TRAFFIC ACROss THe CILICIAN FRONTIeR IN THe NINTH AND TeNTH CeNTURIes: mOVemeNT OF PeOPLe BeTWeeN BYZANTIUm AND THe IsLAmIC NeAR eAsT

IN THe eARLY mIDDLe AGes

Koray DurakBoğaziçi University, Department of History, Istanbul, Turkey

There were many reasons for Near easterners to be in Byzantium. some were taken there against their wishes; some went there willingly to trade, to negotiate, or to die. some entered the Christian empire only once, while others paid frequent vis-its; some had a very fleeting glimpse of Byzantium while others settled there. Con-versely, Byzantines went to the Near east for many of the same reasons, and they travelled under similar conditions. The purpose of this paper is to show how var-ied the movement of people between Byzantium and the Islamic Near east was, us-ing the Cilician frontier in the ninth and tenth centuries as an example. exchanges of people who moved for a variety of reasons across the Cilician frontier have not been examined in the modern scholarship, and those modern works that focus on Islamic Cilicia or the Byzantine-Islamic frontier in Cilicia tend to emphasize the military aspect of the encounter1. more specifically, I ask whether the routes be-tween Byzantium and the Near east that were used, in the words of the late-tenth-century muslim geographer al-mu .kaddasī, to ransom prisoners, send dispatches, invade, or conduct trade were separate from each other, or whether people travel-ling with different motives, such as soldiers, captives, diplomats, exiles, merchants, and pilgrims, shared the same routes2.

The syrian-facing Cilician frontier was one of the major passages between the Byzantine and Islamic worlds in the ninth and tenth centuries, in addition to Trebi-zond and the Armenian routes in the east, and the mediterranean sea routes con-necting Cyprus and Crete to Greece and Anatolia in the south. The Cilician fron-

* I would like to thank the committee of TÜBA-GeBİP (The Turkish Academy of sciences, The Young scientists Award Programme) for its support.

1. haldon and Kennedy (1980) 79-116; lilie (2005) 13-21. Dimitroukas’ work on travel and communications in the Byzantine empire is an exception in this respect. dimitrouKas (1997) I 378-379, 390-392.

2. Al-mu .kaddasī 147.

KORAY DURAK142

tier was formed when the Umayyads consolidated their control of the Cilician plain gradually in the later seventh and early eighth centuries, and it remained stable un-til the mid-tenth century. The border passed roughly through the Taurus/Anti-Tau-rus mountain ranges, starting from the west of Tarsos, passing through muslim-held Germanikeia and melitene in the east. The transition of power in the Islamic world from the Umayyads to the Abbasids in 750 did not bring about drastic changes to the syrian frontier, but the frontier districts facing Byzantium were reorganized, start-ing under Hārūn al-Rashīd, into two defensive zones. The outermost zone (thughūr) was characterised by the presence of fortresses that bore the brunt of the Byzantine attacks, while the second, interior zone (‘awās. im) contained fortresses and fortified cities, which acted as elements supporting the thughūr. In the case of the syrian fron-tier facing Byzantine Anatolia, the thughūr extended from Tarsos to Germanikeia in Cilicia, while the ‘awās. im formed a line in northern syria from Antioch to manbidj. Concerning the Byzantine defences, the themata (themes) of Anatolikon and Arme-niakon stood the land incursions of the Arabs while the theme of the Kibyrrhaiotai faced Arab attacks from the sea for the seventh and early eighth centuries. The Klei-sura of Cappadocia, which was a theme by 863, and the Kleisura of seleukeia (mod-ern silifke), which Romanos Lekapenos promoted to theme status c. 930, faced the syrian thughūr by the tenth century, together with the theme of Anatolikon3.

Throughout the ninth and early tenth centuries, Cilicia and syria remained largely under the control the Abbasids, although the syrian thughūr entered Tulu-nid control between 878 and 896. In the mid- and later-tenth century, the Hamda-nid Amir sayf al-Dawla took control of mosul and Aleppo, extended the military confrontation with Byzantium throughout the whole frontier region, and added Cilicia to his domains4. However, this formidable foe of the Byzantines was obliged to resign against the expanding forces of the Byzantines, and his state was made into a tributary principality by 969. The efforts of Byzantine emperor Nikepho-ros Phokas in the 960s not only resulted in the capture of Cilicia but also in retak-ing control of northern syria for the Byzantines after approximately three centu-ries of Abbasid rule5.

3. canard (1966) 696-708; honigmann (1935) 40; ahrweiler (1962) 1-32; BosWorth (1991) 54-62; BosWorth (1992) 268-73; hild and hellenKemPer (1990) 43-53; lilie (2005) 13-21. For a de-tailed description of the frontier on the Taurus line, see Honigmann (1935) 80-85; haldon and Ken-nedy (1980) 85-87, 106-109. For the military organization of the frontier on the Byzantine side, see Honigmann (1935) 42-50; hild and restle (1981) 70-84. For the frontier defences on the Islamic side, see Canard (2012); BosWorth (2012).

