War Against the Pechenegs (1083–1091)

Meško, Marek

doi:10.1007/978-3-031-26296-8_5

Marek Meško⁵

Part of the book series: New Approaches to Byzantine History and Culture ((NABHC))

82 Accesses

Abstract

This chapter contains an equally comprehensive account of the lesser-known war against the nomadic nation of the Pechenegs, identical in structure to the text in Chap. 4. Similarly, it is introduced by the description of the morphology, climate, and road network in the northeastern Balkans, because such pieces of information are important for the subsequent analysis of the military campaigns conducted in this area by Alexios Komnenos and his subordinate military commanders. In the case of this chapter, the level of detail is sometimes much less elaborate given the limited coverage given to this conflict by the Byzantine and non-Byzantine written sources.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Chapter: USD 29.95; Price excludes VAT (USA)

eBook: USD 99.00; Price excludes VAT (USA)

Hardcover Book: USD 129.99; Price excludes VAT (USA)

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

1.
The Pechenegs were a nomadic tribe that played a role of an important ally of the Byzantines on the Pontic steppe since the end of the ninth century. They have become Byzantium’s troublesome neighbors on the lower Danube after 1025 (see text below). There is no room to reiterate the relationship between Byzantium and the Pechenegs here in full, or to present the historiography covering this interesting topic in its entirety. Fortunately, this work has been very recently carried out in much detail by M. Melnyk. See Mykola Melnyk, Byzantium and the Pechenegs. The Historiography of the Problem (Leiden: Brill, 2022). The Byzantine province of Paradounavon (or Paristrion) was established around 1000 in the area of present-day northern Bulgaria and Romanian Dobrudja between the lower Danube and the Haimos mountains (Stara Planina). Its western border probably passed through the town of Vidin and reached the Black Sea coast in the east. Its military commander (stratēgos, later doux) was stationed in the city of Dristra (now Silistra, Bulgaria). Since the administrative reforms of Basileios II in the 1020s, or since Constantine IX Monomach’s reforms in the 1040s, the province was officially named Paradounavon and geographically called in written sources Paristrion. Kühn, Armee, 223–226; Alexandru Madgearu, “The Military Organisation of Paradounavon,” BSl 60, no. 2 (1999): 421–446; Alexandru Madgearu, Organizarea militară bizantină la Dunăre în secolele X-XII (Tărgovişte: Ed. Cetatea de Scaun, 2007), 101–118; Alexandru Madgearu, Byzantine Military Organization on the Danube, 10th–12th Centuries (Leiden: Brill, 2013); Vasilka Tăpkova-Zaïmova, “L’administration byzantine au Bas Danube (fin du X^e-XI^e s.),” BSl 54 (1993): 95–101; Eugen Stănescu, “Beiträge zur Paristrion Frage. Die Benennungen der unteren Donaugebiete im 10. bis 12. Jahrhundert als historisches Problem,” JÖB 17 (1968): 41–64; Marek Meško, “Vývin obranného systému Byzantskej ríše v 11. storočí – príklad témy Paradounavon,” Byzantinoslovaca I (2006): 128–143; Bojana Krsmanović, The Byzantine Province in Change: (On the Treshold between the 10th and the 11th Century) (Belgrade: Serbian Academy of Sciences and Arts. Institute for Byzantine Studies, 2008), 194–198.
2.
Anna Komnene begins her description of the war against the Pechenegs only after the narration of the war against the Normans, which ends with Guiscard’s death on 17 July 1085. Moreover, the description of the war against the Pechenegs does not immediately follow the description of the Norman war, but in the meantime, Anna Komnene describes the struggles against the Seljuk Turks in Asia Minor, which lasted until 1092. As a result, the order of events gives the impression that the Normans, Turks, and Pechenegs successively attacked the Byzantine Empire. This impression is reinforced by the diction of the opening sentence of the chapter on the war against the Pechenegs, “βουλομένη δὲ δεινοτέραν καὶ μείζονα τῆς προλαβούσης κατὰ τῆς τῶν Ῥωμαίων ἀρχῆς ἔφοδον διηγήσασθαι.” Alexias, VI.14.1 (p. 199). For reasons for this method of reporting historical events, see Howard-Johnston, “Alexiad,” 296–299.
3.
For example, Michael Attaleiates and Anna Komnene use this name. See Attaleiates, 37; Alexias, X.2.6. (p. 286).
4.
María Nystazopoúlou-Pelekídou, “Οι Βαλκανικοί λαοί κατά τους μέσους χρόνους,” (Thessalonica: Ekdóseis Vánias, 1992), 17.
5.
Koder, Γεωγραφία, 28.
6.
One of those fords on the lower Danube was opposite Dervent in Dobrudja. Emil Condurachi, Ion Barnea, and Petre Diaconu, “Nouvelles recherches sur le limes byzantin du Bas-Danube aux X^e et XI^e siècles,” Eight International Congress of Byzantine Studies, Oxford 5–10 September 1966 (Oxford: Oxford University Press, 1967), 190; Diaconu, “Păcuiul lui Soare – Vicina,” Byzantina 8 (1976): 416. It was also possible to cross the Danube near today’s Garvăn and Issacea. Madgearu, Organization, 109.
7.
Diaconu, Petchénègues, 125–126; Madgearu, Organization, 126.
8.
Koder, Βυζάντιο, 28.
9.
Peter Soustal, Thrakien (Thrakē, Rhodopē und Haimimontos), TIB, Band 6 (Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 1999), 54; Nystazopoúlou-Pelekídou, Λαοί, 19; Kyriazópoulos, Θράκη, 51.
10.
Koder, Γεωγραφία, 28.
11.
Soustal, Thrakien, 54.
12.
Soustal, Thrakien, 279. If sufficient troops were available to the Byzantine emperors, the defense of these mountain passes was relatively simple and effective. Haldon, Byzantium at War: AD 600–1453 (London: Osprey, 2002), 11.
13.
Attaleiates, 37.
14.
Veselin Beševliev, “Die byzantinische Heerstrasse Adrianopel – Verigava,” Bulgarisch-byzantinische Aufsätze (London: Variorum Reprints, 1978), 127; Soustal, Thrakien, 141, 143, 441.
15.
Asdracha, Rhodopes, 14.
16.
Ibid.
17.
Soustal, Thrakien, 55; Asdracha, Rhodopes, 15.
18.
See Klaus Belke, “Roads and Travel in Macedonia and Thrace in the Middle and Late Byzantine Period,” Travel in the Byzantine World – Papers from the Thirty-fourth Spring Symposium of Byzantine Studies, Birmingham, April 2000, ed. by Ruth Macrides (Aldershot: Ashgate, 2002), 79–82.
19.
Asdracha, Rhodopes, 15.
20.
Soustal, Thrakien, 55.
21.
See text below.
22.
Asdracha, Rhodopes, 3–4.
23.
Nystazopoúlou-Pelekídou, Λαοί, 20.
24.
Asdracha, Rhodopes, 3, 8–12.
25.
Kyriazópoulos, Θράκη, 53.
26.
For more, see Paul Magdalino, “The Grain Supply of Constantinople, 9th to 12th Centuries,” Constantinople and its Hinterland, ed. by Cyril Mango, Gilbert Dagron, and Geoffrey Greatrex (Aldershot: Variorum, 1995), 36–37.
27.
Koder, Γεωγραφία, 98; Belke, “Roads,” 73–74; Haldon, Warfare, 56; Haldon, Πόλεμοι, 21.
28.
Catherine Asdracha, “La Thrace orientale et la Mer Noire: Géographie ecclésiastique et prosopographique (VIII^e-XII^e siècles),” Géographie historique du monde mediteranéen, ed. by Hélène Ahrweiler (Paris: Presses Universitaires de France, 1988), 255; Annie Pralong, “Remarques sur les fortifications byzantines de Thrace orientale,” Géographie historique du monde mediteranéen, 181; Andreas Külzer, Ostthrakien (Eurōpē), TIB, vol. 12 (Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2008), 684–687.
29.
Külzer, Ostthrakien, 264–267.
30.
Soustal, Thrakien, 223; Asdracha, “Thrace,” 264.
31.
The exact location of this Byzantine city was unknown for a long time; it was assumed that it lay approximately 50 kilometers west of the present-day city of Tekirdağ (in the Byzantine period named Rhaidestos). Asdracha, “Thrace,” 235. However, Ananiev, and most recently Külzer, argues that Apros is today’s Kermiyan. See Ananiev, “Byzantine Seals Found in the Republic of Macedonia,” SBS 7 (2002): 59; Külzer, Ostthrakien, 255–256.
32.
Asdracha, “Thrace,” 236. Village of Rusköy no loger exists. See Külzer, Ostthrakien, 620–622.
33.
Soustal, Thrakien, 330–331; Asdracha, “Thrace,” 237–238.
34.
Koder, Γεωγραφία, 98; Belke, “Roads,” 73; Haldon, Warfare, 54–55; Haldon, Πόλεμοι, 21. See also text and notes 28 and 31 in Chap. 4.
35.
Kyriazópoulos, Θράκη, 54; Haldon, Warfare, 56.
36.
Soustal, Thrakien, 348–349.
37.
Ibid., 143, 271.
38.
This road also represented the shortest overland connection between Constantinople and Dristra. Beševliev, “Heerstrasse,” 126–127; Soustal, Thrakien, 143–144.
39.
Soustal, Thrakien, 239–240.
40.
Koder, Γεωγραφία, 95.
41.
The determination of proper chronology is therefore much more complicated than it was in the case of the war against the Normans, where data from Anna Komnene could be confronted with information from Norman chroniclers, especially William of Apulia. See text above.
42.
See Karl Dieter, “Zur Glaubwürdigkeit der Anna Komnena,” Byz. Zeitsch. 3 (1894): 386–390. The only exception is my article Marek Meško, “Notes sur la chronologie de la guerre des Byzantins contre les Petchénègues (1083–1091),” BSl LXIX (2011): 134–148, as well as Belke’s study, which, however, focuses on events related to the encounters with the Seljuks in Asia Minor. See Klaus Belke, “Byzanz und die Anfänge des rumseldschukischen Staates. Bemerkungen zur Chronologie von Anna Komnēnēs Alexias in den Jahren 1084 bis 1093,” JŐB 61 (2011): 65–79.
43.
Alexias, VII.2.8. (p. 207–208).
44.
Dieter, “Glaubwürdigkeit,” 388–389; Chalandon, Essai, 114. This hypothesis, based on purely astronomical knowledge, was challenged by D’Occhieppo, who believes that the eclipse occurred on 2 October 1084. See K. F. D’Occhieppo, “Zur Identifizierung der Sonnenfinsternis während des Petschenegenkrieges Alexios’ I Komnenos (1084),” JÖB 23 (1974): 182–184. However, the solidly substantiated opinion of this author is difficult to defend from a historical point of view. For example, a widely based invasion of the Pechenegs led by Çelgü and with the participation of the former Hungarian King Solomon, which was unsuccessful and killed both of his leaders, would then have to take place in the spring of the same year. But according to reliable reports of the contemporary analysts Saxo and Bernold, King Solomon died in 1087. See Saxo Grammaticus, Annalista Saxo, MHG SS, VI (Leipzig, 1926), 724; Bernold, Bernoldi chronicon, MGH SS, V (Leipzig, 1926), 446. The date of 20 June 1088, proposed by Petre Diaconu, has no solid basis, and de facto is a wrongly reproduced date from Vasil’evskij, who argues that the day of the solar eclipse was 20 July 1088. See Petre Diaconu, Les Petchénègues au Bas-Danube (Bucarest: Académie de la République Socialiste de Roumanie, 1970), 117, and Vasil’evskij, “Византія и Печенеги,” ZhMNP 164 (1872): 160. This last date must also be dismissed, since, according to more recent astronomical calculations, the solar eclipse of 20 July 1088 was not visible at all in the territory of present-day northeastern Bulgaria. See D’Occhieppo, “Sonnenfinsternis,” 180.
45.
Book VI depicts the Pecheneg invasion in the previous year (spring 1086), during which megas domestikos Gregorios Pakourianos died, and the campaign of the Byzantine general Tatikios during the summer of the same year. See Alexias, VI.14.1–7 (p. 199–202). See also text below.
46.
Events from the beginning of the invasion of the Pechenegs under Çelgü’s command in the spring of 1087 to the defeat of the Paradounavon Pechenegs by their former allies the Kumans during the autumn/winter of 1087. Alexias, VII.1.1. (p. 203–218).
47.
Alexias, VIII.1.1. (p. 236); Vasil’evskij, “Печенеги,” 253; Chalandon, Essai, 128; Paul Gautier, “Diatribes de Jean l’ Oxite contre Alexis I^er Comnène,” REB 28 (1970), 9.
48.
Alexias, VIII.1.5. (p. 238); Vasil’evskij, “Печенеги,” 253; Chalandon, Essai, 128; Gautier, “Diatribes,” 9.
49.
Alexias, VIII.2.4. (p. 239); Vasil’evskij, “Печенеги,” 255; Chalandon, Essai, 128; Gautier, “Diatribes,” 9. Dieter dates this short expedition against the Pechenegs to February 1090, that is, a year earlier. This assumption was not accepted by other researchers. See Dieter, “Glaubwürdigkeit,“ 390.
50.
Alexias, VIII.3.1. (p. 241).
51.
Ibid., 240; Chalandon, Essai, 129. Vasil’evskij provides a wrong date when he indicates 2 March 1091 as the day of the holiday. Vasil’evskij, “Печенеги,” 255.
52.
Alexias, VIII.3.4. (p. 242).
53.
Ibid., 249; Vasil’evskij, “Печенеги,” 283; Dieter, “Glaubwürdigkeit,” 386; Chalandon, Essai, 133.
54.
For references to the coming of the springtime, see Alexias, VII.1.1. (p. 203), VIII.7.1. (p. 220), and VIII.3.4. (p. 242), and of the wintertime VIII.6.6. (p. 220) and VIII.11.6. (p. 235). Dieter notes that Anna Komnene mentions the arrival of winter three times (first sub anno 1086), but I managed to find only two mentions in the text of the Alexiad. See Dieter, “Glaubwürdigkeit,” 387, 390.
55.
Vasil’evskij, “Печенеги,” 156; Dieter, “Glaubwürdigkeit,” 389; Chalandon, Essai, 108; Vasil N. Zlatarski, История на Вългарската държава презъ срђднитђ вђкове. Томъ ΙΙ. Вългария подъ византийско владичество (1018–1187) (Sofia, 1934), 185; Karagiannópoulos, Ιστορία, Γ΄, 47.
56.
Alexias, VI.14.3. (p. 199–200).
57.
See text and note 2 above.
58.
Alexias, VI.4.4. (p. 174), VI.14.2. (p. 200). See text below.
59.
See text and note 187 in Chap. 4.
60.
Alexias, VI.3.2. (p. 172).
61.
Alexias, VI.3.3. (p. 172): “τὰς γὰρ τῶν Περσῶν ἐπελεύσεις καὶ τὰς τῶν Σκυθῶν ἐκδρομὰς”; Arutjunova, “Βзаимоотношения,” 118.
62.
Gautier, Paul. “Le typikon de Grégoire Pakourianos,” REB 42 (1984): 43. See also Arutjunova, “Βзаимоотношения,” 116; Lemerle, “Τypikon,” 172; Peter Frankopan, “A Victory of Gregory Pakourianos against the Pechenegs,” BSl 57 (1996): 279; Stephenson, Frontier, 101; Madgearu, Organization, 137.
63.
Lemerle, “Τypikon,” 173–174; Gautier, “Τypikon,” 19.
64.
See text above.
65.
Frankopan, “Victory,” 280.
66.
Gautier, “Τypikon,” 43: “διὰ τῆς μεγάλης βοηθείας καὶ δυνάμεως τῆς θείας αὐτοῦ δεξιᾶς.”
67.
Arutjunova, “Βзаимоотношения,” 116.
68.
Alexias, V.1.4. (p. 142–143): “τὸ δὲ ἄτερ χρημάτων οὐκ ἐνὴν.”
69.
Gautier, “Diatribes,” 8; Arutjunova, “Βзаимоотношения,” 118.
70.
See text above.
71.
Alexias, V.2.5. (p. 145): “ἐπεὶ καὶ αὖθις ἄλλο νέφος ἐχθρῶν, τοὺς Σκύθας φημί, κατ’ αὐτοῦ ἐξορμῶν ἤδη μεμαθήκοι καὶ διὰ τοῦτο καὶ ἡ συλλογὴ τῶν χρημάτων […] ἐφ’ ὁμοίαις ἀιτίαις ἐσπουδάζετο.”
72.
Gautier dates this passage to the second half of 1087; see Gautier, “Diatribes,” 8–9.
73.
Moreover, this information is organically included in the text. On the contrary, the passage that does not belong chronologically to this section is the description of the trial against John Italos. See note 453 in Chap. 4.
74.
See text above.
75.
See text above.
76.
See text above.
77.
Gautier, “Diatribes,” 43.
78.
Heinrich Hagenmeyer, “Der Brief des Kaisers Alexios I Komnenos an der Grafen Robert I von Flandern,” Byz. Zeitsch. 6 (1897): 19, note 5 in Chap. 1; Chalandon, Essai, 117; Angold, Empire, 136. Yet, Vasil’evskij dates it to 1089; see Vasil’evskij, “Печенеги,” 246, note 2; his view was also taken over by Zlatarski; see Zlatarski, История, 195.
79.
Alexias, VII.6.1. (p. 218). See text below.
80.
See Charles Verlinden, Robert I^er le Frison, comte de Flandre (Paris: Édouard champion, 1932), 151; François L. Ganshof, “Robert le Frison et Alexis I^er Comnène,” Byzantion 31 (1961): 60–61, notes 3 and 4 in Chap. 1; ibid., 61, note 1 in Chap. 1; Michel De Waha, “La lettre d’Alexis I^er Comnène à Robert I le Frison. Une revision,” Byzantion 47 (1977): 123–124.
81.
Ganshof, “Robert,” 61; De Waha, “Lettre,” 123–124; Shepard, “Attitudes,” 103; Belke, “Bemerkungen,” 73.
82.
In September 1089, Alexios Komnenos chaired the church synod held in Constantinople. See Walther Holtzmann, “Die Unionsverhandlungen zwischen Kaiser Alexios I und Papst Urban II im Jahre 1089,” Byz. Zeitsch. 28 (1928): 40, 50.
83.
This has already been argued by Dieter. See Dieter, “Glaubwürdigkeit,” 389. However, it should be noted that this historian dated Alexios Komnenos’ meeting with the Flemish Count to 1085.
84.
Ganshof, “Robert,” 59.
85.
Alexias, VII.6.1. to VII.7.4. (p. 218–222) and VII.9.1. to VII.10.6. (p. 227–236).
86.
In the Alexiad, there is another section that is chronologically confusing in a similar way. It is Book XI describing events in Palestine after the conquest of Jerusalem by the Crusaders until 1105, as well as the relations of Byzantium with the Seljuk Turks before 1113. See Howard-Johnston, “Alexiad,” 291, 294.
87.
Dieter, “Glaubwürdigkeit,” 390; Chalandon, Essai, 119; Paul Gautier, “Le discours de Théophylacte de Bulgarie à l’ autocrator Alexis I^er Comnène (6 janvier 1088),” REB 20 (1962): 96; Paul Gautier, “Défection et soumission de la Crète sous Alexis I^er Comnène,” REB 35 (1977): 217.
88.
Howard-Johnston, “Alexiad,” 294.
89.
A typical example is a description of events related to 1088.
90.
Alexias, VII.6.3. (p. 218).
91.
Ibid.
92.
Ibid. This is the de facto second time that the Kumans are present in the Balkans, since they first arrived in early autumn 1087 as allies of the Paradounavon Pechenegs. Alexias, VII.5.1. (p. 216).
93.
Alexias, VII.5.3. (p. 218).
94.
For the identification of the peace treaty mentioned by Theofylaktos of Ochrid with the first of the two peace treaties concluded with the Pechenegs, see Gautier, “Théophylacte,” 98, 113. This hypothesis was criticized two years later by Ljubarskij in his brief annotation to the Gautier’s publication; see Jakov N. Ljubarskij, “Об источниках ‘Алексиады’ Анны Комниной,” BB 25 (1964): 269–271.
95.
This has been argued by Gautier, the main editor of the primary source; see Gautier, “Théophylacte,” 98–99, 104–105.
96.
Ibid., 98.
97.
Ibid., 114.
98.
Ibid., 99.
99.
Alexias, VII.6.5. (p. 219).
100.
Ibid., VII.6.6. (p. 220).
101.
Ibid.
102.
Ibid.
103.
Ibid., VII.7.1.-2. (p. 220–221).
104.
Ibid., VII.7.4. (p. 221–222). For Emir Abu’l-Kasim’s life, see Claude Cahen, “La première pénétration turque en Asie-Mineure,” Byzantion 18 (1946–1948): 45–51; Claude Cahen, La Turquie pré-ottomane, Varia Turcica 7 (Istanbul: Inst. Français d’Etudes Anatoliennes, 1988), 14–15; Moravcsik, Byzantinoturcica II, 71.
105.
The Flemish knights probably traveled to Byzantium via Apulia, sailed across the Strait of Otranto to Dyrrachion, and from there went all the way to Constantinople along the Via Egnatia. The second alternative route through the Hungarian kingdom, and then via Belgrade, Nish, Sofia, and Adrianoupolis, was not open at the time because of the frequent Pecheneg incursions. A similar route was used five years later by some Crusaders of the First Crusade. See Steven Runciman, A History of the Crusades, vol. I, The First Crusade and the Foundation of the Kingdom of Jerusalem (Cambridge: Cambridge University Press, 1954), 142–144, 166–168; Jean Richard, Histoire des croisades (Paris: Fayard, 1996), 55–56.
106.
See Ganshof, “Robert,” 61; De Waha, “Lettre,” 123–124, and most recently Belke, who believes that the Flemish knights came in the second half of 1090; see Belke, “Bemerkungen,” 74. See also Chalandon, Essai, 125; Vasil’evskij, “Печенеги,” 251; Zlatarski, “История,” 199. Zlatarski advocates this dating for other reasons, specifically because it supports the thesis on dating the speech of Theofylaktos of Ochrid to January 1090. On the other hand, Verlinden argues that the Flemish knights arrived in Constantinople only between 10 February and 20 April 1091. This is based on his assumption that Alexios Komnenos’ letter to Robert of Flanders, written in early 1091, contains a request of the Byzantine emperor to the Flemish count to fulfill his commitment and send the promised military assistance. This means that, at the time of writing the letter, the Flemish knights had not yet been present in the territory of the Byzantine Empire and their arrival was induced only by the aforementioned letter. Verlinden, Robert, 163. However, this assumption does not correspond to the known facts in the Alexiad that, after arriving in Byzantium, the Flemish knights were sent by Alexios Komnenos to Nikomedia for some time, and did not engage in the fight against the Pechenegs. If they had come in the spring of 1091, there would have been no time left for their military activities in Asia Minor against the depredations of the Seljuk Turks.
107.
