-Manṣūrī’s account, see PEACOCK 2014, 283. On the Turk-
men unrest at the Seljuk-Byzantine frontier in the early 1260s, see also SHUKUROV 2016, 123–4
and the studies cited there (n. 180).
21 AQSARĀYĪ/TURAN, 71–2; AQSARĀYĪ/IŞILTAN, 55–6.
particular tribal community, be it the Chepni Turkmens or a group of Cuman-

Qïpchaqs arguably living in the region of Sinop. 22 It is difficult to discuss these
hypotheses as they are mainly based on equally speculative assumptions about the
origin of Sarı Saltuk, whose relation to the supposed migration is highly dubious,
as will be demonstrated below.
The contemporary Ibn Bībī pays less attention to the Turkmen involvement in
the events surrounding ʿIzz al-Dīn Kaykāʾūs II’s flight to Constantinople. He fo-
cuses on the activities of ʿAlī Bahādur, one of the emirs who remained loyal to the
dethroned sultan, and to whom Yazıcıoğlu ʿAlī attributes an instrumental role in
securing the nomad migration to Dobruja.23 He attempted an attack on Rukn al-
Dīn in Konya but was defeated and had to find refuge in the frontier zone (ūj)
where, according to Ibn Bībī and in some contrast to Yazıcıoğlu ʿAlī’s account, the
emir did not feel comfortable and feared the “barbarism of the [local] Turkish
tribes” (jahālat-i ṭavāyif-i atrāk). Consequently, ʿAlī Bahādur joined ʿIzz al-Dīn in the
Byzantine capital with only “a handful of his retainers” (ba sharz̲ima az ḥavāsh-i
khvīsh).24 While this evidence hardly speaks in favour of any significant Turkmen
movement across the Seljuk-Byzantine border, Ibn Bībī’s report that ʿAlī Bahādur
was later successfully employed by Michael VIII (r. 1259–82) against the enemies of
the empire suggests that a certain military force may have followed him and ʿIzz
al-Dīn on Byzantine territory, but not necessarily to Dobruja. 25
Before proceeding to the Greek sources, which seem to support this hypothesis,
it should be noted that the enrolment of individuals and whole contingents of Ana-
tolian Turkish origin in the Byzantine army was a common practice dating back to
the 11th century.26 Some of them became Byzantine subjects and settled in the Bal-
22 The Chepni hypothesis was first put forward by KÖPRÜLÜ 1993, 15, n. 57 (originally pub-
lished in Turkish in 1922) and has later been deemed probable by TOGAN 1981, 268 and
OCAK 2016, 52, 92–3, 136. For the Cuman-Qïpchak link, see ÖZTÜRK 2013, 126–7.
23 See n. 6 above.
24 IBN BĪBĪ/DUDA, 283–4, 342, n. 373; SHUKUROV 2016, 124. In fact, Yazıcıoğlu follows Ibn Bībī
in describing ʿAlī Bahādur’s “fear of the wickedness of the frontier Turks” (uç Türkleri
şerrinden ḳorḳuyla yürüdi) just before he adds the story of the Turkmens’ resettlement to
Dobruja: YAZICIZĀDE ʿALĪ/BAKIR 2014, 375a; YAZICIZĀDE ʿALĪ/BAKIR 2017, 631. ʿAlī Bahā-
dur’s “uprising” (khurūj) and its suppression are also reported by Aqsarāyī, but he gives no
information on the later fate of the emir: AQSARĀYĪ/TURAN, 74; AQSARĀYĪ/IŞILTAN, 58.
25 IBN BĪBĪ/DUDA, 284; SHUKUROV 2016, 125. ʿAlī Bahādur could have participated in Byzan-
tine military campaigns between 1262, when he arrived in Constantinople, and the winter of
1264–5, when he was executed on the accusation of having joined ʿIzz al-Dīn’s alleged con-
spiracy against the emperor (see IBN BĪBĪ/DUDA, 284; SHUKUROV 2016, 225). During this peri-
od, the Byzantines fought against the Bulgarian tsar Konstantin Tikh (r. 1257–77) in Thrace,
against Despot Michael (r. ca. 1230–67/8) in Epirus, and against Guillaume de Villehardouin
(r. 1246–78) in Morea. The relevant evidence is discussed below.
26 See BRAND 1989; SAVVIDES 1993, 125–8.
208 DELYAN RUSEV
kans.27 There is also evidence corroborating Yazıcıoğlu ʿAlī’s claim that some of
ʿIzz al-Dīn Kaykāʾūs’ relatives lived in Berroia, or at least in the region, and it has
been surmised that some of the personal and place names with Turkish or Muslim
etymology attested in present-day Northern Greece in the second half of the 13th
century may be related to Anatolian migrants who appeared there alongside the
sultan’s people.28 Against this background, the silence of the contemporary histori-
an George Pachymeres (1242–ca. 1310), a resident of Constantinople after 1261,
about an influx of thousands of Turkmens to the Balkans ca. 1262 is rather puz-
zling, especially if Yazıcıoğlu’s account of their itinerary through the Bosporus is
true.29 The version of the next-generation historian Nikephoros Gregoras (d. ca.
1361) has more points of intersection with that of his Ottoman counterpart.
Gregoras similarly notes that when ʿIzz al-Dīn arrived in Byzantium he requested
from Michael VIII to either help him fight the Mongols or “allot him a piece of
Roman land where he could live permanently with his people”.30 According to
Gregoras, however, the emperor judged both of these options to be risky and un-
wise, and therefore he let the sultan stay in hope without satisfying his request. To
Pachymeres’s first-hand account of the imprisonment of ʿIzz al-Dīn’s family mem-
bers after his flight from Byzantium in the winter of 1264–5,31 Gregoras adds that
the majority of the sultan’s men “who were quite numerous and excelling at war”,
were baptized and incorporated into the imperial army. 32 Gregoras’s narrative of
the period of ca. 1261–1308 is mainly based on that of Pachymeres, but his source
for this interpolation cannot be identified. It is thus impossible to judge its veracity,
but it is undoubtedly related to a much later episode, which will be dealt with be-
low. It may also have to do with Pachymeres’s statement that some Anatolian no-
mads (σκηνίτας), unwilling to submit to the Mongols, appeared on Byzantine terri-
27 Eustathius of Thessalonike (d. 1195/6) claimed in 1178 that so many Turks (“Persians”)
had settled around Thessalonike that it could be called “New Turkey, or the European land
of the Turks” (quoted from BRAND 1989, 13). See also the encomium for John III Doukas
Vatatzes (r. 1222–54) hailing his policy of resettling Cumans from the Balkans to Anatolia
and “substituting [them] for the sons of Persians” (quoted from BARTUSIS 1992, 26).
28 On the fate of ʿIzz al-Dīn’s relatives in Byzantium, see the works cited in n. 9 above. For a
study of Turkic onomastic material from the region of Macedonia, see SHUKUROV 2016, 128–
31, 159–77. The evidence quoted in the previous note here suggests that some of the topo-
nyms might have a much earlier origin.
29 For Pachymeres’ account of ʿIzz al-Dīn’s flight, see PACHYMERES/FAILLER, vol. 1, 180–5 (II, 24).
30 GREGORAS/SCHOPEN, 82: ἢ συμμαχίαν κατὰ τῶν Σκυθῶν, ἢ γῆς ῾Ρωμαϊκῆς ἀποτομήν τινα
καὶ οἱονεὶ κληρουχίαν εἰς κατοικίαν μονιμωτέραν αὐτοῦ τε καὶ τῶν ἅμα αὐτῷ;
GREGORAS/VAN DIETEN, 103–4.
31 PACHYMERES/FAILLER, vol. 1, 312–3 (III, 25).
32 GREGORAS/SCHOPEN, 101: ὁ δὲ περὶ ἐκεῖνον ὄχλος, ἄνδρες δ᾽ οὗτοι μάλα τοι πλεῖστοι καὶ
κράτιστοι τὰ πολέμια, τῷ Χριστιανῶν ἀναγεννηθέντες βαπτίσματι, τῇ ῾Ρωμαίων

συγκατελέγοντο στρατιᾷ; GREGORAS/VAN DIETEN, 114. Cf. the somewhat different translation
by MUTAFČIEV 1943, 6.
tory and the emperor was eager to employ them as a shield against possible Mon-
gol attacks despite the troubles they caused to the local population.33 Although the
historian does not link them explicitly to ʿIzz al-Dīn, he has placed this information
right after his account of the sultan’s flight to Constantinople. It appears that these
nomads were active in the Anatolian border zone, but further evidence reveals that
Turkish detachments also participated in Byzantine campaigns in Europe during
the period under consideration.
Pachymeres mentions Turks, or “Persians” as he calls them, as a part of the
Byzantine army in Morea in 1263.34 As pointed out by Rustam Shukurov, more
information on them is found in the Chronicle of Morea. According to a not entirely
clear passage from its Greek version, at the outset of the campaign Michael VIII
“went to Turkey (Τουρκία) and hired the Turks; he hired a thousand of select
troops and five hundred others, and around another two thousand Anatolians
went with them”.35 Shukurov has further drawn attention to the Chronicle’s account
that two years later, after the Turks had deserted the Byzantines and spent some
time in service of Guillaume de Villehardouin in Morea, one of their leaders called
Melik asked the prince for permission to return to his “patrimonies” (ἰγονικά),
whereupon he “went to Vlachia”.36 Although he concedes the polysemantic mean-
ing of Vlachia at that time and its possible relation to many locations in the Balkans,
Shukurov assumes that it may refer to Dobruja and adds that, “[i]n any case, Me-
lik’s ἰγονικά was located in Europe, but not in Anatolia, which confirms that his
33 PACHYMERES/FAILLER, vol. 1, 184–7 (II, 24). Cf. MUTAFČIEV 1943, 2; SHUKUROV 2016, 123.
34 PACHYMERES/FAILLER, vol. 1, 273 (III, 16); SHUKUROV 2016, 125.
35 CM/SCHMITT, 300 (vv. 4553–5); CM/LURIER, 203–4; I follow the more literal translation by
SHUKUROV 2016, 126, n. 188. Shukurov wonders whether the term Tourkia denotes the Gold-
en Horde or Anatolia, and whether “a Mongol or Cuman detachment is implied here or just
two different groups of Anatolian Turks”. His suggestion that “it is more likely that Tourkia
denotes here Anatolia” is supported by another passage in the Chronicle where the steppe
region to the north of the Black Sea is called “Cumania”, although this account is related to
an earlier period (CM/SCHMITT, p. 70, v. 1038; CM/LURIER, 97). However, Shukurov’s insist-
ence that “probably the Chronicle intends to draw up a distinction between the 1,500 merce-
naries hired in Anatolia/Tourkia, and the 2,000 Anatolian Turks living in the Balkans, that
is, Kaykāwus’ Turks” is unconvincing. Actually, the meaning of the term “Anatolians” in
the Greek text becomes clear from the French version of the Chronicle, which has the Byzan-
tine army consisting of 1,500 Turks and “the Greek forces from the Levant who were used to
how the Turks fought” (translation by CM/VAN ARSDALL – MOODY, 94; French text:
CM/BUCHON, 154–5). That the “Anatolians” were actually Christian soldiers is further sup-
ported and specified by Pachymeres’s account. He reports that the emperor dispatched to
Monemvasia the Sebastokrator Constantine, who was in charge of the campaign, together
with the Turks (τὸ Περσικὸν) and the “Romans from Magedon” (ἐκ Μαγεδῶνος Ῥωμαίων),
i.e. from the Anatolian town of Magidion near ancient Saittai in Lydia: PACHYMÉRÈS/
FAILLER, vol. 1, 272–3 and 290, n. 2 for the identification of the town. Cf. also SAVVIDES 1992.
36 CM/SCHMITT, 371 (vv. 5710–29); CM/LURIER, 236–7; SHUKUROV 2016, 125.
210 DELYAN RUSEV
soldiers came from Kaykāwus’ Turks”. 37 In fact, the term Vlachia is used through-
out the Chronicle of Morea with a consistent meaning denoting roughly Thessaly
and some territories immediately to the north of it.38 Moreover, according to the
French version, Melik actually asked his master to let him “go to Vlachia, in order
to find passage to go to Turkey”, which is more compatible with the account of the
Turks’ recruitment from Anatolia quoted above.39
Turkish detachments from Anatolia were apparently also used in the Byzantine
campaign in Morea in 1270–2, when the imperial forces, including Greeks, Turks,
and Cumans, were recruited from “the region of Nicaea” or “the East” according to
the Greek and the French versions of the Chronicle of Morea, respectively.40 Turks
are also mentioned as members of the Byzantine troops in Albania (1281) 41 and
Epirus (1292)42, yet without evidence regarding their origin or geographical prove-
nance. A little more specific is Pachymeres’s account of “Persians” in the Byzantine
army in Thessaly ca. 1275, when they were led by a certain Rimpsas. If he is identi-
cal with Nikephoros Rimpsas, mentioned by Akropolites as a Christian Turk who
participated in the battle of Pelagonia in 1259, he was certainly not one of ʿIzz al-
Dīn’s men.43 This identification poses the question of whether the Turks under
consideration were not descendants of earlier, pre-1261 Anatolian settlers in the
Nicaean possessions in the Sothern Balkans of whose existence there are some
hints in the sources.44 Indicative in this regard, if correct, is also Gregoras’s use of
the designation Turcopoles (Τουρκόπουλοι, lit. “sons of Turks”) for the Turkish
detachment in the Thessalian campaign, although it may also imply a link to ʿIzz
37 SHUKUROV 2016, 125–6.

38 See e.g. CM/SCHMITT, 70, 175, 232 (vv. 1031, 2588–9, 3504); CM/LURIER, 97, n. 69, 146–7,
175, where the context is quite conclusive. The translators of the French version have even
rendered the original Blaquie as “Thessaly” (see e.g. the reference in the next footnote).
39 CM/BUCHON, 192; CM/VAN ARSDALL – MOODY, 104. On the interrelations between the
different extant versions of the Chronicle of Morea, which seem to derive from a now lost
archetype, probably in Greek, yet partly represented better by the known French recension,
see SHAWCROSS 2009, 31–52.
40 The Greek text actually reads “from the region of Lycia” (ἐκ τῆς Λυκίας τὰ μέρη) –
CM/SCHMITT, 421 (v. 6490), which was translated as “from the region of Nicaea” by
CM/LURIER, 257. In any case, an Anatolian toponym is meant. For the French version, see
CM/BUCHON, 217; CM/VAN ARSDALL–MOODY, 114. On the campaign, see also VÁSÁRY 2005,
116; GEANAKOPLOS 1959, 229–30.
41 This evidence is provided by the Venetian Marino Sanudo Torsello in his Istoria del Regno
di Romania, quoted in GEANAKOPLOS 1959, 332–3, n. 114.

42 See CM/SCHMITT, 589 (v. 9086); CM/LURIER, 321–2; CM/BUCHON, 310; CM/VAN ARSDALL
– MOODY, 142. See also VÁSÁRY 2005, 117–18.

43 PACHYMÉRÈS/FAILLER, vol. 2, 424–5 (IV, 31), and n. 4 there for the proposed identification
with Nikephoros Rimpsas (see also PLP, no. 24292).

44 See n. 27 above.
al-Dīn’s Turks in particular (see next paragraph). 45 The dearth of details leaves all
possible interpretations open, but it is least likely that the Turks came from a dis-
tant area out of direct Byzantine control such as Dobruja, all the more that accord-
ing to Yazıcıoğlu ʿAlī’s own version of the events by the 1270s the Anatolian no-
mads had long migrated to Crimea before returning to the Balkans towards the
end of that decade or the beginning of the next. 46
Turcopoles appear again in the tumultuous events caused by the Catalan Com-
pany in the Southern Balkans in the early 14 th century. They were present at the
Byzantine camp in Adrianople in 1305 and deserted the imperial army during the
Battle of Apros later in the same year to join the opposing Catalans and their Ana-
tolian Turkish allies who were led, according to Gregoras, by a certain Khalīl
(Χαλὴλ).47 Pachymeres introduces the term Tourkopouloi here as an alternative des-
ignation of the “Persian” detachment of old. 48 His rather vague statement suggests
that the Turcopoles were the descendants of Turkish soldiers who had been
fighting in the Byzantine army for a long time. Gregoras supports this notion by
explicitly identifying them with ʿIzz al-Dīn’s soldiers who had remained in Byzan-
tium and had been baptized after the sultan’s flight to the Golden Horde.49 Con-
sidering the chronology and the use of the designation Tourkopouloi itself, it is more
reasonable to assume that these were rather the descendants of ʿIzz al-Dīn’s men,
but it is also possible that such a connection never existed and the historian was
just trying to link two unrelated events by scaling up the number of the sultan’s
following in the first place. In any case, this evidence alone gives no reason to iden-
tify the Turcopoles with the Dobruja Turks.
The Dobruja link seems to be reestablished by Pachymeres, who, in contradic-
tion to his own earlier statement and that of Gregoras, describes the Turcopoles of
1305 as recently converted Christians who had joined the emperor “from the
45 GREGORAS/SCHOPEN, 111 (IV, 9); GREGORAS/VAN DIETEN, 120; BARTUSIS 1992, 61–2; cf.
SHUKUROV 2016, 126, n. 190, who argues that the term is used here retrospectively. The term
Tourkopouloi/Turcopoles is attested since the 11th century to denote Christianized Turks or
descendants of mixed marriages between Turkish men and Christian women, who occa-
sionally served in the Byzantine army but were most widely employed in the Latin states of
the Eastern Mediterranean, where the term gradually acquired a more common, technical
meaning: see SAVVIDES 1993.
46 See nn. 10–11 above and further below in the text, where the political situation in Dobruja
is also discussed.
47 GREGORAS/SCHOPEN, 232 (VII, 4); GREGORAS/VAN DIETEN, 183. For an overview and chro-
nology of the events (contrary to the traditional dating of the Battle of Apros in 1307), see
e.g. BARTUSIS 1992, 80–1; JACOBY 1979, 231. For the Catalan Company see JACOBY 2015 and
the works cited there.
48 PACHYMÉRÈS/FAILLER, vol. 4, 572–3 (XII, 23): τὸ ἐκ παλαιοῦ Περσικόν, οὓς καὶ
Τουρκοπούλους ὠνόμαζον; cf. SHUKUROV 2016, 126, whose translation once again differs
from that of Failler.
49 GREGORAS/SCHOPEN, 229 (VII, 4); GREGORAS/VAN DIETEN, 182.
212 DELYAN RUSEV
northern lands” not many years ago.50 This somewhat obscure account, Gregoras’s
mention of the name Khalīl, and the Thracian conquests of the Bulgarian tsar Teo-
dor Svetoslav in 1304–5 all seem to support Yazıcıoğlu’s statement that the majori-
ty of the Dobruja Turks left the province and crossed to Anatolia under the leader-
ship of Khalīl Ece after “the Bulgarian rulers had risen, overcome the Basileus, and
taken most of Rumelia”.51 Nevertheless, one should keep in mind that the reported
“northern” provenance of the Turcopoles has been interpreted differently by some
scholars. Besides Petar Mutafchiev’s doubt about the accuracy of Pachymeres’ ac-
count,52 Rustam Shukurov has identified these “northern Tourkopouloi” as “Cu-
man or mixed Cuman-Mongol groups”.53 Although Shukurov provides no specific
argumentation for this interpretation, some sources seem to support it. Thus,
Pachymeres states that the “Tatar Koutzimpaxis” (τὸν Κουτζίμπαξιν Τόχαρον),
who had served as “chief magician” at Nogai’s court before entering Byzantine
service and being sent to Thrace to negotiate with the Alans and the Turcopoles,
was “of the same nation and the same language” (ὁμοεθνεῖ τε καὶ ὁμογλώσσῳ) as
the latter.54 Pachymeres’s last account of the Turcopoles concerns the defection of
their leader Taghachar (Ταχαγτζιάρις) from the Catalans in the region of Tzourou-
lou (Çorlu) some time in 1306–7.55 Taghachar is without doubt a Mongol/Tatar
name, a famous and near-contemporary namesake being the mighty commander
50 PACHYMERES/FAILLER, vol. 4, 626–7 (XIII, 4): Οὐ μὴν δὲ καὶ Ἀλανοῖς ἑτέρωθεν, ἔτι δὲ καὶ
τοῖς ἐξ ὑπογύου χριστιανοῖς Τουρκοπούλοις, οἳ δὴ καὶ οὐ πολλῷ πρότερον χρόνῳ ἐκ τῶν
βορείων βασιλεῖ προσεφοίτησαν... From this passage it remains unclear if the Alans and the
Tourkopouloi arrived from the “northern lands” together, since the “northern” provenance
of the former is undisputed and attested by Pachymeres elsewhere—they had migrated to
Byzantium shortly after Nogai’s death: see PACHYMERES/FAILLER, vol. 4, 336–9 (X, 16). MU-
TAFČIEV 1943, 84–5, n. 4 even assumes that the sentence may be syntactically imprecise and
that its latter part (i.e. the account of the arrival from the north) is solely related to the Alans.
51 Rūm-ili’nde Ulgār begleri khurūc idüp fāsilyūs üzerine müstevlī olup Rūm-ili’nüñ eks̲erin
almışlardı (YAZICIZĀDE ʿALĪ/BAKIR 2014, 444a; YAZICIZĀDE ʿALĪ/BAKIR 2017, 741). For the
Bulgarian-Byzantine war and Teodor Svetoslav’s conquests of 1304–5, see LAIOU 1972, 160–
1, 170–1; KRASTEV 2011, 62–73.
52 See n. 50 above.
53 SHUKUROV 2016, 93. Elsewhere (pp. 124–5, 131), he seems to accept the truthfulness of the
Dobruja story.
54 PACHYMERES/FAILLER, vol. 4, 378–9, 602–3 (X, 30; XII, 32). On Koutzimpaxis, or Kocabahşı,
see PLP, no. 13622; ZACHARIADOU 1978, 262–4; SHUKUROV 2016, 232–3. E. Zachariadou’s
interpretation that he was a Turk is hardly tenable in the light of his explicit description by
Pachymeres as a Tocharos, a term used by the Byzantines to denote the Mongols/Tatars: see
MORAVCSIK 1983, II, 329.
55 PACHYMÉRÈS/FAILLER, vol. 4, 694–7 (XIII, 29). Cf. WITTEK 1952, 665, n. 4; PLP, no. 27546;
MORAVCSIK 1983, II, 296.

and kingmaker of the Ilkhanate Taghachar Noyan (d. 1296). 56 Finally, in the Serbi-
an vita of Archbishop Danilo II (d. 1337) “Tatars” (tatari) are mentioned alongside
the Alans (iasi) among the numerous peoples who attacked Mount Athos in 1307–
11, including Franks (frugy), Turks (tur’ky), Almogavars (mogovari), and Catalans
(katalani), i.e. the Catalan Company and its Turkish allies; according to the same
source, numerous Tatars, Turks, and Alans (mnogyie voisky iezyka tatar’ska i tur’ska i
iash’ska) later became mercenaries of the Serbian king Stefan Uroš II Milutin (r.
1282–1321) in the war against his brother Dragutin. 57 It is difficult to identify these
Tatars with the Tourkopouloi considering the latter’s animosity towards the Alans as
well as other related evidence (see next paragraph), but it is clear that Tatars also
took part in the events in the early 14th-century Balkans. These were probably
contingents close to Nogai who fled to the south after his death, much like those
3,000 men who accompanied his grandson Kara Kesek (or Keshek) in the service of
Vidin’s ruler Shishman.58 Pachymeres may well have confounded them with the
Tourkopouloi when pointing to their “northern” provenance.
Gregoras and the Catalan chronicler and participant in the events Ramón Mun-
taner (d. 1336) are in general unanimous with regard to the actions of the Catalans
and their Turkic allies—including ca. 1,000 Turcopoles (Turcoples in the Catalan
text) under the leadership of a certain Melik 59—after their depredations in Thrace
in 1305–7. Together they roamed along the Aegean coast to reach Thessaly, where
the Anatolian Turks and the Turcopoles decided to abandon the Catalans in the
spring of 1311.60 According to Gregoras, the two Turkic groups under the respec-
tive leadership of Khalīl and Melik soon parted ways themselves. Melik and his
men (here, 1,500) submitted to the Serbian king Stefan Milutin and settled in Ser-
bia, while another 2,100 Turks under Khalīl returned to Thrace in order to find
56 See e.g. ATWOOD 2004, 525 (s.v. Ta’achar); HOPE 2016, passim. Other Mongol bearers of the
name (in the form Tokuchar) include a son-in-law of Chinggis Khan (r. 1206–27) and a
grandson of Ögedei Khan (r. 1229–41): see e.g. SHM/DE RACHEWILTZ, vol. 2, 940–1, 1031.
57 DANILO/DANIČIĆ, 341–2, 359; DANILO /MIRKOVIĆ, 259, 273; VÁSÁRY 2005, 108–10.
58 On Kara Kesek/Keshek and his men, see the Arabic sources in TIZENGAUZEN 1884, 94, 119,
140, 162. Cf. VÁSÁRY 2005, 97–8; UZELAC 2015, 246–7.

59 MUTAFČIEV 1943, 88 and WITTEK 1952, 665 have identified this Melik with ʿIzz al-Dīn’s son
Melik Konstantinos whom PACHYMÉRÈS/FAILLER, vol. 4, 674–5 (XIII, 22) mentions as a gov-
ernor of Pegai (Biga) around the same time; other researchers have opposed this identifica-
tion: see e.g. LAURENT 1956, 361, n. 4; SHUKUROV 2016, 190; PLP, no. 17761–2. Indeed, the
latter three publications show that the name/title Melik was adopted by many Turkish
noblemen and military leaders of the time.
60 GREGORAS/SCHOPEN, 244–8 (VII, 5–6); GREGORAS/VAN DIETEN, 190–2. MUNTANER/DE BO-
FARULL, 429–30, 436–58 (cap. CCXXVIII, CCXXXI–CCXLI, passim), MUNTANER/HUGHES, 109–

10, 119–50 (passim). Muntaner mentions the leader of the Turcoples with the name/title of
Melik (Milich) in the context of events datable to 1305: see MUNTANER/DE BOFARULL, 404
(cap. CCXV); MUNTANER/HUGHES, 73. For the dating, see JACOBY 1979, 233–4; OIKONOMIDES
1993, 161–2.
214 DELYAN RUSEV
passage back to Anatolia before being annihilated by the Byzantines and their al-
lies in Gallipoli in 1313.61 The Serbian vita of Danilo confirms both of these ac-
counts and adds that after entering Stefan Milutin’s service, Melik (Melekil) con-
spired against the king, whereupon he was captured and put to “violent death”. 62
A possibly related passage in the vita of Stefan Milutin himself, written by the
same Danilo, narrates how a group of “godless Persians” (bezbozh’nii per’si) entered
Milutin’s service but soon rebelled and were defeated, with some of them killed,
some exiled, and others enslaved.63 If this account is indeed related to Melik’s men,
i.e. the Turcopoles, the designation “Persians” rather precludes their identification
with the Tatars in Stefan Milutin’s service mentioned above. 64 Their “godlessness”
is also problematic with regard to their identification with the Turcopoles whom
both Pachymeres and Gregoras depict as Christians. However, we should keep in
mind both the hagiographic nature of the source, in which the “godless” adver-
saries of the Serbian king were utilized to underline his saintly image, and the like-
ly superficial Christianisation of the Turcopoles, if Pachymeres was right about its
recency. It is also possible that the ethnic and/or religious profile of the military
commanders sometimes differed from that of their soldiers, like in the case of the
Bulgarian Voysil, who was put in charge of the Alans and the Turcopoles in the
Battle of Apros.65 To be sure, the turbulent events of 1305–13 in the Balkans obvi-
ously involved so many different (but often similar) ethnic and religious groups,
and such a frequent change of allegiances, that even the contemporary observers
were prone to confusion.
61 GREGORAS/SCHOPEN, 254–8, 262–9 (VII, 8, 10); GREGORAS/VAN DIETEN, 195–7; 200–4. Mun-
taner does not mention the parting of the Turcoples and the Turks but confirms that the latter
were almost completely annihilated at Gallipoli when trying to cross the Straits back to
Anatolia: see MUNTANER/DE BOFARULL, 458 (cap. CCXLI); MUNTANER/HUGHES, 150–1. An-
other contemporary, the Venetian Marino Sanudo Torsello, notes that 1,800 cavalrymen inter
Turchos, turchopolos et mortatos left the Catalans in Thessaly (Blachia) in 1311 and the majority
of them were soon destroyed because they had separated from each other; later, some Turks
tried to rejoin the Catalans: see JACOBY 1979, 232; JACOBY 2015, 160. On the mortati/mourtatoi,
a term denoting an apostate from Islam (from Arabic murtadd) and applied to Christian
Turks, see BARTUSIS 1992, 276–8. What differentiated them from the Turcopoles is unclear.
62 DANILO/DANIČIĆ, 354; DANILO/MIRKOVIĆ, 269.
63 DANILO/DANIČIĆ, 143–4; DANILO/MIRKOVIĆ, 108–9.
64 Cf. UZELAC 2015, 252–5, who has even suggested that the Tatars in question formed the
personal guard of Milutin, which defeated Melik’s people in Serbia. According to Uzelac,
the same contingent numbering 2,000 Tatars was identical with the one sent against Khalīl’s
Turks in Gallipoli in 1313 at the request of Andronikos II as well as with the 2,000 “Cumans”
(in John Kantakouzenos’s words) whose return the Serbs demanded from the Byzantines in
1320. It should be noted, however, that the term “Persian” has also been used to denote the
Tatar governor of Akkerman in another Slavic hagiographical work of the early 15 th century:
see KONOVALOVA – RUSSEV 1994, 79–80.
65 PACHYMÉRÈS/FAILLER, vol. 4, 598–8 (XII, 32); MORAVCSIK 1983, II, 93; PLP, no. 2926.
Thus, even if we only pick out the contemporary evidence that seems to sup-
port Yazıcıoğlu ʿAlī’s story, namely Gregoras’s account linking the Turcopoles to
both ʿIzz al-Dīn Kaykāʾūs II and Khalīl as well as Pachymeres’s statement about
their recent arrival from the north, their identification with the Dobruja Turks still
poses problems, most notably with regard to their later fate. Yazıcıoğlu maintains
that they were able to cross back to Anatolia together with Khalīl Ece, while all
contemporary sources agree, with minor differences in detail, that Melik’s people,
i.e. the Turcopoles, ended up in Serbia and Khalīl’s Turks were almost entirely
exterminated in the region of Gallipoli. Paul Wittek has attempted to find a way
around these contradictions by suggesting that “Dobruja Turks would have joined
Khalīl rather during his first stay in Gallipoli, in 1307–8” (actually, 1305–7), and
that they were probably able to transfer their kinsfolk to Anatolia during this time
or during Khalīl’s later sojourn on the peninsula in 1310–11 (actually, 1311–13).66 In
fact, after the Battle of Apros the Alans routed the Turcopoles and seized a number
of their women and children, but Wittek assumes that the Turcopoles were soon
able to retrieve their families and send them to Anatolia, which would corroborate
Yazıcıoğlu’s version.67 While this is not impossible, there is no clear evidence
thereof, and it remains unclear why the Turcopoles (if identical with the Dobruja
Turks) would willingly separate from their families and would then not even try
to rejoin them in Anatolia in 1311 but go to Serbia instead. As a whole, Wittek’s
generally ingenious reconstruction of the events is tainted by a number of specula-
tions, ambiguities, and factual inaccuracies, and can thus remain hypothetical at best.
Finally, scholars involved in the debate surrounding the events of 1262–5 have
discussed the question of whether the Byzantine emperor could grant Dobruja as a
place for settlement of the Anatolian nomads, as reported by Yazıcıoğlu ʿAlī, or
not. It is noteworthy that in 1262–3 Bulgaria and Byzantium were at war, which led
to substantial Byzantine territorial gains in Thrace. Pachymeres’s explicit statement
that these conquests were limited “to the south of the Haemus” (i.e. the Balkan
Mountain Range) has given a reason for Petar Mutafchiev to conclude that the
territories to the north were at the time—as well as later—firmly under Bulgarian
control and not at the emperor’s disposal. 68 In response to this claim, Vitalien Lau-
rent provides some evidence of a Byzantine enclave in the Danube delta, which,
according to him, was formed shortly after the reconquest of Constantinople in
1261, more precisely in 1262–3. He bases this assumption on a panegyric for Mi-
66 WITTEK 1952, 665–6. Since Wittek could only rely on Pachymeres’s and Gregoras’s infor-
mation on the events, his chronology is largely inaccurate: cf. the chronology of JACOBY 1979,
230–4; JACOBY 2015, 150–6.
67 WITTEK 1952, 665–6. For the Alan captivity of the Turcopole families, see PACHYMÉ-
RÈS/FAILLER, vol. 4, 648–9.

68 PACHYMÉRÈS/FAILLER, vol. 1, 278–9 (III, 18); MUTAFČIEV 1943, 18–23. For a more detailed
study of the Bulgarian-Byzantine war of 1262–3, see PETROV 1956, 549–63; cf. BOZHILOV –
GYUZELEV 1999, 511–12.
216 DELYAN RUSEV
chael VIII from 1273, which, however, only mentions the submission of “numerous
Paristrian islands” to the emperor without providing a date.69 Laurent refrains
from taking a firm stance on the question of the nomadic migration to Dobruja, but
concludes that it could have indeed taken place via the Byzantine possessions on
the Danube.70 His conclusion is later endorsed by Paul Wittek who further argues,
rather arbitrarily, that Dobruja was then “nominally part of Bulgaria but in reality
more or less a no-man’s-land” used by the Tatars of the Golden Horde as a corri-
dor for their incursions in the Balkans. 71 As pointed out more recently by Georgi
Atanasov, it was precisely the Tatar dominance and interest in these territories that
makes the whole episode look unlikely. The fact that the Horde and the Bulgarian
Tsardom were at that time engaged in an anti-Byzantine alliance, as evident from
their joint campaign of 1264–5 that liberated ʿIzz al-Dīn Kaykāʾūs from Byzantine
captivity, adds further doubt to Michael VIII’s ability to establish his control over
settlements in the Danube delta and transfer a large group of nomads in their hin-
terland as early as 1262–3.72
Things changed in the early 1270s when Nogai consolidated his authority over
the area and married an illegitimate daughter of the emperor called Euphrosyne,
thus joining hands with the Byzantines. 73 It is theoretically possible to assume that
the resettlement of the Anatolian Turks took place at that juncture, but this would
mean to accept the key event of Yazıcıoğlu’s story as a fact while reshaping it from
the bottom up. The Ottoman historian actually dates the second appearance of the
nomads in the Balkans to after the death of ʿIzz al-Dīn Kaykāʾūs II in 678/1279–80,
when they requested to return “to their abode” (yurdlarına) in Dobruja—a rather
strange designation given that they had only spent around two years there in the
1260s, according to the same author. 74 Yazıcıoğlu once again states that they were
69 LAURENT 1945, 188.

70 LAURENT 1945, 198.
71 WITTEK 1952, 654.
72 ATANASOV 2009, 30, 413–14. For details regarding the Tatar-Bulgarian campaign of 1264–5,
see PAVLOV 1995, 123; PAVLOV 1997, 142–4; VÁSÁRY 2005, 74–7 (who dates it in 662/1263–4);
UZELAC 2017, 380.
73 PACHYMÉRÈS/FAILLER, vol. 1, 243 (III, 5). The dating of this marriage varies between 1272
(VÁSÁRY 2005, 79), 1272/3 (ATANASOV 2009, 30; UZELAC 2017, 380), and “ca. 1274” (PAVLOV
1995, 124). If a pre-1273 dating is accepted, it is possible that the reestablishment of Byzan-
tine presence in the Danube Delta actually ensued from this diplomatic act (cf. n. 69).
74 YAZICIZĀDE ʿALĪ/BAKIR 2014, 414a; YAZICIZĀDE ʿALĪ/BAKIR 2017, 698–9; WITTEK 1952, 649;
DECEI 1968, 89. According to Yazıcıoğlu ʿAlī, the nomads gained Berke’s permission to re-
turn to Dobruja. As this passage is located after the account of ʿIzz al-Dīn Kaykāʾūs’s death
(in 678/1279–80) and Berke died in 1266, the latter’s name is either used anachronistically or
as a common noun for the khan of the Golden Horde, cf. WITTEK 1952, 656–7, who neverthe-
less accepts the authenticity of the second migration. According to TOGAN 1981, 268–9, the
Turkmens’ return from Crimea to Dobruja took place in 1280 in the context of Nogai’s cam-
paign against the Byzantine towns of Anchialos and Mesembria on the Western Black Sea
led by Sarı Saltuk. As will be seen, there are enough grounds to accept that Saltuk
did indeed arrive in Northern Dobruja sometime in the 1270s–1280s and carried
out his missionary activities there in the context of the Tatar dominance in the area
under Nogai, a recent convert to Islam. The near-contemporary Arabic authors
who provide the main evidence thereof, however, are silent about the supposedly
large Anatolian following of the sheikh, which further undermines its historicity.
The historical Sarı Saltuk and Barak Baba: Arabic sources
The identities and activities of Sarı Saltuk and his disciple Barak Baba have found
some reflection in Arabic sources from the first half of the 14 th century. The earliest
among them is Tashwīq al-arwāḥ wa’l-qulūb ilā dhikr ʿallām al-ghuyūb, a Sufism-
related book with diverse contents composed between 703/1303–4 and 715/1315–
16 by Ibn al-Sarrāj al-Dimashqī (d. ca. 747/1347), who spent most of his career as a
judge in Mamluk Syria and Southeastern Anatolia. The most detailed account on
Saltuk is found in the third part of the work, which seems to have circulated sepa-
rately and survives only in two self-contained manuscripts under the title of Tuffāḥ
al-arwāḥ wa-miftāḥ al-arbāḥ.75 In Tuffāḥ, Ibn al-Sarrāj retells the miraculous deeds of
various Sufi saints one of whom is called Saltuq al-Turkī and is without doubt iden-
tical with Sarı Saltuk of the Ottoman tradition. It should be noted that the relevant
account was first made known to and almost exclusively utilized by modern schol-
arship through its shortened version in the similarly structured work of Yūsuf al-
Nabhānī (1842–1932), who has paraphrased Ibn al-Sarrāj’s information, leaving out
some of the anecdotes and important details. 76 The additional data on Saltuk con-
tained in parts of the Tashwīq other than the Tuffāḥ has also been largely ignored. 77
It is beyond the purpose of this paper to analyse the fabulous tales depicting
Saltuk as an ecstatic sheikh transcendentally involved in battles with the infidels,
coast, and the campaign was even incited by the Anatolian nomads under his authority.
This reconstruction of the events is untenable as it is based on a wrong dating of Nogai’s
incursion—it actually took place sometime between 1271 and 1273, before his rapproche-
ment with Byzantium and the death of ʿIzz al-Dīn: cf. PAVLOV 1995, 124; VÁSÁRY 2005, 71;
ATANASOV 2009, 30; UZELAC 2017, 380.
75 On Ibn al-Sarrāj and his works, see ÖZTÜRK 2013, 11–21; SARIKAYA et. al. 2013, 76–9. A
printed edition and a Turkish translation of Tuffāḥ’s account on Saltuk are provided in SARI-
KAYA et. al. 2013, 95–108. I have mainly relied on the Turkish rendering but have compared
the key passages discussed here to the Arabic text with the help of Prof. Pavel Pavlovitch,
for which I express my gratitude to him.
76 To my knowledge, the first modern researchers to mention this source were TOGAN 1981,
269–70, n. 385, and via that study, YÜCE 1987, 89, 99. A partial German and a more reliable
English translation of al-Nabhānī’s text were published by KIEL 2000, 261–5 (transl. by B.
Radtke) and NORRIS 2006, 58–61, respectively.
77 These accounts are summarized or quoted in Turkish translation in ÖZTÜRK 2013, 122–39
and SARIKAYA et. al. 2013.

218 DELYAN RUSEV
conversion to Islam, healing, and other lifetime and posthumous miracles. 78 It suf-
fices to point out that some of the characteristic features of his image in later narra-
tives composed in the Ottoman lands can already be found in this early source
either in embryo or fully developed: his ghazi vocation, his proselytizing ability,
his cult’s appeal to both Muslims and Christians, and his many graves. It is also
worth mentioning that Ibn al-Sarrāj’s main objective was to defend Saltuk’s pro-
vocative lifestyle and unorthodox spiritual practice from the accusations of the
contemporary Ulama who apparently saw them incompatible with the Sharia and
even likened the dervish to the Devil. What is of greater interest here, however, is
the factual information provided by Ibn al-Sarrāj. He cites as his main authority on
Saltuk the latter’s disciple Sayyid Bahrām Shāh al-Ḥaydarī whom the author claims
to have met twice in Behisni (Besni) and Damascus in 703–4/1304–5.79 The text
even suggests that in his youth Ibn al-Sarrāj had a personal acquaintance with Sal-
tuk in Damascus before the latter left for the “northern countries”.80 Considering
that the author’s youth can hardly be dated earlier than the 1270s or the 1280s, 81
this would mean that Saltuk moved to the region of the Northern and Northwest-
ern Black Sea only in the last decades of the 13 th century. This account invites some
suspicion, particularly with Ibn al-Sarrāj’s claim to have “advised” the significantly
older Saltuk, but if true, it would largely disprove Yazıcıoğlu’s story of the sheikh’s
migration to the Balkans with Sultan ʿIzz al-Dīn’s Turks in the 1260s.
In the Tashwīq, Saltuk is said to have settled within the walls of a town and to
have established a dervish lodge (zāwiya, Tur. zāviye) after a long vagrancy “in
mountains, vales, and deserts”.82 The town is more concretely specified in Tuffāḥ. It
bore the name of Ṣajī “in the Qïpchaq language” (bi-l-qifjāqiyya) and was located a
month’s journey—two thirds of it by sea—to the west of a big city called Aqḥāker-
mān.83 Despite the somewhat unfitting distance between the two sites, there can be
little doubt that the city of Akça Kerman/Akkerman (Bilhorod-Dnistrovskyi) at the
78 These aspects of his image have been thoroughly studied: see e.g., YÜCE 1987; OCAK 2016;
ALEKSIEV 2012; KARAMUSTAFA 2015.
79 SARIKAYA et. al. 2013, 95, 100, 103, 107.
80 SARIKAYA et. al. 2013, 100, 107. The Turkish translators have rendered the impersonal
expression “in the time of youth” (‫ )في زمن الشبیبة‬as referring to the youth of Saltuk, but this is
difficult to accept as he was apparently a generation older than Ibn al-Sarrāj (on Saltuk’s age
and time of death, see below).
81 Ibn al-Sarrāj’s time of birth is unknown, but he died ca. 747/1347 and the earliest infor-
mation about his career dates from the early 14th century: see ÖZTÜRK 2013, 12–13.
82 ÖZTÜRK 2013, 129.
83 SARIKAYA et. al. 2013, 100, 107. The 19 th-century MS Berlin, Staatsbibliothek, Wetzstein II
396, fol. 109b gives the alternative forms of Ṣajī and Ḍajī. Ibn al-Sarrāj interprets this “Qïp-
chaq” name with the Arabic term m-ḥ-r-n-y-n (‫)محرنین‬, but the latter’s meaning is unclear,
possibly due to an error in the existing manuscripts. The Turkish translators have rendered
it as “a cotton weaving place” (pamuk dokunan yer), but I could not confirm this meaning.
Dniester Estuary and the town of Sakçı (Isaccea) on the Lower Danube are meant
here.84 According to Ibn al-Sarrāj, Saltuk died in 697/1297–8 at the approximate
age of 70, and his tomb (turba, Tur. türbe) lied three hours away from Ṣajī. His large
zāviye with a capacity of 300 people was “still standing” at the time of writing—as
explicitly stated, some eighteen years after the sheikh’s passing—under the guid-
ance of Saltuk’s disciple Talak (Ṭalāq), whose “Qïpchaq” name Ibn al-Sarrāj trans-
lates in the Arabic as “white”.85
Machiel Kiel has suggested that at least two of the fabulous anecdotes in Tuffāḥ
featuring Saltuk also contain a “historical core”. One of them relates how the
sheikh miraculously released and converted to Islam a Christian merchant who
had been taken captive by “the Franks”. Indeed, this story may reflect a real inci-
dent between a member of the local Christian Orthodox population and the west-
ern, mostly Genoese colonists in the ports of the Danube Delta. According to the
second episode, Saltuk correctly prophesized that a ruler would appear seven
years after his death and ask for a valuable rosary that previously belonged to the
sheikh. Ignoring the warnings of Saltuk’s disciples, the ruler would take the rosary
and thus cause great troubles such as war and disorder. Machiel Kiel has linked
this account to the Bulgarian tsar Teodor Svetoslav (r. 1300–22) and the abovemen-
tioned war with Byzantium that he started in 1304, seven years after the date of
Saltuk’s death as reported by Ibn al-Sarrāj.86 Although this interpretation seems
appealing, there are a number of reasons to be cautious with it. First, the symbolic
meaning of the number seven needs no explanation and, as will be seen, other con-
temporary sources provide a different date for Saltuk’s death. Second, Tuffāḥ has
Saltuk involved in other similar stories, which bear no possible trace of real
events—such as the one relating how he prophesized the arrival of “one of the
rulers of Istanbul” five years after the sheikh’s death in order to claim his corpse
and bring it to the Byzantine capital as an object of veneration. 87 And third, the
Bulgarian-Byzantine war of 1304 took place to the south of the Balkan Mountains,
84 In the extant Arabic text, the puzzling expression gharbī al-qawm (‫ )غربي القوم‬is added next to
the name of Aqḥākermān/Akça Kerman. The editors of the text have translated it into Turk-
ish as “the population is western” (Halk Batılıdır). Albeit grammatically unconvincing, this
translation could make sense in the light of the abundant evidence of “western” (mostly
Genoese) presence in that city at the end of the 13 th and the beginning of the 14th centuries
(see e.g. KONOVALOVA – RUSSEV 1994, 75–82; GYUZELEV 2021, 428–33). However, the gram-
matical construction suggests it is more likely that a copyist’s error has confused the original
reading gharbī al-qiram, “to the west of Crimea”, which perfectly describes the location of
Akça Kerman. I owe this observation to Prof. Pavlovitch.
85 SARIKAYA et. al. 2013, 100–1, 107–8.
86 KIEL 2000, 265–6. For the two anecdotes, see KIEL 2000, 263–4 and SARIKAYA et. al. 2013, 98,
105–6.
87 SARIKAYA et. al. 2013, 97, 105. This story is absent in the translation published by M. Kiel.
220 DELYAN RUSEV
i.e. at a significant distance away from the presumable location of Saltuk’s activi-
ties and resting place near the Danube Delta.
Finally, Ibn al-Sarrāj provides some information about Saltuk’s most renowned
disciple, Barak Baba, both in the Tuffāḥ and elsewhere in the Tashwīq. The author
claims to have personally seen Barak, apparently without establishing personal
contact, as well as to have benefitted from the knowledge of one of the Baba’s de-
scendants. However, he reproduces little more than the type of fabulous tale about
the saint’s miraculous involvement in religious purification of the Muslim com-
munity in Syria and, anachronistically, in the 1250 Battle of Manṣūra against the
Crusaders.88 This information hardly allows any verification of Yazıcıoğlu ʿAlī’s
later account of Barak Baba other than the ex silentio argument that Ibn al-Sarrāj
mentions nothing about his alleged royal lineage.
Other Arabic sources, including the historical work of the contemporary al-
Birzālī (d. 739/1339), are much more specific. They report how Barak Baba im-
pressed Ghāzān Khan (r. 1295–1304) and received high esteem at the Ilkhanid court
in Sulṭāniyya, which he retained under the next Ilkhan Öljaitü (r. 1304–16). In
706/1306, he visited Damascus as a member of a Mongol delegation to the Mam-
lukes, accompanied by a number of his followers. It was apparently on this occa-
sion that Barak Baba was noticed by Arab authors. They provide detailed descrip-
tions of the dervishes who impressed the Damascene population with their eccen-
tric look and unorthodox behavior. 89 Al-Birzālī identifies their leader as a disciple
of Sarīq al-Qirīmī (i.e. “the Crimean”) who is said to have died in 692/1292–3.90
Despite the unrecognizable, likely corrupted, first name and the different year of
death as compared to the above-cited account by Ibn al-Sarrāj (i.e. 697/1297–8),
there is no doubt that Barak Baba’s spiritual link with Sarı Saltuk is meant here—a
link corroborated by other near-contemporary sources.91 Furthermore, al-Birzālī
also retells the story of how the dervish had eaten his sheikh’s vomit and had
therefore been called Barak, the “Qïpchaq” word for “dog”. 92
This is, however, where the parallels with the relevant account of Yazıcıoğlu
ʿAlī end. According to al-Birzālī and later authors, Barak Baba was a native of a
village near Tokat, his father being a local notable and his uncle—a renowned
88 ÖZTÜRK 2013, 139–44.

