Byzantium and Bulgaria. A comparative study across the early medieval frontier
1. Introduction 15
2. The Balkans in Late Antiquity and the Origin of Bulgaria
- The Background 21
- The Barbarian Invasions of the Fourth and Fifth Centuries 26
- The Advent of the Slavs 30
- The Origin of the Bulgarian State 45
- Bulgaria and Byzantium in the Early Ninth Century 49
3. Bulgaro-Byzantine Relations in the Ninth and Tenth Centuries 54
4. The Land 79
5. Cities 89
6. Industry and Trade 102
The Byzantine empire and Bulgaria in the ninth and tenth centuries did not resemble each other closely. If one undertakes a comparative study of them, it is not in order to distinguish the differences between two societies which appear to have advanced much the same distance along the same road, as one might compare say, Lübeck and Venice in the thirteenth century, Rome and Carthage in the third century B.C., or the Roman empire and Han China. The obvious counterpart to the middle Byzantine empire in such a comparison is the Abbasid Caliphate. An examination of their resemblances and their differences would be well worth undertaking. But it would be beyond the capacity of the present writer. And some of the essential preliminary researches are still unattempted.
The first reason for examining Bulgaria and Byzantium together is their geographical proximity. At the period in question they divided the Balkan peninsula between them and had a long common frontier. What difference did it make to an individual or a community to be on this or that side of the long frontier?
As we introduce further historical considerations, fresh points of contrast emerge. The Byzantine empire was the direct continuation of the Roman empire of classical times, with a political and administrative tradition running back uninterrupted for many centuries. The Bulgarian Kingdom was a relatively new arrival on the political map of medieval Europe. As a newcomer, it had open options, and could within certain limits choose the role it was to play. For the Byzantines there was little choice.
Another way of looking at the matter is suggested by the fact that the territory of the Bulgarian Kingdom had all—with the exception of the Transdanubian area, which was soon lost—formed a part of the Roman empire until the age of the successors of Justinian, and was reabsorbed into the Byzantine empire again at the beginning of the eleventh century. Did the few centuries of separation from the great Mediterranean world state lead to irreversible changes in the structure and functioning of society in these lost provinces? The question is an interesting one, and the answer may throw light on some aspects of the history of western Europe, where the separation was a permanent one.
If we look at the matter from the viewpoint of the economic historian, fresh questions pose themselves. In medieval terms the Byzantine empire—like the Muslim caliphate—was an economically advanced state, a super-power. Bulgaria belonged to the underdeveloped world. How did the difference in the level of their economic development affect their relations and the internal development of each of them? When we remember that during much of the period under study the frontier between Byzantium and Bulgaria ran within 200 kilometres of Constantinople, and within only 40 kilometres of Thessalonica, the second city of the empire, the steepness of the gradient of economic development becomes apparent. The rich and the poor of the medieval world faced one another across a very narrow gap, which the armies of either side could cross at will.
The sociologically-inclined historian will be struck by the rapid and decisive changes which took place in Bulgarian society in the period under examination. For him Byzantine-Bulgarian relations may be primarily the occasion for a study in acculturation, and in resistance to it. If Bulgaria formed part of what Professor Obolensky so aptly called ‘the Byzantine Commonwealth,’ it certainly did not become a mere carbon copy of Byzantium. There were striking differences. Some of the features of Bulgarian society recall the barbarian west rather than the Mediterranean lands of ancient culture. And even what is taken over directly from Byzantine models often shows a different profile in Bulgaria, though we may be hard put to it to define the difference.
It was the reflection of Byzantine civilisation in Bulgaria which served in its turn as model for Serbia and, more important, for Kievan and later Muscovite Russia. Thus the relations between Byzantium and Bulgaria in the ninth and tenth centuries were of far more than local and contemporary significance. It was through them that certain lasting features of the Orthodox Slavonic world were formed, features which have had an effect upon the development of European history down to our own day.
Looked at from another point of view, the Kingdom of Bulgaria was another barbarian successor state, comparable to Merovingian Gaul, Visigothic Spain, or perhaps Anglo-Saxon England. Yet, quite apart from its later reabsorption into the empire, Bulgaria differed in a number of important respects from most of the western successor states. There was far less continuity than in Gaul or Spain, as witnessed by the fact that Slavonic and not Greek or Latin became the current speech of the people and the language of state and church.
It may be that the most valid comparison is with Anglo-Saxon England. But the differences are many. Bulgaria was always a single, centralised state, not a group of principalities and ephemeral kingdoms. And Bulgaria was not remote from the ancient centre of the empire, like England, but right on its threshold, so that Bulgarian armies could again and again threaten Constantinople itself.
There are thus a number of considerations which might make a comparative study of Bulgaria and Byzantium in the ninth and tenth centuries of more than narrow specialist interest. One is bound next to ask whether there is enough evidence to permit a valid comparison. There are certainly great gaps in the evidence available to us. Neither for Byzantium nor, still less, for Bulgaria, have we the kind of detailed economic documentation that exists, for instance, for contemporary China of the late T’ang period. Can we say anything useful about such matters as taxation, trade, or distribution of wealth? In the present writer’s view we can, by seeing which of a limited number of possible models fit the known facts. This is indeed the only way in which the economic history of any but a few recent and favoured ages can be studied. In fact the historian of ninth-century Bulgaria is no worse off in this respect than the student of ninth-century England.
The archaeologist can provide useful material for the historian in this field. But there are certain snags. Byzantine archaeology has traditionally tended to concern itself above all with churches and their decorations. This is not unexpected, as until recently most Byzantine archaeologists were by training art historians. It has the unfortunate result that we know very little about the material background to everyday life. In a few special cases there has been complete and thorough excavation and recording of large inhabited areas. Examples are the American excavation of Corinth and of the Athenian Agora, or the Soviet excavation of Cherson. But much of the medieval material from these admirable excavations is still unpublished. From Constantinople and its vicinity nothing comparable is available. Bulgarian archaeology is in many ways more highly developed. The Hungarian Kanitz and the Czech Jireček provided reliable descriptions of ancient remains even before Bulgaria obtained its independence from the Ottoman empire. After independence patriotic feeling provided a stimulus to the investigation of the country’s past. Another stimulus was provided by Russian interest in a country which had contributed so much to Russian civilisation and which lay on the direct route to the Bosphorus.
The Russian Archaeological Institute in Constantinople carried out major excavations in a number of Bulgarian sites, and in particular at Aboba, which is generally identified with the ancient Bulgar capital of Pliska. The Czech scholar Karel Škorpil played an important part both in this work and in the training of Bulgarian archaeologists. In fact few Bulgarian ancient or medieval historians had not some archaeological training and experience. Since the Second World War and the establishment of the People’s Republic of Bulgaria the scale and quality of Bulgarian archaeological work has increased still further. And apart from the continuing excavation at Preslav and other major medieval sites there have been several interesting studies of the humble settlements of the Slavs and Proto-Bulgars. Most of this material is published rapidly. The important epigraphic monuments of the Bulgarian Kingdom have been edited in exemplary fashion by V. Beševliev. There is as yet no corpus of Byzantine inscriptions. However there are some difficulties. The most important is that scarcely any site or monument from the early medieval period has so far been reliably dated. Thus all dates established by archaeological evidence must be regarded with some scepticism. And in particular the attribution of one monument to the late Roman period and of another to the Bulgarian period is often dependent on subjective considerations. We cannot yet distinguish with any sureness between the handiwork of provincial Roman architects of the age of Justinian and that of Bulgarians—or Byzantines—working for King Boris.
Another shortcoming of our evidence is that we see Bulgaria almost exclusively through Byzantine eyes. The period under study is covered by a series of Byzantine historians and chroniclers whose narratives survive —Theophanes the Confessor, George the Monk, Genesius, the Continuators of Theophanes—as well as by contemporary Byzantine official writings such as the legal enactments of Basil I and Leo VI and the treatises on foreign policy—De Administrando Imperio—and on palace ceremonies of Constantine VII Porphyrogenitus, and by the letters and occasional compositions of several of the principal participants in the events—Photius, Arethas, Nicolaus Mysticus, Leo Choirosphaktes. From Bulgaria we have no historical narrative, no administrative texts, no personal documents, apart from some diplomatic correspondence of King Symeon in Greek. What survives in Old Slavonic is mainly translations or adaptations of religious works. But the prefaces to these often contain information on contemporary Bulgarian life. There is also a Bulgarian law code in Slavonic.
From western Europe we have odds and ends of minor observations on Bulgaria and the Bulgarians, and the very important but scrappy and disconnected Responsa of Pope Nicholas I to the queries addressed to him by King Boris shortly after his conversion. There is consequently a great deal that we do not know about Bulgaria. Apart from its rulers, few men or women emerge as credible, rounded personalities. The floodlight of history did not play on Bulgaria as it often did on Byzantium.
Denn die einen sind im Dunkeln
Und die andern sind im Licht;
Und man siehet die im Lichte,
Die im Dunkeln sieht man nicht.
The treatment which follows is determined largely by the availability of evidence. Many important topics are passed over in silence or only touched upon. The need to make, not a static comparison of two societies at a moment of time, or during a brief period, but an examination of the changing relations between two communities during a long period, has also affected the form of the treatment. The changing fortunes of the Balkan peninsula in late antiquity and the early Middle Ages are described in an initial chapter, with special attention to ethnic changes and changing settlement patterns. In the following chapter an account is given of the course of Byzantine-Bulgarian relations in the ninth and tenth centuries. These two chapters are diachronic in their treatment rather than synchronic. There follows a series of chapters in which a comparative study is made of aspects of the two societies, beginning with the land and ending with culture and the patterns of everyday life. A brief final chapter takes up again some of the wider questions raised in the introduction.
Source references have not generally been given for most of the events recounted in the narrative portions of the book. They would have enormously increased the bulk of the work, and they are already available in standard works on the history of Byzantium and of Bulgaria. More frequent reference is made both to the sources and secondary works in the comparative chapters, though here again no attempt is made to quote all the evidence for everything that is said. In the critical bibliography at the end of the book I have tried to provide a guide to the formidable literature on the subject rather than to list every book used or quoted. The non-specialist wants to know where to turn for a balanced and authoritative account of this or that topic.
The specialist will already have compiled his own bibliography. And he will certainly have to read Bulgarian and Russian.
In transliterating foreign names I have aimed at ready intelligibility rather than complete consistency. Greek names, whether of persons or places, are rendered by English equivalents, by Latinised forms, or by transcription, in the way most likely to be familiar to English readers: thus John, Athens; Photius, Aetolia; Choirosphaktes, Serrai. Old Slavonic names are transliterated with ch, sh, zh rather than č, š, ž; the yers (very short vowels) are rendered by ŭ and ĭ except at the end of words, where they are omitted, thus Boris and not Borisŭ; yat (a variety of long e) is rendered by ě. Modern Bulgarian names are transcribed according to the same system, and follow the post-1945 orthography; in particular the dull vowel represented in Cyrillic by yer is rendered by ŭ and not by ă. Any departures from these principles are the result of oversight.
2. The Balkans in Late Antiquity and the Origin of Bulgaria
It is necessary to follow the history of the Balkan area in some detail during the period from the fourth century A.D. to the middle of the ninth for several reasons. First, the gradual appearance in the area of the Roman empire of the Slavs and the Bulgars, the two basic ethnic elements in the medieval Bulgarian state, and their changing relations with each other and with the surviving remnants of earlier populations can only be traced if we begin our story in late antiquity. Second, the reasons for the peculiar ethnic mosaic, the cultural collapse, and the demographic decline which characterise the northern Balkans in the early days of the Bulgarian state are buried deep in the earlier history of the region. Third, the political, religious and strategic choices open to Byzantines and Bulgarians in the ninth and tenth centuries become much clearer if we examine their earlier relations, and the gradual development of the confrontation between them in its peculiar configuration.
The account must necessarily be a complex one, in which narrative and analysis are closely interwoven, and in which the main line of the narrative is from time to time interrupted to pick up the distant beginnings of a train of events which later becomes of crucial importance. In particular the early history of the Slav peoples and of the Bulgars calls for such ‘flashback’ treatment.
I have preferred to err by giving too much detail rather than too little. Partly this arises from the conviction that a knowledge of the past of any society is essential for the understanding of its present—the shibboleth which distinguishes the historian from the sociologist. Partly it is doubtless due to an irrepressible interest in the subject.
By the fourth century of our era the whole of the Balkan peninsula had formed a part of the Roman empire for three hundred years. From the point of view of population and history it fell into two main zones. In the south, in the area roughly corresponding to modern Greece, the inhabitants were uniformly Greek-speaking and the city, with its surrounding dependent territory, was virtually the only political organisation. The few non-Greek communities planted there, such as Roman colonies in Corinth, Megara, Patrae, Dyme in Achaea, Cassandreia in the Pallene peninsula, Philippi and Pella and the Italian merchant groups in such cities as Corinth had long ago been Hellenised.
The same is true of the Macedonians, the people of southern Epirus, whose claim to be regarded as Greek had in earlier centuries not been universally recognised. This region — peninsular Greece and the islands — was one of ancient civilisation. Most of its political communities long antedated the conquest which brought them under Roman control. On the whole Roman power had meant stagnation, dwindling population and crumbling economy in Greece. There were exceptions. Athens drew students from the upper classes of the whole Roman empire because of the fame of its schools of rhetoric and philosophy. The transfer of the central administration of the empire by Constantine to his new capital at Constantinople gradually stimulated the economic life of the eastern European provinces.  But by and large Greece was an economic and political backwater in the fourth century, and probably underpopulated in comparison with earlier ages. It is significant, for instance, that after the Gothic raid of 267, most of the area inside the walls of Themistocles at Athens was abandoned, and a new wall was constructed surrounding the Acropolis and a small area to the north, within which settlement was concentrated. 