4. canard (1966) 708-720; hild and hellenKemPer (1990) 53-59. For a detailed treatment of this period, see Canard (1968); BosWorth (1992) 273-278.

5. canard (1966) 720-722; hild and hellenKemPer (1990) 59-62. farag (1990) 44-47; ho-

TRAFFIC ACROss THe CILICIAN FRONTIeR IN THe NINTH AND TeNTH CeNTURIes 143

In the ninth and tenth centuries, political powers on both sides of the fron-tier created a highly militarized frontier through the erection of a network of forti-fications and regular raids to weaken the enemy. However, complete control of the frontier by the central political authorities was an ideological construct rather than a reality6. As Florin Curta shows, the shift in modern scholarship from the concept of “frontier-as-barrier” towards that of “frontier as a permeable zone” allows us to see continuities and connections rather than impermeable boundaries7. exchanges of people, objects, and information, the networks necessary for these exchanges, and the negative or positive impact of state control over these exchanges have be-come the new foci of frontier studies in history.

In a militarized zone that lay on the shortest route connecting the political centres of Byzantium and the Abbasid and post-Abbasid Near east, it is no sur-prise to find various cases of military, political, and diplomatic movement, which would have served to organize the relations between the political entities. The syr-ian thughūr and the central/eastern Anatolian plateau facing it were unquestion-ably the battlefields of Byzantine-Islamic military encounters. The Taurus/Anti-Taurus line, with its mountain passes, stood on the routes used both in seasonal raids and in large expeditions by caliphs8. As Ahrweiler’s study on the routes of Arab raids from the seventh to the ninth centuries painstakingly shows, the over-land routes that were most frequently visited by Arab raiders coming from the syr-ian thughūr were: A. the easternmost Germanikeia-Caesarea route or Northern syria-Caesarea route that passed through the al-Hadath/Adata defile (hereafter called the Adata Route); B. the route that connected Tarsos to Cappadocia via the Cilician Gates (hereafter called the Route of the Cilician Gates); C. the route that started from seleukeia and followed a northern diagonal line that cut across the Taurus mountains through Isaura and Claudiopolis (modern mut), reaching the Anatolian plateau in Laranda (hereafter called the Diagonal Route); D. the medi-terranean Coastal Route that connected Cilica to seleukeia and Attaleia (modern

nigmann (1935) 93-97.6. For various methods of frontier control, see PoWer (1999) 22-23. For frontier as an ideolo-

gical construct, see PoWer (1999) 24-25; eger (2008) 4-5, 20.7. curta (2005) 2-4.8. Two large military campaigns passing through the syrian thughūr occurred during the early

ninth century. The route of Abbasid Caliph al- ma’mūn’s forces in his grand campaign against Theo-philos in 830-831 was as follows: Baghdad-mosul-manbij in northern syria-Antioch-Tarsos-Koron in Cappadocia. The Abbasid army clearly passed through the Cilician Gates. Al-Ţabarī III/2 1103-1104. In the famous campaign of Caliph al-mu‘ta .sim against Amorion in 838, forces under the Caliph and the general Ashnās entered the Anatolian plateau through the Cilician Gates. Al-Ţabarī III/2 1237.

KORAY DURAK144

Antalya)9. These land routes and passes were used in the tenth century too, both by the Byzantine and the Islamic armies. The De Velitatione (Treatise on skirmish-ing), traditionally attributed to emperor Nikephoros Phokas, reflects the condi-tions of the eastern frontier in the early tenth century. This text counts the passes in the themes of seleukeia and Anatolikon, and in the area around Germanikeia and Adata, as routes which the Arab forces used to retreat to their own territory; the Byzantine forces must also have used these routes to pursue the enemy10. The construction of a fire beacon system from Loulon, north of the Cilician Gates, to Constantinople as an early warning system for Arab attacks shows how frequently the Route of the Cilician Gates was used for military purposes11. Byzantine emper-ors in their campaigns used these Cilician routes in order to descend from the Ana-tolian plateau to Cilicia. For instance, Basil I in his syrian campaign of 878 used the Cappadocia-Germanikeia route to transfer his army12.

A major type of exchange that resulted from military confrontation between the Byzantines and the muslims was the carrying of captives by the enemy forces. Innumerable references to the seizure of soldiers in battles and civilians in raids in the written sources of the period demonstrate that involuntary movement of people across the frontier was a very common occurrence. While some of the captives were sold into markets as slaves, others were kept as prisoners for exchange13. We have ev-idence for the occurrence of both cases in the Cilician region. The Abbasid geogra-pher ibn Khurradādhbih, referring to the raids into the Byzantine territories in Cili-cia in his Kitāb al-masālik wa’l-mamālik from the later ninth century, mentions fe-male slaves being carried on the back of she-camels descending from the mountains

9. ahrWeiler (1962) 4, 8-9, 10. For other examples of the use of the Route of the Cilician Gates, and the mediterranean Coastal Route, see avramea (2002) 75-76. For a detailed study of the routes from Cilicia to Cappadocia and Pamphylia that were mentioned above, see hild (1977) 51-59, 97, 132-140; hild and restle (1981) 124-127; hild and hellenKemPer (1990) 128-142.