Clive Foss, “Nicomedia and Constantinople,” Constantinople and its Hinterland, ed. by C. Mango and G. Dagron (Aldershot: Variorum, 1995), 188. Bondoux also claims that Nikomedia was conquered again in 1089/1090 (although he assumes that it fell into the hands of the Seljuk Turks as early as 1085). René Bondoux, “Les villes,” La Bithynie au Moyen Âge, ed. by B. Geyer and J. Lefort, Réalités Byzantines 9 (Paris: P. Lethielleux, 2003), 401. For the question of the conquest of Nikomedia by the Seljuk Turks, see also Cheynet, “Résistance,” 144–145 and Belke, “Bemerkungen,” 74 (Belke assumes that Nikomedia was conquered by the Seljuks in 1086).
108.
Foss, “Nicomedia,” 188.
109.
Alexias, VII.9.1. to VII.11.6. (p. 227–235).
110.
Ibid., VII.8.2. (p. 222–223).
111.
Gautier, “Diatribes,” 13; Gautier, “Défection,” 227.
112.
Chalandon, Essai, 127–128; Vasil’evskij, “Печенеги,” 252–253; Zlatarski, “История,” 200–201.
113.
Alexios Komnenos leaves for the capital at the end of this chapter to spend the winter months there. Alexias, VII.11.6. (p. 235).
114.
Ibid., VIII.1.1. (p. 236).
115.
Ibid.
116.
Chalandon, Essai, 128.
117.
Alexias, VII.11.6 (p. 235). See text below.
118.
The causes of this peculiar lacuna in an otherwise very significant historical work of Anna Komnene were analyzed by Howard-Johnston; see Howard-Johnston, “Alexiad,” 298–300.
119.
A classic example of this approach is Anna Komnene herself, who in one place in her work quite openly admits her reluctance to reproduce barbaric personal names and foreign ethnic groups, unless necessary. Alexias, XIII.6.3. (p. 401): “ὅπου γὰρ βαρβαρικῶν ὀνομάτων ἐξ ἀνάγκης ἀπατεῖται χρῆσης.”
120.
The main reason for this condition is naturally the total absence of written sources of the Pecheneg origin. Alexander Paroń, The Pechenegs. Nomads in the Political and Cultural Landscape of the Medieval Europe (Leiden: Brill 2022), 3. A brief description of the Pecheneg political structure has already been discussed elsewhere; see Marek Meško, “Nomads and Byzantium: Problematic Aspects of Maintaining Diplomatic Relations with the Pechenegs,” On Research Methodology in Ancient and Byzantine History, ed. by J. Bednaříková and M. Meško (Brno: Masaryk University, 2015), 184–186. For more detailed analysis of Pecheneg political and social organization see Paroń, Pechenegs, 132–160.
121.
Hélène Ahrweiler, “Byzantine Concept of the Foreigner: The Case of the Nomads,” Studies on the Internal Diaspora of the Byzantine Empire, ed. by H. Ahrweiler and A. E. Laiou (Washington, DC: Dumbarton Oaks, 1998), 12.
122.
Meško, “Nomads,” 183; Paroń, Pechenegs, 218, 236ff.
123.
See Omeljan Pritsak, “Two Migratory Movements in Eurasian Steppe in the 9th–11th Centuries,” Studies in Medieval Eurasian History (London: Variorum Reprints, 1981), 162–163. Recently, Paroń suggested a more precise time window for the clashes with the advancing Uzes, which he placed after careful analysis to the period between 1019 and 1027. Paroń, Pechenegs, 315.
124.
Peter B. Golden, “Nomads and Their Sedentary Neighbors,” AEMAe 7 (1987–1991): 56, 58; Victor Spinei, The Great Migrations in the East and South East of Europe from the Ninth to the Thirteenth Century (Cluj-Napoca: Romanian Cultural Institute, Center for Transylvanian Studies, 2003), 124–125, 130; Meško, Obnova, 50–51.
125.
The year 1048/1049 mentioned in earlier literature is no longer valid. For the correct dating of the Pecheneg invasion, see Jonathan Shephard, “John Mauropous, Leo Tornicius and an Alleged Russian Army: the Chronology of the Pecheneg Crisis of 1048–1049,” JÖB 24 (1975): 61–89 and Alexander P. Kazhdan, “Иоанн Mавропод, Печенеги и Pусские в середине XIв,” ZRVI 8 (1963), 177–184; Alexander P. Kazhdan, “Once More about the ‘Alleged’ Russo-Byzantine Treaty (ca. 1047) and the Pecheneg Crossing of the Danube,” JÖB 26 (1977): 65–77, as well as Jacques Lefort, “Rhétorique et politique: Trois discours de Jean Mauropous en 1047,” TM 6 (1976): 265–303; Élisabeth Malamut, “L’image Byzantine des Petschénègues,” Byz. Zeitsch. 88 (1995): 118; Paroń, Pechenegs, 325. For the course of the whole attack, see Attaleiates, 30; Skylitzes, 458; Zonaras, 642; Paul Stephenson, “The Byzantine Frontier at the Lower Danube in the Late Tenth and Eleventh Centuries,” Frontiers in Question: Eurasian Borderlands, 700–1700, ed. by Daniel Power and Naomi Standen (Houndmills: Macmillan, 1999), 92; Angold, Empire, 38; Diaconu, Petchénègues, 58; Spinei, Migrations, 133; Paroń, Pechenegs, 326–327.
126.
For more detailed information on the course of the war, see Skylitzes, 465–473, 475–476; Attaleiates, 30–43; Zonaras, 644; Diaconu, Petchénègues, 73–76; Angold, Empire, 38–39; Fine, Balkans, 209–210; Paroń, Pechenegs, 334–341.
127.
Skylitzes, 465–467; Diaconu, Petchénègues, 68–69; Stephenson, Frontier, 93; Madgearu, Organization, 126.
128.
Attaleiates, 43; Angold, Empire, 39; Alexandru Madgearu, “The Pechenegs in the Byzantine Army,” The Steppe Lands and the World Beyond Them, ed. by Florin Curta and Bogdan-Petru Maleon (Iași: Editura Universității Alexandru Ioan Cuza, 2013), 213; Paroń, Pechenegs, 343. Since the Byzantines failed to achieve a decisive victory during the war against the Pechenegs, they attempted to bind the vested interests of the Paradounavon Pechenegs with those of the empire, that is, both Byzantium and the Pechenegs had a common interest in the security of the border on the lower Danube. See Paul Stephenson, “Byzantine Policy towards Paristrion in the Mid-Eleventh Century: Another Interpretation,” BMGS 23 (1999): 56; Stephenson, Frontier, 93; Kyriazópoulos, Θράκη, 124.
129.
Marek Meško, “Pecheneg Groups in the Balkans (ca. 1053–1091) according to the Byzantine Sources,” The Steppe Lands and the World beyond them, 189.
130.
Skylitzes, Synecheia, 166; Attaleiates, 204–205; Meško, “Groups,” 188–189; Stephenson, Frontier, 98–100.
131.
According to Madgearu, the population of the western part of the province did not join the uprising, and the area of northern Dobrudja remained under Byzantine control, with the new administrative center in the ancient Noviodunum (now Isaccea, Romania; Byzantine name of this site is unknown) restored by the Byzantines by the end of the tenth century. The immediate surroundings of the former second Bulgarian capital Preslav and part of the Black Sea coast with a center in Mesembria (now Nesebăr, Bulgaria) also remained in Byzantine hands. Madgearu, Organization, 82–84.
132.
Stephenson, Frontier, 100. One of the most prominent figures was the Pecheneg chief (also having the Byzantine title of exarchon) Tatouch/Tatrys/Chales, who replaced the Byzantine provincial commander (doux/katepano) in Dristra. Attaleiates, 205; Alexandru Madgearu, “The periphery against the Centre: The Case of Paradounavon,” ZRVI 40–41 (2003): 51; Vasile Marculet, “Considérations concernant la fin du pouvoir des Petchénègues du Bas-Danube,” Annales de l’Université Valahia Târgovişte, Section d’Archéologie et d’Histoire 14 (2012): 94, 98. See also Paroń, Pechenegs, 352–353.
133.
Meško, “Groups,” 190. See also Jordanov, Byzantine Seals 1, 141; Madgearu, “Paradounavon,” 123, 125.
134.
For example, the Paradounavon Pechenegs alternately supported various claimants of the imperial throne. Vasil’evskij, “Печенеги,” 147; Malamut, “Ιmage,” 132–133. For instance, these nomads accounted for much of the insurgent forces of Nikeforos Bryennios and their early undisciplined withdrawal from the battle of Kalavrye in March 1078, coupled with the pillaging of Bryennios’ camp, significantly influenced the outcome of the battle. Attaleiates, 290; Haldon, Wars, 129–130; Birkenmeier, Army, 57–59; Cheynet, Pouvoirs, 353. An uprising led by the Paulikian leader Lekas erupted in Philippoupolis in 1078 or 1079 with possible Pecheneg support. Attaleiates, 302; Skylitzes, Synechia, 184; Ioannis Anastasíou, Οἱ Παυλικιανοὶ: Ἡ ἱστορία καὶ ἡ διδασκαλία τῶν ἀπὸ τῆς ἐμφανήσεως μέχρι τῶν νεότερων χρόνων, dissertation (Athens, 1959), 111; Diaconu, Petchénègues, 111; Malamut, “Ιmage,” 134. Another insurgent general prōtoproedros and commander of Dyrrachion, Nikeforos Basilakios, also sought military support from the Paradounavon Pechenegs, sending them letters. Skylitzes, Synecheia, 182–184; Attaleiates, 298–301; Stephenson, Frontier, 101.
135.
See text above.
136.
In his typikon, Pakourianos mentions the Pechenegs and the Kumans as separate enemies. Gautier, “Τypikon,” 43. However, it is very likely that they attacked Byzantium together, as in 1078. Skylitzes, Synechia, 182–184; Attaleiates, 298–301.
137.
It is possible to assume that Alexios Komnenos left at least part of the men from the Makedonian and Thracian tagmata available to Pakourianos, that is, the troops he commanded in all battles since 1081 as megas domestikos. Therefore, the participation of part of the tagmata from Makedonia and Thrace in repulsing the first Pecheneg raid cannot be ruled out, although there is no direct or indirect evidence for it.
138.
Of course, the numerical strength cannot be determined with greater accuracy, as there is no mention of it in the sources. However, prior to Pakourianos, the position of megas domestikos was held by Alexios Komnenos, who commanded a total of about 6500 men at the battle of Kalavrye in 1078. Haldon, Wars, 128. From a certain point of view, the situations of 1078 and 1083 can be regarded as analogous, since in neither case did megas domestikos have access to the main battle units of the Byzantine army.
139.
In almost all cases, where there is a description of military encounters between the nomad and imperial forces in the Byzantine sources, Byzantine historians almost never fail to emphasize the numerical superiority of nomadic opponents. See more in Denis Sinor, “Introduction: the Concept of Inner Asia,” The Cambridge History of Early Inner Asia, ed. by Denis Sinor (Cambridge: Cambridge University Press, 1990), 10; Sinor, “Warriors,” 134; Jonathan Shephard, “Information, Disinformation and Delay in Byzantine Diplomacy,” Byz. Forsch. 10 (1985): 262; Svetlana A. Pletneva, “Печенеги, Торки и Половцы в южнорусских степях,” MIA SSSR 62 (1958): 196; Karasulas, Archers, 55.
140.
Gautier, “Τypikon,” 43: “ἀπὸ τῆς ἁλώσεως τῶν Κομάνων παραγεγονότι”; Lemerle, “Τypikon,” 172.
141.
The financial resources for ransoming Gregorios Pakourianos, as well as for the payment of the tribute to the Kumans, came from the imperial treasury, as mentioned in the text of two golden bulls of the emperor Alexios Komnenos quoted in the typikon. See Lemerle, “Τypikon,” 172–173; Gautier, “Τypikon,” 129. See also Dölger, Regesten, 31, no. 1098.
142.
The lack of coordination of the attacks can be explained by the fact that while the Paradounavon Pechenegs could invade the Byzantine territory at virtually any time, the Kumans did not have such an opportunity because of the long distance of the area of fighting from their settlements. Therefore, both nomadic allies engaged the Byzantines separately (the Kumans probably later than the Pechenegs).
143.
In 1087, the Pecheneg chief Tatouch resorted to a similar solution and decided to negotiate military assistance of the Kumans under the threat of the Byzantine attack. Alexias, VII.3.3. (p. 209); see also the text below.
144.
See note 141 above.
145.
Vasil’evskij, “Печенеги,” 165. See also notes 295 and 528 below.
146.
Gautier, “Τypikon,” 33. The interpretation of this passage clearly depends on the translation of the adjective ἀνίδρυτον. Gautier, the editor of typikon, used the French equivalent turbulent, meaning “turbulent, tumultuous, wild.” Gautier, “Τypikon,” 32. In her recent article, Laiou argues that this expression refers to the Bulgarians; see Angeliki Laiou, “L’étranger de passage et l’étranger privilégié à Byzance,” Byzantium and Other: Relations and Exchanges, ed. by Cécile Morrisson (Abingdon: Routledge, 2012), 77. Other researchers believe that it means the Paulikans (of Armenian ethnic background). See, for instance, Rach Bartikian, “Об «ἀνίδρυτον ἔθνος»-е («неустойчивом народе»), упомянутом в «типике» Григория Пакуриана,” Herald of the Social Sciences 7 (1980): 78–79 (Russian resume of the article written in Georgian language). However, this adjective also means “taking a hostile attitude toward someone.” According to my assumption, the inhabitants of the Philippoupolis area may have been hostile toward the local Byzantine administration on the grounds that it failed to effectively defend them against the incursion(s) of the Paradounavon Pechenegs in 1083.
147.
Pakourianos then founded the Bachkovo monastery on the aforementioned estates. Their inventory can be found in the text of the typikon. Gautier, “Τypikon,” 35 ff.
148.
This important Thracian city and its surroundings were the subject of interest of the Pecheneg raiders also later in 1086 and 1088, when the Paradounavon Pechenegs apparently controlled this city for a short time. See text below.
149.
Alexias, VI.4.2. (p. 174); Anastasíou, Παυλικιανοὶ, 115; Zlatarski, История, 180; Skoulatos, Personnages, 288–289. Traulos apparently entered Alexios Komnenos’ services in 1078 in Philippoupolis, when Alexios commanded the Byzantine troops heading to fend off the Pecheneg raids in the area between Nish and Sardike. Bryennios, 298.
150.
See note 631 in Chap. 4.
151.
Alexias, VI.4.2. (p. 174); Vasil’evskij, “Печенеги,” 155; Chalandon, Essai, 107; Zlatarski, История, 181; Anastasíou, Παυλικιανοὶ, 115; Paroń, Pechenegs, 359.
152.
See Zlatarski, История, 181, no. 1; Chalandon, Essai, 107; Malamut, “Ιmage,” 134. The ruins near the site of Rozovec on the southwestern slopes of the Sărnena Sredna Gora mountains, about 45 kilometers northeast of Plovdiv, are considered to be the remains of Traulos’ fortress. Soustal, Thrakien, 197.
153.
Alexias, VI.4.3. (p. 174); Chalandon, Essai, 107; Anastasíou, Παυλικιανοὶ, 115. It is unclear when Traulos began with his attacks against Byzantine subjects. In my opinion, it was already in the spring of 1084; Chalandon dates his raids to 1085. See Chalandon, Essai, 105.
154.
See text above.
155.
In 1084, the Turkish emir Abu’l-Kasim attacked Byzantine Bithynia and penetrated all the way to the coast of Propontis. See Alexias, VI.9.1 (p. 186); Chalandon, Essai, 108; Vryonis, Παρακμή, 133. Gautier and Belke date this attack to 1086. See Gautier, “Théophylacte,” 103; Belke, “Bemerkungen,” 72–73, 78.
156.
Alexias, VI.4.4. (p. 174); Dölger, Regesten, 33, no. 1120; Anastasíou, Παυλικιανοὶ, 115; Stephenson, Frontier, 101.
157.
The location of Glabinitza remains unknown (not to be confused with Glabinitza in Epirus mentioned in connection with the war against the Normans; see text and note 213 in Chap. 4), but most researchers agree that it was located on the lower Danube near Dristra. Zlatarski, История, 181; Diaconu, Petchénègues, 116.
158.
Alexias, VI.4.4. (p. 174); Vasil’evskij, “Печенеги,” 155; Chalandon, Essai, 107; Anastasíou, Παυλικιανοὶ, 115; Malamut, “Ιmage,” 135. Although Anna Komnene does not give any names of the Pecheneg chieftains, one of the recipients of Traulos’ alliance proposals had to be the aforementioned Pecheneg exarchōn based in Dristra—Tatous. In this context, the above-mentioned find of the bronze seal of a certain Constantine who held the court title bestarchēs and was the commander of the Byzantine fleet (droungarios ton ploimon) is interesting. This seal was found directly in Dristra and, due to its characteristics as well as the command rank, can be dated prior to 1085. Jordanov, Corpus 3, 379–380. Of course, it is not possible to determine what this Constantine discussed with the Pechenegs in Dristra around 1085, but it can be imagined that his letter contained peace proposals of the Byzantine emperor, who in this way may have tried to discourage the Pechenegs from their planned alliance with Traulos.
159.
See note 134 above.
160.
Alexias, VI.4.4. (p. 174); Chalandon, Essai, 107; Zlatarski, История, 182; Malamut, “Ιmage,” 135; Stephenson, Frontier, 101.
161.
Chalandon, Essai, 108.
162.
The Pechenegs probably crossed the Haimos mountains via the Shipka pass (Shipchenski prohod, 1190 meters above sea level), located approximately 36 km northeast of the hypothetical location of Traulos’ fortress near Beliatoba.
163.
Alexias, VI.14.3. (p. 200); Vasil’evskij, “Печенеги,” 155; Chalandon, Essai, 109; Zlatarski, История, 185. For Nikolaos Branas’ career, see Skoulatos, Personnages, 252–253. In 1081, Branas was appointed the second in command (hypostratēgos) by megas domestikos himself and his task was to remain in Adrianoupolis during Pakourianos’ long absence caused by his involvement in the fight against the Normans. Alexias, VI.4.1. (p. 123).
164.
Alexias, VI.14.3. (p. 200).
165.
However, there exists one chrysobull of Alexios Komnenos issued for the monastery of Megisti Lavra on Mount Athos in May of 1086 that contains a list of mercenary units and their ethnic origin in Byzantine service around this time. In this we find the Rus’, Varangians, Koulpings, English (Anglo-Saxons), Franks (Normans), Germans, Bulgarians, and Saracens (possibly Seljuk Turks). Some of them could be transferred under Pakourianos’ command. See Actes de Lavra, p. 258: „Ῥῶς, Βαράγγων, Κουλπίγγων, Ἰγγλίνων, Φράγγων, Νεμίτζων, Βουλγάρων, Σαρακηνῶν.”
166.
This figure is based on Cheynet’s assumption that both tagmata had a total of up to 10,000 men during the battle of Kalavrye in March 1078. However, it should be noted that the overall estimate also includes units of Thracian infantry. See Cheynet, Pouvoirs, 67, 353, note 29. During the battle of Dyrrachion in 1081, their numerical strength is estimated at about half the previously estimated number (5000 men). Haldon, Wars, 134. However, this estimate is not contrary to the above figure, as the battle of Dyrrachion may not have involved all the tagmata in full force, but only their subdivisions. It should also be taken into account that these troops suffered casualties during the war against the Normans in the battles of Dyrrachion (1081), Joannina (1082), and Larissa (1083), but, due to the lack of information in the written sources, it is impossible to quantify these losses more precisely.
167.
The hostilities between Byzantine forces under the command of megas primikerios Tatikios and the Seljuks of Abu’l-Kasim took place in Bithynia. Alexias, VI.9.1. to VI.13.4. (p. 186–199); Chalandon, Essai, 101–101; Gautier, “Théophylacte,” 101; Belke, “Bemerkungen,” 72–73, 78.
168.
Alexias, VI.14.3. (p. 200).
169.
Ibid. Interestingly, Gregorios Pakourianos’ attitude perfectly corresponded to the measures recommended for similar situations by the Byzantine general Kekaumenos in his Strategikon. See Kekaumenos, 89, 91.
170.
Alexias, VI.14.3. (p. 200). Attacks of heavy cavalry formations represented the basis of the Byzantine army’s combat tactics in the late tenth and eleventh centuries. Kühn, Armee, 128.
171.
Alexias, VI.14.3. (p. 200); Chalandon, Essai, 109; Vasil’evskij, “Печенеги,” 156; Stephenson, Frontier, 101; Madgearu, Organization, 137; Paroń, Pechenegs, 360.
172.
Alexias, VI.14.3. (p. 200); Vasil’evskij, “Печенеги,” 156; Zlatarski, История, 185.
173.
Chalandon, Essai, 109.
174.
This apparently happened at the end of spring of 1086. Gautier, “Théophylacte,” 103.
175.
Alexias, VI.14.4. (p. 200). The criticality of the situation is evidenced by the fact that Alexios Komnenos withdrew Tatikios from Asia Minor where he was engaged in a struggle against the Seljuks, and even some military units were redeployed with him to reinforce weakened troops in Thrace at the expense of the defense of Bithynia. Belke, “Bemerkungen,” 72–73. For Tatikios’ earlier career, see note 276 in Chap. 4.
176.
Alexias, VI.14.4. (p. 200–201); Stephenson, Frontier, 101.
177.
Anna Komnene notes the visible relief shown by Tatikios when his predominantly infantry units were joined by experienced Norman cavalrymen. Alexias, VI.14.4. (p. 201).
178.
Alexias, VI.14.4. (p. 201). Nōbelissimos (this supreme court title was used ca. since 1085; see Jordanov, Corpus 2, 312–314; Jordanov, Corpus 3, 491) and doux Constantine Houmbertopoulos was an experienced commander, who had already demonstrated his qualities during the war against the Normans (he fought in the battle of Dyrrachion). See also note 272 in Chap. 4.
179.
Due to the fact that only in July 1085 did the war against the Normans end and a number of Guiscard’s men did not return to southern Italy after the duke’s death, but joined the ranks of the Byzantine army (especially after the conquest of Kastoria in November 1083 and then in July 1085; see text above and note 740 in Chap. 4), it is almost certain that the name Φράγγοι/Κέλτοι, used in this part of the Alexiad, refers to these men.