To the north of Greece lay a geographically and ethnically much more diversified area. The line of separation of the two corresponds closely, though not exactly, with that marking the northern limit of the cultivation of the olive. Though generally mountainous, the northern Balkans also comprised extensive plains, often watered by perennial rivers, some of which were navigable, as well as extensive areas of rolling hill country. The mountain belt running behind the Adriatic coast—the Dinaric Alps—was barren and treeless in ancient times as today. But the Balkan chain, running across the eastern part of the peninsula from east to west, south of the Danube, the Rhodope mountains, the mountains of western Macedonia and Albania were wooded, and often provided high summer pastures above the forest line. The main ethnic groups in the north Balkan area had been Illyrians in the west, Thracians in the east, and Daco-Moesians between the Balkan range and the Danube.  All groups had been subjected for centuries to influences from the Mediterranean world. Greek cities had been founded along the southern and eastern coasts of Thrace. The native Thracian political organisations, which had the form of confederacies or short-lived empires rather than lasting centralised states, had been subjected first by Macedonia, then by Rome (though a client Kingdom of Thrace was permitted to exist until 46 A.D.).
The Thracian ruling class had been thoroughly Hellenised south of the Balkan chain. Between the mountains and the Danube Greek penetration had been much less intense. Political independence lasted much longer. It was not till early in the first century of our era that Moesia, as this region was called, became a Roman province. Roman garrisons were established along the Danube, and colonies of veterans were settled at a number of points. The Thracian upper class here became Romanised rather than Hellenised. By the fourth century both parts of the Thracian area were studded with cities, each with its dependent territory. But there were still large areas which did not belong to any city, but were administered directly by imperial officials. Most of the population south of the Balkans spoke Greek, those north of the Balkans and those west of a line running roughly from Sofia to Durazzo, spoke Latin. Thracian continued to be spoken in the countryside though we cannot estimate the extent of its survival, any more than we can estimate how much Gaulish was spoken in Gaul at the same period. And some sense of Thracian ethnic unity survived. We hear of the suppressing of cult sites of Thracian gods about 400;  the Bessae, a Thracian tribe, are alleged to have been converted to Christianity by Nicetas of Remesiana in 396;  we hear of Thracian monasteries in Palestine and in Constantinople as late as the sixth century.  Thrace served as a reservoir of military manpower in late antiquity, and men with Thracian names, practising Thracian religious cults, are found all over the empire.  On the survival of Thracian speech and ethnic consciousness in the fourth century and later cf. V. Beševliev, Personennamen bei den Thrakern, 1970, pp. 69-137, a valuable collection of material which must however be used with some caution.
In the western section the situation was similar. There had been in the course of the centuries considerable Roman settlement among the Illyrian population. Cities had grown up, not only on the Adriatic coast, where they sometimes occupied the site of Illyrian tribal capitals or of Greek colonies, but also in the interior, particularly along the major river valleys. By the fourth century the bulk of the population had been Romanised in the sense that they spoke Latin in public. But the continuing survival of the Illyrian language, probably in the mountainous area of northern Albania and the neighbouring region of Yugoslavia, is proved by the existence today of the Albanian language, the core of which is Illyrian.
Throughout the northern Balkans there was already great ethnical diversity, pockets of Goths, Bastarnae (probably a Celtic-speaking people), Carpi, Sarmatians, Scythians, Alans (probably all Iranian-speaking) and others being found among the Thracians or Illyrians.
This state of affairs was the result of centuries of the establishment of conquered or refugee communities in these territories by the Romans, and of the settlement of bodies of Roman veterans, as well as of individual migration. It is hard to determine the extent to which these enclaves preserved their language and way of life. Many were certainly absorbed into the basic population of Romanised or Hellenised Illyrians and Thracians. Yet they must have made some contribution to the common culture of the area. And they certainly contributed to the extinction of old tribal and ethnic traditions.
Through the Balkan provinces of the Roman empire ran a number of roads of strategic significance. They generally followed the line of earlier roads or lines of communication. The most important was that which ran south-eastwards from Aquileia via Emona (Ljubljana), Siscia (Sisak), Singidunum (Belgrade), Naissus (Nish), Remesiana (Bela Palanka), Serdica ( Sofia ), Philippopolis (Plovdiv), and Hadrianopolis (Adrianople) to Constantinople. Cutting through the peninsula from west to east ran a Roman military road, the via Egnatia which, starting from Dyrrhachium (Durazzo) on the Adriatic ran through central Albania, rounded lake Ohrid on the north, passing through Lychnis (Ohrid), and Edessa to Thessalonika, then following the low ground between the mountains and the sea to Constantinople. These two main roads were linked by another which, leaving the Belgrade-Constantinople road at Naissus followed the valley of the Morava, passed over the easy col to join the Axius (Vardar) at Scupi (Skopje), and follow its valley via Stobi (near Veles) to Thessalonika. A less important link road left the main diagonal road at Serdica, proceeded up the valley of the Oescus (Iskar), over the col to the headwaters of the Strymon (Struma), and thence down the river valley to join the via Egnatia a little to the north-east of Thessalonika. Another road ran along the south bank of the Danube from Singidunum to Durostorum and on to the Black Sea at Tomis. It was linked to the diagonal road by secondary roads following the passes through the Balkan chain, of which the most important were those running from Serdica down the valley of the Oescus to Oescus (Gigen) and from Philippopolis via Nicopolis (Nikyup) down the Iantrus (Yantra) to the Danube. Finally, a road followed the coast of the Black Sea from the mouth of the Danube via Odessus (Varna), Mesembria (Nesebur), Develtos (near Burgaz), and Salmydessus (Kiyikoy) to Constantinople.
These Roman roads were protected by a system of military posts and forts, particularly when they passed through defiles. Some of these had gathered a population of veterans and traders around them and become veritable cities. Long after the last Roman soldier had left great areas of the Balkans, the Roman roads and their supporting works continued to exist, and to determine the routes of movement and the zones of settlement of the various peoples who entered the Balkans. Those who held the roads could move more or less freely along them. But they often had no control of the areas between the roads, except in so far as they patrolled them in strength.
In the fourth century the Danube was the frontier of the Roman empire. On one side lay Roman provinces, paying taxes to the Roman government and subject to Roman law. On the other lay a patchwork of client states, tribes and ephemeral political groupings over which Roman control was at best indirect, by means of treaties and subsidies, and at times non-existent. For a time, from the beginning of the second century A.D. to late in the third, direct Roman control had been extended north of the Danube into Dacia (Transylvania and western Wallachia and Latin-speakers from various parts of the Roman world had settled there as soldiers or traders. But the growing pressure of peoples on the move north of the Danube forced the partial abandonment of the new province under Gallienus in 256 and its total evacuation under Aurelian in 274.
The Danube frontier was garrisoned in the fourth century by a large army based upon a system of fortifications in depth and supported by a strategic reserve which could be moved from region to region as the situation demanded. The distinction between the limitanei, based on strong points in the frontier zone and maintained by land-holdings, and the comitatenses, the mobile field army under direct command of a general attached to the imperial court, had already begun to be blurred by the transference of units from one category to the other, but the principle upon which it was based still held. Among both categories of troops there were many soldiers recruited as individuals from territories beyond the frontier, as well as bodies of foreign mercenaries serving under their own officers—the foederati. The bulk of the army was probably recruited from the Balkan area, though there would from time to time be units of Asian, African or other origin to be found along the Danube frontier. The permanent garrison on the Danube frontier in the middle of the fourth century has been calculated to have been about 145,000 strong — 64,000 on the lower Danube,
and 81,000 in the upper Danubian provinces, which do not directly concern us here (cf. A.H.M. Jones, The Later Roman Empire, 1964, 682-3). There were also several flotillas based on harbours on the south bank and patrolling the river during the nine months of the year when it was open for navigation.
This formidable defence system, which was naturally complemented by diplomatic activity, distribution of subsidies etc. north of the river, had scarcely ever been penetrated, apart from raiding operations of purely local significance. The only exception was during the period of political instability and protracted civil war in the middle of the third century, when a force of Goths from the middle Dniester area had crossed the Danube and marched through the Balkan provinces collecting booty until they reached Athens, where they sacked the lower city but were unable to take the Acropolis. But these invaders withdrew at once beyond the Danube. Though ready to break through the frontier of the empire when they could, the barbarians took its existence for granted.
The Barbarian Invasions of the Fourth and Fifth Centuries
Such was the situation when in 376 the Alans, an Iranian-speaking nomadic people moving westward through the steppe zone, overwhelmed the kingdom of the Ostrogoths in the Eastern Ukraine and drove their kinsmen, the Visigoths, from their home on Dniester. The Visigoths fled to the Danube and asked the Roman authorities to enrol them as foederati and to grant them land in Moesia south of the Danube. There were probably already some small Gothic communities settled between the river and the Balkan mountains, among whom Christianity had made some progress. After much negotiation, the Visigoths were ferried across the Danube at Durostorum. They numbered several tens of thousands of adult males, together with their families. Once in Roman territory they were treated by local officials with a combination of dishonesty and inefficiency which both aroused their resentment and reduced them to the verge of starvation. In 378 they turned against the Romans, and were joined by a number of Gothic slaves from the cities and estates of the Balkans as well as by men condemned to penal servitude in the mines in various parts of Thrace, who brought with them valuable technical skills. The Gothic army, mainly a cavalry force, swept through Thrace and on 9 August 378 defeated the Roman field army at Adrianople, killing the emperor Valens.
The Goths, now joined by groups of Alans and Huns from beyond the Danube, who scented booty, advanced on Constantinople, but were unable to take the city. The new eastern emperor, Theodosius, by a combination of diplomacy and force, and by taking on large bodies of Goths as foederati, succeeded in averting the most acute danger. But a substantial rump of armed Goths remained at large for a generation in the Balkan provinces, living off the land, destroying the economic infrastructure, and disrupting the established patterns of life. Unlike previous raiders, they were present in force and they meant to stay. After protracted negotiations and a series of abortive campaigns the Visigothic force was finally diverted into Italy, where in 410 it captured and sacked Rome, and thence to southern Gaul and ultimately to Spain. There a Visigothic kingdom was set up which survived until it was destroyed by the Arabs in the early eighth century.
Though no independent Gothic force remained in the Balkans, a great many Goths were settled on the land there as Roman soldiers, adding a new element to the population. We do not know where the main areas of settlement were. More important, a very long period of security had come to an end. The Turkic Huns, who followed upon the heels of the Alans across the steppe zone, invaded Thrace in 408-9, again in 422, and almost every year thereafter until after the middle of the fifth century. The details of their invasions are of little importance, but they continued to undermine the social and economic balance of life in the northern Balkans. Unlike the Visigoths, they made no serious penetration of peninsular Greece. It was not merely a matter of physical damage, which could always be made good: the expectation each year of further invasions in succeeding years made men loth to repair damage as it occurred. The future was heavily discounted. Small holdings thus fell into disrepair and were often in the end abandoned as the cultivator or his heir left to swell the growing numbers of landless men in regions of the empire less exposed to attack. Patterns of trade were broken and never restored, industrial installations destroyed and never replaced. The troops sent by the government in Constantinople, with their demands for supplies at fixed prices, for quarters and for forced labour, were sometimes no less of a scourge than the invading Huns. And the tax-collector, who might be tolerated when taxes bought security, became the most un welcome of visitors when they no longer did so. Some men decided to quit the empire altogether and throw in their lot with the barbarians, among whom both the rate of taxation and the efficiency with which taxes were collected were less than in the Roman empire.
The Huns, whose own way of life was highly specialised for mobility, had need of craftsmen, merchants and men with elementary administrative skills, and they often welcomed Roman defectors’. Priscus of Panium, who accompanied a Roman embassy to the court of Attila in 448 or 449, reports a highly revealing interview with such a defector, who was certainly not the only one of his kind. Finally, some of the inhabitants of the northern Balkan provinces were carried off as captives by the Huns and either kept by their new masters or sold back as slaves to the Romans. So the first half of the fifth century must have seen demographic changes in the area the extent of which it is impossible to estimate, as well as extensive destruction of agricultural and industrial installations and a disastrous fall in public confidence.
Throughout this time, and indeed for a century and a half later, the frontier was still held in a military sense. The chain of forts along the Danube was still garrisoned. Strong points which were taken by raiders were soon rebuilt. Indeed there was a good deal of new military building, repair and strengthening of city walls and so on. But it was clear that the army which for four centuries had protected the peaceful life of the Balkan provinces was now no longer able to do so. It could neither keep invaders out, nor prevent them getting back to their bases north of the Danube, laden with booty. It continued to be an ethnically mixed force. And to the mixture there was added in the second half of the fifth century a new element, the scarcely Hellenised Isaurian mountaineers of the Taurus range in Asia Minor.
The second half of the fifth century saw less frequent and less disastrous raids than the first half. But there was no long period of peace and security during which the damage of the two preceding generations could be repaired. Attila died in 453, and the Hun state, which had depended largely on his personality, broke up. The resulting splinter groups continued to raid the Balkan provinces, though they did not usually penetrate very deep. The Ostrogoths, formerly in subjection to the Huns, were settled in Pannonia and paid by the Romans to act as a buffer against other peoples on the move. When the emperor Leo I tried to reduce the subvention paid to them they invaded the western Balkans and got as far as Dyrrhachium in 461. A truce was negotiated and some years of peace ensued. But from about 470 they began systematic invasions of Illyricum, and were later settled, as unwelcome and uneasy foederati, in Lower Moesia—between the Danube and the Balkan chain — and in the Dobrudja, while another group of Ostrogoths moved about the northern Balkans.