10. Three Byzantine Military Treatises 230; Kazhdan and mcgeer (1991) 615. For the minor routes on the north-south axis connecting Caeserea to Rodandos and sision, see hild (1977) 118-129.

11. Theophanes Continuatus 197; Pattenden (1983) 258–99. The 11th century Byzantine histo-rian John skylitzes writes that the fortress of Loulon faced Tarsos. His remark draws attention to the close relationship between Loulon and Tarsos. Ioannis skylitzes 107-108.

12. Ioannis skylitzes 141. The famous general of Basil I, Andrew the scythian, met the armies of the amir of Tarsus at Podandos on the route of Cilician Gates. Ioannis skylitzes 143-144. Likewise, Nikephoros Phokas passed via the Cilician gates in his Cilician campaign in 964. Ioannis skylitzes 268. ramsay (1962) 348.

13. rotman (2005) 777. For the most recent discussion on Byzantium’s role in interna-tional trade of slaves, see rotman (2009) 57-81.

TRAFFIC ACROss THe CILICIAN FRONTIeR IN THe NINTH AND TeNTH CeNTURIes 145

to Tarsos14. The Persian geographer ibn al-Fa .kīh, too, writing in the early tenth cen-tury when Cilicia was still under Islamic rule, claims that slaves and eunuchs were brought from Byzantium. However, by the later tenth century, when the Byzantine armies controlled northern syria, the situation must have changed, because al-mu .kaddasī laments in the later tenth century that the slaves were no more brought from Byzantium to syria because of the wars15. In addition to being an important venue for the transportation of slaves, Cilicia was a major nexus for the movement of prisoners as well. Tarsos, Aleppo, and Baghdad were major Islamic cities where Byzantine pris-oners were kept, while Ikonion and Constantinople were major Byzantine locations where Islamic captives were imprisoned in the tenth century16. When one looks at the geographical distribution of the cities listed above, one can easily see that Cilicia stood in the middle of the prisoner routes, connecting to Baghdad via Aleppo, and to Constantinople via Ikonion. There are occasional references in the sources to the use of Cilician routes to transport captives. The amir of Tarsos, abu Thābit, who fell prey to the Byzantines in a raid in 900, was carried from the Taurus border to Constan-

14. Ibn Khurradādhbih 99-100. 15. Ibn al-Fa .kīh 148; Al-mu .kaddasī 242. michael Bonner states that plunder, and espe-

cially slaves, contributed to the local economy in Cilicia. Bonner (1996) 138.16. For prisons in the Near east, see Ibn miskawayh I 54-55 and II 193; Ioannis Caminiata 48;

BosWorth (1993) 189. For prisons in Byzantium, see al-Ţabarī III/4 2276-2277. military confron-tation between the Byzantines and the Tarsos people must have been extremely intense, because ibn Rusta in the early tenth century claims that one of the four major prisons in Constantinople was as-signed to those who came from Tarsus. iBn rusta 120-121.

Routes connecting Byzantine Anatolia to Islamic Syria in the ninth and tenth centuries.

KORAY DURAK146

tinople via Ikonion, following the Route of the Cilician Gates17. In another episode, when ‘Abd Allāh ibn Rashīd entered Byzantine territory via the Cilician Gates and was captured by Byzantine forces in a raid in 878, he was taken to Loulon, a city that stood at the entrance to the Cilician Gates, and then to Constantinople. The famous historian al-Ţabarī, who narrates in the early tenth century, adds a precious note re-garding this episode: he [ibn Rashīd] was taken to Lu’lu’an and then on to the tyrant along the post road18. This statement shows that captives were carried inside the Byz-antine empire through postal routes19. The same holds true for the Islamic routes as well. The author of the vita of st. Ioannikios from the later ninth century puts it very clearly when he describes the story of Byzantine prisoners escaping from a prison in an unmentioned location: They [prisoners] <then> followed the saint who led and transported them safely away from all the sentry posts watchtowers that the most wicked Hagarenes are accustomed to place on all the essential routes20. main routes on the Islamic frontier, manned with soldiers, must have been used for multiple pur-poses including transporting captives. moreover, prisoners were exchanged in highly regular and ritualized prisoner exchanges in western Cilicia. The River of Lamis21, which acted as a separation line between Byzantine mountainous Cilicia and the Is-lamic Cilician plain, was the scene of twelve major and six minor exchanges from 768 to 94622. In order to ransom soldiers and civilians taken captive in raids, mus-lims would leave from the city of Tarsos, where Byzantine captives were collected, and proceed to the mouth of the River of Lamis, where the Byzantines on their boats waited with captives of Near eastern origin23. A statement by al-mas‘ūdī reveals the itinerary of the prisoners of Near eastern origin who were brought to Lamis by the Byzantines. He writes that the prisoners were carried from Constantinople to Lamis via Pylai (modern Yalova), which lay to the southwest of the capital city and stood at the beginning of the land routes leading to the Anatolian plateau24. The most via-ble land route connecting Pylai via the Anatolian plateau to Lamis was the Diagonal