89 GÖLPINARLI 1936, 40–7; OCAK 1989, 106–9; OCAK 2002, 81–3.
90 GÖLPINARLI 1936, 39; OCAK 2002, 65; OCAK 2016, 91. Both authors refer to a MS of al-
Birzālī’s history kept at the Topkapı Palace Library, III. Ahmed, no. 2951, II, fol. 105b.
91 These include a verse by Yunus Emre (presenting himself as a member of the same spir-
itual genealogy) and a saying attributed to Barak himself: see Gölpınarlı 1936, 24, 47–8;
OCAK 2002, 78; OCAK 2016, 113–14. Some later Arabic authors such as Ibn Ḥajar and Ibn
Taghrībirdī attach the byname al-Qirīmī to Barak, too, probably due to his sojourn in Crimea
together with his sheikh: see OCAK 1989, 106; GÖLPINARLI 1936, 40. GÖLPINARLI 1936, 39–40
has tried to explain the corruption of Saltūq to Sarīq.
92 GÖLPINARLI 1936, 39.
scribe.93 Although some researchers are not eager to give preference to this account
over Yazıcıoğlu’s version, its greater chronological proximity to the events and its
less legendary nature both speak in its favour. Moreover, the contemporary author
al-Ṣafadī (d. 764/1363) and other sources report that the dervish was killed at the
age of forty during another mission to the ruler of Gīlān in 707/1307–8, whereupon
his remains were buried in Sulṭāniyya.94 Even if the reported age is approximate
and chosen for its sacred connotations, Barak Baba would not have been born
much earlier than 1267. As his alleged father ʿIzz al-Dīn fled from Byzantium in
1264–5, he could not have left his (presumable) still unborn son there as main-
tained by Yazıcıoğlu. All this evidence suggests that the latter invented the family
link between the Seljuk sultan and the renowned mystic in order to entangle the
story of Barak Baba (and hence Sarı Saltuk) in that of ʿIzz al-Dīn and the Anatolian
migrants to Dobruja.
Two long known and disputed accounts that may or may not have a relation to
the legacy of Sarı Saltuk in Northern Dobruja in the first decades after his death
belong to the Arabic authors Abū’l-Fidāʾ (d. 732/1331) and Ibn Baṭṭūṭa (d.
770/1368–9 or 779/1377–8). The geographic work Taqwīn al-Buldān, completed by
Abū’l-Fidāʾ in 721/1321, contains an entry on the town of Ṣaqjī “in Wallachia and
the lands [under the control?] of Constantinople” (min al-Awlāq wa bilād al-Qusṭan-
ṭīniya). Ṣaqjī is said to be a middle-sized town located on the southern bank of the
Danube, at a distance of ca. twenty days from Constantinople and ca. five days
from Aqjā Karmān, which itself was on the Black Sea, “in [the lands of] the Bulgari-
ans and the Turks” (min al-Bulghār wa’l-Turk). The population of Aqjā Karmān was
partly Muslim and partly Christian, while the majority of Ṣaqjī’s inhabitants were
Muslim.95 Once again, it is clear that the modern towns of Isaccea (Sakçı) and Bil-
horod-Dnistrovskyi (Akkerman) are meant here, and some scholars have interpret-
ed the attested Muslim presence in the former as a possible consequence of the
settlement of Anatolian nomads and the activities of Sarı Saltuk in the area as re-
ported by Yazıcıoğlu ʿAlī.96
It should be noted that the Turkic/Muslim names of the two towns have led
Petar Mutafchiev to doubt the authenticity of this account, suggesting that it was a
later interpolation to the original text of Abū’l-Fidāʾ. Alternatively, he proposed
that Sakçı could have been called after an otherwise unknown person by the name
of Isḥāq, possibly a Muslim Tatar warrior who died there in the second half of the
13th century.97 Mutafchiev’s hypotheses have been disproved by written and mate-
93 GÖLPINARLI 1936, 40 and n. 1; OCAK 1989, 105–6; OCAK 2002, 80; OCAK 2016, 115.
94 GÖLPINARLI 1936, 46–7; OCAK 1989, 107; OCAK 2002, 81–2.
95 ABŪ L-FIDĀʾ/REINAUD – DI SLANE, 212–13 (text) and ABŪ L-FIDĀʾ/REINAUD, 316–17 (French
translation).
96 See e.g. ALEXANDRESCU-DERSCA BULGARU 1978, 447; İNALCIK 1991; KIEL 2005, 285.
97 MUTAFČIEV 1943, 75–80.
222 DELYAN RUSEV
rial evidence that has emerged since his study was published in 1943. Albeit equal-
ly preserved in later manuscript copies, Ibn al-Sarrāj’s above-cited work on Saltuk
clearly testifies to the existence of the town names Sakçı (or Ṣajī in his rendering)
and Akça Kerman (Aqjā Karmān) at approximately the same time when Taqwīn al-
Buldān was completed. The contemporary Mamluk historian Baybars al-Manṣūrī
(d. 725/1325) also mentions Ṣaqjī (sic) to indicate the lands that Tokta (r. 1291–
1312), the khan of the Golden Horde, gave to his son Tukul Bugha after the death
of Nogai in 1299.98 Most notably, the name is attested in both Arabic (Sāqchī) and
Latin (the letters S/A/T/Y surrounding a cross) inscriptions on coins with Tatar
tamghas, which were issued there in the late 13th and early 14th centuries and
demonstrate the complex political and ethnic relations in the area. 99
As for the origin and etymology of the name Sakçı, Mutafchiev’s suggestion to
link it with the Muslim name Isḥāq cannot be entirely discarded, particularly in the
light of the 19th-century evidence about the existence of a türbe in the town attribut-
ed by the local population to a certain Isḥāq Baba.100 Much more probable, howev-
er, is the name’s derivation from the Qïpchaq word sakçı, which can be found (in
the composite form ṣakçılı yol) in a 14th-century Turkic–Arabic lexicon with the
meaning of “guard, warden” and perfectly fits the location of the town at the most
convenient place for crossing the Lower Danube in the Middle Ages. 101 The Mus-
lim or, more likely, Turkic/Tatar etymology of the name does not mean that Sakçı
was founded by the Tatars or the Anatolian settlers who, according to Yazıcıoğlu
ʿAlī, built “two-three Muslim towns” in Dobruja. Archaeological excavations have
shown that a medieval fortress existed there without interruption since the 10th or
11th century, on (or near) the site of the ancient Novioudunum. It probably bore the
original Bulgarian name of Obluchitsa, which is preserved in numerous sources
from the 15th–18th centuries parallelly to the “official” Ottoman name of Sakçı.102
Thus, the questions remain of when did the latter form appear and whether it had
98 TIZENGAUZEN 1884, 93 (text), 117 (Russian translation).

99 OBERLÄNDER-TÂRNOVEANU 1995/6, 199–200; OBERLÄNDER-TÂRNOVEANU 2004; For the in-
terpretation of this numismatic material, cf. ATANASOV 2009, 30–8.
100 MUTAFČIEV 1943, 78.
101 I have borrowed this etymology from SARIKAYA et. al. 2013, 81. For the relevant entry in
the lexicon, see İZBUDAK 1936, 38. The lexicon was written in the margins and between the
lines of a 735/1335 copy of the Kitāb al-ʾidrāk li-lisān al-ʾatrāk, another Turkic–Arab lexicon
and grammar composed in the early 14th century: see İZBUDAK 1936, 4, and ERMERS 1999, 24–
8, 41–3.
102 KUZEV 1981; MUTAFČIEV 1943, 79–80. Some researchers have also argued for the identity
between Sakçı and the town of Vicina, attested in numerous Latin and Greek sources from
the 13th–14th centuries: see e.g., KONOVALOVA – RUSSEV 1994, 82–9 and n. 66 there. If this
identification is correct, it would add an interesting facet to Saltuk’s presence there, since
Vicina was also a metropolitan seat under the Patriarchate of Constantinople (see
KONOVALOVA – RUSSEV 1994, 84–6).
any possible relation to the alleged settlement of Anatolian nomads in the area. The
likely Qïpchaq provenance of the name rather invites a negative answer to the
second question. The first one, on the other hand, cannot be definitively answered,
but a so far neglected detail in Abū’l-Fidāʾ’s account may give an idea. The above-
cited information he provides about Sakçı is perhaps based on contemporary in-
formers of his, but he has also indicated a specific written source for the town’s
coordinates: the Aṭwāl.103 This title refers to one of the main sources for his geo-
graphical tables, the anonymous and unfortunately unpreserved Kitāb al-Aṭwāl wa-
l-ʿurūḍ li-l-Furs, whose exact dating is unknown but certainly earlier than the mid-
13th century.104 If Kitāb al-Aṭwāl did indeed refer to the town with its Qïpchaq name
of Sakçı, as it seems probable, then this appellation appeared prior to the Mongol
invasion in Eastern Europe and should have been introduced by the Cumans who
regularly crossed the Danube since the late 11th century for plundering and in-
volvement in the Balkan military theatre. 105
Finally, Abū’l-Fidāʾ’s statement that the population of Sakçı was predominantly
Muslim around 1321 can hardly bear any relevance to the reported settlement of
Anatolian nomads in the area some sixty years earlier. If anything, it contradicts
Yazıcıoğlu ʿAlī’s account that the majority of these settlers migrated back to Anato-
lia and those who remained in Dobruja apostatized from Islam shortly after the
death of their leader Saltuk. Even if we accept that the Tourkopouloi from the Byz-
antine sources were indeed descendants of the same people, Pachymeres and
Gregoras are also unanimous that they had already accepted Christianity by the
early 14th century. Thus, Abū’l-Fidāʾ’s information about Sakçı is certainly related
to the town’s being an outpost of the Golden Horde on the southern bank of the
Danube and the likely presence of Muslim Tatars among its population. This situa-
tion is probably also the key to understanding the next piece of evidence about
Saltuk authored by Ibn Baṭṭūṭa.
Ibn Baṭṭūṭa’s famous travel-book (riḥla) contains a very puzzling description of
his journey from the Golden Horde to Constantinople via the Northeastern Bal-
kans, probably in 1334.106 Twenty-nine days after his departure from Astrakhan
with the large caravan of one of Khan Özbeg’s (r. 1313–41) wives, a Byzantine prin-
103 ABŪ L-FIDĀʾ/REINAUD – DI SLANE, 212.

104 See KING 1999, 42–3, 156–61 who offers a dating in the 12th or early 13th century. In the
mid-13th century, Kitāb al-Aṭwāl was also used by Naṣīr al-Dīn al-Ṭūsī (d. 672/1274) – see
KING 1999, 43–4.
105 The presence of the name Ṣaqjī in Kitāb al-Aṭwāl is even more likely because it appears in
two later Persian works, which seem to have used this source independently from Abu’l-
Fidāʾ: see the references in KENNEDY – KENNEDY 1987, 303. I could only consult one of these
works, the 16th-century Āʿīn-i Akbarī in H. Blochmann’s edition, where the town is given as
Ṣaqjī “near the Sea of Pontus” (qarīb-i baḥr-i Bunṭus): ʿALLĀMĪ 1877, 46.
106 For the chronology and the itinerary of this journey, see in most detail HRBEK 1962, 468–9,
473–83 who refutes the generally accepted dating in 1331–2.

224 DELYAN RUSEV
cess, and after an alleged detour to Sudak in Crimea that sounds unlikely as it
would have unnecessarily prolonged the trip, the traveller reports his arrival at a
town called Bābā Salṭūq, which he describes in the following way:
We came to the town known by the name of Bābā Salṭūq. Bābā in their lan-
guage has exactly the same meaning as among the Berbers [i.e. ‘father’], but
they pronounce the b more emphatically. They relate that this Salṭūq was an
ecstatic devotee, although things are told of him which are reproved by the
Divine Law. This town is the last of the towns possessed by the Turks, and
between it and the beginning of the territory of the Greeks is [a journey] of
eighteen days through an uninhabited waste, for eight days of which there is
no water.107
The caravan then proceeded through the wasteland and eighteen days later
reached the unidentified fortress of Mahtūlī located “at the beginning of the territo-
ry of the Greeks”. The remaining journey to Constantinople lasted twenty-two
days and passed through the ruins of a fortress named after the early Islamic hero
Maslama ibn ʿAbd al-Malik and lying “at the foot of a mountain beside a tumultu-
ous river called Iṣṭafīlī”; then came three large channels (sixteen days after Mahtūlī),
and, on the bank of the last one, another town called al-Fanīka.108 On his way back
to Astrakhan after more than a month spent in the Byzantine capital, Ibn Baṭṭūṭa
once again passed through Bābā Salṭūq, but he provides no further details about
the town in this context.109 Researchers have tried hard to make sense of this itiner-
ary and identify the obscure toponyms without being able to arrive at convincing
conclusions. Some have situated the town of Bābā Salṭūq in the Northern Black Sea
region, particularly in Crimea or near the lower Dnieper (without a specific identi-
fication), while others accept its identity with present-day Babadag in Northern
Dobruja proposed already by J. von Hammer-Purgstall.110 Since the latter hypothe-
sis has gained an increasing scholarly support in more recent studies, it deserves
some greater scrutiny. As dwelling on Ibn Baṭṭūṭa’s obscure itinerary can only lead
107 Translation from IBN BAṬṬŪṬA/GIBB, 499–500; Original text in IBN BAṬṬŪṬA/DEFRÉMERY –
SANGUINETTI, vol. 2, 416.
108 IBN BAṬṬŪṬA/DEFRÉMERY – SANGUINETTI, vol. 2, 417–24; IBN BAṬṬŪṬA/GIBB, 500–3.
109 IBN BAṬṬŪṬA/DEFRÉMERY – SANGUINETTI, vol. 2, 445; IBN BAṬṬŪṬA/GIBB, 514.
110 For a summary of the scholarly opinions up to the mid-20th century, see MUTAFČIEV 1943,
66, and HRBEK 1962, 479. Mutafchiev (1943, 67–9: in Crimea) and Hrbek (1962, 479: “some-
where west of Odessa”) are among the proponents of the town’s northern location, and so
are e.g. Gibb in IBN BAṬṬŪṬA/GIBB, 499, n. 310 and SMITH 1982, 216, n. 1. For Bābā Salṭūq’s
identification with Babadag see e.g. ALEXANDRESCU-DERSCA BULGARU 1978, 448–9;
KONOVALOVA – RUSSEV 1994, 98; KIEL 1978, 2005; SARIKAYA et. al. 2013, 82, 93; OCAK 2016, 59,
139. For some only partially convincing attempts to unriddle the other toponyms and the
general route of Ibn Baṭṭūṭa through the Eastern Balkans, see the notes to Gibb’s translation
(p. 500–3) and NORRIS 1994.
to another dead end, it makes sense to put his account in the context of the earlier
evidence discussed so far, some relevant sources from Ottoman times, and the
available archaeological data about Babadag and its region.
Bābā Salṭūq, Babadag, Isaccea and the archaeological evidence
The most obvious argument for the identification of Ibn Baṭṭūṭa’s town of Bābā
Salṭūq with Babadag is the still existing mausoleum (türbe) of Sarı Saltuk in that
town in modern Romania. The türbe was first examined systematically by Machiel
Kiel in the late 1960s and the 1970s. The simple construction of the present building
has led him to the assumption that it is “either of very ancient date, built immedi-
ately after the death of Sarı Saltuk, i.e. 1300, or it is a reconstruction from after the
time of the Russian invasions, when the lack of economic resources prevented the
construction of a building of greater quality”. Some logical deductions and archi-
tectural elements (such as the use of pendentives) have made him “more inclined
towards a later date, somewhere in the 18th century”, but unable to reach a firm
conclusion.111 Notwithstanding the dating of the building itself, Kiel fully accepted
the veracity of Yazıcıoğlu ʿAlī’s story of Sarı Saltuk’s appearance in Dobruja with
the Seljuk Turks of Sultan ʿIzz al-Dīn Kaykāʾūs and argued that Babadag is identi-
cal with the town of Bābā Salṭūq mentioned by Ibn Baṭṭūṭa as well as “the real place
where Sarı Saltık lived and worked”.112 The latter statement was mainly based on
Evliya Çelebi’s 17th-century account of the reconstruction of the sheikh’s türbe and
the erection of a large religious complex around it ordered by Sultan Bāyezīd II (r.
1481–1512) on his return from the victorious campaign in Moldavia in 1484. Kiel
has further argued for the credibility of Evliya Çelebi’s statement that Saltuk died
in Babadag by pointing to the two earlier but now lost (if they ever existed) sources
cited by the famous traveller: a book of the saint’s exploits (menāḳıb) written by the
15th-century Sufi author Yazıcızāde Meḥmed of Gelibolu and a Saltuknāme com-
posed by Kenʿān Pasha, a governor of Özi (Ochakov) in the 1630s, on the basis of
Yazıcızāde Meḥmed’s treatise and the otherwise unknown Fütūhāt-i Tohtamış Hān,
apparently a work on the conquests of the Golden Horde khan Tokhtamish (d.
1406).113
In two later studies, Machiel Kiel has drawn on the abovementioned work of
Ibn al-Sarrāj to substantiate his thesis that Sarı Saltuk was indeed buried in Ba-
111 KIEL 1978, 218–19.

112 KIEL 1978, 214–15.
113 KIEL 1978, 214–15; EVLIYA ÇELEBI/KAHRAMAN – DAĞLI, 190–4 (p. 192 for the source cita-
tions). The existence of these earlier sources and the nature of Evliya Çelebi’s account in
general will be discussed in more detail in my forthcoming study on the memory of Sarı
Saltuk in Ottoman writings. It suffices to say here that Evliya’s information should be treat-
ed with the utmost reservation. For his sources on Saltuk, see also ANETSHOFER 2012, 295–6.
226 DELYAN RUSEV
badag and this town was the one which Ibn Baṭṭūṭa attributed to Bābā Salṭūq.114
Kiel pretends to quote, but actually summarizes Ibn al-Sarrāj’s account of the
saint’s resting place via its rendering by Yūsuf al-Nabhānī (see above) as follows:
“Saltuk at-Turki was a dervish performing magnificent miracles. He lived in the
town of İsakçe in the Land of the Qıpçak, died in the year 697 (AD 1297/98) and
was buried near the mountain where he had his retreat, some distance away from
İsakçe. His followers erected a zawiye around his grave.” Kiel further clarifies that
al-Nabhānī (and likewise Ibn al-Sarrāj) actually indicated the distance between
Sakçı and Saltuk’s grave as “three hours”, but this is assumed by the Dutch scholar
to be “a slip of the pen for the correct three days”. He also claims that “the only
mountain in the wide surroundings of İsakçe is the ‘Mountain of the Baba’ at Ba-
badağ”.115 The conclusions regarding the location of Saltuk’s original tomb and
zāviye are self-evident: they were located at Babadag. The problem is that these
conclusions are based on M. Kiel’s wishful interpretation and manipulation of the
source text. Unfitting pieces of evidence cannot be branded as a “slip of the pen”
without any argumentation and, moreover, neither al-Nabhānī nor Ibn al-Sarrāj
locates the türbe in a mountain.116 Even if they had done so, the hills at Babadag (ca.
60 km away from Isaccea) actually form part of a low mountain range crossing the
whole of Northern Dobruja, another part of which (with an even greater altitude
than that at Babadag) is located in the immediate hinterland of Isaccea, the so-
called Niculițel Hills (Dealurile Niculițelului).117 All of this makes Machiel Kiel’s
conclusions rather unconvincing.
114 KIEL 2000, KIEL 2005. I pay special attention to these studies because later research has
mainly used Ibn Sarrāj’s evidence on Saltuk through their mediation.
115 All quoted passages are from KIEL 2005, 286. In the German translation of al-Nabhānī’s
text provided by Kiel the distance between Isaccea and Saltuk’s resting place is even given
as “three days” without any notice that the original text actually reads “three hours”: KIEL
2000, 264. Furthermore, in both of Kiel’s publications the Arabic expression ba-l-qifjāqiyya
(“in the Qïpchaq [language]”) is wrongly rendered as “in the Land of the Qıpçak” (or “in
Kiptschakien”). For a correct English translation of the passage, see NORRIS 2006, 60–1. For
the Arabic original of Ibn al-Sarrāj, see SARIKAYA et. al. 2013, 107.
116 The only reason for such a hypothesis can be one anecdote, according to which “when he
was seated upon the prayer rug of his authority (sajjāda), after a long period of long resi-
dence and contemplation in the mountains [ba-l-jibāl], and within his isolated retreat, [Sal-
tuk] was visited by a certain person. The Shaykh spoke to him, saying, “Cast your mind
back to that time when you came to me in a mountain named such and such, and when in a
lowly state.” (transl. by NORRIS 2006, 60, from al-Nabhānī, whose text here closely follows
that of Ibn al-Sarrāj: cf. SARIKAYA et. al. 2013, 98, 106). As pointed out by Norris (2006, 143,
nn. 23–4), “the ‘mount’ (dağ) at Babadag cannot be excluded”, but “this story may relate to
Sari Saltik’s life at some distance from the Dobrudja region, possibly in Anatolia, or in the
southern part of the Crimean peninsula”.
117 This fact has already been pointed out by ALEKSIEV 2012, 21, who was misled by M. Kiel’s
introduction of a mountain in Ibn al-Sarrāj’s account.

In the 1990s, archaeological excavations were carried out on two sites in Ba-
badag: around the present-day türbe of Sarı Saltuk and in the courtyard of the ʿAlī
Ghazi Pasha Mosque built in 1610 and located ca. 200 m away from the türbe, on
the opposite bank of the brook passing through the town. The earliest structures
were located at the latter site and dated to the late 10 th and the first half of the 11th
century on the basis of the deposited ceramics. 118 This layer has been significantly
damaged by the foundations of a much later monumental building, which existed
next to the mosque until the early 20 th century and was contemporary with it. The
building was apparently constructed as a part of the same early 17th-century pious
foundation and has been identified with one of the eight inns seen by Evliya Çelebi
when he visited Babadag several decades later.119 Meanwhile, the excavations
around the türbe of Sarı Saltuk have confirmed the late emergence of the present
building, probably after the town’s desolation in the Russo-Ottoman War of 1828–
9, but have also revealed the remains of two earlier phases of the tomb’s existence.
One was contemporary to the ʿAlī Ghazi Pasha Mosque and the adjacent inn, i.e.
datable to the 17th–18th centuries, while the earliest level marked the erection of the
original türbe and the religious complex around it by Sultan Bāyezīd II in the late
15th century. No earlier traces of occupation of the site have been found on that
side of the brook.120 This contradicts Evliya Çelebi’s legendary story of how
Bāyezīd, after a miraculous dream on his way through Babadag, unearthed a mar-
ble sarcophagus with the name of Sarı Saltuk written on it “in the Tatar script” and
built there the new mausoleum.121
Thus, although the archaeological evidence is limited, it suggests that a Byzan-
tine settlement existed on the territory of present-day Babadag from the late 10th to
the mid-11th century, when it seems to have been destroyed by the nomad incur-
sions from the north.122 It was only four centuries later that Bāyezīd II revived it by
building a pious complex devoted to Sarı Saltuk on bare soil. A new settlement
grew around it and soon expanded towards the site of the old one, on the other
side of the nearby brook, where a new commercial centre of the town emerged in
the early 17th century around the newly built mosque of ʿAlī Pasha.
The limited scope of the archaeological excavations leaves the possibility open
that the original late 13th-century tomb and zāviye of Sarı Saltuk were still located
somewhere in the vicinity of present-day Babadag, but this can only remain a hy-
pothesis. The specific ideological motivation behind Bāyezīd II’s “reconstruction”
118 VASILIU 1996a.

119 VASILIU 1996b, and especially p. 206 for the identification; EVLIYA ÇELEBI/KAHRAMAN –
DAĞLI, 192.
120 IOSIPESCU – IOSIPESCU 2004, 321–2. I am grateful to Dr. Aurel-Daniel Stănică for drawing
my attention to that study as well as to some of the other archaeological materials used in
this section.
121 EVLIYA ÇELEBI/KAHRAMAN – DAĞLI, 193.
122 VASILIU 1996a, 178.
228 DELYAN RUSEV
of the mausoleum will be discussed elsewhere, but the “rediscovery” of sacred

places related to earlier Muslim presence in Rūm was a characteristic feature of the
Ottoman expansion. It suffices to remind of the equally miraculous discovery of
the tomb of Abū Ayyūb al-Anṣārī after the conquest of Constantinople by Meḥmed
II (r. 1444–6, 1451–81), an act that his son Bāyezīd was surely eager to replicate in
the context of his own victorious campaign in Moldavia in 1484. 123 As for the
choice of location and its possible relation to the name of Babadag (if it had existed
earlier as claimed by Evliya Çelebi), one should keep in mind that the tomb of an-
other “Baba” is still to be found, even more fittingly, on the hill overseeing the
town—the tomb of Koyun Baba. As things stand, it is impossible to find out when
this shrine was established and if it is dedicated to an otherwise unknown local
saint or the more famous Koyun Baba (d. 873/1468) who was buried in the Anato-
lian town of Osmancık but, like Sarı Saltuk, is venerated in other places, too. 124
While the archaeological evidence is insufficient to confirm the original location
of Sarı Saltuk’s tomb, it is much more conclusive regarding the non-existence of
any notable settlement at Babadag between the mid-11th and the late 15th centuries.
This once again brings us to the question of the identification of Ibn Baṭṭūṭa’s town
of Bābā Salṭūq, if we put aside the unverifiable hypothesis about its being a now
non-existent settlement somewhere in the steppe. The traveller’s statement that it
was “the last of the towns possessed by the Turks”, as he calls the Tatars through-
out the narrative of his journey in the Golden Horde, 125 coupled with the incontest-
able evidence about Tatar sovereignty over Sakçı (Isaccea) and Saltuk’s residence
there in the late 13th century, makes that town the most suitable candidate for the
Tatar town of Bābā Salṭūq. Indeed, the generally confused itinerary of Ibn Baṭṭūṭa’s
travel through the Eastern Balkans and the unrecognizable place names that he
provides suggest that he was writing out of memory some time after the journey
itself. He must have simply reproduced what he remembered about the town in
question, i.e. its links to the locally famous saint Saltuk. The latter’s name may
123 Cf. ALEKSIEV 2012, 62; VEINSTEIN 2005, 516–17 and passim.
124 On this saint and his cult, see OCAK 1999, 94–6; ŞAHIN 2002; DOĞANBAŞ 2015. Koyun Ba-
ba’s velāyetnāme mentions nothing about Babadag or Dobruja, but the 15 th-century vita of his
contemporary Otman Baba (d. 1478) retells how Koyun Baba miraculously helped the latter
cross the sea at Terkoz and thus it suggests Koyun Baba’s presence in the Balkans:
OBV/KILIÇ, 22. (Perhaps by chance, the next episode features Otman Baba as a reincarnation
of Sarı Saltuk: OBV/KILIÇ, 23.) Shrines by the name of Koyun Baba can also be found in a
number of other places in Anatolia as well as in Edirne and near present-day Tetovo in
Northern Macedonia: see GÜREL 2019.
125 IBN BAṬṬŪṬA/DEFRÉMERY – SANGUINETTI, vol. 2, 363–4, 372, 375, 379, 484, 489, 496, etc.; IBN
BAṬṬŪṬA/GIBB, 473–4, 478–9, 481, 484, 489, 496, etc.

have even been informally attached to the settlement by the traveller’s Ta-
tar/Muslim informers.126
A relevant, albeit later, source for the location of Sarı Saltuk’s activities in
Northern Dobruja is the Saltuknāme, a book of the saint’s miraculous exploits, orig-
inally composed by Abu’l-Khayr-i Rūmī between 1473 and 1480 for the Ottoman
prince Cem (d. 1495). The earliest dated manuscript copies were made in 985/1577
and 1000/1591, respectively, and scholars have noted that the now-available text of
the work was reshaped in the 16th century to accommodate a number of allusions
to the Sunni-Shia strife between the Ottomans and the Safavids after the rise of
Shah Ismāʿīl (r. 1501–24).127 Much work remains to be done in order to disentangle
the numerous historical, hagiographic, epic, and mythological layers of this volu-
minous source as well as to establish their origin, but it is relatively safe to believe
Abu’l-Khayr-i Rūmī’s own statement that he compiled much of the information on
the basis of oral accounts circulating in Ottoman Rumelia in the 1470s—notably,
before Bāyezīd II gave rise to the Ottoman town of Babadag in the next decade.128
Although the narrative traces Saltuk’s encounters with people—mostly to-be-
Islamized “unbelievers”—and supernatural creatures in various parts of the then-
known world and beyond, Crimea and the Balkans stand out as his most traversed
areas, with several settlements (e.g. Adrianople and Kaffa) more specifically asso-
ciated with his exploits. One of them is a place called Tūnā Baba, i.e. “[the town of
the] Baba on the Danube”, which some researchers tend to identify with Ba-
badag.129 Theoretically, this identification is not impossible considering Babadag’s
relative proximity to the Danube (ca. 35 km), but a closer reading of the text once
again makes present-day Isaccea the better fit. First of all, Tūnā Baba is where
Prince Cem reportedly arrived after hunting down a monstrous wolf and, hearing
about Sarı Saltuk’s glorious deeds from the sheikh’s followers (mürīdler) living
126 A similar case can be found in the same itinerary, where Ibn Baṭṭūṭa mentions a “fortress
of Maslama ibn ʿAbd al-Malik”, somewhere in the Southern Balkans, “at the foot of a moun-
tain beside a tumultuous river called Iṣṭafīlī”: see IBN BAṬṬŪṬA/DEFRÉMERY – SANGUINETTI,
vol. 2, 419; IBN BAṬṬŪṬA/GIBB, 501. The name of the river and some other considerations
have led H. NORRIS 1994, 218–9 to reasonably identify the place with the town referred to by
the 12th-century geographer al-Idrīsī as Istilīfinūs—an Arabicized form of the Byzantine
name (ἡ Στίλβνος) of present-day Sliven in Bulgaria: cf. NEDKOV 1960, 87, 105, 144. Whatever
the identification of the town, it is impossible that it bore the Muslim hero’s name in the first
half of the 14th century, when the region was firmly under Christian control.
127 MÉLIKOFF 1994, 232–3; KARAMUSTAFA 2015, 363. On the MSS, see YÜCE 1987, 5–15; SMITH
1982, 217; and the forewords to the 3-volume edition cited below. I have also consulted a
copy of MS Hazine 1612 from the Topkapı Palace Library in Istanbul.
128 EBÜ’L-HAYR-I RÛMÎ/AKALIN, vol. 3, 365–6; Hazine 1612, fol. 616a. For a discussion of the
historical layers in the Saltuknāme, see KÖPRÜLÜ 1992, 43–51. Unfortunately, M. F. Köprülü’s
announced monograph on the topic was never published.
129 See e.g. ALEXANDRESCU-DERSCA BULGARU 1978, 448, n. 27; OCAK 2016, 86, 95–6, 139. A. Y.
Ocak strangely seems to equate Babadag with Isaccea or at least with Ibn al-Sarrāj’s Ṣajī.
230 DELYAN RUSEV
there, commissioned Abu’l-Khayr-i Rūmī to collect and write down all the stories
(menāḳıb) about him.130 The author was apparently a man of Cem’s entourage and
thus should have had first-hand knowledge of the place in question, if we accept
the veracity of his account. On several occasions in the text, he refers to the town as
being “on the banks of the Danube” (Tūnā kenārında), while other episodes also
suggest its immediate juxtaposition to the river. 131 Most telling is the story about
how it acquired its name: On his way from the Balkans to Kaffa in Crimea with a
thousand men, Saltuk charged one of them with the task to hurry forward and find
a place for an “easy” (āsān) crossing of the Danube. On “this” (i.e. the southern)
side of that place, there was a fortress whose ruler welcomed the sheikh and of-
fered him his allegiance. His name was Pāpā Dīmītrī, but Saltuk called him Ib-
rāhīm and left three hundred of his men with him. The town itself acquired the
name of Baba and the local church was turned into a masjid where the Friday
prayer was held.132 This description leaves little doubt as to the identity between
the town of Tūnā Baba and Sakçı, with the latter located at one of the most suitable
fords on the Lower Danube in the Middle Ages. If any further argumentation is
needed, it ought to be noted that no possible form of the name Sakçı is encountered
throughout the text, i.e. there is no reason to assume that there was a place of that
name different from Tūnā Baba. As a side note, the account of three hundred men
left there by Saltuk is the closest to a possible Turkic/Muslim colonisation in the
area to be reported in the Saltuknāme—an account that hardly corroborates
Yazıcıoğlu ʿAlī’s story of a mass migration of Anatolian nomads. To be sure, ex-
tracting any reliable historical information from a work of that kind is a risky task
requiring a systematic analysis of the text, which is impossible to conduct here.
What is obvious, however, is that in Abu’l-Khayr-i Rūmī’s terms Tūnā Baba was
identical with Sakçı and not with Babadag.
The evidence discussed so far suggests that until the late 15 th century there was
no town in Dobruja associated with Sarı Saltuk other than Sakçı/Isaccea. What
was, then, the location of his original tomb, built at a distance of “three hours”
away from that settlement according to the contemporary Ibn al-Sarrāj? Even if we
take this account as an approximation, it is reasonable to locate the türbe in the
close vicinity of the Danubian town. In fact, it was long before Ibn al-Sarrāj’s narra-
tive became known to modern scholarship that Petar Mutafchiev drew attention to
the frequent identification of Sarı Saltuk with the Christian saint Nicholas in a
number of Ottoman-era sources and holy sites (especially in the Eastern Balkans),
on the one hand, and the fact that the present-day village of Niculiţel, ten kilome-
tres to the south of Isaccea, probably drew its name from a now non-existent mon-
astery devoted the St Nicholas and attested in 16 th- and 17th-centrury sources (as a
130 EBÜ’L-HAYR-I RÛMÎ/AKALIN, vol. 3, 365–6; Hazine 1612, fol. 616a.

131 EBÜ’L-HAYR-I RÛMÎ/AKALIN, vol. 2, 58, 62, 68, 186; EBÜ’L-HAYR-I RÛMÎ/AKALIN, vol. 3, 266.
132 EBÜ’L-HAYR-I RÛMÎ/AKALIN, vol. 1, 153–4; Hazine 1612, fol. 95a.
place called “Monastery” only), on the other.133 Later archaeological research at the
nearby site of an early medieval earthen rampart has revealed the remains of a
monastic complex datable, according to Petre Diaconu, to the second half of the
12th century.134 A reconsideration of the numismatic finds has led Ernest Oberlän-
der-Târnoveanu to shift the dating to the first half of the 13 th century and to sug-
gest that the destruction of the complex, and the small trefoil-plan church in par-
ticular, was caused by a late Cuman incursion or the Mongol invasion of 1241.
However, Oberländer-Târnoveanu also points to a Golden Horde coin “from the
13th–14th centuries” demonstrating that the place was still inhabited some time after
the disappearance of the Christian monastery.135 This scarce evidence, coupled
with the widespread practice of Sufi shrines appropriating former Christian sanc-
tuaries in the Balkans—a practice particularly pronounced in the cult of Sarı Sal-
tuk—suggests that the area of Niculiţel was indeed a good candidate to house the
sheikh’s tomb and the dervish lodge founded around it. 136 Needless to say, the
proposed location cannot but remain hypothetical until the discovery of further
written or archaeological evidence.
The consequences and the (lacking) evidence
Considering the shaky grounds of Yazıcıoğlu ʿAlī’s story of a largescale migration
of Anatolian Turks to 13th-century Dobruja demonstrated so far, it is unnecessary
to enter here into a detailed discussion of that migration’s supposed consequences.
A note is due, however, on some theses circulating in scholarship without suffi-
cient argumentation, such as those in the opening quotation of this paper. Most
striking in this regard is the representation of the 14 th-century Principality of Do-
bruja as a state formation of the Anatolian migrants from the previous century. 137
To begin with, it should be highlighted that there is no direct evidence regarding
133 MUTAFČIEV 1943, 54–9; On the identification of Sarı Saltuk with St Nicolas see also HAS-
LUCK 1929, vol. 1, 51, 54–6; vol. 2, 429–34, 576–9; ROHDEWALD 2017, 85–8; OCAK 2016, 143–6;
MUTAFOVA 2008.
134 DIACONU 1972, 312–15 and passim.
135 OBERLÄNDER-TÂRNOVEANU 1980.
136 The possible relation of Niculiţel to the activities of Sarı Saltuk has also been hinted at, on
the basis of Ibn al-Sarrāj’s account, by COSTAN 2014, 109, n. 29; COSTAN 2016, 22–3. G. Cos-
tan’s (mostly mediated) use of the sources is at times questionable, but his idea to search for
some connection between the still-standing 14th-century church of St Athanasius in the vil-
lage of Niculiţel itself and the reintroduction of Christianity in the area is noteworthy, albeit
perhaps far-fetched (see COSTAN 2014). For the church of St Athanasius, cf. MOISESCU 2000.
137 This thesis was first developed by the Bulgarian historian G. BALASCHEV (1930) and did
later enter some standard reference works such as TDV İslâm Ansiklopedisi (KARPAT 1994b,
1996) and The Encyclopaedia of Islam (ZAJĄCZKOWSKI 1991). In the latter, it acquired absurd
dimensions in denial of all available sources: according to W. Zajączkowski, the principality
was founded by ʿIzz al-Dīn Kaykāʾūs II himself in the 13th century. See also KIEL 1994, 166–7;
KIEL 2009, 141.
232 DELYAN RUSEV
the origin of the short-lived dynasty that ruled the principality. Its first three
known representatives, the brothers Balik, Dobrotitsa, and Theodor, first appear in
the sources in the mid-1340s. In 1346, the latter two were sent by “Balik, the archon
of Karvona” (the first capital of the principality to the north of Varna) in support of
the Byzantine regency led by the Empress Anna of Savoy in the internecine war
against John Kantakouzenos, who actually provides the evidence. 138 Attempts to
establish the origin of the brothers have been primarily based on their names as
well as on those of Dobrotitsa’s son/s Ivanko and Terter—probably denoting one
and the same person139—and the notable Cholpan (Jolpani), who is only attested
alongside a certain Kosta (Costa) as a diplomatic representative of Ivanko (Iuan-
chus) in his 1387 treaty with Genoa.140 The available onomastic evidence is in fact
quite conclusive that the dynasty and some part of the ruling elite in the principali-
ty had a Cuman origin. The most compelling proof thereof is the rare name of Ter-
ter141, which is identical with that of the Terterids who ruled as Bulgarian tsars in
the late 13th and the early 14th centuries. The originator of that dynasty, George I
Terter (r. 1280–92), was explicitly referred to by Pachymeres as being “from the
Cumans” (Τερτερῆς ἐκ Κομάνων ἦν), and was probably related to the Cuman clan
of Terter-oba, one of whose leaders is attested in a Rus’ chronicle among the partic-
ipants in the battle of Kayala River (somewhere in present-day Eastern Ukraine) in
1185.142 Balik and Cholpan are more common Turkic names, which may happen to
appear in an Anatolian/Oghuz context as well, but onomastic compendia show
that they were particularly widespread in Eastern Europe (especially Moldavia),
Central Asia, and, in the case of Cholpan, in the Mamluk lands—all places with
considerable Cuman or Tatar presence in the period under consideration here. 143
Furthermore, on the basis of the Slavic/Bulgarian name of Dobrotitsa and the
Christian/Greek names of Theodor and Ivanko—the latter in a Bugarianized di-
138 KANTAKOUZENOS/SCHOPEN, 584; KANTAKOUZENOS/FATOUROS – KRISCHER, 389–90; ATA-

NASOV 2009, 90–1.
139 See BILARSKY 1992, 18–21.
140 Among the notable studies on this matter, see MUTAFČIEV 1927, 1931; IORGA 1928; BILIAR-
SKY 1992, 6–7; DIACONU 1994; ATANASOV 2009, 26, 82–3, 112–13, 183–4, 421–2 (and other
works cited in n. 1 on pp. 9–10 there). For the text of the 1387 treaty, see GYUZELEV 1995, 127–132.
141 This name is attested in a Byzantine short chronicle (no. 22 in SCHREINER 1975, p. 182/19)
and some coins (see BILIARSKY 1992, 3–6; ATANASOV 2009, 291–5 and the works cited there).
142 PACHYMERES/FAILLER, vol. 3, 290–1 (IX, 26); PSRL, vol. 2, c. 641. See also RÁSONYI 2006,
247; GOLDEN 1995–7, 119; MORAVCSIK 1983, II, 306–7. I would like to thank Dr. Konstantin
Golev for his useful observations on that matter.
143 See RÁSONYI – BASKI 2007, vol. 1, 117, 208; RÁSONYI 2006, 174, 198–9. On the etymological
evolution of the Turkic word çolpan with some further examples of its usage as a name, see
ŞIRIN USER 2014. Since the 14th–15th centuries, Tatar became a common designation for the
heterogeneous Turkified population of the western steppes and the Northern Black Sea
region, demographically dominated by the remnants of the Cuman-Qïpchak tribes: see e.g.
GOLDEN 1992, 297–302, 317, 388–9.
minutive form—it is safe to conclude that they were all descendants of a Cuman
family well integrated in the Bulgarian elite of the time by means of cross-
marriages, very much like the royal dynasties of the Terterids and the Shishmanids
(1323–96) who may have been their kinsmen. 144
Two more vague arguments of ecclesiastic and archaeological nature, respec-
tively, have been put forward to substantiate the alleged link between the Princi-
pality of Dobruja and the 13th-century migration. The first of them, most thorough-
ly developed by G. Balaschev, is based on the fact that for some time in the 1320s
and, ultimately, since 1340 the territories that came to be consolidated under the
control of the principality and the metropolitan seat of Varna and Karvuna were
subjected to the Patriarchate of Constantinople. 145 While the secession from the
Bulgarian Church has rightfully been interpreted in scholarship as an indication
for the chronology of the principality’s formation,146 it hardly gives any clues about
the dynasty’s origin or allegiance. This process was typical for the separatist
tendencies that dominated political life in the Balkans in the 14 th century and has
its most obvious parallel in the Tsardom of Vidin, whose ruler Ivan Sratsimir (r. ca.
1356–96), a brother of the Bulgarian tsar in Tarnovo Ivan Shishman (r. 1371–95),
also subjected the local church hierarchy to Constantinople. 147
The dubious archaeological evidence consists of two stone plates with reliefs
that once decorated the entrance to the fortress of Kaliakra, which served as the
principality’s capital under its second ruler Dobrotitsa and housed a zāviye associ-
ated with Sarı Saltuk in Ottoman times.148 One of the reliefs depicts a fight between
two riders, while the other one has a horseman holding a mace or a sceptre with a
bull head on the top. Their discoverer Maria Yosifova has seen in them a Seljuk
tradition on the basis of some partial parallels with 12 th–14th-century stone reliefs
from the Caucasian town of Kubachi (Dagestan) and 13th–14th-century engravings
144 The name of Dobrotitsa’s son Terter suggests a probable link to the Terterid dynasty,
while the location of the principality also allows to hypothesize the kinship between Balik’s
family and that of the Shishmanid tsar of Bulgaria Ivan Alexander (r. 1331–71). The latter’s
father was half-Cuman, half-Bulgarian, while his mother is attested in a 1337 letter by Pope
Benedict XII (1334–42) as “ducisse Carnonen”, which some researchers have rendered as
“duchess of Karvona”, the later domain of Balik: see DUYCHEV 1937; ATANASOV 2009, 78–83.
145 BALASCHEV 1930, 15–17.
146 See e.g. ATANASOV 2009, 67–72, 76–8.
147 BOZHILOV – GYUZELEV 1999, 651.
148 For the zāviye, see HASLUCK 1929, vol. 1, 51; vol. 2, 578; KIEL 2000, 281–2; DIMITROV 1994,
76–88; OCAK 2016, 142–3. It is attested since the 16th century, but 15th-century sources such as
the Saltuknāme and the velayetnāme of Hacı Bektaş also feature Kaliakra (Keligra) as a place
where Sarı Saltuk performed some of his miracles: see EBÜ’L-HAYR-I RÛMÎ 1988, II, 31–2;
GÖLPINARLI 1990, 46. See also YÜCE 1987, 33–4, 197; ALEKSIEV 2012, 43–9, 55, 58; OCAK 2016,
70–5.
234 DELYAN RUSEV
on metal objects from Mosul (Iraq).149 These locations, which can be deemed pe-
ripheral to the Rūm Seljuk cultural sphere at best, as well as the lack of identified
parallels from Anatolia proper, both speak against any relation to the supposed
Turkmen migration to Dobruja. Moreover, Georgi Atanasov has convincingly ar-
gued for a significantly later, Ottoman-era dating of the plates. 150 The lack of any
other archaeological material that can be related to pre-Ottoman settlers with Ana-
tolian, let alone Seljuk, background in the generally well-studied medieval settle-
ments along the Danube and Black Sea coasts of Dobruja is also telling.
Despite the absence of the 13th-century Anatolian migrants from the archaeolog-
ical map of the region, a number of researchers have argued that they had a lasting
impact on its ethnodemographic setting by presenting them as forefathers of the
Gagauz—a Turkic-speaking Christian population that was mainly based in Dobru-
ja until the late 18th and the early 19th centuries, when a large share of it migrated to
Bessarabia and Southern Ukraine in the context of the Russian-Ottoman wars.151
Besides Yazıcıoğlu ʿAlī’s statement that some part of Sarı Saltuk’s men remained in
Dobruja after his death and apostatized, i.e. became Christians, the main argu-
ments underlying this theory rest on the name “Gagauz” itself, which some see as
a derivative of “Kaykāʾūs”, as well as on some linguistic features of the Gagauz
language. The etymology of “Gagauz” is an object of numerous alternative inter-
pretations that cannot be dealt with here, but it ought to be noted that the name
appears in no pre-19th century source.152 This is surely not to say that it did not
149 YOSIFOVA 1978.

150 ATANASOV 2013. His main arguments are the continued use of the Kaliakra fortress until
the 18th century; the stylistic parallels with reliefs from other Ottoman fortifications in the
region such as Silistra and Hârșova; and the strong resemblance between the depiction of
the sceptre-holding horseman and miniatures from a 16th-century copy of Firdausī’s
Shāhnāma as well as with a number of similarly dated physical sceptres with bull heads from
museum collections in the Middle East. With respect to the last argument, it could be added
that the Persianate miniature art of the Shāhnāma style, and copies of the Shāhnāma itself,
gained tremendous popularity among the Ottoman elite in the 16 th century: see e.g.,
SCHMIDT 2012; ULUÇ 2012.
151 The first one to propose this thesis was once again G. BALASCHEV (1930, 19). See also WIT-
TEK 1952, 668; WITTEK 1953; DECEI 1968, 108–11; İNALCIK 1991, 610; ZAJĄCZKOWSKI 1991; KAR-
PAT 1994a, 1994b, 1996; GÜNGÖR 1996; KIEL 2009, 141; GANGLOFF 1998, 14–22, who also pro-
vides a relatively comprehensive bibliography.
152 The first known mention of the designation “Gagauz” dates from 1837 in the context of
the largescale Christian migrations from Ottoman Dobruja to the Russian Empire. Moreo-
ver, a number of 19th and early 20th-century observers have noted that this name was origi-
nally an exonym with pejorative connotations: see MATEEVA 2006, 20, 73–5. Its alternative
interpretations usually go hand in hand with the various theories about the origin of the
Gagauz. For an outline of the major theses, see MATEEVA 2006, 23–36; SHABASHOV 2006, 8–14;
ATANASOV 2009, 401; GUBOGLO 2011; KAPALÓ 2011, 58–63 (in English, with some useful ob-
exist in earlier times, but it is risky at best to link it to a particular 13 th-century per-
sonality that, moreover, in no way survives in the well-researched historical
memory of these people.
As for the linguistic arguments, they are entirely based on the classification of
the Gagauz as belonging to the Oghuz subgroup of the Turkic languages as well as
on a study published in 1933 by Tadeusz Kowalski, who argued that the Gagauz of
Dobruja and the Muslims in the neighbouring Deli Orman region spoke a common
tongue, termed by him “Danubian Turkic”. He rather cautiously proposed a three-
stage chronology of the evolution of this language: “the earliest [layer] was formed
by the remains of a northern Turkic people, the second – by a strong southern
group going back to a time before the arrival of the Ottomans, and, finally, the
third layer was established by Turkic colonists and Turkified elements from the
Ottoman age.”153 The second layer, which, in Kowalski’s view, imposed the
“southern” (i.e. Oghuz) linguistic character on that Turkic dialect, has naturally
been interpreted by some later researchers as stemming from the 13 th-century
“people of Kaykāʾūs” who reportedly settled in Dobruja. However, while Ko-
walski provides numerous examples of “northern” (mostly Cuman-Qïpchak) ele-
ments in the Gagauz tongue, he gives no particular argumentation for the “second
layer” other than the inference that it cannot have been brought by Ottoman-era
settlers, part of whom (the Gagauz) would then become Christians by the 19 th cen-
tury. Indeed, mass Christianisation was hardly thinkable in a Muslim state, but a
pre-Ottoman migration from Anatolia is not the only possible scenario for the
“Oghuzisation” of the Gagauz language. Kowalski actually admitted in a later
study that the Gagauz may be descendants of “Turkic elements who came from the
north, from beyond the Danube”, but oddly maintained that the neighbouring
Muslim Turks of the Deli Orman migrated to the Balkans in pre-Ottoman times
and, in time, their language mingled with that of the northern settlers. 154 In fact, the
steppe nomads who repeatedly entered Dobruja from the north in the Middle Ages
included Oghuz groups while, on the other hand, Ottoman archival sources have
made it clear that the majority of the Turkish speaking Muslim population in Do-
bruja and the Deli Orman was formed in the late 15 th and the 16th centuries by
means of Turkmen colonisation from Anatolia, either directly or via the previously
colonized Southern Balkans.155 The question of which elements of the highly eclec-
tic Gagauz tongue can be deemed “original” and which were a product of “exter-
servations on the national/political biases underlying some of the theories). For arguments
against the Gagauz–Kaykāʾūs theory and an alternative hypothesis, see also CAHEN 1982.
153 KOWALSKI 1933, 27.
154 KOWALSKI 1938, 73.
155 See DIMITROV 1997–9; ANTOV 2017, esp. 115–24. Given that a large share of the settlers
migrated from Eastern Anatolia in the context of the Ottoman-Safavid struggle, it probably
accounted for the parallels between the “Danubian Turkic” and some dialects from North-
eastern Asia Minor, identified by KOWALSKI 1933, 24–5.
236 DELYAN RUSEV
nal” influences remains in the hands of the linguists who have actually proposed
theories for its evolution differing from that of Kowalski. 156 Undoubtedly, this
purely linguistic issue cannot be isolated from the abundant ethnographic and
historical scholarship on the Gagauz in which the thesis presenting them as “the
people of Kaykāvūs” is neither the only nor the dominant one.
Conclusion
Returning to the goals of this study as outlined in the introduction, it may be use-
ful to first highlight the undisputed historical data that can be extracted from the
diverse source material discussed so far. The most obvious fact is that a dervish
called (Sarı) Saltuk was active in the region of the Danube Delta in the last decades
of the 13th century. More precisely, he was based in the Danubian town of Sakçı
(present-day Isaccea), which was at the time under direct Tatar control. Sakçı was
most likely the “town of Bābā Salṭūq” visited by Ibn Baṭṭūṭa in the 1330s. Attempts
to identify the latter with present-day Babadag have been proven futile by the writ-
ten and archaeological evidence, according to which no noticeable settlement exist-
ed on that site between the 11th and the 15th centuries.
The pre-Ottoman sources unanimously locate Sarı Saltuk in a “Qïpchaq” envi-
ronment, i.e. an environment dominated demographically, culturally, and politi-
cally by the Tatars of the Golden Horde, including the Crimean Peninsula where
he undoubtedly spent some time. It is clear that this saintly figure was involved in
the Islamisation of the Horde and its southwestern section under the autonomous
ruler Nogai, in particular.157 There is also enough contemporary evidence, mainly
of Arabic provenance, to prove that the equally famous mystic Barak Baba, who
ended up at the Ilkhanid court, was Saltuk’s disciple. He was, however, not a son
of the former Seljuk Sultan ʿIzz al-Dīn Kaykāʾūs II. This link seems to have been
invented by Yazıcıoğlu ʿAlī in an attempt to incorporate the two saints in his story
of a largescale migration to Dobruja of Anatolian nomads associated with the sul-
tan. Indeed, it was already Paul Wittek who convincingly showed that the passag-
es featuring Sarı Saltuk formed one of four plot lines artificially stitched together
by the Ottoman historian and the sheikh’s figure was interpolated in the migration
156For an outline of the various theses, see MATEEVA 2006, 36–42.

