Byzantine emperors desired stability and security in the peripheral regions of the empire so as to continue controlling and exploiting the productive lands which provisioned the principal cities, most importantly Constantinople; these also yielded tax revenues to support the apparatus of government. In the Balkans the vital regions were the rich lands of Thrace and the hinterland of Constantinople in the east, and Thessaly and the lands around Thessaloniki in the west. Security required direct supervision of major communication routes, by land and water, and of strategic cities across the peninsula, but only a stabilising influence in the mountainous interior, the north-eastern plains and the north-western littoral. Control of the Black Sea ports between Constantinople and the lower Danube, notably Anchialos, Mesembria and Varna, was considered essential, as was command of the major mountain passes through the Haemus mountains. Minor paths remained in the hands of locals, largely Vlachs, whose allegiance was assiduously cultivated. A similar situation prevailed to the west, where Albanians (Arbanoi) and Vlachs (Blachoi) had intimate knowledge of the tracks and defiles of mountains known to the Byzantines as the Zygos. Close regulation of the Egnatian Way, the principal land route between Constantinople and Thessaloniki which ran on to the Adriatic coast at Avlona and Dyrrachium, was a priority; so was control over the main land roads to the north, along the course of the Maritsa, Vardar and Velika Morava rivers. This required supervision of such cities as Skopje, Sofia, Niˇs and Braniˇcevo. However, there was never an attempt to establish a centralised administrative structure across the whole Balkan peninsula: stability was best ensured by retaining the allegiance of regional potentates and populations, including Serbs and Bulgarians, through a combination of force and favour. This would also, in principle, prevent insurrection or defection in the event of foreign invasion. Basil II (976–1025) was well able to assert his will in the periphery by virtue of the formidable reputation he established through regular campaigning, and by occasional acts of great brutality, for example mass blinding, which instilled fear in those who might contemplate resistance or rebellion. There is little evidence that Basil intended to conquer Bulgaria before 1014; rather, he used it as an arena for exercising his troops and as a source of booty and slaves.1 Following the death of Samuel of Bulgaria (987/8–1014), no individual was able to dominate the Bulgarian magnates, and the consequent instability forced Basil’s hand. In four years of hard, not always successful campaigning, he established his military superiority and received the submission of the magnates.2 Basil’s annexation of Bulgarian lands in 1018 involved primarily the military occupation of strategic towns and fortresses. Civilian administration was left to local potentates, who received stipends and honours from the emperor, and whose sons and daughters were married to those of Byzantine aristocrats. Members of the Bulgarian royal family were taken to Constantinople and absorbed into the hierarchy and ceremonial life of the imperial court. A few powerful chieftains were transferred with their retinues to the eastern frontier. Most of the lesser Bulgarian nobles were left in place, where they continued to levy taxes in grain and wine, now as representatives of the Byzantine emperor. These they passed on, in part, to the local garrisons. In his determination not to innovate in this matter, Basil will have considered the best way to supply an army.3 Supreme authority over the Byzantine forces of occupation rested with the strat¯egos autokrat¯or of Bulgaria, based in Skopje, the patrikios David Areianites.4 His most important subordinate was the patrikios Constantine Diogenes, who from 1018 was designated commander in Sirmium, which he had captured himself, as well as the neighbouring territories.5 It is possible that Diogenes’ title was strat¯egos of Serbia at this time.6 Diogenes later bore the elaborate title ‘anthypatos, patrikios and doux of Thessaloniki, Bulgaria and Serbia’.7 Further subordinate strat¯egoi, stationed in key cities as commanders of garrisons, were responsible for liaising with and monitoring local potentates. George Ostrogorsky noted astutely that ‘. . . the mention of a strat¯egos in any particular town by no means implies that this town was the centre of a theme’.8 We may take this a stage further, as did H´el`ene Ahrweiler, and state that the presence of a strat¯egos need not imply the establishment of a theme, except insofar as the term signifies military control of a locality or region. John Skylitzes uses the term in exactly this limited sense.9 Basil II’s efforts to consolidate military control of the north-western limits of Bulgaria have left clear traces in the archaeological record. At Sirmium renovations were undertaken on the walls, and a