180.
Alexias, VI.14.4. (p. 201); Vasil’evskij, “Печенеги,” 156; Belke, “Bemerkungen,” 72–73; Madgearu, Organization, 137.
181.
The information on the size of the gathered army is in Anna Komnene’s work contradictory at best. On the one hand, the Byzantine princess notes that Tatikios’ army was strong enough (“ἱκανὸν … στράτευμα”), and, on the other hand, she claims that Tatikios did not consider his force adequate to successfully face the more numerous Pechenegs in a direct pitched battle. Alexias, VI.14.4. (p. 201) and VI.14.7. (p. 202).
182.
The size of 400 to 500 men was a common average in the newly formed units of the Norman and other Latin (i.e., Latin West) mercenaries during the second half of the eleventh century. See Haldon, Warfare, 104, as well as Cheynet, “Rôle,” 117 and Cheynet, “Effectifs,” 323–324, and also related literature found therein. In my personal consultations with Haldon and Cheynet, both scholars repeatedly confirmed this estimate.
183.
Various Norman units serving in Byzantium during the eleventh century were highly praised by their employers, and thanks to their heavy cavalry equipment became the spearhead of all the Byzantine armies in the West and in the East since the 1040s. It is only natural that in the critical situation after the defeat at Beliatoba, Norman cavalry tagma was sent by the emperor to stem the Pecheneg tide. For the eminent role of the Normans in Byzantine army, see Jonathan Shepard, “The Uses of the Franks in Eleventh-Century Byzantium,” Anglo-Norman Studies 15 (1993): 276; Georgios Theotokis, “Rus, Varangian and Frankish Mercenaries in the Service of the Byzantine Emperors (9th–11th c.): Numbers, Organisation and Battle Tactics in the Operational Theatres of Asia Minor and the Balkans,” Vyzantina symmeikta 22 (2012): 143ff; Christos Markrypoulias, “‘Our Engines are Better than Yours’: Perception and Reality of Late Byzantine Military Technology,” Byzantium and the West: Perception and Reality (11th – 15th c.), ed. by Nikolaos G. Chrissis, Athina Kolia-Dermitzaki, and Angeliki Papageorgiou (London: Routledge, 2019), 307.
184.
In the text describing Tatikios’ encounter with the Pechenegs, which will be analyzed below, Anna Komnene suggests the division of the Byzantine combat formation into three sections (the right and left wings and the center). Alexias, VI.14.7. (p. 202). According to Byzantine military customs, each section ideally consisted of at least one taxiarchia of infantrymen, that is, a section of 1000 men (see note 55 in Chap. 2).
185.
Vasil’evskij, “Печенеги,” 157.
186.
Chalandon, Essai, 109.
187.
Alexias, VI.14.5. (p. 201).
188.
The river on the bank of which the Byzantines built the camp is probably the Sazlijka river. Blisnos was within three days march west of Adrianoupolis, with its exact location uncertain. Researchers locate it somewhere between today’s Bulgarian towns of Galǎbovo and Simeonovgrad. Soustal, Thrakien, 210; Zlatarski, История, 185.
189.
Anna Komnene states that Tatikios himself spotted the Pechenegs. This assertion does not seem to be true, otherwise the Pechenegs and the Byzantines would have seen each other already when marching, and the subsequent skirmish would have taken place quite differently. Rather, Anna Komnene offers a simplified description of events with noticeable shortcuts. The spies and scouts were a permanent and indispensable component of the Byzantine army while on a campaign; besides monitoring the opponent’s movements, they were to look for the best places for camping. See Haldon, Warfare, 150, 152.
190.
Alexias, VI.14.5. (p. 201); Vasil’evskij, “Печенеги,” 156.
191.
This is a typical example of surviving Roman military traditions. Another is, the building of a marching camp, as well as the above-mentioned division of the army into the part that builds the camp and the part that guards the baggage train and the builders. A detailed description of the procedure for building a camp can also be found, for example, in the Byzantine military manual Anonymou biblion taktikon, which is, considering the time of its creation (around 1020), one of the manuals closest to the events described and analyzed here. See Dennis, Treatises, 243, 247 ff.
192.
Alexias, VI.14.5. (p. 201).
193.
Ibid., Zlatarski, История, 186.
194.
The dispersal of warriors into small mobile units when moving through the enemy territory was a typical feature of the steppe warfare. It was advantageous in several ways, since, on the one hand, it allowed even a large number of nomads to be able to live off the local resources and plunder larger territory, thus increasing the quantity of spoils obtained, and, on the other hand, it confused enemy forces in terms of main direction and target of the nomadic attack, while allowing for a highly effective monitoring of the enemy movements. If the nomads came across a larger enemy host, their scattered groups could join and attack shortly. This tactic was perfected by the Mongols in the thirteenth century. See Karasulas, Archers, 53; George T. Dennis and Ernst Gamillscheg, Das Strategikon des Maurikios in Mauricii Strategicon, intro., ed. and index by George T. Dennis, CFHB, vol. XVII (Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 1981), 362.
195.
There was a similar situation during the battle between the Byzantines and the Pechenegs at Diakene in 1049. The then Byzantine commander attacked the nomads without allowing his exhausted soldiers to rest, which resulted in a disastrous defeat of the Byzantine army. See Skylitzes, 468–469, 475–476, and, in particular, Kekaumenos, 93, 95.
196.
Alexias, VI.14.5. (p. 201).
197.
For a contrasting assessment of this skirmish, see Stephenson, Frontier, 101.
198.
Alexias, VI.14.6. (p. 201); Zlatarski, История, 186; Vasil’evskij, “Печенеги,” 156. Tatikios’ measure fully agrees with the advice given for a similar situation by Kekaumenos. Kekaumenos, 53, 55.
199.
Alexias, VI.14.6. (p. 201); Malamut, “Ιmage,” 136.
200.
Ibid.
201.
The claim about pitching the camp immediately after crossing the river is based on Anna Komnene’s information that after the whole day of waiting, the army returned to the (already built) encampment. See Alexias, VI.14.7. (p. 202).
202.
Alexias, VI.14.7. (p. 201–202); Zlatarski, История, 186; Vasil’evskij, “Печенеги,” 156.
203.
Alexias, VI.14.7. (p. 202).
204.
Exactly the same arrangement is recommended to military commanders in the anonymous military manual known as the Strategikon of Maurice, written at the beginning of the seventh century (probably between 592 and 602) when facing the “Scyths” (the nomadic peoples of the steppe). See Dennis and Gamillscheg, Strategikon, 366.
205.
Alexias, VI.14.7. (p. 202).
206.
“Scythian” formation is referred to in the Strategikon of Maurice. See Dennis and Gamillscheg, Strategikon, 218 (reference to the use of this type of combat formation in the Byzantine army), 362 (description of its characteristic features). The division of the nomadic (Pecheneg, or Kuman) forces into smaller groups of warriors is also mentioned in the late eleventh-century written accounts from Kievan Rus’, which state that each such group carried its own combat standard or sign. Pletneva, “Печенеги,” 197–198. For information on Arabic sources describing the Pecheneg combat tactics, see also András Pálóczi Horváth, Petschenegen, Kumanen, Jassen: Steppenvölker im mittelalterichen Ungarn (Budapest: Corvina, 1989), 17–18.
207.
Alexias, VI.14.7. (p. 202); Vasil’evskij, “Печенеги,” 156–157; Zlatarski, История, 186.
208.
Alexias, VI.14.7. (p. 202).
209.
Alexias, VI.14.7. (p. 202); Vasil’evskij, “Печенеги,” 157; Zlatarski, История, 187.
210.
Alexias, VI.14.7. (p. 202); Zlatarski, История, 187; Paroń, Pechenegs, 360. There are several reasons to assume that the Pecheneg troops were following exactly this route. First, it was the shortest way from the surroundings of Philippoupolis to the Sidera pass and it led through flat terrain, which would certainly facilitate the Pechenegs to transport large volumes of spoils and captives. Second, the Pecheneg commanders probably wanted their troops to live off the resources from the Byzantine territory for as long as possible on the way back to their settlements in Paradounavon.
211.
Alexias, VI.14.7. (p. 202); Vasil’evskij, “Печенеги,” 157; Chalandon, Essai, 109–110; Zlatarski, История, 187.
212.
Alexias, VI.14.7. (p. 202); Vasil’evskij, “Печенеги,” 157; Chalandon, Essai, 110; Zlatarski, История, 187.
213.
Anchialos, lying on the spur protruding into the waters of the Gulf of Burgas, represented a significant hub of overland and maritime communications during the Byzantine period and was therefore of great military as well as economic importance. Soustal, Thrakien, 175–177. See also text below.
214.
Ibid., 355–359.
215.
Ibid., 454–456.
216.
Ibid., 234–235.
217.
Alexias, VI.9.6. (p. 188); Kühn, Armee, 168; Kyriazópoulos, Θράκη, 242; Belke, “Bemerkungen,” 67. This idea was further developed by Madgearu, who assumes that a new administrative-military unit was created in early 1087 in connection with the fending off of the great Pecheneg incursion (see text below). Madgearu, Organization, 85. My assumption of its establishment in 1086 is based on the fact that it was a purely defensive measure and, therefore, logically occurred after the repulsion of the first serious Pecheneg attacks taking place in the second half of 1086. If the thema of Anchialos had been created in early summer 1087, its establishment would have had to be rather offensive in nature (supporting the major Byzantine attack against the Paradounavon Pechenegs; see also text below) on an ad hoc basis, since the great Pecheneg invasion in the spring of 1087 was, in my opinion, unpredictable (see text below). At the same time, the establishment of a new thema would have taken place only shortly before (or at the same time as) the campaign itself, which is not very logical if it were to effectively support the Byzantine war effort north of the Haimos mountains. Anchialos later proved its worth and exceptional strategical significance during the Kuman invasion in 1095 (see text below).
218.
Madgearu, Organization, 85.
219.
The first commander of the newly formed province (thema) of Anchialos was a Turk named by Anna Komnene as Siaous (Çavuş), who, as an envoy (çavuş) of the Seljuk ruler of Damascus, Tutuş (1078–1092), respectively his older brother Sultan Malikshah (1072–1092), defected to the Byzantines and received baptism, and eventually enabled Alexios Komnenos to regain control of the north coast of Asia Minor with the important port of Sinope. Alexias, VI.9.6. (p. 188) and VI.12.1. (p. 194); Moravcsik, Byzantinoturcica ΙΙ, 274; Kyriazópoulos, Θράκη, 242; Soustal, Thrakien, 176; Belke, “Bemerkungen,” 67; Brand, “Element,” 4–5; Skoulatos, Personnages, 281.
220.
There is no doubt about the dating of this invasion to spring 1087, as it precedes the aforementioned and firmly dated solar eclipse that took place on 1 August 1087. See text and note 44 above.
221.
The hostilities were limited only to the area north of Philippoupolis and its surroundings, and the numbers of defenders and attackers were not too high either. The same applies to the scale of military operations carried out. Further evidence of the relatively restricted nature of the conflict is the fact that even though Alexios Komnenos considered the situation critical after the defeat at Beliatoba, he did not assume command of the units fighting off the Pechenegs, as was the case during the key campaigns of the war against the Normans.
222.
Alexias, VII.1.1. (p. 203). These nomads were distant relatives of the Pechenegs south of the Danube who, after the great Pecheneg invasion of 1046/1047, decided to remain in the territory north of the Danube. For more detailed information on this Pecheneg group in the late eleventh century, see Meško, “Groups,” 194–196.
223.
Alexias, VII.1.1. (p. 203). See also Moravcsik, Byzantinoturcica ΙΙ, 311; Paroń, Pechenegs, 361.
224.
For possible reasons for Solomon’s participation in this expedition and his fate between 1074 and 1087 elsewhere, see Marek Meško, “Pečenežsko-byzantské dobrodružstvo uhorského kráľa Šalamúna (1083–1087),” Konštantínove listy 4 (2011): 77–94.
225.
See also text and note 118 above.
226.
Today’s Turkish town of Hayrabolu. Asdracha, “Thrace,” 250; Külzer, Ostthrakien, 308–310.
227.
Alexias, VII.1.1. (p. 203). See Chalandon, Essai, 112; Diaconu, Coumans, 36.
228.
This was first argued by Dieter, “Glaubwürdigkeit,” 387, and later by Madgearu, “Paradounavon,” 430.
229.
Apart from a few mentions in the Alexiad, Gregorios Maurokatakalon is a less-known military commander. See Skoulatos, Personnages, 111–112. Besides, the sigillographic material associated with his person published so far is not very rich and does not cover his entire cursus honorum. According to the latest findings, he seems to have been the strategos of the northern part of Paradounavon (i.e., those territories that were still under Byzantine control after 1072). Several lead seals from the lower Danube region with the titles of patrikios, anthypatos, and katepanō seem to confirm this assumption, although there is no name of the province inscribed on them. See Madgearu, “Paradounavon,” 83, as well as Costel Chiriac, “Un nouveau sceau de Grégoire Mavrokatakalon découvert à Oltina (départ. de Constanta),” Études byzantines et post-byzantines 4 (2001): 113–121. However, it has been wrongly believed until recently that these seals should be dated back to the earlier period (the 1050s or 1060s). For this, see Jordanov, Byzantine Seals 2, 280–281.
230.
See Alexias, VII.2.3. (p. 205); Dieter, “Glaubwürdigkeit,” 387.
231.
Madgearu, “Paradounavon,” 430.
232.
The only Byzantine fortress on the lower Danube, the destruction of which is probably linked to the turbulent events of the 1080s, is Garvăn (ancient Dinogetia; its Byzantine name is unknown), lying on the bend of the Danube, where its watercourse turns from the north and runs eastward to the Black Sea. Perhaps it was due to the proximity of the Pechenegs in the spring of 1087 that the buildings outside the perimeter of the walls of this fortress were abandoned. Madgearu, “Paradounavon,” 106, 132. Therefore, it is possible to imagine the situation that in the spring of 1087 the Pecheneg host crossed the Danube somewhere nearby and continued further south, attacking the existing Byzantine enclaves around Issacea and Nufărul (east of Garvăn), where the center of the Byzantine administration was located (see text above), or in the southwest direction toward today’s Oltina, where the seal of Gregorios Maurokatakalon was uncovered (probably attached to a letter in which its inhabitants and garrison troops were warned of the approaching Pechenegs). See note 229 above.
233.
This hypothesis was formulated by Madgearu, “Paradounavon,” 430. See notes 131 and 229 above.
234.
Madgearu, “Paradounavon,” 430, note 35 in Chap. 2; Jordanov, Byzantine Seals 2, 282–283; Jordanov, Byzantine Seals 3, 647.
235.
Madgearu, “Paradounavon,” 430.
236.
This hypothesis is based on the fact that the seals were part of the letters, or orders, that Maurokatakalon sent from his command post in Isaccea. Their purpose may have been to warn of an impending Pecheneg invasion, or to give instructions to the Byzantine garrisons stationed at the sites where the seals were uncovered. Alexios Komnenos used a similar strategy only a few years later, in 1095, during the Kuman raid. See text below. In connection with these findings, the discovery of the seal of megas domestikos Adrianos Komnenos in Isaccea should also be mentioned. Although the seal is dated to the period between 1087 and 1105, Barnea does not rule out that the find may be related precisely to the Pecheneg invasion of 1087, or the Kuman invasion of 1095. See Ion Barnea, “Sceaux byzantins du Nord de la Dobroudja,” RESEE 23, no.1 (1985): 31–32. Since Adrianos Komnenos apparently fell into disgrace with his brother Alexios Komnenos (see text below) after the failed 1094 conspiracy of Nikeforos Diogenes, it can be assumed that the seal is related to the invasion of the Pechenegs in 1087, thus representing further probable evidence of communication (originally containing orders, dispatches, or warnings) between the supreme commander of all the Byzantine units in the Balkans and his subordinate (doux/katepanō) in the remaining Byzantine outpost in Paradounavon.
237.
If this assumption is correct, the nomads led by Çelgü effectively sought to bypass the territory of the supposedly newly formed province of Anchialos. See text above.
238.
For a basic overview of this province, which was created by merging the separate provinces of Makedonia and Thrace in the first half of the eleventh century, see Kyriazópoulos, Θράκη, 195–199.
239.
See Kühn, Armee, 206–207. For an overview of the city’s rich history during the eleventh century, see Soustal, Thrakien, 163, and also text and note 164 in Chap. 6. Unfortunately, the list of military commanders of the province during this period is still far from complete.
240.
Alexias, VII.1.1. (p. 203). The location of this site is unknown; see Külzer, Ostthrakien, 654. However, it is very likely that Skoteinon was located to the west of the nearby town of Charioupolis, that is, on the road between this town and the town of Pamfylos.
241.
Alexias, VII.1.1. (p. 203); Vasil’evskij, “Печенеги,” 158.
242.
Byzantine troops were unlikely to move to the area of fighting from the Byzantine capital, as the city of Adrianoupolis (the main base of the tagma of Thrace and Makedonia) was closer. Moreover, if the Byzantine army were to arrive from Constantinople, it would approach the Pechenegs from a different direction, from the northeast or east, not from the northwest.
243.
Alexias, VII.1.1. (p. 203). Nikolaos Maurokatakalon was probably one of the commanders of the Byzantine troops stationed in Thrace (theoretically, he may have been the doux of the entire province, but this is not evidenced in the period sources, nor can this presumption be confirmed by the existing sigillographic material uncovered so far). A little later (based on Anna Komnene), he may have held a post under megas domestikos Adrianos Komnenos similar to that of Nikolaos Branas before him under Gregorios Pakourianos, that is, he was his aide, deputy, and second-in-command. See text below and also Skoulatos, Personnages, 256–257. Several lead seals of this Byzantine general dating back to the last quarter of the eleventh century have also been preserved, but none, as already stated, confirms the above assumptions (these seals are from his later life, when Maurokatalon no longer held a military position). See Jordanov, Byzantine Seals 2, 283. It is also likely that he was somehow related to the general Gregorios Maurokatakalon (based on the same last name), but there is no definite proof of this.
244.
Alexias, VII.1.1. (p. 203). We have very little data about this Byzantine commander, not knowing even his first name. His last name is usually derived from the location of Mempetz or Bempetz (ancient town of Hierapolis) in the Euphrates basin (today’s Manbij, Syria). This is also the only mention of him in the entire historical work of Anna Komnene. Skoulatos, Personnages, 46. Based on the lead seal finds, a link can be established between this Byzantine commander and a certain Theodoros Bempetziotes, whose seal, without specifying a rank, dating back to the late eleventh century or the first quarter of the twelfth century, is displayed in the Numismatic Museum of Athens. See Christos Stavrakos, “Korrekturen zu Lesung einiger Siegel des Numismatischen Museums Athens,” SBS 2 (1990): 39–47.
245.
Alexias, VII.1.1. (p. 203); Vasil’evskij, “Печенеги,” 158; Diaconu, Petchénègues, 117; Diaconu, Coumans, 36. The town of Pamfylos is believed to have been located near today’s town of Uzunköprü near the Ergene river. Asdracha, “Thrace,” 253; Külzer, Ostthrakien, 560–562.
246.
Alexias, VII.1.1. (p. 203); Vasil’evskij, “Печенеги,” 158; Zlatarski, История, 189.
247.