17. Al-Ţabarī III/4 2193.18. Al-Ţabarī III/3 1917. For the translation, see Al-Ţabarī, The Revolt of the Zanj 190.19. For the postal system and roads in Byzantine Anatolia, see avramea (2002) 59-60, 74-75.20. Byzantine Defenders of Images 330. For the Greek text, see Acta Sanctorum Novembris 2.1, 425-426.21. modern Lamas su in the district of modern day erdemli, mersin.22. For the list of prisoner exchanges between the Abbasids and the Byzantines, see Kolia-der-

mitzaKi (2000) 614-620. There were later prisoner exchanges when the Byzantine-Islamic frontier mo-ved towards syria in 960s, such as the exchange of 966 on the bank of euphrates. honigmann (1935) 82. For more information on these exchanges, see campagnolo pothitou (1995) 1-55; Khouri al odetallah (1983); Patoura (1994). For the performative nature of these exchanges, see duraK (forthcoming-a).

23. Al-Ţabarī III/4 227; Al-I.st.akhrī, 69; Ibn H. aw .kal 201.24. Al-mas‘ūdī 139-140.

TRAFFIC ACROss THe CILICIAN FRONTIeR IN THe NINTH AND TeNTH CeNTURIes 147

Route (the route that started from the Byzantine city of seleukeia, which lay to the west of Lamis, went north to cut across the Taurus mountains through Claudiopolis, and reached the Anatolian plateau in Laranda).

The routes connecting the syrian thughūr to Byzantine Anatolia were fre-quently used in two other types of politically-motivated movements: movements of diplomats and deserters. As Ioannis Dimitroukas’ review shows, the routes through Cilicia were major highways used by diplomats moving between Constantinople and Umayyad Aleppo or Abbasid Baghdad25. For example, ‘Abdal- Bā .kī, a frontier leader of the early tenth century from Tarsus, escorted the Byzantine envoys twice, in 918 and 924, from the frontier to Baghdad via “the euphrates Route”26. The euphrates Route was traditionally used to connect Baghdad to Byzantium via northern syria along the euphrates River. The route would start from Baghdad, proceed through al-Rakka, and enter Byzantium in Germanikeia or melitene.

However, the syrian thughūr was not simply a stop for diplomats travelling be-tween capital cities. It was occasionally a target of Byzantine diplomatic missions, while envoys from Cilicia visited Constantinople for negotiations. For example, in 859, emperor michael III sent a certain patrician with the title of logothete to the city of Loulon because the Arabs of Loulon decided to deliver the fortress to the Byzan-tines27. In the reverse direction, as we learn from the De Ceremoniis, Tarsiot envoys were present in Constantinople in the tenth century, and in 965, the people of mop-suestia and Tarsos sent an envoy to the emperor Nikephoros Phokas requesting to pay tribute in return for a truce. The emperor received the envoy, but did not accept the request. The local envoys must have travelled through the Cilician Gates28.

Routes through Cilicia also provided a major venue for deserters to escape into enemy territory. Wā .sif, the eunuch of governor of Azerbaijan, sought refuge in 900 in Byzantium when the plans of Wā .sif and his master ibn Abī al sādj to over-take egypt were revealed by the Abbasid Caliph al-mu‘ta .did. The Caliph captured Wā .sif in Ain Zarba/Anazarbus in Cilicia while he was preparing to escape into the Byzantine territory29. similarly, manuel, the strategos of Anatolikon Theme, escaped through the “syrian Kleisurai” to Baghdad during the reign of Theophi-los around 82930. Finally, the Byzantine general Andronikos Doukas, who served

25. Dimitroukas (1997) I 244-246.26. Ibn miskawayh I 53-54, 139.27. Al-Ţabarī III/3 1448.28. Constantine Porphyrogenitus, De Cerimoniis I 570-82; Ibn miskawayh II 210. 29. Al-Ţabarī III/4 2198.30. Georgius monachus Continuatus 796.

KORAY DURAK148

under Leo VI, revolted against the Byzantine state, and escaped to the Arabs in 906/907. The general left Ikonion (or Kabala near Ikonion) with his Christian fel-lows and two hundred muslim prisoners, whom he freed, and approached the mus-lims in Cilicia31. In none of examples mentioned above is there a reference to a spe-cific route that connected Cilicia to the Anatolian plateau. However, given the starting points and destinations of the deserters, it is very likely that the Route of the Cilician Gates or the Diagonal Route was traversed by the deserters in question.