Both groups needed land above all, and until they could get it they used their military strength to blackmail their titular commander, the East Roman Emperor. In fact those who most suffered were the inhabitants of Thrace, Moesia and the Illyrian provinces. Not only was the countryside devastated, the Ostrogoths, no doubt with the aid of Roman engineers in the first place, soon learned how to take cities. In the late seventies or early eighties of the century they captured and destroyed Stobi and Heraclea Lyncestis (Bitola) in western Macedonia. The former city was abandoned by its surviving inhabitants and never rebuilt. Other Ostrogothic raids penetrated as far as Thessaly and to the walls of Constantinople. The Ostrogoths, having no lands of their own to cultivate, could live only by depredation at the expense of the agricultural population of the Balkan provinces. That after so many years of war and devastation and the accompanying depopulation the provinces could still maintain such a formidable body of unwelcome guests for nearly a generation is a measure of how great their productivity must have been a century earlier.
In 488 the Roman government managed to get rid of the Ostrogoths from the Balkans by sending them under their king Theodoric, who had spent many years of his youth as a hostage in Constantinople, into Italy to expel the usurper Odoacer. Ostensibly an officer of the East Roman empire, Theodoric set up what was in reality an independent kingdom of Italy. The departure of the Ostrogoths gave only temporary relief to the hard-pressed inhabitants of the northern Balkan provinces. In 493 the Bulgars, a Turkic people who had been useful allies of the empire in the reign of Zeno, and who had earlier probably formed part of the Hun confederation, crossed the Danube, advanced into Thrace and killed a Roman general in battle. This raid was followed by others by the Bulgars and by other peoples from north of the frontier. In 499 a Bulgar raiding force in Thrace destroyed two thirds of a Roman army of 15,000 which had been sent to halt them. In 502 Thrace was once more invaded, and great quantities of booty and prisoners were taken across the Danube without any attempt at resistance by the field army; the frontier army was usually helpless once the frontier had been penetrated. It is significant of the new military situation in the second half of the fifth century that by 469 a new wall was built reaching from the Black Sea to the Sea of Marmara about forty miles from Constantinople, following the same line as the present-day Çatalca fortifications.
The capital was no longer protected by the army on the Danube, and the territory between the Danube and the new Long Walls was regarded as expendable if the worst came to the worst. In the meantime the short-term danger represented by the Bulgars was evaded by the customary method of enrolling some of them as foederati in the Roman army and paying the others to make war on a people living to the east of them in Bessarabia or the nearby Ukraine, the Antae. Though the Antae may have owed much of their political organisation and some of their ruling class to the Iranian Sarmatians, who had long lived north of the steppe zone in eastern Europe, the testimony of contemporary witnesses and the evidence of personal names make it clear that actors now appear upon the Balkan scene who will play the principal part in subsequent events there. A brief digression on their previous history may therefore be appropriate.
The Advent of the Slavs
The speakers of Slavonic had been settled in the area of the upper Vistula and middle Dnieper since the fourth millennium before our era. In this region, where forest and steppe mix, they slowly developed their peculiar culture. The use of bronze was gradually adopted during the second millennium by diffusion from their southern neighbours. They practised a mixture of stock-breeding and agriculture adapted to the conditions of their homeland, which provided little surplus to be accumulated in the hands of a ruling class. Their settlements were small, often on river terraces, their humble houses were often partly sunk in the ground for warmth in the long cold winter. From the beginning of the first millennium B.C. the Slavs were in contact with Iranian-speaking peoples of the steppe zone, who may have exercised some kind of overlordship over them, first the Cimmerians, and later the Scythians. They are probably to be identified with the ‘Scythian ploughmen’ of whom Herodotus speaks (4.17). It was during the period of Scythian predominance, about the middle of the first millennium B.C. that the Slavs appear to have learnt the use of iron. After this long period of contact with Cimmerians and Scythians, to which are to be attributed a number of Iranian loanwords in common Slavonic, there followed a period during which the Slavs were subject to penetration and displacement by invaders.
First came the Iranian-speaking Sarmatians, who unlike their Scythian predecessors left the steppe and entered the forest and steppe zone, where they exchanged their pastoral way of life for agriculture, which they probably learned from their Slav subjects.
There was probably a good deal of intermingling of Slavs and Sarmatians. Large villages are found, with buildings adapted to different economic purposes. The presence of uniform wheel-made pottery throughout the Ukraine, Moldavia, Wallachia and Transylvania in this period indicates that the craftsman was beginning to be differentiated from the peasant. The existence of both cremation and inhumation all over the area suggests some mingling of religious beliefs. By about 200 A.D. new invaders appeared on the scene, the Germanic-speaking Goths advancing up the Vistula. They probably subjected at least some of the Slavs to their dominion. Their influence is attested by a stratum of very old Germanic loanwords in common Slavonic. The disturbance caused in the peaceful and rather static life of the Slav peasants by these successive penetrations and by their close symbiosis with Sarmatian and Goth may have led to the beginning of an eastward and southward spread of the Slavonic tribes. It probably had much to do with the establishment of the mysterious state of the Antae between the Don and the Volga, and later west of the Don, in which there was at least a strong Slavonic element.
The advent of the Huns towards the end of the fourth century A. D. shattered the uniform Slavo-Gothic culture of the forest-steppe zone and drove the Goths towards the Danube. The break-up of Gothic power led to a slow spread of Slavonic farmers and stock-breeders outwards from their original area of settlement towards the east, the south and the west. The dense forest zone, to which the Slavs' way of life was ill-adapted, was left to the Baltic and Finno-Ugrian tribes who had lived there for millennia. The penetration of the forest by the eastern Slavs belongs to a later period. The reasons for this Slavonic expansion are obscure. The population was evidently increasing. But this could only occur if more food could be produced. Perhaps improvements in agricultural technique, new breeds of animals, new varieties of crops, enabled these peasant communities to bring into cultivation land which would previously have been left wild. In any case the expansion is a fact, though we know scarcely anything of the way in which it took place. It was not an invasion. The Slavs were not swiftly-moving armed horsemen, like the nomads of the steppes or the Goths. They formed no large political units, had no kings. They spread by walking, or by sailing on the great rivers, for they were superb boatmen. They must have taken over much land which had been left untilled after the devastations of the Huns.
Where they evicted the previous cultivators, it was after hostilities on a small scale—scuffles rather than battles.
By the end of the fifth century Byzantine sources report the presence of this new people, whom they call Sklavenoi, all along the Danube frontier. At the same time a recognisable cultural profile emerges over an area extending from the Elbe in the west to the Dnieper in the east, from the Baltic Sea and the forest zone in the north to the Danube in the south. It is characterised by small, unfortified villages of square wooden houses, partly sunk into the ground and with a stone stove in one corner. Near the villages are burial grounds containing cremation burials either in urns or in pits. Hand-made pottery of a greyish clay mixed with sand or crushed sherds has rounded shoulders and often a slightly out-turned rim. The distinction between craftsman and peasant, which began to develop in the homeland, was given up by the Slavs in their period of migration. Byzantine contemporaries tell us something about this people, whose humble remains have only recently been clearly distinguished by the archaeologist. They had no political organisation higher than the tribe, and even the power of the tribal chief was very limited. There was little social differentiation among them. They moved on foot, or in the dug-out boats, which they built and handled expertly. They swam well, and could swim undetected under water, breathing through a reed as through a snorkel. They wore no body-armour, but carried shields and spears, or bows with which they shot poisoned arrows. In warfare they avoided open plains and stuck to hilly wooded land, in which they ambushed their enemies with great skill. They were specialists in night attacks. Their pale complexion and reddish-blond hair attracted the attention of Mediterranean observers, Greek and Arab alike. They were of course pagans, and more will be said about the religion of the Slavs later. What struck the Byzantines who first came into contact with them—and who saw non-Christian religions in the light of Hellenic paganism— was the existence of one god among others who was lord of the thunder and to whom animal sacrifices were offered at special shrines; the worship of a number of female deities connected with vegetation and the countryside; and the absence of any belief in destiny.
Following their usual practice, the Byzantines soon began enrolling Slav detachments as auxiliary forces in their army, and we hear of such Slavonic troops taking part in Justinian’s campaigns in Italy and on the Persian front. In this way the Slavs would first be brought into direct contact with the civilisation of the late Roman empire and with the Christian religion,
and would learn something of the topography of the land facing them across the Danube, where they were prevented from settling by the Roman garrisons of the frontier defences.
We may now return to our narrative of events in the Balkans in the sixth century. Two new peoples are recorded by our sources north of the lower Danube, the Kotrigurs and the Utigurs. They appear to be Turkic speakers, part of the debris left by the dissolution of the Hun empire. Whether, as seems probable, they are to be identified with the Bulgars whose raids into Roman territory have recently been mentioned cannot now be determined. Still less certain is their identification with the Bulgars who a little later appear north of the Danube delta and in Pannonia, and with whose fortunes this book will largely be concerned. It is well to bear in mind that the pastoralists of the steppe moved quickly over large distances; that in the absence of a powerful state like that of the Huns, the Khazars or the Mongols, clans and tribes are continually making alliances and subjecting their neighbours, thus forming ephemeral quasi-states often known by the name of the tribe in a position of leadership at the time, and that Greek writers are generally extremely vague in their identification of the steppe peoples, often calling them by the name of their distant predecessors, which had become part of the literary tradition. Be that as it may, the first half of the sixth century is marked by raids of growing severity made into Roman territory by these peoples, and we hear of Slavs both accompanying them—perhaps providing supporting infantry to the steppe people’s cavalry—and fighting against them. It is generally useless to try to determine exactly who was involved in each raid. In 517 Macedonia and Thessaly were ravaged as far as Thermopylae, presumably by raiders following the Morava-Vardar route. A thousand pounds of gold did not suffice to ransom the prisoners, many of whom were taken back north of the Danube by their captors. Shortly afterwards Germanus, nephew of the emperor Justin I, inflicted a crushing defeat on a Slav force, and peace was restored south of the Danube for a number of years.
In 528 the Bulgars raided Thrace, in 529 we hear of Slav invaders. The probability is that in both cases the two peoples were acting in concert. Such invasions continued in succeeding years. In 533 the Roman commander-in-chief in Thrace, Chilbudius, himself allegedly a Slav, was killed by a Slav raiding force. Once again the Byzantine authorities took strong action. There was a major defeat of the Bulgars somewhere between the Balkan chain and the Danube.
At the same time Sirmium (Sremska Mitrovica) was recaptured from the Ostrogoths. In a law promulgated in 535 Justinian boasts of the restoration of peace on the Danube (Novel 11 pr.). It did not last long: Sirmium was lost to the Germanic Gepids of Pannonia in the next year and in 540 there was a major invasion by the Bulgars or Kotrigurs, accompanied by Slavs. Justinian’s concentration of units of the field army in Africa and Italy in the course of his reconquest of these territories had probably led to some weakening of the defence in depth ory the Danube frontier. One invading force reached the walls of Thessalonika, then turned eastwards along the Via Egnatia, forced the Long Walls and appeared outside the walls of Constantinople itself. Unable to take the city, they withdrew to the Danube taking with them 100,000 prisoners. The second invading force entered peninsular Greece and got as far as the Isthmus of Corinth before returning to the Danube laden with prisoners and booty. No invasion on this scale had been seen since the operations of the Ostrogoths two generations earlier.
From 540on there was little peace in the northern Balkans, despite the extensive programme of military works to protect the main Roman lines of communication, the strengthening of the huge military bases on the Danube itself, the reinforcement of the Danube flotillas, for which all invaders had a healthy respect, and the efforts of the diplomatists of Constantinople to persuade more distant peoples of the steppe zone to fall on the rear of the Bulgars (or Kotrigurs) and Slavs. But the frontier itself was held, and the pressure of the invaders seems to have slackened after 551, thanks to an alliance between the Romans and the Utigurs, who lived east of the Kotrigurs. There was no settlement by peoples from north of the Danube in the Balkans except for limited settlement of foederati by the Romans themselves. In 558-9 major invasions began again. The Kotrigurs, with the Slavs in their train, crossed the frozen Danube at the beginning of 559 and fanned out in three directions: through Macedonia and Greece as far as Thermopylae, into the Gallipoli peninsula, and towards the Long Walls which protected Constantinople. Only the threatened cutting off of their retreat by the fleet on the Danube forced them to return, once again with many prisoners and much booty. There was a fresh invasion of Thrace on a smaller scale in 562. What strikes the reader of contemporary accounts of these massive invasions of the mid-sixth century is the virtual absence of any mention of the countless strong points and fortified positions occupied by the Romans.
The invaders seem simply to have bypassed them. They certainly did not stop to assault them. And once they have passed through the defensive zone, they seem to have no difficulty in going where they like until they reach the fortifications of one of the great cities of the empire. When they fight a regular battle in open ground they are generally defeated. And they keep well out of the way of the Roman fleet. The immediate motive of the invasions is to take booty and prisoners. But behind this there probably lay the need to get land to settle on. They are constantly being pushed on from behind, both by the continuing demographic expansion of the Slavs and by the movement from east to west of yet further pastoralists of the steppe. But the Roman authorities have learnt from their experience with the Visigoths and the Ostrogoths, to say nothing of what had happened in the western empire, where the foederati had ultimately taken over province after province from them. Only small numbers from beyond the Danube were allowed to settle, and then under strict Roman control. So long as the Romans hold the fortified cities and the strong points which guard the main roads, the invaders cannot take over the land, even if it is lying uncultivated as a result of devastation. And they cannot take fortified cities and strong points; their technology is not up to the task. This was the situation of stalemate which existed in the closing years of the reign of Justinian.