The actual site of the fort of Koule is unknown, but historians believe that it corresponds to today’s location of Chelebiköy, located about 22 km southeast of the port of Enez (in Byzantine, Ainos). However, Soustal rightly observes that this belief is in principle unfounded and the identification of the Koule castle with Chelebiköy is practically an expression of willfulness, rather than an assumption based on an analysis of historical data. Soustal, Thrakien, 328.
248.
See Vasil’evskij, “Печенеги,” 158; Chalandon, Essai, 113. The hypothesis of these two historians was also adopted by Diaconu. See Diaconu, Petchénègues, 117.
249.
Alexias, VII.1.1. (p. 203).
250.
The distance between the two locations could not have been great (up to 10 km), as Anna Komnene notes that a small stream was running between them. Alexias, VII.1.2. (p. 204).
251.
Nikolaos Maurokatakalon, an experienced commander, must have known that he was most likely observed from the heights by the Pecheneg scouts. For this, see Alexias, VII.1.1. (p. 203).
252.
Alexias, VII.1.2. (p. 204).
253.
Alexias, VII.1.2. (p. 204).
254.
Ibid., 203–204. Perhaps Maurokatakalon had the recent fateful deaths of Gregorios Pakourianos and Branas in his mind, and did not want to commit the same errors.
255.
Ibid., 204. This experienced commander had previously fought in a series of armed encounters against Bohemund’s Normans at Larissa in 1083; see text and note 520 in Chap. 4. Kourtikios may have found himself in the ranks of the troops facing the Pechenegs as a military counsellor sent by Alexios Komnenos from Constantinople, in the company of his younger brother Adrianos Komnenos. See note 265 below.
256.
Ibid., 204.
257.
Ibid., 203.
258.
This assumption is based on the theory of the existence of three Pecheneg groups in the Balkans at the end of the eleventh century (two north of the Danube and one in the province of Paradounavon), which I have discussed elsewhere. See Meško, Obnova, 144–149; Meško, “Groups,” 188–197. According to this theory, Çelgü directly controlled only about half of all the Pechenegs living north of the Danube, while the second chief Kutesk, who for some time provided refuge to the deposed Hungarian king Solomon and, therefore, is referred to in the Chronicon Pictum, controlled the rest of the territory. However, Kutesk’s Pechenegs did not take part in this expedition. At the same time, I have voiced the assumption that between seven and nine Pecheneg tribes could have been settled beyond the Danube (minus a tribe that moved to Paradounavon between 1074 and 1077). Meško, “Groups,” 185. The total number of warriors of these two nomad groups could have been between 60,000 and 80,000. My estimate is based on Pritsak’s assumption, according to which one tribe = one tümen = 10,000 warriors. See Omeljan Pritsak, “The Pečenegs: A Case of Social and Economic Transformation,” Studies in Medieval Eurasian History (London: Variorum Reprints, 1981), 21. Thus, if Çelgü had half of this total at his disposal, there could have been between 30,000 and 40,000 men in his host.
259.
Alexias, VII.1.1. (p. 203): “καὶ ἀπὸ τοῦ δακικοῦ στρατεύματος οὐκ ὀλίγους, ὧν ὁ οὕτω καλούμενος Σολομὼν δημαγωγὸς ἦν.” For the identification of the Hungarians under the archaic ethnonym Dacians in the contemporary Byzantine chronicles, see Peter Rokai, “Дачани као име Маћара у византијским изборима,” ZRVI 38 (1999–2000): 230; Vasilka Tăpkova-Zaïmova, “L’emploi des ethnica et les problèmes de la communicaton à Byzance,” in Επικοινωνία στο Βυζάντιο – Πρακτικά του Β΄ διεθνούς συμποσίου, 4-6 Οκτωβρίου 1990, ed. by Nikos G. Moschonas (Athens: Kéntro vyzantinnon erevnon/Ethnikó Ídryma, 1993), 705.
260.
A similar impression of a small number of men accompanying Solomon arises from the description of this expedition in the Chronicon Pictum. See Alexander Domanovszky, “Chronici hungarici compositio sæculi XIV,” in Scriptores rerum hungaricarum, Vol. I, ed. by E. Szentpetery (Budapest: Academia Litter. Hungarica, 1937), 408, 410.
261.
Alexias, VII.1.1. (p. 203).
262.
See text above.
263.
The inclusion of this detachment in the Byzantine battle order in the spring of 1087 is based on the assumption that the defeat suffered by the tagmata of Makedonia and Thrace under Pakourianos’ leadership was not as severe as it might seem from Anna Komnene’s account, and that some of the soldiers and their mounts survived the slaughter at Beliatoba unscathed. Of course, defeated tagmata that suffered severe casualties could not have been battle-ready for Tatikios’ campaign in the autumn of 1086, but some of their subunits could already be able to take part in the fighting during the spring of 1087.
264.
This hypothesis is based on Anna Komnene’s brief mention about soldiers who returned to the capital after the battle against the Pechenegs. Alexias, VII.1.2. (p. 204). If soldiers of units stationed near Constantinople are really referred to in this passage, this information provides further support for my other theory that the Pecheneg attack was not as surprising as described, and that the Byzantines had time to react and take the necessary defensive measures (again, according to the analogy to the Kuman invasion in 1095).
265.
Alexias, VII.1.2. (p. 204). Adrianos Komnenos previously played a dangerous role in the battle of Larissa against the Normans in 1083 (see text above).
266.
Alexias, VII.1.2. (p. 204); Vasil’evskij, “Печенеги,” 158; Chalandon, Essai, 113; Diaconu, Petchénègues, 117; Spinei, Migrations, 142; Stephenson, Frontier, 102; Paroń, Pechenegs, 361.
267.
Chronicon pictum, 409: “Cum enim vidissent Cuni magnam multitudinem loricatorum, timuerunt valde ceperuntque festinare (…) priusquam ab hostibus conluderentur.”
268.
Alexias, VII.1.2. (p. 204); Madgearu, Organization, 139.
269.
Solomon’s death in 1087 is confirmed by brief mentions in western Latin sources. See Saxo, 724; Bernold, 446; Vasil’evskij, “Печенеги,” 158. See also note 44 above.
270.
During the eleventh century, the Byzantine border on the Danube was perceived by the social elite in Constantinople not only as a political boundary between Byzantium and its northern neighbors, but also as the northern boundary of Orthodox Christianity. Such a notion of a dividing space did not exclude the establishment of new ethnicities in the Byzantine territory, but new arrivals were required to embrace Christianity in its Orthodox form, which, in its consequences, represented a form of integration. See Stephenson, “Lower Danube,” 97–98.
271.
In her description of the events of spring 1087, Anna Komnene focused only on the military clashes of the main forces. See Alexias, VII.1.1. (p. 203–204). However, following the defeat of the Pecheneg horde led by Çelgü, she mentions very briefly that the Pechenegs plundered the adjacent territory in Makedonia and around Philippoupolis during their retreat and even after that continued in their inroads. Since this territory was quite far from the main area of military operations (between the town of Pamfylos and the Koule fort on the lower Maritsa), it could have been the Paradounavon Pechenegs who took advantage of the incursion of their relatives from beyond the Danube. See Alexias, VII.2.1. (p. 204).
272.
Stephenson assumes that Çelgü and his host passed through their territory with the consent of the leaders of the Paradounavon Pechenegs. Stephenson, Frontier, 102.
273.
Alexias, VII.2.1. (p. 204); Vasil’evskij, “Печенеги,” 158–159; Chalandon, Essai, 113.
274.
For this dating of the start of the campaign, see note 276 below.
275.
Alexias, VII.2.1. (p. 204); Vasil’evskij, “Печенеги,” 259; Chalandon, Essai, 113; Zlatarski, История, 190; Diaconu, Petchénègues, 117. Anna Komnene says that Lardea was located between the towns of Goloe (today’s Lozarevo) and Diampolis (today’s Jampol), but its more precise location is unknown. Identification of Lardea with today’s site of Lozevec is purely hypothetical, though highly likely. See Soustal, Thrakien, 333.
276.
This dating is derived from the fact that, according to Anna Komnene, at least forty days passed between the arrival of the Byzantine army in Lardea and the arrival of the Pecheneg envoys on 1 August 1087. See Alexias, VII.2.2. (p. 204). In order to reach Lardea in mid-June, the Byzantine army, which also included cavalry units, would have had to start their march from Constantinople as early as the end of May or at the latest in early June 1087. As a basis for this estimate, I used data from the Arabic geographer Idrísí, who mentions various routes in Thrace, along with information concerning the distance between individual locations in miles or days of march. For example, the route between Arkadioupolis and Tzouroulon (roughly 45 km as the crow flies) corresponds in his work to forty miles, or two days of march. See Idrîsî, Géographie, 407. Some researchers argue that one day of march equals 15 to 30 km. See, for example, Veselin Beševliev, “Zur Geographie Nord-Ost Bulgariens in der Spätantike und im Mittelalter,” Bulgarisch-Byzantinische Aufsätze (London: Variorum Reprints, 1978), 69. Haldon provides a figure derived from the military manual De velitatione of the emperor Nikeforos II Fokas stating that one day of march corresponded to a distance of 16 miles (24 km). John Haldon, “The Organization and Support of an Expeditionary Force: Manpower and Logistics in the Middle Byzantine Period,” Byzantium at war (9th–12th c.), ed. by K. Tsiknakis (Athens: Goulandri-Horn Foundation, 1997), 122, note 42. Above, in connection with the transfer of the Byzantine army to Dyrrachion in autumn 1081 (see note 310 in Chap. 4), I mentioned the possibility of observing the late ancient marching standards, that is, iter iustum (10 Roman miles = ca. 15 km) and iter magnum (15 Roman miles = 22.2 km). Dimitroúkas, “Ενδείξεις,” 16. Based on this data, the march from Constantinople to Adrianoupolis could have taken the Byzantine army eight to nine days and from Adrianoupolis to Lardea at least four days, meaning that the Byzantine military forces would need at least two weeks to move from Constantinople via Adrianoupolis (with a possible one-day stop to gather additional troops from the Adrianoupolis area) to Lardea. If the Byzantine army consisted only of cavalry units (which of course was not the case), it could have moved twice as fast, as evidenced later in March 1206, when a detachment of approximately 140 knights led by the emperor Baldouin of Flanders (1204–1207) covered the distance between Constantinople and Adrianoupolis in just three days. Geoffroi de Villehardouin, La conquête de Constantinople, ed. and transl. by Edmond Faral, vol. II (Paris: Les Belles Lettres, 1961), 158, 160; Asdracha, Rhodopes, 48–49.
277.
Alexias, VII.2.1. (p. 204). Georgios Euforbenos is one of the relatively little-known Byzantine commanders and appears here in the historical narration of Anna Komnene for the first time. Skoulatos, Personnages, 94–95. The Byzantine princess naturally does not mention either his rank or court title, and since none of his lead seals have been uncovered, it can only be assumed that this Byzantine commander held the position of komēs tou stolou usually reserved for a squadron commander. For this, see Böhm, Flota, 219.
278.
Alexias, VII.2.1. (p. 204) and VII.2.7. (p. 207); Vasil’evskij, “Печенеги,” 159; Chalandon, Essai, 113–115; Zlatarski, История, 190; Diaconu, Petchénègues, 117. Byzantine Dristra is today’s city of Silistra, Bulgaria, originally ancient Durostorum/Dorostolon (the Slavic or contemporary Bulgarian name of the town is Drăstăr). The fortification of the acropolis consists of walls erected in the early ninth century during the reign of the Bulgarian khan Krum (803–814). After the Byzantines recaptured this important fortress in 1002, a second acropolis was built in the eleventh century in the southern part of the city. Dimitar Angelov and Boris Cholpanov, Българска военна история от втората четъврт на X до втората половина на XV в (Sofia: Izdavatelstvo na Balgarskata akademija na naukite, 1989), 25; Madgearu, Organization, 112. After the establishment of the province of Paradounavon during the reign of Basileios II Bulgaroktonos, Dristra became the headquarters of its military commander (stratēgos and since the mid-eleventh century doux/katepanō). Krsmanović, Province, 194–195, 198.
279.
See note 277 above.
280.
For my estimate of losses in November 1084, see above. A visible manifestation of this reorganization, initiated after the end of the war against the Normans, was also the creation of a new command rank of doux tou stolou (or after 1092, megas doux) by the emperor Alexios Komnenos in 1085 or 1086, which relegated the original commander of the Byzantine navy (droungarios tou stolou, later megas droungarios) to the second position in the command hierarchy. See Alexias, XII.8.8. (p. 381); Böhm, Flota, 217; Guilland, “Marine,” 220. The first Byzantine fleet commander with this rank was probably Manuel Boutoumites. For his appointment, see Alexias, VI.10.5. (p. 190). For Manuel Boutoumites’ person as well as his later rich career in Alexios Komnenos’ service, see Skoulatos, Personnages, 181–185.
281.
In line with my previous estimates and after counting the projected losses, this could amount to approximately twenty dromons. See text above.
282.
See note 280 above.
283.
Alexias, VI.10.5. (p. 190–191); Belke, “Bemerkungen,” 68; Böhm, Flota, 140–141, 222.
284.
Alexias, VII.2.1. (p. 204); Vasil’evskij, “Печенеги,” 159. The emphasis on additional training of all divisions just before the expedition may point to the fact that there was a large number of fresh recruits in the ranks of the Byzantine army, for whom this campaign was to be the baptism by fire.
285.
Alexias, VII.2.2. (p. 205).
286.
Ibid., VII.2.2. (p. 204).
287.
Ibid., VII.2.2. (p. 205). Nikeforos Bryennios the Elder is one of the important figures in the Byzantine history of this period. It is sufficient to say that he participated in the fateful campaign of the emperor Romanos IV Diogenes of 1071, which ended in the Byzantine defeat at Manzikert. During the uprising against the emperor Nikeforos III Botaneiates, he crossed arms with the then domestikos of the West Alexios Komnenos at the battle of Kalavrye and was defeated. Subsequently, he was blinded and stripped of his property, although some of his assets were later returned to him. After Alexios Komnenos became the emperor, he was awarded with elevated court titles of nōbelissimos and later of prōtonōbelissimos. See Seibt, Bleisiegeln, 288–289. His expertise in military matters was so highly regarded by Alexios Komnenos that Bryennios, despite his blindness, accompanied the emperor in the campaign against the Pechenegs as his personal advisor. See Skoulatos, Personnages, 218–224.
288.
Alexias, VII.2.3. (p. 205). See note 243 above.
289.
Ibid. See also text above.
290.
Ibid.; Chalandon, Essai, 114; Vasil’evskij, “Печенеги,” 159; Zlatarski, История, 190.
291.
It should be reiterated that, even when addressing this dilemma, it was Alexios Komnenos who ultimately had the decisive say. His strong determination to resolve the Pecheneg problem once and for all is also reflected in the opening sentence, which, according to Anna Komnene, her father said to the military council. See Alexias, VII.2.2. (p. 204): “‘οὐ χρή’, λέγων, ‘ἐκεχειρίαν ὅλως τοῖς Σκύθαις δίδοσθαι’.”
292.
As stated above, Anna Komnene often avoids using personal names of those considered by her barbarians. The same applies to the Pecheneg (or other) rulers in Paradounavon. Besides Tatu, we know the name of a chief called Satzas settled in Bitzina located somewhere in the Danube delta (also probably of nomadic origin) and another leader named Časlav (Anna Komnene calls him Sesthlavos, which indicates the Slavic/Bulgarian origin). Alexias, VI.14.1. (p. 199); Madgearu, “Paradounavon,” 132–133.
293.
The main operational base of the Byzantine squadron could, according to Madgearu, be Isaccea (ancient Noviodunum) with the residence of katepanō still controlling the Byzantine enclaves in Paradounavon. In connection with the 1087 campaign against the Pechenegs, it is worth noting that the lead seal of the emperor’s younger brother megas domestikos Adrianos Komnenos was uncovered at this site. See Madgearu, “Paradounavon,” 83–84, 104–105, 138.
294.
Alexias, VII.2.7. (p. 207).
295.
The Kumans and their settlement areas in the late eleventh century will be mentioned in more detail later. See note 528 below.
296.
Alexias, VII.3.3. (p. 209); Vasil’evskij, “Печенеги,” 165; Petre Diaconu, Les Coumans au Bas-Danube aux XI^e et XII^e siècles (Bucarest: Académie de la République Socialiste de Roumanie, 1978), 38. In this context, a question arises: Why did Tatu undertake such a long way to the Kumans, who were also ethnically related to the Pechenegs, but nonetheless to a lesser degree than the Pechenegs living north of the Danube? In my opinion, this only demonstrates that the Byzantine victory in the spring of 1087 over the Pechenegs led by Çelgü living north of the Danube was absolute—the Pechenegs living beyond the Danube were weakened militarily to such a degree that their allies living south of the Danube could no longer find any support from them and had to seek new allies at the much more distant Kumans.
297.
Alexias, VII.2.7. (p. 207); Vasil’evskij, “Печенеги,” 159–160; Chalandon, Essai, 114; Diaconu, Petchénègues, 117; Malamut, “Ιmage,” 136; Paroń, Pechenegs, 362. It was apparently an attempt to promise a return to the status quo before 1072, when parts of the Paradounavon province, under the rule of the Pecheneg chiefs, became independent and the terms of the 1053 peace treaty were nullified.
298.
Alexias, VII.2.8. (p. 207); Paroń, Pechenegs, 362.
299.
Ibid., VII.2.9. (p. 208); Vasil’evskij, “Печенеги,” 160; Chalandon, Essai, 114; Diaconu, Petchénègues.
300.
Alexias, VII.2.9. (p. 208). Leo Nikerites was an eunuch, but some sources present him as an experienced and courageous military commander. Enough data is known about his career, both from historical and sigillographic sources. In the 1070s, he held the title of bestarchēs and a little later, at the turn of the 1070s and 1080s, he was a proedros. By 1087, he became prōtoproedros and anagrafeus of Pelopones and after the final victory of the Byzantines over the Pechenegs in 1091 he was awarded the court title of kouropalatēs and at the same time became the commander (doux) of the restored and unified province of Paradounavon. See Skoulatos, Personnages, 179–180; Ivan Jordanov, “The Katepanate of Paradounavon according to the sphragistic Data,” SBS 8 (2003): 67–68; Jordanov, Byzantine Seals 2, 308–309; Valentina S. Shandrovskaja, “Некоторые исторические деятели ‘Алексиады’ и их печати,” Palestinskij sbornik 23 (1971): 41–42; Madgearu, “Paradounavon,” 84.
301.
Alexias, VII.2.9. (p. 208).
302.
For this date see note 304 below.
303.
Little Nikaia, or Nike, was located on the route between Adrianoupolis and Boulgarofygon in Thrace (today’s Babaeski); today, the Turkish city of Havsa is located there. Soustal, Thrakien, 374–375
304.
Alexias, VII.2.9. (p. 208). At this point, Anna Komnene’s text contains contradictory pieces of information, because the reader first gets the impression that the Byzantine army was still encamped in Lardea when Alexios learned of the escape of the Pecheneg prisoners, and that the start of the campaign was actually the direct result of this unfavorable news (Alexias, VII.3.1. (p. 208)). However, on the basis of the same text, Leo Nikerites personally informed the emperor of the failure of his mission in Goloe (Ibid., VII.2.9. (p. 208)). This means that the military expedition against the Pechenegs began independently of the outcome of Leo Nikerites’ mission, and that the Byzantine units were already on the march, but probably only for one day, because Goloe is only about 24 km away from Lardea. Based on the fact that the bad news was reported to the emperor by Leo Nikerites himself, that the distance between Lardea and Little Nikaia was so short, and that Nikerites and his captives traveled on horseback, we can assume that the entire mission, including the escape of prisoners and the return of Nikerites from Little Nikaia to the emperor in Goloe, could have taken him approximately six days. Finally, if Leo Nikerites were to set off with the captives as early as 1 August 1087, he would inform the emperor of the prisoners’ escape some six days later, on 7 August. If these assumptions were correct, then it would be hypothetically possible to determine the date of departure of the Byzantine army from Lardea, and thus the start of the whole campaign against the Pechenegs on 6 August 1087.