157The Golden Horde has already been highlighted as the main environment of Sarı Saltuk’s
missionary activities by MUTAFČIEV 1943, 34–63 and TOGAN 1981, 267–70, although the latter
accepts the veracity of the migration story. The Tatar link is further evident in the wide-
spread legends (e.g. in the Saltuknāme and Evliya Çelebi’s Seyāḥatnāme) about Saltuk’s ex-
ploits in places with significant Tatar and no Anatolian-Turkish population such as Poland
and Lithuania: see e.g. NORRIS 2006, 64–6; ROHDEWALD 2017. Sarı Saltuk’s association with
the lands of the Golden Horde and the significant intertextuality between his legendary
cycle (esp. the Saltuknāme) and Tatar narratives of conversion are explored by DEWEESE
1994, 207–8, 251–6.
story, itself likely based on oral accounts. 158 Nevertheless, Wittek accepted the gen-
eral reliability of the narrative and later researchers have tended to view it as a
homogeneous entity. Ensuing attempts to identify the Dobruja nomads with a par-
ticular Anatolian tribal community such as the Chepni Turkmens as well as to pre-
sent Sarı Saltuk as their tribal chief are therefore highly speculative and have no
footing in the pre-Ottoman sources.159
The relation of Sarı Saltuk to the supposed Anatolian settlers in Dobruja, and
indeed the authenticity of the migration story itself, can at this stage remain in the
sphere of the hypotheses only. The contemporary and near-contemporary sources
are quite inconclusive in this regard. The Seljuk historians mention nothing, while
the information of the Byzantine authors is rather circumstantial. The most rele-
vant accounts concern the origin of the so-called Tourkopouloi who, according to
Gregoras, were descendants of the Turks who came to Byzantium together with
ʿIzz al-Dīn Kaykāʾūs, and, according to Pachymeres, had been recently Christian-
ized when they joined the tumultuous events in Thrace of the early 1300s from “the
northern regions”. While this evidence may indeed be related to a group of Anato-
lian nomads who had settled in Dobruja, this is not the sole possible interpretation
and, moreover, the accounts of the Byzantine authors contradict each other as well
as that of Yazıcıoğlu in a number of details. The latter is also tainted by factual
inaccuracies,160 apparent manipulations (e.g. the ancestry of Barak Baba), and in-
ternal contradictions such as his statement that the nomads were sent to Dobruja
due to their unsuitability to urban life and need for pastureland, but when they
arrived there, they not only established their camps but populated “two-three
Muslim towns” as well.
Despite the outlined shortcomings of Yazıcıoğlu ʿAlī’s account, some of its ele-
ments do sound more trustworthy. He certainly had some relatively reliable in-
formation on the fate of ʿIzz al-Dīn Kaykāʾūs’s descendants in Byzantium extracted
from people who were, or claimed to be, among them. There is also no obvious
reason for him, as a Muslim historian, to record the apostasy of the Muslim Turks
who remained in Dobruja after the death of Sarı Saltuk if there was no grain of
truth in the story. As noted, however, there is a problem of a geographical-cum-
chronological nature with some researchers’ attempts to identify these apostates
with the Tourkopouloi of the Byzantine sources: it is exactly in the late 13 th and the
early 14th centuries that contemporary evidence attests to Muslim presence in some
settlements of Northern Dobruja as well as to the area’s location in the Tatar sphere
of influence. This was certainly not a suitable environment for the Christianisation
158 WITTEK 1952, 652–3. According to Wittek, the amalgamation of the four stories resulted in
what he called a “Destān of ʿIzzeddīn’s and his people’s exile in the Dār ul-Ḥarb” (WITTEK
1952, 667).
159 The image of Sarı Saltuk as a tribal chief is most thoroughly developed by OCAK 2016, 92–8.
160 See e.g. n. 74 above.
238 DELYAN RUSEV
of a significant portion of the local population. Nevertheless, it is still possible that

Yazıcıoğlu ʿAlī only developed and exaggerated a genuine core of information
about a group of Anatolian Turks who accompanied either ʿIzz al-Dīn Kaykāʾūs or
Sarı Saltuk in the Golden Horde and ended up in Dobruja. One should also keep in
mind the possibility that the Ottoman author manipulated or invented the migra-
tion story in order to explain the presence of Turkish-speaking Christians in the
Northeastern Balkans, if such a population already existed in his time.
Whatever the truth, the relation of the supposed migration to the Gagauz is
equally hypothetical. If such a migration did in fact take place and some of the
Turkish settlers in Dobruja were baptized, it is logical to assume that they played a
certain role in the ethnogenesis of these people but were hardly the decisive ele-
ment in it. Historical, linguistic, and ethnographic data all show that this was a
complex process spanning multiple centuries and involving a substantial northern
Turkic substratum. Against this background, the conception of the 14th-century
Principality of Dobruja as a state formation of the Anatolian Turks is a fiction that
runs counter to the scarce evidence about the origin of the ruling dynasty.
At the present stage of research, it is impossible to provide a definitive solution
to the puzzle set up by Yazıcıoğlu. What is clear is that his interpolations to the
original narrative of Ibn Bībī form a masterful compilation of historical facts and
personalities, dubious oral traditions, and outright historical mystifications. The
particular motivation behind this compilation and its intended message as well as
its reception and developments in later times deserve a separate study within the
context of Ottoman politics and historiography.
Bibliography
ABU L-FIDAʾ/REINAUD – DI SLANE (1840): Géographie d’Aboulféda. Texte arabe. Publié
par M. Reinaud et M. le Baron Mac Guckin di Slane, Paris.
ABU L-FIDAʾ/REINAUD (1848): Géographie d’Aboulféda. Trad. par M. Reinaud, Paris.
ALEXANDRESCU-DERSCA BULGARU, M. M. (1978): Ştiri despre Babadag în Evul Me-
diu. Revista de istorie 31/3, 445–463.
ATWOOD, C. P. (2004): The Encyclopedia of Mongolia and the Mongol Empire, New York.
ALEKSIEV, B. (2012): Sedem myusyulmanski svettsi ot Balgariya. 2nd ed., Sofia.
ʿALLĀMĪ (1877): The Áín-i-Akbarí by Abul-Fazl-i-ʾAllámí. Ed. H. Blochmann, vol. 2,
Calcutta.
ANETSHOFER, H. (2012): Legends of Sarı Saltık in the Seyahatnâme and the Bektashi
Oral Tradition, in: Tezcan, N. – Tezcan, S. – Dankoff, R. (ed.:) Evliyâ Çelebi: studies
and essays commemorating the 400th anniversary of his birth. Ankara, 292–300.
ANTOV, N. (2017): The Ottoman “Wild West”. The Balkan Frontier in the Fifteenth and
Sixteenth Centuries, Cambridge.
AQSARĀYĪ/IŞILTAN (1943): Işıltan, F. Die Seldschukengeschichte des Aḳserāyī, Leipzig.
AQSARĀYĪ/TURAN (1999): Aksaraylı Mehmed oğlu Kerîmüddin Mahmud.

Müsâmeret ül-Ahbâr. Moğollar Zamanında Türkiye Selçukluları Tarihi, ed. O. Tu-
ran, Ankara.
ATANASOV, G. (2009): Dobrudzhanskoto despotstvo. Kam politicheskata, tsarkovnata,
stopanskata i kulturnata istorya na Dobrudzha prez XIV vek, Veliko Tarnovo.
ATANASOV, G. (2013): Za datirovkata i mnimiya seldzhukski proizkhod na sred-
novekovnite relefi ot Kaliakra. Prinosi kam balgarskata arkheologiya, vol. 7, 159–172.
BALASCHEV, G. (1930): O autokratōr Michaēl o Palaiologos kai to idrythen tē syndrome
autou kratos tōn Ogouzōn para tēn dytikēn aktēn tou Euxeinou, Sofia.
BARTUSIS, M. L. (1992): The late Byzantine army: arms and society, 1204–1453, Phila-
delphia.
BILIARSKY I. (1992): Despot Yoan Terter (40-te – 90-te godini na XIV stoletie). Is-
toricheski pregled 48/10, 3–23.
BOZHILOV, I. – GYUZELEV, V. (1999): Istoriya na srednovekovna Balgariya VII-XIV vek,
Sofia.
BRAND, Ch. M. (1989): The Turkish Element in Byzantium, Eleventh-Twelfth Cen-
turies. Dumbarton Oaks Papers 43, 1–25.
CAHEN, C. (1982): Points de doute sur les Gagaouzes, in Gallotta, A. – Marazzi, U.
(ed.:) Studia Turcologica memoriae Alexii Bombaci dicata, Napoli, 89–90.
CM/ BUCHON (1845): Le livre de la conqueste de la princée de Morée. Publ. par J. A.
Buchon, t. I, Paris.
CM/ LURIER (1964): Crusaders as Conquerors. The Chronicle of Morea. Transl. H. E.
Lurier, New York–London.
CM/SCHMITT (1904): The Chronicle of Morea. Ed. by J. Schmitt, London.
CM/VAN ARSDALL – MOODY (2015): The Old French Chronicle of Morea: an account of
Frankish Greece after the Fourth Crusade. Transl. by A. Van Arsdall, H. Moody,
Burlington, VT.
COSTAN, G. A. (2014): Contextul ridicării bisericii cu planul în cruce grecă înscrisă
de la Niculiţel (jud. Tulcea). Buletinul Cercetărilor Ştiinţiice Studenţeşti 20, 105–121.
COSTAN, G. A. (2016): The Rifa’iyya Order and the Romanians during Pre-Ottoman
and Ottoman Times, in Nitu, F. et al. (ed.:) Turkey & Romania. A History of Part-
nership and Collaboration in the Balkans (IBAC, Book Series, 4). Istanbul, 21–32.
DANILO/DANIČIĆ (1866): Životi kraljeva i arhiepiskopa srpskih, napisao arhiepiskop
Danilo i drugi. Ed. by Đ. Daničić. Zagreb.
DANILO/MIRKOVIĆ (1935): Životi kraljeva i arhiepiskopa srpskih. Transl. L. Mirković.
Belgrade.
DECEI, A. (1968): Le problème de la colonisation des Turcs seljoukides dans la Do-
brogea au xiiie siècle. Tarih Araştırmaları Dergisi 10–11, 85–111.
DEWEESE, D. (1994): Islamization and native religion in the Golden Horde: Baba Tükles
and conversion to Islam in historical and epic tradition, Pennsylvania.
DIACONU, P. (1972): Despre datarea „circumvalaţiei“ şi a „bisericii treflate“ de la
Niculiţel. Studii şi Cercetări de Istorie Veche şi Arheologie XXIII/2, 307–319.
240 DELYAN RUSEV
DIACONU, P. (1994): Cumanii şi familiei lui Dobrotiţă. Revista istorică 3–4, 283–288.
DIMITROV, I. K. (1915): Preselenie na seldzhukski turtsi v Dobrudzha okolo sredata
na XIII v., Spisanie na BAN, vol. X. Klon istoriko-filologichen i filosofsko-
obshtestven, 6. Sofia, 23–40.
DIMITROV, S. (1994): Kam istoriyata na dobrudzhanskite dvuobredni svetilishta.
Dobrudzha 11, 76–97.
DIMITROV, S. (1997–9): Novi danni za demografskite otnosheniya v yuzhna Dobru-
dzha prez parvata polovina na XVI v. Dobrudzha 14–16, 278–333.
DOĞANBAŞ, M. (2015): Koyun Baba Velâyetnamesi, Istanbul.
DUDA, H. W. (1943): Zeitgenössische islamische Quellen und das Oġuznāme des
Jazyǧyoġlu ʿAlī zur angeblichen türkischen Besiedlung der Dobrudscha im 13.
Jhd. n. Chr. Balgarska akademiya na naukite i izkustvata LXVI, 131–145.
DUYCHEV, I. (1937): Neizdadeno pismo na papa Benedikt XII do maykata na tsar
Ivan Aleksandar. Izvestiya na istoricheskoto druzhestvo v Sofiya XIV–XV, 205–210.
EBÜ’L-HAYR-I RÛMÎ/AKALIN (1987–1990): Saltuk-nâme. Vol. 1–3, ed. Ş. H. Akalın.
Ankara.
ERMERS, R. (1999): Arabic Grammars of Turkic, Leiden.
EVLIYA ÇELEBI/KAHRAMAN–DAĞLI (1999): Evliyâ Çelebi Seyahatnâmesi. Vol. 3, ed. by
Kahraman S. A. – Dağlı, Y., Istanbul.
GANGLOFF, S. (1998): Les Gagaouzes : état des recherches et bibliographie. Turcica
30, 13–61.
GOLDEN, P. B. (1992): An Introduction to the History of the Turkic Peoples: Ethnogenesis
Wiesbaden.
GOLDEN, P. B. (1995–7): Cumanica IV: The Cumano-Qıpčaq Clans and Tribes. Ar-
chivum Eurasiae Medii Aevi 9, 99–122.
GENAKOPLOS, D. J. (1959): Emperor Michael Palaeologus and the West, 1258–1282. A
Study in Byzantine-Latin Relations, Cambridge, MA.
GÖLPINARLI, A. (1936): Yunus Emre. Hayatı, Istanbul.
GÖLPINARLI, A., ed. (1990): Menakıb-ı Hacı Bektâş-ı Velî. “Vilâyet-nâme”, Istanbul.
GREGORAS/SCHOPEN (1829): Byzantina historia. Ed. by L. Schopen, vol. 1, Bonn.
GREGORAS/VAN DIETEN (1973): Rhomäische Geschichte. Historia Rhomaïke. Übers. u.
erl. v. Jan Louis van Dieten. I. Teil, Stuttgart.
GUBOGLO, M. N. (2011): Vvedenie, in Guboglo, M. N. – Kvilinkova, E. N. (ed.:)
Gagauzy. Moskva, 8–21.
GÜNGÖR, H. (1996): Le problème de l’origine des gagaouzes d’après les chroniques
seldjoukides et ottomans. Belleten LX/229, 695–701.
GÜREL, Z. (2019): Hacı Bektaş Veli’nin Halifelerinden Koyun Baba’nın Balkan-
lar’daki İzleri. Türk Kültürü ve Hacı Bektaş Velî Araştırma Dergisi, 91 (2019), 29–42.
GYUZELEV, V. (1995): Ochertsi varkhu istoriyata na balgarskiya Severoiztok i Cherno-
morieto (kraya na XІІ – nachaloto na XV vek), Sofia.
GYUZELEV, V. (2021): Balgarskata diaspora i prisastvie v Genuezkata republika prez

XIII–XV v., in Gyuzelev, V. Sedmoknizhie za srednovekovnata istoriya na Balgariya
(IV – XV vek). Vol 2, Balgariya i srednovekovniyat svyat. Sofia, 414–438.
HASLUCK, F. W. (1929): Christianity and Islam under the Sultans, vol. 1–2, Oxford.
HOPE, M. (2016): Power, Politics, and Tradition in the Mongol Empire and the Ilkhanate
of Iran, Oxford.
HRBEK, I. (1962): The Chronology of Ibn Baṭṭūṭa’s Travels. Archiv orientální 30, 409–486.
IBN BAṬṬŪṬA/DEFRÉMERY–SANGUINETTI (1854): Voyages d’Ibn Batoutah. Éd. par B.
Defrémery, R. Sanguinetti. 2 tomes, Paris.
IBN BAṬṬŪṬA/GIBB (1959): The Travels of Ibn Baṭṭūṭa, A.D. 1325–1354, vol 2. Transl.
H. A. R. Gibb, Cambridge.
IBN BĪBĪ/DUDA (1959): Die Seltschukengeschichte des Ibn Bībī. Übers. v. H. W. Duda,
Kopenhagen.
İNALCIK, H. (1991): Dobrud̲j̲a. Encyclopaedia of Islam, 2nd ed., vol. 2, 610–613.
IORGA, N. (1928): Dobrotitsch (Dobrotica, Dobrotici) : quelques observations. Revue
historique de Sud-Est européen, V/4–6, 133–136.
IOSIPESCU, R. – IOSIPESCU, S. (2004): Stefan cel Mare si Dobrogea de nord în vremea
marelui razboi cu Imperiul Otoman, in Stefan cel Mare si Sfânt. Atlet al credintei
crestine. Sfânta Mănăstire Putna, 307–322.
İZBUDAK, V. (1936): El-İdrâk Haşiyesi, Istanbul.
JACOBY, D. (1979): Catalans, Turcs et Vénitiens en Roumanie (1305-1332) : un nou-
veau témoignage de Marino Sanudo Torsello, in Jacoby, D. Recherches sur la
Méditerranée Orientale du XIIe au XVe siècle. London, V, 217–261.
JACOBY, D. (2015): The Catalan Company in the East: the Evolution of an Itinerant
Army (1303-1311), in Halfond, G. (ed.:) The Medieval Way of War: Studies in Me-
dieval Military History in Honor of Bernard S. Bachrach. Farnham, 153–182.
KANTAKOUZENOS/SCHOPEN (1830): Ioannis Cantacuzeni eximperatoris historiarum libri
IV. Graece et Latine, vol. 2. Ed. L. Schopen, Bonn.
KANTAKOUZENOS/FATOUROS – KRISCHER (2011): Johannes Kantakuzenos. Geschich-
te, Buch III. Übers. u. erl. v. G. Fatouros, T. Krischer, Stuttgart.
KAPALÓ, J. A. (2011): Text. Context and Performance. Gagauz Folk Religion in Discourse
and Practice, Leiden–Boston.
KARAMUSTAFA, A. T. (2015): Islamisation through the Lens of the Saltuk-name, in
Peacock, A. C. S. – de Nicola, B. – Nur Yıldız, S. (ed.:) Islam and Christianity in
Medieval Anatolia. Farnham, 349–364.
KARPAT, K. (1994a): Selʹdzhuksko-anatoliiskoe proiskhozhdenie gagauzov. Etno-
graficheskoe Obozrenie 4, 36–43.
KARPAT, K. (1994b): Dobruca. TDV İslâm Ansiklopedisi, vol. 9, 482–486.
KARPAT, K. (1996): Gagauzlar. TDV İslâm Ansiklopedisi, vol. 13, 288–291.
KENNEDY, E. S. – KENNEDY, M. H. (1987): Geographical Coordinates of Localities from
Islamic Sources, Frankfurt a. M.
242 DELYAN RUSEV
KIEL, M. (1978): The Türbe of Sarı Saltık at Babadag – Dobrudja. Brief Historical
and Architectonical Notes. Güney-Doğu Avrupa Araştırmaları Dergisi 6–7, 205–225.
KIEL, M. (1994): Mevlana Neşri and the Towns of Medieval Bulgaria, historical and
topographical Notes, in Heywood, C. – Imber, C. (ed.:) Studies in Ottoman His-
tory in Honour of Professor V. L. Ménage. Istanbul, 165–187.
KIEL, M. (2000): Sarı Saltuk: Pionier des Islams auf dem Balkan, im 13. Jahrhundert,
mit Materialien von Berndt Radtke, in Franz, E. – Engin, İ. (Hrsg.:) Alevîler /
Alewiten. Hamburg, 253–286.
KIEL, M. (2005): Ottoman Urban Development and the Cult of a Heterodox Sufi
Saint: Sarı Saltuk Dede and Towns of İsakçe and Babadağ in the Northern Do-
bruja, in Veinstein, G. (ed.): Syncrétismes et hérésies dans l’Orient seljoukide et ot-
toman (XIVe-XVIIIe siècle). Paris, 284–298.
KIEL, M. (2009): The Incorporation of the Balkans into the Ottoman Empire, 1353–
1454, in Fleet, K. (ed.): The Cambridge History of Turkey. Vol. 1: From Byzantium
to Turkey. Cambridge, 138–191.
KING, D. A. (1999): World-Maps for Finding the Direction and Distance to Mecca. Inno-
vation and Tradition in Islamic Science, Leiden.
OBV/KILIÇ (2007): Otman Baba Velâyetnâmesi. Ed. by F. Kılıç, M. Arslan, T. Bülbül,
Ankara.
KONOVALOVA, I. – RUSSEV, N. D. (1994): Zolotaya Orda v zhizni gorodov Dnestrov-
sko-Dunaiskogo regiona (Po svedeniyam pisʹmennykh istochnikov), in Evul
mediu timpuriu în Moldova (Probleme de istoriografie şi istorie urbană). Chişinău,
70–108.
KOWALSKI, T. (1933): Les Turcs et la langue Turc de la Bulgarie de Nord-Est, Krakow.
KOWALSKI, T. (1938): Les éléments ethniques turcs de la Dobroudja. Rocznik Orien-
talistyczny 14, 66–80.
KÖPRÜLÜ, M. F. (1992): The Seljuks of Anatolia: their history and culture according to
local Muslim sources. Transl. and ed. G. Leiser, Salt Lake City.
KÖPRÜLÜ, M. F. (1993): Islam in Anatolia after the Turkish Invasion (Prolegomena).
Transl. and ed. G. Leiser, Salt Lake City.
KRASTEV, K. (2011): Balgarskoto tsarstvo pri dinastiyata na Terterevtsi (1280-1323),
Plovdiv.
KUZEV, A. (1981): Obluchitsa – Isakcha, in Kuzev, A. – Gyuzelev, V. (ed.:) Balgarski
srednovekovni gradove i kreposti. Vol. I, Varna, 211–215.
LAIOU A. E. (1972): Constantinople and the Latins: The Foreign Policy of Andronicus II,
1282–1328, Cambridge, MA.
LAURENT, V. (1956): Une famille turque au service de Byzance : les Mélikès. Byzan-
tinische Zeitschrift 49/2, 349–368.
LAURENT, V. (1945): La domination byzantine aux bouches du Danube sous Michel
VIII Paléologue. Revue historique de Sud-Est européen 22, 184–198.
MATANOV, H. (2013): Oshte po vaprosa za “seldzhukskoto preselenie“ v Dobru-
dzha prez 60-te godini na XIII vek, in Nikolov, G. – Nikolov, A. (ed.:) Balgarsko
srednovekovie: Obshtestvo, vlast, istoriya. Sbornik v chest na prof. d-r Miliyana Kay-
makamova. Sofia, 373–384.
MATEEVA, V. (2006): Gagauzite. Oshte edin pogled, Sofia.
MELIKOFF, I. (1994): Qui était Sarı Saltuk? Quelques remarques sur les manuscrits
du Saltukname, in Heywood, C. – Imber, C. (ed.:) Studies in Ottoman History in
Honour of Prof. V. L. Ménage. Istanbul, 231–238.
MOISESCU, C. (2000): Noi considerații asupra vechimii și arhitecturii bisericii Sf.
Atanasie din Niculițel (jud. Tulcea), in Istro-Pontica: Muzeul tulcean la a 50-a ani-
versare (1950–2000). Tulcea, 363–370.
MORAVCSIK, G. (1983): Byzantinoturcica, 2 vols, Leiden.
MUNTANER/DE BOFARULL (1860): Crónica Catalana de Ramon Muntaner. Texto origi-
nal y traduccion Castellana por A. de Bofarull, Barcelona.
MUNTANER/HUGHES (2006): The Catalan Expedition to the East: from the Chronicle of
Ramon Muntaner. Transl. R. D. Hughes. Barcelona–Woodbridge.
MUTAFCIEV, P. (1927): Dobrotic, Dobritica et la Dobrudza. Revue des études slaves 7,
27–35.
MUTAFCIEV, P. (1931): Encore sur Dobrotica. Godishnik na Sofiyskiya universitet –
Istoriko-filologicheski fakultet 27, 1–14.
MUTAFČIEV, P. (1943): Die angebliche Einwanderung von Seldchuk-Türken in die
Dobrudscha im XIII. Jahrhundert. Balgarska akademiiya na naukite i izkustvata 66,
1–129.
MUTAFČIEV, P. (1973): Mnimoto preselenie na seldzhushki turtsi v Dobrudzha prez
XIII v., in Mutafčiev, P. Izbrani proizvedeniya, ed. D. Angelov, vol. 2, Sofia, 607–745.
MUTAFOVA K. (2008): Utrakvistichnite svetilishta, posveteni na Sara Saltak na Bal-
kanite i kodat na konfesionalno sazhitelstvo, in Balkanite – ezik, istoriya, kultura
(Materiali ot mezhdunarodnata nauchna konferentsiya, 2007), Veliko Tarnovo, 280–
289.
NEDKOV, B. (1960): Balgariya i sasednite i zemi prez XII vek spored "Geografiyata" na
Idrisi, Sofia.
NORRIS, H. T. (1994): Ibn Baṭṭūṭa’s Journey in the North-Eastern Balkans. Journal of
Islamic Studies 5/2, 209–220.
NORRIS, H. T. (2006): Popular Sufism in Eastern Europe. Sufi brotherhoods and the dia-
logue with Christianity and ‘Heterodoxy’, London–New York.
OBERLÄNDER-TARNOVEANU, E. (1980): Pentru o nouă datare a bisericuței cu plan
treflat de la Niculițel (jud. Tulcea). Peuce 8, 451–457.
OBERLÄNDER-TARNOVEANU, E. (1995/6): Byzantino-Tartarica. Le monnayage dans
la zone des bouches du Danube à la fin du XIIIe siècle. Il Mar Nero 2, 191–214.
OBERLÄNDER-TÂRNOVEANU, Е. (2004): The Coinage of the Genoese Settlements of
the Western Black Sea Shore and on the Danube. Peuce, S. N. 2 (15), 285–296.
OCAK, A. Y. (1989): La révolte de Baba Resul, ou la formation de l’hétérodoxie musulmane
en Anatolie au XIIIe siècle, Ankara.
244 DELYAN RUSEV
OCAK, A. Y. (1999): Osmanlı İmparatorluğu'nda Marjinal Sûfîlik: Kalenderîler (XIV-

XVII. Yüzyıllar). 2nd ed., Ankara.
OCAK, A. Y. (2002): Sarı Saltık: Popüler İslam'ın Balkanlar'daki Destanî Öncüsü (XIII.
yüzyıl), Ankara.
OCAK, A. Y. (2016): Sarı Saltık: Popüler İslamın Balkanlar'daki Destani Öncüsü (13.
yüzyıl), revised 3rd ed., Istanbul.
OIKONOMIDES, N. (1993): The Turks in Europe (1305-13) and the Serbs in Asia Mi-
nor (1313), in Zachariadou, E. (ed.:) The Ottoman Emirate (1300-1389). Halcyon
Days in Crete I. A Symposium Held in Rethymnon 11-13 January 1991. Rethymnon,
159–168.
ÖZTÜRK, E. (2013): Velilik ile delilik arasında: İbnu's-Serrâc'ın gözünden muvelleh der-
vişler, Istanbul.
PACHYMERES/FAILLER, vol. 1 (1984): Pachymérès, G. Relations historiques. I. Livres I-
III. Éd. par A. Failler, trad. par V. Laurent, Paris.
PACHYMERES/FAILLER, vol. 2 (1984): Pachymérès, G. Relations historiques. II. Livres
IV-VI. Éd. par A. Failler, trad. par V. Laurent, Paris.
PACHYMERES/FAILLER, vol. 3 (1999): Pachymérès, G. Relations historiques. III. Livres
VII-IX. Éd. et trad. par A. Failler, Paris.
PACHYMERES/FAILLER, vol. 4 (1999): Pachymérès, G. Relations historiques. IV. Livres
X-XIII. Éd. et trad. par A. Failler, Paris.
PAVLOV, P. (1995): Tatarite na Nogay, Balgariya i Vizantiya (okolo 1270–1302 g.), in
Todorov, P. (ed.:) Balgarite v Severnoto Prichernomorie, vol. 4, Veliko Tarnovo,
121–130.
PAVLOV, P. (1997): Balgariya i pohodite na “Zlatnata orda” sreshtu Vizantiya prez
XIII–XIV v. (S ogled na politicheskite sabitiya v Severnoto Chernomorie i Bal-
garskiya Severoiztok), in Todorov, P. (ed.:) Balgarite v Severnoto Pricher-nomorie,
vol. 6, Veliko Tarnovo,
PEACOCK, A. C. S. (2013): Court and Nomadic Life in Saljuq Anatolia, in Durand-
Guédy, D. (ed.:) Turko-Mongol Rulers, Cities and City Life. Leiden, 191–222.
PEACOCK, A. C. S. (2014) The Seljuk Sultanate of Rūm and the Turkmen of the Byz-
antine frontier, 1206–1279. Al-Masaq: Journal of the Medieval Mediterranean 26/3,
267–287.
PETROV, P. (1956): Balgaro-vizantiyskite otnosheniya prez vtorata polovina na XIII
v., otrazeni v poemata na Manuil Fil “Za voennite podvizi na izvestniya
chutoven protostrator”. Izvestiya na Instituta za balgarska istoriya 6, 545–575.
PLP (2001): Prosopographisches Lexikon der Palaiologenzeit. Erstellt von E. Trapp. CD
ROM-Version, Wien.
PSRL (1908): Polnoe sobranie russkikh letopisei. Vol. 2, Ipatʹevskaia letopisi, S. Peters-
burg.
RASONYI, L. (2006): Kuman Özel Adları, in Rásonyı, L. Doğu Avrupa’da Türklük.
Istanbul, 151–269.
RÁSONYI, L. – BASKI, I. (2007): Onomasticon Turcicum. Turkic personal names, 2 vols.

Bloomington, Ind.
ROHDEWALD, S. (2017): Sarı Saltuk im osmanischen Rumelien, der Rus’ und Polen-
Litauen. Zugänge zu einer transosmanischen religiösen Erinnerungsfigur (14.–
20. Jh.), in: Jobst, K. S. – Hüchtker, D. (Hrsg.): Heilig. Transkulturelle Verehrung-
skulte vom Mittelalter bis in die Gegenwart. Göttingen, 67–98.
SARIKAYA, M. S. – BARDAKÇI, M. N. – GÜRKAN, N. (2013): İbnü’s-Serrâc’a Göre Sarı
Saltuk. Tasavvuf 32/2, 75–110.
SAVVIDES, A. G. C. (1992): Ē proeleusē kai o rolos tōn tourkophōnōn misthophorōn
ston Morea kata ton Vyzantinophrankiko polemo tou 1263–64, in Praktika tou
D’ diethnous synedriou peloponnēsiakōn spoudōn (Korinthos 9-16 Septemvriou 1990).
Vol. I, Athens, 165–188.
SAVVIDES, A. (1993): Late Byzantine and Western Historiographers on Turkish
Mercenaries in Greek and Latin Armies: the Turcoples/Tourkopouloi, in Bea-
ton, R. – Roueché, Ch. The Making of Byzantine History. Studies dedicated to D. M.
Nicol. Aldershot, 122–136.
SCHMIDT, J. (2012): The Reception of Firdausi’s Shahnama Among the Ottomans, in
Melville, Ch. – Van den Berg, G. (ed.:) Shahnama Studies. Vol. 2, The reception of
Firdausi’s Shahnama. Leiden–Boston, 121–139.
SCHREINER, P. (1975) Die byzantinischen Kleinchroniken. 1. Teil: Einleitung und Text.
Wien.
SHABASHOV, V. A. (2006): Etnogenez, in Bulgar, S. S. (ed.:) Istoriya i kulʹtura gagau-
zov. Komrat-Chișinău, 8–43.
SHAWCROSS, C. T. M. (2009): The Chronicle of Morea: Historiography in Crusader
Greece, Oxford–New York.
SHM/DE RACHEWILTZ (2004): The Secret History of the Mongols. A Mongolian Epic
Chronicle of the Thirteenth Century. Transl. and comment. I. de Rachewiltz. 2
vols., Leiden–Boston.
SHUKUROV, R. M. (2008): Semeistvo ‛Izz al-Dina Kai-Kavusa II v Vizantii. Vi-
zantiiskii Vremennik 67 (92), 89–116.
SHUKUROV, R (2016): The Byzantine Turks, 1204–1461, Leiden.
SMITH, G. A. (1982): Some Türbes/Maqāms of Sarı Saltuq, an Early Anatolian Turk-
ish Ġāzī-Saint. Turcica 14, 216–225.
ŞAHIN, H. 2002: Koyun Baba. TDV İslâm Ansiklopedisi, vol. 26, 229–230.
ŞIRIN USER, H. (2014): Čolpan ‘The Planet Venus’ in Turkic. Studia Etymologica
Cracoviensia 19, 169–178.
TIZENGAUZEN, V. (1884): Sbornik” materıalov”, otnasyashchikhsya k” istorıi Zolotoi ordy,
vol. 1, St Petersburg.
TOGAN, Z. V. (1981): Umumi Türk Tarihine Giriş. 3rd ed., Istanbul.
ULUÇ, L. (2012): The Shahnama of Firdausi in the Lands of Rum, in Melville, Ch. –
Van den Berg, G. (eds.:) Shahnama Studies. Vol. 2, The reception of Firdausi’s
Shahnama. Leiden–Boston, 159–180.
246 DELYAN RUSEV
UZELAC, A. (2015): Pod senkom psa: Tatari i južnoslovenske zemlje u drugoj polovini XIII
veka. Beograd.
UZELAC, A. (2017): The Golden Horde and the Balkans (13–14th Cenrturies), in
Khakimov, R. – Favereau, M. (ed.:) The Golden Horde in World History, Kazan,
376–394.
VÁSÁRY, I. (2005): Cumans and Tatars. Oriental Military in the Pre-Ottoman Balkans,
VASILIU, I. (1996a): Consideraţii finale asupra locuirii feudale timpurii de la Baba-
dag. Peuce 12, 169–186.
VASILIU, I. (1996b): Ştiri istorice şi date arheologice referitoare la oraşul Babadag în
evul mediu. Peuce 12, 195–224.
VEINSTEIN, G. (2005): Le rôle des tombes sacrées dans la conquête ottomane. Revue
de l’histoire des religions 222/4, 509–528.
WITTEK, P. (1934): La descendance chrétienne de la dynastie Seldjouk en Macé-
doine. Échos d’Orient 33/176, 409–412.
WITTEK, P. (1952): Yazijioghlu 'Ali on the Christian Turks of the Dobruja. Bulletin of
the School of Oriental and African Studies 14/3, 639–668.
WITTEK, P. (1953): Les Gagaouzes : Les gens de Kaykāūs. Rocznik Orientalistyczny
17/1951–2, 12–24.
WITTEK, P. (1964): Das Datum von Yazıcıoğlu’s Oğuznâme. Türkiyat Mecmuası 14,
263–265.
YAZICIZĀDE ʿALĪ/BAKIR (2014): Selçuḳ-nāme. İndeksli tıpkıbasım. Ed. A. Bakır, Ankara.
YAZICIZĀDE ʿALĪ/BAKIR (2017): Tevârîh-i Âl-i Selçuk. Ed. A. Bakır, Istanbul.
YOSIFOVA, M. (1978): Srednovekovni relefi ot Kaliakra. Arkheologiya XX/3, 50–58.
YÜCE, K. (1987): Saltuk-Nâme’de Tarihi, Dinî ve Efsanevî Unsurlar, Ankara.
ZACHARIADOU, E. (1965): Oi christianoi apogonoi tou Kaikaous V’ stē Veroia. Make-
donika 6, 62–74.
ZACHARIADOU, E. (1978): Observations on Some Turcica of Pachymeres. Revue des
études byzantines 36, 261–267.
ZAJACZKOWSKI, W. (1991): Gagauz. The Encyclopaedia of Islam, 2nd ed., vol. 2, 971–972.
ZHAVORONKOV P. I. (2006): Tyurki v Vizantii (XIII–seredina XIV v.). Chast’ per-
vaya: tyurkskaya aristokratiya. Vizantiiskii vremennik 65, 163–177.
Experiencing Alterity: Italian Merchants
and the Local Population in the
14th Century Venetian Azov Sea
Changes and Continuity
LORENZO PUBBLICI
Abstract
After the Fourth Crusade (1204), Western merchants began to frequent the Black
Sea basin. Genoa and Venice established commercial settlements in the region
called emporia, which over time became an indispensable reference point for travel-
lers from both Western Europe and from the East. The commercial history of the
two Italian cities soon intersected with the rapid formation process of the Mongol
Empire that by the mid-13th century extended from China to Eastern Europe. The
constitution of a homogeneous and vast political entity facilitated communications
and partly promoted them, guaranteeing Western merchants a “safer” space with-
in which to move and cover distances previously unimaginable. In the emporia
founded on the Black Sea—and in this context—the Western urban mercantile class
encountered the local element and other migrants from substantially unknown
geographical areas. This paper examines relationships between Venetian citizens
and foreigners as well as the local population who lived in Tana on the Azov Sea.
Tana was the easternmost settlement of the entire Latin trading system and is stud-
ied here from the mid-14th century until the end of the century, a period that was
both politically and economically problematic and characterized by international
tensions, economic crises but also extraordinary opportunities for commercial ex-
pansion and profit.
Keywords: Venice, Mongol Empire, Nomads, Azov Sea, Tana, Golden Horde, Eur-
asian trade
 Lorenzo Pubblici is Assistant Professor of History and Cultures of Premodern Central

Asia, Department of Asian, African, and Mediterranean Studies, L’Orientale, University of
Naples. The article is a revised version of the keynote lecture delivered at the 8th Interna-
the Middle Ages, held in Sofia, November 20–3, 2019.
248 LORENZO PUBBLICI
1. The 14th-century Venetian merchant: growth, crisis, and consolidation

Between the 11th and 14th centuries, Western Europe experienced demographic and
economic growth that is traditionally associated with the concept of “commercial
revolution”. This phenomenon, whose breadth varied from region to region and
did not affect all social classes equally, changed western society just as much as the
industrial revolution did in the 18th and 19th centuries.1 The expansive tendency
was general and rearranged the social hierarchies without upsetting the order pre-
ordained by divine will.2 The protagonists of this new course were the Italian cities
and, in particular, those that, for a number of reasons that are out of the scope of
our investigation here, were better prepared to take advantage of the opportunity:
Genoa, Milan, Venice, and Florence.
Genoa and Venice’s maritime power reached its acme in the 13 th century. After
the Fourth Crusade, both city-states established trade posts on the Black Sea coasts
and their merchants regularly travelled to the Pontic regions. The two cities experi-
enced very different historical trajectories, and their governing sought to dominate
the Mediterranean and Levant maritime routes. If in Genoa the private initiative
took precedence, in Venice the State supervised everything without overwhelming
or depressing the capabilities of the individual by limiting their opportunities.
Especially after the Serrata del Maggior Consiglio (February 1297),3 Venice’s aristo-
cratic government organized the market by centralizing its control. In Venice, trade
was the primary source of wealth since the birth of the city. The nobility was per-
sonally engaged in commerce, but the humble classes were never entirely excluded
from it. The State financed the shipbuilding industry, controlled the navigation
routes, and assigned the ships to private auctions: a system called incanto. The State
regulated the navigation calendar and managed the convoys voyages (conserva) to
ensure the ships’ safety. The State supervised the contracts with which a company
was established. The State also deterred the creation of monopolies and guaranteed
high wages even to the popular professions. Finally, the State regulated the flow of
goods in and out of Venice, from the cheapest to the most valuable merchandise.
From the middle of the 13th century, and in some areas of Europe even earlier,
the dominant classes increased their prosperity and the nouveaux wealthy (gente
nova) pushed up the demand for luxury and exotic products. The changes in the
mercantile profession represented a natural outlet to this condition. Merchants
specialized, professionalized, and created increasingly sophisticated tools to max-
imize their business profits while reducing risks. In the 14 th century, the merchant
gradually became a sedentary entrepreneur at the head of a company capable of
producing profits in far regions, counting on a well-organized and consistent
workforce. Accountancy and banking techniques were developed, and new forms
1 GIES – GIES 1972; LOPEZ 1976, 103ff.; LOPEZ 1955, 46–7; CIPOLLA 1980, 142–62; POUNDS 1978.
2 See for example KEDAR 1976, esp. ch. 2, and SPUFFORD 2006.
3 RAINES 2003, 6ff.
EXPERIENCING ALTERITY: ITALIAN MERCHANTS… 249
of association created. There was no law in Venice that excluded the individual
from commercial activity. The filter consisted of the resources available. That is
why the nobility represented the majority in the Venetian emporia of the Levant.
Nonetheless, the aristocracy completely belonged to the urban mercantile class that
populated all the late medieval Italian cities.
Venetian exceptionalism was also expressed in the relative closure of the city
towards foreigners. In Venice, one could enter, live, and trade, but the authorities
imposed strict control on the activities of non-citizens. It is not a surprise that there
was a closed and well-controlled German factory (Fondaco dei Tedeschi) just as it is
no coincidence that the first Jewish ghetto in Europe was established in Venice.
Still, in the first decades of the 14th century, it was necessary to reside in the city
for twenty-five years to obtain Venetian citizenship. 4 This rigidity was an obstacle
to Venice’s commercial development. 5 The restrictions that applied to foreigners in
the Black Sea emporia were somewhat more flexible, in particular in Tana, where
all those who travelled on Venetian ships and resided in the Venetian quarter were
considered part of the community.
The Black Sea region was central to the Eurasian trade system. Most of the lead-
ing international transit routes, both before and after the Mongol conquest, passed
through the Pontic area. However, the Mongol Empire’s formation in the 13 th cen-
tury stretched the barycentre of international trade further eastwards. The Mongols
created the conditions for the integration of the Mediterranean and Pontic (from
the Volga to Tabriz) trade systems. Whether we believe in the so-called Pax Mon-
golica or not, there is no doubt that the Mongol Empire guaranteed coherent politi-
cal interlocutors and an efficient and secure communication system over unprece-
dented distances for at least fifty years.6 From the second half of the 13th century,
Italian merchants penetrated deeper and deeper into Asia, and Muslim merchants
from Transoxiana frequented the Pontic markets assiduously. It is also worth
pointing out that goods travelled more than people, both in frequency and dis-
tance. The extraordinary experiences described in celebrated works by Marco Polo,
Plano Carpini, or Giovanni di Montecorvino were exceptions and not the rule. The
Black Sea ports became crucial junctions of this international trade system,
which—in the face of the 14th-century downturn in the continental economy—
constituted the indispensable prop of a new process of economic “bilateral” expan-
sion. The Italian “colonization” did not have a dominant actor that imposed its
rhythm on local politics, but two subjects that dialogued to exploit the commercial
activity from which both drew profits.
This balance underwent a crisis starting in the 1330s and travelling over long
distances became increasingly difficult. To overcome the impediments, the mer-
4 MUELLER 1992, 37–8; MOLA 1994, 36–44.

5 LOPEZ 1955, 49.
6 DI COSMO 2009, 99–100.
chants invested more and more in local intermediaries and transitional stages.
Trade was a primary resource for the nomads both for subsistence and the repro-
duction of their social hierarchies. Latin and Asian merchants were the two faces of
the same coin, bearers of two similar exigences that met halfway. The encounter of
the Western urban mercantile class with the nomads gave life to new economic
cooperation that also developed into human and social interactions.
2. Venice, the Azov Sea and the “levels” of interactions
The Italian frequentation of the Black Sea ports began right after the Fourth Cru-
sade. The Venetian conquest of Constantinople of 1204 led to the Eastern Roman
Empire’s dismantling, sealed in the Partitio Romaniae.7 Venice used the port of Su-
dak/Soldaïa in Crimea as a trade hub, but it was only in 1261 that the Italians
planned to settle on the Black Sea shores. The Treaty of Nymphaeum, 8 stipulated in
March 1261 between Emperor Michael VIII (r. 1261–82) and the Genoese authori-
ties, shifted the political balance in favour of the Tyrrhenian city, ensuring the pos-
sibility of exploiting the Pontic region, a centre between the Christian West and the
Eastern markets. The stipulation of the treaty and the subsequent political fortunes
of the Byzantine Emperor Michael VIII put Venice in a condition of inferiority in
relation to its rival. For decades the city tried to overcome the disadvantage. After
alternating clashes, negotiations, and compromises, Venetian diplomacy concluded
peace with the Byzantine emperor Andronicus II in 1310. Venice could expand in
the Black Sea region once more.
Tana was located on the Don’s left bank, within the vast delta that the river
forms when it flows into the Sea of Azov. 9 Built near the ancient Muslim Azak, the
settlement was for almost two centuries the easternmost outpost of the Latin mari-
time cities’ entire Eastern trading system. Since the second half of the 13th century,
Tana obtained extraordinary economic and strategic importance. Here, interactions
between the indigenous element (Turkic–Cuman and Mongol) and the Latin mer-
chants was constant. Tana became one of the major centres for the supply of slaves
and a crucial hub for Eurasian commerce, being on the route to the major Eurasian
trading centres: Astrakhan, Saray, Urgench, Bukhara, Samarkand, and of course,
China.
According to the sources, the first Latin officer who worked in Tana is the Gen-
oese consul Ansaldo Spinola, active since 1304.10 The first mention of commercial
galleys sent by the Venetian Senate to the Black Sea (Galee di Romania-Mar Nero) is
in 1306.11 Western merchants resided and worked in Tana from the second half of
7 CARILE 1965, 129–35; CARILE 1965-6, 168–9.

8 BALARD 2017, 529–32.
9 KARPOV 1995, 225–8; PAPACOSTEA 1979, 205–7; BRATIANU 1969, 160-5; SKRŽINSKAJA 1968, 3–10;
BERINDEI – VEINSTEIN 1976, 110–11.

10 BALARD 1978, 151; LOPEZ 1986, 463.
11 STÖCKLY 1995, 105.
the 13th century until 1308, when the khan of the Golden Horde, Toqta (r. 1291–
1312), expelled all the Westerners from the region. In this period, Tana grew and
became an important international trade centre, benefiting from a relatively peace-
ful international context.
Venetian Tana’s history began its second phase characterized by the Genoese
return to Crimea and the Sea of Azov in 1315 and continued until the 1350s Geno-
ese–Venetian war. The Venetian Senate expressly forbade their merchants to dock
in Tana as late as 1330.12 In fact, Venice established its organized community on the
Azov Sea coast in 1332, after a long and exhausting negotiation with Özbeg Khan
(r. 1313–41).13 These decades marked the culmination of the Italian presence in Tana;
Genoa and Venice settled in the Northern Pontus permanently and with a stable
administrative structure.
A third phase goes from the expiry of the ban to frequent Tana (devetum Tane),
imposed by Genoa on its rival after the peace of Milan (1358), until the attack of
Timur (1370–1405) in 1395. In between, there were the well-known clashes of 1343.14
In these years, the Venetian galleys resumed their journeys to both the Black Sea
and the Azov Sea. The frequentation of the settlements was slowly re-established,
and the internal transit routes towards Transoxiana were exploited again. In the
second half of the 14th century, Tana did not lose its strategic and economic im-
portance.
Throughout the 14th century, relations between Venice and the local authorities
evolved on three levels, often overlapping each other. A first institutional level
consisted of dialogue at the highest state level: the Venetian Senate on one side and
the Mongol Khan of the Golden Horde on the other. A second institutional level on
which the local authorities, the Venetian consul in Tana, and the Mongolian gover-
nor of Solgat interacted with one another. Finally, there was a personal level of
daily relations between Venetian citizens and the local population. Commercial
transactions and personal ties were part of this type of relationship.
3. Interactions: from political to personal
On June 19, 1347, the Venetian Senate sent two ambassadors to Tana, to whom a
third ambassador was added directly from the bailo of Constantinople.15 The Vene-
tian diplomats obtained from Janibeg (r. 1342–57), through the mediation of the
Mongol governor of Solgat Sichilbey, the confirmation of the concessions made by
Özbeg in 1332.16 Furthermore, Janibeg allowed the Venetians to use a plot of land
12 STÖCKLY 1995, 106; CESSI 1937, 233.

13 THOMAS 1880, 243–4; TAFEL – THOMAS 1964, 243ff.; HEYD 1913, 751.
14 KARPOV 1994, 121–6; BALARD 1978, 154; PUBBLICI 2017: 25–47; DI COSMO – PUBBLICI 2022: 113–122.
15 THIRIET 1958, n. 201. Janibeg agreed to let the Venetians return to Tana, but the concession
was ratified only in December 1347. THOMAS 1880, 311.

16 On the treaty between Venice and Özbeg, see PUBBLICI 2005, 447–8.
separate from the Genoese, where they could trade.17 Janibeg confirmed the judi-
cial prerogatives to the Venetian consul in Tana, but exclusively for the community
he presided over. Any non-Venetian citizen was excluded from his jurisdiction.
Judicial disputes should be held in the curia, where on September 9, 1363 the con-
sul Niccolò Basilio threatened to fine the banker Pietro Bembo if the latter did not
pay 1,800 besants he owed to him. 18
Janibeg confirmed a large area (ca. 12 square kilometres) where the Venetians
could reinstall their neighbourhood. The Mongol khan granted the Venetian com-
munity living and operating in Tana a tax reduction of 5% to 3%, while gold and
silver were exempted altogether.19 The consul had to manage a potentially danger-
ous situation due to the tensions between Venetians and the local population;
maybe that is why the officer received a substantial salary increase from the 30
grossi in 1334 to 70. Furthermore, the Venetian Senate established that the consul
should receive a 1% on commercial transactions from all Venetian merchants or
those “qui pro Venetis tractarentur”. The levy was justified by the onerous expens-
es incurred for the frequent embassies that the Venetian Comune had to sustain in
recent years and the expenses related to the restoration of damaged infrastructure
and the construction of new houses.
The surviving notarial deeds contain precious information about restoration
and new construction. The Venetian community in Tana was eager for a new
start.20 In their neighbourhood, the Venetians constituted a stable community.
They had built a hospital,21 a public bath, and several churches. One of them was
dedicated to Santa Maria and run by the Franciscans 22 and so was a church dedi-
cated to San Francesco.23 Another church was named after San Marco24 and hosted
17 «Luogho diviso da quello de Zenoessi, da poder far le suo mercadantie»: THOMAS 1880, 311.
18 Archivio di Stato di Venezia (from now on ASV), Cancelleria Inferiore (from now on CI),
Notai 181, fasc. 5, f. 68r. PUCCI DONATI 2019, 163. I have personally studied most of the doc-
uments cited. Where I did not see the original, I cite from the excellent and precious work
conducted by F. Pucci Donati (2019).
19 THOMAS 1880, 243; PEGOLOTTI 1936, 24: «Oro e argento e perle non pagano né comerchio né
tamuga né nullo diritto alla Tana». The Mongols always encouraged the commerce of pre-
cious stones and metals.
20 ASV, CI, Notai 19, fasc. 7, reg. 1, f. 42v.
21 Dedicated to S. Antonio. ASV, CI, Miscellanea Notai, 20, carta 361; Notarile, Testamenti,
Testamenti 924 of 30 July 1364; ASV, CI, Notai, 130, fasc. 7/B, f. 19v.
22 ASV, Notarile, Testamenti, Testamenti, 361, fasc. 32°, f. 129r; ASV, CI, Notai 19, fasc. 7, reg.
3, f. 3r–v; 5v–6v; 6v–7v; 13v–14r; 18r–19r; 130, fasc. 7/B, f. 20r.