In 558 a new people appeared on the frontier of the empire in the Caucasus. The Avars had previously established their domination over a large area north and west of China. Overthrown by the revolt of their Turkic subjects, many were put to the sword. The rest swept across the steppe zone of Eurasia with incredible rapidity, and now demanded land in the Roman empire for settlement. The government in Constantinople, first under Justinian then under his nephew and successor Justin II, temporised. The Avars were given a subsidy and asked to deal with certain enemies of the empire north of the Black Sea. This they did with such cold-blooded efficiency that they were soon on the Danube, demanding land once more. The Byzantines drew out the negotiations as long as possible. For a time the Avars were put off with an offer of land north of the Sava, if only they would expel the Herules who were then occupying it. Justin II suspended payment of subsidies to them without provoking any hostile reaction. The Avars seem to have been at first very unsure of themselves in their new environment. They even allowed themselves to be attacked and defeat ed by a Byzantine army.
But the news of Byzantine defeats on the Persian front in 573 or 574 changed the situation. The Avars crossed the Danube, eluding the Byzantine navy, and defeated a Byzantine army. The emperor Tiberius, his hands tied in Persia, signed a treaty ceding to them the region of Sirmium, but not the city itself. The chronology of events in the next few years is very uncertain. The Avars seem to have turned against the Slavs north of the Danube, perhaps acting ostensibly as allies of the empire, and to have subdued them. They then may have urged the Slavs to raid Byzantine territory south of the river; we hear of a major offensive about 580. The Avars asked the emperor Tiberius II for permission to cross the Danube and ‘punish’ the Slavs, but the emperor saw the trick and refused. But soon the Avars no longer needed his permission. In 582 they besieged and took Sirmium, the key to the Balkans, and in the next few years Slavs and Avars poured into the peninsula, meeting with scarcely any resistance. The Avar crossing of the Danube was carried out on boats built by their Slav subjects or allies, cf. Theophylact Simocatta Hist. 6.3.
The problem of the relation of the Avars and the Slavs is a difficult one. Now we hear of Avar invasions, now of Slav invasions, again of Avaro-Slav invasions; the earliest express mention of Slavs as allies of the Avars is in raids on Roman territory in 576-7. Let us get some points clear. First, the Avars, who had long been in the sphere of influence of both Chinese and Sasanian culture, were at a much higher level of political development than the Bulgars, Kotrigurs and Onogurs, and a fortiori than the Slavs. The Avar community was a state, with institutions, an organised army, and a king or khagan. The powerful personality of the khagan Baian emerges from the Byzantine and western sources. They conducted diplomatic negotiations and signed treaties not only with the Byzantines, but with the Lombards, the Franks and other peoples. They had long given up the relatively unprofitable business of herding animals for that of herding men. The Slavs had no political cohesion; each little community seems to have acted for itself, and we never hear of diplomatic negotiations or treaties. The Avars were not numerous, though whether they were more or less than 100,000 strong we do not know; the Slavs were to be counted in millions. The Avars did not cultivate the soil; the Slavs were excellent farmers. The Avars were uninterested in settlement outside of the steppe zone, in which they knew how to live; the Slavs needed the agricultural land of the Balkans. It was in the interest of the Avars to cream off the agricultural surplus of the Slavs, and the more land they cultivated the more there would be to cream off ;
it was in the interest of the Slav communities to attach themselves to the Avar armies, whose military technology was superior to their own. This is probably the basis of their uneasy collaboration. And the Slavs doubtless did not confine themselves to following in the wake of Avar armies, but moved forward into unoccupied or poorly defended territory whenever they could. In 582 Slavs and Avars reached Anchialos, on the Black Sea. A few years later they were pouring into Greece, and it may be that they besieged Thessalonika in 586. So long as the Byzantines were occupied in war with Persia, Slavs and Avars could go almost anywhere they liked in the Balkans. And when the Avars withdrew, as they did on several occasions, the Slav peasants stayed.
The emperor Maurice forced the Persians to concede victory in 591, and turned his attention to the reconquest of the Balkans. Ten years of hard-fought war ensued, punctuated by a short peace with the Avars. Many of the new settlers were driven back north of the Danube, but many no doubt stayed on in the more remote areas. In 602 the Byzantine army was ordered to winter north of the Danube in order to keep up the pressure on the retreating enemy. This order provoked a military revolt, which soon spread to the capital. Maurice was murdered, and Phocas, the leader of the revolt, proclaimed emperor. Phocas had his hands full dealing with his own subjects, and the Persians seized the opportunity to reopen hostilities. The Danube frontier was left without a garrison, and Avars and Slavs once more poured unhindered across the river and spread through Illyria, Moesia, western Thrace and peninsular Greece.
Both had learned a great deal from the Romans. The Avars were keen students of siegecraft, and we hear of a Roman defector, no doubt a military engineer, acting as their instructor. The Slavs too had learned how to take fortified positions. They had long had in their midst numerous captives and defectors from Roman territory, including craftsmen of every kind. A contemporary military handbook records the interesting fact that the Slavs did not keep their captives in permanent slavery, as did other peoples, but after a fixed interval allowed them either to return home or to remain as free men with their former captors. In this way Roman technicians of all kinds must have settled among the Slavs and passed on their skill. And the Slavs themselves were apt pupils. Their expertise in making boats, which is often commented on, implies widespread acquaintance with wood working and the use of tools. The details of this process of acculturation remain unclear.
What is clear is that between the middle of Justinian’s reign and the end of the sixth century the Slavs had learned siegecraft and acquired or constructed the necessary engines. When they invaded the Balkans in the eighties of the sixth century and again after 602 they took city after city. They were thus in a position to settle permanently and till the land. During some thirty years the ethnic composition of the Balkan population was completely changed.
Although it is evident that a radical transformation took place, it is difficult to establish with any clarity exactly what happened. In the west virtually all territory north of the present Greco-Yugoslav frontier with the exception of the coastal cities seems to have passed out of Byzantine control. Kastoria remained in Byzantine hands, though the surrounding countryside was largely occupied by Slavs. The great city and port of Thessalonika was never lost by the Byzantines in spite of a series of sieges, some by Slavs alone, others by Slavs and Avars acting in concert. The first of these may have been in 586, but was more likely in 597, and another probably took place during the reign of Phocas (602-610). Shortly before 626 there was an attack in force on the city by a number of Slav tribes acting in common under a leader called Chatzon, and operating by sea in their dug-out boats as well as on land: this seems to have formed part of a general thrust into the Aegean area by the Slavs. A storm destroyed the Slav fleet and the city was saved. Its hinterland, however, was in Slav hands, and a kind of symbiosis between the Byzantine city and the surrounding Slav peasants began quite early. In the eastern part of the Balkan peninsula the area north of the Balkan chain was completely lost to the Byzantines, except for some strong points in the Dobrudja. The cities along the Black Sea coast were in general held, since the Byzantine fleet had control of the sea, but individual cities from time to time fell temporarily to the Avars or Slavs. South of the Balkan chain the situation is less clear. Naissus fell shortly before the great siege of Thessalonika by Chatzon, and Serdica appears to have fallen at the same time. But it was soon recaptured and remained in Byzantine hands, though dwindling from a great city to a garrison-town clustered round the sixth century church of the Holy Wisdom, and the surrounding territory became entirely Slav.
In 626 the Avars under the khagan Baian, supported by a large contingent of Slavonic troops, besieged Constantinople. They were equipped with a variety of sophisticated siege engines, and accompanied by a fleet of dug-out boats manned and presumably built by Slav sailors.
Their attack coincided with an advance by the Persians under Chosroes II to the Asiatic shore opposite Constantinople, and it was generally believed that the Avars were acting in concert with the Persians. In the outcome the siege was unsuccessful, and both Avars and Persians had to withdraw. This defeat marked a turning-point in the fortunes of the Avars, whose interest in the Balkans diminished as they were faced with revolts of their subjects and pressure from the Franks in Pannonia. The Byzantine government had to deal henceforth with the Slavs on their own. No doubt the long-term perspective was their absorption into the empire as tax-paying Christians. In the meantime Constantinople itself and its Thracian hinterland had to be held. The latter became particularly important when Egypt fell, first to the Persians and then to the Arabs, and Egyptian corn no longer reached the city. Hence Byzantine policy was to hold on to a series of fortresses protecting eastern Thrace—Develtos, Adrianople, Traianopolis (today Alexandroupolis) —to retain control of the coast road along the Black Sea shore and of the Via Egnatia as far as Thessalonika, and to try to establish a Byzantine presence in as much of the diagonal road as was feasible. On the whole this policy was realised. The Via Egnatia could only be kept open by patrolling in force, and land communications with Thessalonika were never secure. But the ring of inner fortresses was held, Philippopolis remained Byzantine, Serdica was recaptured, from the Slavs or Avars, though we do not know when. And the coastal cities of the Black Sea remained Byzantine possessions.
Further south, in peninsular Greece, there was extensive Slav settlement in the last decades of the sixth and the early decades of the seventh century. Its extent is very hard to estimate. Fallmerayer’s thesis, recently revived by Jenkins, that the original population of Greece was wiped out or expelled by the invading Slavs, and that the present Greeks are the descendants of Hellenised Slavs, has bedevilled discussion of the problem for more than a century. It is based essentially on a series of lapidary statements by sources who, if contemporary with the events, were far removed geographically, like Isidore of Seville and Willibald of Mainz, or who wrote many generations later, like the epitomator of Strabo’s Geography, the emperor Constantine VII, the patriarch Nicholas III and the Chronicle of Monemvasia. We have to check the sweeping statements of these authorities against surviving narrative and hagiographical sources, episcopal lists, known changes in the administrative structure of the area, the evidence of place-names and the like. From these it is clear that certain of the major cities remained in Byzantine hands: Athens and Patrae certainly, Corinth except for one possible brief occupation, Thebes almost certainly.
We hear of Greek inhabitants of areas seized by the Slavs taking refuge in various strong points on the coast and elsewhere, or emigrating to Byzantine southern Italy. In the eighth century there was a Slavonic archon in Thessaly who conspired with one of the sons of the empress Irene against his mother. The Peloponnese was heavily settled by Slav peasants, and centuries later mount Taygetus was still inhabited by Slavs who lived under a special régime. Place-names of probable Slavonic origin— and it must be emphasised that there is a great deal of uncertainty in some of the etymologies—are dense in Epirus and Thessaly and in the western Peloponnese, much less dense in central Greece, and hardly existent in Attica and some other regions of eastern Greece.
All in all we can legitimately conclude that there was extensive displacement of the original population, that much of western Greece and the Peloponnese passed out of Byzantine control for a time, but that the Aegean coast and parts of central Greece, together with coastal cities elsewhere, were never overrun. The areas of Slav settlement in peninsular Greece were probably more broken up than in the northern Balkans. None of them was far from a Byzantine city or from the sea in which the Byzantine fleet could cruise at will. And we must be careful not to assume that because an area was effectively out of the range of the Byzantine government all its inhabitants were of foreign stock. Many Greek-speaking peasants probably remained on their land off the main routes of Slavonic advance, only too glad to be protected from the Byzantine tax-collector by the new invaders. There is a further factor relevant to the Slav settlements in Greece, to which little attention has been paid in the past. The agricultural technique of the Slavs, developed over several millennia in the forest-steppe zone of central Europe, was relatively well adapted to areas with heavy soil, perennial rivers, moderately warm summers, forests to conserve the water, etc. In the strange Mediterranean world of olive and vine, bare limestone hills, light soil, and long, hot, dry summers the invaders must have been at a disadvantage. In particular their pattern of settlement along river banks was impossible in peninsular Greece, where the larger intermittent rivers with their changing course turn the lower part of their valleys into uninhabitable malarial swamps, and the smaller rivers are dry for most of the year. They would, where they could, settle in areas most adapted to their own pattern of agriculture. One such was the plain of Thessaly, another the plain of Laconia.
Epirus, with its relatively heavy precipitation and its background of snow-capped mountains, suited them better than arid Attica. They would also avoid the high mountain pastures characteristic of some parts of Greece, for they were essentially farmers, not transhumant stock-breeders. So it is likely that Greece after the Slavonic settlement was ethnically a rather complex mosaic, with Slav and Greek in contact along a number of lines.
The long-term Byzantine reaction to the occupation of so much territory within the historical frontiers of the empire was to seek to bring the newcomers under imperial control bit by bit. We hear of settlement of Slavs from areas in Thrace which the Byzantines reconquered as soldiers in Asia Minor. The experiment was not a complete success. Many of the new soldiers deserted to the Arabs. There are traces of ad hoc administrative arrangements to deal with Slav areas over which Byzantine control had been successfully reasserted. An active policy of conversion of Slav communities to Christianity was begun by church and state. And conversion to Christianity at this time implied the use of Greek as a liturgical language. Patterns of trade developed between the various Sklaviniai, as the Byzantines called areas inhabited by Slavs on theoretically imperial territory, and the cities held by the imperial authorities. At the same time, the tribal nobility of the various Slav communities felt the pull of Byzantine culture and Byzantine ways of life. We hear of the prince of one of the Slav tribes near Thessalonika, Perbund, who was accustomed to live in the city, dress like a Greek, and speak Greek. When he was arrested and taken to Constantinople on suspicion of plotting revolt, a joint mission of Slavs and citizens of Thessalonika went to the capital to intercede in his favour. The case was probably quite typical. The main effort in this process of absorbing the Slavs was probably made in Thrace, near Thessalonika, and along the line of the diagonal road. Peninsular Greece could wait, and in any case the Sklaviniai were less favourably situated there than further north. The re-establishment of Byzantine control can be traced by the setting up of themes—provinces under a regular government, with a locally raised army, taxes, courts etc. The theme of Thrace, comprising the immediate hinterland of Constantinople, was set up between 680 and 685, that of Hellas in central Greece before 695. The next group of themes was established about a century later, that of Macedonia (in fact western Thrace) towards 800, that of Thessalonika early in the ninth century, that of Strymon later in the century, that of the Peloponnese some time before 812 and perhaps already in the eighth century,
that of Dyrrhachium early in the ninth century and that of Nicopolis in Epirus a little later.