All of the movements mentioned above resulted from political and/or military encounters. However, two categories of movement, those of merchants and pilgrims, took place outside the sphere of military confrontations or political negotiations. The role of trade, and especially international trade, on the syrian frontier is a debate on which there is no consensus in the modern scholarship. However, as I attempt to show in another article, Arabic and Greek sources from the ninth and tenth cen-turies contain direct and indirect references to merchants, commercial routes, cara-vans, and infrastructure for commerce (such as markets and inns in cities and on the roads), all of which point to the Cilician passes as major commercial routes, if not the only ones, between the Islamic and Byzantine realms32. Byzantine merchants were present in Cilician and syrian cities in the mid- and later-tenth century, and mus-lims from syria visited frontier towns of the Byzantines in order to obtain various commodities during the same period33. As the following story demonstrates, some merchants travelled further inland, beyond the immediate frontier. A Byzantine do-mestikos from Cappadocia, probably an important military official, made numerous appeals to ransom his son, who was captured by the muslims and kept in Aleppo, sometime around the mid-tenth century. Failing to free his son, the domestikos ap-pealed to a Christian perfume merchant from Aleppo, asking him to arrange the kill-ing of his son, rather than leaving him in captivity34.

The routes that were used for military attacks and carrying captives were also used for commercial purposes. Ibn Khurradādhbih presents the Route of the Cilician Gates as the main route connecting Baghdad to Constantinople in the later tenth cen-tury35. He does not specifically say for what purposes the route was used, but the fact that he draws the attention of the reader to the transportation of vegetables from Ni-caea to Constantinople on the same route tells us that he was aware of the commer-

31. Al-Ţabarī III/4 2276-2277.32. duraK (forthcoming-b).33. Three Byzantine Military Treatises 152, 162; Sowing the Dragon’s Teeth 156. 34. Ibn shaddād 196.35. Ibn Khurradādhbih 99-102, 113.

TRAFFIC ACROss THe CILICIAN FRONTIeR IN THe NINTH AND TeNTH CeNTURIes 149

cial use of the route. moreover, the presence of a funduq (inn/hostel) to the north of Bozanti (Podandos/al-Badandun) on the same route demonstrates that there were merchants carrying commodities on this route. The eastern Germanikeia-Caesarea route was also a corridor for trade, because we know that there was funduq between Germanikeia and samandu (Tzamandos)36. Finally, there must have been at least one route connecting syria to Byzantium via the Byzantine city of seleukeia, which was a central entry point for goods (specifically silk) coming from syria in the early tenth century37. After arriving in seleukeia, merchants and commodities could have followed 1. the Diagonal Route that was used for transporting captives; 2. the medi-terranean Coastal land Route from seleukeia to Attaleia through the land38; 3. the sea route connecting seleukeia to Attaleia and finally to Constantinople.

Finally, pilgrims traveled on trade routes in peaceful periods, and their pres-ence supports the possibility of trade connections in the same region. As michael mcCormick shows, the western european pilgrims of the early middle Ages (up to the year 800) followed the late antique pilgrimage route that passed through Con-stantinople or ephesus, and then connected to the Holy Land via Cyprus, while those traveling to Palestine after 800 preferred the southern route via egypt39. While west-erners preferred sea routes rather than the Cilician land routes, the ancient pilgrim-age route connecting Palestine to Constantinople via the Cilician Gates was used by the Byzantines from the Late Antique period to the central middle Ages, as Ioan-nis Dimitroukas shows40. Among the saints who travelled to the Near east, we have Gregory Akritas who passed from seleukeia to the holy land for pilgrimage in 78041; michael synkellos (d. 846), who travelled back from the holy land via seleukeia42; and Constantine the Jew, who travelled to Cyprus via Attaleia, and on his way back used the Diagonal Route from seleukeia to Nicaea sometime before 867 or 87443. These pilgrims may have made their way into syria from seleukeia either using Cilician routes or the sea routes. An interesting remark in .Ηudūd al- ‘ālam, a later tenth cen-

36. Dahabī in vasiliev (1950) 241. 37. Eparchikon Biblion 94: τὰ ἐκ Συρίας ἐξερχόμενα ἐσθήματα, ὁποῖα δ’ ἂν εἶεν, καὶ τὰ ἀπὸ

Σελευκείας καὶ ἀλλαχόθεν εἰσερχόμενα χαρέρια. 38. For the use of this route for trade, see Ibn .Haw.kal 201; Al-Idrīsī II 647.39. mccormicK (2001) 148, 171-173.40. dimitrouKas (1997) i 375-392.41. AASS Novembris 372-374: …ἀπάρας ἐπὶ τὴν Σελεύκειαν ἔρχεται, κἀκεῖ χρόνον οὐκ ὀλί-

γον ἐνδιατρίψας, ἄρτῳ βραχυτάτῳ καὶ ὕδατι ἔζη. τῷ δὲ εἰκοστῷ ἕκτῳ χρόνῳ τῆς ἡλικίας αὐτοῦ, ὅτε καὶ Λέων ὁ εἰκονομάχος τὸ ζῆν ἀπέρρηξε καὶ τὸ ὀρθόδοξον ἐπαρρησιάζετο, τὰ Ἱεροσόλυμα καταλαμβάνει…; mccormicK (2001) 197.