23 ASV, CI, 19, fasc. 7, reg. 1, c. 31bis; Miscellanea, Notai, 22, c. 774; PUCCI DONATI 2019, 206,
n. 592.
24 ASV, CI, Notai, 19, fasc. 7, reg. 3, f. 1r-v; 4v-5v; 9r-11r; 14v-15r; ASV, CI, Notai 117, fasc.
Marco Marzella, f. 2r. PUCCI DONATI 2019, 154, n. 435; 170, n. 481 and 178, n. 487. In hist tes-
tament, the Venetian Primasio di Ragusa writes that it was customary for the Venetians to
the confraternity of Santa Maria and San Antonio. 25 A church of San Giacomo was
popular, too.26 Most of the religious buildings had a fabrica in charge of renovations
and restorations, such as the church of San Domenico. 27
The church of San Marco in Tana was damaged at least until May 1363 when
the merchant Benedetto di Romagna dictated his testament and left two silver
sums for its restoration. He also left one sum for the fabrica of the churches of San
Domenico and Santa Maria. On July 1364, Gerardo di Castello left the fabrica of
Santa Maria (where he wanted to be buried) three sums. 28 In general, it was up to
the consul to cover all the costs for the city “remittendo residuum Venecias per
incantum dominio vel officialibus furmenti”. 29
In the late 1340s, tension between Venice and Genoa increased. It was the prel-
ude to the Venetian–Genoese war of the 1350s. On May 19, 1348, the Venetian Sen-
ate ordered Giustiniano Giustiniani, captain of Romania’s galleys, to take shelter. 30
The State auctioned the Romania-Black Sea galleys that year, but only on the condi-
tion that a prohibition to pass the straits was put in place. 31 All the attempts to
avoid the war failed, and on August 28, 1350, the Venetian Senate ordered the con-
sul in Tana to send a negotiator to Janibeg and warn the merchants who were still
in Caffa, controlled by Genoa, to secure themselves in the areas under Venetian
authority.32
leave one silver sum to the fabrica of the church of S. Marco in Tana: PUCCI DONATI 2019, 179,
n. 488.
25 ASV, CI, Notai 19, fasc. 7, reg. 3, ff. 7v-8v; 12r-13r; PUCCI DONATI 2019, 171, n. 482.
26 ASV, CI, Notai 19, fasc. 7, reg. 3, f. 2r-v; 7v-8v; 20r-23r; PUCCI DONATI 2019, 140, n. 413.
27 PUCCI DONATI 2019, 142, n. 415 and 18, n. 489.
28 ASV, CI, Notai 19, fasc. 7, reg. 3, ff. 25r–26r; ASV, CI, Notarile, Testamenti 924, 30 luglio
1364; PUCCI DONATI 2019, 170. The sum (sommo) was the most diffused means of exchange
among the merchants who operated in the Ulus of Jochi and constituted the unit of measure
of monetary reference and weight. The sum was the expression of the partially monetized
economy of the Golden Horde and consisted of an ingot of silver of variable weight but
settled around 200 grams (at the time of Pegolotti the sommo weighed 206 grams). It circulat-
ed also in China and was produced by local mints. PEGOLOTTI 1936: 41. See on this MUELLER –
LANE 1985: 163 and n. 6; KURODA 2009: 261.
29 ASV, Senato Misti, XXIV, f. 114; THOMAS 1880, 340–1.
30 ASV, Secreta, reg. B, ff. 7v e 8r; THIRIET 1958, n. 211. Tension is confirmed by another doc-
ument of November 1349, in which Pietro Tagliapietra, captain of Bitici Niddo’s galley, is
condemned in Venice because he had attacked the Mongols in the port of Varango (Varan-
golimen in Western Crimea). Tagliapietra believed that those Mongols had robbed his ship,
but he was wrong (KARPOV 2000, 181).
31 THIRIET 1958, n. 239.
32 In August 1349 the notary Marino Grifoni was active in Tana, where he stipulated a con-
tract between Nicoletto Gato and Bonavere Albani (ASV, CI, Notai 19, fasc. 7, reg. 1, ff. 25.v–
26r). THIRIET 1958, n. 247. On the Venetian-Genoese war see also ASV, Procuratori di San
Marco, Misti, Commissarie, busta 2: it contains letters and receipts by Marco Nani. They are
The war between the two republics lasted for five years with alternating phases.
After the battle of the Bosporus, in early 1352, the two fleets met for the final clash
in 1354. In Portolungo, near Modone on the Peloponnese, the Genoese admiral
Paganino Doria defeated the Venetian galleys of Nicola Pisani. Genoa and Venice
signed the peace treaty in Milan in 1355 (Genoa was under the Visconti rule). The
Peace of Milan decreed that Venetians be expelled from the Black Sea region for
three years, but Genoa proved incapable of imposing its conditions despite having
won the war.
At first Genoa and Venice respected the agreement reached in Milan 33 and
avoided sending their merchants to Tana. Nevertheless, it was too expensive for
both of them. The massive investments made in the region to guarantee an infra-
structural system adequate for their communities and the considerable profits from
the trade with the Levant forced the two Italian republics to re-establish contact
with the Mongol authorities.
On July 28, 1355, Venice sent an ambassador to the Mongol noyon Ramaḍān,
who supervised the region’s affairs. The Venetian diplomats asked to return to
Soldaïa, and also for the confirmation of the commercial taxes at 3%, and the libera-
tion of two Venetian merchants arrested by Ramaḍān.34 Janibeg was cold in the
response and granted only the small port of Provato, west of Tana and too close to
Caffa to represent a gain on the Venetian side. 35 However, the Venetian ambassa-
dors had no other choice but to accept, and the captain of the galleys was entrusted
with controlling the territory of Provato where the Venetian emporium would rise.
It was a situation still in evolution, and the Senate forbade the galleys of Romania-
Black Sea, auctioned in May 1357, to cross the straits.
Venice and Genoa returned to Tana only in 1358, after the expiry of the devetum.
Both cities benefited immediately from the problematic political situation within
the Golden Horde. Janibeg had died in 1357, and his successor was his son Ber-
dibeg (r. 1257–9) who was well disposed towards Westerners.
Berdibeg confirmed to the ambassadors Giovanni Quirino and Francesco Bono
the concessions made by Özbeg and Janibeg and renewed them, 36 but increased
taxation on commercial transactions to 5% for everyone. Equally, the 5% tax re-
mained in force for all goods to the weight (per cantaro). Every ship entering the
port had to pay three silver sums to the Mongol governor. 37 To avoid further prob-
lems with the Venetian community Berdibeg reduced the judicial prerogatives of
very interesting because Nani, a merchant with a prevalence of interests in Cyprus, was
captain of Pancrazio Giustiniani's galley during the war.
33 F. THIRIET 1953, 219–45; THIRIET 1959, 176–7; BALARD 1978, 85–6.
34 THIRIET 1958, n. 273; THOMAS 1880, 24.
35 THIRIET 1958, n. 299; the concession is dated March 1356. Ramaḍān confirms to the Vene-
tian the port of Provato «che a nome Citade Nova»; see also THOMAS 1880, 25.
36 THOMAS – PREDELLI 1899, 47–51; BALARD 1978, 154; THIRIET 1958, n. 311.
37 THOMAS – PREDELLI 1899, 50.
the consul who, in case of a conflict, had to consult first with the Mongol governor
of Solgat (“lo signor de la tera e lo consolo insembre”). In May 1358, the Senate
auctioned Romania’s galleys that this time were allowed to cross the straits and
enter the Black Sea. It was the first time after almost a decade.38 The captain of the
galleys sent to the Black Sea in September 1359 was Giovanni Priuli. 39
The new consul in Tana was most likely Leone Bembo 40 who was replaced in
the next year by Pietro Caravello. The Venetian consul in Tana worked with his
familia41 and a modest military garrison capable of guaranteeing the consular pre-
rogatives, among which the administration of justice and public order were promi-
nent. The familia consisted of an interpreter, a cellarius42, one or more bastonarii43, one
or more famuli, a council,44 a nuncio (commandador),45 and a doctor (phisicus).46 The
interpreter was a crucial function in Tana as the consul needed to communicate
with the local population. In 1359, the function was performed by Guglielmo Bon. 47
The communities living in Tana had their own living quarters, with the Vene-
tians and the Genoese having a fenced or fortified area under the consul’s authori-
ty. In the 1360s, in Tana there were a Jewish, Armenian, and even an Alan neigh-
bourhood.48 Merchants from other regions were allowed to settle in the Venetian
38 THIRIET 1958, n. 328. This time the incanti were expensive, more than 160 liras of grossi,
while the average was rarely above 50 liras in 1348). Maybe the Venetian merchants partici-
pated in the auction because the Mongol concessions had generated optimism.
39 ASV, CI, Notai 19, fasc. 7, reg. 1, ff. 4r of 16 September, 5v of 19 September and 6r of 20
September 1359.
40 ASV, CI, Notai, 19, fasc. 7, reg. 1, f. 5r of 18 September 1359.
41 One of Caravello’s famuli in December 1359 was Giacomo Furlan, son of Francesco di
Valvasone who was related to the consul. Giacomo replaced Marco Grimani, a Greek from
Crete, and received a salary of 130 aspers. ASV, CI, Notai 19, fasc. 7, reg. 1, f. 14r.
42 In 1359, this important task was carried out by Francesco de’Garfaldi di Bologna. ASV, CI,
Notai 19, fasc. 7, reg. 1, fol. 12r of 30 October 1359. In 1383, the interpreter of the Venetian
Curia in Tana was Pietro detto Gata (ASV, CI, Notai 130, fasc. 7/B; PUCCI DONATI 2019, 184.
43 ASV, CI, Notai 19, fasc. 7, reg. 1, f. 5r of 18 September 1359.
44 In 1359, the Council consisted of eight nobles: Giacomo Contarini, Giovanni Bembo, Nic-
colò Renier, Bartolomeo Loredan, Leonardo Falier, Andreolo Bragadin, Luca Contarini, and
Guglielmo Bon, who was also the interpreter. ASV, CI, Notai 19, fasc. 7, reg. 1, f. 13v of 14
December 1359.
45 In 1360, the nuncio of the Venetian curia in Tana was Benedetto di Romagna.
46 ASV, CI, Notai 181, fasc. 5, f. 66bis; PUCCI DONATI 2019, 159.
47 ASV, CI, Notai 19, fasc. 7, reg. 1, f. 4v; 7v; 8v and 12v.
48 In September 1360, the tabernarius Ianixinum lived near the Alan bath (ASV, CI, Testamen-
ti 17). The Alan bath is also mentioned in the testament of the Genoese Andalò Basso, draft-
ed in Tana in November 1362 (ASV, CI, Notai 19, fasc. 7, reg. 3, ff. 20r-23r.). A contrata iu-
daorum is mentioned in a document from July 1366. ASV, CI, Miscellanea, Notai Diversi,
busta 134bis, Contratti di schiavi, Notaio Francesco di Boninsegna di Strada di Mantova and
PUCCI DONATI 2019, 174, n. 485. A contrata arminorum is mentioned in two documents: ASV,
CI, Notai 19, fasc. 7, reg. 1, f. 36r and PUCCI DONATI 2019, 158, 441.
neighbourhood. In September 1360, Ser Marco di Ser Viviano sold a house in Tana
for an exuberant amount of money: 24 sums. The property bordered an Armeni-
an’s house. Coza Azillyas, Saracenus, gave for rent a plot of land with a cellar (fo-
vea) to Niccolò Cornaro. The property was next to Azillyas’s house. 49
Consul Caravello travelled to Tana with his son Luca. 50 It was customary for the
Venetian merchants to travel with their sons and teach them the profession they
would have inherited from their fathers. On September 25, 1364, Lucia Zabacha
entrusted her son Andrea to Giacomello Marino to teach him the mercantile pro-
fession. Giacomello undertook to provide the young man with food, accommoda-
tion, clothes, and footwear for ten years.51 The Florentine merchant Brandaia di
Brandaia, Venetian resident and living in Tana, made a testament on August 7,
1359. He wanted his two sons Cristoforo and Domenico to frequent the scholae
gramamticae, to learn how to read, write, and do accounting. 52
For the Venetians, it was necessary to dialogue with a “coherent” interlocutor.
The assassination of Beridibeg in 1359 opened a political crisis within the Golden
Horde, culminating with the coup of Nawrūz in February 1360. 53 The Venetian
authorities became concerned and worked to obtain guarantees from the Mongols
on the settlement’s safety. In December 1359, the consul’s council nominated two
nobles, Giacomo Contarini and Giovanni Bembo, as representatives to Taydula
Khatun, grandmother of Berdibeg.54 Contarini and Bembo were charged to address
the empress to obtain guarantees on the settlement’s security and the annulment of
the penalty imposed as a condition to return to Tana. 55 In September, Giacomo
Cornaro is mentioned as ambassador to “the Tartar khan”.56
Despite the difficult political situation, the Venetian return to Tana in the late
1350s revived trade. The volume of business set up by the merchant communities
in Tana and the growing difficulties of travelling East of the Volga required the
exploitation of local resources and collaboration became inevitable. The notarial
deeds produced in Tana contain numerous cases of contracts stipulated between
Latin merchants and locals or Muslim merchants, likely from Transoxiana and the
Ilkhanate. The latter were the majority, together with the Armenians. On July 31,
49 ASV, CI, Notai 19, fasc. 7, reg. 1, carta sciolta.

50 ASV, CI, 19, fasc. 7, reg. 1, f. 7r of 21 September 1359 and f. 9r of 22 September 1359. On
the same day, Luca nominated his father procurator of all his business (omnia sua negotia) in
Tana. Ibid., f. 9r. In 1362, the new consul was Giacomo Contarini. ASV, CI, Notai 19, fasc. 7,
reg. 3, f. 20-23r.
51 ASV, CI, Notai 130, fasc. 7/B, f. 15r.; PUCCI DONATI 2019, 189–90, n. 528.
52 ASV, CI, Notarile, Testamenti, Testamenti, 361, fasc. 32°, f. 129r.
53 KARPOV 2018, 532.
54 Taydula was Özbeg’s wife. THOMAS – PREDELLI 1899, 47–50; DE NICOLA 2017, 149; FAVE-
REAU 2016, 45.

56 ASV, CI, Notai 19, fasc. 7, reg. 1, f. 43r.
1360, the Avar Corthoha del fu Manzar from casale Balgima in imperio Gazarie was
sold to the forty-five-year-old Paduan Michele Coperi.57 On August 25, 1360, Coza
Mansuth Saracenus sold to the Venetian Pietro Penzi 870 “salted” and 348 dried
cowhides.58 As we have seen above, on September 13, 1360, Coza Azillyas Sara-
cenus and habitator in Tana rented to Niccolò Cornaro a house, a cellar (fovea),
warehouses, and a plot of land. Among others, the property bordered with another
Muslim merchant’s home, Coza Monsafil, whose land “is not cultivated”. Another
Muslim, Coza Mansuth, received a share of the income generated by the proper-
ty.59 On September 18, 1360, the Venetian Giovanni di Pistoia purchased some fab-
ric from Urgench (Organza) from Muslim merchants who were selling it in a public
bazaar.60 On September 21, 1360, Ziachmach Thoulu, Saracenus and habitator in
Tana, sold to Pietro Bembo a twelve-year-old Tatar slave.61 In his will, the Genoese
merchant Andalò Basso nominated an Armenian from Solgat, Gabrieto Balochzi, as
his partner.62 On September 13, 1363, Macometh Coza, Saracenus, sold to Niccolò
Basilio a big property consisting of a non-cultivated land and a bath in Tana’s Ve-
netian neighbourhood.63
On July 21, 1365, Costanzo di Lodi di Candia dictated his last will and testa-
ment to the Venetian notary Nascimbene Scarena in Tana. He had a credit of 50
aspers with Tolach and Mansafir Saracenos for some cowhides.64 In August 1365, an
Alan named Satalario received a loan of 100 sums from Pietro della Fontana, Vene-
tian consul in Tana, to purchase merchandise. 65 On August 14, 1365, Usuf Saracenus
sold a cellar (fovea) to the Venetian Giovanni.66 Naturally, cases of solidarity be-
tween people with a common ethnic background were frequent. On August 12,
1365, a Jewish man called Thomas nominated his procurator Calli Bila, another
Jew, to collect 100 aspers from a certain Anastasio Pangalo, his debtor. 67 On Sep-
tember 12, the same Anastasio Pangalo gave a loan of 3 sums to Mosé Calazi; both
were Jewish.68
57 ASV, CI, Notai 19, fasc. 7, reg. 1, f. 29v. Pucci Donati 2019, 65.
58 Cuoia salate and vaccine. ASV, CI, Notai 19, fasc. 7, reg. 1, f. 34r.
59 ASV, CI, Notai 19, fasc. 7, reg. 1, carta 36bis.
62 ASV, CI, Notai 19, fasc. 7, reg. 3, ff. 20–23r.
63 ASV, CI, Notai 181, fasc. 5, f. 68v.; PUCCI DONATI 2019, 164. Basilio bought the property for
a total of 130 sums and sold it almost immediately (22 September) to the Venetian consul in
Tana Pietro della Fontana for the same amount: ASV, CI, Notai 181, fasc. 5, f. 69r; PUCCI
DONATI 2019, 165.
64 ASV, Notarile, Testamenti, Testamenti, 924, 21 luglio 1365; PUCCI DONATI 2019, 172 n. 483.
65 PUCCI DONATI 2019, 160 n. 451.
66 PUCCI DONATI 2019, 161 n. 454.
67 PUCCI DONATI 2019, 161, n. 452.
68 ASV, CI, Notai 181, fasc. 5, f. 68v; PUCCI DONATI 2019, 164 n. 464.
Intermediation among merchants of diverse origin and cooperation was fre-

quent not only for buying commodities but also for renting facilities. On April 28,
1360, Ser Grava di Stanzi, from Samsun, gave his boat with 21 sailors for rent to the
Venetian Giovanni Bembo. Giovanni needed the boat to go to the charger port
(caricatoio) called Aziachon to load 460 moggia of grain and trade them in Pera. 69
Trade between Latins and local merchants was widespread, but non-Venetians
preferred to have a fellow citizen or coreligionist in front of the notary. On July 22,
1360, Marino de Ruosa agreed to pay the loan he had with Eza Aly Saracenus. Two
Venetians and three Saracenos de Tana: Adula, Coza, and Azibey were witnesses to
this act.70
Nevertheless, the friction between Venetians and the local population did not
completely go away. In the testament she dictated to the notary Benedetto Bianco
on August 11, 1360, Elisabetta Alban destined two sums for the liberation of her
son Giovanni, indebted to certain Tartari who had captured him.71
In 1363, the commercial activity in Tana focused more and more on slave sales.
In particular, the sales of slaves by locals to Venetian merchants intensified. On
August 17, Donna Erda, widow of the Tartar Bephramir, sold to the Venetian mer-
chant Sandro Lovato, a twelve-year-old slave. The next day, the Jewish merchant
Burdoch del fu Nodin sold a twelve-year-old Tartar slave on behalf of Ser Giovan-
ni Selvaggio to Lovato. On the same day, the Tartar Brunach del fu Zerzi sold a
Tartar slave to a Venetian merchant whose name is not readable in the document. 72
On August 20, 1363, the Tartarus Chonach del fu Chotluboga sold his twelve-year-
old daughter to Sandro Lovato.73
In these years, the Muslim merchants appear to be the most active as intermedi-
aries in the sales of slaves. On September 26, 1362, Chalil Charchaulli Choicholba,
Saracenus, sold a Circassian girl to Lorenzo Querini. 74 Zarubey del fu Emin, Sara-
cenus, sold an eleven-year-old Tartar slave to the Venetian merchant Stefanello
Bonfiglio.75 On September 4, Assam del fu Audul, Saracenus, sold a seven-year-old
Mongol slave (genere mongalorum) to the merchant Francesco del fu Bruzese from
Rimini.76 Ramadan del fu Cusma, Saracenus, sold a whole Mongol family—father,
wife, and a six-year-old child—to the Venetian merchant Giovannino Bembo. 77 On
September 6, Adia, widow of Audul, Saracenus, sold a thirteen-year-old Tartar
69 The freight is established in 940 perpers payable within twelve days from the arrival of the
grain in Pera. ASV, CI, Notai 19, fasc. 7, reg. 1, f. 18r.
71 ASV, CI, Notai 19, fasc. 7, reg. 1, carta 31bis.
74 ASV, CI, Notai 117, reg. Marco Marzella, f. 2r.
slave to the Venetian merchant Niccolò Barbo. 78 On September 8, Cheder del fu

Azo Saracenus sold a sixteen-year-old slave to Enrico Barbarigo, who was a Vene-
tian.79 On September 9, Baziman del fu Bachomuth, Saracenus, sold a ten-year-old
Tartar slave to the Venetian merchant Fantino Zorzi.80 On September 10, Adia,
widow of Charchibas, Saracenus, sold a fourteen-year-old slave to Marco Carlo; the
same day, Besboga del fu Siaban Saracenus sold a fifteen-year-old slave to Pietro
Bon.81 On September 11, Jaylmis del fu Hacboga, Saracenus, sold a ten-year-old boy
to Ser Tommaso Falier.82 On September 17, Macometh del fu Ozi Saracenus sold an
eleven-year-old Tartar boy to Giovanni di Vidor. The same day, Abdura Coman
del fu Mamussia Saracenus sold four slaves, two girls and two boys, in four sepa-
rate documents to the Venetian merchant Niccolò Baseggio, for a total sum of 1,375
aspers.83 Chotur del fu Chozuboga, Saracenus, sold an eleven-year-old boy to
Bonaccorso di Vanni.84 On September 18, Monsafor del fu Azibey, Saracenus, sold a
twelve-year-old boy to Niccolò Baseggio. 85 On September 19, Hachboga del fu
Siansadin, Saracenus, sold a fifteen-year-old girl, a ten-year-old boy, and a fourteen-
year-old girl—all Tartari—to Niccolò Baseggio.86
Venetian merchants and local people often came into contact during the sale of
slaves not only with professional traders, but with families in the necessity to sell
their children to alleviate the hardship they lived in. In notarial documents the
cases are numerous. On June 3, 1360, a Russian woman, Ocholinato, married to
Dmitrij (de casale Insbleymamat de Rusia), sold her fifteen-year-old daughter to the
Venetian merchant Bertolino Magnamosto. 87 On June 29, Apanas, a Tatar woman,
sold her thirteen-year-old sister to Marco Contarini. 88 On July 2, Anecoza, a Tatar
who lived in the Tuman Melicheli just outside Tana (in districtu Tane), sold a four-
teen-year-old son to Marco Zaccaria.89 Bech del fu Thuboga sold his fourteen-year-
old niece to Albertino de’Conti.90 Donna Cotlucaton del fu Zaccaria di Tabriz and
Sarchi del fu Cherin sold their twelve-year-old son to Giorgio Colito from Thessa-
loniki.91

83 ASV, CI, Notai 19, fasc. 7, reg. 4, f. 14v and 15r.
86 ASV, CI, Notai 19, fasc. 7, reg. 4, f. 18v, 19r, and 20v.
The Venetians who were established for a longer period in Tana bought serv-
ants. They generally “rented” the services of a young boy or girl from locals in
need of money. For example, on October 16, 1359, Anchau and Daniele, both de-
fined as Alani, received 300 aspers from Giovanni Bembo for a young person who
was described as being “protected” by Anchau for the time Giovanni would stay
and work in Mongol Crimea (in imperio Gazarie).92 Elya, Alanus, received from the
Venetian merchant Andrea Armino a 300 aspers loan for which he left his fifteen-
year-old son as a pledge. The boy had to stay in service in Andrea’s house until the
loan was repaid.93
Mixed marriages were not too frequent but occurred in Tana and all over the
Latin settlements in the Black Sea region. The Genoese Giorgio, resident in Soldaïa,
married Teodora, who was Greek. 94 Francesco di Bonavere Alban, Venetian citizen,
married Cotlu.95 On April 14, 1385, an Armenian named Antonio declared that he
had received the entire dowry of his wife, Bartolomea di Graziano from Treviso.96
On October 21, 1385, Domenico Balta affirmed that he had received the entire dow-
ry of his wife, whose name is Beymolich.97
The mouth of the Kuban river and the entire area around the Kerch Strait had
been of fundamental importance since the time of the Greek colonies. 98 For centu-
ries, Kievan Rus’ exploited the fluvial arteries (the Dnepr above all) for interna-
tional trade with the Byzantine Empire to the south and the Baltic markets to the
north. For the Georgian kingdom, the Black Sea outlet was, together with the land
routes to Persia, a formidable instrument for commercial development.
The area around Solgat was known for wine and the Venetian merchants pur-
chased it in Tana from local producers or intermediaries and resold it in Venice.
On December 14, 1359, Perino d’Ognibene bought two barrels of wine mosti de
sorcati from Coza Macomuth di Sadradin, from Solgat. 99 On March 2, 1360, the
Greek Giorgio di Enno agreed to pay back a loan of 220 aspers within three months

94 It is not surprising that the witnesses of the document are all Greek: Marco Grimani from
Crete, Giorgio Colito from Thessaloniki, and Sana di Caraboga Paraschiva from Trebizond.
ASV, Notai 19, fasc. 7, reg. 1, f. 14v.
95 ASV, CI, Notai 19, fasc. 7, reg. 1, c. 31bis; PUCCI DONATI 2019: 69. The woman was likely
baptized, since here name changed into Elizabetta.

96 ASV, CI, Notai 130, fasc. 7/B, f. 15v; PUCCI DONATI 2019: 193.
97 ASV, CI, Notai, 130, fasc. /B, f. 17v; PUCCI DONATI 2019: 201. They had a daughter named
Culmelich.
98 From October 2 to 15, 1362, the Venetian galleys were docked in the port of Kuban, where
the notary Marco Marcello drafted seven documents: ASV, Notai 117, reg. Marco Marzella,
ff. 2v–3r; 3v; 4r and 4v.
to Teodoro, a merchant from Solgat, for a certain quantity of wine he had

bought.100
Despite the difficulties and the scarce convenience of long journeys some mer-
chants went further and settled deeper into the Golden Horde, entrusting on in-
termediaries the purchase of merchandise farther East. Marco dell’Orsa was a very
active merchant in Tana, but lived in Saray. 101 On September 24, 1360, Giovanni
Barozzi declared in front of the notary Benedetto Bianco that he had left precious
goods in Saray at the merchant Coza Azi, who was a banker. Among the commodi-
ties, there were silk, pearls, and jewels. 102 On May 22, 1362, Benedetto di Romagna
dictated his will to notary Benedetto Bianco on the occasion of his journey to
Saray.103 In August 1378, Niccolò d’Arpino—a Venetian merchant active in Saray—
sold honey and slaves for a total value of 840 besants to Marco Nani, Venetian as
well, who was supposed to sell it in Tana. Marco had paid only 251 besants, but
never paid the rest to Niccolò. The consul, after having evaluated the case, con-
demned Marco to the payment of all his debts and expenses for a total of 689 bes-
ants.104
It was uncommon for the Venetian merchants to travel farther than the capital
of the Golden Horde. Nonetheless, in the 1380s they still had business in the Volga
area and used intermediaries to transport their merchandise. In his will drafted in
Tana on May 14, 1384, the Venetian merchant Primasio di Ragusa arranged for a
consignment of carobs to be sent to Astrakhan with the first available caravan.105 In
March 1383, the Venetian merchant Giovanni Servodio resided in Astrakhan. 106
Another interesting case is that of the Venetian merchant of Florentine origin Do-
menico, who purchased and sold a lot of slaves. One slave, an eleven-year-old girl,
was from China (genere cataynorum).107
Sometimes things went wrong and merchants could lose everything. On Sep-
tember 23, 1359, Ser Micheletto Emo, merchant in Tana, appointed as his procura-
tor Ludovico da Molin, his fellow citizen. Ludovico had to manage the seizure of
Micheletto’s goods, who had lost everything in the shipwreck of the galley of Ser
Nicoletto Bernardo.108 In his rich testament, the Genoese merchant Andalo Basso
mentioned a bottomry stipulated for 50 sums with Cosma dello Levante, likely a
Greek shipowner. The ship charged to transport the commodities to Constantino-

101 ASV, CI, Notai 19, fasc. 7, reg. 1, f. 24r-v.
103 ASV, CI, Notai 19, fasc. 7, reg. 3, ff. 25r-26r.
104 ASV, CI, Notai 189, n. 5, 24 aprile 1380; PUCCI DONATI 2019, 176 n. 486.
105 ASV, CI, Notai 130, fasc. 7/B, f. 19v.; PUCCI DONATI 2019, 179, n. 488.
106 ASV, CI, Notai 130, fasc. 7/B, f. 13v.; PUCCI DONATI 2019, 184, n. 505.
ple shipwrecked and all the cargo was lost. 109 On September 18, 1363, Ser Marco
Mariga nominated Pietro Morosini as his procurator to recover all his goods from
the shipwrecked cog named Dolfina.110 According to the document mentioned
above, the Muslim banker Coza Azi Aza had to guard Giovanni Barozzi’s goods
(silk, pearls, and jewels) and transport them from Saray to Tana, but during the
trip he was robbed of everything for a total value of 50 sums. 111
The political difficulties within the Golden Horde affected the life of the com-
munities living in Tana. The Mongol defeat at Kulikovo in 1380 at the hands of
coalition of Rus’ principalities marked the beginning of a process of internal weak-
ening that did not stop in the years to come. The Mongols gradually lost the regu-
lar inflow of money from the tributes paid by Russian princes. Yet, as we have seen
from notarial documents in the 1360s, the Venetian community in Tana along with
the others who used to interact with it was still well organized and lively. The situ-
ation changed, however, in the following decade.
After the expulsion of all the Westerners from Tana by Janibeg in 1343, 112 the
navigation in the Black Sea was interrupted; in this period the problems accumu-
lated after the war between Genoa and Venice in the 1350s. When, after 1358, the
voyages resumed, the average price of the incanti, after an initial enthusiasm, was
maintained on an average level (about 60 liras), but already in 1363 it increased
considerably (with more than 94 liras on average for each galley).113 In 1368 and
1369, the galleys directed to the Black Sea diminished from 6 to 5, and the follow-
ing year from 5 to 4, which were eventually auctioned.
The central power of the Horde, led in these years by Mamai, imposed a 4% tax
on merchandise to the Venetians, whereas the Genoese paid 3%. On June 3, 1369,
the Venetian Senate sent the consul of Tana to the Mongols asking for fair treat-
ment.114 To deal with the political instability, the Venetian authorities in Tana in-
creased their defence expenses. In July 1370, the Senate decided to send arms and
money to the consul to have the lobia repaired and the fortifications restored. Fur-
thermore, he was given 36 sums to pay for an interpreter. 115 Between 1374 and
1376, the entity of the incanti collapsed; the news of poor security within the bor-
ders of the Golden Horde had probably reached Venice and the participants in the
auctions.116
109 ASV, CI, Notai 19, fasc. 7, reg. 3, ff. 20–3r.

112 PUBBLICI 2017: 25–47.
113 THIRIET 1958, n. 540 and 561; STÖKLY 1995, 110–11 and 371–4 /tables); KARPOV 2000, 91.
114 THIRIET 1958, n. 476.
115 THRIET 1958, n. 488.
116 KARPOV 2000, 102; for a general description of events, see KARPOV 1978, 102–9.
The 1370s were also marked by tensions between Venice and the Empire of
Trebizond, the other fundamental port for the Venetian Black Sea trade.117 A new
conflict between Genoa and Venice worsened the situation. The war, known as the
war of Chioggia, lasted from 1376 to 1381 and ended with the peace of Turin. 118 In
this period, navigation in the Black Sea was almost completely interrupted. In June
1377, only 3 galleys were auctioned in Venice. Only two of them were assigned,
but on August 7 the Senate decided not to let them leave because it was considered
too dangerous.119 Nonetheless, Venice maintained its community in Tana whose
consul in April 1380 was Donato Moro. 120 According to the agreements signed in
Turin, Genoa and Venice imposed a blockade of navigation in the Black Sea for
three years,121 but already in 1382 the ban was broken by Venice, which sent two
armed galleys across the Straits. The trips to Tana resumed regularly only from
1383.122
On July 24 of that year the Venetian doge sent two ambassadors to the Tartar
Governor of Crimea (“excellentissimum dominum Imperatorem Tartarorum”) to
ask for confirmation of the agreements still in place formally. The diplomats had
the order to obtain a reduction of the commerchium to 3% and if this was not possi-
ble, to fix it at maximum 5%.123
Relations between Venetian merchants and the local community in the 1380s
were regular, with a marked prevalence of the buying and selling of slaves. In De-
cember 1383, Karomerot, habitator in imperio Gazarie, sold a Russian girl to Pietro
Premarin.124 In July 1384, Caterina Alana was hired by the Venetian merchant Gio-
vanni Vettori as his servant.125 Giovanni wanted to open a tavern and put Caterina
there to work. Venetians and locals continued to collaborate for many diverse
businesses. On September 5, 1384, Francesco Catalano from Barcelona rented his
boat (grippo) to three Circassian merchants—Giorgio Laterano, Sgromalio, and
Caloiano—for 70 perpers. According to the deal, once the caravan with caviar and
salted fish arrived in Tana, they had to load 200 foschi of caviar on Francesco’s ship
(grippo) within ten days.126 On January 16, 1385, the Circassian Reboga and a Tartar
117 KARPOV 1986, 99–101.

118 See the Chronicle of Daniele Chinazzo: CHINAZZO 1958, in particular 200–35.
119 THIRIET 1958, nn. 588, 589 and 593. The previous year, the galleys of Romania had to be
escorted to Tana by the galleys of Marco Giustiniani, who then waited for their return to
Constantinople; between September and October 1376, the casus belli that would cause the
war of Chioggia was being created: CHINAZZO 1958, 19–20.
120 ASV, CI, Notai 189, n. 5, documento del 24 aprile 1380; PUCCI DONATI 2019, 176.
121 CHINAZZO 1958, 209–10.
122 ASV, Senato Misti, XXXVIII, f. 34v; THIRIET 1958, n. 649; KARPOV 1986, 101.
123 THOMAS – PREDELLI 1889, 188-90.
124 ASV, CI Notai 130, fasc. 7/B, f. 13vr; PUCCI DONATI 2019, 184, n. 502.
125 ASV, CI Notai 130, fasc. 7/B, f. 14r; PUCCI DONATI 2019, 187, n. 518.
126 ASV, CI, Notai 130, fasc. 7/B, f. 14v.; PUCCI DONATI 2019, 188–9, n. 524.
merchant whose name is illegible in the document signed a contract to deliver to

Matteo Bugari from Crete five thousand maghrebi (salted fish) for 15 aspers per one
hundred. The payment is divided into two parts: half now and half in a month.127
On January 16, 1385, Sich Achsicoza and Assan, both Tartars, sold 2000 maghrebi to
Matteo Bugari of Candia. The price was 15 aspers per 100. Temir Colunzach and
Bonzos, both Tartars, sold to Matteo Bugari of Candia 2000 maghrebi for 15 aspers
per 100. On May 15, 1385, Cozichar, son of Susut, and his brother Ianbas sold to
Giovanni de Besagna a relative of theirs as a servant for seven years. 128 On August
16, 1385, the Florentine merchant Matteo bought a property from Donato di Porto,
a Venetian. The property was located outside the Venetian district in Tana, by the
moat, and it bordered, among others, with the homes of Chotulboga, evidently a
Mongol.129 In 1386, the consul was Francesco Bragadin and the interpreter Pietro
Gata. The service of Bragadin ended in the summer of 1386.130
The analysis of the documents shows a progressive localization of relations in
the 1380s. Interactions between Venetians and Mongols continued to be steady and
fruitful on a personal level but became increasingly complicated on an institutional
one. In these years the trips of the galleys were regular, and the amount of the in-
canti confirm a lively interest in trade with Tana. Timur’s attack on the settlement
in 1395 disrupted everything again, but did not discourage the Western merchants
who rebuilt their area and continued to frequent Tana until the Ottoman invasion
of 1475.
Conclusions
The encounter of the “new ethical system” of the Italian medieval mercantile class
was particularly fruitful with those peculiar realities whose ethnic and cultural
diversity was more accentuated, for example, in Spain, where “the Islamic world...
was mediator between Christianity and Judaism”.131 The dominant culture of
Christian-Roman Europe was both the source and the limit of the new mercantilist
tendencies. In the Pontic region, a hinge between cultures, where an ethnic com-
plex coexisted, the consequences were even more differentiated. The merchant’s
utilitarian conception, a predominant element of Western emigration to the East,
was faced with peoples whose cultural background was extremely fragmented.
The nomadic element coexisted with the Muslim merchant from Central Asia; the
Minorite and the Dominican lived in convents in the heart of the Steppe, etc.
The Western merchant was confronted with this stratification. It was mainly the
emerging urban bourgeoisie—in Venice, often families of ancient noble origin—
127 ASV, CI, Notai 130, fasc. 7/B, f. 15r.; PUCCI DONATI 2019, 191–2, n. 537.
128 ASV, CI, Notai 130, fasc. 7/B, f. 15v.
129 ASV, CI, Notai 130, fasc. 7/B, f. 16v.
130 ASV, CI, Notai 130, fasc. 7/B, ff. 17v and 18r.
131 FUMAGALLI 2003, XII.
who emigrated to the East, and they had nothing in common with the steppe peo-
ples and their traditional roots. In the Levant came the cives, citizens, with their
urban associative model.
Nomadic solidarity—based on the clan—found itself in front of a new associa-
tive principle in turn based on the city of origin. Western merchants, representing
the urban growth of the late 13th century, gathered around the consul (or podestà) as
the representative of that collective power in which they identified themselves and
drew the synthesis of their status as cives. In comparison, in the nomadic model
there was a heterogeneous social fabric in which the “unstable” dynamic element
was closely linked to that of the atavistic bond of descent; in both cases, it was a
constructed model.
The merchant of the 14th century was professionally prepared and equipped to
live away from home. The Eurasian nomads knew the trade well. The establish-
ment of the Mongol Empire brought Asia closer to Christian Europe and attracted
new capitals from the East. From the second half of the 13 th century, the Asian
trade system, with its western centre of gravity in the Black Sea region, became
increasingly integrated into the Mediterranean trade system. The construction of
emporia on the coasts of the Black Sea and the travels of Venetian merchants to dis-
tant lands brought people with different histories and cultures into contact. Com-
munities lived together peacefully, traded, met, married, fought, and widened each
other’s horizons. After the 14th century, the Eurasian continent would never be the
same again.
Bibliography
BALARD, M. (1978): La Romanie Génoise (XII–debut du XV siècle), Genua.
BALARD, M. (2017): 1261: Genova nel mondo: il trattato di Ninfeo. Quaderni della
Società Ligure di Storia Patria 3, 29–49.
BERINDEI, M. – VEINSTEIN, G. (1976): La Tana-Azaq de la présence italienne à
l’emprise ottomane (fin XIIIe-milieu XVIe siècle). Turcica 8, 110–201.
BRATIANU, G.I. (1969): La Mer Noire. Des origines à la conquête ottomane, Munich.
CARILE, A. (1965-6): La Partitio Terrarum Imperii Romanie del 1204 nella tradizione
storica dei Veneziani. Rivista di Studi Bizantini e Neoellenici II-III, 167–179.
CARILE, A. (1965): Partitio terrarum Imperii Romanie. Studi Veneziani 7, 125–305.
CESSI, R. (1937): Documenti finanziari della Repubblica di Venezia. «Accademia dei
Lincei», serie IV, vol. 1, Padua.
CHINAZZO, D. (1958): Cronica de la guerra da Veniciani a Zenovesi, ed. by V. Lazzerini,
Venice.
CIPOLLA, C. M. (1980): Before the industrial revolution: European Society and Economy,
1000–1700, New York.
DE NICOLA, B. (2017): Women in Mongol Iran: The Khātūns, 1206–1335, Edinburgh.
DI COSMO, N. (2009): Black Sea emporia and the Mongol empire: A reassessment of
the Pax Mongolica. Journal of the Economic and Social History of the Orient 53/1–
2, 83–108.
DI COSMO, N. – PUBBLICI, L. (2022): Venezia e i Mongoli. Commercio e diplomazia sulle
vie della seta nel medioevo (secoli XIII-XV), Rome.
FAVERAU, M. (2016): Sources vénitiennes pour l’histoire de la Horde d’Or : nou-
velles perspectives de recherche. Zolotordynskoe obozrenie 1, 39–54.
FUMAGALLI, V. (2003): L’uomo e l’ambiente nel Medioevo, Rome–Bari.
GIES, J. – GIES, F. (1972): Merchants and Moneymen. The Commercial Revolution, 1000–
1500, London.
HEYD, W. (1913): Storia del commercio del Levante nel Medio Evo, Torino.
KARPOV, S. P. (1978): Venetsiansko-trapezundskii konflikt 1374–1376 gg. i
neizvestnyi mirnyi dogovor 1376 g. Vizantjnskii Vremennik 39, 102–109.
KARPOV, S. P. (1986): L’impero di Trebisonda, Venezia, Genova e Roma: 1204–1461, Roma.
KARPOV, S. P. (1994): Krizis Tany 1343g. v svete novykh istochnikov. Vizantiiskii
Vremennik 55, 121-126.
KARPOV, S. P. (1995): On the origin of medieval Tana. Bizantino-slavica LVI, 227–235.
KARPOV, S. P. (2000): La navigazione veneziana nel Mar Nero, secoli XIII-XV, Ravenna.
KARPOV, S. P., (2018): Nachalo smuty v Zolotoi Orde i perevorot Navruza. Zolotor-
dynskoe obozrenie 6/3, 528–536.
KEDAR, B. (1976): Merchants in crisis. Genoese and Venetian men of affairs and the four-
teenth-century depression, London.
KURODA, A. (2009): The Eurasian silver century, 1276–1359: commensurability and
multiplicity. Journal of Global History 4/2, 245–269.
LOPEZ, R. S. (1955): Venezia e le grandi linee dell’espansione commerciale nel seco-
lo XIII, in La civiltà veneziana del secolo di Marco Polo, Firenze.
LOPEZ, R. S. (1976): The commercial revolution of the middle ages, 950–1350, Cam-
bridge.
LOPEZ, R. S. (1986): Nelle terre dell’Orda d’Oro: tre documenti genovesi inediti, in
Collucci, M. et al. (ed.:) Studia Slavica Mediaevalia et Humanistica Riccardo Picchio
dicata, vol. 2, Roma, 463–474.
MOLÀ, L. (1994): La comunità dei lucchesi a Venezia: immigrazione e industria della seta
nel tardo Medioevo, Venezia.
MUELLER, C. R. (1992): Mercanti e imprenditori fiorentini a Venezia nel tardo me-
dioevo. Società e Storia LV, 29–60.
MUELLER, C. R. – LANE, F. (1985): Money and banking in medieval and Renaissance
Venice, Baltimore.
PAPACOSTEA, S. (1979): Quod non iretur ad Tanam. Un aspect fondamental de la
politique Génoise dans la Mer Noir au XIVe siècle. Revue des Études Sud-Est Eu-
ropéenne 17/2, 201–217.
PEGOLOTTI (1936): La pratica della mercatura di Francesco Balducci Pegolotti, ed. A.
Evans, Cambridge (MS).
POUNDS, N. J. G. (1978): An Economic History of Medieval Europe, London.

PUBBLICI, L. (2005): Venezia ei l mar d’Azov. Alcune considerazioni sulla Tana nel
XIV secolo. Archivio Storico Italiano 163/3, 435–483.
PUBBLICI, L. (2017): Antagonism and Coexistence. Local Population and Western
Merchants on Venetian Azov See in the 14th century, in Pieralli, C. –Delaunay,
C. – Priadko, E. (ed.:) Russia, Oriente slavo e Occidente europeo. Fratture e integ-
razioni nella storia e nella civiltà letteraria, Biblioteca di Studi slavistici 36, Floren-
ce, 25–47.
PUCCI DONATI, F. (2019): Ai confini dell’Occidente. Regesti degli atti dei notai veneziani a
Tana nel Trecento, 1359–1388, Udine.
RAINES, D. (2003): Cooptazione, aggregazione e presenza al Maggior Consiglio: le
casate del patriziato veneziano, 1297-1797. Storia di Venezia, vol. 1, 1–64.
SKRŽINSKAJA, E. Č. (1968): Storia della Tana. Studi Veneziani 10, 3–47.
SPUFFORD, P. (2006): Power and Profit. The Merchant in Medieval Europe, London.
STÖCKLY, D. (1995): Le système de l’Incanto des galées du marché à Venise (fin 13 e–milieu
15e siècle), Leiden.
TAFEL, G. L. F. – THOMAS, G. M. (1964): Urkunden zur älteren Handels- und Staats-
geschichte der Republik Venedig: mit besonderer Beziehung auf Byzanz und die Le-
vante: vom neunten bis zum Ausgang des fünfzehnten Jahrhunderts, Amsterdam.
THIRIET, F. (1953): Venise et l’occupation de Ténédos au XIV e siècle. Mélanges
d’archéologie et d’histoire 65, 219–245.
THIRIET, F. (1958): Régestes des délibérations du Sénat de Venise concernants la Romanie,
vol. I: 1329–1399, Paris.
THIRIET, F. (1959): La Romanie vénitienne au Moyen Age, le développement et l’exploi-
tation du domaine colonial vénitien, XIIe–XVe siècles, Paris.
THOMAS, G. M. (1880): Diplomatarium Veneto-Levantinum sive Acta et Diplomata res
Venetas, Graecas atque Levantis illustrantia, vol. 1: 1300–1350, Venice.
THOMAS, G. M. – PREDELLI, R. (1899): Diplomatarium Veneto-Levantinum sive Acta et
Diplomata res Venetas, Graecas atque Levantis illustrantia, vol. 2: 1351–1454, Venice.
Some Preliminary Notes on the
Aftermath of the Eurasian Steppe’s
“Mongol Moment”
ISHAYAHU LANDA
Abstract
The paper singles out the decades following the Great Chinggisid Crisis, i.e. the
period between the later 14th century and the early 15th century, as a “transition
phase” between the Mongol and post-Mongol eras of Eurasian history. Looking at
the political, religious, and ethnogenetic characteristics, the paper on the one hand
attempts to stress the transitory nature of this specific period of time and, on the
other, to assess the role of the Mongol Eurasian age in the history of the Eurasian
Steppe belt.
Keywords: Mongol Moment, Great Chinggisid Crisis, post-Mongol Eurasia, Steppe
ethnogenesis, 14th–15th centuries, Eurasian Steppe belt
Starting with Temüjin’s unification activities in the Mongolian steppes in the early
13th century, Eurasia entered a new historical phase. Around the mid-14th century
this phase came to an end. The last decades of Mongol Eurasia, often called the
Great Crisis, are usually defined as ranging from the 1330s to ca. 1370s, depending
on the geographical location. The turmoil started in the Ilkhanate in November
1335, triggered by the death of Ilkhan Abū Saʿīd (r. 1316–35), the last Abaqaid rul-
er, who had no male descendant. The Ulus Chaghatai was split into two parts in
1347, and this marked the beginning of the political division of the Central Eura-
sian Steppe for more than three centuries, until it was eventually split up between
the sedentary superpowers starting in the late 16th century. In 1259, the death of
Berdibek (r. 1257–9), the last confirmable Batuid khan, destabilized the Jochid ulus
in the north. In the early 15th century, this process, known in the Russian letopisi as
 University of Bonn, ilanda@uni-bonn.de, ORCID 0000-0001-5342-9175. This research was

made possible by the German Research Foundation (DFG, project no. 429873935). The article
is a revised version of a paper delivered at the 8th International Conference on Medieval
History of the Eurasian Steppe Nomads and their Neighbours in the Middle Ages, held in Sofia,
November 20–3, 2019.
270 ISHAYAHU LANDA
“Velikaya zamyatnya” (Great Turmoil), entered a qualitatively new stage follow-

ing the death of the last Jochid unifier, Tokhtamysh (r. 1378–1406). The Qa’an ulus,
ruled by the Qubilaid Yuan dynasty (元朝, 1271–1368/70) in China, Mongolia, Ti-
bet, and Korea, collapsed in 1368. The death of the last Yuan Emperor Toghon Te-
mür (r. 1333–70) from dysentery in 1370 and the rise of Temür (Tamerlane, r. 1370–
1405) in Central Eurasia finalized the Crisis. 1 Mongol Eurasia was no more, even
though the Chinggisid families continued to play a role in the Steppe and, partly,
in the adjoining regions in the following centuries. 2
As part of the general reassessment of the Chinggisid influence and relevance
for Eurasian history, this paper singles out the decades that followed the Great
Crisis up to the early 15th century as a separate historical phase, structurally differ-
ent both from the Mongol Eurasian period and from the early modern period of
history from the 15th (or, according to other perspectives, from the 16 th) century
onward.3 The paper aims to provide some very initial remarks on approaching the
highly complex “Mongol Moment’s” impact on the Eurasian steppe belt’s history
through a perspective that focuses on the immediate aftermath of the Great Crisis. 4
Concentrating on the territorial stretch from today's Mongolia to the northern
steppes of the Black Sea, it aims to single out the major features characterizing the
aftermath, the “transition phase”, while underlining Chinggisid Eurasia’s rele-
vance for the historical developments that followed its disintegration. Due to the
limited scope of the publication, the discussion will highlight developments in
three domains: geopolitical, religious, and ethnic. All of the issues discussed below
were rooted in the pre-Crisis history of the Chinggisid domination in Eurasia and
some of them became visible or prevalent in the Steppe and the adjoining areas
also before the Crisis. For our discussion, however, it is important that it was main-
ly during the “transition phase” that those issues and developments could be ob-
served better, with the Chinggisid rule either absent or struggling for survival.
One of the basic hypotheses underlying the following arguments is that during
most of its existence Chinggisid Eurasia could be characterized as a sort of the
“Mongol Commonwealth” (another term would be “Pax Mongolica”), in which the
blood relations of the ruling families, trade, and the circulation of knowledge and
religious personnel provided an intricate network uniting (politically and ideologi-
cally) the various domains in one geopolitical whole. 5 Despite the differences be-
1 For brief depictions of the Crisis see MORGAN 2009; ALLSEN 2015a.
2 E.g. MCCHESNEY 2009; VEIT 2009, 165–8.
3 See, e.g., SUBRAHMANYAM 1997, 736–7; for a broader discussion see GOLDSTONE 1998.
4 For the term “Mongol Moment”, see SUBRAHMANYAM 1997, 737, as far as I know, the first
one to discuss it. Subrahmanyam uses it in order to stress the singularity of the period of
Mongol conquests and rule in Eurasian history. While he does not explicitly include the
Crisis decades in this term, it seems reasonable to do so.
5 For a broader discussion, see KIM 2009; note also KHAZANOV 1993, 467 and the discussion
in KOTKIN 2007, 503–8.

SOME PRELIMINARY NOTES ON THE AFTERMATH… 271
tween the areas under Chinggisid control, the ruling branches shared, at least nom-
inally, the same Chinggisid political ideology. Similar political institutions were
present in the various Chinggisid polities. The co-existence of different economic
models under one imperial rule, the tension between hierarchical and collective
rulership as well as the importance of the nomadic military characterized the
Chinggisid space. One major common feature of the post-Crisis Steppe was the
preservation of the Chinggisid priority to rule: the “Chinggisid principle”, to use
Miyawaki’s term.6 This was not the case everywhere, and outside of the Steppe this
principle was mainly replaced by the (at least seemingly) non-Chinggisid legitima-
cies.7 But the Steppe areas have also witnessed a number of variations and devia-
tions from the principle’s “pure form” (in which a ruler of Chinggisid origin pos-
sessed real power).8 Especially the decades after the Great Crisis were dominated
by such variations of rulership and control across the Steppe that co-opted ele-
ments rooted in both Chinggisid and non-Chinggisid ideological and political real-
ities. These developing forms included, e.g., the model in which the Chinggisid
puppet khans were installed by a certain military commander or a group of com-
manders. The Chaghadaid ulus can serve as a specific example. The case of Temür
and his Chinggisid puppet khans, the Ögedeid Soyūrghatmīsh b. Dānishmanjī and
his son Sulṭān Maḥmūd, is well-known.9 Somewhat similar developments can be
observed in the Eastern Chaghadaid domains, the so-called “Moghulistan”,10
where a group of powerful Dughlat commanders enthroned a number of Chinggis-
6 E.g. MIYAWAKI 1992; note also AMITAI – BIRAN 2015, 6.