garrison installed.10 On the opposite bank of the Sava, at modernMaˇcvanksaMitrovica, a newepiscopal church was built, the third on the site.11 Similarly, a sixth-century church was renovated at Veliki Gradac, some way to the east of Sirmium.12 The restoration of ramparts, and relatively large number of bronze coins found there, suggest that Basil also installed a garrison at Belgrade.13 Excavations at Margum, at the confluence of the Velika Morava and Danube, have turned up seals and several coins from Basil’s reign.14 A new fortress was constructed at Braniˇcevo, at the confluence of the Mlava and Danube, and it grew in importance through the eleventh century.15 In contrast, there are no clear indications that Basil established garrisons in the interior highlands south of theDanube andwest of theVelikaMorava, namely in Raˇska and Bosnia. Instead, a line of small watch-towers studded the passes through the Zygos mountains west of the VelikaMorava corridor between Skopje and Niˇs. Excavations or surveys have identified several fortresses constructed or rebuilt in the eleventh century, including those at, from south to north, Lipljan, Zveˇcan, Galiˇc, Jeleˇc, Ras and Brvenik.16 In the later eleventh century we know that a no-man’s-land stretched to the west of these fortresses.17 Serbia, lying beyond this, was never to become a Byzantine administrative district. A seal struck by a katepan¯o of Ras has been convincingly dated to the reign of John I Tzimiskes (969–76), and may indicate that he enjoyed a brief period of recognition in Raˇska.18 This is apparently confirmed by the Chronicle of the priest of Duklja.19 Moreover, a seal has demonstrated that a command known as Serbia existed briefly, perhaps related to the recovery of Sirmium in 1018. However, these brief periods of intensified Byzantine presence never compromised the local Slavic power structures. Instead, Constantinople sought to work through local rulers, to whom titles and stipends were distributed. Thus, in a charter issued in July 1039, the Slavic ruler of Zahumlje boasted a string of imperial titles: ‘Ljutovit, pr¯otospatharios of the Chrysotriklinos, hypatos and strat¯egos of Serbia and Zahumlje’.20 Beyond Serbia, in Croatia and Dalmatia, authority was similarly exercised by local notables who were willing to recognise Byzantine overlordship. A seal has come to light which bears the legend ‘Leo, imperial spatharokandidatos and [arch¯on] of Croatia’.21 The use of the name Leo may suggest that the Croat in question had taken a Byzantine name, or a bride, or had been baptised by the emperor, or by one of his subordinates.We have examples of all such eventualities in the Balkan lands recovered by Basil II.22 A certain Dobronja, who also went by the name Gregory, accepted Byzantine money and titles in recognition of his authority in the northern Dalmatian lands. Charters preserved in Zara show that he had been granted the rank of pr¯otospatharios and the title strat¯egos of all Dalmatia. Kekaumenos records that he travelled twice to Constantinople as arch¯on and toparch of Zara and Split before 1036, when he was taken prisoner and later died in gaol.23 Dobronja’s change of fortunes is likely to have been the result of a change in Byzantine policy towards the western Balkans. Rather than patronise regional potentates and continue to raise taxes in kind, a policy was devised to extract gold from the periphery through taxation in coin. This initiative was taken in response to cash shortages within the state economy generally,24 and in particular to demands to redirect both coinage and manpower to the north-eastern Balkans, which were threatened by the Pechenegs in the 1030s.25 While the medium-term result was the increased monetisation of the region, the immediate consequence was a series of rebellions between 1040 and 1042. The first was started by a certain Peter Deljan in Belgrade and Margum. The second was a direct result of the first, and involved the troops raised in the theme of Dyrrachium to fightDeljan. Instead of fighting Deljan, they and their leader Tihomir joined him. The third rebellion was also a response to Deljan’s activities: a Bulgarian prince, Alusjan, observed his success with envy from Constantinople and determined to seize control of the rebellion to further his own interests. He succeeded in dividing support for Deljan and, ultimately, replacing him at the head of the rebel army. He then surrendered to the Byzantines. The fourth rebellion was an entirely distinct affair which arose in Duklja under Stefan Vojislav, who went by the title arch¯on and ‘toparch of the kastra in Dalmatia, Zeta and Ston’. According to Kekaumenos, Vojislav invited the strat¯egos of (Ragusa) Dubrovnik, named Katakalon, to act as godfather to his son, but kidnapped him en route to the ceremony.26 This confirms that a close working relationship – albeit one subject to arbitrary violation – existed between local potentates and Byzantine officers in this peripheral