But before these themes were set up, marking the gradual restoration of Byzantine administration and influence in the areas occupied by the Slavs, a new development had taken place which was to change for ever the face of the northern Balkans, namely the establishment of the Bulgars south of the lower Danube. This will be the next main topic in our survey of the history of the Balkans in the Dark Ages. But before turning to discuss this topic it would be well to glance at the problem of what happened to the original inhabitants of the areas occupied by the Slavs. First, it is likely that many areas, especially in the northern Balkans, were thinly populated and uncultivated when the Slavs arrived, as a result of the two centuries of warfare and invasion which had gone before. In many cases there was no one for the newcomers to displace. Then we hear of refugees moving to areas nearer Constantinople, or to the islands, or to Byzantine southern Italy. A third situation, and one arising particularly when the Avars were in command of operations, was that of the wholesale evacuation of the inhabitants of a newly-conquered area to some other region under the invaders’ control. For instance we hear of a large body of Greek or Latin speakers removed from Illyria and Thrace during the first Avar advance into the Balkans and transported to the neighbourhood of Sirmium, where they retained their language, their ethnic identity, and their Christian religion for sixty years. Later they took part in a revolt against the Avars, were granted and somewhere in Macedonia by the leader of the revolt, and finally made their way to Constantinople. There must have been many such cases of mass deportation, most of which would end in the absorption of the deportees into their new milieu.
Lastly, it is clear that some of the original inhabitants took up ecological niches unattractive to the Slavs, and in this way played their part in the new economy of the Balkans. This is particularly true of transhumant stock-breeding. Leaving out of account the Rumanians north of the Danube, who may or may not be descendants of Latin speaking inhabitants of the Balkans, there were until recently all over the Balkans wherever there are high mountain pastures groups of sheep and goat breeders practising transhumance and speaking a language related closely to Danubian Rumanian. Sometimes they were based on villages in the valleys and went up in summer to their mountain pastures. Sometimes their base was in the high ground and they had no permanent dwellings in the valleys.
These Macedo-Rumanians, Meglenites and Istro-Rumanians are presumably descendants of Latin-speaking inhabitants of the Balkans, mainly Romanised Thracians and Illyrians, who took to this specialised way of life when driven off their agricultural land by the incoming Slavs, and they were without doubt more extensively settled in the past than in recent years.  Another group which may have originated at the same time and in the same way are the Greek-speaking transhumants of Epirus and elsewhere, the Saracatsani, whose way of life and system of values is almost identical with that of the Romance-speaking transhumants, and who are probably in the main the descendants of Hellenised Thracians. Almost confined to Epirus today, the Saracatsani once spread over a wider area. There are Greek-speaking Saracatsan communities who have long given up the transhumant way of life in various parts of present-day Bulgaria.  Lastly, the ancestors of the Albanians seem to have been specialist stock-breeders, who only later, and in specially favourable circumstances, took up agriculture. They are generally thought to have been descendants of unromanised Illyrians who held out in the high mountains of northern Albania and the neighbouring regions of Yugoslavia. But V. Georgiev has recently argued, with some plausibility, that they are descendants of the Daco-Moesian speakers from north of the Balkan chain. The question is as yet unresolved and of little importance for the theme of this book.
In general, then, even after the great Slav invasions of the late sixth and early seventh century the Balkans must have been an ethnic mosaic. Eastern Thrace and the Black Sea coast, as well as Thessalonika and parts of peninsular Greece remained solidly Greek, their population strengthened by refugees. Elsewhere there survived pockets of Greeks, Latins, and no doubt unromanised Thracians, Illyrians, Daco-Moesians and Dalmatians. Some kind of Byzantine presence was maintained along the great diagonal road probably as far as the defile of Dragoman. Cities and city life did not survive in the conquered areas. The Slavs learned in time how to take cities, but they were not interested in living in them. The Avars were capable of using cities—there is a story of the khagan Baian forbidding the demolition of the baths in captured Anchialus. But the Avars did not settle in the Balkans, and by the middle of the seventh century their influence there was at an end. So the ruined cities dotted the landscape, uninhabited except by occasional squatters, serving as quarries for stone,
while the Slav peasants built their settlements some distance away from them.
This formulation requires some qualification, as conditions varied from region to region. There is possible evidence of continuity of occupation at Bononia (Vidin) and Durostorum (Silistra) on the Danube; Nicopolis ad Istrum (Nikjup), which was off the main invasion routes of the Avars, may have been in continuous occupation until the end of the Middle Ages; in southern Bulgaria there was more continuity of occupation, even in the territories long lost to the Byzantine empire. But what did not continue was the antique pattern of city life, by which an urban population of rentiers, merchants and craftsmen was maintained by the surplus produce of the surrounding countryside, and which found its expression in a whole structure of institutions, from city councils and bishoprics to theatres and baths. If some of the buildings continued to be occupied, it was by squatters rather than citizens. However, these squatters are likely to have been survivors from among the original inhabitants rather than Slav immigrants. So far as the evidence goes, early Slav settlements were always made at some distance from Greco-Roman cities, whose towering walls and massive buildings were useless and perhaps disquieting to the newcomers. 
Immediately after his decisive victory over the Persians in 626 the emperor Heraclius turned his attention to the former Balkan provinces of the empire. Certain of the measures then taken had a lasting effect upon the history of the north-western Balkans. The Croats, a people probably of Sarmatian origin and living among Slavonic tribes in Bohemia and Silesia, were encouraged by the Byzantines to revolt against the Avars and to settle in the region south of the Save and behind the Dinaric Alps, where they speedily established themselves as the dominant power. Their rulers accepted Christianity from Byzantium, though they later lapsed, and it was from Rome, via the surviving Dalmatian cities, that the later conversion of the Croat people proceeded. Another half-Slavicised Sarmatian people, the Serbs, also revolted against the Avars, left their home somewhere in Saxony and were settled by Heraclius in the neighbourhood of Thessalonika, which was being threatened by the neighbouring Slav tribes. They were later, perhaps at their own request, moved to the area south of Singidunum and settled among the Slav tribes there as allies of Byzantium. In this way a large part of the north-western Balkans was brought under the control of powers loosely allied with the empire, a fact which in Byzantium political thinking was interpreted as the restoration of Byzantine sovereignty.
That this Byzantine counter-attack took place so far from Constantinople is to be explained by the consideration that in Heraclius’ eyes the enemy was the Avars, whose army was able to defeat Byzantine forces, and whom he saw as master-minding the invasion and occupation of imperial territory. The far more numerous Slavs were a nuisance, but not a mortal danger. They could be absorbed or dispersed later. So the emperor was bent on settling allies on whom he could rely as close to Pannonia, the centre of Avar power, as possible, while large parts of the eastern Balkans and peninsular Greece were left in the hands of the Slav invaders. Had Heraclius in his last years not directed his attention entirely to resisting the advance of the Arabs in Syria, Palestine and Egypt, the policy of settling a cordon sanitaire of allies on the Avar frontier might well have been successfully continued.
The Origin of the Bulgarian State
The Bulgars have appeared already in our survey of the fortunes of south eastern Europe in the Dark Ages. They were a Turkic people, who first emerge as an identifiable group after the collapse of the Hun empire. They were probably related to the Kutrigurs and Utigurs who occupied much of the steppe zone north of the Danube delta and the Black Sea in the sixth century, and indeed may be mainly their descendants. This is the common view. Beshevliev has recently argued with some plausibility that they are to be distinguished from the Kutrigurs and Utigurs, and that it was only when they appeared north of the Danube that they absorbed these remnants of the Huns, probably few in number by this time, and with them some of their traditions, including the memory of Attila and his son Irnak.  The matter is of little importance for the subject of this book. The main body of the Bulgars probably spent some time living as pastoralists in the steppe north of the Caspian Sea. Some of them appear on the left bank of the lower Danube in the last decades of the fifth century, either as raiders or as mercenaries in the Roman service. The sixth century seems to have been one of relative calm for the Bulgars—unless they are to be identified with the Kutrigurs and Utigurs, who engaged in one another’s mutual destruction at this time. In spite of the military weakness of the Roman empire in the latter part of the sixth century, Roman influence upon this people settled beyond the frontiers of the empire was considerable. We hear of a Bulgar prince coming to Constantinople for baptism about 600, no doubt as a result of Byzantine diplomatic reaction to the threat of the Avars and their Slav subjects.
The establishment of the Avars in Pannonia apparently changed the balance of power in the steppe zone in such a way as to permit— or oblige — the Bulgars to expand. About 600 they seem to have split into several groups. One went to the Kuban steppes; another migrated far to the north-east and settled on the banks of the lower Kama, before its confluence with the Volga, where they played a part in the ethnogenesis of the Chuvash people, a third group attached themselves to the Lombards, took part in their invasion of Italy, and ended up near Benevento. But the largest group resettled or remained in the lower Danube area, from which they took up their old habit of raiding the Roman provinces to the south. There were apparently Bulgar contingents along with the Avars and Slavs at the sieges of Thessalonika and Constantinople. One of their leaders, Kubrat, seems to have broken with the Avars and formed an alliance with the local Slavs.
The history of the Bulgars in this period is extremely obscure, the names of their leaders are badly transmitted in our scanty sources, and any reconstruction of the course of events is hypothetical. The mysterious Kuver, a high officer under the Avars who, with the aid of a body of Romans who had been settled for two generations north of the Danube and of some Slav tribes, rebelled against the khagan, moved south into former Roman territory and gave siege to Thessalonika, has been thought by some to be a son of this Kubrat; his date is about the third quarter of the seventh century. The mention in the rock inscription at Madara, probably dating from the reign of Khan Tervel, of ‘my uncle in Thessalonika’ suggests that Kuver was indeed a Bulgarian and a son of Kubrat. If so, his establishment in former Roman territory overrun by the Slavs is a close parallel to what happened in the north-east Balkans. It was there that Asparukh, according to tradition another son of Kubrat, ruled over the main body of the Bulgars just north of the Danube delta. Pressed by the growing power of the Khazars to the east, he and his people were forced to take refuge in an island in the delta. The emperor Constantine IV saw in this move a threat to the Black Sea coastal strip still held by the Byzantines, and perhaps to Thrace. A campaign was mounted, perhaps in concert with the Khazars, to oust the Bulgars from their refuge. The campaign was bungled, the emperor withdrew to Mesembria with gout, the army broke up in disorder, and the Bulgars were able to break out of the delta and establish themselves in eastern Moesia. A late source puts their numbers at 10,000 fighting men.  There they subjected the Slav tribes of the area, obliging them to pay tribute and settling them on the frontiers of the territory which they controlled. 
The Byzantines were in no position to do anything about this new invader of what had been their territory. They may indeed have welcomed the Bulgars as a counterpoise to the Avars, much as Heraclius had welcomed the Croats and the Serbs. At any rate they came to terms with the Bulgars, and signed a treaty granting to Asparukh all the land between the Danube, the Balkan range and the sea, as well as an annual subsidy. This treaty, probably signed in 680, marks the first formal recognition by the Byzantine government of the existence of a foreign state on what had been imperial territory, but perhaps too much should not be made of this. It is likely that the organisation of the first European theme, that of Thrace, was a reaction to the new situation created by the arrival of the Bulgars in the Balkans.
The Bulgars seem to have inherited from their nomad past a complex and stable social and military hierarchy, recalling that of the Orkhon Turks at the same period. They were organised into tribes, and the hereditary chief of one of the tribes, that of Dulo, acted as leader of the whole confederacy. Under the leader or Khan were a number of ministers belonging to the tribal nobility bearing different titles and with more or less clearly defined responsibilities. There were two classes of nobles, though the precise distinction between them escapes us. The army was the whole Bulgar male population, presumably organised in tribal contingents, and supported by the forces of the Slav tribes which the Bulgars controlled and with whom they were very closely linked from the start. For their strong sense of ethnic unity did not prevent them mingling with the Slav communities, with whom in any case they must have been in contact since the early sixth century. It is possible to distinguish Slav and Bulgar sites from the Dark Ages. The Bulgars seem to have lived in round, yurt-like houses inherited from their steppe past, and to have inhumed their dead, while the Slavs lived in square, half-underground houses with a stove in one corner, and cremated their dead. Different types of pottery are associated with the two peoples. But we soon find evidence of mixed settlements; as none of these settlements can be dated with any certainty, too much importance should perhaps not be attached to this archaeological evidence, but it should not be altogether neglected. 
The Bulgars established their capital at Pliska, between the Danube and the Balkan chain, where the ruins of a late Roman city provided building material. There they built a complex of public buildings for the Khan and his court, surrounded by an earthwork perimeter wall within which the whole Bulgar people could probably find shelter.
They no doubt used the forced labour of Byzantine captives and of the surviving Roman population, especially to construct the heavy stone buildings. The original Bulgar state was confined to a corner of the north-eastern Balkans, from the Danube to the Balkan chain, and from the Black Sea to the Avar frontier, which may have followed the line of the Iskŭr. It is not known how much territory Asparukh controlled to the north of the Danube but it may have been quite extensive.