7 Such were, e.g., the cases of the Ming in China, Ottoman Anatolia, the Muzaffarids and
Injuids in southern Iran, and the Sarbadarids in Khurasan. Note, however, that in some
cases, e.g. under the early Ming dynasty, the anti-Chinggisid rhetoric coexisted with the
new powerholders’ usage of the Chinggisid legitimacy for their own rule (e.g. ROBINSON
2020, 129–57).
8 Surely, the rule of the Chinggisid khan was dependant on the support of the powerful
military commanders even before the Crisis, during Chinggisid Eurasia’s heyday. In some
cases, these commanders, many of whom were Chinggisid sons-in-law, served as kingmak-
ers, somewhat similarly to the situations during and after the Crisis. At the same time, while
the usage of puppet khans became a political norm during and especially after the Crisis, it
was more of an exception before that.
9 On Temür, see MANZ 1988, 110 (on the name of Soyūrghatmīsh’s father, see MUʿIZZ AL-
ANSĀB, fl. 44b–45a; note further MANZ 1998, 23; JACKSON 1990, 400–1; SELA 2004, 28); on the
further Timurid matrimonial connections with this Chinggisid lineage, see, e.g., WOODS
1993, 28, 33. Generally speaking, in this regard Temür followed the example of his predeces-
sor in Western Central Asia, the Qara’unas commander Qazaghān, who became the factual
ruler of the ulus in 1347 and enthroned two Chinggisid puppet khans, Soyūrghatmīsh’s
father Dānishmanjī and Bāyān Qulī, Du’a’s grandson (MUʿIZZ AL-ANSĀB, fl. 44b–45a; AL-
YAZDĪ, 1957: 22; JACKSON 2018, 98).
10 On the name note KIM 1999, 290, fn. 1.
272 ISHAYAHU LANDA
id khans throughout the decades following the Chaghadaid split of 1347. 11 Looking
further eastward, to the Mongolian steppes, one should also not ignore the long-
term relations between the Oyirads and the first khans of the Northern Yuan, who,
in contrast with the Yuan Qubilaids, were connected to the descendants of
Qubilai’s younger brother and contender for power, Arigh Böke. 12 Often, the
commanders behind the Chinggisid puppets claimed a relation to the Chinggisid
family as well. In many cases this was based on a (real or invented) marriage with
a Chinggisid female, through which one rose to the prestigious position of a gü-
regen, a Chinggisid son-in-law.13 This type of power acquisition was especially
typical in the areas centred in the Steppe or adjoining it, the best-known example
again being Temür of the Barlas tribe.14 It is often overlooked that he was not
alone. Similar cases could be observed in Crimea and the western Steppe in gen-
eral, where various power actors were either married to Chinggisid women or
faked connections with such, as well as in Moghulistan, the Eastern Chaghadaid
realm, where a certain Namun, a Chinggisid in-law of unclear tribal origin, made
an attempt to enthrone the Chaghadaid Gunashiri in cooperation with the Ming in
the end of the 14th century.15
The appearance of the mixed form of ideological post-Chinggisid (or, perhaps,
more correctly, Chinggisid-related) authority claims is closely connected with an-
other issue, namely a change in the composition of the ruling elites, especially the
military. The Crisis shattered the ruling establishment of the Chinggisid khanates
across Eurasia. Deliberate purges (as in the Jochid and the Chaghadaid cases),16
migrations, and refugee waves (as in the case of the Yuan collapse), 17 various big-
ger and smaller warfare conflicts, in which most of the Chinggisid military was
involved, as well as epidemics and the early beginning of the so-called “Little Ice
Age”,18 which coincided with the Crisis processes: all these factors arguably led to
11 PISHCHULINA 1977, 41–60 (but note the Soviet Marxist perspective of her discussion); BIRAN
2009, 59–60.
12 On the Northern Yuan, see, e.g., HONDA 1958, esp. 247–8.
13 On the Chinggisid güregens, see LANDA 2016a, 2018b.
14 MANZ 1988, 110.
15 Thus, e.g., Mamai, one of the most powerful Jochid military commanders of the Turmoil
period, appears to have married Berdibek’s sister (IBN KHALDŪN 1865, vol. 5, 538); on Na-
mun, see KIM 1999, esp. 291–9.
16 On the purges of the early 1360s in the Jochid domains, see, e.g., NAṬANZĪ 1957, 85–6; on
Qazan Sulṭān (r. 1343–7), the last “bad” Chaghadaid ruler of the united ulus (a perspective
on which we do not possess any alternative opinion of the contemporaries), see AL-YAZDĪ
1957, 21; BIRAN 2009, 59.
17 On the ex-Yuan domains, see e.g. ROBINSON 2012, 113ff.
18 On the early phase of the so-called “Little Ice Age”, the cool phase that started around the
mid-14th century, and its interplay with the “Black Death” epidemic, see, e.g., BROOKE 2014,
380–90.
the disappearance of a significant part of the elite layers and to the restructuring of
power hierarchies in a short period of time. It is in this context that we observe the
appearance of new power actors in all post-Mongol Eurasian domains. While, as
noted above, the “Chinggisid principle” generally kept its importance across the
Steppe, especially in the Mongolian areas, a number of attempts to distance oneself
from it can be observed since very early. Such attempts can be seen in the Crimea-
Azov area, where Mamai, the omnipotent Jochid beglerbegi19 of the Qiyat, arguably
made an attempt to proclaim himself khan very early, around 1361; 20 in the Volga
Bulgaria and the Moksha-Narovchat areas, where a number of (predominantly)
copper coins indicate attempts by certain personalities to establish themselves as
local rulers with and without references to the Batuid legacy; 21 in Hajji-Tarkhan,
today’s Astrakhan, during the 1360s, where a non-Chinggisid legitimacy was
claimed by a certain Hajji-Cherkes and, later on, by Salchey. 22 While those attempts
were less visible in the beginning of the “transition phase”, during which the
Chinggisid “principle” and its power claim remained broadly accepted, the Ching-
gisid puppet khans became quite a burden for a number of charismatic personali-
ties towards the early 15th century. While Temür himself kept his Chinggisid rul-
ers, his descendants started regarding them as unnecessary shortly after the found-
er’s death in 1405.23 The same happened in the Manghit realm in the Lower Volga
and Yaik basin, where Edigü’s descendants developed their own ruling legitimacy,
connecting themselves in their power rhetoric both to Baba Tükles, the famous Sufi
converter of the Steppe, as well as to the Righteous Caliph Abū Bakr. 24 Neverthe-
less, even if many of the new elites and power groups did (partially or fully) break
with the Chinggisid legacy and legitimacy, their very appearance and power ac-
quisition was in many cases still connected with the pre-Crisis decades and their
ancestors’ position at the Chinggisid courts. The Jalayirids, one of the important
Ilkhanid in-law families, who ruled in Tabriz during the second half of the 14 th
19 On beglerbegi, i.e. the amīr-i ulūs, one of the leading positions in the Jochid administration,
see TREPAVLOV 2014.
20 Note PACHKALOV 2012; for a contesting opinion, see, however, PETROV 2013, 184–5.
21 On Taghay Bek, the Mokhsha ruler, see, e.g., PAVLIKHIN 2014, 36–7; on the Volga Bulgarian
ruler Bulad-Temür, see, e.g, SAFARGALIEV 1960, 118–9 (note also POCHEKAEV 2019, 353, fn. 494).
22 IBN KHALDŪN 1865, vol. 5, 538–9; PSRL, vol. 11 [Nikonovskaya letopis’], 24. On the discus-
sion of these two personalities and their possible identification, see ZAYTSEV 2004, 17–21.
23 MANZ 1998, 35. The final break with the practice of appointing a puppet khan did not
occur until Ulugh Beg’s rule (note SELA 2004, 28, fn. 41).
24 DEWEESE 1994, 336, 386–7; TREPAVLOV 2020, 109–10; also, LANDA 2020, 99.
274 ISHAYAHU LANDA
century25, and the Sufi-Qonggirad ruling family in Khwarazm 26 are two exam-
ples.27
As a rule, the geopolitical changes of the “transition phase” showed a stronger
tendency towards dissolution of power. This was not a linear development, as the
very idea of the unification of the domains under the (ex-)Chinggisid rule re-
mained during the second half of the 14th and early 15th centuries. It was only after
the death of two major unifiers, Temür and Tokhtamysh, that the ex-Chinggisid
areas were further decentralized both in Central Asia and in the western Steppe. At
the same time, moving to our second point of interest and observing the religious
dimension of the “transition phase”, we cannot but notice that the collapse of
Mongol Eurasia impacted Central Eurasia’s religious landscape quite differently.
In general, the religious diversity in the Steppe and the adjoining areas observable
in the previous centuries (also during the Chinggisid phase) was greatly reduced
during the 14th century. There were winners, of course. The Chinggisid conquest of
Eurasia led to a remarkable spread of Islam. Supported both by state and non-state
(mainly Sufi) authority and conversion channels, various forms of Islam spread
from west to east during the Mongol Eurasia period, both via land and maritime
routes, replacing or pushing aside other religious forms and movements. 28 Starting
with the end of the 13th century, multiple power actors across Western and Central
Asia turned to the Islamic legitimacy; this development becomes even more visible
with the Great Crisis and during the “transition phase”, when a broad number of
non-Chinggisids turned to Islam in order to reinforce (or justify) their rule by and
through Islamic rhetoric. But this was not only a question of power acquisition.
The “Mongol Moment” helped Islam spread across various societal layers
throughout the Eurasian Steppe belt, becoming one of the most important identity
markers towards the beginning of the 15th century.29
Islam was not the only winner. The Orthodox Church in the Rus’ principalities
(knyazhestva) was another one. Surely, its impact and dispersion area were much
smaller than that of Islam, and its primary domains lay way to the north of the
25 E.g., WING 2016, 101–146.

26 LANDA 2018a.
27 But there were more, even though the other cases have been studied less. Thus, the Oyirad
support of several Northern Yuan rulers during the second half of the 14 th century was pos-
sibly connected to the high position the Oyirads of the Altai Mountains held in the Ögedeid
ulus of Qaidu about a hundred years before (see below; on the relations between the West-
ern Mongolian Oyirads and Arigh Böke, see OKADA 1987, 185; on the relations between
presumably the same Oyirad groups and Qaidu, see BAI 2006: 8). A similar situation can also
be observed in Hajji-Tarkhan. One of the apparent rulers of the mid-1370s, Salchey of Hajji-
Tarkhan, mentioned above, seems to have originated from the in-laws of the Jochid Özbek
(MUSTAKIMOV 2009, 123–5 and see above, p. 271, fn. 22).
28 For a general introduction to this process, see DEWEESE 2009.
29 For a detailed discussion, see LANDA 2022.
Steppe, but it is important to keep in mind that the Church significantly profited
from the Chinggisid rule. The Chinggisids famously followed a rather practical
attitude towards the multitude of religious groups in their domains. 30 Only in the
aftermath of a conversion fever did they sometimes persecute other religions or
limit them in their rights.31 The exemption of the religious personnel from taxation
also strengthened their position; the Orthodox Church was no exception. 32 During
the first half of the 14th century the prerequisites for the growing power of the
Church and its tightening relations with Muscovy were developed further. Thus,
the metropolitan seat was transferred from Vladimir to Moscow in 1325, only 25
years after it had been moved to that location from Kiev in 1300. 33 Generally, start-
ing with the 1310s, the continuous (though not without setbacks) rise of the Mus-
covy princes (knyazes) can be observed together with their status vis-à-vis the Jo-
chid authorities and the other knyazhestva.34 Notably, this process started with the
Great Knyaz Yuri Danilovich (r. 1303–25), who became a güregen, an imperial in-
law, through his marriage to Konchaka, Özbek's (r. 1313–41) sister, in the 1310s.35
Even though no further marriages were concluded after Yuri's death in 1325, it is
highly plausible that in the Chinggisid eyes the high status of the family was trans-
ferred to the following generations.36 It is also in this context that we should under-
stand the rise of Muscovy during the 14 th century. The Great Crisis and, especially,
the “transition phase”, did not immediately entail the breaking away from the
Jochids of the Rus’ principalities in general or that of Muscovy in particular, nei-
ther militarily nor politically. Indeed, the Great Knyazes and Metropolitans were
still obliged to get the Horde’s endorsement for their appointment. This period
witnessed a gradual estrangement of Moscow from the Jochid domains. Dmitry
30 JACKSON 2005.
31 The first years after Ghazan’s (r. 1295–1304) conversion (1295) are a good example: the
Ilkhanids seem to have suppressed non-Islamic faiths up to some degree. These events are
often attributed to the influence of the powerful commander and son-in-law Nawrūz, who
possessed an exceptionally impressive influence on Ghazan up until the former’s execution
in 1297 (e.g., DASHDONDOG 2011, 197).
32 For a general discussion of the Chinggisid impact on the position of the Church in the
Rus’ domains, see GALIMOV 2016; on the ambiguous relations between the two, see also
HALPERIN 1987, 113–6.
33 BORISOV 1999, 209; BORISOV 2017.
34 See, e.g., HALPERIN 1987, 53–6.
35 Such is, at least, the claim of the chronicles (e.g., PSRL, 10 [Nikonovskaya letopis’], 180).
36 As it happened, e.g., with the Jalayirid family of the Ilkhanid commander Aq Buqa, a
supporter of the Ilkhans Aḥmad Tegüder (r. 1282–4) and Geikhatu (r. 1291–5). Aq Buqa was
himself a güregen, as he married Öljetei, Ilkhan Arghun’s (r. 1284–91) eldest daughter, who
was later given to Aq Buqa’s son Ḥusayn (the one whose son Ḥasan-i Buzurg became the
founder of the Jalayirid dynasty). On Aq Buqa’s marriage, see RASHĪD AL-DĪN 1373/1994–5,
vol. 2, 1153; on that of his son, see RASHĪD AL-DĪN 1373/1994–5, vol. 2, 1153; further see WING
2016, 63–73.
276 ISHAYAHU LANDA
Donskoy’s (r. 1359–89) attempt to mint his own coins, known as den'ga mos-
kovskaya, during the mid-second half of the 14th century is one clear sign of this.37
Importantly, the coins include a reference to the Batuid legacy, arguably citing
Özbek's name and including the issuer in the broader hemisphere of Chinggisid
legitimacy. They did not connect the Knyaz to the contemporary developments in
the Steppe, however, and should be seen as a clear sign for Muscovy's carefully
increased autonomy.38 The Battle of Kulikovo Field (September 8, 1380) is well
known, and even though it certainly did not lead to a major systemic change in the
Rus’-Horde relations, it also clearly indicates that changes were imminent. The
personality of the (ethnically Bulgarian) Metropolitan Cyprian (ca. 1330-1405), one
of the most important and powerful Orthodox Metropolitans of the 14 th and 15th
centuries, a major opponent of the Jochids and a protégé of St. Sergy Radonezhsky
(1314–92), is beyond our discussion, but the very appearance of such a figure in the
Great Moscow Knyazhestvo during the “transition phase” (even though Cyprian
entered Moscow as late as May 23, 1381)39 indicates the increasing role of the
Church in supporting the Muscovy positions vis-à-vis the Horde and in general.
While the Chinggisid legacy remained crucial in the Muscovy ideology for the next
centuries, so did the Byzantine political tradition and tzardom ideology as well,
provided and strengthened by the Church; it was the mixture of both ideological
trends that formed the forthcoming Russian imperial ambitions and self-
representation.40
There were also losers. One was Buddhism. As we know, it was deeply en-
trenched in the Chinggisid political landscape, both in the Steppe and beyond,
throughout the 13th century. The primary sources tell us about the richness of the
37 The issue remains complicated. Starting with the 13th century and until the mid-14th centu-
ry at the earliest the Rus’ knyazhestva went through a period known as “bezmonetnyi” (i.e.
“without the usage of coins”, see SPASSKY 1962, 53–4). While many authors suggest that
Dmitry began to mint his own money around the Battle of Kulikovo Field (i.e. around 1380),
Gaydukov stresses that the Great Knyaz was not the first one to issue his own coins. Various
copies and imitations of the Jochid coinage appeared in the Rus’ territories since the early
1360s, i.e. since the beginning of the Turmoil in the Jochid ulus. Later on, the imitation coins
appear with the tamghas of the Rus’ knyazes, i.e. various identifying signs similar to the
Chinggisid tamghas (on this, see PETROV 2005), with such a usage clearly being an imitation
of the Jochid monetary policies. Further on, the Rus’ knyazes, the most important of them
being Dmitry Donskoy, started issuing their own coins; parts of their legend were clearly an
imitation of the Chinggisid coins, notably not of the last Batuids or the contemporary khans,
but those of Özbek, Yuri Danilovich’s long deceased brother-in-law (for the further discus-
sion, see GAYDUKOV 2007).
38 On the first Muscovy coins of the second half of the 14 th century, see, e.g., GAYDUKOV 2009.
39 OBOLENSKY 1978, 91.
40 On Cyprian, the “Metropolitan of Kiev and All Rus’” since 1375, see OBOLENSKY 1978;
DMITRIEV 1963, esp. 216–28. On the Muscovy ideology and its post-14th-century Chinggisid-
Byzantine mixture, see, e.g., CHERNIAVSKY 1959; HALPERIN 1976.
Buddhist monasteries in Azerbaijan and Northern Iran under the first Hülegüids. 41
As stressed by Elverskog, the figurative depictions of the Prophet Muhammad, the
first ones in the history of Islam, occurred not least under the influence of the
Ilkhanid Buddhist tradition.42 On the other side of the continent, Buddhism played
an important role in connecting the Qubilaids of the Yuan and the populations of
China. The new Tibetan teaching, until then alien to the Chinese Buddhism move-
ment, was brought to China by the Chinggisids and rose in power, both extending
the support networks for the dynasty and creating frictions inside the local Bud-
dhist establishment.43 The 14th century changed the rules of the game, even though
not on the same scale and in the same way everywhere. Buddhism did not disap-
pear from most of Central Eurasia immediately, but the religious landscape gradu-
ally changed. We do not hear anything about Buddhism in the late Ilkhanate or, in
any form, during the “transition phase” (but at least we know that most of the
Ilkhanid bakhchis, the Buddhist holy men, were either expelled or converted to
Islam in the early 14th century following the Ilkhanid conversion in 1295).44
Moving northward and eastward, to the areas of our primary interest, we wit-
ness a much more complex and much less clear picture. On the one hand, despite
the conversion of both the Jochid and Chaghadaid uluses to Islam, and despite the
persecutions against Buddhism, the latter remained in some form in the northern
Steppe and in Central Asia.45 It was especially in Moghulistan that the Buddhist
presence can be traced also after the Crisis and until the early 15 th century. A num-
ber of recently published late Chagatai (Moghul) decrees reveal not only the
preservation of Buddhism in Eastern Turkestan, 46 but also mention intensive Bud-
dhist pilgrimage and trade routes between the Chinese Dunhuang ( 敦煌, in today's
Gansu province) and the Moghul centres of the late 14 th century.47 We also possess
valuable descriptions of a Buddhist temple in Turfan in the early 1420s, preserved
in the diary of a Timurid diplomatic mission to the Ming court, and are aware that
it was only in the 1430s that the city of Turfan was completely purged of Bud-
41 Note PRAZNIAK 2014, esp. 661. For an example regarding the alleged Ilkhanid Buddhist
temple to the east of the Ilkhanid city of Sulṭāniyya, see HATEF NAIEMI 2020, 23–5.
42 ELVERSKOG 2010, 162–74. See KADOI 2009 for a detailed discussion of the Ilkhanid aware-
ness of the Buddhist art tradition and its influences on the Iranian cultural heritage.
43 For the general discussion on the relations between the Yuan dynasty and the Tibetan
Buddhist establishment, see PETECH 1983; PETECH 1990, esp. 5–84; for the developments in
the Yuan South and the Yuan Buddhist administration, see WANG 2015.
44 RASHĪD AL-DĪN 1373/1994–95, vol. 2, 1259, 1356; AMITAI 1999, 41.
45 On the Jochid conversions to Islam, see, e.g., DEWEESE 1994, 90–142; on those of the Cha-
ghadaids, e.g., BIRAN 2002.

46 On this terminology, see LANDA 2020b, 507, fn. 5.
47 E.g. DAI 2008.
278 ISHAYAHU LANDA
dhists.48 The “transition phase” witnessed Buddhism’s continuous retreat in most

of Central Eurasia. It was only in the Mongolian steppes where its Tibetan varia-
tion was revived during the 16th century, already after the “transition phase” came
to its own end.49 During the second half of the 14th century, however, one could
clearly see the process that started, depending on the location, probably a couple of
decades before that—Buddhism was slowly losing its attractiveness for the broader
populations and the powerholders when the old Chinggisid power hierarchies that
had kept it alive, or at least tolerated it, disappeared.
Buddhism was not alone. The Church of the East, also (wrongly) known as the
Nestorian Church, went through its last significant revival during the Chinggisid
rule, especially in the early Ilkhanate and under the Yuan, but suffered large-scale
setbacks and disappeared almost everywhere, except for a number of locations in
Western Asia (primarily Mosul and the adjoining areas), after the Great Crisis at
the latest.50 In contrast with Buddhism, it never succeeded in finding a new patron.
In the Ilkhanate, the Church suffered severely starting from the late 1320s. Saint
John the Baptist Monastery near Maragha was desecrated and converted into a
mosque in the last years of the Ilkhanate. 51 Of primary importance for us are the
following decades, coinciding with the pontificate of the Syrian Patriarch Denha II
(1336–81).52 While the Church was tolerated by the Jalayirid rulers of parts of Iran
and Iraq in Denha's time, the East Syrians almost completely disappeared from
Central Asia and, it seems, North China.53 Up to some degree the devastation of
the Syriac communities started with the plague (a “Black Death” outbreak), that
wiped out multiple areas in the Semirechye in the late 1330s. 54 The last known Syr-
iac tombstones, found near today’s Chinese-Kirgiz border, are dated to the early
1370s.55 The Castilian ambassador Clavijo, who visited Temür's Samarkand in the
early 15th century, claims to have seen East Syrians among the citizens. Judging
from the available data, however, those were most probably merchants from West-
48 The embassy arrived at Turfan at the end of Jumada al-ʾĀkhir 823 AH, i.e. the end of July
1420 (ḤĀFĪZ-I ABRŪ 1993/1372–73, 821); note MUSTAFAEV 2015, 105 on the embassy, as well as
ELVERSKOG 2010, 203 and KIM 1996 for the broader discussion. The Buddhist-Islamic contacts
became of relevance again much later, since the 16th century at the earliest (on this see, e.g.,
SELA 2014).
49 VEIT 2009, 170–73, further see KOLLMAR-PAULENZ 2003.
50 For the post-Chinggisid history of the Eastern Syriac Christianity in West Asia (until 1552,
the year of the Church’s major split over the question of the patriarchate), see WILMSHURST
2000, 346–48.
51 BORBONE 2017, 118.
52 WILMSHURST 2000, 18–19.
53 WILMSHURST 2000, 19.
54 HALBERTSMA 2015, 67; SLAVIN 2019.
55 Note KOKOVTSOV 1906, 196; also, SLAVIN 2019, 86, as well as esp. SPYROU et al. 2022; SLAVIN
2022.
ern Asia, and not the remnants of the native Syriac communities.56 We see a similar
picture towards the east. Despite the flourishing of the Syriac populations (still as a
minority faith) under the Yuan, especially in today’s Inner Mongolian areas, their
further destiny is very unclear. Some scholars claim that the East Syrians were
forced to leave the Ming domains after the dynastic change. 57 A contrary position
holds that its communities could have merged with the other (more dominant)
religious movements, such as the Daoists and Buddhists, in the decades following
the collapse of the Yuan.58 Be that as it may, as far as I am aware, no Nestorian
graveyard dated after the early 1370s has been found on the territory of today’s
China, and it is plausible (even though not completely confirmed) that the Yuan’s
Syriac settlements in today’s Inner Mongolia were abandoned during the second
half of the 14th century at the latest.59
The “transition phase” from the Mongol to the post-Mongol Eurasian Steppe
was characterized by political fragmentation, a multiplication of legitimacy chan-
nels as well as a reduction of religious diversity. It was, at the same time, not char-
acterized by an immediate division between the “old” and the “new”, but by a
gradual process with multiple setbacks, most of which can be only partially recon-
structed from the sources. The same can also be said about the phase’s third di-
mension of interest for this discussion, namely the gradual change of the Central
Eurasian ethnic landscape that became visible during the Crisis and especially in
the following decades of the “transition phase”. Indeed, these periods of the Cen-
tral Eurasian history witnessed the revival of multiple old nomadic identities
across the Steppe and the adjacent regions. 60 The Jalayirids in Iran, the Oyirads in
Iraq, Iran, and Western Mongolia, the Qonggirad of Khwarazm, the Timurid Bar-
las, Temür’s multiple commanders, originating from the Merkit, the Naiman, the
Sulduz, the Qipchaq, the Moghul Dughlats—a multitude of tribal ethnonyms are
found in the primary sources narrating the “transition phase”.61 In fact, the use of
ethnonyms was not new. When one compares two major genealogical compendia,
56 DE CLAVIJO 2009, 171; note DICKENS 2015, 30.

57 E.g. WILMSHURST 2000, 19; VAN MECHELEN 2001, 97; also, HALBERTSMA 2015, 68–9, who
does not come to any conclusive suggestion. Note HALBERTSMA 2015, 67, who suggests that
the East Syriac Christianity did not survive in Central Asia also due to conflicts with Islam.
This argument is given by the author without any confirmation and, though plausible, re-
mains open for discussion.
58 E.g. WANG 2014 and note HALBERTSMA 2015, 145 (in this regard note also a much older
hypothesis of WILLIAMS [1883, 289] that the Eastern Syrians could have converted to Islam
and Buddhism following the collapse of the Chinggisid Yuan).
59 HALBERTSMA 2015, 145 (citing GAI 1991, 394ff. that I was not able to reach); further see
KLEIN 2000, 56, 283–9 for the detailed discussion.

60 A process termed “retribalization” by Togan (1998, 13), further see LANDA 2017, 1203, esp.
fn. 81.
61 Note, e.g, the description of Temür’s commanders, found in the MUʿIZZ AL-ANSĀB, fl. 97a–99a.
280 ISHAYAHU LANDA
Rashīd al-Dīn’s Shuʿab-i Panjgāna and the anonymous Muʿizz al-ansāb, that have
about one hundred and fifty years in between, one cannot but stress the explicit
attention the compilers paid to mentioning the tribal ethnonyms in their discussion
of both the Chinggisid and Timurid military.62 This is not surprising. The major
steppe vocabulary was directly connected to the differentiation between various
“tribes”, and while “tribal identity” and “tribe” themselves were surely very flexible
and unstable categories, one cannot simply curtail the complexity of the traditional
steppe power lexicon, tribal in its core, only to the existence of the aristocratic line-
ages of the nomadic elites, as David Sneath has done in his famous and controver-
sial research.63 The tribal vocabulary itself is not peculiar. What is especially inter-
esting is the reappearance of the tribal ethnonyms as major identity markers for
new powerholders and military elites in the “transition phase”. On the local levels,
these often replaced the Chinggisid ones. As a rule, notably, those of the tribal
identities that had flourished during the Chinggisid rule survived the Chinggisid
period and were reinforced after its end (e.g. the Qonggirads, the Oyirad, etc.).64
Connecting this point to the discussion of the post-Crisis political elites, we
should stress that Mongol Eurasia became a watershed for the ethnic history of the
Steppe, both in Mongolia and in the Turkic world up to Crimea. The post-
Chinggisid history of the Steppe, including the periods beyond the first transition
decades, witnessed the appearance of multiple new tribal and ethnic identities that
had not existed before the Chinggisid unification of the Steppe or before its end.
The 15th and 16th centuries witnessed the formation and development of several
new identities, such as the Crimean and Volga Tatars, the Yaik Manghits, the Uz-
beks, the Kalmyks, etc. While the roots of many of these developments can be
traced way back, in many cases to the heyday of the Mongol Empire, it is mainly
during the Crisis and the transition decades that many of them became visible. In
this sense, the “transition phase” stands between the Chinggisid and the post-
Chinggisid periods in terms of the tribal and ethnic history of the Eurasian Steppe.
It is in this period that we witness both the preservation of old, but still relevant
tribal and ethnic markers, and the first stages of the formation of new ones. The
disappearance of the Batuid lineage and the dramatic decline in power of the Jo-
chid Right Wing led to a significant rise of the Blue Horde, ex-Orda domain, the
future key area of the Uzbek ancestors. 65 According to a Mamluk source, the Blue
Horde also became known as the bilād uzbek already during the second half of the
62 Cf. e.g. the lists of Chinggis Khan’s commanders in both sources (RASHĪD AL-DĪN (n.a.), fl.
105b–106a and MUʿIZZ AL-ANSĀB, fl. 13b–17a).
63 SNEATH 2007. Sneath is supported e.g. by ATWOOD 2015, esp. 16–17 and MUNKH-ERDENE
2011, 212; for critical voices, see, e.g., ABASHIN et al. 2009, KHAZANOV 2010, DURAND-GUÉDY
2011, KRADIN 2012.
64 E.g. LANDA 2016b, 168–176; LANDA 2018a.
65 For the overtaking of the Jochid Left Wing by the Toqa-Temürid descendants during the
1360s, see e.g. VÁSÁRY 2009.

14th century.66 Similarly, Niẓām al-Dīn Shāmī called Urus Khan “Urūs Khān-i
Ūzbīkī”.67 While it is quite clear that the Uzbeks as a fixed tribal identity or people
did not exist at that time, the appearance of this political marker, in the future an
ethnic one, is characteristic for the “transition phase” from the old world order to
the new one.68 Another interesting case, that of the Yaik Manghit, shows us a dif-
ferent development, namely how a new ethnic formation was constructed around
the old tribal one. It cannot be completely repudiated that the Yaik Manghit identi-
ty was constructed not only around an old respectable ethnonym, but also based
on the nucleus of the “original” Manghit of Mongol origin, who underwent Turki-
fication and settled between the Yaik and Emba towards the end of the 14 th centu-
ry.69 Again, this process exemplifies the formative nature of the “transition phase”,
during which the old Chinggisid characteristics and processes developed inde-
pendently in the absence of the strong unifying Chinggisid rule.
Another point to keep in mind is that the rise of the “old” and “new” tribal
markers in the Steppe lexicon around the “transition phase” is closely connected to
the cross-Eurasian migrations of the nomadic groups that took place both during
the heyday of Mongol Eurasia and following its decline since the mid-14th centu-
ry.70 It is in this context that both the “retribalization” issue as well as the appear-
ance of the new tribal and ethnic identities in the post-Crisis Steppe get their spe-
cial importance as parts of a much broader question of ethnic continuity and
change in the Eurasian Steppe during the 13th and 14th centuries. It is a well-known
fact that the famous military reform of Chinggis Khan broke the borders of many
tribal groups and gathered various peoples under one military unit. 71 With time, it
seems, many of those military units became major reference points for the peoples
involved, and so the new ethnic-military groups, such as the Negüderi-Qara’unas
in Chaghadaid Afghanistan or Ja’un-i Qurban in the post-Ilkhanid Khurasan, came
to life.72 Importantly, however, not all tribal structures were broken or eliminated.
66 Thus, e.g, AL-QALQASHANDĪ (n.a.), fl. 69v [today’s MS Qq. 36, Cambridge University Lib.,
see BROWN 1900, 113], as quoted in TIESENHAUSEN 1884, 402 (the manuscript was not availa-
ble to me). One wonders whether the name “Uzbek” is more than a reference to Özbek
Khan, who was long dead at that time.
67 SHĀMĪ, 1363/1984–85, 75.
68 Cf. AKHMEDOV 1965, 15–16, who insists on calling the Turkic speaking population of the
whole Desht-i Qipchaq “nomadic Uzbeks” (kochevye uzbeki) already during the period under
discussion. Most probably this should be seen as a political (but much less, if at all, an eth-
nic) identity marker (for a more careful but not completely contradictory approach, see
GOLDEN 1992, 330).
69 Note TREPAVLOV 2020, 67–85; LANDA 2020, 98.
70 Note esp. ALLSEN 2015b.
71 E.g. BIRAN 2007, 41–3.
72 For the first ones, see JACKSON 2018, for the second ones PAUL 2011; LANDA 2016b, 171–2,
174–5. Golden points out the very personalized nature of the new ethnic formation process-
282 ISHAYAHU LANDA
In some notable cases, especially when the specific tribal elite became involved in
matrimonial relations with the Chinggisids and got the status of imperial in-laws,
the tribal framework and affiliation seem to have been preserved even after the
tribes were dispersed across Eurasia. 73 Surely, we cannot claim that the post-
Mongol Eurasian Qonggirad, Oyirads, Jalayir, Manghit, etc. had any substantial
connection with the old groups with the same ethnonyms. The very survival and
also the dispersion of the Chinggisid ethnonyms all across the Steppe after the 14 th
century suggest, however, the lasting relevance of the Chinggisid domination for
the post-Chinggisid nomadic Eurasia.74
The paper suggested to single out the first decades of the Eurasian Steppe histo-
ry after the Great Crisis, i.e. the second half of the 14 th through the early 15th centu-
ry, as a separate phase, the analysis of which might highlight the historical devel-
opments that took place amid the decline or disappearance of the Chinggisid pow-
er. This “transition phase” of the Eurasian Steppe history is different from Ching-
gisid Eurasia, as it came after the Crisis decades that shattered the Chinggisid dyn-
asties in all four major uluses. Though partly overlapping with it, this phase is also
different from the Crisis as such, facing already the new political realities that fol-
lowed the major breakup of the Chinggisid power. It is, however, also different
from the post-early-15th century historical realities, as many of the actors and the
political structures participating in the developments of this phase were born or
already appeared during the heyday of Chinggisid Eurasia. Three dimensions of
change have been addressed, the geopolitical-ideological, the religious, and the
ethnic. As shown, all these dimensions combined the examples and precedents
based on the Chinggisid principle and/or the historical contexts of Mongol Eurasia
with the developments mirroring both the Crisis of the Chinggisid rule and new,
alternative modes of power applied by both the non-Chinggisid and Chinggisid
actors across the Steppe. Still not completely “new”, but also not completely “old”,
the historical events of the first decades after the Great Crisis make up a unique
period for research, during which the decline and the transformation of the old
Chinggisid realities coincided both with radically new historical processes, as well
as with circumstances that intricately encompassed both pre-Crisis and post-Crisis
realities, institutions, and actors.
es in the aftermath of the Crisis. In many cases the self-naming of the new entities was relat-
ed to specific “tribal or confederational names of clearly anthroponymic origin” (e.g.
Nogays, Chaghadai, Uzbek, etc., see GOLDEN 2000, 39). Golden connects this development to
the importance of the nökör/nöker and keshig institutions in Chinggisid Eurasia (GOLDEN 2000,
38–41; on the nökör [companions], see DOERFER 1963, 521–26, § 388; on the keshig, see DOERFER
1963, 467–70, §331–34). I thank Dr. Golev for bringing my attention to this phenomenon.
73 For some examples, see LANDA 2016b, 164–71.
74 For a broader discussion, see, e.g., LANDA 2017, 1201–5.
Bibliography
ABASHIN, S. – KRADIN, N. N. – SKRYNNIKOVA, T. D. – KIVELSON, V. A. – MUNKH-
ERDENE, L. – LIM, R. – EDGAR, A. – GLEBOV, S. (2009): Debating the Concepts of
Evolutionist Social Theory: Responses to David Sneath. Ab Imperio 4, 110–163.
AKHMEDOV, B. A. (1965): Gosudarstvo kochevykh uzbekov, Moscow.
ALLSEN, T. T. (2015a): Eurasia after the Mongols, in Bentley, J. H. – Subrahmanyam,
S. – Wiesner-Hanks, M. E. (ed.:) The Cambridge World History, vol. 6: The Con-
struction of a Global World, 1400–1800 CE, Part 1: Foundations. Cambridge, 159–181.
ALLSEN, T. T. (2015b): Population Movements in Mongol Eurasia, in Amitai, R. –
Biran, M. (ed.:) Nomads as Agents of Cultural Change. The Mongols and Their Eura-
sian Predecessors. Honolulu, 119–151.
AMITAI, R. (1999): Sufis and Shamans: Some Remarks on the Islamization of the
Mongols in the Ilkhanate. JESHO 41/1, 27–46.
AMITAI, R. – BIRAN, M. (2015): Introduction, in Amitai, R. – Biran, M. (ed.:) Nomads
as Agents of Cultural Change. The Mongols and Their Eurasian Predecessors. Hono-
lulu, 1–9.
ATWOOD, Ch. (2015): The Administrative Origins of Mongolia’s ‘Tribal’ Vocabu-
lary. Eurasia: Statum et Legem 4, 7–45.
BAI, C. (2006): Wala shi 瓦剌史 [The History of the Oyirad], Guilin.
BIRAN, M. (2002): The Chaghadaids and Islam: The Conversion of Tarmashirin
Khan (1331–34). Journal of the American Oriental Society 122/4, 742–752.
BIRAN, M. (2007): Chinggis Khan, Oxford.
BIRAN, M. (2009): Central Asia from the Conquest of Chinggis Khan to the Rise of
Tamerlane: The Ögödeid and Chaghadaid Realm, in Di Cosmo, N. – Frank,
A.J. – Golden, P. (ed.:) The Cambridge History of Inner Asia. The Chinggisid Age.
Cambridge, 46–66.
BORBONE, P. G. (2017): Marāgha Mdittā Arškitā: Syriac Christians in Marāgha Un-
der Mongol Rule. Egitto e Vicino Oriente 40, 109–143.
BORISOV, N. S. (1999): Politika moskovskikh knyazei, konets XIII - pervaya polovina XIV
vv., Moscow.
BORISOV, N. S. (2017): K voprosu o perenose mitropolichʹei kafedry iz Kieva vo
Vladimir, in Borisov, N.S. et al. (ed.:) Rusʹ, Rossiya: Srednevekovʹe i Novoe vremya.
Issue 5, Pyatye chteniya pamyati akademika RAN L.V. Milova, Moscow, 44–51.
BROOKE, J. L. (2014): Climate Change and the Course of Global History: A Rough Journey,
New York.
BROWN, E. G. (1900): A Hand-list of the Muhammadan Manuscripts, Including All Those
Written in the Arabic Character, Preserved in the Library of the University of Cam-
bridge, Cambridge.
CHERNIAVSKY, M. (1959): Khan or Basileus: An Aspect of Russian Mediaeval Politi-
cal Theory. Journal of the History of Ideas 20/4, 459–476.
284 ISHAYAHU LANDA
DE CLAVIJO, R. G. (2009): Narrative of the Embassy of Ruy González de Clavijo to the

court of Timour, at Samarcand, A.D. 1403–6, transl. by C.R. Markham, Cambridge
[repr.].
DASHDONDOG, B. (2011): The Mongols and the Armenians (1220–1335), Leiden.
DAI, M. (2008): A Mongolian Decree from the Chaghataid Khanate Discovered at
Dunhuang, in Zieme, P. (ed.:) Aspects of Research into Central Asian Buddhism: In
Memoriam Kōgi Kudara. Turnhout, 159–178.
DEWEESE, D. (1994): Islamization and Native Religion in the Golden Horde, University
Park, PA.
DEWEESE, D. (2009): Islamization in the Mongol Empire, in Di Cosmo, N. – Frank,
A. J. – Golden, P. (ed.:) The Cambridge History of Inner Asia. The Chinggisid Age.
Cambridge, 120–134.
DICKENS, M. (2015): Le christianisme syriaque en Asie centrale, in Borbone, P.G. –
Marsone, P. (ed.:) Le christianisme syriaque en Asie centrale et en Chine. Paris, 5–39.
DMITRIEV, L. A. (1963): Rolʹ i znachenie mitropolita Kipriana v istorii drevnerusskoi
literatury: K istorii russko-bolgarskikh literaturnykh svyazei XIV—XV vv.,
Trudy Otdela drevnerusskoi literatury 19, 215–254.
DOERFER, G. (1963): Türkische und mongolische Elemente im Neupersischen. Unter be-
sonderer Berücksichtigung älterer neupersischer Geschichtsquellen, vor allem der
Mongolen- und Timuridenzeiten, vol. 1. Wiesbaden.
DURAND-GUÉDY, D. (2011): Review: The Headless State, Aristocratic Orders, Kin-
ship Society and Misrepresentations of Nomadic Inner Asia. By David Sneath.
New York: Columbia University Press, 2009. International Journal of Asian Stud-
ies 8/1, 119–122.
ELVERSKOG, J. (2010): Buddhism and Islam on the Silk Road, Philadelphia.
GAI, SH. (1991): Yinshan Wanggu 阴山汪古 [The Önggüd of the Yin Mountain],
Hohhot.
GALIMOV, T. R. (2016): Vliyanie Ordy na kanonichesko-pravovoi status mi-
tropolichʹei kafedry i russkoi tserkvi vo vtoroi polovine XIII - nachala XIV vv.
Drevnyaya Rusʹ: vo vremeni, v lichnostyakh, v ideyakh 6, 428–452.
GAYDUKOV, P. G. (2007): Novye materialy o nachale russkoi monetnoi chekanki v
XIV veke. Vestnik Rossiiskogo gumanitarnogo nauchnogo fonda 3, 22–35.
GAYDUKOV, P. G. (2009): Imennye denʹgi velikogo knyazya Dmitriya Ivanovicha
Donskogo (Tipologiya i khronologiya), in Makarov, N.A. (ed.:) Velikii Novgorod
i Srednevekovaya Rusʹ. Sbornik statei: K 80-letiyu akademika V.L. Yanina. Moscow,
323–364.
GOLDEN, P. (1992): An Introduction to the History of the Turkic Peoples: Ethnogenesis
Wiesbaden.
GOLDEN, P. (2000): “I Will Give the People unto Thee”: The Činggisid Conquests
and Their Aftermath in the Turkic World, Journal of Royal Asiatic Society 10:1,
21–41.
GOLDSTONE, J. (1998): The Problem of the “Early Modern” World. Journal of the Eco-
nomic and Social History of the Orient 41/3, 249–284.
ḤĀFĪZ-I ABRŪ, ʿAbd Allāh ibn Luṭf Allāh (1993/1372–1373): Zubdat al-Tawārīkh, ed.
by Sayyid Kamāl Hājj Sayyid Jawādī. Tehran.
HALBERTSMA, T. H. F. (2015): Early Christian Remains of Inner Mongolia. Discovery,
Reconstruction and Appropriation, Leiden.
HALPERIN, Ch. (1976): The Russian Land and the Russian Tsar: The Emergence of
Muscovite Ideology, 1380-1408. Forschungen zur osteuropäischen Geschichte 23, 7–103.
HALPERIN, Ch. (1987): Russia and the Golden Horde. The Mongol Impact on Medieval
Russian History. Bloomington and Indianapolis.
HATEF NAIEMI, A. (2020): The Ilkhanid City of Sultaniyya: Some Remarks on the
Citadel and the Outer City. Iran (published online), 1–30. DOI:
10.1080/05786967.2020.1744469.
HONDA, M. (1958): On the Genealogy of the Early Northern Yüan. Ural-Altaische
Jahrbücher 30, 232–248.
IBN KHALDŪN, ʻAbd al-Raḥmān al-Mughrabī (1865): Kitāb al-ʻibar wa dīwān al-
mubtadaʼ wa al-khabar fī ayyām al-ʻarab wa al-ʻajam wa al-barbar wa man ʻāṣarahym
min dhawī al-sulṭān al-akhbar wa-huwa tarīkh waḥīd ʻaṣrih, Bulaq.
JACKSON, P. (1990): Review of “The Rise and Rule of Tamerlane by Beatrice Forbes
Manz”. Journal of the Royal Asiatic Society 2, 399–401.
JACKSON, P. (2005): The Mongols and the Faith of the Conquered, in Ryan, J.D.
(ed.:) The Spiritual Expansion of Medieval Latin Christendom: The Asian Missions.
Surrey, 297–342.
JACKSON, P. (2018): The Mongols of Central Asia and the Qara’unas. Iran 56/1, 91–103.
KADOI, Y. (2009): Buddhism in Iran under the Mongols: An Art-historical Analysis,
in Gacek, T. – Pstrusińska J. (ed.:) Proceedings of the Ninth Conference of the Euro-
pean Society for Central Asian Studies. Newcastle-upon-Tyne, 171–180.
KHAZANOV, A. M. (1993): Muhammad and Jenghiz Khan Compared: The Religious
Factor in World Empire Building. Comparative Studies in Society and History
35/3, 461–479.
KHAZANOV, A. M. (2010): Review of David Sneath, The Headless State: Aristocratic
Orders, Kinship Society, and the Misrepresentation of Nomadic Inner Asia. So-
cial Evolution and History 9/2, 206–208.
KIM, H. (1996): The Rise and Fall of the Hami Kingdom (ca. 1389-1513), in Desert
Route Expedition International Seminar (1990: Urumqi, China): Land Routes of
the Silk Roads and the Cultural Exchanges Between the East and West Before the 10 th
Century. Beijing, 89–95.
KIM, H. (1999): The Early History of the Moghul Nomads – The Legacy of the Cha-
ghatai Khanate. In Amitai-Preiss, R. – Morgan, D.O. (ed.:) The Mongol Empire
and its Legacy. Leiden, 290–318.
KIM, H. (2009): The Unity of the Mongol Empire and Continental Exchanges over
Eurasia. Journal of Central Eurasian Studies 1, 15–42.
286 ISHAYAHU LANDA
KLEIN, W. (2000): Das nestorianische Christentum an den Handelswegen durch Kyrgyz-

stan bis zum 14 Jh. Turnhout.
KOKOVTSOV, P. (1906): Khristіansko-sirіiskіya nadgrobnyya nadpisi iz” Almalyka.
Zapiski Vostochnogo Otdeleniya Rossiiskogo Arkheologicheskogo Obshchestva 16,
190–200.
KOLLMAR-PAULENZ, K. (2003): Der Buddhismus als Garant von »Frieden und
Ruhe«. Zu religiösen Legitimationsstrategien von Gewalt am Beispiel der ti-
betisch-buddhistischen Missionierung der Mongolei im späten 16. Jahrhun-
dert. Zeitschrift für Religionswissenschaft 11/2, 185–207.
KOTKIN, S. (2007): Mongol Commonwealth?: Exchange and Governance across the
Post-Mongol Space. Kritika: Explorations in Russian and Eurasian History 8/3,
487–531.
KRADIN, N. N. (2012): The Headless State: Aristocratic Orders, Kinship Society, and
Misrepresentations of Nomadic Inner Asia by David Sneath (review). Asian
Perspectives 51/1, 130–132.
LANDA, I. (2016a): Imperial Sons-in-law on the Move: Oyirad and Qunqirad Dis-
persion in Mongol Eura-sia. Archivum Eurasiae Medii Aevi 22, 161–198.
LANDA, I. (2016b): Oyirads in the Ilkhanate and the Mamluk Sultanate in the Thir-
teenth to Fourteenth Centuries: Two Cases of Assimilation into the Muslim
Environment. Mamluk Studies Review 19, 149–191.
LANDA, I. (2017): Türaqai Güregen (d. 1296–7) and His Lineage: A History of a
Cross-Asia Journey. Asiatische Studien/Études Asiatiques 71/4, 1189–1211.
LANDA, I. (2018a): From Mongolia to Khwarazm: The Qonggirad Migrations in the
Jochid Ulus (13c.–15c.). Revue des mondes musulmans et de la Méditerranée 143,
215–231.
LANDA, I. (2018b): Reconsidering the Chinggisids’ Sons-in-Laws: Lessons from the
United Empire. Chronica 18, 212–225.
LANDA, I. (2020a): Manghit tribal groups. Encyclopaedia of Islam THREE 2, 97–102.
LANDA, I. (2020b): Márton Vér, Old Uyghur Documents Concerning the Postal Sys-
tem of the Mongol Empire (Berliner Turfantexte XLIII). 263 pp. Turnhout, Bel-
gium: Brepols, 2019. ISBN: 978-2-503-58417-1. EUR 55,00 excl. tax (review).
Zentralasiatische Studien 48, 505–508.
LANDA, I. 2022. The Islamization of the Mongols, in May, T. – Hope, M. (ed.:) The
Mongol World. London/New York, 642–661.
MANZ, B. (1988): Tamerlane and the Symbolism of Sovereignty. Iranian Studies
21/1–2, 105–122.
MANZ, B. (1998): Temür and the Problem of a Conqueror’s Legacy. Journal of the
Royal Asiatic Society 8/1, 21–41.
MCCHESNEY, R.D. (2009): The Chinggisid Restoration in Central Asia: 1500–1785, in
Di Cosmo, N. – Frank, A.J. – Golden, P. (ed.:) The Cambridge History of Inner
Asia. The Chinggisid Age. Cambridge, 277–302.
MIYAWAKI, J. (1992): The Chinggisid Principle in Russia. Russian History 19/1, 261–277.
MUʿIZZ AL-ANSĀB: Muʿizz al-ansāb, MS Persan 67, Bibliothéque nationale de France.

MUNKH-ERDENE, L. (2011): Where Did the Mongol Empire Come From? Medieval
Mongol Ideas of People, State and Empire. Inner Asia 13, 211–237.
MUSTAFAEV, Sh. (2015): Timuridskoe posolʹstvo v Kitai i ego opisanie na azer-
bajdzhanskom yazyke soglasno rukopisi XV v. Izvestiya NAN Azerbaidzhana.
Seriya obshchestvennyh nauk 3, 98–115.
MUSTAKIMOV, I. A. (2009): Ob odnom spiske ‘Daftar-i Chingis-name’, in Khakimov,
R.S. (ed.:) Srednevekovye tyurko-tatarskie gosudarstva. Kazan, 122–131.
NAṬANZĪ, Muʿīn al-Dīn (1336/1957–58): Muntakhab al-tawārīkh-i muʿīnī, ed. Jean
Aubin, Tehran.
NIKONOVSKAYA LETOPIS’ (1885): Nikonovskaya letopis’, PSRL, vol. 10, Saint-
Petersburg.
NIKONOVSKAYA LETOPIS’ (1897): Nikonovskaya letopis’, PSRL, vol. 11, Saint-
Petersburg.
OBOLENSKY, D. (1978): A "Philorhomaios Anthropos": Metropolitan Cyprian of
Kiev and All Russia (1375–1406). Dumbarton Oaks Papers 32, 77–98.
OKADA, H. (1987): Origins of Dörben Oyirad.” Ural-Altaische Jahrbücher 7, 181–211.
PACHKALOV, A. V. (2012): K voprosu ob imennykh monetakh Mamaya. Numizmat-
ika Zolotoi Ordy 2, 117–119.
PAUL, J. (2011): Zerfall und Bestehen. Die Ǧaun-i Qurban im 14. Jahrhundert. Asia-
tische Studien / Études Asiatiques 65/3, 695–733.
PAVLIKHIN, A. V. (2014): Klad zolotoordynskikh ukrashenii i pozolochennykh mo-
net s I Vyshenskogo gorodishcha. Numizmatika Zolotoi Ordy 4, 32–38.
PETECH, L. (1983): Tibetan Relations with Sung China and the Mongols, in Rossabi,
M. (ed.:), China Among Equals: The Middle Kingdom and Its Neighbours, 10 th–14th
Centuries. Berkeley, 173–203.
PETECH, L. (1990): Central Tibet and the Mongols: The Yüan – Sa-Skya Period of Tibetan
History, Rome.
PETROV, P. N. (2005): Tamgi na monetakh mongolʹskikh gosudarstv XIII-XIV vv.
kak znaki sobstvennosti, in Trudy mezhdunarodnoi numizmaticheskoi konferentsii
“Monety i denezhnoe obrashchenie v mongolʹskikh gosudarstvakh XIII-XV vekov”,
Murom 6-12 sentyabrya 2003. Moscow, 170–177.
PETROV, P. N. (2013): Nekotorye aspekty srednevekovoi vostochnoi numizmatiki—
istoricheskogo istochnika. Zolotoordynskoe obozrenie 1, 177–206.
POCHEKAEV, R. Yu. (2019): Tsari ordynskie: biografii khanov i pravitelei Zolotoi Ordy, 2d
ed., Saint-Petersburg.
PRAZNIAK, R. (2014): Ilkhanid Buddhism: Traces of a Passage in Eurasian History.
Comparative Studies in Society and History 56/3, 650–680.
RASHĪD AL-DĪN, Faḍl-Allāh (n.a.): Shuʿab-i panjgānah, MS Ahmet III 2937, Topkapı-
Sarayı, Istanbul.
RASHĪD AL-DĪN, Faḍl-Allāh (1373/1994–5): Jāmiʿ al-Tawārīkh, edited by M. Rawshan
and M. Mūsawī, 4 vols., Tehran.
288 ISHAYAHU LANDA
ROBINSON, D. M. (2012): Mongolian Migration and the Ming’s Place in Eurasia.

Journal of Central Eurasian Studies 3, 109–129.
ROBINSON, D. M. (2020): In the Shadow of the Mongol Empire. Ming China and Eurasia,
Cambridge.
SAFARGALIEV, M. G. (1960): Raspad Zolotoi Ordy, Saransk.
SELA, R. (2004): Central Asia in the 18th Century: The Age of Introspection. PhD Disser-
tation, Indiana University. Unpublished.
SELA, R. (2014): Central Asian Muslims on Tibetan Buddhism, 16 th–18th Centuries,
in Vitali, R. (ed.:) Trails of the Tibetan Tradition: Papers for Elliot Sperling. Dhar-
amshala, 345–358.
SHĀMĪ, Nizām al-Dīn 1363/1984–1985: Zafarnāmah: Tārīkh-i futūhāt-i Amīr Timūr
Gūrkānī, Tehran.
SLAVIN, Ph. (2019): Death by the Lake: Mortality Crisis in Early Fourteenth-Century
Central Asia. Journal of Interdisciplinary History 50/1, 59–90.
SLAVIN, Ph. (2022): A Rise and Fall of a Chaghadaid Community: Demographic
Growth and Crisis in ‘Late-medieval’ Semirech’ye (Zhetysu), circa 1248–1345.
Journal of Royal Asiatic Society, 1–32. doi: 10.1017/S135618632200044x
SNEATH, D. (2007): The Headless State. Aristocratic Orders, Kinship Society, and Misrep-
resentations of Nomadic Inner Asia, New York.
SPASSKY, I.G. (1962): Russkaya monetnaya sistema, Leningrad.
SPYROU, M. et al. (2022): The Source of the Black Death in Fourteenth-century Cen-
tral Eurasia. Nature 606, 718–724.
SUBRAHMANYAM, S. (1997): Connected Histories: Notes towards a Reconfiguration
of Early Modern Eurasia. Modern Asian Studies 31/3, 735–762.
TIZENGAUZEN, V. G. (1884): Sbornik materialov otnosyashchikhsya k istorii Zolotoi Ordy,
vol. 1, Saint-Petersburg.
TOGAN, İ. (1998): Flexibility and Limitation in Steppe Formations: The Kerait Khanate and
Chinggis Khan, Leiden.
TREPAVLOV, V. V. (2014): Beklerbek v strukture mongolʹskoi i tyurkskoi gosudar-
stvennosti. Zolotoordynskoe obozrenie 2, 21–34.
TREPAVLOV, V. V. (2020): Istoriya nogaiskoi ordy, Moscow.
VAN MECHELEN, J. (2001): Yuan, in Standaert N. (ed.:) Handbook of Christianity in
China, Volume One: 635–1800, Leiden.
VÁSÁRY, I. (2009): The Beginnings of Coinage in the Blue Horde. Acta Orientalia
Academiae Scientiarum Hung. 62/4, 371–385.
VEIT, V. (2009): The Eastern Steppe: Mongol Regimes after the Yuan (1368–1636), in
Di Cosmo, N. – Frank, A. J. – Golden, P. (ed.:) The Cambridge History of Inner
Asia. The Chinggisid Age. Cambridge, 157–181.
WANG K. (2014): Ming dai jingjiao de daojiao hua – xin faxian yi pian daojiao bei-
wen de jiedu 明代景教的道教化—新發現一篇道教碑文的解讀 [Syriac Christian-
ity of the Ming Period and Its Getting Daoist – Deciphering of the Newly Re-
covered Daoist Epigraphic Inscription]. Xueshujie 196, 194–198.
WILLIAMS, S. W. (1883): The Middle Kingdom, vol. 2, London.