305.
However, according to the hypotheses described above, the Byzantine army was still, at least in theory, marching on the Byzantine territory, as these enclaves in Paradounavon remained under Byzantine administration even after Nestor’s uprising. See text above.
306.
Not to be confused with the town of Bitzina, which was probably located somewhere near the mouth of the Danube. Madgearu, “Paradounavon,” 135–136. The Ticha river, or the Golyama Kamchija in present-day northeastern Bulgaria, has been reliably identified as Bitzina (in Bulgarian apparently as Dičina). Beševliev, “Geographie,” 69; Zlatarski, История, 192; Madgearu, “Paradounavon,” 134.
307.
Alexias, VII.3.1. (p. 208–209); Vasil’evskij, “Печенеги,” 161; Zlatarski, История, 192.
308.
Alexias, VII.3.1. (p. 209); Zlatarski, История, 192. The route along which the imperial army advanced from Adrianoupolis to Dristra is also indicated by the imperial seals and depots of gold Byzantine coins found on the sites of Golyam Izvor, Melnitsa, Zlati Voyvoda, Preslav, Kirkovo, Drandar, Vodno, Gurgendzhik, Ishirkovo, and Păcuiul lui Soare. Madgearu, “Paradounavon,” 138.
309.
Alexias, VII.3.2. (p. 209). This Danube tributary cannot be further identified. The distance noted by Anna Komnene of 24 stadia is equivalent to 4.8 km, provided that one ancient στάδιον equals 200 meters. See Malamut, “Ιmage,” 137; Madgearu, “Paradounavon,” 138. However, this distance is too short, considering that the total length of the route between Dristra and Pliska is about 90 km. So, if the Byzantine army were to reach a point about 4.8 km from Dristra after one day of march, it would have to march an incredible distance of 85.2 km. However, as explained elsewhere (see note 313 in Chap. 4), Anna Komnene seems to have used the ancient στáδιον in the meaning of the Roman mile, which was used as a unit to measure distance even in the Byzantine period. In this case, the distance of 24 stadia (in fact Roman miles) would suddenly increase to 35.52 km. Yet, despite this alteration, Byzantine soldiers would still have to cover a considerable distance of around 55 km in one day. Also, the 36 km distance of the camp from Dristra is at best illogical, and perhaps, after all, the Byzantine camp was located less than 5 km from the city. Based on these facts, it can be assumed that Anna Komnene, in her detailed description of the route of the Byzantine army, failed to count at least one complete day, and that in order to cover the distance between Pliska and Drista, Byzantine soldiers needed not two but at least three or more days of marching.
310.
Alexias, VII.3.2. (p. 209); Cheynet, Pouvoirs, 392; Birkenmeier, Army, 72.
311.
Alexias, VII.3.2. (p. 209); Vasil’evskij, “Печенеги,” 161; Zlatarski, История, 193.
312.
Chalandon, Essai, 116.
313.
See note 278 above.
314.
In 971, the Byzantine camp was probably located on the same hill during the campaign of the emperor John I Tzimiskes against the Prince Svyatoslav of Kievan Rus’ (964–972). See Haldon, Wars, 123.
315.
Alexias, VII.3.3. (p. 209–210).
316.
Malamut, “Ιmage,” 137; Alexias, VII.3.3. (p. 210). In my older article, I assumed that the Byzantine army retreated from Dristra westward. Meško, “Výprava,” 113, note 42. This assumption was based on the fact that the Vetren fortress on the bank of the Danube, mentioned by Anna Komnene, is located west of Dristra. The emperor sent his imperial tent and baggage train under the leadership of Georgios Koutzomites the evening before the battle to Vetren, apparently to board the ships of the imperial squadron. See Alexias, VII.3.6. (p. 211). I wrongly supposed that the rest of the army followed suit later. For more information on the Vetren fort, see Madgearu, “Paradounavon,” 112–113. However, recent archaeological findings clearly show that the Byzantines actually withdrew from Dristra to the south (as Vasil’evskij assumed, albeit for a wrong reason; see Vasil’evskij, “Печенеги,” 162) and encamped in the wider surroundings of today’s Dulovo (located approx. 25 to 30 km southwest of Dristra), that is, somewhere between Dulovo and Dristra. Bulgarian archaeologists at this site uncovered several sets of objects indicating the exact location of the Byzantine camp as well as the battle itself, mainly objects of a military nature (various types of weapons, parts of horse harnesses and cavalry equipment, etc.), as well as the complete equipment of the “field” blacksmith’s forge, which apparently belonged to the baggage train of the Byzantine army. See Joto Jotov and Georgi N. Nikolov, “Походъ на Алексий I Комнин към Дръстър (1087 г.): Нови данни и интерпретация,” Пътуванията в средневековна България. Материали от първата национална конференция Пътуване към България, пътуванията в средневековна България и съвременният туризъм Шумен, 8.-11. 5. 2008 г. (Veliko Trnovo, 2009), 438. Unfortunately, the data in this study is only general and does not reveal the exact location of the site. See also Valery Yotov, “The Traces of the Presence of Scandinavian Warriors in the Balkans,” Byzantium and the Viking World, ed. by Fedir Androshchuk, Jonathan Shephard, and Monica White (Uppsala: Uppsala Universitet, 2016), 252.
317.
Alexias, VII.3.4. (p. 210); Chalandon, Essai, 116; Diaconu, Petchénègues, 118; Zlatarski, История, 193. The proposal of Georgios Palaiologos and Gregorios Maurokatakalon does not necessarily imply that Byzantine soldiers would have to conquer the city upon arrival under the walls of Preslav. Conversely, this more or less automatic assumption by most historians that Preslav was in the hands of the Pechenegs in the summer of 1087 seems mistaken, since the lifting of one siege (Dristra), the subsequent march through enemy territory, and, finally, the initiation of a second siege (Preslav) would represent, from a military point of view, a sequence of incomprehensible maneuvers that would worsen rather than improve the position of the Byzantine army. In this context, the verb katalambanein means to “reach” or “enter,” as had already been the case in the aforementioned transfer of the Byzantine army from Constantinople via Adrianoupolis to Lardea, during which the emperor “καταλαμβάνει τὴν Ἀδριανούπολιν,” that is, reaches the city of Adrianoupolis. Adrianoupolis was in Byzantine hands at the time, so the verb here certainly does not mean to “seize” or “conquer.” See Alexias, VII.2.1. (p. 204). There are several such examples in the text of the Alexiad. This detail supports Madgearu’s argument about the Byzantine presence in Preslav before the summer of 1087. According to his hypothesis, Preslav was then under katepanikion with the seat in Mesembria (today’s Nesebăr). See Madgearu, “Paradounavon,” 429. Similar conclusions are made on the basis of archaeological findings in Preslav itself, where no archaeological evidence of interruption of the Byzantine presence during the eleventh century has yet been found. See Peter Frankopan, “The Working of Byzantine Provincial Administration in the 10th–12th Centuries: the Example of Preslav,” Byzantion 71 (2001–2002): 96.
318.
Soldiers were strictly forbidden to make fires or light wooden beams. Alexias, VII.3.6. (p. 211).
319.
Although Anna Komnene does not describe it literally like this, only in this way can we explain the emperor’s actions at night before the battle, the apparent aim of which was to prevent a hostile night attack. See Alexias, VII.3.6. (p. 211). Therefore, it can be assumed that the Pechenegs had already been in close proximity to the Byzantine camp.
320.
This dating is only indicative and valid if the Byzantine army actually left Lardea on the morning of 6 August 1087. See note 304 above. As a result, it would be possible to date the siege of Dristra to approximately 12 and 13 August, and finally the battle of Dristra would take place on the day mentioned above, that is, on Saturday, 14 August 1087.
321.
Alexias, VII.3.6. (p. 211); Zlatarski, История, 194.
322.
The emperor’s personal guard consisted of six members—both brothers Leo and Nikeforos Diogenai, Nikolaos Maurokatakalon, Basileios Kourtikios, Nampites, who was the commander of the Varangian guard, and finally Goules, one of the long-time servants of the imperial family. See Alexias, VII.3.6. (p. 211); Vasil’evskij, “Печенеги,” 162; Zlatarski, История, 194; Birkenmeier, Army, 73.
323.
Alexias, VII.3.6. (p. 211).
324.
Ibid.; Vasil’evskij, “Печенеги,” 162. The aforementioned Adrianos Komnenos was appointed to this command post to fight the Pechenegs after the death of megas domestikos Gregorios Pakourianos in 1086. Skoulatos, Personnages, 5–8. A seal on which Adrianos Komnenos is titled prōtosebastos and megas domestikos pasēs dyseōs has been preserved. Georges Zacos and Alexander Veglery, Byzantine Lead Seals, vol. 1, part 3, no. 1497–1500. For sigillographic material related to his person, see Jordanov, Byzantine Seals 1, 79–80; Jordanov, Byzantine Seals 2, 218–220; Jordanov, Byzantine Seals 3, 211, 366; Nesbitt and Oikonomides, Catalogue, 6–7.
325.
Their presence in the center of the Byzantine formation stems from the above episode, when they fought in the immediate vicinity of the emperor during one of the Pecheneg attacks on the Byzantine camp. See note 310 above.
326.
Anna Komnene does not mention the Varangian guard as such during the expedition; she only mentions the presence of its commander Nampites. However, the real presence of at least part of the Varangian guard was unambiguously confirmed only recently by archaeological finds directly on the battlefield, where weapons (battle axes, swords, spears, lances, etc.) or other objects (buckets, forgings, blacksmith’s utilities, ceramics, etc.) of Scandinavian origin were uncovered. Jotov and Nikolov, “Дръстър,” 438; D’Amato and Rava, Guard, 36, 39, 45.
327.
Alexias, VII.3.6. (p. 211); Vasil’evskij, “Печенеги,” 162; Birkenmeier, Army, 73.
328.
Alexias, VII.3.6. (p. 211); Vasil’evskij, “Печенеги,” 162; Birkenmeier, Army, 73. For more on Niketas Kastamonites, see Skoulatos, Personnages, 249–250. The course of his career can be traced from the preserved lead seals. Kastamonites held the honorary title of bestarchēs until 1094, and then he began to move up the court hierarchy. The highest rank he achieved was that of prōtokouropalatēs. See Jordanov, Byzantine Seals 2, 196–198.
329.
The assumption that the right wing of the Byzantine army consisted of Makedonian and Thracian units is based on the fact that it was commanded by Tatikios, who seems to have been in charge of this tagma even earlier (in the summer of 1086). See Birkenmeier, Army, 73. As for the left wing, the tagma from Thessaly may have been under the command of Nikeforos Melissenos based on the fact that he was the doux of Thessalonica, the then administrative center of the Byzantine Thessaly. By virtue of his status, he was able to put the unit on alert and march with it to Adrianoupolis, where he met the rest of the Byzantine army led by the emperor.
330.
Vasil’evskij, “Печенеги,” 162; Zlatarski, История, 194. Birkenmeier assumes that these were light cavalry units of unknown origin, trained to fight in the fashion of the nomadic peoples. Birkenmeier, Army, 73.
331.
Alexias, VII.3.6. (p. 211). In the course of this campaign, Argyros Karatzas commanded the units of the ethnikoi, that is, groups of foreign warriors (in this case, the Uzes) settled on the Byzantine territory. Both Karatzas and Ouzas came from the nomadic peoples of the Sarmatians (Sauromats), as Anna Komnene anachronically names them, who can be quite safely identified as the Turkish Uzes. See text above. The career of this nomad warrior in Byzantine services is known in rough contours thanks to the findings of his lead seals. Until 1092, Argyros Karatzas held a fairly exalted court title of kouropalatēs. See Jordanov, Byzantine Seals 2, 188–190; Jordanov, Byzantine Seals 3, 175. See also text below.
332.
Alexias, VII.3.6. (p. 211). Ouzas had previously distinguished himself in the battle of Larissa in 1083 against the Normans; see note 590 in Chap. 4.
333.
It is clear from the calculation of the Byzantine troops that could have taken part in the battle of Dristra that Alexios Komnenos managed to assemble the majority of the Byzantine forces for the expedition against the Pechenegs. When we compare the composition of the Byzantine army at Dristra in August 1087 with the army at Dyrrachion in 1081, we can conclude that, except for the Exkoubitai and the Bestiaritai (2000 men), the Armenian infantry, and the Seljuk allies (2000 men), all other troops from the battle of Dyrrachion also took part in the battle of Dristra. Since the size of the Byzantine army at the battle of Dyrrachion is estimated between 18,000 and 20,000 men (see text above), it is very likely that, this time, the Byzantine army was about 15,000 men strong.
334.
Alexias, VII.3.7. (p. 211); Malamut, “Ιmage,” 137.
335.
Alexias, VII.3.7. (p. 211).
336.
Alexias, VII.3.8. (p. 212). The nomads traveled over the steppe on these two- to four-wheeled wagons, which they used mainly as their homes. Many were so large that their decks could hold round ger tents, often mistakenly referred to as yurt. Pletneva, “Печенеги,” 203. For more details on ger tents, see Karasulas and McBride, Archers, 16–17. It is possible that the Pecheneg wagons carried these tents also during the battle of Dristra. A detailed description of such large wagons laden with tents can be found in the work of the Arab geographer and traveler Ibn-Battûta from the mid-fourteenth century: “Les habitants de cette contrée les apellent ‘arabahs, et ce sont des chariots dont chacun est pourvu de quatre grandes roues. Il y en a qui sont traînés par deux chevaux, ou même d’avantage; des bœufs et des chameaux les traînent également, selon la pesanteur ou la légèreté du char. (…) On place sur le chariot une espèce de pavillon, fait de baguettes de bois liées ensemble avec des minces lanières de cuir. Cette sorte de tente est très legère, elle est recouverte de feutre ou de drap, et il y a des fenêtres grillés (…) Ceux de ces chariots qui portent les bagages, les provisions de route et les magasins de vivres sont recouverts d’ un pavillon pareil, fermant par une serrure.” Ibn Battûta, Voyages, II, de la Mecque aux steppes russes et à l’ Inde (Paris: La Découverte, 1997), 205–206.
337.
Alexias, VII.3.8. (p. 212); Vasil’evskij, “Печенеги,” 162; Zlatarski, История, 194; Malamut, “Ιmage,” 137. The presence of women and children in the nomadic host does not necessarily mean a migratory movement. See Shepard, “Disinformation,” 262. Women and children were usually tasked with taking care of horses and other domestic animals also during combat operations. Karasulas and McBride, Archers, 55; Dennis and Gamillscheg, Strategikon, 362.
338.
Alexias, VII.3.7. (p. 211–212). The shape of the Pecheneg wagon fort was mostly circular or oval. Birkenmeier, Army, 72.
339.
From the description by the Byzantine princess, it can be assumed that when Alexios saw the Pecheneg formation from afar, he observed it carefully and made several last-minute modifications to the structure of the Byzantine combat formation in order to face the impending onslaught of the nomads more effectively. Yet, the nature of these changes is unknown. See Alexias, VII.3.7. (p. 212). See also Birkenmeier, Army, 73.
340.
Alexias, VII.3.8. (p. 212); Vasil’evskij, “Печенеги,” 162; Birkenmeier, Army, 74.
341.
Alexias, VII.3.8. (p. 212); Vasil’evskij, “Печенеги,” 162; Zlatarski, История, 194.
342.
Alexias, VII.3.8. (p. 212); Vasil’evskij, “Печенеги,” 162; Diaconu, Petchénègues, 119.
343.
Alexias, VII.3.8. (p. 212); Vasil’evskij, “Печенеги,” 162.
344.
Alexias, VII.3.8. (p. 212); Vasil’evskij, “Печенеги,” 162.
345.
Anna Komnene reports that these reinforcements were 36,000-strong, which is clearly exaggerated. Alexias, VII.3.8. (p. 212); Vasil’evskij, “Печенеги,” 162; Chalandon, Essai, 116; Zlatarski, История, 194; Diaconu, Petchénègues, 119. Interestingly, this figure is the only direct numerical reference concerning the size of the opposing force at the battle of Dristra.
346.
Alexias, VII.3.8. (p. 212); Vasil’evskij, “Печенеги,” 162; Birkenmeier, Army, 74.
347.
Alexias, VII.3.9. (p. 212). Prōtostratōr Michael Doukas was the brother of Alexios Komnenos’ wife, the empress Eirene. For his person and career, as well as the findings of lead seals associated with this period of his career, see Skoulatos, Personnages, 202–205; Bulgakova, Bleisiegel, 102–105; Demetrios Polemis, The Doukai: A Contribution to Byzantine Prosopography (London: Athlone Press, 1968), 63–66. He also took part in the battle of Larissa; see text and note 521 in Chap. 2.
348.
Alexias, VII.3.9. (p. 212); Vasil’evskij, “Печенеги,” 162; Diaconu, Petchénègues, 119; Paroń, Pechenegs, 362.
349.
Alexias, VII.3.12. (p. 214); Vasil’evskij, “Печенеги,” 162–163; Chalandon, Essai, 116; Malamut, “Ιmage,” 137. Anna Komnene’s phrasing suggests that the maforion (type of cloth mantle covering shoulders and head) of the Virgin Mary was hung on a pole as a military standard. Before the campaign against the Pechenegs, this rare relic had been stored in the church of St Mary in Blachernae. It was brought to Constantinople in around 466 at the behest of Verina, wife of the emperor Leo I. (457–474). See Malamut, Alexis, 98–99.
350.
Tossed weapons and various parts of horse harnesses (most often stirs and spurs) were uncovered directly on the battlefield. Jotov and Nikolov, “Дръстър,” 438–439.
351.
This is evidenced by the archeological findings of scattered horse harness components (stirs, spurs) in these two locations. Jotov and Nikolov, “Дръстър” 439.
352.
For instance, Georgios Palaiologos was pursued by the nomads for a total of eleven days before finding safe refuge with a widow. Alexias, VII.4.3. (p. 216).
353.
The description of Georgios Palaiologos’ escape is one the most vivid episodes of the Alexiad. See Alexias, VII.4.1. to VII.4.3. (p. 215–216); Vasil’evskij, “Печенеги,” 163.
354.
Alexias, VII.3.12. (p. 214); Vasil’evskij, “Печенеги,” 163; Chalandon, Essai, 116; Zlatarski, История, 194; Diaconu, Petchénègues, 119; Malamut, “Ιmage,” 138; Birkenmeier, Army, 74. The town of Beroe is today’s Stara Zagora, Bulgaria. Soustal, Thrakien, 203–205.
355.
Alexias, VII.4.4. (p. 216); Vasil’evskij, “Печенеги,” 164; Diaconu, Petchénègues, 119.
356.
Alexias, VII.4.4. (p. 216); Vasil’evskij, “Печенеги,” 164; Chalandon, Essai, 117; Zlatarski, История, 194.
357.
Alexias, VII.4.4. (p. 216): Vasil’evskij, “Печенеги,” 164; Chalandon, Essai, 117.
358.
Alexias, VII.4.4. (p. 216); Vasil’evskij, “Печенеги,” 164; Zlatarski, История, 195.
359.
Chalandon, Essai, 116–117; Paroń, Pechenegs, 362.
360.
Alexias, VII.6.1. (p. 218); Chalandon, Essai, 116.
361.
Malamut, “Ιmage,” 138.
362.
Alexias, VII.5.1. (p. 216); Vasil’evskij, “Печенеги,” 164. The distance between the area of the lower Dnieper and the lower Don and the mouth of the Danube is approximately 700 km. In theory, a horseman could cover this distance in two weeks (provided he rides at least 50 km a day). Thus, if Tatu set off from Dristra to seek help at the Kumans at the end of July 1087 (around the same time when the 150-member delegation was negotiating with Alexios Komnenos in his camp near Lardea), he could have arrived at the Kumans by the end of August. Tatu’s journey back with the Kuman host also must have taken about two weeks, so the Kumans seem to have reached the lower Danube no earlier than in mid-September 1087. The timeline of Tatu’s journey also corresponds to the conditions in the steppe north of the Sea of Azov, which is almost impassable in the summer months due to severe droughts and lack of water for horsemen. Moreover, for the same reason, there were almost no people in the area, because the nomads usually spent this time with their flocks on summer pastures lying more to the north, upstream of the Dnieper and the Siverskyi Donets rivers. Therefore, if Tatu had come to the steppe on the lower Siverskyi Donets river in the middle of summer, he simply would not have been able to find the Kumans there (this valuable information was kindly provided to me during a personal discussion with Professor Renata Holod of the University of Pennsylvania in Philadelphia).