42. The Life of Michael the Synkellos 60-61.43. Dımıtroukas (1997) I 385; Kazhdan (1991) 506-507; Malamut (1993) 253.

KORAY DURAK150

tury anonymous geography in Persian, shows that Cilicia was not only a stop along the greater pilgrimage to Palestine, but had its own pilgrimage centers. The coastal city of Avlās/āyās (ancient Aigai) to the east of the Ceyhan River was home to two places which the Byzantines venerate and to which they go on pilgrimage44.

In conclusion, when al-mu .kaddasī wrote that routes leading to Byzantium were used to transport prisoners, to send envoys, to invade, and to conduct trade, he not only informed his reader about the close relations between the Byzantines and Is-lamic empires, but he also indicated that the same routes were used by people whose interests in passing to the other side arose from diverse and occasionally contradic-tory motives. The present study is an attempt to show that even a relatively small frontier region such as Cilicia was home to extremely diverse movements of people in the ninth and tenth centuries. An exhaustive examination of all of Byzantium’s fron-tier regions facing the Islamic Near east, from the Caucasus in the east to sicily in the west, would reveal an even more multifaceted relationship between the Byzan-tines and the Islamic Near easterners of the early middle Ages. The same situation applied to other parts of the Byzantine-Islamic frontier as well. For instance, Hārūn b. Ya .hyā, captured in Ascalon in the later ninth century, was first taken to Atta-leia by boat. He was transported to Constantinople, his route cutting a direct line through Asia minor. A century later, ibn .Hawqal describes this route connecting Attaleia to Constantinople through central-western Anatolia as a busy commer-cial route45. As these two cases show, the paths of merchants and captives coincided. Having answered in the context of early medieval Cilicia whether travelers with dif-ferent motives made use of the same routes, the next step in this inquiry would be how the travelers in question covered the same routes. Did the merchants, pilgrims, and diplomats use the same infrastructure, boarding on the same boats and stay-ing in the same inns? Did commerce and pilgrimage stop when armies traveled the routes in the region? And were there clear-cut distinctions among the categories of people moving between Byzantium and the Near east? In other words, could a dip-lomat or a soldier simultaneously act as a merchant and/or a pilgrim? examining Greek and Arabic sources from the middle Ages in the context of these questions will lead us from pondering about the infrastructure of roads to asking more nu-anced questions about the infrastructure of coexistence on the roads.

44. .Ηudūd al- ‘ālam 147.45. Ibn Rusta 119; Ibn .Hawqal 200-202.

TRAFFIC ACROss THe CILICIAN FRONTIeR IN THe NINTH AND TeNTH CeNTURIes 151

ABsTRACT

Cilicia, between Byzantine Anatolia and Islamic syria, was one of the loca-tions along the Byzantine-Islamic frontier where the movement of people between the two worlds was especially intense. The aim of this paper is first to identify which routes were used on the frontier in question and then to argue that the Cilician fron-tier in the ninth and tenth centuries acted as a venue for the exchange of people, not only for military purposes but for a variety of reasons ranging from diplomacy to pilgrimage. It is important to note that the same routes were used by people mov-ing for different reasons. The present paper focuses on the initial stages of the inter-course across the frontier rather than pursuing the story into the heartlands of the Byzantine and Islamic empires.

BIBLIOGRAPHY

Primary sources

AASS Novembris Propylaeum, Synaxarium Ecclesiae Constantinopolitanae. 1902. Acta Sanctorum Novembris 2.1, J. van den gheyn (ed.). Brussels 1984.Al-Idrīsī, Kitāb nuzhat al-mushtāq fī ikhtirāq al-āfāq. Cairo 1990.Al-I .sT.akhrī, m. J. de goeJe (ed.), Kitāb masālik al-mamālik [Bibliotheca Geogra-

phorum Arabicorum, vol. I]. Leiden 1967.Al-mas‘ūdī, m. J. de goeJe (ed.), Kitāb al-tanbīh wa-al-ishrāf, [Bibliotheca Geogra-

phorum Arabicorum, vol. VIII]. Leiden 1967. Al-mu .kaddasī, m. J. de goeJe (ed.), Aĥsan al-taqāsīm fī ma‘rifat al-aqālīm [Bib-

liotheca Geographorum Arabicorum, vol. III]. Leiden 1967.Al-Ţabarī, m. J. de goeJe (ed.), Ta’rīkh al-rusul wa al-mulūk, series I-III. Leiden

1964-1965.Al-Ţabarī, The Revolt of the Zanj, d. Waines (trans.), The History of al-Tabari,

vol. XXXVI: The Revolt of the Zanj. Albany, NY 1992. Byzantine Defenders of Images: Eight Saints’ Lives in English Translation, a.-m.

talBot (trans.). Washington, D.C. 1998.Constantine Porphyrogenitus, De Cerimoniis, J. J. reisKe (ed.), vol. I-II [Corpus

scriptorum Historiae Byzantinae]. Bonn 1829.