WILMSHURST, D. (2000): The Ecclesiastical Organisation of the Church of the East, 1318–
1913, Louvain.
WANG, F. (2015): Negotiating between Mongol Institutions and “Han Traditions”:
Buddhist Administration in Southeastern China under the Yuan. Journal of
Song-Yuan Studies 45, 339–369.
WING, P. (2016): The Jalayirids. Dynastic State Formation in the Mongol Middle East,
Edinburgh.
WOODS, J. E. (1993): The Timurid Dynasty, Bloomington, IN.
ZAYTSEV, I. V. (2004): Astrakhanskoe khanstvo, Moscow.
The Oriental Coin Hoard from
Ružomberok, Slovakia
MICHAL HOLEŠČÁK
Abstract
A unique find of five copper coins was unearthed during the Second World War
north of the city of Ružomberok in Slovakia. It consists of four identified coins,
minted from the 9th to the beginning of the 15th century mainly in Central Asia, and
one unidentified coin. The composition and nature of the hoard, as well as the de-
scribed circumstances of its discovery, point out that the coins were not in mone-
tary circulation, but rather a souvenir with sentimental value, and they prove some
kind of connection between the Hungarian Kingdom and the polities of Central
Asia at the beginning of the 15th century. The present paper attempts to localize the
find and connect it with known historical sources about the diplomatic relations of
King Sigismund of Luxembourg with the Tatars, Turcomans, and Timurids.
Keywords: coins, Sigismund of Luxembourg, Hungarian Kingdom, Aq Qoyunlu,
Shāh Rukh, diplomatic relations, Central Asia
Circumstances of the finding
A coin hoard got into the Institute of Archaeology of the Slovak Academy of Sci-
ences under peculiar circumstances. In 1961, during a research stay in the Federal
Republic of Germany, A. Točík, who was at that time the head of the Institute, ac-
quired through his acquaintances a set of coins together with a letter describing
their origin. It was found by K. Wendel, a soldier in the German army during the
Second World War, while digging a trench in the vicinity of the city of Ružomber-
ok on the right bank of the Váh river in 1944. The city of Ružomberok is located in
northern Slovakia, in the region of Liptov, surrounded by the mountain ranges of
 Department of Medieval and Early Modern Age Archaeology, Institute of Archaeology of

the Slovak Academy of Sciences in Nitra. An earlier version of this paper was presented at
the 8th International Conference on Medieval History of the Eurasian Steppe Nomads and
their Neighbours in the Middle Ages held in Sofia, November 20–3, 2019. Work was supported
by the grant “Role of material culture in forming of economic, social and interethnic contacts
in Medieval communities” from VEGA agency.
292 MICHAL HOLEŠČÁK
the Greater Fatra and the Low Tatras. The hoard was found at a depth of ca. 1.6 m
in a burned layer together with the remains of some kind of leather, possibly part
of some kind of a pouch around them. The finder took the coins with him back
home, but his conscience troubled him and he tried to contact someone from
Czechoslovakia so that he could return them to their place of origin after the war
ended. The credibility of this statement can be doubted, but the general identifica-
tion seems to be believable, since there is no obvious motive for K. Wendel to lie
and therefore it can be accepted as truth. After their return to Czechoslovakia, the
coins were identified by J. Štěpková from Prague and stored in the treasury of the
Institute of Archaeology in Nitra. They were published in the catalogue of numis-
matic finds from Slovakia,1 but without further inquiry. Recent re-evaluation of the
coins confirmed their provenance, produced up-to-date pictorial documentation,
and analysed the composition of the metal by spectrometry.
Description of the coins
Chronologically, the first coin (Fig. 1: 2) was struck in Samarkand by the amir of
Khurāsān Ghassān b. ʿAbbād during the reign of the ʿAbbāsid caliph al-Maʾmūn
(r. 813–33) in 205 AH (820–1 AD).2 The main components of the metal are copper
(89%), zinc (8%), and lead (2%). The diameter of the coin is 19 mm, its weight—
2.576 g. It bears a little wear on the averse, and the reverse side is well preserved.
The second coin (Fig. 1: 3) was minted in Bilär in 373 AH (983–4 AD) by the
Sāmānid amir Nūḥ II b. Manṣūr (r. 976–97), bearing the name of the ʿAbbāsid ca-
liph al-Ṭāʾiʿ (r. 974–91).3 The metal composition is mainly copper (93.5%), lead
(3%), and tin (2.6%). It has a diameter of 28 mm and a weight of 4.436 g. The coin
bears a little wear on both sides.
The third identified coin (Fig. 1: 1) was struck in the name of Khan Janibeg (r.
1342–57 AD). The reverse is not fully readable, but the closest analogy is a
Khwārezmian pūl from 757 AH (1356 AD).4 The metal composition is nearly pure
copper (99.3%). It has a diameter of 17 mm and a weight of 3.183 g.
The fourth coin (Fig. 2: 2) was previously not identified, but it is possible to
connect it with the coin of Shāh Rukh (r. 1405–47), struck in Samarqand AH 819
(1416–7 AD).5 Although heavily worn, it seems that it bore the three-circle sign of
Timur (r. 1370–1405) (Fig. 2: 2b).6 The metal composition is again nearly pure cop-
per (99.5%). It’s shape is oval, with a diameter of 28x23 mm and a weight of 3.358 g.
1 KOLNÍKOVÁ – HUNKA 1994, 114.

2 NÜTZEL 1898, 379, cat. nr. 2212.
3 TORNBERG 1848, 241, cat. nr. 577.
4 FEDOROV-DAVYDOV 2003, 185, cat. nr. 175.
5 I express my gratitude to the late V. Novák from the Náprstek Museum for the help with the
identification: DAVIDOVICH 1983, Fig. 4: 5; https://www.zeno.ru/showphoto.php?photo=181149

6 KADOI 2010, 149–51; NYAMAA 2005, 82.
THE ORIENTAL COIN HOARD FROM RUŽOMBEROK, SLOVAKIA 293
The last coin (Fig. 2: 1) is countermarked, with the original coin being heavily
worn out. The only visible feature is the hexagram with a dot in each point, not an
uncommon sign on the Jochid coinage. The countermark (Fig. 2: 1b) is unreadable,
but it was identified by J. Štěpková as belonging to Ariq Böke (r. 1259–64) of the
Toluid family. The tamgha of Ariq Böke is usually found on the coins of Berke,
struck in Bulghar or sometimes in Chagatayid mints, 7 but not on countermarks,
according to my knowledge. Also, this specimen is currently too damaged to be
read and the original identification unfortunately lacks any reference to scholarly
literature to corroborate this claim. The coin is almost purely copper (99.4%) with a
diameter of 25 mm and a weight of 3.715 g.
As all the coins are of oriental provenance, which makes them rather unique for
the region they were found in, there is no doubt that they were buried in this par-
ticular place in one deposition, regardless of their wide time span. The earliest date
of the hoard deposition can be based on the youngest identified coin in the set. If
the unidentified coin is not younger, the latest coin is the Timurid pūl of the early
15th century, possibly connected to Shāh Rukh. The attempts at finding a plausible
context in which they got to the area of the Hungarian kingdom, specifically to
Ružomberok, are therefore focused on this time period. Also, as all of them are
made of copper, they most likely weren´t part of the currency in use, as their value
would have been minimal. It can be assumed, therefore, that they had a more sym-
bolical or personal value than an economic one.
Localization of the hoard
According to the finder of the hoard, the coins were found on the right bank of the
river Váh, close to Ružomberok, a free royal town and a semi-important merchant
hub in the region, established at the beginning of the 14 th century by German colo-
nists from the surrounding areas.8 The city itself lies on the left bank of the river, so
the possible location of the find lies to the north of it. For the purpose of the locali-
zation, we must also consider the military activities of the German army in 1944,
when the hoard was uncovered. The political situation was very complex as it
happened during the Slovak National Uprising, an anti-fascist movement against
the Slovak Republic, a client state of Nazi Germany. The uprising started by the
Soviet-supported local guerrillas fighting mostly in the hilly areas of the central,
northern, and eastern parts of Slovakia, followed by a part of the Slovak army turn-
ing rebellious, which led to the arrival of the German army, whose ranks probably
included the finder of the hoard. Ružomberok was liberated from the small num-
ber of Germans deployed in the city even before the official beginning of the upris-
ing, and the rebel forces continued towards the east to assist in the liberation of the
city of Liptovský Mikuláš. After the uprising officially erupted, these forces had to
7 NYAMAA 2005, 63–5.

8 RUŽOMBEROK 2009, 44.
turn back, as they expected a German offensive led by a full SS battalion from east
and north. The first major battle was at the village of Lisková and continued
through Martinček, after which the Germans took the city of Ružomberok very
quickly. After that, they effectively blocked the approach of the rebellious forces
from the north, possibly in the vicinity of the village of Likavka. When the situa-
tion stabilized, German forces focused on destroying the guerrilla outposts and
hideouts in the hills around, mainly to the northeast of the city. 9
Unfortunately, there are no archaeological excavations in the described locality,
with only two sites worth mentioning as exceptions to this trend. The first one is on
the prehistoric hill fort Konislav in the cadastre of Lisková. The excavations re-
vealed pottery shards from the 14th – 15th centuries10 which corresponds with the
timeline of the possible deposition of the hoard. The presence of the German army
may be explained by the anti-guerrilla operations in the woods, as the hill fort is
situated on a hilltop rising 735 metres above sea level. It is, however, questionable
if the Germans dug trenches in such a terrain and circumstances. The second, more
plausible location is the cadastre of Likavka. During the archaeological survey
caused by the need of highway construction, a circular feature identified as possi-
ble WW2 military entrenchment was spotted, but not excavated.11 It is very possi-
ble that the digging of the entrenchment was connected to the German garrison of
Ružomberok blocking the passage of rebellious troops from the north. The current
village of Likavka lies under the castle of Likava, an important defence and control
point on the crossroad of north-south and east-west commercial roads built in the
first quarter of the 14th century. Likava was conquered by the Hussites in 1431
through the treachery of the castle warden, and a Hussite garrison stayed there
until the end of 1434, despite the attempts of its official owner Queen Barbara of
Cilli to gather an army and retake it. The city of Ružomberok and its surroundings
were threatened by two more Hussite armies passing through the region at this
time. The castle of Likava was then given to John Hunyadi, who started a massive
reconstruction, followed by Peter Komorowski, whose rebellion against King Ma-
thias Corvinus led to open military activity in this region and the capture of the
castle of Likava among others.12
Contacts between Hungary and the Orient at the beginning of the 15 th century
The Hungarian Kingdom had a major clash with the powers of Central Asia during
the Mongol invasion of 1241–2, preceded by the missions of Friar Julian to search
for the eastern Magyars, which were also caused by the growing Mongol threat.
After the establishment of the western part of the Mongol Empire, the Ulus of Jo-
9 RUŽOMBEROK 2009, 119–21.

10 HANULIAK 1976, 122.
11 BARTÍK et al. 2011, 27.
12 RUŽOMBEROK 2009, 51–2.
chi, at the border of the Hungarian Kingdom, the contacts of both military and
diplomatic nature continued. However, the policy of the Hungarian kings of the
House of Árpád was focused more on the regional struggles of the European pow-
ers than on the areas behind the vast steppes to the east. The situation slightly
changed with the coming of the new Anjou dynasty, 13 but a real diplomatic con-
nection was established only during the rule of Sigismund of Luxembourg (r.
1387–1437). It was caused by the rise of the Ottoman Empire, against which the
king was trying to find allies, the Tatars being one of them. 14 Tatar and other non-
Christian emissaries were received in Hungary as well as by the Polish king in
1412, and a Hungarian embassy was sent to Crimea later that same year, however
this time mainly focused on another enemy, the Republic of Venice. This embassy
was led by three ambassadors, Nicolaus of Geretz, Ladislaus of Salimavar, and
Jacob of Haugar. They travelled to Kaffa to ask the Genoese merchants for help in
the negotiations with the son and successor of Khan Tokhtamysh Jalāl al-Dīn (r.
1411–12), unfortunately with an unknown result, since by that time the khan could
have been already dead.15 A Tatar delegation was also present at the Council of
Constance (1414–19). The next eastern embassy was led by the brothers Johannes
and Conrad Fisher in 1418, members of an important merchant family with posses-
sions also in Košice, in today’s eastern Slovakia. Merchants of Košice, or Cassa, as
it was called, had strong trading connections with Crimea at the beginning of the
15th century.16 Merchants and diplomats once again travelled to Kaffa with the aim
of firmly establishing eastern trading routes that would bypass the Venetians.17
The coins, however, originate from much farther east than the Crimean Peninsula
and these lands were reached a year later by two adventurers and diplomats, Nico-
laus the Saracen and Joseph the Turk. In a royal diploma issued in the city of
Trnava in 1428, they were rewarded for their diplomatic mission to the eastern
steppe and Asia Minor. 18 Nicolaus of Geretz, who had already participated in the
first mission, was called “the Saracen” due to his twelve-year captivity after the
battle of Nikopol (1396).19 He was an envoy to the Turcoman state of Aq Qoyunlu,
which was ruled at the time by Qara-Yülük (r. 1378–1435), who was an ally of Ti-
mur and also of his son, Shāh Rukh, with whom he attacked the Ottomans. 20 Jo-
seph, called “the Turk”, was a diplomat of Turkish or Turkic21 descent who was
13 TARDY 1978, 5–6.

14 TARDY 1978, 10–11.
15 GULEVICH – KHAUTALA 2018, 200–2.
16 TARDY 1978, 49–50, ref. 13.
17 GULEVICH – KHAUTALA 2018, 203.
18 TARDY 1978, 12.
19 ÁGOSTON 1995, 273.
20 TARDY 1978, 14–15.
21 He later became the Judge of the Cumans in Hungary and was a practicing Muslim. TAR-
DY 1983, 14.
sent as an envoy to Ulugh Muḥammad (r. 1419–23, 1428–37), one of the strongest
rulers of the slowly decentralizing Golden Horde. Two years later, in 1430, a letter
from Qara Yülük to Sigismund speaks about his successful dealings with Shāh
Rukh and mentions again Nicolaus and Joseph, who evidently went on a second
diplomatic mission after the success of the first one. 22 The diplomatic relations be-
tween the two rulers continued, and Qara Yülük´s envoy Johannes Tartarus, or
John the Tatar, reached the city of Pozsony (nowadays Bratislava) probably via the
Danube, from where he went to meet the king in the city of Tyrnau (nowadays
Trnava) in 1435.23 Another famous traveller who was captured at the battle of Ni-
kopol, Johann Schiltberger, visited the eastern steppe as well as Siberia, while in
service of Timur, Shāh Rukh, and the Tatar prince Chakra. On his return from cap-
tivity in 1427, he passed close to the Hungarian borders, travelling from Moldavia,
through the city of Lviv, where he fell ill for three months, then further through
Cracow to his native Germany.24
After the death of King Sigismund, the Orient became distant and the threat of
the Ottomans grew even closer. The next chapter of the fight with the Turks was
overseen by the already mentioned John Hunyadi, who for a short time owned the
castle of Likava north of Ružomberok—from 1435 to 1449 at the latest. 25 The full-on
war with the Ottoman Empire hampered the attempts to maintain contacts with
Central Asia, and the few allies of Hungary were now to be found mostly among
the local European powers.
Conclusions
The presented coin hoard is unique on many levels, even though the circumstances
of its acquisition are not entirely clear. There is, however, no apparent motive for
the alleged finder to lie about the situation of the find, and the fact that he looked
after it after its discovery in 1944 and before handing it over to the officials of the
Institute of Archaeology in 1961 makes his statements seem rather reliable. There
are no analogous compositions of coins from the region of Central-Eastern Europe.
The absence of other coins from the closer geographical area and the fact that the
coins were all made of copper makes it possible to say that they were not part of a
monetary exchange or a source of precious metal that got to the region for its eco-
nomic value. The composition of coins spanning from the 9 th to the 15th century can
be interpreted as some sort of a collection, again underlining the sentimental char-
acter of the deposition.
No definitive conclusion can be made regarding the origin and the meaning of
this hoard, but the hints described in the present paper make the following hy-
22 TARDY 1978, 15–17.

23 TARDY 1978, 24–5, ref. 40–1.
24 SCHILTBERGER 1966.
25 RUŽOMBEROK 2009, 51.
pothesis the most plausible one in light of the current state of research. The coins
were acquired by the person who deposited them in Ružomberok right in their
place of origin, in Central Asia at the beginning of the 15 th century, based on the
terminus post quem of the youngest identified coin. Even though the contacts of the
Hungarian Kingdom and specifically its north-eastern part, with the Crimean cit-
ies, mainly Kaffa, are well documented, since there is no identified coin from the
Pontic area the hypothesis that they came from this milieu can be doubted. As
there are records of more than one exchange of ambassadors between the powers
of the Middle East and Central Asia and the Hungarian Kingdom during the reign
of king Sigismund of Luxemburg, it is more possible that the collector of the hoard
was taking part in at least one of them. People mentioned in diplomas, such as
Nicolaus of Geretz and Joseph the Turk, had no specific ties to the region of Liptov,
but it is very possible that they did not travel alone and that the royal embassy
coming from Hungary consisted of more people and hired local guides or transla-
tors on their way. One of these people could have resided in Ružomberok, bring-
ing with him a souvenir from his travels—not economically valuable, but for sure a
sentimentally important memento of his journey—in the form of exotic coins. The
hoard was buried possibly during the military clashes with the Hussite armies or
Peter Komoriwski during the 15th century, when the danger of armies in the vicini-
ty of Ružomberok and the villages north of the Váh River may have forced the
owner of the coins to hide them from the enemy. Centuries later, they were uncov-
ered by a German soldier digging a trench during a punitive expedition of the Nazi
army against the rebels of the Slovak National Uprising. Only further archaeologi-
cal excavations in the area can bring forth more data, either for or against this hy-
pothesis. A possible identification of the last of the coins by analogies can also
change the current view on this unique find.
The present article aims at gathering the whole available archaeological and his-
torical information about the coin hoard and put it in the scope of the relations
between Europe and Asia in the 15th century, when it was most probably deposited
in the ground. It does not dare to fully examine the coins from a numismatic point
of view, but an important facet of the study is the publication of photographical
evidence of the coins, previously unpublished in a graphical way, with the aim to
present them to the professional scientific community. Also, the identification of
the last coin could provide valuable information, as it is evidently an original coin
that was later countermarked. Last but not least, a spectroscopic analysis in the
future could add to the research on the metallurgical aspect of the oriental coins.
Acknowledgements
I have to sincerely thank Konstantin Golev for the organization of the conference,
István Zimonyi for the creation of the compendium of papers, and Yuka Kadoi and
Ishayahu Landa for the constructive consultation on the topic of Oriental coinage.
Last, but not least, I would like to thank and dedicate this article to the memory of
Ján Hunka (1958–2021), a friend, a colleague and a numismatist.
Bibliography
ÁGOSTON, G. (1995): Karamania, the Anti-Ottoman Christian diplomacy and the
non-existing Hungarian-Karamanid diplomatic relations of 1428. Acta Orien-
talia Academiae Scientiarum Hungaricae 48/3, 267–274.
BARTÍK, J. – BENEDIKOVÁ, L. – ČAMBAL, R. – CHOMA, I. – SLIACKA, S. (2011):
Doplňujúci archeologický výskum na trase diaľnice D1 Hubová – Ivachnová.
Archeologické výskumy a nálezy na Slovensku 2008, 27–28.
DAVIDOVICH, E.A. (1983): Istoriia denezhnogo obrashcheniia srednevekovoi Srednei Azii,
Moscow.
FEDOROV-DAVYDOV, G. A. (2003): Denezhnoe delo Zolotoi Ordy, Moscow.
GULEVICH, V. P. – KHAUTALA, R. (2018): Maloizvestnyi istochnik o posol’stve
Sigizmunda Liuksemburgskogo v Kaffu 1412 g. Zolotoordynskoe obozrenie =
Golden Horde Review 6/1, 199–211.
HANULIAK, V. (1976): Prieskum hradiska Konislav pri Liskovej. Archaeologické
výskumy a nálezy na Slovensku 1975, 122.
KADOI, Y. (2010): On the Timurid flag, in RITTER, M. – KORN, L. Beiträge zur
Islamischen Kunst und Archäologie. Wiesbaden, 143–162.
KOLNÍKOVÁ, E. – HUNKA, J. (1994): Nálezy mincí na Slovensku IV, Nitra.
NÜTZEL, H. (1898): Katalog der orientalischen Münzen, Berlin.
NYAMAA, B. (2005): The Coins of Mongol Empire and Clan Tamgha of Khans (XIII–XIV),
Ulaanbaatar.
RUŽOMBEROK (2009): Ružomberok. Monografia mesta, Banská Bystrica.
SCHILTBERGER, J. (1966): Bondage and Travels of Johann Schiltberger, transl. by J.
Buchan Telfer, Ayer Publishing.
TARDY, L. (1978): Beyond the Ottoman Empire, 14th – 16th century Hungarian diplomacy
in the east, Szeged.
TORNBERG, C. J. (1848): Numi Cuffici Regii Numophylocii Holmiensis, Uppsala.
Figures
Figure 1. Coin hoard from Ružomberok: 1. Janibeg; 2. Ghassān b. ʿAbbād; 3. Nūḥ II

b. Manṣūr
Figure 2. Coin hoard from Ružomberok: 1. unidentified countermarked coin; 2.

Shāh Rukh
Rakb al-Ḥajj al-Shāmī: Relations between
Nomads and Pilgrims during the
Mamluk and Ottoman Periods
HATIM MAHAMID & CHAIM NISSIM
Abstract
This paper deals with the relations between nomads (Bedouins) and the pilgrimage
caravans from Syria to the Hijaz (Rakb al-Ḥajj al-Shāmī) in the Late Middle Ages and
the early Ottoman period. It tries to follow the changes and developments in the
relations between nomads and pilgrims on the road between Syria and Mecca.
Before Islam, agreements and alliances (īlāf) were made between rulers of Mecca
and Bedouin tribes in Arabia to preserve security and safety in the area. In Islamic
times, these norms continued and even developed. In the Mamluk era and then
under the Ottomans, many arrangements for securing the pilgrimage routes and
the pilgrims’ stay in Mecca were renewed. The Bedouins’ dealings with pilgrims
ranged from peace and security to aggression and looting, according to the politi-
cal and economic arrangements that prevailed between the Bedouins and the rul-
ing authorities.
Keywords: nomads/Bedouins, Rakb al-Ḥajj (pilgrim caravans), Syria (al-Shām),
Mecca, raids, Mamluks, Ottomans
Historical background: relations between nomads and others in Arabia
Since the building of the Kaʿba, which Islamic tradition ascribes to the Prophet
Ibrāhim and his son Ismāʿīl, the people of Arabia, and then all Muslims, have con-
tinued to make the pilgrimage to Mecca, which developed as an important reli-
gious and commercial centre in the area. Even for the pre-Islamic period, the
Dr. Hatim Mahamid is a lecturer at Sakhnin College for Teachers Education, and The Open
University of Israel; Dr. Chaim Nissim is an emeritus of the Department of History, Philoso-
phy and Judaic Studies, The Open University of Israel. This article is a written version of a
paper presented at the 8th International Conference on Medieval History of the Eurasian
Steppe Nomads and their Neighbours in the Middle Ages, Sofia, Bulgaria, November 20–3, 2019.
The research for this article was funded by the Open University of Israel, grant no. 509829.
302 HATIM MAHAMID & CHAIM NISSIM
Quran mentions riḥlat al-shitā’ wal-ṣayf (travel of winter and summer). The roads to
Mecca pass through mountainous and desert areas inhabited by scattered Bedouin
tribes. According to the accepted custom, those who controlled the holy sites in
Mecca were responsible for order and security. In pre-Islamic times, it was agreed
that the pilgrimage seasons would be a period of peace and security called the
sacred months (al-ashhur al-ḥurum), during which wars were forbidden in Mecca
and its surroundings. But sometimes, fighting and conflicts broke out among local
tribes, and they were known as “the sinful wars” (ḥurūb al-fujjār), because they
violated the sanctity of the sacred months. (In the Islamic period, the Prophet
Muḥammad turned Yathrib [Medina] into a sacred [ḥaram] land as well.)1 To that
purpose, the people of Mecca in the pre-Islamic centuries had ensured the peace of
visitors to the city, both pilgrims and merchants, in three main areas: water supply
(siqāya), food (rifāda), guarding and security (ḥijāba/sadāna).2 Therefore, the central
tribe that controlled Mecca at the time (Quraysh) had good relations with the Bed-
ouin along the routes to Mecca, drawing up mutual agreements in which the Bed-
ouin received money and livelihood in return for peace and security. 3 With the rise
of Islam, these relations between Mecca and the pilgrims were maintained, ex-
panding to include other visitors for religious and commercial purposes, as indi-
cated in the Quran.4
The desert and arid geographical region of Arabia, between Yemen and the Le-
vant (Syria and Iraq), affected the nomadic population in the region both in life-
style and customs. The ancient customs continued to influence the lives of the no-
mads after their conversion to Islam, whereby the patterns of attitude toward “the
other” were still enshrined in past traditions. Some habits of raiding on agricultur-
al and settled areas or raids on the trade routes to obtain property held by mer-
chants or pilgrims to the holy places in the Hijaz continued. During the Islamic
times, and without respect to religious considerations, the Bedouins continued to
carry out attacks on trade and pilgrimage routes, exploiting periods of drought or
weak government.5
1 On al-ḥaram in Medina, see AL-SIJISTĀNĪ 2009, 3, 378–83.

2 See AL-AZRAQĪ 1983, 155, 197, 200.
3 On the ḥajj before Islam, see WEBB 2013, 6–14.
4 See the Quran, sūra 106 (Quraysh). There were two journeys to and from Mecca: the winter
one to Yemen and the summer one to Syria (al-Shām).

5 Quranic verses mention the habits and behaviour of the Bedouin, such as: “The aʿrāb
(ʿurbān/nomads) are the worst of disbelief and hypocrisy and should not know the limits of
what God (Allah) has revealed […]”. “(?)While among the aʿrāb (Bedouins), there are those
who consider what they spend in the cause of God (Allah) […]”. See Surat al-Tawba, (Chap-
ter 9), no. 97–8; In this sense, a saying of the Prophet says: “man sakana al-bādiya fa-qad jafā,
wa-man ittabaʿa al-ṣayd fa-qad ghafal”. That means that whoever sits in the desert remains dry,
and whoever hunts without attention focuses on the reward, so he neglects good deeds. See
AL-ALBĀNĪ 2002, 213–14.
RAKB AL-ḤAJJ AL-SHĀMĪ: NOMADS AND PILGRIMS RELATIONS… 303
All roads lead to Mecca

Even before Islam in Arabia there was a famous saying: “the people of Mecca
know its routes best” (ahl Makka adrā bi-shiʿābihā).6 This hinted at the difficulty of
coming to Mecca, accessing it, and identifying its ways, due to the desert area. This
made strangers vulnerable to wandering and possibly getting lost. So, pilgrims
needed assistance and used to travel accompanied by guides from Mecca or from
among the Bedouins of the area. As Islam spread outside Arabia, the need for ac-
cess to pilgrimage routes to Mecca increased. The importance of these routes was
in line with the changing political centres of Muslim rule and the centre of the Ca-
liphate from Medina to Damascus, Baghdad, Cairo, and Istanbul. The main pil-
grimage routes to Mecca that developed over time were:7
1. the Syrian pilgrimage route (Rakb al-Ḥajj al-Shāmī), which later included the
Ottoman pilgrims;8
2. the Iraqi route (Rakb al-Ḥajj al-‘Irāqī), including Persians and other Asian pil-
grims;
3. the Egyptian route (Rakb al-Ḥajj al-Maṣrī), including African, Maghrebi, and
Andalusian pilgrims;
4. the Yemenite pilgrims (Rakb al-Ḥajj al-Yamanī).
The Tabūk route (to the north of Arabia) is known as the oldest road in the ear-
ly Islamic period used by Muslim pilgrims. It connects Damascus to Medina and
passes a number of camps and stations in Syria through Jordan via Tabūk to Hijaz.
Caliphs and Muslim rulers in various Islamic periods were interested in conduct-
ing many construction projects along the routes, including the creation of pools,
canals, forts, mosques, bridges, markets, and other rest stations. After Tabūk, a
number of routes dispersed and headed to Medina, such as the Taymā’ and Khay-
bar roads linking the Levant to the holy cities of Mecca and Medina. 9
Continuing to take care of pilgrims’ affairs after the early Islamic period, caliphs
and sultans later on adopted the duty to protect and preserve the Two Holy
Mosques in the Hijaz, and were known by the title “The Protector of the Two Holy
Mosques” (ḥāmī al-ḥaramayn al-sharīfayn). These duties focused on several im-
portant aspects in order to ensure peace and security for the pilgrims to Mecca,
back and forth all the way:
1. Road protection and construction of ponds and wells in the ḥajj routes
2. Organizing and securing the pilgrimage caravans and removing obstacles
and risks by all means needed
6 See the electronic dictionary: https://www.almaany.com/ar/dict/ar-en

7 On the main routes to Mecca, see AL-ʿUMARĪ 2010, 2: 338–48.
8 On the Syrian Ḥajj, see MILWRIGHT 2013, 28–35.
9 See in detail about the Syrian route to Mecca with its stations during the Mamluk era: AL-
ʿUMARĪ, 2010, 2: 343-5; Many inscriptions and memorial Islamic writings still describe the
pilgrimage routes in Jordan to the Hijaz, see TUTUNCU – PETERSEN 2012, 155–63.
3. Providing means of subsistence for pilgrimage as well as necessary services

and food
4. Carrying out necessary repairs in the two Holy Mosques (al-Ḥaramayn)
5. Responsibility for the annual supplies sent with the convoy of al-Ḥajj al-
Shāmī to the poor people in Mecca and needy pilgrims (al-mujāwirūn), who
spent their time in the two Holy Mosques in Mecca and Medina
6. Sending the maḥmal annually with the kiswa (cloth covering of the Kaʿba) and
al-ṣurra (a package of money)10
7. Appointing the mission leader of the pilgrimage (amīr al-ḥajj)
In the Late Middle Ages and the Ottoman era, the ḥajj caravan was accompa-
nied by a large staff and guards, headed by amīr al-ḥajj, who represented the caliph
or the sultan in implementing his duties. He was the foremost official and the
commander-in-chief of the pilgrims’ mission, with several tasks to fulfil regarding
care for the mission:11
 Administrative, security, scientific, medical, catering, and guidance duties
 Several assistants and deputies for managing the caravan’s affairs, such as a
judge, court witnesses, organizer of the camel loaders, preacher to the pil-
grims, and others
The pilgrimage was a manifestation of the exercise of state authority, particular-
ly with regard to the protection and security of the pilgrims in the face of thirst,
hunger, and the dangers of Bedouin attacks throughout the journey. After the early
Islamic expansion in Syria, the Bedouin followed, and they even preferred nomad-
ic life on the borders of independent lands. They were not interested in living in
cities or in villages or moving to a life of stability.
From the early Islamic era to the Umayyad period (660–750), and the early Ab-
basid period, ensuring security was a reason for the multiplication of paths to Mec-
ca. As mentioned, the main route from Syria to Mecca was the Tabūk route, but
there were other ways, including the coastal route from Gaza to Ayla (ʿAqaba) and
Taymā’. Travellers picked the route of their choice depending on the season and
the arrival point. Pilgrimage routes had various days of glory depending on which
state was dominant. From their centre in Damascus, the Umayyads’ policy regard-
ing the Bedouins was formed by the Caliph Muʿāwiya b. Abī Sufyān (r. 660–80). In
times of good relations with the central authorities in Damascus, the nomad tribes
used to fight and act in the border areas, and their chiefs received the tax revenue
in the areas for which they were responsible. Muʿāwiya cared for important as-
10 The maḥmal is the camel that was sent during the ḥajj season by the sultan carrying the
kiswa (covering) for the Kaʿba and other gifts. On the maḥmal, see PORTER 2013, 195–205;
VROJLIK 2013, 206–13; and regarding the kiswa, see NASSAR 2013, 175–83.
11 On the role of amīr al-ḥajj and his accompanying staff, see ʿANKAWI 1974, 151–66.
pects of the pilgrimage including the kiswa, with which the caliph was concerned in
order to personalize a political act.12
In contrast, during the Abbasid Caliphate in Baghdad (750–1258), payments of
taxes stopped, and the Bedouin gradually increased their attacks on the adjacent
areas (cities and agricultural areas), especially in Syria due to its peripheral loca-
tion. The Iraqi pilgrimage, then, was the site where the pilgrims of the East (Iraq,
Persia, and Central Asia) and the Levant met, due to its safety and the availability
of services such as wells (ponds) and rest houses, at least until the collapse of the
Abbasid Caliphate in Baghdad. During the Fatimid period in Cairo (969–1171),
followed by the Seljuk control over Syria, and then the effect of the Crusades in the
area (1099–1291), the Syrian ḥajj routes were interrupted, allowing the Bedouin to
invade the adjacent areas, exploiting the political conflicts in Syria. 13 As a result of
those conflicts, the holy Black Stone (al-ḥajar al-aswad) in the Kaʿba was moved to
the Persian Gulf by the Qarāmiṭa (a Shiʿi sect) for 20 years after they raided Mecca
in 929 CE, looting and killing many pilgrims.14
From the time of the Fatimid rule in Cairo, and later under the Ayyubids (1171–
1250) and the Mamluks (until 1517), the Egyptian pilgrimage was the most im-
portant one, and it included the Maghrebi pilgrims, the Africans (Takrūr), and the
Andalusians.15 During the 10th to 12th centuries, the endless wars severely damaged
the economic situation in Syria. Trade relations with the outside world were cut
off, roads were unsafe for pilgrims, the livelihoods of farmers, artisans, and traders
were disrupted, even peasants abandoned their land. Therefore, as a result of the
weakness of political authorities in Syria, the Bedouin tribes of Banū Hilāl and
Banū Salīm entered Palestine from Arabia and moved towards Sinai, Egypt, and
North Africa (Tunisia). In the other direction, the Yemeni pilgrimage had the
strength and importance under the Banū Rasūl dynasty to provide the kiswa and
other supplies to the people of Mecca in the 12th century.
After the Ayyubids expelled the Crusaders from Jerusalem and the area, the fu-
ture Ayyubid sultan al-Muʿaẓẓam ʿĪsā (r. 1218–27) renewed the Tabūk route again
for the pilgrimage of Damascus in 1214, and it reached its golden age in the Mam-
luk period.16 Following the conquest of Syria, Egypt, and the Hijaz by the Ottoman
Empire (1517), glory returned to the Syrian pilgrimage route because it was fol-
12 See AL-AZRAQĪ 1983, 1: 254.

13 As a result of the Crusader conquest of the region and the building of their castles in Pal-
estine and Jordan, the Syrian route to Mecca was interrupted and the Muslim pilgrims were
forced to change their ḥajj routes. See: PETERSEN 2013, 21-27. On Muslim-Crusader conflicts,
see: AMITAI 2016, 324-45; HOLT 2005, 170–9. See also AL-KASASBA 2018, 2.
14 On this attack of the Qarāmiṭa and its outcome, see IBN KATHĪR 1991, 11: 160–2.
15 See ʿANKAWI 1974, 146–8.
16 On the favourable conditions for pilgrim caravans to Mecca from Cairo and Damascus
under the Mamluks, see ʿANKAWI 1974, 146–51.

lowed by Ottoman pilgrims. Thus, the Egyptian and al-Shāmī pilgrimage caravans
seemed like “small towns on the move”.17
Motives of nomads’ raids on pilgrimage routes
A question should be asked about the Bedouins’ motives for attacking pilgrim
caravans along the roads to Mecca despite the religious goals of this travel. It is
known that the harsh living conditions of the nomads living in the deserts and
their severe need for money and food led them to attack pilgrim caravans and
traders passing by, but sometimes these raids occurred for no reason. A review of
relevant sources of the period leads to the conclusion that there were various caus-
es of these Bedouin attacks. Ignorance and tribal fanaticism (ʿaṣabiyya) among the
Bedouins motivated them to loot the pilgrims for food and clothing without any
religious motives for their deeds. Money, goods, and property held by pilgrims
and merchants passing through Bedouin areas back and forth to Hijaz stimulated
the Bedouins’ ambitions to attack their caravans as part of their custom of raiding
(ghazw). The major causes of such raiding can be summarized as follows:
1. Provocation or conflicts between the princes of Mecca, or between them and
the pilgrims, led some tribesmen to seize the opportunity to raid the pilgrims’
armed guards and loot them. In cases of lack of wisdom by the pilgrimage
commander (amīr al-ḥajj) in dealing with the Bedouins, this led to conflicts with
them—for instance, when he failed to perform his job properly by being negli-
gent or unjust toward the Bedouins.18
2. Failure to deliver the funds agreed upon between the central authority and the
Bedouins in exchange for guarding and protection of roads and maintaining
water sources on the pilgrimage routes.
3. Conflicts and instability in the central authority and internal disagreements
between the central authority or state governors led to a loss of security, i.e. to
attacks on pilgrimage routes. For example, the nomads in Egypt and Syria be-
gan rebellions and attacked their surrounding areas, exploiting the political
vacuum during the transition from Ayyubid to Mamluk rule (1250–60).
4. Severe cruelty used by the authorities to punish the Bedouins who sometimes
attacked pilgrims, as well as military campaigns against them that led to cap-
turing of women and children, seizing livestock, looting homes, and destroy-
ing properties also deepened the Bedouins’ eagerness for revenge, as hap-
pened in the Damascus region in 1501.19
17 See DAUPHIN – BEN JEDDOU – CASTEX 2015.

18 On the effect of conflicts among the emirs of Mecca and Medina, see AL-MAQRĪZĪ 1997, 1:
481, 486, 487, 491, 2: 44, 59–60; IBN ʿABD AL-ẒĀHIR 1976, 359–60.
19 See IBN ṬŪLŪN 1998, 198.
Mamluk policy towards Bedouins

As already mentioned, in 1250–60 the Bedouins exploited the political vacuum and
began rebelling and attacking surrounding areas in Egypt and Syria, claiming
Mamluk authority to be illegal, and that they (the Bedouin) have the right to rule,
considering themselves the “owners of the country” (aṣḥāb al-bilād).20 Sultan Bay-
bars acted severely against them, taking special care to secure the pilgrimage
routes to the holy places in Mecca and Medina. The main aims of Baybars’s acts
against Bedouins were:
 to restore security along the roads;
 to ensure the nomads’ loyalty and their obedience;
 to preserve the borders in Egypt, Syria, and the Hijaz areas by committing the
Bedouins to keeping and controlling their territories;
 to obligate the Bedouins to pay duties and taxes (zakāt and ʿushūr) like his oth-
er subjects.
From the data in the tables 1 and 2, it can be concluded that most of the Bedouin
raids occurred in times of interruptions and conflicts among the political rulers or
weakness of authority. That can be understood from al-Maqrīzī’s description of the
Bedouins’ situation after Sultan Baybars had subdued them during his reign (1260–
77) by saying that the Bedouins had been dispersed and their revolt was sup-
pressed, and they were treated with violence and oppression until they were re-
duced and humiliated.21
As a result of the revival of the Abbasid Caliphate in Cairo in 1261 by Sultan
Baybars, the Mamluk sultans assumed a special status of “the Servitor of the Two
Holy Sanctuaries” (khādim al-ḥaramayn al-ṣharīfayn). So, the Mamluk duty of main-
taining the pilgrimage became a political as well as a religious obligation. The rul-
ers became responsible for all matters related to the safety of the pilgrimage along
the roads and in the holy places, including sending the annual kiswa of the Kaʿba
and the departure of the richly decorated maḥmal that led the annual pilgrimage
caravan from Cairo. The emir of the ḥajj was appointed by the Mamluk sultan to
manage pilgrimage matters until returning back home. He was to be a man of
competence, strength, courage, and diplomacy in order to protect the pilgrims,
which made him frighten the nomads on the way. This responsibility of the Mam-
luk sultans imposed upon them the duty of maintaining the pilgrimage in a proper
and safe manner. To this end, the sultans took various measures against violators
20 On these events, see IBN ʿABD AL-ẒĀHIR 1976, 51, 120; AL-MAQRĪZĪ 1997, 1: 479–81; AL-
NUWAYRĪ 2004, 30: 33–4, 68–9; AL-DHAHABĪ 1999, 49: 23.
21 See AL-MAQRĪZĪ 1997, 1: 479–81: …ḥattā tamma tabdīd shaml al-badw wa-ikhmād thawratihim,
wa-muʿāmalatahum bi-l-ʿunf wa-l-qahr ḥattā qallū wa-dhallū.

of the order of the holy places in Mecca and Medina, and against the Bedouins
along the ḥajj routes.22
Table 1: Baybars’s acts against Bedouins during his reign
Year Events and results of Baybars’s acts
 Sultan Baybars provided the Bedouins with their livelihood.
 He handed over to them the guarding of the country.
1261  He obligated them to secure the borders with Iraq.
 A proclamation to Sharaf al-Dīn ʿĪsā ibn Muhannā as emir of all Bedouins in
Syrian areas (mashyakha).23
 Sultan Baybars obligated the Bedouins to secure the country and safeguard it
1263
until the Hijaz.
 Baybars sent troops (khafar) to Khaybar to occupy it to guard the area and the
1264
ways to Hijaz.24
1265  Twenty-one Bedouin leaders (sheikhs) were executed.
 Baybars succeeded in collecting both the zakāt andʿushūr taxes. As a result,
many of the Bedouins, such as Banū Ṣakhr, Banū Lām, Banū ʿAnza, and other
1268
tribes of the Hijaz submitted to him and committed to pay their zakāt dues of
sheep and camels.
Baybars had previously punished the Bedouins by taking their children
hostage. When the Bedouin leaders apologized to him, he freed their chil-
1272
dren and reintroduced the previously decided-upon aid in money and live-
stock from neighbouring villages and cities.25
 The Bedouins of Banū ʿUqba were obliged to provide horses and camels to
1276 Sultan Baybars.26 This apparently became the usual relationship between the
Mamluk state and the nomads (ʿurbān).
 Sultan Baybars succeeded in subduing the nomad tribes after implementing
By the end strict policies against their actions. The nomads were obliged to pay their
of Baybars’ duties of zakāt from all regions under his influence, even from the Maghreb
reign – 1277 and the Hijaz. Then, the Sultan used to send a special mamluk to the Hijaz
area to collect the zakāt.27
22 B. Walker, for example, discussed the changing relations between Mamluk authorities and
the Bedouin tribes in Transjordan and its importance for several tasks, including defence
against foreign invasion and preserving security on the main ḥajj route from Damascus to
Mecca. See WALKER 2009, 83–105.
23 AL-NUWAYRĪ 2004, 30: 26; AL-MAQRĪZĪ 1997, 1, 538.
24 See on this policy of Baybars: IBN ʿABD AL-ẒĀHIR 1976, 165, 220; AL-MAQRĪZĪ 1997, 1, 557;
AL-NUWAYRĪ 2004, 30: 152.

25 On this event see: AL-NUWAYRĪ 2004, 30, 119–20.
26 AL-NUWAYRĪ 2004, 147.
27 See AL-NUWAYRĪ 2004, 96.
Table 2: Main Bedouin raids in the Mamluk period by years:28

Bedouin raids Bedouin raids
1st Mamluk Era (Turk) 2nd Mamluk Era (Circassian)
659–76 AH (1260–77 CE) (Sultan 804–10 AH (1401–6 CE) 899–900 AH (1493–4
Baybars’ fighting against Bedouins.) 814–29 AH (1411–26 CE) CE)
683–7 AH (1284–8 CE) 841–2 AH (1437–8 CE) 903–4 AH (1497–8 CE)
698–705 AH (1299–1305 CE) 843–66 AH (1439–61 CE) 906–7 AH (1500–1 CE)
714–51 AH (1315–51 CE) 872–98 AH (1468–92 CE) 921 AH (1515 CE)
Sending money (al-ṣurra) to the Bedouins living near the pilgrim routes in ex-
change for securing the area, in addition to arranging money, clothes, and means
of dressing for the Bedouin leaders, was a practice that continued uninterrupted
every year. Occasionally, the authorities seized known violators of the order
among the Bedouins and held them as a precaution until the ḥajj convoys moved
away from their areas and then released them. Sultan Baybars dealt with the Bed-
ouins in different ways, ranging from tolerance to harsh punishments. He was
wary of the treachery of the Bedouins and punished them when needed by taking
their children as hostages.29
Many times, Mamluk sultans or their regional governors arranged for armed
escorts guarding the pilgrim and trade caravans to repel attacks against them.
They used to send military aid to the pilgrims if they were attacked for two pur-
poses: to help the pilgrims or to punish the Bedouins. In 1304, during the reign of
Sultan Qalāwūn (r. 1293–4, 1299–1309, 1310–41), Emir Sallār, for example, pun-
ished some of the aggressors against the pilgrims by cutting off their hands and
feet, in addition to providing charitable aid, money, and assistance to pilgrims and
other people in Mecca and Medina.30 Another example is Sultan Khushqadam (r.
1461–7): in 1467, he acted to punish the attacking tribe of Banū ʿUqba in the ʿAqaba
area by a military campaign in which their leader Shaykh Mubārak and sixty other
men from his tribe were arrested and executed in Cairo.31
In the second Mamluk (Circassian) era, which was characterized by relative
weakness, the Bedouin raids on pilgrimage routes and their surrounding areas
increased.32 That is why the Mamluks worked to build rest stations and guard
28 For examples of those raids in the late Mamluk period, see IBN ṬŪLŪN 1998, 32, 69, 132,
267, 276; IBN ḤIJJĪ 2003, 777.
29 Al-ʿUmarī reports peaceful conditions on the way to Mecca, and good relations between
pilgrim missions and the Bedouins in the first Mamluk period. See AL-ʿUMARĪ 2010, 338–9.
30 See AL-MAQRĪZĪ 1997, 2, 378.
31 AL-MAQRĪZĪ gives many examples of such conflicts with the Bedouin in 1437–8, see, for
example AL-MAQRĪZĪ 1997, 7, 372–3.

32 See Table 2 above to compare the number of conflicts between the two Mamluk eras. In
his essay, Ibn Ṭulūn mentions how corruption spread in Damascus among the sultans and
local rulers in the late Mamluk era and how it affected the maintenance of al-Ḥajj al-Shāmī
and resulted in its neglect, see IBN ṬŪLŪN 2009. On the events in 900 AH/1494 CE, see IBN
posts between Damascus and the Hijaz, including forts and inns (khans), to provide
comfort and safety for pilgrims.33 The function of these stations was very im-
portant for safety and security as they provided shelter and nourishment to the
poor and needy. Sometimes, pilgrims deposited and stored their goods in the forts,
including some food for their return trip, or they provided dwelling for those
forced to leave the caravan due to illness or injury.
Forts or khans on pilgrimage routes served as housing for guards controlling
wells, pools, and other properties. They were also important because of the mar-
kets organized there between local people, pilgrims, and traders in general that
Bedouins took advantage of. A good example of such a market is the one that the
historian of the Mamluk era, Ibn Faḍl Allah al-ʿUmarī, described in Ayla (ʿAqaba)
on the royal road, at the meeting of the pilgrimage routes from Gaza and Egypt.
He says: “It is a destination intended for the traders of the Levant to come to, and
great, extended and branched markets are established, that are not found in the
main provinces and the big cities. Perhaps it does not lack anything in it: horses,
camels, sheep, flour, barley, fodder and types of food, stripes, bearings… weapon,
cloth, bedding, luggage, etc.”34
Relations between nomads and pilgrims under the Ottomans
After the Mamluks, the Ottoman sultans took upon themselves the special care for
the holy sites and the safety of the ḥajj in order to preserve their religious status in
the Islamic world. This duty was the principal reason for the successful protection
of the Hijaz from Damascus, which was maintained by effective road security.35
The pilgrimage caravan known as the Syrian Caravan included pilgrims from the
Levant and Anatolia and was usually accompanied by the emir of the ḥajj, the
maḥmal, and the Kaʿba cloth. Often, the ḥajj caravans would be combined with a
military expedition to enable the Ottomans to maintain an active presence in the
Holy Places.36
ṬULŪN 1998, 132. In the year 916 AH/1510 CE, the pilgrimage from Syria via Tabūk was
renewed after four years of interruption because of Bedouin raids and insecurity, see IBN
ṬULŪN 1998, 267, 276.
33 On early 16th-century forts on the Syrian ḥajj route, see PETERSEN 2013, 24–7. On establish-
ing khans in Egypt, see also ʿABD AL-MALIK 2013, 42–54.