Asparukh’s son and successor Tervel formed an alliance with the exiled Byzantine emperor Justinian II in 705 and helped to restore him to Constantinople. This was a gamble, but one which came off, and it increased the power and influence of the Bulgar state. Tervel was welcomed to Constantinople and given the rank of Caesar, a title which only recently had marked out its bearer as heir apparent to the reigning emperor. It still implied a close relationship with the emperor: Tervel was no doubt baptised, but there was no wholesale conversion of his people. More important, Tervel was given the area of Zagoria, probably the region between the Balkan range and the Gulf of Burgas. But the coastal cities of Mesembria, Develtos and Anchialos remained in Byzantine hands. So the situation remained throughout most of the eighth century. The iconoclast emperor Constantine V fought a series of campaigns against the Bulgars without dislodging them from their land, though he may have recaptured part of the territory ceded by Justinian II. In 784 the empress Irene was able to make a peaceful progress from the Byzantine naval base at Anchialos to Berrhoea (Stara Zagora), without provoking any hostile reaction from Bulgaria. In the west Bulgar expansion was firmly blocked by the Avars. But they did make tentative moves south-westwards, towards Macedonia, where Kuver had tried to establish himself in the previous century, and where the Slav population was relatively independent of Byzantine and Avar power alike. In 789 a Byzantine general was killed by the Bulgars while reconnoitring disputed territory in the Struma valley. However, in the early years of the ninth century two events took place which transformed the situation of the Bulgar state. The crushing defeat of the Avars in Pannonia in 796 by Charlemagne’s son Pippin led to the speedy disintegration of the Avar state, and opened the way to westward expansion for the Bulgars. And the old royal house of Dulo was replaced, we do not know exactly how, by a new dynasty, that of the Pannonian Bulgars, who had revolted against their Avar lords after 796. The new ruler, Krum, succeeded Khan Kardam in 802 or 803.
Bulgaria and Byzantium in the Early Ninth Century
The immediate effect of the collapse of Avar power was the extension of Bulgarian territory far to the west, as far as the Theiss, where it marched with the land of the Franks. Most of this new territory was inhabited principally by Slavs, over some of whom the Pannonian Bulgars had established overlordship. Frightened by this increase in Bulgarian power, and perhaps hoping by a show of strength to divert Bulgaria westwards, the Byzantine emperor Nicephorus set out in 807 to invade Bulgaria. A further motive for the Byzantine resumption of hostilities was probably fear of Bulgar influence on the Slav populations of peninsular Greece and Macedonia. There had recently been a revolt by some of the Slav tribes of the Peloponnese which had been put down with great severity. However Nicephorus got no further than Adrianople before he had to return to the capital to deal with a plot against him. Krum, whether he had harboured aggressive intentions previously or not, realised that he had better strike first. Rather than throw himself against the ring of fortresses protecting eastern Thrace, he moved against Macedonia, where he could probably count on the support of much of the Slav population. In 808 he swept down the Struma valley, defeated a Byzantine army, and took 1,100 pounds of gold intended as army pay. But the way to Macedonia was blocked by a series of Byzantine strong points in western Thrace, of which Serdica was the principal. In 809 Krum captured Serdica, killed its garrison of 6000 and demolished its walls. As yet the Bulgars were not interested in holding fortified positions themselves. Nicephorus’ reaction was to march through the mountains and sack and burn Krum’s capital of Pliska in the autumn of the same year. The main theatre of operations was transferred to Thrace. In 811 Nicephorus again captured Pliska, destroying the royal palace, which was probably of wood, and making a display of barbarity—we hear of Bulgarian babies being thrown into threshing machines. Krum apparently made overtures for peace, but Nicephorus was obdurate in the euphoria of victory.
However, the Bulgarian army had not been destroyed. As Nicephorus and the Byzantines, laden with booty, wound their way through the mountains on their return journey they were caught in an ambush and the whole army, with its commander, perished. It was the first time for nearly six hundred years that a Roman emperor had been killed in battle. The Byzantines had not only lost their best fighting men, they were utterly demoralised and could not understand what had gone wrong.
The Bulgars did not waste time. Krum attacked the key Byzantine cities around the Gulf of Burgas. Develtos was taken, the inhabitants of Anchialus fled. Panic spread into Thrace, where there was wholesale flight from the fortress city of Philippopolis. Krum, anxious not to overstretch his resources, proposed peace. It is interesting to note that his envoy, Dargomer, bore a Slavonic name. The emperor asked for the restoration of the status quo ante bellum. Krum pressed on, and by autumn 812 had taken Mesembria. the major port on the west coast of the Black Sea, with magnificent natural and artificial defences. He did not remain there, but dismantled the walls and retired to Pliska for the winter. In the next year he defeated the main Byzantine field army, and by July was at the walls of Constantinople. Though the Bulgars had made great progress in siegecraft, they were unable to tackle the superb fortifications constructed by Theodosius II four centuries earlier. And as they had no fleet, they could not hope to starve the city to surrender. Nevertheless Krum maintained the siege for nearly two years. It may be that he did not realise the impasse into which he had got himself. But it is possible that he hoped to profit by a coup d’état within the beleaguered city which would bring to power an emperor willing to make concessions to the Bulgars. He seems to have been well aware of the disaffection of some units of the army, and may have had contacts within the walls. What is clear is that he had no intention of taking over Constantinople. There were abortive negotiations, ending in an attempt by the Byzantines to assassinate Krum. More important, while the Khan besieged Constantinople, his generals were mopping up the remaining Byzantine cities of Thrace. Adrianople surrendered in autumn 813, Arcadiopolis (Lüle Burgaz) in the ensuing winter. How events might have turned out we shall never know, for Krum died suddenly in April 814 while preparing a new attack on Constantinople, and the Bulgarian army withdrew.
Krum was succeeded by his son, Omurtag, but apparently only after quelling a revolt of the old Bulgar nobility. The details escape us. It is this which explains Omurtag’s failure to press home the favourable military position established by his father. In the winter of 815-16 he concluded a treaty with the emperor Leo, valid for thirty years, which arranged the exchange of prisoners—in particular those taken from the Thracian cities and transported to Bulgaria—and fixed the frontier, which was more or less that determined by Justinian II and Tervel a century earlier. Incidentally this treaty is hardly mentioned by the Byzantine historians and chroniclers,
but is imperfectly preserved on a stele which probably stood originally in the palace at Pliska, and is now in the Archaeological Museum in Sofia. 
In the main the Thirty Years Peace was maintained. The Bulgars did not seek to expand into eastern Thrace, which would inevitably bring them up against the problem of the capital and its impregnable fortifications. They dug a great ditch and rampart along the frontier from Develtos to the Maritsa near Simeonovgrad, which they kept permanently manned. The cities which Krum had taken from the Byzantines—Serdica, Philippopolis, Develtos, Anchialos etc.—were left deserted, in a kind of no-man’s-land. Only Mesembria and Adrianople were rebuilt by the emperor. Presumably this reflects a lost clause of the treaty. In the meantime the Bulgars directed their attention westward, into territory not covered by the treaty, and to building a great new royal residence at Preslav in the years following 821. Omurtag concerned himself principally with the northern part of his kingdom, which lay north of the Danube, and with his relations with the Franks. In his reign, a modern Bulgarian historian has written, Bulgaria became one of the Great Powers of Europe.
Omurtag’s successor, his youngest son Malamir, took up hostilities against the empire, provoked by a Byzantine plan to sail up the Danube and remove a community of Greek prisoners settled north of the river. In 836 he annexed the unfortified city of Serdica and put in a Bulgarian garrison. He was clearly establishing himself on the road to Macedonia. On the termination of the Thirty Years Peace he sent troops into the Struma valley at a time when the Byzantines were hampered by a revolt by the Slavs of the Peloponnese. There seems to have been a large-scale westward expansion into the largely ungoverned country of northern Macedonia, which led to a war with Serbia in 839-42, but the details are inaccessible to us. In Thrace itself Philippopolis and Philippi were annexed by the Bulgars, and the surrounding Slav populations absorbed into the Bulgarian state. There seems to have been a truce which probably granted to the Bulgars the right to expand into Macedonia, an expansion which the Byzantines could not in any case hinder. Khan Malamir died in 852 and was succeeded by his nephew Boris, the son of his elder brother. With the accession of Boris begins the century which forms the subject of the present comparative study. Before closing this historical introduction we must glance again at the situation in the southern part of the Balkans, peninsular Greece.
In the seventh century the Byzantines were in no position to devote attention to Greece, which lay off the main lines of communication of the empire.
We hear of Sklaviniai in various parts. The effective range of Byzantine control was confined to a zone in central Greece around Thebes, Athens and Corinth, and to a series of ports with their immediate hinterland—Thessalonika, Nauplia, Monemvasia, Patras. The islands of the Aegean and the Ionian islands also remained firmly in Byzantine hands, at any rate until the loss of Crete to the Arabs in about 826. Areas outside Byzantine control were not necessarily inhabited entirely by Slavs, as has been remarked. The Slav communities in Greece neither established a political unity of their own nor had one superimposed upon them from outside, as happened to those of the northern Balkans under the Avars and Bulgars. They lived in tribal units, each governed by its council, and forming ephemeral alliances among themselves for particular purposes. Within this tribal society the beginnings of a military aristocracy were forming, and this new ruling class was often strongly attracted by the prestige of the Greek way of life. Even for the ordinary tribesman, contact with Greeks must often have been close. The needs of trade brought Slav peasants to the Greek cities and Greek merchants to the Slav villages. And there were certainly Greek peasant communities surviving in various parts of Greece out of the reach of Byzantine state power. So a gradual process of Hellenisation and naturally of Christianisation, whose steps we cannot trace, began early and continued uninterrupted. Byzantine society was not racist. A Slav who spoke Greek and was an orthodox Christian found few doors closed to him. Thomas, a member of one of the Slav communities settled by the Byzantines in Asia Minor, became a leading officer under Leo, and led a revolt against his successor Michael II, in the course of which he was himself proclaimed rival emperor. The revolt was unsuccessful. The Patriarch Nicetas (766-80) is said to have been a Slav, but the evidence is uncertain. We hear of missionaries working among the Slavs of Greece during the seventh century. The liturgical language used was Greek, and Christianisation must have soon brought Hellenisation in its wake. Nevertheless the bulk of the Slavs in Greece were at the beginning of the eighth century both pagan and independent of the Byzantine government. There was a major attack by the Peloponnesian Slavs on Patras in about 807. Its defeat was followed by punitive expeditions and a serious effort to bring the Peloponnese once again under Byzantine control. At the same time the Slavs near the mouth of the Vardar were still pagan and practised piracy. But in the early years of the ninth century the area under Byzantine control round Thessalonika was extended,
and by the middle of the century a group of new themes had been organised in this region. New episcopal sees make their appearance, particularly in the Peloponnese, testifying to the absorption of the Slav settlements into the Byzantine community.
By the middle of the ninth century, then, the Slavs of peninsular Greece and southern Macedonia had formed no political society of their own, and were well on the way to absorption into the life of the Byzantine world, which soon led to loss of their national identity and their language, except for a small pocket in the mountains of the southern Peloponnese. In the northern Balkans the situation was very different. There the Slav settlers had been united under the power of the Bulgar state, a state in which from the beginning the Slavs had played a major role, and in which by the middle of the ninth century the fusion of the Bulgar and Slav ruling groups had made considerable progress; the Bulgars had partly forgotten their original Turkic language and had adopted that of their Slav fellow-citizens. It is probably more than accidental that the line dividing the two regions of the Balkan peninsula coincides closely with that marking the northern limit of the cultivation of the olive. It is a line which separates two very different styles of life, the Mediterranean and the central European.
3. Bulgaro-Byzantine Relations in the Ninth and Tenth Centuries
When Boris succeeded to the throne of Bulgaria in 852 he found on his southern frontier a Byzantine empire which was rapidly recovering from the long eclipse following on the Arab conquests and the Slav and Avar occupation of the Balkans. There was an economic recovery which filled its treasury with gold. A new interest among the ruling class in its own past led to an educational and cultural renaissance. New public buildings, in a style which combined motifs from late antiquity with innovations, some probably of oriental origin, were constructed everywhere. And the Byzantine army turned from a defensive to an aggressive role. Recovery and restoration were the watchwords. Boris must have viewed with disquiet the growing military strength of his powerful southern neighbour. Fortunately for him the first priority for the Byzantines was to win and hold the strategic passes through which the Arabs had made their regular raids into Asia Minor. Asia Minor, not Thrace, was the economic and demographic centre of the empire.
To enable it to concentrate its forces in the east the Byzantine government of the empress-regent Theodora and the logothete Theoctistus was prepared to make concessions to the Bulgarians. Shortly after Boris’ accession the Byzantines ceded to Bulgaria a belt of territory some 25 miles wide south of the old frontier in Thrace, and including the now ruined fortresses of Develtos and Anchialos. Encouraged by this apparent weakness Boris concentrated his military effort in the far west, pushing through Macedonia to reach the mountains of Albania and the northernmost peaks of Pindus. This was all territory to which Byzantium laid claim, but in which there had been no effective Byzantine authority for two and a half centuries. There was little the Byzantines could do to stop Boris if they would.
In 856 Michael III and Bardas replaced Theodora and Theoctistus at the head of affairs in Constantinople. The new ruler pursued a vigorous policy of aggression in the east, against both the Arabs and the Paulicians. But in the far west he was unable to stem the Arab advance in Sicily and southern Italy. At the end of his reign only Syracuse and Taormina remained in Byzantine hands in Sicily. Michael had therefore an interest in continuing his mother’s policy of peace with Bulgaria.