KORAY DURAK152

Eparchikon Biblion, J. Koder (ed.), Das Eparchenbuch Leons des Weisen. Vienna 1991.Georgius monachus Continuatus, Chronicon, i. BeKKer (ed.), Theophanes Contin-

uatus, Ioannes Cameniata, Symeon Magister, Georgius Monachus [Corpus scriptorum Historiae Byzantinae, XXXIII]. Bonn 1838.

h. udūd al-‘ālam, The Regions of the World, a Persian Geography 372 A.H. – 982 A.D., v. minorsKy (trans.). London 1937.

Ibn al-Fak. īh, m. J. de goeJe (ed.), Mukhtaşar kitāb al-buldān [Bibliotheca Geogra-phorum Arabicorum, vol. V]. Leiden 1967.

Ibn H. awqal, m. J. de goeJe (ed.), Kitāb S.ūrat al-ard. [Bibliotheca Geographorum Arabicorum, vol. II]. Leiden 1967.

Ibn Khurradādhbih, m. J. de goeJe (ed.), Kitāb al-masālik [wa-] al-mamālik [Bib-liotheca Geographorum Arabicorum, vol. VI]. Leiden 1967.

Ibn miskawayh, d. s. margoliouth and h. f. amedroz (eds. and trans.), Tajārib al-umam (The experience of Nations), vol. I-VII. Oxford 1920.

Ibn Rusta, m. J. de goeJe (ed.), Kitāb al-a‘lāq al-nafīsah [Bibliotheca Geographo-rum Arabicorum, vol. VII]. Leiden 1967.

Ibn shaddād, m. canard and h. gregoire (eds.), La dynastie Macédonienne, Ex-trait des sources arabes, vol. II/1. Brussels 1935.

Ioannis Caminiata, g. Böhlig (ed.), Ioannis Caminiatae de Expugnatione Thessa-lonicae. Berlin 1973.

Ioannis skylitzes, h. thurn (ed.), Ioannis scylitzae Synopsis historiarum. Berlin 1973. Sowing the Dragon’s Teeth: Byzantine Warfare in the Tenth Century, e. mcgeer

(ed.). Washington, D.C. 1995.The Life of Michael the Synkellos, m. B. cunningham (ed.). Belfast 1991.Theophanes Continuatus, i. BeKKer (ed.), Theophanes Continuatus, Ioannes Cam-

eniata, Symeon Magister, Georgius Monachus [Corpus scriptorum Historiae Byzantinae, XXXIII]. Bonn 1838.

Three Byzantine Military Treatises, g. t. dennis (ed.). Washington, d.c 1985.

secondary sources

ahrWeiler, h. (1962) L’Asie mineure et les invasions arabes (VIIe-IXe siècle). Re-vue Historique 227/1, 1-32.

avramea, a. (2002) Land and sea Communications, Fourth-Fifteenth Centu-ries, in a. e. laiou (ed.), Economic History of Byzantium From the Seventh through the Fifteenth Century, vol. I, 57-90. Washington, d.c.

Bonner, m. (1966) Aristocratic Violence and Holy War: Studies in the Jihad and the Arab-Byzantine Frontier. New Haven.

TRAFFIC ACROss THe CILICIAN FRONTIeR IN THe NINTH AND TeNTH CeNTURIes 153

BosWorth, c. e. (1991) Byzantium and the syrian Frontier in the early Abbasid Period, in m. a. BaKhit (ed.), Bilād al-Shām during the Abbasid Period: Pro-ceedings of the Fifth International Conference on the History of Bilād al-Shām, 54-62. Amman.

BosWorth, c. e. (1992) The City of Tarsus and the Arab-Byzantine Frontiers in early and middle ‘Abbāsid Times. Oriens 33, 268-273.

BosWorth, c. e. (1993) Abu Amr ‘Uthman al-Tarsusi’s siyar al-Thughur and the Last Years of Arab Rule in Tarsus (Fourth/Tenth Century). Graeco-Arabica 5, 183-195.