34 See AL-ʿUMARĪ 2010, 341. On the role of the ʿAqaba khan (Ayala) as the meeting place be-
tween the Egyptian and the Syrian Ḥajj from Gaza, see ʿABD AL-MALIK 2013, 56–8.
35 On the early Ottoman forts on the ḥajj routes, see PETERSEN 1989, 97-118. Sultan Selīm I
ordered the construction of fortresses, and then other Ottoman sultans followed his policy
as much as they could, see PETERSEN 2008, 31–46.
36 In her study, S. Faroqhi focusses on the pilgrimage routes and securing the ḥajj missions
by the Ottomans, see FAROQHI 1994, especially chapters 2 and 3. On early Ottoman Arabia
and the ḥajj routes, see PETERSEN 2012, 9–19; on the Syrian Ḥajj see also PETERS 1994, 144–61.
During the golden age of the Ottoman Empire two major pilgrim caravans were
organized regularly: the Syrian (al-Shāmī), joined by the pilgrims from Istanbul and
Anatolia, and the Egyptian ḥajj caravan, joined by pilgrims from North Africa (the
Maghreb). The Ottoman sultan himself oversaw the exit of the pilgrims of Istanbul
and accompanied them from the Ḥaram area in Üsküdar. The convoy was loaded
with gifts of gold, antiques, and camels. He also sent with them money from en-
dowment funds and gifts to the Two Holy Mosques (al-ḥaramayn al-sharīfayn),
while for the people of Mecca and Medina he sent food, lamps, and carpets. Sultan
Selīm I (r. 1512–20) was the first Ottoman Sultan to obtain the submission of the
emirs (sharīfs) of Mecca and Medina to Ottoman power in 1517. He appointed as
amīr al-ḥajj from Egypt Muṣliḥ al-Dīn who was the first Ottoman official to carry
the kiswa for the Kaʿba. On January 22, 1518, the four judges and several Sufis par-
ticipated with the Emir Muṣliḥ al-Dīn to bid farewell to the ḥajj caravan in Damas-
cus (Maʿlā).37
Sultan Süleymān (r. 1520–66) was the first sultan to pay attention to the safety
of the ḥajj and took several steps, including building forts and sending garrisons to
guard the pilgrimage, that the Ottomans continued to implement in order to facili-
tate the pilgrimage routes. In 1520–1, Sultan Süleymān completed the renovation of
Mecca’s water supply system. In general, the Ottomans made efforts to minimize
risks to trade and pilgrimage by repairing and building khans, caravan stations,
and forts, in addition to the mobilization of many of their soldiers as guards, and
by making agreements with the Bedouin, which contributed further protection. In
his study, Andrew Petersen states that the ḥajj forts had a variety of functions that
served to increase Ottoman power in the region, in addition to their ideological
services in guarding the ḥajj routes.38
In times of security, good relations spread between Bedouins and the others on
the routes, such as pilgrims, traders, authorities, inhabitants of villages and neigh-
bouring cities. The Bedouin benefited from this calm atmosphere in several fields,
improving their living conditions and economic and commercial perspectives at
the same time. They performed "guide" and "service" functions for the pilgrims’
caravans, so they earned from services like renting camels and carrying pilgrims,
providing them with food and drinks. As mentioned above, pilgrimage camps and
stations turned into mobile markets with items for sale to Bedouins, locals, and
pilgrims. Meanwhile, European merchants also benefitted from the hospitality and
good treatment by the Mamluks and then the Ottomans that encouraged them to
trade in the area, and they established their ports and commercial centres wherever
37AL-JAZĪRĪ 2002, 496–7.

38See PETERSEN 2008, 31–50. On the special care and provision of protection for the pilgrims
between Damascus and the Hijaz in the early Ottoman period, see AL-KASASBA 2018, 5–7, 11-
13; BAKHIT 1982, 223; AL-GHAZZĪ 1979, 3, 157.
Mamluks and Ottomans had erected khans on the roads and spread safety and
protection along the pilgrimage routes.
Under weaker Ottoman governments, after Sultan Süleymān, governors in-
creasingly neglected their duties so that many of these achievements were corrupt-
ed. Forts were ruined and the soldiers lacked commanders. As a result, the gover-
nors took advantage of the weakness of the central authority and began to seize the
funds (al-ṣurra) allocated to the Bedouins, which led the latter to react by attacking
the pilgrim caravans and looting them. Since the early 18 th century, it became a
practice for the governor (nā’ib) of al-Shām to assume the position of amīr al-ḥajj as
well. Thus, Damascus became the important centre for the ḥajj in the Ottoman Em-
pire, gathering pilgrims from the neighbouring regions and even from Iraq and
Anatolia.39
The influence of Ottoman janissaries on the post of amīr al-ḥajj made matters
worse. When the ḥajj leadership was taken over by one of those janissaries, the
emir of the ḥajj started to manipulate the money of al-ṣurra and to prevent it from
being spent for the benefit and appeasement of the Bedouins. Those changes led to
instability among the Bedouins, and then pushed them to attack pilgrim caravans.
In the late Ottoman period, governors were unable to prevent Bedouin attacks on
populated areas adjacent to the desert in the Levant in general. The Bedouin loot-
ing of pilgrimage routes spread, and adversely affected the surrounding popula-
tion, causing sometimes disruption of trade and impediment for pilgrim cara-
vans.40 Most Bedouin attacks on the al-Ḥajj al-Shāmī in the Ottoman period were
carried out in the 18th century, while the 16th and 17th centuries witnessed only a
few raids, which indicates the strength and prestige the state had enjoyed in the
first two centuries after the Ottoman conquest of Syria and Egypt in 1516–7.41 The
worst and most miserable conditions of pilgrims and their guards are described by
al-Bidīrī during their return to Damascus in 1757. He mentions that “al-Ḥajj al-
Shāmī had been stolen by ʿArabs (Bedouins) and looted, and the Bedouins had
taken away women and men with their money and their belongings”. Al-Bidīrī
added that “women and girls were barefoot and without clothing and horrors that
happened to the pilgrims caused by the evil of the Bedouin unbelievers were acts
that even fire worshipers do not do, things that were never heard of from ancient
times, and not even from idolaters”.42
As a result, the next year (1758), the leader of the ḥajj mission, ʿAbd al-Raḥmān
Pasha, the governor of Tripoli, and his janissary troops acted with cruelty by kill-
ing many Bedouins in the Hijaz together with their leader (sheikh). 43 During the
39 AL-KASASBA 2018, 10–11.

40 PETERSEN 2008, 41–4; FAROQHI 1994, chapter 3.
41 See FAROQHI 1994, chapter 3.
42 See AL-BIDĪRĪ 1959, 206–8.
43 AL-BIDĪRĪ, 221.
rule of the Ottoman governor and de facto ruler of Egypt Muḥammad ʿAlī Pasha
(r. 1805–48), he dealt cruelly with the Bedouins in Egypt, Hijaz, and Syria: He tried
to restrain them and change the course of their nomadic lives, their aggressive
habits, and their traditions.44
Conclusion
The nomads’ (Bedouin) relationship with others, especially with the pilgrim cara-
vans, was characterized by changeable situations according to the status of the
central authority. Sometimes it was distinguished by maintaining peace and securi-
ty, while on other occasions it involved violent attacks and looting. Dealings with
the Bedouins were related to maintaining the security and safety of the pilgrims
and their caravans. The changes in relationship with the Bedouins were an im-
portant aspect of the reality on the pilgrimage and trade routes, especially between
the Levant and the Hijaz. In times of strict rulers who implemented organized ad-
ministrative systems, safety and security spread on pilgrim routes. As a result,
caravans flourished in the path of pilgrimage, for the benefit of both pilgrims and
traders. So, the authorities established khans, forts, and rest stations for them. This
situation supported the development of seasonal markets on pilgrimage routes,
and busy permanent markets and khans in major cities, which benefitted the local
people in towns and villages as well as the Bedouins, merchants, and pilgrims.
But in times of weakness of the central authority or in states of confusion, the al-
Shāmī pilgrimage was interrupted and joined the Egyptian pilgrimage through the
Gaza route to ʿAqaba, because insecurity, wars or attacks on pilgrims, strife or
disasters prevented them from travelling directly from Damascus to Mecca. Power
struggles between princes of Mecca and Medina were usually accompanied by
inter-tribal conflicts over areas of influence. These events affected pilgrims’ travels,
causing the looting of funds and then commercial stagnation. The Bedouins react-
ed with attacks, taking captives and killing travellers in order to obtain financial
and material resources. On the other hand, the authorities acted to punish the at-
tacking Bedouins by taking their children and leaders hostage as a means of sub-
duing them and recovering stolen money and goods. As a compromise, signing
agreements between Bedouins and the authorities led to the appointment of
sheikhs, obedience, and the protection of roads in return for money.
Bibliography
ʿABD AL-MALIK, S. (2013): The Khans of the Egyptian Hajj Route in the Mamluk
and Ottoman Periods, in Porter, V. – Saif, L. (ed.:) The Hajj: Collected Essays, Re-
search Publications 193. London, 52–64.
AL-ALBĀNĪ, M. (2002): Ṣaḥīḥ Sunan Abī Dāwūd (8), Kuwait.
AL-AZRAQĪ, M. (1983): Akhbār Makka (1), Beirut.
44 On Muḥammad ʿAlī Pasha’s treatment of the Bedouins in Syria, see SĀLIM 1990, 216–23.
AL-BIDĪRĪ, A. AL-Ḥ. (1959): Ḥawādith Dimashq al-Yawmiyya, Cairo.

AL-DHAHABĪ, SH. M. A. (1999): Tārīkh al-Islām (49), Beirut.
AL-GHAZZĪ, N. (1979): Al-Kawākib al-Sā’ira (3), Beirut.
AL-JAZĪRĪ, ʿA. M. (2002): Al-Durar al-Farā’id al-Munaẓẓama fī Akhbār al-Ḥajj wa-Ṭarīq
Makka al- Muʿaẓẓama (1), Beirut.
AL-KASASBA, H. F. (2018): The Role of the Ottoman State in Protecting Hajj Cara-
vans. Route Educational and Social Science Journal, vol. 5/8, 1–17.
AL-MAQRĪZĪ, A. b. ʿA. (1997): Al-Sulūk li-Maʿrifat al-Duwal wal-Mulūk, Beirut.
AL-NUWAYRĪ, SH. A. (2004): Nihāyat al-Arab fī Funūn al-Adab (30), Beirut.
AL-SIJISTĀNĪ, ABŪ D. (2009): Sunan Abī Dāwūd (3), Damascus.
AL-ʿUMARĪ, IBN F. ALLAH (2010): Masālik al-Abṣār fī Mamālik al-Amṣār (2), Beirut.
AMITAI, R. (2016): The Early Mamluks and the End of the Crusader Presence in
Syria (1250–1291), in Adrian, J. B. (ed:) The Crusader World. London–New-York,
324–345.
ʿANKAWI, A. A. (1974): The Pilgrimage to Mecca in Mamlūk Times, in Serjeant, R.B.
– Bidwell, B.L. (ed.:) Arabian studies, vol. 1. London, 146–170.
BAKHIT, M. (1982): The Ottoman Province of Damascus in the Sixteenth Century, Beirut.
BURAK, G. (2017): Between Istanbul and Gujarat: Descriptions of Mecca in the Six-
teenth-Century Indian Ocean. Muqarnas 34, 287–320.
COLEMAN, S. – ELSNER, J., ed. (1995): Pilgrimage. Past and Present in the World Reli-
gions, Cambridge, Mass.
DAUPHIN, C. – BEN JEDDOU, M. – CASTEX, J-M. (2015): To Mecca on Pilgrimage on
Foot and Camel-back: The Jordanian, Darb al-Hajj. Bulletin of the Council for Brit-
ish Research in the Levant 10/1, 23–36, DOI: 10.1179/1752726015Z.00000000029
EPHRAT, D. – KABHA, M. D. (2003): Muslim Reactions to the Frankish Presence in
Bilād al-Shām: Intensifying Religious Fidelity within the Masses. Al-Masaq 15
(March), 47–58.
FAROQHI, S. (1994): Pilgrims and Sultans: The Hajj Under the Ottomans, New York.
HOLT, P. M. (2005): The Treaties of the Early Mamluk Sultans with the Frankish
States, in Gerald, H. (ed.:) Muslims, Mongols and Crusaders. London, 170–179.
IBN ʿABD AL-ẒĀHIR, M. (1976): Al-Rawḍ al-Zāhir fī Sīrat al-Malik al-Ẓāhir, Riyadh.
IBN KATHĪR, I. (1991): Al-Bidāya wal-Nihāya (11), Beirut.
IBN ḤIJJĪ, SH. (2003): Tārīkh Ibn Ḥijjī, Beirut.
IBN ṬŪLŪN, M. B. ʿA. (1998): Mufākahat al-Khillān, Beirut.
IBN ṬŪLŪN, M. B. ʿA. (2009): Al-Barq al-Sāmī fī Tiʿdād Manāzil al-Ḥajj al-Shāmī, Da-
mascus.
LEWIS, N. (2009): Nomads and settlers in Syria and Jordan, 1800–1980, London.
LUZ, N. (2020): Pilgrimage and religious tourism in Islam. Annals of Tourism Re-
search 82 (May), 1–9, DOI: 10.1016/j.annals.2020.102915
MILWRIGHT, M. (2013): Trade and the Syrian Hajj between the 12 th and the Early
20th Centuries: Historical and Archaeological Perspectives, in Porter, V. – Saif,
L. (ed:) The Hajj: Collected Essays, Research Publications 193. London, 28–35.
MUSLU, C. Y. (2014): The Ottomans and the Mamluks: Imperial Diplomacy and Warfare
in the Islamic World, London.
NASSAR, N. (2013): Dar al-Kiswa al-Sharifa: Administration and Production, in
Porter, V. – Saif, L. (ed.:) The Hajj: Collected Essays, Research Publications 193.
London, 175–183.
PETERS, F. E. (1994): The Hajj: the Muslim Pilgrimage to Mecca and the Holy Places,
Princeton.
PETERSEN, A. (1989): Early Ottoman Forts on the Darb al-Hajj. Levant 21, 97–118.
PETERSEN, A. (2008): The Ottoman Hajj Route in Jordan: Motivation and Ideology.
Bulletin d’études orientales 57, 31–50.
PETERSEN, A. (2012): The Medieval and Ottoman Hajj Route in Jordan; an Archaeological
and Historical Study, London.
PETERSEN, A. (2013): The Lost Fort of Mafraq and the Syrian Hajj Route in the 16th
Century, in Porter, V. – Saif, L. (ed:) The Hajj: Collected Essays, Research Publica-
tions 193. London, 21–27
PORTER, V. – SAIF, L., ed. (2013): The Hajj: Collected Essays, Research Publications 193,
London.
PORTER, V. (2013): The Mahmal Revisited, in PORTER, V. – SAIF, L. (ed:) The Hajj:
Collected Essays, Research Publications 193. London, 195–205.
RAHIMI, B. – ESHAGI, P., ed. (2019): Muslim Pilgrimage in the Modern World (Muslim
Civilization and Muslim Networks), Chapel Hill.
SĀLIM, L. M. (1990): Al-Ḥukm al-Maṣrī fī al-Shām 1831–1841, Cairo.
TÜTÜNCÜ, M. – PETERSEN, A. (2012): Arabic and Turkish Inscriptions on the Darb
al-Hajj al-Shami, in Petersen, A. The Medieval and Ottoman Hajj Route in Jordan;
an Archaeological and Historical Study. London, 155–163.
VAN STEENBERGEN, J. (2016): Caliphate and Kingship in a Fifteenth-Century Literary
History of Muslim Leadership and Pilgrimage, Leiden.
VROJLIK, A. (2013): An Early Photograph of the Egyptian Maḥmal in Mecca, in Por-
ter, V. – Saif, L. (ed.:) The Hajj: Collected Essays, Research Publications 193. Lon-
don, 206–213.
WALKER, B. J. (2009): The Tribal Dimension in Mamluk-Jordanian Relations. Mam-
luk Studies Review 13, 83–105.
WEBB, P. (2013): The Hajj before Muhammad: Journeys to Mecca in Muslim Narra-
tives of Pre-Islamic History, in Porter, V. – Saif, L. (ed.:) The Hajj: Collected Es-
says. London, 6–14.
Vlachs and Yörüks as Military Groups
under Ottoman Rule and the Evolution
of Their Roles
HARUN YENI
Abstract
This paper discusses the changing role of the semi-nomadic groups within the
Ottoman military structure. First, the position of the yörüks—the Turkish semi-
nomadic group—will be underlined during the foundation of the Ottoman State. It
is argued that they were a significant source for the Ottoman army both as infantry
and cavalry. Then, through the imperial building process which peaked during the
reign of Mehmed II, they were less needed and turned into auxiliary forces. Simi-
larly, the Vlachs were another group of semi-nomadic nature that was utilized by
the Ottomans from their first encounter onwards. Being a demographic pool for
the military class of the Balkans before the Ottoman arrival, they were granted
their previous status by the Ottomans as well. Yet, in time, in accordance with their
needs, the Ottomans turned them into auxiliary forces, too.
Keywords: Yörüks, Vlachs, nomads, Ottomans
Even though the foundation and expansion of the Ottoman state in its early years
have attracted historians for many years, some of its structural components such as
the semi-nomadic groups have been neglected as separate areas of research.1 De-
spite the remarkable existing efforts by some scholars devoted to the study of this
subject,2 the discussions on the foundation process evolve around similar points
 PhD, Lecturer at TOBB University of Economics and Technology, Ankara, Turkey. This
article is a written version of the paper entitled “Nomads and The Ottomans: Evolution of
Serving the Ottoman Military” presented at the 8th International Conference on Medieval
History of the Eurasian Steppe Nomads and their Neighbours in the Medieval Ages, which took
place on November 20–3, 2019, in Sofia, Bulgaria. I would like to thank Delyan Rusev for his
valuable comments and editing.
1 R. P. Lindner’s monograph is an exception, see LINDNER 1983.
2 For a close study of the basic arguments on the question of the foundation of the Ottoman
state, see KAFADAR 1995.

318 HARUN YENI
and yet they seem to be far from opening new avenues for research. 3 Within this
context, this paper analyses the roles of the semi-nomadic groups in the early Ot-
toman period. It first argues that the yörüks played a significant part in the early
phases of the Ottoman expansion both demographically and militarily. In time,
with the transformation of the Ottoman state, the importance of the yörüks dimin-
ished. Another group of semi-nomads, the Vlachs, with whom the Ottomans met
during their advance in the Balkans, was employed for the imperial expansion as
well. Their connection with other already existing military groups such as voynuks
and martoloses increased their value. The changes in the Ottoman state structure
resulted initially in them becoming auxiliary forces, and later in the abolishment of
their organization.
We know almost for certain that the early Ottoman society was predominantly
of semi-nomadic and pastoralist nature. 4 Therefore, the constructive sections of the
state apparatus, i.e. the military, would inevitably reflect such a picture as well.
The changes in the state structure and the expansion in the Balkans resulted in a
departure from the dominance of its semi-nomadic culture. On the other hand, the
pragmatism of a state on the verge of evolving into an empire made it possible to
make use of the semi-nomads who recently came under Ottoman rule. Employing
already existing institutions was an established practice for the Ottomans during
the expansion period. The evolution from a frontier beylik to an imperial state was
mainly realized through the institutional changes, especially in the military, intro-
duced between the reigns of Orhan (r. 1324–62) and Mehmed II (r. 1444–46, 1451–
81). Thus, the role of the semi-nomadic groups in the Ottoman army should be
traced accordingly. As Irène Mélikoff states, “the early Ottomans were apparently
closer to the nomadic Turcomans both in terms of their lifestyle and religious prac-
tices, at first Osman and Orhan being simple chiefs of such a tribe.” 5 Within the
same scope, the fact that the first ruler of the Ottomans, Osman Gazi (d. 1324), was
a son-in-law of a follower of Baba İlyas, a religious leader of the Turcomans with a
heterodox Islamic belief, strengthens such a picture. According to a 15 th-century
chronicle, Orhan, when contemplating the establishment of a military detachment,
consults his brother Ali, who plans to retire by becoming a dervish. Ali suggests to
him that Orhan and his new soldiers wear white headgears, unlike the soldiers of
other beys wearing red headgears, which itself is another indication of the domi-
nant Turcoman presence.6
3 For a recent criticism of the literature on the process of imperial foundation, see DEMIRHAN
2019.
4 İNALCIK 1993, 25–6; İNALCIK 1980, 74–5, İNALCIK 1986, 41–2.
5 MELIKOFF 1993, 154.
6 ORUÇ 2014, 18–19; ANONIM TEVÂRIH-I ÂL-I OSMAN 1992, 16; AŞIKPAŞAZADE 2013, 55. Aşık-
paşazade does not mention a permission from Hacı Bektaş.

VLACHS AND YÖRÜKS AS MILITARY GROUPS UNDER OTTOMAN RULE… 319
In this respect, it would be useful to give an outline of the Ottoman military

structure and its evolution until the end of the 15 th century. As stated above, the
social structure of the early Ottomans was dominantly of semi-nomadic nature
beside the fact that the Seljukid state tradition had a formative impact in the ad-
ministrative background of the Ottoman principality. The military nature of the
Ottomans, therefore, was shaped in parallel with their social structure. The semi-
nomadic groups were both subjects and soldiers of this body, which was domi-
nantly of tribal nature. It was “very difficult to distinguish between society and
military among the Turkomans living in the Bithynian marches.” 7 The Bithynian
frontier principality thus became a centre attracting wandering groups in Anatolia
with the prospects it promised. It should be added that the combatant semi-
nomads were led by a bey and his retinue, consisting of alps—or alp erens—and
their comrades (yoldaş, nöker), who were bound by a sort of chivalric codes, and
these warriors were rooted in the Eurasian steppes. 8 Aşık Paşa described the alp of
his time in his Garibnāme (1330) as “a soldier having a good horse, carrying a keen
edged sword (kılıç), bow (yay), arrows (ok) and lance (süngü), and bearing an im-
penetrable and awe-inspiring coat of mail (ton).”9 Thus, it can be said that during
the period of frontier march while the beylik was a part of the Seljukid Empire un-
der the Ilkhanid rule, the Ottoman forces mainly consisted of mounted archers,
who were dominantly of semi-nomadic background and were under the command
of alps.
During the process of becoming an independent principality, the Ottomans’
popularity gained momentum among the Turcoman warriors, and in the same
period Osman Beg became a more dominant figure among his fellow alps.10 Then,
in the period of expansion into Rumelia, these fellow alps turned into leading fron-
tier lords of the Ottoman principality. The mounted Turcoman warriors were set-
ting raids against the land of the so-called infidels, thus paving the way for con-
quests. These groups later constituted the akıncı (raider) forces that seem to have
gained an organized structure in the 15th century.11 The reign of Orhan was the
period of institutionalization of the military in parallel with the new territorial
gains and these gains increased the need for a regular army. Yaya müsellem (infan-
try and cavalry) corps were recruited from the peasantry and semi-nomadic
groups in return for a farm to cultivate and/or for some tax exemptions. The in-
come from a certain amount of land was granted to the timarlı sipahis, who were
professional mounted warriors, whose background varied from previous lords of
the newly acquired lands to Turcomans. Land tenure in return for military service
7 FODOR 2009, 192.

8 See PRITSAK 2012 and GÜNAL 2007.
9 FODOR 2009, 194.
10 For a recent and comprehensive account of his career, see ALTUĞ 2020.
11 KIPROVSKA 2004, 11–28.
320 HARUN YENI
was not a novelty in the medieval Mediterranean, and the Ottomans utilized it
from the early years onwards. Azeb soldiers were another part of the infantry, and
they were in time replaced by the janissary corps, which were first introduced dur-
ing the time of Orhan. Yaya-müsellem corps together with azebs would be turned
into auxiliary forces later. Until the time of Mehmed II, there was no structural
change. The reigns of Murad I (r. 1362–89) and Bayezid I (r. 1389–1402) are known
for the Ottomans’ efforts to consolidate their power over their tributary states in
the Balkans and in Anatolia. Bayezid I’s aggressive attempt to dominate the pe-
riphery of the Ottoman state resulted in a period of interregnum as a result of the
heavy defeat at the hands of Timur in the Battle of Ankara in 1402. After Mehmed I
(r. 1413–21) rebuilt the Ottoman state, the reign of Murad II (r. 1421–44, 1446–51)
paved the way for the military and organizational accomplishments of Mehmed II.
It is a well-known phenomenon that Mehmed II, referred to as the Conqueror,
developed the Ottoman state from a beylik into an empire. Apart from his reputa-
tion for putting an end to the Eastern Roman Empire by taking Constantinople into
Ottoman possession, he is also credited for the codification which enabled him to
transform the Ottoman state structure into an empire. He regulated some laws
including the exaltation of the sultan by his officers, increased the number of janis-
saries to a great extent, and began choosing his viziers from the kuls, i.e. slaves. All
these reforms are seen as signs of “the imperial turn” of the Ottoman state. 12 Be-
sides, the integration of the already existing military class in the Balkans gained
momentum through the annexation of the Bosnian lands. Some military remnants
of the formative period were turned into auxiliary forces within the same context
of imperial transformation, which will be discussed below and is important within
the scope of this paper.13
The nature of the bulk of the units within the Ottoman army was thus shaped
between the early years of the beylik and the reign of Mehmed II. In this context, it
could be argued that the evolution of the Ottoman army over a period of around
one hundred and fifty years was thus realized in accordance with two basic princi-
ples. First, the early Ottoman military was dominantly of semi-nomadic nature in
terms of manpower, which was also a result of the demographic flow accompany-
ing the military movements in the Balkans. Second, the Ottomans, as a late medie-
val/early modern formation, did not hesitate to adopt the pre-existing military
institutions and their human resources. The Turkish semi-nomadic groups, the
yörüks, and the Vlachs, as already existing semi-nomadic groups in the Balkan
lands that came under the Ottoman rule, were both effectively utilized within the
Ottoman military machine. Their roles within the military changed in time in ac-
cordance with the evolution of the Ottoman state and its needs. Both groups will
Here, I use the term to define the evolution of the Ottoman state into an empire.
12
For a brief survey of the early Ottoman military organization and its evolution in time, see
13
KÁLDY-NAGY 1977.
be analysed in terms of their roles within the Ottoman military structure and the
changes in time.
As one of the demographic elements of the Ottoman movement, the semi-
nomadic yörüks groups were moved into the Balkans both voluntarily and by force
in an intensive wave during the 14th and 15th centuries. These yörüks formed the
man force of the frontier leaders, akıncı beyleri, in the very early stages of the Otto-
man state, as touched upon by Maria Kiprovska:
... keeping in mind that those deportees came to Rumeli along with the he-
reditary akıncı leaders of the late fourteenth and early fifteenth century, it
would not be mistake if we supposed that exactly these companions of the
frontier beys performed the service of akıncıs on the Balkans, for which ser-
vice, on the one hand, they were granted lands and certain privileges on the
beys’ part, and, on the other hand, they themselves were loyal to their com-
manders in chief – the akıncı leaders.14
In other words, the semi-nomadic groups shaped the nature of the Ottoman
military during the early years of the Ottoman state. It is estimated that the human
source of these raiders (akıncıs) were yörüks. In the earliest known fiscal and mili-
tary register from the 1430s, the cavalry with fiefs (timarlı sipahi) under the com-
mand of Ali Bey, who was the son of the famous frontier lord Evrenos Bey, are
mentioned as people with Anatolian and semi-nomadic origin: around 30 sipahis
are recorded to be “Saruhanlu” or “Saruhanludan sürülüb gelmiş”, or “Sar-
uhan’dan sürülüb gelmişin oğlı”. These expressions can be translated as “from [the
Principality of] Saruhan”, “deported from Saruhan”, “son of а deportee from Sar-
uhan”. The deportation from Saruhan is a famous one recorded in the early chroni-
cles. In the late 1300s and the early 1400s, the yörüks of Saruhan did not obey the
prohibition on retail of salt (for the benefit of the state monopoly on salt trade), and
they were deported to Rumelia, predominantly to Plovdiv (Filibe) and Serres (Si-
roz). It is clearly stated that in 1385 “there were nomads in the land of Saruhan,
[Kara Temürtaş] deported them to the province of Serres.” 15 Another record per-
taining to the reign of Bayezid I also clearly refers to the deportation of Saruhanlı
yörüks: “The province of Saruhan had nomadic people, they were spending winter
in the Menemen plain. And there was prohibition on the trade of salt. They were
unwilling to consent. Bayezid Khan was informed about it, and he let his son Er-
tuğrul know about the case and ordered him to deal with these people accordingly
by assigning his trustworthy men and to deport them to Filibe. Ertuğrul received
his father’s order and sent these nomads to Filibe.”16 When these lines from the
chronicles are evaluated together with the records from the early 15 th century tapu
14 KIPROVSKA 2004, 58.

15 AŞIKPAŞAZADE 2013, 84.
322 HARUN YENI
tahrir defters, one may see that not only the raiders but also the cavalry with fiefs
included a good number of warriors with semi-nomadic origins.
In this way, one may notice that in the very early stages of the Ottoman expan-
sion in the Balkans, the frontier marches attracted many people for booty and gaza,
and the yörüks were dominant among them. Yet, in time—probably in the early 15th
century—the military organization of the Ottoman state began to change and the
forces under the command of the frontier leaders were reorganized. The number of
the cavalry with fiefs increased in this period. From what is implied in the sources,
we know that the raider groups did not consist only of yörüks, yet they were domi-
nantly of this origin. According to the 15 th-century chronicles, the deportees from
the Saruhan region were led by Paşa Yigit Bey.17 As a frontier leader he led this
group of nomads to Filibe, which is a strong sign of the military nature of the
yörüks in the Balkans.18 Besides, once these groups were in the Balkans, they began
to live along the frontier of the Ottoman marches, thus forming a pool for the mili-
tary forces of the frontier leaders. It is instructive to follow the direction of the early
Ottoman conquests in the late 14th and early 15th centuries and compare it to the
yörük auxiliary organization centres from the 16th century.19 The settlements con-
quered by the Ottoman forces in the early phases of the Ottoman movement in the
region, a significant number of whom were of semi-nomadic origin, were the plac-
es where one can notice the density of the yörük ocaks in the 16th century.
How can we trace the military legacy of the yörüks in the Ottoman Balkans? In
the second half of the 15th century, the central and provincial armies became more
professional through the increase of fief holding sipahis and the janissaries. This
period coincides with the infantry yaya, azeb soldiers, and cavalry müsellems becom-
ing auxiliary forces due to the centralization and imperial policies of Mehmed II
and being replaced by timarli sipahis and janissaries. Curiously, this is also the pe-
riod when we have first records of the yörüks as auxiliary forces. The law codes
from the reign of Mehmed II outline the structure and duties of yörük auxiliaries. In
accordance with them, they were organized in units called ocaks (literally meaning
“hearth”), and each ocak had 24 people. Four of them were assigned as the serving
ones, one of whom would participate in a campaign or will conduct state service
each year. The remaining three would serve alternately. Twenty of the hearth’s
members were named yamak, and they were to pay for the expenses of the incum-
bent. The incumbent was “to be equipped with his own corselet, and with no lack
in his pole, iron, arrow, bow and shield.” 20 There were some minor changes in the

18 AŞIKPAŞAZADE 2013, 100: “Paşa Yiğit Bey ol kavmin ulusuydı”.
19 For a detailed discussion of the origins of the yörük organization, see YENI 2013, 152–212.
Also, see a very recent study on the issue, KALIONSKI 2020. Unfortunately, I did not have the
chance to see this study before the submission of my manuscript.
20 AKGÜNDÜZ 1990, 354–5; cf. YENI 2013, 168.
organization in the 16th century, yet their role as auxiliary forces was defined in the
law codes more clearly.21 In the records of the Grand Council (mühimme defterleri of
the divan-ı hümayun), there are various entries where yörüks, yaya-müsellems, and
voynuks were all summoned to act as auxiliary forces doing such jobs as construct-
ing bridges over rivers or preparing roads when needed. When they were not at-
tending campaigns, they worked in mines, repaired fortifications, and performed
other works of that kind. As for the transformation of the semi-nomadic groups
that were within the state mechanism from the foundation process onwards both
demographically and militarily, it can be deduced that the early Ottoman society
and military—their polity being still a kind of frontier principality—were hugely
dominated by these semi-nomadic groups. Yet, from the second half of the 15 th
century onwards, a centralist and imperial policy began to hold sway. Thus, the
frontier lords in the Balkans turned into ordinary provincial governors, and the
position of the yörüks in the military diminished accordingly.
There was also another semi-nomadic group with whom the Ottomans met
during their military movements in the Balkans. It was the Vlachs. As it was prov-
en to be valid for some other groups, they were employed by the Ottomans within
their military structure as they had some previously established military organiza-
tion. 22 It is well known that the Ottomans were ready to adopt pre-existing institu-
tions and regulations, at least in the early years of their control of a region. 23 On the
basis of the early Ottoman laws concerning the Vlachs, it is assumed that “Otto-
man legislation concerning the Vlachs was adopted and revised legislation of pre-
vious periods.”24 The adoption of the established administrative applications
opened the way for a smooth transition, thus in some cases, facilitating the integra-
tion of a region. In this way, a good number of the existing Balkan aristocratic elite
and military officers were utilized in accordance with their status and ranks. Stud-
ies on the 15th-century tax registers reveal many examples of it. 25
While a part of the Balkan aristocracy became a part of the Ottoman army with
fiefs as income in return for their services, there were members of the lesser nobili-
ty who were also integrated. Among them, there were voynuks, martoloses, and the
Vlachs. The Vlachs were also in a way related to the other military groups, in terms
of supplying manpower for voynuks and martoloses. Roughly speaking, the Vlachs
had a military organization by themselves besides being a human source for the
21 YENI 2013, 168–70.

22 N. Isailović provides a picture of the Vlachs before the arrival of the Ottomans with details
on the taxation of their pastoral production, ISAILOVIĆ 2017, 30.
23 Kiprovska discusses the integration of the already existing groups of population with
military functions into the Ottoman system and its reflection on the Ottoman conquests in
the Balkans throughout her article, KIPROVSKA 2016. For the earliest study on the issue, see
İNALCIK 1953.
24 ISAILOVIĆ 2017, 39.
25 The earliest Ottoman tax register has many examples of it, see İNALCIK 1987.
324 HARUN YENI
other local groups.26 In this aspect, they were similarly employed as the yörüks. As
noted above, in the medieval Balkans the Vlachs were already a group performing
military services, thus enjoying a specific legal status differentiating them from the
rest of the population. They were also used for the inhabitation of depopulated
regions. In time, the impact of the feudal structure within the Vlachs began to loos-
en their position in favor of sedentary life, and thus some of them were no longer
part of the military structure. Yet, their leaders continued to be associated with the
military.27 When the Ottomans came to the Balkans in the 14th century and moved
towards the Bosnian territories in the early 15 th century, they encountered the mili-
tarily-associated Vlachs, and agreed to include them into the Ottoman military and
administrative structure. As already mentioned, the way of their registration into
the Ottoman fiscal and military registers reveals the fact that they had already been
associated with the military even before the arrival of the Ottomans. The chiefs of
the Vlachs were recorded as “kadîmî sipâhî oğlı” or “sipâhî neslinden” (trans. sons of
old cavalry, descendants of cavalry).28 We know that the Vlachs were present in the
conquest of the towns of Maglaj, Tešne, Doboj, Zvornik, Teočak, and Tuzla in
northeastern Bosnia.29 It can thus be asserted that the Ottomans made use of these
nomadic groups as active combatant forces. In this sense, they can be seen as non-
Muslim akıncıs, raiders.
How were the Vlachs organized according to the early Ottoman registers? Ac-
cording to the first kanuns for the Vlachs of Braničevo and Vidin in the 1460s and
Smederevo in the 1470s, the basic military obligation of the Vlachs was to provide
one voynuk from five households. The kanuns for Hersek from 1477 and for the
Maglaj and Kral districts of the 1480s mention their obligations to provide one
incumbent (eşkünci) for every ten or fifteen households to participate in campaigns.
In the early 16th century, they were required to provide one voynuk or gönder for
guarding unsafe regions, and they were to join the campaigns as cavalrymen. The
Vlachs were organized into knezliks in parallel with their settlement. The term
katun was also used to refer to a fiscal unit consisting of 20 households in the be-
ginning, but then this term disappeared together with the decreasing importance
of the Vlachs for the Ottoman military. A knez was the chief of the Vlachs in his
knezlik/nahiye, while a premikür was the leader in his village and the unit was called
premikürlük.30 The voynuks and martoloses were related to the Vlachs because they
were mostly of Vlach origin. Then in time they began to be distinguished from
them through the sedentarisation of the Vlachs. 31
26 ANTOV 2013, 227; KURSAR 2013, 137.

27 KURSAR 2013, 118.
28 KIPROVSKA 2013, 88–9; cf. KURSAR 2013, 143.
29 KURSAR 2013, 130.
30 KURSAR 2013, 139.
31 KURSAR 2013, 138.
In the 1530s, the Vlachs lost their importance for the Ottoman army. This was
because the Ottoman borders were changed, and thus the Vlachs were not as use-
ful as they used to be as active military groups. But the Vlachs living along the
Danubian route were an exception. Thus, most Vlachs lost their military status due
to the changes in the borders, except for the ones along the Danubian route after
the Battle of Mohacs. Their communal leaders such as knezes, premikürs, voyvodas,
and katunars continued to serve, albeit in the auxiliary forces in return for some tax
exemptions.
They are mentioned in a 16th-century provincial law of Pojega in the following
way:
Vlachs are settling in desolated arable fields in the border-province, making
them inhabited and prosperous. Some Vlachs are cultivating fields, while
others are pasturing goats and sheep. In other provinces, they pay 83 akçes
per household in return for cultivation of fields, and cattle tax according to
the Vlach custom. If this records in the register, these hearths (ocaklar) will
become contractors for the performance of the imperial services, defense and
security.32
This quote reveals the fact that the Vlachs had also been turned into auxiliary
forces and they were mainly serving in mines, mountain passages, clearing the
roads for the army, repairing the fortresses, etc., just like the semi-nomads who
came with the Ottomans, the yörüks. The Vlachs were summoned for the imperial
services either during the campaigns to assist the army or for other services in re-
turn for their tax exemptions. There are many examples of such calls for service in
the imperial registers of the Imperial Council.
The conservative nature of the early modern Ottoman state can be seen in this
evolution. Since some of the Vlachs were of semi-military character, the Ottomans
continued to employ them as auxiliary forces and work forces. Around the 1620s,
the Vlachs along the borders also lost their privileges and became ordinary subjects
because the fortress system was introduced for the defence of the borders, 33 while
the yörüks continued to be used as auxiliaries until the end of the 17 th century un-
der the yörük organization. In 1691, they were reorganized under the name of evlâd-
ı fâtihân yet with similar obligations.34 In short, the early Ottoman society and mili-
tary, whose political entity was still a kind of frontier principality, were hugely
dominated by the semi-nomadic groups, and the structure of the military organiza-
tion reflected this through the domination of these groups in the army. Yet, from
the 15th century onwards, and especially during the reign of Mehmed II, the Otto-
mans followed a centralist and imperial policy. Thus, the frontier lords in the Bal-
32 KURSAR 2013, 132.

33 KURSAR 2013, 136.
34 GÖKBILGIN 1957, 255.
326 HARUN YENI
kans evolved into ordinary provincial governors, and the position of the semi-
nomadic groups in the military diminished. However, the pragmatism of an early
modern state continued to make use of these groups by transforming them into
auxiliary forces and using them according to the needs of the state.
Bibliography
AKGÜNDÜZ, A. (1990): Osmanlı Kanunnameleri ve Hukukî Tahlilleri, vol. 1, Osmanlı
Hukukuna Giriş ve Fatih Devri Kanunnâmeleri, İstanbul.
ALTUĞ, U. (2020): Cihan İmparatorluğunun Kurucusu: Osman Beg, İstanbul.
ANONIM TEVÂRIH-I ÂL-I OSMAN (1992): Anonim Tevârih-i Âl-i Osman, ed. N. Azamat,
İstanbul.
ANTOV, N. (2013): The Ottoman State and Semi-Nomadic Groups along the Otto-
man Danubian Serhad (Frontier Zone) in the late 15 th and the first half of the
16th Centuries, Hungarian Studies 27/2, 219–235.
AŞIKPAŞAZADE (2013): Tevârih-i Âl-i Osman, ed. N. Öztürk, İstanbul.
DEMIRHAN, A. (2019): ‘Kuruluş Sarmalı’ndan Kurtulmak, İstanbul.
FODOR, P. (2009): Ottoman Warfare 1300–1453, in Fleet, K. (ed.:) The Cambridge His-
tory of Turkey, vol. 1, Byzantium to Turkey 1071–1453. Cambridge, 192–226.
GÖKBILGIN, M. T. (1957): Rumeli’de Yürükler, Tatarlar ve Evlâd-ı Fâtihân, İstanbul.
GÜNAL, Z. (2007): “Nöker“, in Türkiye Diyanet Vakfı (TDV) İslâm Ansiklopedisi, vol.
33, İstanbul, 216–217.
ISAILOVIĆ, N. (2017): Legislation Concerning the Vlachs of the Balkans Before and
After Ottoman Conquest: An Overview, in Rudić, S. – Arslantaş, S. (ed.:) State
and Society in the Balkans: Before and After Establishment of Ottoman Rule. Bel-
grade, 25–41.
İNALCIK, H. (1953): Stefan Duşan’dan Osmanlı İmparatorluğuna XV. Asırda Ru-
meli’de Hıristiyan Sipahiler ve Menşeleri, in Melanges, V. (ed.:) 60. Doğum Yılı
Münasebetiyle Fuad Köprülü Armağanı. Ankara, 207–248.
INALCIK, H. (1981–82): The Question of the Emergence of the Ottoman State, Inter-
national Journal of Turkish Studies 2/2, 71–79.
İNALCIK, H. (1986): The Yürüks: Their Origins, Expansion and Economic Role, in
Denny, W. B – Pinner, R. (ed.:) Oriental Carpet & Textile Studies. London, 39–65.
İNALCIK, H. (1987): Hicri 835 tarihli Sûret-i Defter-i Sancak-i Arvanid, Ankara.
İNALCIK, H. (1993): Dervish and Sultan: An Analysis of Otman Baba Vilâyetnâmesi,
in İnalcık, H. (ed.:) The Middle East and the Balkans under the Ottoman Empire: Es-
says on Economy and Society. Bloomington, 19–36.
KAFADAR, C. (1995): Between Two Worlds: The Construction of the Ottoman State.
Berkeley–Los Angeles–London.
KÁLDY-NAGY, G. (1977): The First Centuries of the Ottoman Military Organization,
Acta Orientalia Academiae Scientiarum Hungaricae 31/2, 147–183.
KALIONSKI, A. (2020): Yürüks in the Ottoman Balkans, Sofia.
KIPROVSKA, M. (2004): The Military Organization of the Akıncıs in Ottoman Rumelia,

Unpublished MA. Thesis, Bilkent University, Ankara.
KIPROVSKA, M. (2016): Ferocious Invasion or Smooth Incorporation? Integrating the
Established Balkan Military System into the Ottoman Army, in Schmitt, O. J.
(ed.:) The Ottoman Conquest of the Balkans: Interpretations and Research Debates.
Vienna, 79–102.
KURSAR, V. (2013): Being an Ottoman Vlach: On Vlach Identity(ies), Role and Status
in Western Parts of the Ottoman Balkans (15 th–18th centuries), Osmanlı Tarihi
Araştırma ve Uygulama Merkezi Dergisi (OTAM) 34/Autumn, 115–161.
LINDNER, R. P. (1983): Nomads and Ottomans in Medieval Anatolia, Bloomington.
MELIKOFF, I. (1993): İlk Osmanlılar’ın Toplumsal Kökeni, in Zachariadou, E. (ed:)
Osmanlı Beyliği 1300–1389. İstanbul, 149–158.
ORUÇ BEĞ (2014): Tevârih-i Âl-i Osman, ed. N. Öztürk, İstanbul.
PRITSAK, O. (2012): Alp, in Encyclopaedia of Islam, Second Edition,
http://dx.doi.org/10.1163/1573-3912_islam_SIM_0558
YENI, H. (2013): The Yörüks of Ottoman Western Thrace in the Sixteenth Century, Un-
published Ph.D. Dissertation, Bilkent University, Ankara.
The Battle of the Epic Hero with Death:
The Problem of the Relationship between
Medieval Turkic and Byzantine Folklore
TATIANA A. ANIKEEVA
Abstract
The fifth tale of the Turkic Book of Dede Korkut about Deli Domrul has a strong con-
nection with the Greek legends and Byzantine songs about Digenis Akritis. In
terms of its plot, this story stands apart from the rest of the tales of the Book of Dede
Korkut and, at first glance, from the tradition of the Turkic epic in general. It com-
bines two motifs that were distributed primarily in ancient and modern Greek
folklore: the struggle of the hero with death and the wife who gives her life as a
ransom for her husband’s death. But at the same time, there is a similar motif in the
Bashkir folklore in the oral epic Ural-Batyr as well as in a legend (known to us from
a later literary study from the 19th century) about the hero, who, boasting to his
friends, tries to deceive and kill Death, but God punishes him for that.
Keywords: Oghuz Turks, epic, Byzantine folklore, Turkic folklore
The Turkic medieval written epic The Book of Dede Korkut (Kitāb-i dedem Ḳorḳud,
BDK further in the text) is undoubtedly the most important source on the social
and cultural life of the early Oghuz Turks. Customarily, the composing of the twelve
stories of which the BDK consists is said to have taken place in the 11 th century, but
the stories were fixed in writing only later, approximately in the 15 th century.
The BDK (according to the Dresden manuscript1) consists of twelve
songs/legends, which tell of the exploits of the Oghuz heroes. At the core of the
 PhD (Philology), senior research fellow, Institute of Oriental Studies RAS. The article is a
written version of a paper presented at the 8 th International Conference on Medieval History
of the Eurasian Steppe Nomads and Their Neighbours in the Middle Ages, 20–3 November, 2019,
Sofia.
1 In 2018, a new manuscript containing the BDK epic cycle was found in Iran (the so-called
“Gonbad manuscript”); a brief research on it as well as the manuscript itself was published
in June 2019, shortly before completion of the original version of this paper (see: SHANGOLI
et al. 2019).
330 TATIANA A. ANIKEEVA
epic, which is constructed by these stories, lies the struggle of the Oghuz tribes
against the infidels, non-Muslims (kāfir), in the lands of Asia Minor, as well as the
constant internecine strife among the Oghuz themselves. This text reflects both the
events of early Turkic semi-legendary history (not only historical facts, but also a
set of mythological beliefs) and later events related to the spread of their power in
Asia Minor and their contacts with Byzantium. The stories that comprise the BDK
display a clear connection with both common Turkic literary and folk traditions
and more recent strata. A monument of a written epos, the book lies on the border
between oral and literary tradition as well as between folk narrative and historical
chronicle.
There are several tales in this epos which are closely connected with Byzantium
and the Empire of Trebizond: “Deli Domrul, Son of Dukha Koja”, “How Bisat
killed Depegöz” and “Kan Turali, Son of Kanlı Koja”. Those tales are chronologi-
cally the latest in the BDK, they had been formed already on the territory of Asia
Minor after the migration of the Oghuz tribes to the South Caucasus. The fifth tale
of the Turkish BDK about Deli Domrul probably dates back to the Greek legends
and Byzantine songs about Digenis Akritis: the main hero named Deli Domrul
(“Wild Domrul”) challenges the angel of death Azrail. Allah sends Azrail to take
the soul of Domrul. Domrul engages him in battle, then begs for his life, and then
asks his father and mother to give their souls in his place—but they don’t agree.
After that, Domrul asks his wife and she is ready to make this sacrifice. God gives
life to Domrul and his wife and takes the lives of his old parents.
In terms of its plot, this story stands apart from the rest of the tales of the BDK
and, at first glance, from the tradition of the Turkic epic in general. 2 It combines
two motifs that were distributed primarily in Greek folklore: the struggle of the
hero with death, and the wife who gives her life as a ransom for her husband’s
death—after the mother and the father of the hero refuse to do that.3
2 Later, the story of Deli Domrul in the Turkish literary tradition was closely connected with
the image and legends of Haji Bektash Veli (see GÖKYAY 2000, LXXIII). O. Ş. Gökyay, with
reference to P. N. Boratav, reports, in particular, about the legends about the Holy Tuğrul
(Doğrul/Domrul) that existed among the Bektashis. It is known that the tales related to the
cycle of the BDK for a long time—up to the first half of the 20th century—existed in oral form
in Turkey and Transcaucasia, often within a different, later genre of Turkish traditional
literature: the genre of the folk story (Tur. halk hikayesi).
3 A fairly detailed analysis of the second motif with quite extensive comparisons with mon-
uments of world folklore is contained in the unpublished article of Wolfram Eberhard “Über
die Erzählungen des Dede Korkut” (it is stored in the archive of the Folklore Society of the
Finnish Academy of Sciences) (see EBERHARD, n.d.). In his research, W. Eberhard, who is
primarily interested in the origin of the plot of each of the stories of the BDK, derives the
image of Domrul and the tale about him from Admetos and Alkestis of the ancient Greek
mythology.
THE BATTLE OF THE EPIC HERO WITH DEATH… 331
The epic poem “Digenis Akritis” is a monument of the Byzantine heroic epic
and has come down to us in several versions (six are known) and is referred to as
“the only epic monument of its kind created in Greek after Homer”;4 most likely, it
is founded on the basis of folklore material. The so-called “Acritic songs” belong to
medieval and modern Greek folklore and praise the military exploits of the akri-
tai—a special class of warriors, which emerged in Byzantium roughly in the 9th–
10th centuries for the defense of the eastern frontiers of the Empire.
Tales about Digenis Akritis and his exploits were extremely widespread in the
territory of Asia Minor and the South Caucasus.5 The original version apparently
went back to the end of 10th and the beginning of 11th century; the number of layers
in the surviving versions indicates different periods of their emergence from the
second half of the 11th to the 14th century.
The image of Charon (Χάρων)—the demon of death on a black horse—passed in-
to Byzantine folklore from ancient Greek mythology and for a long time remained in
Greek folklore and the beliefs of the Middle Ages, and even modern times. In ancient
Greek mythology, Charon was a boatman who ferried the shadows of the dead
across the underground river Acheron to the underworld. In many modern Greek
folk songs, Charon appears as a demon of death riding a black horse.
The motif of the struggle of the epic hero with Charon is presented in the 8th
book of the poem “Digenis Akritis”, as well as in a number of the Acritic songs:
And many other things too, for your love,
My soul, I did, to win you utterly;
And yet I missed my aim, I lost my hope.
For be assured certainly I am dying.
Me, the invincible, Charon quite routs,
Hades parts, dear, from my much love of you,
And the tomb covers me with all my pain
And grief unbearable for your widowing.
(Digenis Akritis, 8, 124-5)6
See also:
If Digenes the Borderer’s name was heard
Horror seized all and greatest cowardice;
Such favour had the youth received from God
His name alone would rout his adversaries.
For if the hero ever went out hunting
All beasts would run to cover in the swamp.
4 SYRKIN 1960, 131.

5 The Armenian legend of Kaguan Arslan and his bride Margrit is also known as a version of
the Greek song about Digenis and his struggle with Charon (ZHIRMUNSKII 1962, 199).
6 MAVROGORDATO 1970, 239.
Now he is held down by a little tomb,

Vain, ineffectual for all to see.
Who is it greatly dared to bind the strong?
Who had strength to subdue the undefeated?
Bitterest Death, accessory of all,
Charon, thrice-cursed and common taker-off,
Hades insatiate, these three man-killers,
The three unpitying, and every age,
All beauty withering, wasting all glory.
(Digenis Akritis, 8, 260-70)7
Here, in the poem of Digenis Akritis, the “battle” of the main epic hero Digenis
with Charon ending with the hero’s death is referred to metaphorically, whereas in
the Acritic songs this motif is much more developed: Digenis fights with Charon
more than once.8
The question of the relationship between Byzantine folklore and the Turkic
epos remains unexplored to a significant extent, but it is obvious that the formation
of the BDK was strongly influenced by both Byzantine Greek folklore and some
archaic epic themes and motifs.
The motif of the fight or outwitting between a man and Death/Azrail from the
Turkic epics can also be found in Crimean Tatar folklore and—quite unexpected-
ly—in the Bashkir epics.
At the end of 19th century, Wilhelm Radloff collected some works of the oral ep-
ic folklore (epic destans, songs, short stories) of different dialects of the Turkic peo-
ples of Crimea (Crimean Tatars of the southern coast of Crimea, Tatars of the north
and steppe parts of Crimea, and the Karaite Turks); the materials were published
in the 7th volume of his Proben der Volkslitteratur der nördlichen türkischen Stämme
(1888, VII. Theil “Die Mundarten der Krym” 9). Among them, there was the short
story “Azraiilnɪŋ dostu” (“The Friend of Azrail”) from Miskhor: 10 a man who once
became a friend of Azrail cheats him when the latter comes for his soul. 11 As a re-
sult of a trick, the man succeeds in cheating the angel of death and almost defeats
him: when Azrail comes after the man’s soul, the man starts reading a book and
asks Azrail to wait for a while (until he finishes the book) and Azrail agrees with
him; when the angel of death comes for a second time the man says that he would
finish reading the book in seventy years––thus, the man manages to deceive Azrail
and get another seventy years of life. This legend, of course, is not directly related
7 MAVROGORDATO 1970, 249.