The unexpected Russian attack on Constantinople itself in 860 directed Byzantine attention to the danger from this new state emerging far away in the heart of north-eastern Europe, and made friendly relations with Bulgaria all the more desirable. Nevertheless when Boris, anxious to build up Bulgaria into a great power, formed a military alliance with the Franks, the Byzantine reaction was swift and sharp. A Byzantine army invaded Bulgaria, supported by the fleet in the Black Sea, and no doubt in the Danube delta. Boris was caught at a disadvantage. In any case he wanted at all costs to avoid a full scale military confrontation with the empire, particularly since a series of striking victories by the general Petronas on the eastern frontier had enabled troops to be transferred from there to Thrace. He at once accepted Byzantine terms. These were not onerous from the military point of view. But they required that Boris and his people should accept the Christian faith and that the Bulgarian church be subordinate to that of Constantinople. For a further analysis of the implications of this for Bulgaria, see Chapter Eight. For some years before, the Byzantine authorities had been deeply interested in conversion as a means of extending Byzantine influence. The concept arises naturally out of the Byzantine idea of the empire as a unique instrument of God’s plan for the salvation of mankind, destined to last until the Second Coming. Acceptance of the spiritual authority of the church implied in principle acceptance of the temporal authority of the emperor and vice versa. It may have been brought to the forefront by recent Byzantine experience in the east, where mass conversion of Moslems, Paulicians or Monophysites often followed on Byzantine conquest. Missions had recently been sent to the Khazars, to Russia and to Moravia. As it turned out, none of these produced lasting results. But in 864 this was not evident. Bulgaria could not be allowed to remain in pagan independence. So Byzantine insistence on its conversion as a condition of withdrawal of its invading army was only to be expected. It is a neat example of the intimate collaboration of Church and state in the Byzantine empire. As a counterpart to the conversion the Bulgarians probably received some territory in Thrace. And the frontier was left vague in the west, where Bulgarian expansion continued unchecked.
The ensuing rebellion of boyars hostile to a rapprochement with Byzantium nearly cost Boris his life. But it was suppressed, and the opportunity taken of replacing many of the traditionalist Bulgar and Slav clan-leaders by men sympathetic to Boris’ plans and dependent on his patronage.
Boris himself realised the danger of becoming a Byzantine satellite, and for a time flirted with the church of Rome, which was at that moment anxious to assert its power in the Balkans. The Byzantine clergy were expelled from Bulgaria for some years. But in the end the realities of power told. Bulgaria returned in 870 to the Byzantine obedience but with a degree of internal autonomy in church affairs which was the price paid by the Byzantines for the abandonment by Bulgaria of its connection with the church of Rome and its search for allies in western Europe.
A new spirit of cooperation prevailed between Bulgaria and Byzantium. Young Bulgarians were sent to Constantinople for education. Byzantine craftsmen were sent to Bulgaria to build churches and palaces. But Boris and his colleagues were well aware that Byzantine ideas could be as dangerous as Byzantine arms. A Greek speaking church whose clergy in the last resort owed allegiance to the emperor in Constantinople could in a short time sap the self-confidence of the Bulgarian people and alienate those very groups in Bulgarian society whose support was essential. When in 885-6 a group of pupils of Cyril and Methodius, who had been sent by the Byzantine church to evangelise Moravia, arrived in Bulgaria with liturgical books in Slavonic, Boris welcomed the opportunity to form in his kingdom a Slavonic church, whose clergy would be native Bulgarians, who would preach in the language of the people, and who would neither feel in themselves nor inspire in others an overriding loyalty to the Byzantine empire. Slavonic clergy were trained by Naum in north eastern Bulgaria and above all by Clement in Macedonia, and liturgical and other works were translated into Slavonic under Boris’ patronage.
Meanwhile the death of Basil I in 886 and the consequent deposition of Photius and restoration of Ignatius to the Patriarchate reduced the immediate likelihood of direct Byzantine intervention in Bulgaria. In 889 Boris, who must by now have been in his sixties, abdicated in favour of his eldest son Vladimir or Rasate and entered a monastery. Vladimir, who had probably been kept too long in the background, was closely linked with those backward-looking elements in the Bulgarian ruling class who had revolted against Boris in 864. There was a violent swing against Boris’ policy of building up an independent church without provoking the Byzantines and against the austerity which he had imposed upon the Bulgarian court. Some sort of steps seem to have been taken towards the formal restoration of paganism. There may also have been overtures made to the Pope.
The whole policy which had avoided open war with Byzantium for nearly sixty years, while Bulgaria transformed itself from a federation of tribes and clans into a centralised feudal state was in danger of being frustrated. The Byzantines could not have stood by while an anti-Christian and anti-imperial régime was installed at Pliska. Boris, who may have tried to exercise some kind of control over Vladimir from his monastery, emerged from his cell in 893, rallied his old guard and, making the most of his immense personal moral authority ousted his son from power and had him blinded. In the delicate situation in which he found himself he convened a council of boyars, attended by provincial governors and high officers of state as well as by Bulgar and Slav tribal leaders. He justified his action to them and obtained their agreement to accept his younger son Symeon as ruler. Symeon had received a Greek literary education in Constantinople and was at the time of his accession a monk in Bulgaria. He had probably been destined by his father for high ecclesiastical office. His education had given him an intimate knowledge of Byzantine life and a deep hatred of the Byzantine aristocracy who had treated him contemptuously as ἡμίαργος — half-Greek, ‘métis’. And his years of contemplation in the monastery had given him a taste for action. Two other matters settled at the council of boyars in 893 represented the fruition of much of Boris’ work. The capital was transferred from Pliska, which had many pagan and Proto-Bulgar associations, to the new royal residence of Preslav. And Greek was replaced by Slavonic as the language of liturgy and teaching of the church and the language of internal administration of the state. At the same time a Slav archbishop was probably appointed.
The Byzantine government might have reacted sharply. But Leo VI and his advisers were probably glad to see the unpredictable Vladimir replaced by a brother whom they regarded as a Byzantine protégé. In any case Leo, though a learned man, prolific legislator and active re former of Byzantine administration, had no taste for military matters, and worse still, no foreign policy. And he was under the influence of his father-in-law the basileopator Stylianos Zaoutzes, an able man who shared his lack of interest in foreign affairs and who was too ready to reward his supporters.
Symeon seems from the outset of his reign to have decided to reverse his father’s policy of avoiding a confrontation with the Byzantine empire. He was aware of the opposition in many quarters in Bulgaria to what looked like weakness. And he calculated that unless the balance of power was changed, Bulgaria was bound to be brought into the political orbit of Constantinople and to lose her freedom of action, whether or not she succeeded in maintaining an independent Slavonic church.
Bulgaria’s strength, he believed, lay in her army, a formidable force of peasant soldiers well equipped and led, while the Byzantine army was stretched by its commitments in the east and weakened by the collapse of the system of military land holdings. Boris had sought to avoid a military conflict, Symeon awaited one eagerly. His long-term aims are not always easy to discern. At his accession he probably hoped to establish Bulgaria as a state equal in standing to the Byzantine empire, equal in military power and owing no debt of ultimate submission to the emperor at Constantinople. This was a direct challenge to the whole Byzantine concept of a unique oecumenical empire, destined to last until the end of the world. Symeon understood this concept very well, but he rejected it utterly.
The casus belli for which Symeon was waiting was not long in coming. Bulgarian merchants visiting Constantinople to sell their wares were housed in an official lodging, a mitaton, as were merchants from other foreign lands. Their stay in the capital was limited to three or six months at a time, and their goods had to be sold in their mitaton, not in the open markets of the city. It was in the mitaton, too, that the ten per cent tax was levied on the goods they brought, of which they had to pay one half while the other half was payed by the purchaser. In return for this limitation of their freedom they were guaranteed against harassment by Byzantine local authorities and exploitation by private merchants. These matters were settled in commercial treaties between the Byzantine and foreign governments, and were part of the system of international trade. A lively exchange of Bulgarian hides, wax, flax, honey and other goods for Byzantine products was conducted under these arrangements. In 894 two Greek merchants, Staurakios and Cosmas, used their influence with the all powerful Stylianos Zaoutzes to have this system set aside. The mitaton for Bulgarian merchants was suddenly transferred from Constantinople to Thessalonika and the collection of the taxes farmed out to Staurakios and Cosmas, who promptly raised the duty on Bulgarian goods. This was a step completely at variance with Byzantine practice at the time. According to our sources, which are all hostile to Zaoutzes, Leo VI later repented of his unwise decision and punished Staurakios and Cosmas severely, while only his impending death saved Zaoutzes from sharing their disgrace. The true motive of Leo in sanctioning such a departure from normal usage can never be discovered. It may be the mixture of corruption and muddle which the sources suggest. It could be a step taken for security reasons. Be that as it may, it was a severe blow to Bulgarian trade which followed the Belgrade-Constantinople road, the Via Egnatia or the Black Sea coastal road to Constantinople.
Medieval sovereigns were not usually sensitive to mercantile interests. But Symeon reacted at once. A personal request to the emperor to annul the offending order met with refusal. Immediately the Bulgarian ruler—who only a year earlier had still been a monk— marched at the head of his army into the Byzantine theme of Macedonia, i.e. western Thrace. The main Byzantine forces were in Asia Minor. The troops sent out from Constantinople were crushingly defeated by Symeon and their commander Procopius Crenites killed, along with many of his officers. Symeon seized every opportunity to humiliate the Byzantine emperor in the eyes of his subjects, for example by cutting off the noses of soldiers of the imperial guard whom he captured and returning them to their master. This was the first battle between Bulgarian and Byzantine troops for more than half a century, and the outcome was a signal Bulgarian victory.
The Byzantines dared not disengage their main field army in the east to face the new threat in the west. But though they were militarily inferior their diplomatic possibilities were much greater. The patrician Nicetas Skleros was sent by sea with Byzantine gold to the Magyars, at that time settled north of the Danube delta, on the frontier of Bulgaria. He persuaded the Magyar leaders Árpád and Koursanis to attack the Bulgars in the rear. This was the classic move of Byzantine diplomacy, and one no doubt prepared in advance as a contingency plan.
While the Magyars attacked from the north, the Byzantine land army marched through Thrace under Nicephorus Phocas and the fleet sailed to the Danube mouth under the patrician Eustathios. Leo probably expected Symeon to yield to this threat without a battle, and even sent an envoy to his camp to offer peace terms. But he had misjudged his man. Symeon in vain tried to block the Danube crossings, but Eustathios’ fleet helped the Magyars to cross the river. They swept through Bulgaria, pillaging and burning, and Symeon’s army fled before them without offering combat. Symeon himself took refuge in the Danube fortress of Silistra while the Magyars went on to sack his capital of Preslav before withdrawing across the Danube. Symeon asked the Byzantines for peace terms, and Leo unwisely withdrew his forces from Bulgaria.
When Leo Choirosphaktes, the Byzantine ambassador, reached Symeon’s court he was thrown into prison. Symeon had meanwhile followed the Byzantine example and was negotiating with the Pechenegs, a Turkic people then established in the steppes north of the Black Sea.
Bulgarians and Pechenegs fell on the Magyars from both sides, inflicting upon them a terrible defeat and ravaging their territory. The demoralised Magyars crossed the Carpathians into the plain of Hungary, -where they soon became the terror of western Europe. Symeon now felt himself in a stronger position vis-à-vis the Byzantines, and refused to make peace unless the Bulgarian prisoners, whom the Byzantines had bought from the Magyars, were returned to him. The emperor Leo, who had signally failed to realise the radical change of Bulgarian policy since the accession of Symeon, agreed. The prisoners were returned and negotiations on peace terms begun. But Symeon, having recovered his lost subjects, broke off the negotiations and once again imprisoned the hapless Leo Choirosphaktes. The emperor decided at last to transfer troops from the east and sent a large army into Bulgaria. Symeon caught it in an ambush at Bulgarophygon near Adrianople and defeated it with crushing losses.
By now it was 896. It was the turn of the Byzantines to seek to ransom their prisoners. We have some of the correspondence between Symeon and Leo Choirosphaktes on this matter. Symeon, who had been educated in Constantinople, propounded to the Byzantine ambassador a problem well known to Greek logicians. He would return the prisoners if the emperor answered correctly one question. That question was: ‘Did he intend to return them?’ Leo recognised the trap and made a suitably ambiguous reply. For some time the exchange of letters went on, Symeon and the ambassador alike displaying their familiarity with the finer points of classical philosophy. The point of this curious correspondence was no doubt to give Symeon the pleasure of beating the Greeks at their own game and to gain time. He probably also hoped to impress the Byzantine dignitaries with his own sophistication. Bulgaria could no longer be treated as a land of barbarians. In the end terms were agreed upon: 120,000 Byzantine prisoners were returned, and the emperor agreed to pay an annual tribute to Bulgaria. There was also a territorial adjustment in Thrace in Bulgaria’s favour. Symeon had succeeded in defeating the Byzantines in the field, and then humiliated them at the conference table. He was being treated by the Byzantine emperor as a near equal, as the King of Persia had once been treated, or the Caliph of Islam. He was well pleased with his achievements, though they were bought at the cost of fearful devastation of much of his country. Whether old Boris in his monastery was equally pleased by the new turn in policy made by his bookish son, the erstwhile monk, is another matter.
The treaty of peace remained in effect until the death of Leo VI in 912. These were difficult years for the Byzantine empire. The war with Symeon had given a breathing space to the Arabs, and Armenia was at once lost to them. Taormina, the last Byzantine strong point in Sicily, fell in 902. And Arab naval strength in the Aegean grew until their fleets could sail where they wished. Demetrias in Thessaly was sacked in 902. In 904 a great Arab fleet under the command of Leo of Tripoli, a Greek renegade, entered the Aegean, captured Abydos, then turned west and seized Thessalonika, the second city of the empire. There was a massacre of the citizens, and measureless booty and count less prisoners were taken. Apart from these losses, Byzantine prestige suffered a serious blow. Symeon did not fail to profit by the empire’s defeat. The western frontiers of Bulgaria were probably not defined by the treaty. At any rate Symeon pushed south and west in Macedonia. An inscribed stone dated 904 from Nea Philadelpheia 22 km north of Salonica marks the ‘frontier between the Romans and the Bulgarians’ (Beševliev No. 46). In the far west Bulgarian territory reached the Adriatic between Hagioi Saranta and Dyrrhachium, on the coast of present-day Albania. This extension of territory was obtained by a series of local advances, each of which the Byzantine authorities were prepared to ignore, since they were unwilling to face the possibility of open war with Bulgaria.