363.
Alexias, VII.5.1. (p. 216); Vasil’evskij, “Печенеги,” 164; Chalandon, Essai, 117; Zlatarski, История, 195; Diaconu, Petchénègues, 119; Diaconu, Coumans, 39; Paroń, Pechenegs, 362.
364.
Alexias, VII.5.2. (p. 216); Vasil’evskij, “Печенеги,” 165; Chalandon, Essai, 117; Moravcsik, Byzantinoturcica II, 228; Spinei, Migrations, 143. Diaconu originally identified Ozolimne with the area around the towns of Pliska and Preslav in today’s northeastern Bulgaria, which was much more humid in the Middle Ages, with more lakes and marshes located there than today. See Diaconu, Petchénègues, 121–129; Diaconu, Coumans, 39. However, younger historians agree that the site of Ozolimne should be searched for in a marshy area on the lower Danube, where the large island of Balta Ialomiţei (with an area of 831.3 km²) is located. This island covered with marshes and lakes could have served better as a refuge for the defeated Pechenegs than the area around Preslav. Moreover, it better matches Anna Komnene’s description. See Madgearu, “Paradounavon,” 138–139.
365.
Alexias, VII.6.3. (p. 218); Vasil’evskij, “Печенеги,” 165; Chalandon, Essai, 117; Shepard, “Disinformation,” 262–265.
366.
Chalandon, Essai, 118; Szilvia Kovács, “The Cuman Campaigns in 1091,” Golden Horde Review 1, no. 3 (2014): 176–177.
367.
Alexias, VII.6.1. (p. 218); Chalandon, Essai, 118; Zlatarski, История, 196.
368.
Alexias, VII.6.2. (p. 218); Vasil’evskij, “Печенеги,” 247; Chalandon, Essai, 118; Zlatarski, История, 196; Diaconu, Petchénègues, 130; Paroń, Pechenegs, 363.
369.
Alexias, VII.6.2. (p. 218). In the text of the Alexiad, Synesios is referred to only in connection with the events of the war against the Pechenegs, and his main domain seems to have been diplomacy. See also Skoulatos, Personnages, 285–286. Seals of a certain Nikolaos Synesios (the last name is almost illegible, however), dating back to the last third of the eleventh century, have been uncovered. One of them was found on the territory of present-day Bulgaria in the village of Melnitsa, not far from the site of the Byzantine-Pecheneg encounter. Jordanov assumes that the above-mentioned prōtoproedros Nikolaos Synesios could be identical to Synesios referred to by Anna Komnene. See Ivan Jordanov, “Byzantine Seals from the Village of Melnitsa,” SBS 7 (2002): 49; Jordanov, Byzantine Seals 3, 210–211.
370.
Alexias, VII.6.2. (p. 218); Vasil’evskij, “Печенеги,” 247; Zlatarski, История, 196; Spinei, Migrations, 144; Stephenson, Frontier, 113. There is an interesting mention in the peace treaty with the Pechenegs, stating that the Pechenegs were to retain those territories in Paradounavon they had conquered so far (“μένειν ἐν ᾧ προκατέλαβον τόπῳ”), which, in my view, certainly did not apply to the surroundings of the town of Markellai, as the territory south of the Haimos was to remain in Byzantine hands, as the treaty stated. I believe that the territory recently conquered by the Pechenegs represents the Byzantine enclaves in Paradounavon (around the town of Preslav and northern Dobrudja). See also Chalandon, Essai, 118–119. See also text above.
371.
Alexias, VII.6.2. (p. 218); Zlatarski, История, 196; Malamut, “Ιmage,” 139. It is interesting to see how, even in a very disadvantageous strategic situation, the Byzantine emperor tried, holding the “Kuman card,” to force the Paradounavon Pechenegs to normalize their mutual relations, or to revert the status quo to the period before the Nestor’s uprising.
372.
Alexias, VII.6.3. (p. 218); Zlatarski, История, 197; Diaconu, Petchénègues, 130.
373.
I arrived at this time data in a similar way to note 362 above. The Kumans were able to travel back and forth from the lower Danube to their settlements in about a month. The hypothetical date at the turn of October and November 1087 also takes into account the fact that if the Kumans arrived later than on the proposed date, they would have had to deal with harsh weather conditions at the beginning of winter.
374.
Alexias, VII.6.3. (p. 218); Zlatarski, История, 197.
375.
Alexias, VII.6.3. (p. 218).
376.
Alexias, VII.6.3. (p. 218); Vasil’evskij, “Печенеги,” 248; Chalandon, Essai, 118; Zlatarski, История, 197; Diaconu, Petchénègues, 130.
377.
Alexias, VII.6.3. (p. 218); Vasil’evskij, “Печенеги,” 248; Chalandon, Essai, 119; Zlatarski, История, 197; Diaconu, Petchénègues, 130.
378.
There is no mention of the Byzantine-Kuman alliance in the relevant passage of Anna Komnene’s text (see note above). However, a brief mention in the speech of Theofylaktos of Ochrid, in which it is clearly stated that Alexios Komnenos concluded an alliance with the Huns against the Skyths, provides a firm foothold for this claim. In this context, the Huns were the Kumans, whereas the Skyths were the Paradounavon Pechenegs. See Gautier, “Théophylacte,” 111: “Ἐκεῖνά σε, ὦ γενναῖε, ἐκεῖνα καὶ τοῖς Οὔννοις φοβερὸν ἀπειργάσαντο (…) εἰς συμμάχους ἑαυτοὺς κατὰ Σκυθῶν τῶν καταράτων ἐνέγραψαν.” Moreover, in my opinion, it would be difficult to otherwise explain the “coincidental” arrival of the Kumans in Thrace after several years of absence, just before the decisive encounter of the entire war with the Pechenegs (battle of Lebounion) in the spring of 1091. See Vasil’evskij, “Печенеги,” 248; Jonathan Shepard, “‘Father’ or ‘Scorpion’? Style and Substance in Alexius’ Diplomacy,” Alexios I Komnenos—Papers of the Second Belfast Byzantine International Colloquium, 14–16 April 1989, 83, note 77. See also text below.
379.
The same kind of cautiously optimistic spirit is reflected in the speech of Theofylaktos of Ochrid of 6 January 1088. See Gautier, “Théophylacte,” 114.
380.
Alexias, VII.6.4. (p. 219); Chalandon, Essai, 119.
381.
Alexias, VII.6.4. (p. 219); Vasil’evskij, “Печенеги,” 248; Diaconu, Petchénègues, 131; Paroń, Pechenegs, 364. Some historians reject or completely question this brief reference to the domination of the city by the Pechenegs; see Soustal, Thrakien, 400.
382.
Alexias, VII.6.4. (p. 219).
383.
Ibid.; Vasil’evskij, “Печенеги,” 248; Chalandon, Essai, 119; Diaconu, Petchénègues, 131; Malamut, “Ιmage,” 141; Kólias, “Πολιτικὴ,” 257–258.
384.
This premise is based on the birth of the third daughter of the emperor Alexios Komnenos and Eirene Doukaina, Eudoxia, on 14 January 1089. See Schreiner, Kleinchroniken, 55: “μηνὶ ἰαννουαρίῳ ιδ΄ ἰνδικτιῶνος ιβ΄, ἐγεννήθη ἡ πορφυρογέννητος κυρὰ Εὐδοξία, ἡμέρᾳ <α΄>.”
385.
Good relations between the emperor Michael Doukas and the pope Gregory VII were directed against the rise of Robert Guiscard, who in 1071 seized the last Byzantine foothold in southern Italy, the city of Bari, and intended to extend his expansion plans to the other side of the Adriatic Sea. See also Charanis, “Crusade,” 22–23. In the early 1080s, however, relations with the papal curia cooled down, and, therefore, the 1081 Norman invasion into the Byzantine territory, as mentioned in detail above, took place with full papal support. In the early years of Alexios Komnenos’ reign, the emperor Henry IV, who was an adversary of both the pope and Guiscard, was therefore the main ally of Byzantium in the Latin West.
386.
According to the information contained in the Chronicle of Ordericus Vitalis, the initiative was taken by the pope Urban II. The chronicler notes that this pope sent messengers to “the French and Greeks” with letters asking them for support against his opponent Clement III. See Orderic Vitalis, 166: “Vrbanus (…) missit legatos et epistolas Romanæ auctoritatis Francis et Grecis.” This thawing of mutual relations between the papal curia and Byzantium became more apparent only in September 1089, when Alexios Komnenos pushed through the matter of restoration of the pope’s name in the diptychs of the Orthodox Church at a church synod in Constantinople. Holtzmann, “Unionsverhandlungen,” 47–50. The slow progress of negotiations between the pope and the Byzantine emperor can also be partly explained by the fact that Byzantine diplomacy initially sided with the antipope Clement III, who was supported by the main Byzantine ally in the West during the war against the Normans, the Roman-German emperor Henry IV.
387.
Alexias, VII.6.5. (p. 219): “τὸ ἐλπιζόμενον μισθοφορικὸν οὔπω κατέλαβε.” This effort can also be seen as a kind of forerunner of numerous requests for assistance at the beginning of 1091 and later, which Alexios Komnenos addressed to Urban II and which ultimately represented one of the important causes of the genesis of the First Crusade.
388.
Ian S. Robinson, The Papacy 1073–1189: Continuity and Inovation (Cambridge: Cambridge University Press, 1990), 374.
389.
Alexias, VII.6.4. (p. 219); Chalandon, Essai, 119; Zlatarski, История, 198.
390.
Alexias, VII.6.5. (p. 219); Vasil’evskij, “Печенеги,” 248; Chalandon, Essai, 119; Malamut, “Ιmage,” 141. This second peace treaty with the Pechenegs is not specifically dated in the text of Anna Komnene either. A clue is the dating of the first peace treaty with 6 January 1088 as terminus ante quem, because this treaty is mentioned in the speech of Theofylaktos of Ochrid. See Gautier, “Théophylacte,” 98, 113. Therefore, all the events mentioned in the text of the Alexiad following the conclusion of this treaty, including the capture of Philippoupolis by the Pechenegs, had to take place in the spring of 1088. Since the reference to the second peace treaty predates the arrival of the winter of 1088, it can be assumed that the second round of peace talks took place in autumn 1088, because the Pechenegs managed to break their part of the new (second) agreement by the wintertime. See Alexias, VII.6.2ff (p. 218–220).
391.
Alexias, VII.6.5. (p. 219).
392.
Ibid. For more details on this Pecheneg leader, see Skoulatos, Personnages, 217–218; Moravcsik, Byzantinoturcica II, 210.
393.
Alexias, VII.6.6. (p. 219).
394.
Alexias, VII.6.6. (p. 220); Vasil’evskij, “Печенеги,” 250; Chalandon, Essai, 124; Diaconu, Petchénègues, 131; Paroń, Pechenegs, 364. The exact location of the town (or mountain ridge with numerous peasant settlements) of Taurokomos is unknown, but it was apparently located southeast of Adrianoupolis. Soustal, Thrakien, 472; Zlatarski, История, 198. See also text below.
395.
Alexias, VII.7.1. (p. 220); Chalandon, Essai, 125; Zlatarski, История, 198.
396.
The negative effect of this strategy was that the Pechenegs had unlimited freedom of movement throughout Thrace.
397.
Alexias, VII.7.1. (p. 220); Vasil’evskij, “Печенеги,” 250; Chalandon, Essai, 125; Zlatarski, История, 198; Diaconu, Petchénègues, 131; Haldon, Warfare, 93; Paroń, Pechenegs, 364.
398.
Alexias, VII.7.3. (p. 221). The emperor’s personal guard probably consisted, as during the battle of Dristra in 1087, of Alexios Komnenos’ close relatives and servants of the imperial household. Haldon, Warfare, 93.
399.
Alexias, VII.7.1. (p. 220); Vasil’evskij, “Печенеги,” 250; Chalandon, Essai, 125; Zlatarski, История, 198; Diaconu, Petchénègues, 131.
400.
The basic scheme of this battle plan is very similar to the tactics used by Alexios Komnenos against the Normans at Larissa in 1083. When one part of the Byzantine army (the “bait”) lured the main Norman forces out of their camp, the emperor attacked it with the rest of his forces and practically wiped out its few defenders. See text above.
401.
Alexias, VII.7.2 (p. 220–221).
402.
Alexias, VII.7.2 (p. 221).
403.
Alexias, VII.7.2 (p. 221); Vasil’evskij, “Печенеги,” 250–251; Zlatarski, История, 198–199; Diaconu, Petchénègues, 131; Malamut, “Ιmage,” 141; Paroń, Pechenegs, 364.
404.
Alexias, VII.7.3. (p. 221); Vasil’evskij, “Печенеги,” 251; Chalandon, Essai, 125; Zlatarski, История, 199.
405.
Alexias, VII.7.3. (p. 221); Vasil’evskij, “Печенеги,” 251; Chalandon, Essai, 125; Diaconu, Petchénègues, 131.
406.
Alexias, VII.7.3. (p. 221).
407.
Alexias, VII.7.3. (p. 221); Vasil’evskij, “Печенеги,” 251; Zlatarski, История, 199; Diaconu, Petchénègues, 131.
408.
Birkenmeier, Army, 75.
409.
Alexias, VII.6.1. (p. 218). For the dating of the meeting of Alexios Komnenos and the Flemish count to the end of 1089, or the beginning of 1090, but not to the autumn of 1087, see text above.
410.
Alexias, VII.6.1. (p. 218). See Ganshof, “Robert,” 58–64.
411.
For the dating of the arrival of the Flemish knights, see text above.
412.
Alexias, VII.7.4. (p. 221–222); Ganshof, “Robert,” 71–72; Vasil’evskij, “Печенеги,” 251; Zlatarski, История, 199; Diaconu, Petchénègues, 131; Shepard, “Substance,” 103.
413.
Ganshof, “Robert,” 72; Krijnie Ciggaar, “Flemish Mercenaries in Byzantium: Later History in an Old Norse Miracle,” Byzantion 51 (1981): 45, note 6.
414.
Alexias, VII.7.4. (p. 222); Janin, “Francs,” 69; Ganshof, “Robert,” 73; Belke, “Bemerkungen,” 73–74; Bondoux, Bithynie, 401; Frankopan, Crusade, 57.
415.
For this date, see above.
416.
The exact location of Polyboton is unknown. Külzer, Ostthrakien, 599. However, it probably lay north of the Rousion-Kypsella line. Anna Komnene’s text also clearly suggests that the town was located nearby, perhaps 8 to 10 km west of the town of Rousion, and that a small river flowed between the two towns. A hilly area rolled to the north of the town. Alexias, VII.9.1. (p. 229). See also Zlatarski, История, 200, note 2 in Chap. 1.
417.
Alexias, VII.9.1. (p. 227); Zlatarski, История, 200. Rousion can be identified with the abandoned town of Rusköy, located 6 km southeast of Keşan in Turkish Thrace. Asdracha, “Thrace,” 236; Külzer, Ostthrakien, 620–622.
418.
During the first half of 1090, Alexios Komnenos, while in Constantinople, was busy organizing the operations aimed against Çaka, the Turkish emir of Smyrna, mentioned in the Alexiad before these events. Alexias, VII.8.1.ff (p. 222–226). Perhaps this is the main reason behind his apparent impatience to return to Constantinople and deal with Çaka as soon as possible.
419.
Alexias, VII.9.1. (p. 227).
420.
Alexias, VII.9.1. (p. 227).
421.
Ibid.
422.
Ibid., VII.9.2. (p. 227). See note 232 in Chap. 4.
423.
Alexias, VII.10.4. (p. 231). Georgios Pyrrhos took part in the 1083 battle against the Normans at Larissa, in which he excelled as an expert archer. See note 523 in Chap. 4.
424.
Alexias, VII.9.1. (p. 227). This Pecheneg joined the service of the Byzantine emperor in 1088 after the second peace treaty with the Paradounavon Pechenegs. See text above.
425.
Alexias, VII.9.1. (p. 227); Skoulatos, Personnages, 159; Moravcsik, Byzantinoturcica II, 149.
426.
Tatranes/Katranes or Tatran was, like Neantzes, of Pecheneg origin, and defected to the service of the Byzantine emperor. Alexias, VII.9.1. (p. 227); Skoulatos, Personnages, 292–293; Moravcsik, Byzantinoturcica II, 302; Paroń, Pechenegs, 229.
427.
Alexias, VII.9.7. (p. 229).
428.
Alexias, VII.9.7. (p. 229). Anna Komnene does not give any further details about Monastras’ origin, but scholars assume that he was of unspecified Turkish, most probably Kuman, origin. See Moravcsik, Byzantinoturcica II, 192; Skoulatos, Personnages, 213–216. Recently, the seals of an individual named Michael Monastras and a rather modest court rank of prōtobestiarios or prōtobestēs have been uncovered in the town of Shumen in northwestern Bulgaria. Although it would be very tempting to identify this individual with Monastras, who had certainly been baptized before entering the imperial service (perhaps during the reign of Michael VII), this possibility remains unresolved. Jordanov, Corpus 2, 271.
429.
Alexias, VII.9.7. (p. 229). It needs to be remembered that Synesios, as the Byzantine envoy, lived with the Pechenegs in Paradounavon for some time and thus probably had the opportunity to become familiar with their everyday life and customs.
430.
See text above. See also Meško, “Nomad Influences,” 70.
431.
Alexias, VII.9.2. (p. 227).
432.
Ibid. The location of this place is unknown, but it had to lie somewhere between the towns of Rousion and Polyboton, somewhat closer to Rousion, separated by a river. See Alexias, VII.10.4. (p. 231); Külzer, Ostthrakien, 387.
433.
Alexias, VII.9.2. (p. 227).
434.
Alexias, VII.9.5. (p. 228).
435.
Alexias, VII.9.6. (p. 227).
436.
Alexias, VII.9.6. (pp. 227–228).
437.
See Malamut, “Ιmage,” 141.
438.
Alexias, VII.9.7. (p. 229).
439.
That is, the Byzantines to the town of Rousion and the Pechenegs to the location known as Hades. Alexias, VII.9.2. (p. 227).
440.
Alexias, VII.9.6. (p. 229).
441.
This assumption is based on the presence of Ouzas on the battlefield. During the battle of Dristra in 1087, he was the second-in-command of this unit, which at that time was under the command of megas hetaireiarchos Argyros Karadja, and it is possible that in the autumn of 1090 he was acting as its leader again, either as the deputy of the absent Karadja or as its new commander.
442.
Alexias, VII.10.1. (pp. 229–230).
443.
Alexias, VII.10.2. (p. 230). In this case, it is again a very similar tactic of indirect charge to that used by Alexios Komnenos against Bohemund’s Normans in the battle of Larissa. See text above.
444.
Alexias, VII.10.3. (p. 231).
445.
Alexias, VII.10.3. (p. 231).
446.
Alexias, VII.10.4. (p. 231).
447.
Alexias, VII.10.4. (p. 231).
448.
Alexias, VII.10.3. (p. 231).
449.
Alexias, VII.11.1. (p. 232); Chalandon, Essai, 128; Zlatarski, История, 200.
450.
Malamut, “Ιmage,” 141. Tzouroulon is today’s Turkish city of Çorlu. Asdracha, “Thrace,” 255; Pralong, “Fortifications,” 181, 186; Külzer, Ostthrakien, 684–688.
451.
Shepard, “Disinformation,” 257. Parts of the walls on the western side of the town have been preserved to this day and in some places reach as high as four meters. The construction technique is typically Byzantine and is characterized by alternating use of layers of worked blocks of stone and fired bricks. As for the dating of the fortification, it is possible that its origins still fall into the Roman period. See Pralong, “Fortifications,” 185; Külzer, Ostthrakien, 687.
452.
Alexias, VII.11.1. (p. 232).
453.
Today’s Çorlu Suyu river, left tributary of the Ergene river. Külzer, Ostthrakien, 691.
454.
Alexias, VII.11.1. (p. 232).
455.
Ibid. The Pechenegs and other nomadic peoples rarely besieged fortified towns. Pletneva, “Печенеги,” 198.
456.
Alexias, VII.11.2. (p. 232); Zlatarski, История, 200.
457.
Alexias, VII.11.2. (p. 232).
458.
Alexias, VII.11.2. (p. 232); Vasil’evskij, “Печенеги,” 253; Zlatarski, История, 200; Shepard, “Disinformation,” 257.
459.
Alexias, VII.11.3. (p. 233).
460.
Alexias, VII.11.3. (p. 233).
461.
Alexias, VII.11.4. (pp. 233–234); Vasil’evskij, “Печенеги,” 253; Shepard, “Disinformation,” 257; Zlatarski, История, 200.
462.
Alexias, VII.11.4. (p. 234). Interestingly enough, a similar trick (albeit unsuccessful) was used by the Goths in 540 against the Byzantine army led by Belisarios, which besieged the Goth fortress of Auximon located on a high hill. Identical tactics (with no success, however) against the Makedonian phalanx of Alexander the Great were used in 335 BC by the Thracians. See Alexopoulos, “Handbooks,” 57–58.
463.
Alexias, VII.11.5. (p. 234).
464.
It is very likely that his decision-making was influenced by the fact that Tzouroulon now housed the town residents, as well as peasants from the surrounding area seeking refuge from the Pechenegs and the entire Byzantine army, and, in the event of a prolonged siege, there was a possibility that they would soon run out of food and water supplies and that epidemic diseases would eventually break out. Shepard, “Disinformation,” 257.
465.
Alexias, VII.11.5. (p. 234).
466.
Alexias, VII.11.5. (p. 234); Zlatarski, История, 200.
467.
Alexias, VII.11.5. (p. 234); Birkenmeier, Army, 75.
468.
Birkenmeier, Army, 75; Diaconu, Petchénègues, 132.
469.
Alexias, VII.11.6. (p. 235); Vasil’evskij, “Печенеги,” 253; Chalandon, Essai, 128; Zlatarski, История, 200; Malamut, “Ιmage,” 141; Paroń, Pechenegs, 365. The area where the Pechenegs retreated lay about 80 km northwest of Tzouroulon at the foot of the Υıldız dağları heights.
470.
Alexias, VII.11.6. (p. 235).
471.
This date is based on the assumption that Alexios Komnenos left with his troops for the town of Choirobakchoi the very next morning. See text below.
472.