8 POPOVA 1974, 90.
9 See: RADLOFF 2010.
10 Miskhor (Tur. Mishor) is a village (and now a locality) on the southern coast of Crimea, 10
km southwest of Yalta.
11 RADLOFF 2010, 88.
to the Turkic motif of the hero's struggle with death, which was discussed above,
but probably represents one of its very late interpretations.
However, at the same time, there is a similar motif in Bashkir folklore in the
oral epic “Ural-Batyr” and in a late legend about the hero who, boasting to his
friends, tries to deceive and kill Death, but God punishes him for this.12
In the Bashkir “Ural-Batyr” epic the hero is also looking for a struggle with
Death in order to kill it.13 But according to some researchers, the status of this epic
itself is still subject to discussion14—it is known that the epos exists in a single edi-
tion from 1910, which was compiled by the collector of the epos. The other men-
tioned Bashkir legend is known to us only from a late literary and dramatic narra-
tive of the 19th century (though its folklore basis is indisputable), which was pub-
lished by I. I. Zheleznov15—this is probably the latest development of the theme of
Ural-Batyr.
Ioasaf Ignatievich Zheleznov (November 12, 1824 – June 10, 1863) was a Russian
writer, researcher of the Ural (Yaitsky) Cossacks, collector of folklore, historian,
and ethnographer. Being himself from a family of Ural Cossacks, in 1862 he was
elected as a member of the military office, which was responsible for the affairs of
the Ural Cossack community. Zheleznov was famous as a defender of the rights
and interests of the community. In 1858, he published his ethnographical work
Uraltsy — Essays on the Life of the Ural Cossacks («Уральцы — очерки быта
уральских казаков»), which immediately drew attention to the author and re-
vealed his remarkable literary talent. After that, he started to collect material on the
ethnography and history of the Ural Cossacks. A significant part of these materials
has not been treated and edited by I. I. Zheleznov and was published only after his
death; some part was issued by him in Legends and songs of the Ural Cossacks
(«Предания и песни уральских казаков») that was reprinted with some addi-
tions in 1888 and then in 1910 (both after the death of I. I. Zheleznov). 16
In Uraltsy, there is a legend represented as a story told in Russian to Zheleznov
by an old man somewhere in the steppes. The man tells of “the holy hero” (without
name) that lived in Bashkiria about 200 years ago and was famous as a brave de-
fender of all poor and underprivileged people. The hero was also well known for
his exploits as a successful and skilled hunter. Once, at a feast, he decides to pre-
tend to be dead in order to experience what death is and tell all the living people
12 I am very grateful to professor Sergey Yu. Neklyudov who drew my attention to this
publication.
13 See: ZARIPOV 1987, MIRBADALEVA et al. 1977.
14 NEKLYUDOV 2019, 91.
15 See: ZHELEZNOV 1910, 220–6.
16 In 1888, his three-volume collected works with the first biographical essay about him were
published in Saint Petersburg. The author of the essay was N. A. Borodin, a well-known
Ural scholar who was acquainted with the local tradition. In 1910, this three-volume edition
was reissued under the general title The People of Ural (Uraltsy) (see: ZHELEZNOV 1910).
about it, and at the same time “to punish Death properly”. However, he does not
succeed in deceiving Death, and his fellows, the djigits, shortly find the hero actual-
ly dead in his fake grave, with the handle of his axe on his feet and a look of horror
on his dead face. The legend finishes with a moralizing maxim by the old man
about the punishment of Allah for mocking death. The man also claims the happi-
ness of the Bashkir people was buried in the grave of the hero:
Struck with fear and barely alive, the friends of the hero, having filled up his
grave tightly, return to the nomad camp and tell about his death to the peo-
ple, and the people, seeing in this event the punishment of the Almighty
God for the daring desire of a mortal to learn the secret of fate, give them-
selves up to grief and pray for the soul of the hero. The grave of the hero,
concludes the legend, buried in itself, together with him, also the happiness
of the entire Bashkir people.17
Thus, the motif of the hero’s struggle with Death/the Angel of Death/Azrail is
represented in a few different Turkic epic traditions sometimes with a strong liter-
ary influence or editing. If we talk about the plot of the Turkic epic (the tale of Deli
Dumrul from the Book of Dede Korkut), then the influence and the relationship with
medieval Greek folklore can undoubtedly be seen, but at the same time it is possi-
ble that such a motif of confrontation, dispute, and cheating of Death by the main
epic hero may also have a Turkic origin.
Bibliography
EBERHARD, W. (n.d.): Über die Erzählungen des Dede Korkut (manuscript).
MAVROGORDATO, J. (1970): Digenes Akrites, ed., transl. Mavrogordato, J. Oxford.
MIRBADALEVA A. S. et al. (1977): Bashkirskii narodnyi epos, ed. Mirbadaleva A. S.,
Sagitov, M. M., Kharisov, A. I. Moscow.
NEKLYUDOV, S. Yu. (2019): Folʹklornyi landshaft Mongolii. Vol. 1, Mif i obryad, Moscow.
GÖKYAY, O. Ş. (2000): Dedem Korkudun Kitabı, ed. Gökyay, O. Ş. İstanbul.
POPOVA, T.V. (1974): Vizantiiskii narodnyi i knizhnyi epos, in Averintsev S. S. (ed.:)
Vizantiiskaya literatura. Moscow, 77–121.
RADLOFF, W. (2010). Proben der Volkslitteratur der nördlichen Türkischen Stämme,
gesammelt und übersetzt von Dr. W. Radloff. VII Theil. Die Mundarten der Krym.
Ed. T. Çulha, Istanbul.
SHANGOLI, N., YAGHOOBI, V., AGHATABAI, SH., BEHZAD, S. (2019): Dede Korkut
Kitabı’nın Günbet Yazması Modern Türklük Araştırmaları Dergisi 16/2, 147–379.
SYRKIN, A. Ya. (1960): Digenis Akrit, transl. Syrkin, A. Moscow–Leningrad 1960.
ZARIPOV, N. T. (ed.) (1987): Bashkirskoe narodnoe tvorchestvo. Vol. 1, Epos, Ufa.
17 ZHELEZNOV 1910, 226.

ZHELEZNOV, I. I. (1910): Uralʹtsy. Ocherki byta uralʹskikh kazakov. Polnoe sobranie

sochinenii Iosafa Ignatʹevicha Zheleznova. 3rd ed., ed. Borodin, N.A. Vol. 1, St. Pe-
tersburg.
ZHIRMUNSKII V.M. (1962): Oguzskii geroicheskii epos i “Kniga Korkuta”, in Zhir-
munskii, V. M. – Kononov, A. N. (ed.:) Kniga moego deda Korkuta. Oguzskii ge-
roicheskii epos. Transl. Bartol‘d, V. V. Moscow–Leningrad, 131–258.
Early Mediaeval Nomads between Iran,
Central Asia and Europe:
The Issues of the Archaeological Evidence
BRUNO GENITO
Abstract
In this article, the author deals with the themes of the centrality of the Iranian plat-
eau and the archaeological cultures that have succeeded therein in relation to the
issues of the ancient and mediaeval Eurasian nomadism. Spanning briefly from the
Iron Age to the late Middle Ages, the author identifies and analyses the main as-
pects of this centrality crossing also the intricate question of the ethnical and cul-
tural affiliation. The whole phenomenon of ancient Eurasian nomadism was active
and operational thanks to numerous ethnically Iranian population groups. And
even in the Middle Ages there was a conspicuous number of Iranian populations,
although for the most part they were replaced by those of Paleo-Turkic stock.
Keywords: Archaeology, Nomads, Iran
Introduction
Among the methodological approaches used by most of the archaeologists dealing
with the Eurasian nomadic groups, the most popular one is the attempt at singling
out their identity on the basis of material evidence. While ethnocultural identities
are usually well documented in historiographical evidence, they always appear
difficult to recognize from the point of view of the archaeological documentation.
 Professor at the Department of Asia, Africa and the Mediterranean – Università degli Studi
di Napoli, L’Orientale. The article is a revised version of a paper delivered at the 8th Interna-
the Middle Ages, held in Sofia, November 20-3, 2019. I would like to thank the organisers of
the conference and in particular, I would like to thank Dr. Konstantin Golev, Assistant Pro-
fessor at the Institute for Historical Studies at the Bulgarian Academy of Sciences, and István
Zimonyi, Head of the Department of Mediaeval History at the University of Szeged and
Head of the Department of Altaic Studies at the same University. I would also like to thank
Dr. Roberto Esposito and Dr. Francesco Dall’Aglio for the English translation of this paper.
338 BRUNO GENITO
Moreover, the complex relationship between the concepts of ethnos and culture
further complicates the issue, especially with regard to the medieval period, both
in Europe and Asia.1 Without delving into an analysis of the theories proposed
during the last two centuries in the archaeological scientific debate, we will just
mention some of the most important and recent contributions to this issue. 2 All
cited authors have examined the differences and the overlaps of those terms ac-
cording to their ideological and political perspectives, and the different socio-
political and cultural environment where they lived.
Generally speaking, the cultural-historical approach is certainly predominant in
the field of archaeological studies, and the identification and the reconstruction of
ethnicities is thus one of the main aims of the related research activities. The origi-
nal significance and functions of items and objects, even when they remained sub-
stantially unknown, have often been used in order to construct a possible ethnic
background.3 Step by step, ethnicity has been conceptually transformed into a real
culture, moving from an abstract indeterminacy to a real identity. Many archaeo-
logical interpretations have, thus, been oriented to consider every culture, i.e., the
set of traits widespread over a given territory within a definite time span, as the
bearer of a certain identity. Thus, the scientific debate in the archaeological field
has determined a sort of distinction between those who consider ethnicity as the
original character of a people and those who regard it simply as the marker of cul-
tural changes. Today, most of the scholars refer to ethnos as a human group de-
fined by physical type, customs and cultural traits, without considering whether an
actual blood relationship existed among its members. Much more problematic,
from a properly archaeological point of view, is to understand the value and the
function of a single item of material culture, and associate it with a sufficient de-
gree of approximation to an ethnos or population.
As a matter of fact, attempts to identify through their material culture the typo-
logical and chronological elements pertaining to ethnic groups have been devel-
oped together with those used in other fields of research in order to identify lan-
guages and homelands. Therefore, the concept of culture has been generally con-
sidered in each of its elements a reflection of sorts of a “national” spirit. This in
turn led scholars to consider material culture as bearing the “footprint” of the eth-
nic group responsible for its production. The detailed topographic maps increas-
ingly widespread in archaeological research are generally used to identify the dis-
tribution of the patterns of object types, viewed as localised in highly homogene-
ous and well-defined macro cultural areas. Such areas are then tentatively com-
bined with the mosaic of the ethno-historical sources. This approach has occurred
in particular within the field of studies and analyses of the nomadic peoples, con-
1 GEARY 1983.
2 LUCY 2005; BIERBRAUER 2009; CURTA 2002; POHL – MEHHOFER 2009; BÁLINT 2009.
3 BRATHER 2004.
EARLY MEDIAEVAL NOMADS BETWEEN IRAN, CENTRAL ASIA AND EUROPE… 339
sidered bearers of a set of cultural traits that could be also somewhat recognized in
their historical background. Archaeological cultures were thus equated, in particu-
lar political contexts, to ethnic groups, and this was also seen as a legitimisation of
modern claims over a given territory. Since cultures were regarded as a homoge-
neous compact unity, the direct consequence of this cultural-historical approach to
single out ethnic identities was the idea that distinct types of objects were actually
ethnic indicators. Ethnic groups could be recognized, then, from archaeological
records based on their specific aspects, parts of which had been turned into partic-
ular types of pottery, tools, brooches, swords, arrow heads, belts, belt fittings or
architectural remains. Searching for ethnic groups was thus reduced to the identifi-
cation of typologies, with the assumption that archaeological classifications could
reproduce the categories that ancient producers and consumers had originally
elaborated.
In this regard, there is no great difference between the approaches used to iden-
tify, especially in the mediaeval world, a population, be it nomadic or sedentary, or
a political dynasty, and the context of the archaeological studies of Iran in the late
Iron Age, without taking into consideration the details of the obvious distinction
between different interpretative levels in the dynastic, ethnic, political and chrono-
logical horizons.4
The Iranian Plateau
Within the methodological framework of the study of mediaeval and earlier no-
madic realities in Eurasia, the crucial role played by the Iranian plateau in regard
to those people probably originates in the presumably mostly nomadic character of
the original Iranian groups that moved to Iran from Central Asia, and that only in
a later moment would be transformed, albeit gradually, into sedentary popula-
tions.
This long and perhaps complicated process is far from being fully understood,
both due to the lack of direct sources, and a rather insufficient archaeological doc-
umentation. The interpretation of that process, furthermore, lends itself to the in-
terpretative misunderstandings we have mentioned earlier. In any case, the Iranian
centrality is evident in all the most important phases in which the history of Iran
and that of the nomadic populations are analysed. This role has remained the same
also in much later periods, including the late mediaeval, even if with very different
and less fluid cultural and ethnic connotations. This is also due to the fact that the
plateau was geographically located between the east and the west, i.e., the two
poles of the nomadic movements from the late Bronze Age to the middle of the 1st
Millennium AD.
Within the context of the early and late nomadism in the Eurasian Steppes, the
centrality of the Iranian plateau and of the cultures that moved across it is quite
4 GENITO 1998.
340 BRUNO GENITO
well known in its general features. It will be enough to consider, for instance, most
of the nomads of the Iron Age historically known as Cimmerians, Sauromatians,
Sarmatians, Scythians, Saka, Massagetes, Arimaspes, Issedones, etc. They were
thoroughly described by Herodotus and other important historians, but unfortu-
nately, they are much less identifiable through archaeological evidence. Historians
have always correlated those populations to an ethnic Iranian origin and to Iranian
languages. The cultural relationship between the nomadic world and the Iranian
plateau can also be clearly noticed at a later stage, by many features and elements
chronologically related to the historical “state/imperial” Iranian dynasties. Such is
the case, for example, of the nomadic social nature and origins of the first dynasty,
the Achaemenids in the 6th–5th century BC. That character is at times evident within
the epigraphic texts written in Old Persian, Accadian, and Elamite. The communi-
cation system of the dynasty reminds of its original nomadic social structure, espe-
cially in the form and titles, more than we may expect.5 The same may be observed
in the later nomads, starting from the habit of the Byzantine sources to call all the
nomads of their time Scythians. 6 The name of the Scythians, in fact, outlived their
own historical period; some authors continued to use the term for many groups
unrelated to the original Scythians, such as Huns, Goths, Türks, Avars, Khazars,
and other unnamed nomads.7 Others frequently applied the name “Scythian” to
nomadic groups in general who never had any ethnic relation to the earlier Iranian
nomads bearing the same name.
During the Parthian period (2nd century BC – first half of the 2nd century AD),
we can observe that the same Arsacid dynasty originally belonged to the nomadic
tribe of the Dahae8 and that during the Sasanian period (mid-3rd and mid-6th centu-
ry AD) a dynamic link continued to exist between the cultural realities already
5 ROSSI 2003, 681–700.

6 Although the relations between the nomads and Byzantium have been the subject of many
studies, there is to date no comprehensive treatment of the topic. Byzantine sources provide
abundant information about how the imperial government in Constantinople dealt with the
peoples inhabiting the steppe lands north of the Black and Caspian Seas. Conclusions drawn
from written sources, however, differ from those resulting from archaeological excavations
(CURTA 2001). New evidence from excavations has enriched our understanding of relations
between nomadic and settled communities and has questioned the pigeonholing of diverse
communities into preconceived ethnic categories such as “Pechenegs,” “Oghuz,” and “Cu-
mans”.
7 HARMATTA 1996, 181–182; NICHOLSON 2018, 1346–1347.
8 The Dahae or Daha were a group of Scythian tribes of Iranian origin. Strabo places the
Dahae nomads roughly in present-day Turkmenistan (DE BLOIS – VOGELSANG 1993). Together
with other Scythian tribes they fought alongside Darius III in the final battle with Alexander
at Gaugamela (331 BC) in Modern Iraq. They later joined Alexander the Great on his Indian
expedition. A branch of the tribe, the Parni, migrated south into Parthia and joined the locals
who were also Iranians. They founded the Arsacid dynasty which reunified Persia under
Iranian hegemony.
sedentarized for centuries and the continuous nomadic movement from the North
and North-East of the plateau. Over the course of a few centuries, at the turn of the
1st millennium BC and the 1st millennium AD, that movement will experience one
of the most significant and great ethno-cultural and linguistic transformations. This
step was, among other things, signalled by the use of the expression “Hun-
Sarmatian period”,9 emphasising the ethno-cultural Iranian/Turkic mixture – be-
fore the increasingly decisive advent of the Paleo-Turkc peoples in place of the
Iranians in the Eurasian steppes.
It is not the case to analyse here in detail the numerous archaeological and his-
torical proofs of the cultural relationship between the Parthian and Sasanian peri-
ods and Central Asian cultures, especially with nomadic peoples such as the Huns,
the eastern Goths, the Avars, the Khazars, etc. There are many examples that could
be given in this regard, from simple items of material culture (representations on
coins, seals, etc.) to the macroscopic architectural and figural remains (rock-reliefs,
architectural and painted decorations, etc.). I will refer here, for example, to the
well-known remains of the painted walls of the Afrasiyab palace in Samarkand
(Uzbekistan)10 where the iconographic details of the images are incredibly similar
9 BOTALOV – GUTSALOV 2000.

10 One of the pictorial decorations most useful for our discussion is that of Room 1 in the
palace discovered at Afrasiyab: see Azarpay, Belenitsky, Marshak, Dresden 1981. It was, in
all probability, the audience hall of the palace of Varkhuman (Vargoman), one of the sover-
eigns of Samarkand (REMPEL’ 1984; PACHOS 1966; AFRASIAB 1969–75). From a purely icono-
graphic point of view, the subjects depicted can be summarised as follows: 1. Southern wall:
a long procession of richly dressed characters advances on foot or on horse and camelback,
representing the procession accompanying the princess sent in marriage to the sovereign of
Samarkand; 2. Northern wall: the pictorial remains are divided into two parts: eastwards, a
group of horsemen is represented fighting with wild animals attacking them; westwards, a
river with fishes, water birds, lotus flowers, fantastic animals, on which two boats advance,
one of which is crowded with only female characters; 3. Eastern wall: southwards, a sea-
scape featuring birds and fishes is depicted; three naked children shoot bows, a man is at-
tached to the tail of a bovine; northwards, two very fragmentarily preserved figures, per-
haps Indians; 4. Western wall: forty characters, in which it is possible to recognize different
ethnic groups due to their various costumes and facial features, advance in procession from
several points towards the central part of the composition, which was originally supposed
to be occupied by an image unfortunately totally destroyed. On the garments of one of these
characters (AL’BAUM 1975, pl. 15) appears a long inscription, consisting of two horizontal
lines of Bactrian cursive writing and sixteen vertical lines of Sogdian cursive. The ambassa-
dor of the principality of Ciaghānyān (basin of Surkhān-dāryā) and that of Čač (the region of
modern Tashkent) speak to the king of Samarkand, Varkhuman (Vargoman). The complex
historical situation that arose in Sogdiana after his death does not allow to believe that the
paintings were made to commemorate him, therefore they must have been painted in his
reign, more precisely the third quarter of the 7 th century. Shishkin (SHISHKIN 1966) believed
that the paintings illustrated legendary stories, according to a Sogdian tradition attested,
e.g., also in Pendžikent, ‘building’ VI, 41, where the literary cycle of Rustam is depicted.
342 BRUNO GENITO
to most of the documented Avar objects, like dresses, belts, belt-fittings, swords,
etc.,11 also similar to other representations that the Sasanian dynasty has carved on
the rock reliefs in the gorges and mountain passes of Fars and other regions of the
plateau.12
Those iconographic peculiarities arose as a consequence of the increasingly
growing role of the Sasanian dynasty in the geopolitical balance of the time, which
somehow connected geographically the plateau and the world of the steppes. The
Sasanian state political-formation was also in close contact with the Roman-
Byzantine world in the West, and in the East with the new political realities that
were beginning to appear on the international stage in the areas of the steppes. The
horse was certainly one of the most important figures in this series of contacts: it
seems to have played a particular and decisive role due to its technical and tactical
military importance, the latest consequence of the primitive equestrian nomadism
dating back to the Iron Age. The presence in the Roman imperial army of the
clibanarii and catafractarii derived from the nomadic world 13 reveals an unexpected
Al’baum (AL’BAUM 1975) concluded that the subject depicted on the western wall was the
reception ceremony of the foreign ambassadors at the court of Varkhuman (Vargoman) and
that the journey of the legations to Samarkand was illustrated on the other walls. Different
interpretations have been advanced recently (CRISTOFORETTI – COMPARETI 2007, 33–71) in
relation to two Chinese and Iranian festivities that in a certain chronological period occurred
probably in the same time of the year.
11 LÁSZLÓ 1955; BÓNA 1979; 1980; 1988; GARAM 1990; GARAM 2000.
12 HERMANN 1980–81. The ancient Near Eastern tradition of carving rock sculptures became
extremely widespread during the Sasanian period with a long series of reliefs, most of
whom are in fact located in the mountain ranges of Fars. The commemorative and laudatory
character of this type of sculpture (with scenes of royal investiture and victories) always
found its main representation on the plateau. This particular form of figurative expression,
conveying peculiarly distinctive meanings and values aimed at the memory of the observer,
can be read at several levels. These rock-reliefs have often been observed and studied as
isolated phenomena of great artistic importance and great commemorative and celebrative
significance. On the other hand, their role as significant elements of a successful attempt to
institutionalise or, if this term is to be preferred, sacralise a territory, has been much less
emphasized. We are especially referring to their presence on mountain passes, river bends,
etc., as a way of occupying a territory which may be dated back, as we know, to very ancient
times. It is well known that the Zagros and the Elburz mountain chains have always been
key physical features of a difficult, impervious landscape, sometimes characterised by for-
bidding heights as well, which could also be ideologically and religiously celebrated, mak-
ing it co-essential to all the political events which occurred on the areas of the plateau. These
remains are some of the most successful and scenically impressive examples of figurative
artistic production; everything was designed to plan, and these works were cut into the
most durable material support, rock, turning it into the most sensational instrument of polit-
ical propaganda. The study of the territory certainly remains one of the most neglected as-
pects in the traditional approach to the archaeology of the Sasanian period.
13 GAMBER 1968; BIVAR 1972; NIKONOROV 1988, 127-28; ERAMO 2018.
predecessor of mediaeval tactics. The earlier Iranian peoples like the Sarmatians,
and more recent dynasties like the Parthians and the Sasanians, continued to large-
ly influence the western habit of warfare. Other findings supposedly originating
from Iran and more recent discoveries give evidence of the mixed cultural context
of its nomadic origins much larger than it was previously expected, 14 owing to the
application of a new, up to date physical anthropological approach which seems to
provide fresher perspectives to the matter.
Within this context the Parthians, and especially the Sasanians, make the horse
the main iconographic instrument of their imperial propaganda; the first rock re-
liefs at Firuzabad realised by Ardashir I (224–41 AD), and those of Shapur in
Bishapur, will always feature representations of the sovereign on horseback, in
investiture scenes, military victories, or even in combat.
These kinds of representations have been uncovered on graffiti, paintings, figu-
rines, rock reliefs, and also on items like belts, belt fittings, cuirasses, weapons,
arrowheads, etc., in a very wide spatio-temporal distribution in Eastern Europe
and Hungary, Doura Europos, Palmyra (Syria), Iran, Central Asia, and China be-
tween the end of the 1st millennium BC and the middle of the 1st millennium AD.
Their enormous geographical and cultural diffusion has also raised some perplexi-
ties about their origins and chronological location; nevertheless, they should not be
interpreted as a consequence of the transmission of original Sasanian iconographic
themes and motifs all around the Euro-Asiatic world, as proposed for a long time
with many nuances by different schools of studies, particularly in Hungary,15 Chi-
na,16 and Central Asia.17 They may rather be interpreted as aspects of a more gen-
eral and undifferentiated framework of cultural references in which the Iranian
cultural centrality of the period continued to play a significant role. This centrality
consists precisely of the extraordinary complex capacity of the Iranian culture to
absorb, re-elaborate and transmit values, symbols, techniques, and habits of life
and warfare from all around the world, including the nomadic cultures of the
steppes, well beyond the territorial limits of their own sphere of political influence.
In short, following a concept already stated some time ago, we can reasonably
speak of the so-called “external” Iran. Furthermore, that centrality was favoured by
the enormous physical diffusion of the Iranian populations (from China to Eu-
rope), and by the nomadic character of most of them.
The iconographic evidence found on the rock reliefs of the Sasanian period un-
equivocally clarifies the central role of the horse as well as representing objects and
material evidence related to the use of this animal, such as harness, trousers, cloth-
14 BÁLINT 1978; GENITO 1993.

15 BÁLINT 1978.
16 LERNER 2005; 2011; QI DONFANG 1999.
17 MARSHAK 2001.
344 BRUNO GENITO
ing, horse harness, bites, belts, belt fittings.18 One important exception is represent-
ed by the lack of any iconographic and material reference to the use of stirrups,
which is instead present in many of the contemporary pictorial and engraved rep-
resentations of special armour and fittings in Central Asia. Contemporarily to the
Sasanian period during the 420s and 430s, the Huns created an enormous imperial
political formation, probably the largest nomadic political unit during this time. 19
The socio-political evolution of the nomadic society which developed under such
political unity created a contradictory phenomenon: an imperial political formation
(which represented the super-state stage in the long lasting state evolution) com-
posed of nomads (who traditionally represented a non-state community). Previ-
ously obscure peoples were thus able to constitute a social and political system
based on a royal lineage and, at a lower level, on differing kinds of aristocracy.
These political structures were strong enough to triumph at the moment, but so
fragile as to collapse in a relatively short space of time. This was the case of the so-
called “Empire” of Attila which disappeared almost immediately after reaching the
pinnacle of its history in AD 454.20 From an archaeological point of view, a charac-
terization of the material culture of this period based on ethnic origin is practically
impossible.
There are also other archaeological finds, basically treasure troves with all their
dubious related baggage of chronological uncertainties and ethno-cultural affilia-
tions. For their typology and their technical, stylistic, and iconographic characteris-
tics, their attribution remains uncertain, fluidly shifting between the Hun, Iranian-
Sasanian, Avar, and Byzantine world, without neglecting the presence of other
cultural contributions. We are referring to the cemetery of Nagyszéksós21 which
has, in fact, revealed many golden objects for which Greek, Roman, Hun, and Byz-
antine craftsmanship has been variously claimed, 22 or to the famous diadem of
Csorna, with a symmetrical row of stones, reflecting different styles and techniques
which can be connected to the so-called Sarmatian or Alan tradition.23 The object
has, furthermore, many close typological parallels with others found over the
whole Hun political imperial-formation, such as the extremely significant ritual
cauldrons,24 of which those of Törtel25 and Várpalota26 are the best preserved in
Hungary.
18 BÁLINT 1992; BÁLINT 1993.

19 MAENCHEN HELFEN 1973; MAENCHEN HELFEN 1978.
20 WERNER 1956.
21 FETTICH 1953.
22 KISS 1982.
23 KOVRIG 1985.
24 ZASECKAJA – BOKOVENKO 1994; ÉRDY 1994.
25 PÁRDUCZ 1963.
26 KOVRIG 1973.
In 567 AD, the Avars reached the Great Hungarian Plain, where they created a
new state political-formation which was nomadic in character. 27 The Avar political
system, organised in tribes, immediately had a strong political impact on the area.
But its most significant cultural aspect was the creation of a new military tactic
which represented the final stage of the warfare evolution among the nomadic
peoples of the steppes. The military system was based on light cavalry employing
saddles, stirrups, and leaf-shaped lance heads with an incredible penetrating force.
The men fastened their dresses with belts from which weapons hung; the belts,
decorated with characteristic embossed metal plaques constituted the sign of social
rank, according to the techniques and the value of the material used. 28 Amongst
the finds of this period, the hundreds of horsemen graves discovered by Hungari-
an scholars constitute one of the most characteristic elements. Also, these graves
represent the final stage in the evolution of a burial rite which had started long
before, but they also contain an innovation: the simultaneous deposition of the
horseman with his animal.
The study of the archaeology of the nomadic people of the Eurasian steppes en-
compasses a wide range of research, both in geographical and chronological terms.
Within this very extended perspective, the nomadism of the early Middle Ages
presents some particular aspects, on account of which it is appropriate to approach
the topic in a more nuanced manner, especially when considering the many recent
and less recent discoveries that are not always adequately known, and which may
open new research perspectives. Beyond the macroscopic cases of the treasures of
Vrap in Albania,29 of Mala Pereshchepinа in Ukraine,30 of Sânnicolau Mare/Nagy-
27 POHL 1988.
28 LÁSZLÓ 1955.
29 The treasure was discovered in 1902; the objects were produced in approximately 700 AD
and given to the Metropolitan Museum of Art by J. P. Morgan Jr. in 1917. The various ves-
sels making up the ensemble have been attributed to the Avars, who subjugated much of the
local populations and occasionally clashed with the Byzantine Empire. Through these con-
quests, the Avars were able to amass considerable amounts of wealth, some of which was
buried in-mass near Avar settlements. The origin of the treasure troves is disputed; some
maintain that the Avars were themselves skilled metalworkers, while others believe that the
valuable objects (including gold jars, cups, and dishes) found in Avar hoards were produced
in Byzantium and then either looted or given as tribute to the Avars. The ensemble consists
of several gold cups, a silver bucket, several drinking dishes, and a jug.
30 The treasure hoard of Mala Pereshchepinа was found by accident by a cow herder in May
1912, in the Poltava Governorate of the Russian Empire. The villagers destroyed all the ar-
chaeological context, so the researchers who arrived two days later could only study the
traces of its destruction. In total, the treasure consists of approximately 25 kg of gold and 50
kg of silverware, including pots (16 gold and 19 silver), weapons (straight swords, daggers),
jewellery (belts, earrings, bracelets, necklaces, rings), coins (the so-called heavy solids of
Mauricius (582–602) and Constant II (641–668)) and hand tool ornaments. Numbering more
than 800 pieces, the trove was first studied by I. A. Zareckij and N. Makarenko (ZARECKIJ –
346 BRUNO GENITO
szentmiklós,31 variously traced back to the proto-Bulgarians, the Avars, the Khaga-
nate of Khazaria and even to local artisans, we can also mention the belt fittings
MAKARENKO 1912) and by A. A. Bobrinsky (BOBRINSKY 1914): see FONJAKOVA ET AL. 1996;
WERNER 1984. The origins of the objects may be traced back to the artistic production of
Byzantines, Bulgarians, and Avars. The initials M. P. engraved on the bottom of one of the 5
kg, 4th-century bowls, and the signature on one of the rings, have been variously interpreted.
The treasure was considered a Slavic spoil of war, or the tomb of the Bulgarian Khan Ku-
vrat. It is certain that some items, such as the solidi and a belt set, were made in the imperial
workshops in Constantinople. The identification of the trove as a burial site received wide
international resonance after the renowned German researcher Joachim Werner read, in
1984, the name Seibt as Kuvrat, the prominent leader of the early Bulgarian principality (in
the form of ΠΟΒΡΑΤΟΥ or ΠΡΟΒΑΤΟΥ). Several experts had already disputed the identifi-
cation of the find at the time of the reading, as the absence of bones or coffin remains sug-
gests that the treasure should not be interpreted as a burial find.
31 BÁLINT 2010. A group of precious objects found in 1799 (Romanian Sânnicolau Mare),
along the lower course of the Maros river, constitutes the most important treasure (9.9 kg) of
the Carpathian basin for the early Middle Ages (ALFÖLDI 1951; 1952; 1954; LÁSZLÓ – RÁCZ
1977a; RUSU 1985–86). It is, unfortunately, now partially dismembered: twenty-three gold
objects converged, in fact, in the Imperial Treasury in Vienna, while various other pieces
ended up missing or were melted down. The original ensemble is preserved in Vienna,
Kunsthistorisches Mus., Schatzkammer. The ornamental repertoire adopted on these objects
appears very diverse. Among all, the most critically debated are two containers (a jug and a
bottle) both decorated with figurative scenes, however very different from each other in
quality and execution. The most important artefact of the treasure, from an artistic and tech-
nical point of view, is a bottle, on which the depiction of the victorious sovereign is config-
ured as an iconographic motif derived from the art of the steppes; the scene of the abduction
to heaven and that of the hunting sovereign are instead Sassanid and proto-Islamic motifs,
while that of the fight between animals is of ancient and Byzantine inspiration. The dating
and place of production of the treasure, as well as the ethnic origin of the owner, are still the
subject of discussion, both because it is difficult to establish precise parallels – if not to a
limited extent – for the shapes and decorations of the various artefacts, and because the
production of Byzantine metal vessels of the 8th–10th centuries is very little known, whereas
it must have had an influence, directly or indirectly, on the objects in question (BENDA 1965;
BÁLINT 2010). In the specialised literature the hypothesis (MAVRODINOV 1943) is that the
treasure should be connected to late Sasanian and proto-Islamic production, although the
formal affinities and chronological relationships are unsatisfactory. According to the cur-
rently most widespread hypotheses – but not demonstrated in a methodologically convinc-
ing way – the treasure should have been assembled in the Avar Khaganate, in the 7 th–8th
century, or in Bulgaria in the 9th; consequently, its burial could have occurred at the begin-
ning of the 9th century – perhaps following the collapse of the Avar Khaganate (796–803) – or
at the end of the same century, under the pressure of the Hungarian “conquest of the home-
land” (896). The possibility of a Hungarian origin of the treasure, with a dating to the 10 th
century, has also been proposed (LÁSZLÓ – RÁCZ 1977b), yet based on arguments questiona-
ble on a methodological, typological, and chronological level. The thesis of a Bulgarian
origin, on the other hand, does not seem to be supported by the scarce analogies found in
from Uč Tepe in Azerbaijan32 or those from the Saltovo-Mayaki culture in

Ukraine33 and from the Martynovka treasure in Ukraine. 34 These less striking, but
not less meaningful findings, are still waiting to be published in full, and will pro-
vide more exhaustive answers to the many still unsolved questions about the ar-
chaeology of the early mediaeval nomads, and also on the background of the Irani-
an culture.
Bibliography
AFRASIAB. (1969–1975): Afrasiab, 4 vol., Tashkent.
AL‘BAUM, L.I. (1975): Zhivopis‘ Afrasiaba, Tashkent.
ALFÖLDI, A. (1951): Etudes sur le trésor de Nagyszentmiklós. Cahiers Archéologiques
5, 123–149.
6, 43–53.
7, 61–67.
AZARPAY, G. – BELENITSKII, A. M. – MARSHAK, B. I. – DRESDEN, M. J. (1981): Sogdian
Painting, Berkeley–Los Angeles–London.
BALINT, Cs. (1978): Vestiges Archéologiques de l’époque tardive des Sassanides et
leurs relations avec le peuples des steppes. Acta Archaeologica Hungarica 30,
173–211.
BÁLINT, Cs. (1989): Die Archäologie der Steppe. Steppenvölker zwischen Volga und Do-
nau vom 6. bis zum 10. Jahrhundert, Wien, Köln.
artefacts from the territory of Bulgaria; the comparisons proposed up to now suggest that
the Proto-Bulgarian goldsmiths had relations not so much with the metalworkers of the
steppes, but above all, to a substantial extent, with the Byzantine ones. Given its nature as a
treasured complex, the set of artefacts is, on several levels, composite: in fact, concerning its
dating it is possible to indicate, for many elements, similarities with works of the 7 th century.
Most of the decorative motifs, on the other hand, can be compared to the late Avar art of the
8th century. Conversely, from a typological point of view, it seems possible to establish some
connections with other works from the treasure found in Vrap, near Durres, Albania.
32 BÁLINT 1992.
33 PEKARSKAJA – KIDD, 1994; KIDD 1995; KIDD 1996; BÁLINT 1989, 88-92; BALOGH 2004. Saltovo-
Mayaki is the name given by archaeologists to the early mediaeval culture of the Pontic
steppe region located approximately between the Don and the Dnieper rivers. This culture
was a melting pot of Onogur, Khazar, Pecheneg, Magyar, Alan, and Slavic influences. Dur-
ing the 9th century, the Saltovo-Mayaki culture was closely associated with the Khazar
Khaganate, and archaeological sites from this period are one of the possible ways in which
historians can track the geographic extent of the Khazar influence.
34 The objects of this treasure are also very problematic, including the belt fittings that are
usually used as a term of comparison in order to find stylistic and technical confirmations of
the various proposed attributions.
348 BRUNO GENITO
BÁLINT, Cs. (1992): Kontakte zwischen Iran, Byzanz und der Steppe. Das Grab von
Üč Tepe (Sowj. Azerbajdžan) und der beschlagverzierte Gürtel im 6. und 7.
Jahrhundert. Awarenforschungen I, 309–496.
BÁLINT, Cs. (1993): Probleme der archäologischen Forschung zur awarischen Land-
nahme, in: Müller-Wille, M. – Schneider, R. (ed.): Ausgewählte Probleme europäi-
scher Landnahmen des Früh- und Hochmittelalters. Part I, Sigmaringen, 195–273.
BÁLINT, Cs. (2009): A contribution to research on ethnicity: a view from and on the
east, in: W. Pohl & M. Mehofer (eds.) Archaeology of identity – Archäologie der
Identität. Wien, 145–182.
BÁLINT, Cs. (2010): Der Schatz von Nagyszentmiklós, Archäologische Studien zur früh-
mittelalterlichen Metallgefässkunst des Orients, Byzanz’ und der Steppe. (Varia Ar-
chaeologica Hungarica XVIb), Budapest.
BALOGH C. (2004): Martinovka-típusú övgarnitúra Kecelről. A Kárpát-medencei
maszkos veretek tipokronológiája. Móra Ferenc Múzeum Évkönyv – Studia Ar-
chaeologica, 241–303.
BENDA, K. (1965): Současny stav studia zlatych nádob pokladu ze Sannicolāul mare
(Nagyszentmiklós). Slovenská archeológia 13, 399–414.
BIERBRAUER, V. (2009): Zur ethnischen Interpretation in der frühgeschichtlichen
Archāologie, in: W. Pohl (ed.): Die Suche nach den Ursprüngen. Von der Bedeu-
tung des frühen Mittelalters. Wien 45–84.
BIVAR, A.D.H. (1972): Cavalry Equipment and Tactics on the Euphrates Frontier.
Dumbarton Oaks Papers 26, 271–291.
BÓNA, I. (1979): A Szegvár-sápoldali lovassír (Das Reitergrab von Szegvár-Sápolda.
Archaeologiai Értesítő 106, 3–32.
BÓNA, I. (1980): Studien zum frühawarischen Reitergrab von Szegvár. Acta Archaeo-
logica Hungarica 32, 31–95.
BÓNA, I. (1988): Die Geschichte der Awaren im Lichte der archäologischen Quellen
Popoli delle Steppe: Unni, Avari, Ungari. Settimane di studio del Centro italiano di
studi sull’alto medioevo, 35, 437–481.
BOTALOV, S. G. – GUTSALOV – S. JU. (2000): Gunno-sarmaty uralo-kazakhstanskikh ste-
pei, Chelyabinsk.
BRATHER, S. (2008): Archaeology and identity. Central and East Central Europe in the
earlier Middle Ages, Bucharest.
COMPARETI, M. – CRISTOFORETTI, S. (2009): The Chinese Scene at Afrāsyāb and the Ira-
nian calendar, (Eurasiatica 78) (Editrice Cafoscarina) Venezia.
CURTA, F. (2001): The Making of the Slavs: History and Archaeology of the Lower Danube
Region, ca. 500–700, Cambridge.
CURTA, F. (2002): From Kossinna to Bromley: ethnogenesis in Slavic archaeology,
in: A. Gillett (ed.): On Barbarian Identity. Critical Approaches to Ethnicity in the
Early Middle Ages. Turnhout, 201–218.
DE BLOIS, F. – VOGELSANG, W. (1993): Dahae, Encyclopaedia Iranica, Originally Pub-
lished: December 15, 1993, Last Updated: November 11, 2011, This article is
available in print. Vol. VI, Fasc. 6, 581–582, and available online at:
https://iranicaonline.org/articles/dahae
ERAMO, I. (2018): Ammiano, l’Historia Augusta e uno strano caso di fornai a caval-
lo. Bollettino di Studi Latini XLVIII – fasc. I- Gennaio-Giugno, 173–182.
ÉRDY, M. (1994): An Overview of the Xiognu Type Caudron Finds of Eurasia in
three Media, with Historical Observations, in: B. Genito (ed.): The Archaeology of
the Steppes. Methods and Strategies. Napoli: Istituto Universitario Orientale, 379–438.
FETTICH, N. (1953): A Szeged-nagyszéksósi hun fejedelmi sírlelet: I-LVIII táblával és két
térképpel (La Trouvaille de tombe princière hunnique a Szeged-Nagyszéksós: avec 58
planches et 2 cartes géographiques), Budapest.
FONJAKOVA, N. A. et al. (1996): Der Schatz von Mala Pereščepina (Malaja Perešče-
pina), in: Reitervölker aus dem Osten - Hunnen + Awaren, Bad Vöslau, 209–224.
GAMBER, O. (1968): Kataphrakten, Clibanarien, Normannenreiter. Jahrbuch der
kunsthistorischen Sammlungen in Wien 64, 1968, 31–32.
GARAM, É. (1990): Bemerkungen zum ältesten Fundmaterial der Awarenzeit, in:
Friesinger, H. – Daim, F. (ed.): Typen der Ethnogenese unter besonderer Berücksich-
tigung der Bayern II, Wien, 253–272.
GARAM, É. (2000): Gürtelverzierungen byzantinischen Typs im Karpatenbecken
des 6.-7. Jahrhunderts. Acta Archeologicae Hungaricae 51, 379–391.
GEARY, P. J. (1983): Ethnic Identity as a Situational Construct in the Early Middle
Ages. Mitteilungen der Anthropologischen Gesellschaft, vol. 113, 15–26.
GENITO, B. (1998): The Achaemenids in the History of Central Asia, in: Gubaev, A.
– Košelenko, G. – Tosi, M. (ed.): The Archaeological Map of the Murghab Delta.
Preliminary Reports 1990-95, Reports and Memoirs Series Minor, III, 149–158.
HARMATTA, J. (1996): The Scythians, in: Hermann, J. – Zurche, E. (ed.): History of
Humanity: From the seventh century B.C. to the seventh century A.D, 3, 181–182.
HERRMANN, G. (1980/81): The Sasanian Rock Reliefs at Bišapūr, Part 1, Iranische
Denkmäler, Iranische Felsreliefis, Ser. F, Berlin; Part 2. Berlin.
KIDD, D. (1995): New Insight into the hoard of 6th-7th century silver from Marty-
novka (Ukraine). Association Française d’Archéologie Mérovingienne, 351–360.
KIDD, D. (1996): Der Silberschatz von Martinivla (Martinovka). Amt der Burgenlän-
dischen Landesregierung, Abt. XII/1-Landesmuseum, 204–208.
KISS A. (1982): Die goldbeschlagenen Schalen der Fürstengräber von Szeged-
Nagyszéksós und Apahida I-II. Folia Archaeologica 33, 163–185.
KOVRIG I. (1973): Hunnischer Kessel aus Umgeburg von Várpalota. Folia Archaeolo-
gica 23, 95–127.
KOVRIG, I. (1985): Das Diadem von Csorna. Folia Archaeologica 36, 107–147.
LASZLÓ, Gy. (1955): Études archéologiques sur l’histoire de la société des avars (Archaeo-
logia Hungarica, XXXIV), Budapest.
LASZLO, Gy. – Rácz, I. (1977a): A nagyszentmiklósi kincs, Budapest.
LÁSZLÓ, Gy. – Rácz, I. (1977b): Der Goldschatz von Nagyszentmiklós, Budapest.
350 BRUNO GENITO
LERNER, A. J. (2005): Aspects of Assimilation: The Funerary Practices and Furnishings of

Central Asians in China (Sino-platonic Papers 168), Philadelphia.
LERNER, A. J. (2011): Zoroastrian Funerary Beliefs and Practices Known from the
Sino-Sogdian Tombs in China”. Silk Road, Vol. IX, 18–25.
LUCY, S. (2005): Ethnic and Cultural Identities, in: Dı´az-Andreu, M. – S. Lucy (ed.:)
The archaeology of identity. Approaches to gender, age, status, ethnicity, and religion.
London/New York, 86-109.
MAENCHEN-HELFEN, Otto J. (1973): The World of the Huns: Studies in Their History
and Culture, ed. by Max Knight. Berkeley – Los Angeles.
MAENCHEN-HELFEN, Otto J. (1978): Die Welt der Hunnen: Eine Analyse ihrer histori-
schen Dimension. Hermann Böhlaus Nachf. Wien–Cologne–Graz.
MARSHAK, B. (2001): La thématique sogdienne dans l’art de la Chine de la seconde
moitié du VIe siècle. Comptes rendus des séances de l’Académie des Inscriptions et
Belles-Lettres, 145e année, n. 1, 227–264.
MAVRODINOV, N. (1943): Le trésor protobulgare de Nagyszentmiklós (Archaeologia
Hungarica, 29), Budapest.
NICHOLSON, O. (2018): Scythians (Saka), The Oxford Dictionary of Late Antiquity,
1346–1347.
NIKONOROV, V. P. (1988) Cataphracti, catafractarii and clibanarii: Another Look at the
Old Problem of their Identification. Military Archaeology. Weaponry and Warfare
in the Historical and Social Perspective. Materials of the International Conference, 2-5
September 1998. St. Petersburg, 131–138.
PACHOS, M. K. (1966): Oboronitel'nye sooruzheniya Afrasiaba, Tashkent.
PÁRDUCZ, M. (1963): Die etnischen Probeme der Hunnenzeit in Ungarn (Studia Archaeo-
logica, 1), Budapest.
PEKARSKAJA, L.V., KIDD, D. (1994): Die Silberschätze von Martynovka (Ukraine) aus
dem 6. Und 7. Jahrhundert (Monographien zur Frühgeschichte und Mittelalterarchäo-
logie), I, Taf., Innsbruck, 45–53.
POHL, W. – M. MEHOFER, ed. (2009): Archaeology of identity – Archäologie der Identität,
Wien, 145–182.
POHL, W. (1988): Die Awaren. Ein Steppenvolk in Mitteleuropa 567 – 822 n. Chr, Mün-
chen.
PRITSAK, O. (1975): The Pechenegs: A Case of Social and Economic Transformation.
Archivum Eurasiae Medii Aevi 1, 211–235.
PRITSAK, O. (1976): The Pečenegs: a case of social and economic transformation, Lisse.
QI DONGFANG (1999): Researches on golden and silver wares of Tang Dynasty, Beijing.
ROSSI, A. V. (2003): Archeologia, storia e filologia a Susa, in: Fontana, M.V. – Geni-
to, B. (ed.:) Studi in onore di Umberto Scerrato per il suo settantacinquesimo com-
pleanno, Series Minor XLV, Napoli, Istituto Universitario Orientale, Istituto Ita-
liano per l’Africa e l’Oriente, 681–700.
REMPEL’, L. I. (1984): Epos v zhivopisi Srednei Azii, Iz istorii zhivopisi Srednei Azii,
Tashkent.
RUSU, M. (1985-86): Tezaurul de la Sînnicolau Mare. Noi puncte de vedere, Anuarul

Institutului de istorie si archeologie 27, 31–66.
SHISHKIN, V. A. (1966) Afrasiab – sokrovishchnica drevnei kul’tury, Tashkent.
ZASECKAJA, P. I. – BOKOVENKO, N. (1994): The Origin of Hunnish Cauldrons in
East-Europe, in: B. Genito (ed.:) The Archaeology of the Steppes. Methods and Stra-
tegies, Napoli, Istituto Universitario Orientale, 701–724.
WERNER J. (1956): Beiträge zur Archäologie des Attila-Reiches, München, Bayerischen
Akademie der Wissenschaften.
WERNER, J. (1984) Der Grabfund von Malaja Pereščepina und Kuvrat, München.

Chronica_2023-21

Uploaded by

Copyright:

Available Formats

Chronica_2023-21

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chronica_2023-21

Uploaded by

Copyright:

Available Formats

CHRONICA

Nomads and their Neighbours in the Middle Ages

Editors: Konstantin Golev – Delyan Rusev

KONSTANTIN GOLEV – DELYAN RUSEV

2ZIMONYI, I. Medieval Nomadic conference in the shadow of the Egyptian revolution.

Prof. Valery Stoyanov

14 ZUMSCHLINGE 1990, 53–4.

II. Pannonia and Valeria

18 ZUMSCHLINGE 1990, 54.

Table 3. The first list of the geographical names in Pannonia

Table 4. The second group of the geographical names in Pannonia 23

23 ZUMSCHLINGE 1990, 57.

Table 5. The third list of the toponyms in Pannonia

Twenty-eight geographical names are not identifiable: Siclis, Ecclavia, Sanigion,

25 ZUMSCHLINGE 1990, 57.

The second group of settlements in the province of Valeria consists of seven

Anonymous of Ravenna additionally mentions that the border of Valeria was

31 ZUMSCHLINGE 1990, 58.

36 ZUMSCHLINGE 1990, 45, 48–9.

 Nazarbayev University & University of Wisconsin-Madison (Emeritus). An earlier version

(SARS-CoV-2). It is a reminder that pandemics have been with us throughout our

4. Cultural and technological regression is a well-known phenomenon for me-

works in the First Phase, 1453–1517).

17 ANTOV 2017, 37, 41–7.

2015; DIMITROV 2016; KOVACHEV 2017.

25 LITTLE 2006, 18–21.

28 STATHAKOPOULOS 2006, 102–3; KOVACHEV 2017. On the Avars see SZÁDECZKY-KARDOSS

33 GOLDEN 1992, 244–58; RÓNA-TAS 1999, 215–28.

ters.38 This would be an example of cultural and technological regression. Of

38 VASILIEV 1958, 181–2.

its inscriptions to the mid-7th to 8th centuries, see RÓNA-TAS 2016.

GOLDEN, P. B. (1980): Khazar Studies. An Historico-Philological Inquiry into the Origins

SCHAMILOGLU, U. (2018b): Reflections on Regional Variation in the Impact of the

6 SUI SHU 1997, 1844–5, 1879–80.

9 SUI SHU 1997, 89, 1582, 1876–9.

12 PAN 1997, 119.

16 PAN 1997, 117, 209.

20 YILDIRIM 2015, 21–39.

a Central Asian vessel, which is often interpreted as a representation of Sasanian or

Saryg-Bulun in Tuva, as well as at Voznesenka on the Dnieper River in Ukraine.5

Figure 2. A Byzantine coin, Bayannuur tomb

Figure 3. Shiveet Ulaan ritual complex

Figure 4. Shiveet Tolgoi sculpture

10 BOYLE 1997, 55‒6.

Table 1. Origin and transcendence

Table 2. Origin from a tree/the tree is a symbol of origin

CLAUSON, G. (1972): An Etymological Dictionary of Pre-Thirteenth Century Turkish,

TOGAN, Z. V. (1939): Ibn Fad̟lān’s Reisebericht, Leibzig.

were collected at different times and, possibly, processed by different authors.

16 MALAMUT 1995, 114.

MAREI, A. V. (2000): Osobennosti sotsialʹno-politicheskoi organizatsii pechenegov,

Naturally, this does not mean—as indeed is unconditionally accepted by a

khagan-bek Joseph from the 950s–960s in KOKOVTSOV 1932.

14 See STOYANOV 2005, 6–9.

15 DIMITROV 1998, 110–120, 235–9.

19 IVANOV 2017, 157.

the extensive bibliography provided in the text.

6 HILENDARSKI 2012, 80–5.

Bulgarian state, see note 45 below.

especially STEPHENSON 1996; STEPHENSON 2000; TŬPKOVA-ZAIMOVA 1976; TŬPKOVA-ZAIMOVA