BosWorth, c. e. (2012) al-Thughūr. Encyclopaedia of Islam, second edition. Brill Online. Reference. Harvard University. 07 June 2012 http://referenceworks.brillonline.com/entries/encyclopaedia-of-islam-2/al-thughur-COm_1214.

camPagnolo-Pothitou, m. (1995) Les échanges de prisonniers entre Byzance et l’Islam aux IXe et Xe siècles. Journal of Oriental and African Studies 7, 1-55.

canard, m. (1966) Byzantium and the muslim World to the middle of the elev-enth Century, in J. hussey (ed.), The Cambridge Medieval History, vol. IV/1, 696-735. Cambridge.

canard, m. (1968) Les relations politiques de Byzance et des Arabes à l’époque de la dynastie Macédonienne, première période: 867 à 959, vol. I. Brussels.

canard, m. (2012) al-‘Awās. im. Encyclopaedia of Islam, second edition. Brill On-line. Reference. Harvard University. 07 June 2012 http://referenceworks.bril-lonline.com/entries/encyclopaedia-of-islam-2/al-awasim-sIm_0867;

curta, f. Introduction, in f. curta (ed.) Borders, Barriers, and Ethnogenesis. Frontiers in Late Antiquity, 1-9. Turnhout.

dimitrouKas, ch. (1997) Reisen und Verkehr im Byzantinischen Reich vom An-fang des 6. Jh. bis zur Mitte des 11. Jahrhunderts, vol. I-II. Athens.

duraK, K. (forthcoming-a) Performance and ideology in the exchange of prisoners be-tween the Byzantines and the Islamic Near easterners in the early middle Ages, in Performance in the Medieval and Early Modern Eastern Mediterranean.

duraK K. (forthcoming-b) The Cilician Frontier: A Case study of Byzantine-Is-lamic Trade in the Ninth and Tenth Centuries, in Centre and Periphery in the Age of Constantine VII Prophyrogennetos.

eger, a. a. (2008) The Spaces Between The Teeth: Environment, Settlement, And In-teraction On The Islamic-Byzantine Frontier. Ph.D.Thesis, University of Chicago.

farag, W. a. (1990) The Aleppo Question: A Byzantine-Fatimid Conflict of In-terest in Northern syria in the Later Tenth Century. Byzantine and Modern Greek Studies 14, 44-60.

haldon, J. and Kennedy, h. (1980) The Arab-Byzantine Frontier in the eighth

KORAY DURAK154

and Ninth Centuries: military Organization and society in the Borderlands. Recueil des travaux de l’institute d’études Byzantines 19, 79-116.

hild, f. (1977) Das byzantinische Strassensystem in Kappadokien. Vienna.hild, f. and hellenKemPer, h. (1990) Kilikien und Isaurien, Teil 1 [Tabula Impe-

rii Byzantini 5]. Vienna.hild, f. and restle, m. (1981) Kappadokien: (Kappadokia, Charsianon, Sebaste-

ia und Lykandos) [Tabula Imperii Byzantini 2]. Vienna.honigmann, e. (1935) Die Ostgrenze des Byzantinischen Reiches von 363 bis

1071. Brussels. Kazhdan, a. P. (1991) Constantine the Jew, in a. P. Kazhdan (ed.), The Oxford

Dictionary of Byzantium, vol. I, 506-507. New York.Kazhdan, a. P. and mcgeer, e. (1991) De Velitatione, in a. P. Kazhdan (ed.), The

Oxford Dictionary of Byzantium, vol. I, 615. New York. Khouri al odetallah, r. a. (1983) Άραβες και Βυζαντινοί. Το πρόβλημα των

αιχμαλώτων πολέμου. Thessalonica. Kolia-dermitzaKi, a. (2000) some Remarks on the Fate of Prisoners of War in

Byzantium (9th-l0th Centuries), in g. ciPollone (ed.), La liberazione dei “captivi” tra Cristianità e Islam, 614-620. Vatican.

lilie, r.-J. (2005) The Byzantine-Arab Borderland from the 7th to the 9th Century, in f. curta (ed.), Borders, Barriers, and Ethnogenesis. Frontiers in Late An-tiquity, 13-21. Turnhout.

malamut, e. (1993) Sur la route des saints byzantins. Paris.mccormicK, m. (2001) Origins of the European Economy: Communications and

Commerce A.D. 300–900. Cambridge.Patoura, s. (1994) Οι αιχμάλωτοι ως παράγοντες επικοινωνίας και πληροφό-

ρησης (4ος-10ος αι.). Athens. Pattenden, P. (1983) The Byzantine early Warning system. Byzantion 53, 258–299. PoWer, d. (1999) Introduction. Frontiers: Terms, Concepts, and the Historians of

medieval and early modern europe, in d. PoWer and n. standen (eds.), Frontiers in Question: Eurasian Borderlands, 700-1700, 1-31. New York.

ramsay, W. m. (1962) Historical Geography of Asia Minor. Amsterdam.rotman, y. (2005) Byzance face à l’Islam arabe, VIIe-Xe siècle. D’un droit terri-

torial à l’identité par la foi. Annales-Histoire, sciences sociales 60/4, 767-788.rotman y. (2009) Byzantine slavery and the Mediterranean world. Cambridge, mass.VasilieV, α. α. (1950) Byzance et les Arabes, vol. II.2: La dynastie Macédonienne

(867-959), Extraits des sources arabes (H. Grégoire and m. Canard eds.). Brussels.