Alexias, VIII.1.1. (p. 236); Vasil’evskij, “Печенеги,” 238; Chalandon, Essai, 128; Zlatarski, История, 201; Gautier, “Diatribes,” 9. Choirobakchoi was situated about 20 km west of Constantinople on the plain near the present-day towns of Büyükçekmece (in the Byzantine area known as Athyra) and Küçükçekmece (known as Rhegion). As two waterways ran through this plain (the Athyras and the Mollas rivers), and the Via Egnatia also passed through this area, it was a very convenient and strategically important place for the gathering of the Byzantine troops (infantry and cavalry alike). Jean-Claude Cheynet, “Episkeptitai et autres gestionnaires des biens publics (d’après les sceaux de l’IFEB),” SBS 7 (2002): 101–102. This fact also explains the increased Pecheneg interest in this location (an excellent site for winter encampment), as well as the immediate reaction of Alexios Komnenos. Currently, the southern part of this plain is located below the water surface of the Büyükçekmece gölü reservoir. Most recently, this location has been definitely confirmed by Külzer, who puts the town of Choirobakchoi to a place between the aforementioned rivers near the present-day site of Bahşayıs. See Külzer, Ostthrakien, 313.
473.
Alexias, VIII.1.1. (p. 236): “ἐπὶ τῆς φρουρᾶς τῆς πόλεως τεταγμένους.” Thanks to this formulation, it could be assumed that Bigla was still a separate and functional army unit in the spring of 1091.
474.
Alexias, VIII.1.1. (p. 236).
475.
Ibid.
476.
Alexias, VIII.1.2. (p. 236–237); Vasil’evskij, “Печенеги,” 238; Chalandon, Essai, 128; Zlatarski, История, 201.
477.
Alexias, VIII.1.3. (p. 237); Vasil’evskij, “Печенеги,” 253; Zlatarski, История, 201.
478.
Anna Komnene reports 6000 Pecheneg horse archers moving toward Dekaton in total. Alexias, VIII.1.3. (p. 237). Vasil’evskij, “Печенеги,” 253; Zlatarski, История, 201. In that case, however, the ratio of forces between the Byzantines and this Pecheneg unit would be a staggering 12:1. The second group was probably much smaller and it is possible that it roughly matched the number of the Byzantines. The reasons why I assume the numerical strength of “only” 1000 fighters for the bigger Pecheneg unit are as follows: in the afternoon clash, when this larger group was returning to Choirobakchoi, only a part of the Byzantine force pretending to be the Pechenegs (wearing the clothes of the dead nomads from the first clash) was sufficient to inflict defeat upon them. The Byzantine force could not be large because of the limited number of stolen nomadic clothing and mounts available on the spot. Alexias, VIII.1.5. (p. 238). It is possible to assume that there were no more than 500 of them (the total number of Byzantine soldiers) because even a few days later on their way back to Constantinople, when Alexios Komnenos, as part of a “Canadian prank,” used soldiers dressed as the Pechenegs to scare off the other Byzantine troops on their way from the Byzantine capital to join him, not all soldiers were wearing Pecheneg garments or brandishing Pecheneg standards—only the Byzantine soldiers in the vanguard. Alexias, VIII.2.2. (p. 239). Therefore, it can be assumed that the entire Pecheneg host moving to the town of Choirobakchoi (before splitting into two) could not have exceeded 1500 or 2000 men, so they outnumbered the Byzantine forces under the emperor’s command “only” by a ratio of 3:1 to 4:1.
479.
Alexias, VIII.1.3. (p. 237); Vasil’evskij, “Печенеги,” 253; Zlatarski, История, 201. The name Dekaton is derived from its location 10 Roman miles, that is, 14.8 km (1 Roman mile = 1480 m), from Constantinople. See Külzer, Ostthrakien, 325. Anna Komnene uses the ancient unit of length of stadion in her archaic style, as it had not been used since the time of the Roman Empire. See Schilbach, Metrologie, 32–33.
480.
Alexias, VIII.1.3. (p. 237); Vasil’evskij, “Печенеги,” 253; Zlatarski, История, 201.
481.
Alexias, VIII.1.3. (p. 237); Vasil’evskij, “Печенеги,” 253.
482.
Alexias, VIII.1.3. (p. 237); Vasil’evskij, “Печенеги,” 253; Zlatarski, История, 201.
483.
Alexias, VIII.1.5. (p. 238).
484.
Alexias, VIII.1.5. (p. 238); Vasil’evskij, “Печенеги,” 254; Zlatarski, История, 201.
485.
Alexias, VIII.1.5. (p. 238). It is today’s river Karas (aforementioned river Melas). See Vasil’evskij, “Печенеги,” 254; Zlatarski, История, 202. In other sources, this river is known by the name Mauropotamos. Külzer, Ostthrakien, 518. All these names always mean dark/black river/water.
486.
Alexias, VIII.1.5. (p. 238); Vasil’evskij, “Печенеги,” 254; Zlatarski, История, 201. At the same time, this event demonstrates how much the Byzantine soldiers at the time had to resemble the Pechenegs in terms of external appearance and how powerful the influence of the nomadic style of warfare was at the time on the Byzantine military tradition. as the Byzantines had practically the same weaponry and equipment, it was enough to change only the garments and mounts (or the horse harnesses) to look alike. See also Meško, “Nomad influences,” 70.
487.
Alexias, VIII.1.5. (p. 238); Vasil’evskij, “Печенеги,” 254–255; Zlatarski, История, 202.
488.
Alexias, VIII.2.5. (p. 240); Vasil’evskij, “Печенеги,” 255; Chalandon, Essai, 128; Zlatarski, История, 202.
489.
Compare with a critical remark allegedly uttered by kaisar Nikeforos Melissenos in the aftermath of Alexios Komnenos’ victory at Choirobakchoi: “This victory was a profitless joy to us, to them a harmless pain.” Alexias, VIII.3.1. (p. 241): “ἡ νίκη αὕτη χαρὰ μὲν ἀκερδής, λύπη δὲ ἀζήμιος.”
490.
Anna Komnene’s description gives a similar impression. However, two speeches by the patriarch of Antioch John VI the Oxite may evoke a slightly different perception of the military standing of the Byzantine Empire in the spring of 1091. The patriarch claims that Alexios Komnenos had gradually lost his entire army in the battles against the Pechenegs. Gautier, “Diatribes,” 35. For the person and career of John the Oxite, see Grumel, “Patriarches,” 294. Most recently, Ryder commented on the role of both his speeches in the context of imperial court politics. See Judith R. Ryder, “The Role of the Speeches of John the Oxite in Komnenian Court Politics,” Reading in the Byzantine Empire, ed. by Teresa Shawcross and Ida Toth (Cambridge: Cambridge University Press, 2018), 93–115.
491.
Alexias, VIII.3.1. (p. 241); Gautier, “Diatribes,” 35; Spinei, Migrations, 144.
492.
The feast of St Theodore Teron is celebrated in the Orthodox Church on the first Saturday of the Great Lent, which in 1091 fell on 7 March. Alexias, VIII.3.1. (p. 241); Chalandon, Essai, 129. Vasil’evskij gives an erroneous date of 2 March 1091. See Vasil’evskij, “Печенеги,” 255.
493.
Külzer, Ostthrakien, 282–283.
494.
Alexias, VIII.3.1. (p. 241); Chalandon, Essai, 129; Gautier, “Diatribes,” 10.
495.
Gautier, “Diatribes,” 35.
496.
Ibid.
497.
Shepard, “Disinformation,” 251, 260–269. The issue of hypothetical Byzantine “operational code” was last dealt with in great detail by Luttwak, who eventually summarized it in seven principal points. If we thoroughly recap the actions and decisions of Alexios Komnenos during the war against the Pechenegs, we find that they strikingly match Luttwak’s theoretical reconstruction of maxima of Byzantine military strategy. See Edward Luttwak, The Grand Strategy of the Byzantine Empire (Cambridge, MA: Harvard University Press, 2009), 415–418; Kaegi, “Logistics,” 47.
498.
Shepard, “Disinformation,” 256–257. In addition to large and well-fortified cities (e.g., Adrianoupolis), there were many smaller walled towns and castles in Thrace, such as Tzouroulon, Brysis (today’s Pinarhisar), Bizye (today’s Vize), and Medeia (today’s Kiziköy). See Pralong, “Fortifications,” 179–200. An exception was the city of Philippoupolis, which the Pechenegs probably seized in the spring of 1088. However, this important city did not seem to have remained in Pecheneg hands for long, since a Byzantine inscription was recently uncovered in the church near the village of Batkun (Pasardzhik District, west of Plovdiv, now Bulgaria), dating to 1090/1091 and giving the name of doux of Philippoupolis, Gregorios Kourkouas. This means that by 1090 at the latest, the city was again controlled by the Byzantines. See Jordanov, Byzantine Seals 2, 240–241; Luisa Andriollo, “Les Kourkouas (IXe – Xie siecle),” SBS 11 (2012): 84–85.
499.
Shepard, “Disinformation,” 252.
500.
For the name of Çaka, see Moravcsik, Byzantinoturcica II, 310. Çaka’s life career was very variegated—during the reign of the emperor Nikeforos III Botaneiates (1078–1081), he fell into Byzantine captivity, where he was probably baptized. Soon he managed to gain the trust of the emperor and with it he achieved high status at the imperial court and an exalted court title of prōtonōbelissimos. After Alexios Komnenos took the throne, the situation changed and Çaka was stripped of his duties and titles and released. See Alexias, VII.8.7. (p. 225); Brand, “Element,” 17; Gautier, “Défection,” 218. Greatly disappointed, Çaka then headed to the city of Smyrna in Asia Minor, which he seized, and started to build his independent power base there (a type of a pirate emirate). Böhm, Flota, 143–144.
501.
Alexias, VIII.3.2. (p. 241); Vasil’evskij, “Печенеги,” 256, 277; Chalandon, Essai, 129; Zlatarski, История, 203; Kólias, “Πολιτικὴ,” 258; Gautier, “Diatribes,” 10; Alexandrís, Δύναμις, 324; Aléxios G. C. Savvídis, “Ο Σελτζούκος εμίρης της Σμύρνης Τζαχάς (Çaka) και οι επιδρομές του στα μικρασιατικά παράλια, τα νησιά του ανατολικού Αιγαίου και την Κωνσταντινούπολη (c. 1081–1106),” B’ 1090-c.1106, Βυζαντινοτουρκικά μελετήματα (Athens: Iródotos, 1991), 52; Angold, Empire, 133; Malamut, “Ιmage,” 141; Madgearu, “Paradounavon,” 139; Kovács, “Campaigns,” 178; Frankopan, Crusade, 59; Paroń, Pechenegs, 365.
502.
Çaka’s fleet consisted mainly of small and fast oared pirate warships with little displacement, suitable for rapid attacks on coastal sites and for ambushing larger and slower merchant vessels. Alexias, VII.8.1. (p. 222); Alexandrís, Δύναμις, 323; Ahrweiler, Mer, 184; Böhm, Flota, 144; Gautier, “Défection,” 217–218.
503.
Gautier, “Diatribes,” 12, 35. Compare with the data in Alexios Komnenos’ famous letter to Count Robert of Flanders in Heinrich Hagenmeyer, Die Kreuzzugsbriefe aus den Jahren 1088–1100 (Hildesheim: Olms, 1973), 132–133: “insulae principales Chios et Mitylena et multae aliae regiones et insulae, quas non ualemus modo enumerare, usque Thracias”; Ahrweiler, Μer, 184; Gautier, “Défection,” 217; Cheynet, “Résistance,” 145–146. For information on the Aegean islands occupied by Çaka, see Malamut, Îles, vol. 1, 232 (Mytiléné), 236–237 (Chios), 237–239 (Samos).
504.
Niketas Kastamonites took part in the battle of Dristra in 1087. See note 328 above. Given that he commanded the squadron of the Byzantine fleet, it can be assumed that at the turn of 1090 he held the rank of komēs tou stolou.
505.
Alexias, VII.8.2. (p. 223); Böhm, Flota, 144.
506.
Constantine Dalassenos was a relative of Alexios Komnenos by the empress mother Anna Dalassene. Alexias, VII.8.3. (p. 223). He was appointed as the commander of the fleet (doux tou stolou) no earlier than spring 1090. Skoulatos, Personnages, 61–62; Guilland, “Marine,” 220. His lead seal, which also includes the title of prōtokouropalatēs, was published by Cheynet. See Cheynet, “Dalassenoi,” 101–103. His other brother-in-law, sebastos John Doukas, who had personally participated in the naval operations against Çaka only after 1092, was appointed by the emperor as the commander-in-chief (doux tou stolou and later megas doux). For the appointment of John Doukas as the supreme commander of the fleet, see Alexias, VII.8.8. (p. 226); Guilland, “Marine,” 220. For his career, see Skoulatos, Personnages, 145–150. For the preserved sigillographic material of this distinguished member of the high Byzantine aristocracy, see Jordanov, Byzantine Seals 2, 140–142.
507.
Alexias, VII.8.3. (p. 223). Magistros Constantine Opos took part in the battle of Dyrrachion, where he commanded the unit of the Exkoubitai. See note 269 in Chap. 4. During the naval operations against Çaka in the spring of 1090, he acted as a subordinate commander of Constantine Dalassenos and was entrusted with part (a squadron) of the Byzantine fleet. It is possible that he held the rank of komēs tou stolou, although this fact has not yet been confirmed either by written sources or by uncovered sigillographic material. Remarkably, his entire later career is firmly tied to the Byzantine fleet, of which he eventually became the commander-in-chief (megas doux) sometime after 1100. See Jordanov, Byzantine Seals 2, 444.
508.
Alexias, VII.8.10. (p. 226); Alexandrís, Δύναμις, 323–324; Böhm, Flota, 147–148.
509.
Alexias, VIII.3.2. (p. 241); Savvídis, “Τζαχάς,” 52; Alexandrís, Δύναμις, 325; Kólias, “Πολιτικὴ,” 258; Gautier, “Diatribes,” 10.
510.
Madgearu, “Paradounavon,” 139. The siege of Constantinople in 626 has been most recently systematically discussed by Hurbanič. See Martin Hurbanič, Konstantinopol 626: Poslední bitka antiky (Prague: Academia, 2016).
511.
Based on his short experience at the imperial court in Constantinople, Çaka was definitely aware that the Byzantine capital could only be conquered by a combined attack of land and naval forces. Since he had a war fleet and could attempt to perform a naval blockade of Constantinople, he only needed the Pechenegs to carry out a similar blockade on land. Vasil’evskij, “Печенеги,” 256.
512.
Chalandon, Essai, 129; Kólias, “Πολιτικὴ,” 258.
513.
Alexias, VIII.3.3. (p. 242); Vasil’evskij, “Печенеги,” 256; Zlatarski, История, 203.
514.
Vasil’evskij, “Печенеги,” 257–258, 271–273; Chalandon, Essai, 129–131; Charanis, “Crusade,” 27–28; Shepard, “Attitudes,” 103–104. One of these letters was probably served to count Robert of Flanders, which, as the heading of the letter says, was not addressed only to him, but also to other princes in the kingdom (i.e., other important feudal lords); see Hagenmeyer, Kreuzzugsbriefe, 130: “Domino et glorioso comiti Flandrensium Rotberto et omnibus totius regni principibus.” See also Heinrich Hagenmeyer, “Der Brief des Kaisers Alexios I Komnenos an der Grafen Robert I von Flandern,” Byz. Zeitsch. 6 (1897): 1–32; Marek Meško, “Otázka pravosti a datovania listu Alexia I. Komnéna grófovi Róbertovi z Flámska,” Byzantinoslovaca 3 (2010): 64–71.
515.
The snowfall in Constantinople was so heavy that many of its inhabitants remained trapped in their homes. Similar weather is not uncommon in Istanbul even today. Alexias, VIII.3.3. (p. 241–242); Vasil’evskij, “Печенеги,” 256.
516.
Alexias, VIII.3.4. (p. 242). Nikeforos Melissenos had to leave Constantinople sometime between Alexios Komnenos’ victory at Choirobakchoi and the spring equinox. His mission is also confirmed by a letter from bishop Theofylaktos of Ochrid, dated to the period of Easter Lent in 1091. Margaret Mullett, Theophylact of Ochrid: Reading Letters of a Byzantine Archbishop, Birmingham Byzantine and Ottoman Monographs, vol. 2. (Aldershot: Variorum, 1997), 86, 298.
517.
Alexias, VIII.3.4. (p. 242); Vasil’evskij, “Печенеги,” 278; Chalandon, Essai, 132; Zlatarski, История, 204; Spinei, Migrations, 145; Kovács, “Campaigns,” 178. Today, Turkish Enez is located on the site of the Byzantine port of Ainos; see Soustal, Thrakien, 170–173. Very little is known about the history of Ainos during the Byzantine era, except that it was a medium-sized fortified commercial harbor dominated by a castle (ancient acropolis). Recently (from 2012 to 2018), investigations focused on ancient and medieval building structures were conducted on the site, but its finds remain still rather sketchy at best, because they were unable to provide a general idea about the layout of this Byzantine harbor. In any case a portion of the surviving city walls was built in the Middle Byzantine period, probably not before the eleventh century. See: Thomas Schmidts, “Fortifying Harbour Cities on the Southern Thracian Coast in the Early Byzantine Era – Case Studies on Ainos and Anastasioupolis,” in Seasides of Byzantium, Harbours and anchorages of a Mediterranean Empire, edited by Johannes Preiser-Kapeller, Taxiarchis G. Kolias, and Falko Daim (Mainz: Verlag des Römisch-Germanischen Zentralmuseums, 2021), 222–224.
518.
Anna Komnene does not specify this date either. However, in the text of the Alexiad, she notes that the Kumans appeared on the lower Maritsa a week after Alexios Komnenos’ arrival in Ainos (the emperor spent the first day after his arrival looking for a suitable place to build a camp for the Byzantine army; the next day, the Byzantines pitched the camp in Choirenoi; on the third day, the Pechenegs appeared in the area; and finally, after four days, the Kuman host also arrived). Alexias, VIII.3.5.ff (p. 242–243). Apparently, on the day after their arrival, the Kumans agreed with Alexios Komnenos on a ten-day period during which they were to attack the Pechenegs. Alexias, VIII.4.3. (p. 243). However, we certainly cannot take this whole timeframe into account, because the Kumans “forced” the Byzantines into a decisive battle a little earlier. Alexias, VIII.5.1. (p. 245–246). As described by Anna Komnene, the events that took place after the conclusion of the agreement cover a total of about five days (the Byzantine army crossed the Maritsa river on the third day after the arrival of the Kumans, on the fifth day there was a skirmish with the Pechenegs at the ford of Philokalos and the camp was built, and finally on the fifth day, i.e., the last day before the battle, the Byzantines moved to Mount Lebounion (Alexias, VIII.4.4ff (p. 244–245)). Therefore, since his arrival in Ainos until 29 April 1091, when the battle of Lebounion took place, Alexios Komnenos stayed on the lower Maritsa for approximately two weeks.
519.
Alexias, VIII.3.5. (p. 242); Vasil’evskij, “Печенеги,” 278; Chalandon, Essai, 132; Ciggaar, “Flemish,” 45; Frankopan, Crusade, 57; Paroń, Pechenegs, 365.
520.
Alexias, VIII.3.4. (p. 242); Ahrweiler, Μer, 190; Kólias, “Πολιτικὴ,” 259.
521.
Alexias, VIII.3.5. (p. 242); Vasil’evskij, “Печенеги,” 278.
522.
Alexias, VIII.3.5. (p. 242); Vasil’evskij, “Печенеги,” 278; Zlatarski, История, 204; Kovács, “Campaigns,” 178. The location of Choirenoi is unknown. The assumption that it was located on the right bank of the Maritsa river relies on Anna Komnene’s claim that Alexios Komnenos had to sail across the river on his way back to Ainos, which lies on the left bank of the Maritsa estuary into the Aegean Sea. Alexias, VIII.3.5. (p. 242). The description of events also shows that the Byzantine army crossed the river twice in total and after the second crossing it remained on the left bank, where Mount Lebounion is also located. See text below.
523.
Alexias, VIII.4.1. (p. 243); Vasil’evskij, “Печенеги,” 278.
524.
Alexias, VIII.4.1. (p. 243); Vasil’evskij, “Печенеги,” 278.
525.
Alexias, VIII.4.1. (p. 243).
526.
Ibid.
527.