zone of the empire into the 1040s. Eventually, Vojislav’s rebellion was crushed with the aid of neighbouring Slavic potentates. If we can trust the Chronicle of the priest of Duklja, Ljutovit ‘princeps of Zahumlje’ mentioned above, as well as the ban of Bosnia and the ˇzupan of Raˇska, all welcomed Byzantine ambassadors offering piles of imperial silver and gold in return for military assistance against Vojislav.27 AfterVojislav’s demise,Duklja’s loyalty was to be ensured by the marriage of Vojislav’s son and successor Michael to a relative of Constantine IX (1042–55).28 Michael’s four sons by this marriage, and seven by another, extended their authority over neighbouring regions, including Zahumlje, Travunija and Raˇska, without provoking serious complaints or interference from Constantinople. In contrast, an increased Byzantine presence was introduced into the lands around Ohrid and Skopje, which comprised the secular and ecclesiastical centres of the theme of Bulgaria. In the 1040s we learn of a prono¯et¯es of (all) Bulgaria, the eunuch and monk Basil, who was also called the satrap¯es of Bulgaria.29 Subsequently a prait¯or operated alongside the military commander, now known as the doux of Bulgaria or Skopje. The prait¯or John Triakontaphyllos held the elevated rank of pr¯otoproedros, which was introduced c. 1060,30 and he may well have been a contemporary of Gregory, pr¯otoproedros and doux of Bulgaria.31 The autocephalous status of the Bulgarian church, with its prelate in Ohrid, had been guaranteed in directives issued by Basil II in 1019–20. Basil had left the church under a local archbishop, who could conduct services in Slavonic and communicate effectively with his subordinates; the latter were left in post throughout the Bulgarian lands, from the Adriatic to the Danube and Black Sea.32 In effect, the ecclesiastical jurisdiction of the archbishop of Bulgaria in 1020 matched that of the strat¯egos autokrat¯or of Bulgaria in military affairs. And just as the role of the latter diminished over time, so did that of the archbishop. By the mid-eleventh century a metropolitan of Dristra was appointed for the bishoprics of the lower Danube, a parallel to the emergence of a theme known as Paradounabon.33 In 1037 Leo, a Greek-speaking archbishop, was installed in Ohrid. He conducted services in Greek in the church of St Sophia, a domed basilica named after theGreat Church in Constantinople, but whose wall-paintings have parallels in contemporary structures in Thessaloniki,34 and it has been suggested that ‘the program of decoration provides an explicit statement of the imperial agenda for the newly reintegrated province’.35 There can be no doubt that the theme of Bulgaria was treated differently from semi-autonomous lands left in the hands of local rulers. However, the extent to which Bulgarian culture was hellenised under Byzantine rule is impossible to quantify, and there is no evidence for systematic suppression of the Slavic language, which continued to exist alongside Greek as a language of worship and literary production. A number of Old Church Slavonic liturgical codices, for example the mid-eleventh-century Epistolarium eninensis and twelfth-century Euchologium sinaiticum, attest the vitality of the language. Indeed, it is only in the twelfth century, not the tenth, that Slavic appears to have replacedGreek as the principal language of sung liturgies throughout Bulgaria.36 If the number of original Slavonic works of the highest order was small, translations of Greek works into Slavic continued uninterrupted into the twelfth century, including those of the Cappadocian fathers Basil of Caesarea, Gregory Nazianzen and Gregory of Nyssa. This production suggests an increasingly bilingual community of scholars in the major centres, but also in provincial monastic settings, for example the new Slavic foundations of St Joachim atMount Osogovo and St Gabriel at Lesnovo.37 There emerged also, for the first time, a genre with no contemporary parallels in Byzantine Greek: popular Slavic saints’ Lives composed in a simple style without rhetorical introductions and conclusions or long theological digressions, for example the Lives of the forementioned Joachim and Gabriel.38 The fifth Greek-speaking archbishop of Ohrid, Theophylact, wrote extensively from and on Bulgaria, elucidating the balance between secular and ecclesiastical affairs. From his letters we learn that taxation and conscription were pursued ruthlessly in the lands around Ohrid, and he appeals on numerous occasions to Byzantine administrators on behalf of locals.39 Although he often appears scornful of their rusticity, Theophylact took the care of his Bulgarian flock seriously and showed grudging respect for distinctive Bulgarian institutions in the Cyrillo-Methodian tradition.40 It would appear that the principal agents of hellenisation, such as Theophylact, had no desire to eradicate Slavonic, but rather allowed it to develop in an enriched cultural context.