Ibid. The information about 40,000 Kumans, as reported by Anna Komnene, is clearly exaggerated. Unfortunately, there is no way to arrive at a more accurate estimate. Kovács, “Campaigns,” 180, note 25. It can be assumed that there could have been no more than half or a quarter of the number mentioned by the Byzantine princess. The arrival of the Kumans was probably not as much of a surprise to the Byzantines as Anna Komnene pretends, because the Kumans had had strong motivation to fight the Pechenegs (see text above) since 1087 at the latest. Moreover, Alexios Komnenos’ letters with a request for military aid, which the Byzantine emperor had sent not only to the Latin West but “to all sides,” also reached the Kumans. See note 513 above. According to Shepard, this call for help could be related to the lead seal of the emperor’s brother-in-law sebastos and prōtostratōr Michael Doukas, found in Kuban in 1912. For this seal, see Bulgakova, Bleisiegel, 105. In the eleventh century, this territory was known as Zichia and belonged to the principality of Tmutorakan (see Pritsak, “Tmutorokan,” ODB, 2090), where Oleg Sviatoslavich was the ruling Riurikid prince during this period (1083–1094). He had friendly relations with the Kumans and since 1094 was on amicable terms with the khan Tugorkan. See Nestor, The Russian Primary Chronicle: Laurentian Text, transl. and ed. by Samuel Hazard Cross and Olgerd P. Sherbowitz-Wetzor (Cambridge, MA: Cambridge University Press, 1953), 179–180. Although the seal cannot be accurately dated, it could be seen as a possible clue that the emperor’s request for assistance had reached the Kumans precisely through this channel. Jonathan Shepard, “Mingling with Northern Barbarians: Advantages and Perils,” The Steppe Lands and the World beyond them, 230. Nevertheless, the overall context of this seal may be quite different; see the text below. The claim that Alexios Komnenos had contacted the Kumans directly has been challenged recently by Kovács. In particular, in line with Anna Komnene’s testimony, she argues that the arrival of the Kumans was surprising to the emperor and that both Kuman khans were not too readily inclined to the idea of fighting the Pechenegs together at first. She also refers to the emperor’s enduring distrust of the Kuman khans and the fact that there seem to have been secret negotiations between the Pechenegs and the Kumans (see text below). Kovács, “Campaigns,” 179. As already indicated, the astonishment of Anna Komnene may indicate a sort of pretense so as not to admit that her father—the Byzantine emperor—had to beg some barbarians for help. In this vein, it was far easier to note that the Kumans arrived out of the blue. Moreover, it would have been even less acceptable for the Byzantine princess to identify the Kumans as the real saviors of the Byzantine Empire. For this reason, too, she presented their arrival as coincidental, and fully attributed the credit for luring them into the alliance and making them participate in the battle on the Byzantine side to her father (but only as a momentous diplomatic ruse, not as a result of long-term diplomatic ties and negotiations dating back to 1087 at the latest). Therefore, in my opinion, it is most likely that the Kumans came at the direct request/invitation of the Byzantine emperor. This view is also held by Shepard, who formulated it back in 1996; see Shepard, “Substance,” 83–84.
528.
Alexias, VIII.4.2. (p. 243): “προαγοὶ δὲ πάντων ὁ Τογορτάκ, ὁ Μανιὰκ”; Chalandon, Essai, 132; Zlatarski, История, 204; Aléxios G. C. Savvídis, “Οι Κομάνοι (Κουμάνοι) και το Βυζάντιο 11^ος-13^ος αι. μ. Χ.,” Vyzantina 13, no. 2 (1985): 946. The two supreme commanders of the Kuman host are well-known khans Tugorkan (Тугоркань) and Böňek (Бонякъ) mentioned in Russian chronicles, for example in Nestor’s chronicle The Tale of Bygone Years. Nestor, 179 ff (Tugorkan), 182 ff (Böňek); Vasil’evskij, “Печенеги,” 279, note 2, Kovács, “Campaigns,” 185. Khan Tugorkan probably came from the Kuman tribe of Terter-oba and ruled over the right wing of the Kuman tribal union (west of the Dnieper). Omeljan Pritsak, “The Polovcians and Rus,’” AEMAe 2 (1982): 376. The main center of this horde lay between the lower Dniester and Dnieper. Peter B. Golden, “Imperial Ideology and the Sources of Political Unity amongst the Pre-Cinggisid Nomads of Western Eurasia,” AEMAe 2 (1982): 68. For reconstruction of the original form of his name, see Moravcsik, Byzantinoturcica II, 316. As a legendary figure, he was later adopted into Russian heroic epic poems, the so-called byliny. Peter B. Golden, “Cumanica V: the Basmils and Qipčaqs,” AEMAe 15 (2006–2007): 38–39. The other Kuman khan Böňek (Anna Komnene transcripts his name as Maniak; see text above) probably came from the Toks-oba tribe and, as the khan of the Kuman left-wing tribes (east of the Dnieper; since the mid-twelfth century this Kuman horde was in the Russian chronicles referred to as Polovci dikii, meaning “wild Kumans”), he had probably the highest position in the Kuman tribal union and was superior even to Tugorkan. Peter B. Golden, “The Polovci Dikii,” HUS 3–4 (1979–1980): 300; Golden, “Cumanica V,” 39. Pritsak argues that Böňek came from the Kaj-oba tribe. Pritsak, “Polovcians,” 368. The core territory of this horde was located on the central and lower course of the Siverskyi Donets river. Golden, “Ideology,” 68. For an overview of all Kuman tribal groups, see Peter B. Golden, “Cumanica IV: The Tribes of the Cuman-Qıpčaqs,” AEMAe 9 (1995–1997): 99–122.
529.
Alexias, VIII.4.3. (p. 243); Vasil’evskij, “Печенеги,” 280; Chalandon, Essai, 133; Zlatarski, История, 205; Angold, Empire, 133. The invitation of the Kumans to an opulent banquet in the presence of the emperor highlights the desperate situation in which Alexios Komnenos found himself. In fact, he badly needed to confirm the validity of the “alliance” concluded in 1087, and because of this he put aside all the recommendations of the court ceremony in dealings with barbarian chieftains, and basically negotiated with the Kuman khans as his equals. See Shepard, “Substance,” 92.
530.
Alexias, VIII.4.3. (p. 243); Vasil’evskij, “Печенеги,” 280; Chalandon, Essai, 133; Zlatarski, История, 205; Kólias, “Πολιτικὴ,” 259; Spinei, Migrations, 145.
531.
Alexias, VIII.4.4. (p. 244). Before the Byzantines crossed the Maritsa, there had already been minor skirmishes between the Kumans and the Pechenegs. It was only this fact that apparently convinced Alexios Komnenos of the credibility of the Kumans and the true nature of their hostile attitude toward the Pechenegs. However, he was still cautious and continued to distrust them. Shepard, “Substance,” 84.
532.
Alexias, VIII.4.4. (p. 244).
533.
Ibid. For Alexios Komnenos’ distrust of the Kumans, see above. Another reason may have been the reports of secret invitations to the negotiations that, according to Anna Komnene, were sent to the Kumans by the Pechenegs in those days. Alexias, VIII.5.1. (p. 245); Vasil’evskij, “Печенеги,” 281; Kólias, “Πολιτικὴ,” 259.
534.
Alexias, VIII.4.5. (p. 244); Vasil’evskij, “Печенеги,” 280–281; Chalandon, Essai, 133. Kaisar Nikeforos Melissenos himself was delayed by the gathering of troops and arrived at the battlefield only after the decisive battle at Lebounion. Alexias, VIII.6.3. (p. 250).
535.
Alexias, VIII.4.6. (p. 245); Zlatarski, История, 205
536.
Alexias, VIII.4.6. (p. 245); Vasil’evskij, “Печенеги,” 281.
537.
Alexias, VIII.4.6. (p. 245). Zlatarski mistakenly believes that the Kumans encamped at the Byzantine camp near Choirenoi. See Zlatarski, История, 205. The fact that the Kumans eagerly used the abandoned Byzantine field fortifications as their camping site seems to indicate that they set off for the military campaign against the Pechenegs under great time pressure and, therefore, did not bring with them wagons typical of nomadic peoples of the steppe (the Pechenegs had them, as described below), which usually served as the fortification of the nomadic encampment.
538.
Alexias, VIII.4.6. (p. 245). The main reason for the move of the Byzantines to the foot of Mount Lebounion was probably the local rugged terrain (especially difficult for the cavalry), which minimized the possibility of a surprise attack by the Pechenegs from the rear.
539.
Soustal, Thrakien, 333–334.
540.
Alexias, VIII.4.6. (p. 245).
541.
The described transfers of the Byzantine army before the battle also signify extreme caution on the part of Alexios Komnenos. Frequent changes of the campsite of the Byzantine army reduced the risk of a Pecheneg surprise attack.
542.
Alexias, VIII.5.1. (p. 245–246): “Οἱ δὲ Κόμανοι (…) ἑσπέρας μηνύουσι τῷ βασιλεῖ «μέχρι πόσου τὴν μάχην ἀναβαλώμεθα; ἴσθι τοίνυν ὡς ἐπὶ πλέον οὐκ ἐγκαρτερήσομεν, ἀλλ΄ ἡλίου ἀνατέλλοντος λύκου ἢ ἀρνειοῦ κρέας ἐδόμεθα”; Chalandon, Essai, 133; Zlatarski, История, 205. Vasil’evskij explained the meaning of the message, arguing that the expression lykou kreas (the flesh of wolf) was a reference to the Pechenegs (pagans) and the expression arneiou kreas (the flesh of lamb) to the Byzantines (Christians). See Vasil’evskij, “Печенеги,” 281.
543.
Alexias, VIII.5.1. (p. 245); Vasil’evskij, “Печенеги,” 260, 273, 281; Chalandon, Essai, 129, 133; Zlatarski, История, 205.
544.
Alexias, VIII.4.6. (p. 246); Vasil’evskij, “Печенеги,” 281; Chalandon, Essai, 133.
545.
Alexias, VIII.5.2. (p. 246). Anna Komnene refers to them as the “highlanders.” There have been sharp nationalistic disputes over their ethnicity. See, for example, Mátyás Gyóni, “Le nom de Βλάχοι dans l’Alexiade d’Anne Comnène,” Byz. Zeitsch. 44 (1951): 241–252; Vasil’evskij, “Печенеги,” 282, note 1; N. Bănescu, “Ein ethnographisches Problem am Unterlaufe der Donau aus dem XI. Jahrhundert,” Byzantion 6 (1931): 303–305; Chalandon, Essai, 133, note 3; Karagiannópoulos, Ιστορία, Γ΄, 59, note 220. Currently, it is mostly agreed that they were indeed the Vlachs. See, for example, Cheynet, “Effectifs,” 325. See also the latest summary of this issue in Madgearu, “Paradounavon,” 140; Alexandru Madgearu, “Vlach Military Units in the Byzantine Army,” Samuel’s State and Byzantium: History, Legend, Tradition, Heritage – Proceedings of the international scientific symposium ‘Days of Justinian I’, Skopje, 17–18 October 2014 (Skopje: “Euro-Balkan” University, 2015), 50.
546.
Alexias, VIII.5.5. (p. 247); Birkenmeier, Army, 76; Paroń, Pechenegs, 366.
547.
Alexias, VIII.5.5. (p. 247); Birkenmeier, Army, 76. See text above. The archaeological finds uncovered in Bulgaria (the sites of Dristra, today’s Kalugerovo northeast of Philippoupolis, and today’s Dimitrovgrad lying on the direct line between Philippoupolis and Adrianoupolis in the Maritsa valley) include the seals of Constantine Dalassenos with the titles of proedros and doux. See Jordanov, Corpus 2, 119–122. Once again, it would be very tempting to connect these findings to the battle of Lebounion, or to the restoration of the Byzantine administration in the provinces of Makedonia and Thrace immediately afterward.
548.
Alexias, VIII.5.5. (p. 247); Birkenmeier, Army, 76.
549.
See the description of the deployment of the united Byzantine-Kuman army in Alexias, VIII.5.5. (p. 247).
550.
This idea was first expressed by Birkenmeier. Birkenmeier, Army, 76.
551.
Anna Komnene does not explicitly mention the division of the Kuman army into two halves, but such a solution does not seem entirely improbable, since there were two supreme commanders of the Kuman host—khans Tugorkan and Böňek. Some support for such an explanation is provided in the passage of the text describing how the infantry ranks were surrounded by cavalry units; see note 548 above.
552.
Alexias, VIII.5.5. (p. 247).
553.
Alexias, VIII.5.5. (p. 247).
554.
Ara E. Dostourian, The Chronicle of Matthew of Edessa (New Brunswick, NJ, 1972), 90: “the emperor Alexius attacked the Pecheneg forces with his army, comprising three hundred thousand Roman, Latin, Armenian and Bulgar troops.” The number of soldiers under the emperor’s command is, of course, exaggerated.
555.
Anna Komnene does not mention the members of the Varangian guard at Lebounion specifically, but there is no reason to assume their absence. The Varangians fought in the battle of Dyrrachion in October 1081 against the Normans and also in September 1087 against the Pechenegs at Dristra. See text above.
556.
The emperor’s personal guard is not mentioned in the description of the Byzantine battle formation, but, due to its participation in the battle of Dristra in 1087 and its involvement in the skirmishes with the Pechenegs in Thrace between 1089 and 1090 (see text above), it is more than likely that it was also present at Lebounion.
557.
There is no mention of this Norman unit in the Byzantine service in the Alexiad either. Maniakatoi Latinoi took part in the battles against the Pechenegs in Thrace and at the end of 1090 their command was entrusted to Basileios Kourtikios. His presence in the battle of Lebounion (although mentioned only in connection with events after the battle; see Alexias, VIII.6.4. (p. 251)) also probably indicates the involvement of the whole unit in the battle.
558.
These were fresh recruits dispatched by Nikeforos Melissenos to reinforce the Byzantine army. Bulgarians in the ranks of the Byzantine army are also referred to by Matthew of Edessa; see note 554 above. However, it is impossible to determine whether these men were ethnic Bulgarians or whether they were members of the tagma of the province of Bulgaria. It is also possible to add 5000 Vlachs to these fresh recruits.
559.
The presence of the Paulikians can be assumed on the basis of the passage of Matthew of Edessa quoted above (see note 554 above), where they are referred to as the Armenians. The Paulikians represent the only element in the Byzantine ranks that could be perceived by Matthew of Edessa in this manner.
560.
The cavalry unit of the Archontopouloi had been constantly on the battlefield since its combat premiere in 1089 (see text above) and were most likely present at the battle of Lebounion.
561.
The presence of Norman mercenaries in the ranks of the Byzantine army was persistent during this period. Despite the fact that a tagma of Norman mercenaries was virtually wiped out in the battle of Dristra in 1087, another similar unit was again active under Tatikios’ command in 1090 (see text above). Therefore, the presence of such a unit in the ranks of the Byzantine army during the battle of Lebounion is more than probable, especially since we know that nōbelissimos Constantine Houmbertopoulos was on the battlefield that day and led a unit of the “Celts” (mercenaries from the Latin West or the Normans). See Alexias, VIII.5.5. (p. 247). It needs to be mentioned here that Houmbertopoulos, thanks to his Norman origin, usually commanded Norman mercenaries, or mercenaries from the Latin West in general.
562.
This identification is based on the fact that Anna Komnene consistently identifies Ouzas and Monastras as commanders of these units. Both of them had been in command of the Uzes in previous military clashes. See text above.
563.
This is confirmed by the order that Alexios Komnenos sent to Nikeforos Melissenos before the start of the entire campaign, in which he instructed him not to recruit into the army, if possible, men already enlisted in the army registers, veterans, and servicemen who served as garrisons of cities and fortresses, but to levy as many freshmen as possible and thus increase the numerical strength of the Byzantine military. Alexias, VIII.3.4. (p. 242).
564.
Alexias, VIII.5.5. (p. 247); Vasil’evskij, “Печенеги,” 283.
565.
Alexias, VIII.5.6. (p. 247); Birkenmeier, Army, 76.
566.
Birkenmeier, Army, 76. The number of the Pecheneg warriors is difficult to estimate; yet, Anna Komnene notes that each Byzantine soldier had thirty or more captives after the battle. Alexias, VIII.6.1. (p. 249). Nevertheless, this figure cannot be used because, besides the Pecheneg warriors, also their wives and children took part in the campaign, and their ratio to the number of fighters is very difficult, if not impossible, to assess. In general, however, it can be assumed that the number of men in the Pecheneg host was far greater than in the Byzantine army, but smaller than in the allied Byzantine-Kuman host.
567.
Paroń, Pechenegs, 366. Anna Komnene notes that soon after the start of the battle, the Pecheneg battle formation fell apart. Alexias, VIII.5.7. (p. 248). I believe that this did not mean the disintegration of their units, but only that, in view of the crushing numerical superiority of the Byzantine-Kuman forces, the Pechenegs had to abandon their positions, retreat, and seek refuge behind the wagon fort. From there, they continued to resist until sunset, when they were completely surrounded and their wagon fort was breached.
568.
Matthew of Edessa, 90: “Then, at a signal from the emperor Alexius, the Romans set fire to the chariots and burned them.” Anna Komnene does not mention the use of fire to break up the Pecheneg wagon fort. In terms of military tactics, however, breaking it was the best way to defeat the Pechenegs, otherwise their resistance would have lasted much longer and the Byzantine victory would have been more costly in terms of casualties. A similar combat strategy was used later by Alexios Komnenos’ son and successor, John II Komnenos, against the Pechenegs in the battle of Beroe in 1122. See Marek Meško, “Bitka pri Beroé 1122: Posledný boj Pečenehov,” Vojenská história 11, no. 4 (2007): 21. On the other hand, the forces of Kiev’s grand duke Mstislav Romanovich (1212–1223), surrounded by the Mongols after the battle at the Kalka river in 1223, were able to resist the repeated Mongol attacks for several days thanks to the use of the wagon fort until forced to surrender due to an acute shortage of drinking water. Nicolle, Shpakovskij, and Korol’kov, Kalka River 1223: Gengiz Khan’s Mongols Invade Russia (London: Osprey, 2001), 74–75, 81.
569.
Alexias, VIII.5.8. (p. 248); Vasil’evskij, “Печенеги,” 284.
570.
Alexias, VIII.5.9. (p. 249); Paroń, Pechenegs, 366.
571.
Alexias, VIII.5.8. (p. 248–249); Vasil’evskij, “Печенеги,” 284; Zlatarski, История, 207; Chalandon, Essai, 134; Kólias, “Πολιτικὴ,” 260; Angold, Empire, 133; Paroń, Pechenegs, 366.
572.
See text and note 501 in Chap. 4.
573.
Zonaras, 740–741; Ephraemus, Chronografia: Κείμενο – μετάφραση – σχόλια Οδ. Λαμψίδη. Τόμος Α’ (Athens: Kéntron ekdóseos ellínon syngrafeón, 1984), 123; Angold, Empire, 133; Birkenmeier, Army, 158; Mărculeţ, “Petchénègues,” 101. The first commander of the Pechenegs settled in Moglena was apparently tourmarchēs Joseph Maniakes, whose seal was published by Jordanov. See Jordanov, Corpus 1, 127; Jordanov, Corpus 2, 273–274; Jordanov, Corpus 3, 459–460.
574.
See text below.
575.
Vasil’evskij, “Печенеги,” 287; Angold, Empire, 133.
576.
Vasil’evskij, “Печенеги,” 322–323; Chalandon, Essai, 331; Shepard, “Substance,” 84–85, 123, 128. One of the many examples of their new role is the skirmish that took place between the Pechenegs and the crusaders, led by the count Baldouin of Bouillon in the spring of 1097, near the Byzantine capital. Steven Runciman, A History of the Crusades. Vol. I, The First Crusade and the Foundation of the Kingdom of Jerusalem (Cambridge: Cambridge University Press, 1954), 285.
577.
Kovács, “Campaigns,” 185. During the war against the Pechenegs, the province of Makedonia and Thrace certainly suffered great damage, and it can be assumed that its basic military organization and garrisons in many settlements were also in need of restoration. Perhaps it is precisely the renewal and organization of local military units that were to ensure security at various locations that are linked to the findings of two seals of megas domestikos of the West, prōtosebastos Adrianos Komnenos, uncovered at today’s sites of Kazanlak and Tsareva poliana. For these seals, see Jordanov, Corpus 1, 79–80; Jordanov, Corpus 2, 218–220; Jordanov, Corpus 3, 211, 366; Nesbitt and Oikonomides, Seals, 6–7.
578.
The first military commander (doux) of the restored province of Paradounavon became the aforementioned prōtoproedros Leo Nikerites (see note 300 above) who, in connection with this new assignment, was also awarded a higher court title of kouropalatēs. Zlatarski, История, 210; Madgearu, “Paradounavon,” 84; Madgearu, Organization, 87; Stephenson, Frontier, 103; Kühn, Armee, 226. For Leo Nikerites’ preserved lead seal with the title doux Paradounabou, see Shandrovskaja, “Деятели,” 41–42. Another measure that may reflect the reliability of the Vlachs incorporated in the Byzantine army ranks just before the battle of Lebounion is the likely creation of a separate military unit (tagma) composed and recruited exclusively from among the Vlachs coming from the Haimos mountains and deployed in the province of Paradounavon. This hypothesis is supported by the existence of a lead seal of a certain Georgios Dekanos, the general of the Vlachs (stratēgos tōn Blachōn), found in Issacea. For more detailed information, see Madgearu, “Paradounavon,” 140; Madgearu, “Vlach,” 50. For further information about the career of Georgios Dekanos, see Skoulatos, Personnages, 93–94.

Author information

Authors and Affiliations

University of Hradec Králové, Hradec Králové, Czech Republic
Marek Meško

Authors

Marek Meško
View author publications
You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

Copyright information

About this chapter

Cite this chapter

Meško, M. (2023). War Against the Pechenegs (1083–1091). In: Alexios I Komnenos in the Balkans, 1081–1095. New Approaches to Byzantine History and Culture. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-031-26296-8_5

Download citation

DOI: https://doi.org/10.1007/978-3-031-26296-8_5
Published: 15 June 2023
Publisher Name: Palgrave Macmillan, Cham
Print ISBN: 978-3-031-26295-1
Online ISBN: 978-3-031-26296-8
eBook Packages: HistoryHistory (R0)

Publish with us

Policies and ethics