Leo VI’s laissez-faire policy in regard to Bulgaria was not to the taste of all his subjects. When Leo died in 912 his successor, his brother Alexander, a man of little judgement or experience of affairs, decided to seek popularity by taking a strong line with Symeon. He peremptorily and insultingly cut off the tribute due to Preslav under the treaty signed by Leo. Symeon, who had consolidated his territorial gains and replaced the soldiers lost in the previous war, was waiting for just such an opportunity to be offered him. In 913 he marched the main body of the Bulgarian army right through Byzantine Thrace without opposition and encamped before the walls of Constantinople itself. Inside the city all was in disarray. Alexander had died in June, and a council of regency acted for the seven-year-old son of Leo VI, Constantine VII. Its leading members were the Patriarch Nicolaus Mysticus and the empress Zoe, widow of Leo VI. They had no concerted policy, and they were faced with an attempt at usurpation by Constantine Ducas, the commander-in-chief of the army, which they put down with difficulty.
Symeon was no longer bent on booty or the accretion of territory.
From aiming at establishing a Bulgarian state parallel to the Byzantine empire he had by now advanced to a new ambition, that of becoming himself ruler of an enlarged empire which would include Bulgaria, and of putting on his own shoulders the mantle of legitimacy of the Christian Roman emperors. It was not an absurd ambition. His ethnic origin might provoke a raised eyebrow or a curled lip in some Byzantine milieux, but Byzantine society was not racist. Men who were not Greek by descent had ruled the empire before. What was required was orthodox religious belief, Greek culture, and success. Symeon had the first two of these, and the third seemed within his grasp. The walls of Constantinople, however, proved impregnable from the land, and Symeon had no fleet. He was forced to negotiate, but he could at least negotiate from strength. He and his escort were received within the walls with every mark of respect by the regency council and the boy emperor. The concessions made to him were unprecedented. He was naturally given his arrears of tribute. The young emperor was betrothed to one of his daughters. And a coronation ceremony of some kind was held in the church of the Virgin at Blachernae, at which the Patriarch put upon Symeon’s head not a diadem but some part of his own liturgical head-dress and declared him Basileus. The interpretation of this ceremony has been disputed ever since it took place. Some believed then, and believe now, that Nicolaus Mysticus crowned Symeon as co-emperor with Constantine VII. If he did so, it must have been without the agreement of the empress Zoe. Others believe that the whole ceremony was a meaningless sham designed to impress a gullible barbarian. But Symeon was far too well acquainted with Byzantine ways to be easily taken in, and far too dangerous to provoke. Some degree of ambiguity was essential to the Patriarch if he was to retain the confidence of his fellow regents; and it was acceptable to Symeon, who imagined he would only have to wait a little for his daughter to become empress and he himself the power behind the throne, and in due course for his grandson to become emperor. He probably accepted something which to Byzantine eyes could be represented as coronation of a Basileus of Bulgaria—and the Byzantines had recognised no other sovereign as Basileus since the Sassanid monarchy of Persia had been overthrown by the Arabs - and to Bulgarian eyes might be taken to imply partnership in the Byzantine empire. The whole matter is highly mysterious and already overlaid in our sources with much self-exculpation. 
However we are to interpret what took place, it appeared to put the mastery of the empire in Symeon’s hands. He returned to Preslav in autumn 913 to await the full fruits of his victory. However he had misjudged the strength of the opposition to his plans in Byzantine ruling circles. The empress Zoe felt that she and her son had been betrayed, and many of the clergy regarded Nicolaus as uncanonically appointed, and remained loyal to the deposed Patriarch, Euthymius. As the attempted usurpation by Constantine Dukas had shown, Nicolaus could not count on the unconditional support of the army or its leaders. The concessions made to the Bulgarian ruler united the various elements hostile to Nicolaus. He was removed from the council of regency, and the dowager empress Zoe assumed charge of affairs. The projected marriage between Constantine VII and Symeon’s daughter was cancelled and Symeon’s coronation declared to be null and void. Success seemed to be slipping from Symeon’s grasp. But his military power remained undiminished. In 914 his armies once again invaded Thrace, while Symeon demanded that the Byzantine population should recognise him as legitimate emperor. In the following campaigning seasons he made similar incursions into the regions near Dyrrhachium and Thessalonika, demonstrating the power of Bulgarian arms and doubtless demanding recognition of his coronation in 913.
The government of empress Zoe dared not continue to allow Symeon to march where he liked through Byzantine territory proclaiming himself to be a legitimate emperor. If it was to hold on to power it had to counter-attack. But recent experience had made Byzantine commanders cautious about challenging the Bulgarian army in the field. So a combined operation by land and sea was carefully planned. The land army, under Leo Phocas, was to march up the Black Sea coast road, while the fleet, under Romanus Lecapenus, was to go ahead of the army at sea, attack behind the enemy lines and cut his communications. The aim of the expedition was to strike into the centre of Bulgarian power in north-eastern Bulgaria, and if possible to capture Preslav, as Nicephorus I had captured Pliska a hundred years before. There were negotiations with the Pechenegs to induce them to attack Bulgaria from the north. In summer 917 the expedition moved off. Leo Phocas, however, was no match for Symeon. The Bulgarian army fell on the Byzantines by surprise by the river Achelous, near Anchialos on the Gulf of Burgas, on 20 August. The Byzantine army was utterly routed. Leo Phocas only just succeeded in fleeing to the well-defended city of Mesembria. The bones of the dead were still to be seen scattered on the ground fifty years later.
Byzantine plans were completely frustrated. When Romanus Lecapenus and his fleet reached the Danube mouth the Pechenegs refused to play their part. In Constantinople Nicolaus Mvsticus came to the fore again. In a long letter to Symeon, written immediately after receiving the news of the defeat at the Achelous, he expatiates on human liability to error, declares the loss of life to be the fault of Byzantines and Bulgarians alike, points out that he himself had no hand in the military plans, argues that in any case the Byzantines had not seriously intended to attack Bulgaria, only to relieve the pressure round Thessalonika and Dyrrhachium, and conjures Symeon in the name of Father, Son and Holy Ghost not to pursue his military advantage. It is a curious letter, and no doubt Symeon took careful note of it when it was delivered to his camp by the abbot of a well-known monastery. But he continued to pursue the survivors of the Byzantine army down the coast road towards Constantinople, and inflicted a second crushing defeat on them in a night-attack at Catasyrtae not far north of the capital.
He realised that it was useless to sit down once again before the walls of Constantinople. Access to the city depended on political rather than military victory and the Patriarch was clearly hesitating. Symeon no doubt had adherents within the city, who could put pressure on him and on others in high places. In the meantime he withdrew with his army to Bulgaria, in part to deal with the situation in Serbia. Byzantine diplomacy had long been working there to create an anti-Bulgarian group among the various tribal princes, and had had some success. A Bulgarian punitive expedition broke up the hostile coalition. In 918 Symeon again allowed himself to be distracted from his principal objective, and invaded northern Greece, pushing south as far as the Gulf of Corinth. This seems the most probable date for his Greek campaign. It is said in the Life of St Luke the Younger  to have taken place ‘ten years’ before Symeon’s death. This would put it in 917, an impossible year, as Symeon was then engaged in the Achelous campaign. Scholars have sought various ways out of the difficulty. At any rate the campaign must be dated some time between 917 and 921, not 916 as Runciman suggested. The question poses itself why Symeon allowed himself to be diverted from his main objective of keeping up the pressure on Constantinople. It may be the result merely of the restlessness of this man of action. He may have hoped to induce the Byzantines into transferring troops from the capital to peninsular Greece; if so he failed in his intention.
The most fundamental explanation is probably to be sought in the essentially aggressive character of the Bulgarian state. The heavy cost of maintaining a large army in the field could not be met out of the economic surplus normally available within Bulgarian society. Booty and new territory to exploit were essential, particularly as more and more of the available economic surplus was taken by feudal proprietors instead of being at the disposal of the state.
Whatever the explanation of Symeon’s deviation from his objective, it proved fatal to his success. In Constantinople the ineffectual regency council, dominated by priests and women, steadily lost the support of the ruling class. A strong man with a military background was what they wanted. The two principal candidates were Leo Phocas, son of the great commander Nicephorus Phocas and commander-in-chief of the army, and Romanus Lecapenus, commander of the imperial fleet. Phocas belonged to one of the most powerful families of territorial magnates and Lecapenus was the son of an Armenian peasant who had worked his way up in the imperial service. In the end it was Lecapenus who manoeuvred himself into power. Perhaps he was merely a better intriguer than the aristocratic Phocas; perhaps the series of defeats which the army had suffered at the hands of Symeon made its commander an implausible saviour of the state; perhaps men understood that what gave the empire a decisive advantage over Symeon’s Bulgaria was its command of the sea, which must be held on to at any cost. Be that as it may, Romanus Lecapenus had himself appointed a member of the regency council and gradually ousted the dowager empress Zoe, who favoured Leo Phocas, and her friends and advisers from positions of power. In May 919 he consolidated his position by marrying his daughter Helena to the fourteen-year-old Constantine VII and by taking the title of Basileopator, formerly held by the all powerful Stylianos Zaoutzes under Leo VI. In the next year he was raised by Constantine first to the rank of Caesar then to that of co-emperor, and effective ruler of the empire.
Symeon had lost the political battle that might have brought him to the head of a revitalised Roman empire of which Bulgaria would have been an essential part. A few years before he had hoped to be father-in-law of the emperor and the power behind the throne. Now Romanus Lecapenus was in that position, with the support of the mass of the citizens, even if many of the landed aristocracy were lukewarm or sullenly hostile. From 920 until his death Symeon continued military operations against Byzantium, using up more and more of Bulgaria’s human and material resources in what was by now a vain endeavour.
He could defeat Byzantine armies and take provincial cities. But without a fleet he could not bring Constantinople to capitulate. He could devastate the European provinces, but he could not carry the war into Asia Minor, the demographic and economic heartland of the empire. And he had not the charismatic prestige, the countless personal connections, and the reserves of gold and luxury artefacts which enabled the empire to build up a network of alliances surrounding its enemies. Symeon, unlike his father, thought in military rather than diplomatic terms.
In 920 the Bulgarian army was engaged in Serbia, where once again the Byzantines had succeeded in establishing an anti-Bulgarian coalition. At the same time an attempt was made to seize key positions in Thrace giving control of the Dardanelles. Symeon was evidently testing the possibility of taking the war into Asia Minor by by-passing Constantinople. In 921 the Bulgarians appeared once more before the walls of Constantinople, and again in 922. In 923 Adrianople, the key to Byzantine Thrace, fell to Symeon. Throughout these years he had been in correspondence with the Patriarch Nicolaus Mysticus, through whom he proposed that Romanus Lecapenus abdicate, that a marriage alliance be established between his own family and that of Constantine VII, and so on. Needless to say these overtures were rejected. With Adrianople in his hands, however, he was in a stronger negotiating position, for there was great fear in Constantinople. But Romanus lecapenus counted on the walls of Constantinople and his own diplomacy to save him. He had been for some time negotiating with the Magyars, Pechenegs and Russians to create a grand alliance favourable to Byzantium on the northern frontier of Bulgaria. Symeon too had been negotiating with the Fatimid Caliph of Africa for naval support. This was an obvious step for him to take, and if successful it would have changed the situation radically in his favour. But it presented religious difficulties which might have seriously weakened the allegiance of many of his subjects. And above all Bulgaria had little to offer that the Caliph wanted. However an agreement was reached in 923-4 and in 924 Symeon led his army again to the walls of Constantinople to await the arrival of the ships of his Moslem ally. By September he realised that the Moslem fleet was not coming and that the Byzantines had outbid him at the Caliph’s court. His hopes dashed, Symeon asked for a meeting with the emperor. For the last time he entered the city in which he had passed his youth.
Byzantine chroniclers recount at length the meeting of the two monarchs, including the significant detail that Symeon was accompanied by a numerous suite of soldiers and civilians who constantly hailed him as Basileus in the Greek tongue. Romanus too did all possible to impress his guest and the bystanders with his power. The same chroniclers are curiously silent about the results of the meeting. Was any kind of truce arrived at? Probably not, though neither side pressed things to a definite rupture. For Symeon the meeting in September 924 marked the failure of his grand project. He had not succeeded in taking the city or in breaking through to Asia Minor. He had not won the adherence of a party in Constantinople. And now the ominous movements on his northern frontier required his immediate return to Preslav. He continued to style himself Emperor of the Romans and Bulgarians and to provoke diplomatic protests from Constantinople. In 926 he raised the archbishop of Bulgaria to the status of a patriarch, independent of and of equal rank with the Patriarch of Constantinople.
In the meantime Byzantine diplomacy, disappointed in the unstable Serbian principalities as instruments of its policy, had turned to the powerful Croatian Kingdom of King Tomislav to the north-west. An alliance was arranged. When Symeon marched through Serbia in 926, overthrowing hostile princes and massacring countless peasants in a show of strength, and then sent a major force into Croatia, it was annihilated by Tomislav’s army. Only the intervention of the Pope, who was worried about Croatian intentions towards the cities of the Dalmatian coast, saved Bulgaria from an invasion by the victorious Croats. It can have been little consolation to Symeon that the Pope, in his correspondence, accorded to him the title of emperor. As he was planning a further campaign against Byzantium he died in his capital of Preslav on 27 May 927. Byzantine chroniclers recount that an astrologer identified a statue in the Xerolophos quarter of Constantinople, near the forum of Arcadius, as the Stoicheion of Symeon, the physical object to which his fate was mysteriously linked. Romanus Lecapenus sent men by night to cut off the head of the statue, and at the same moment Symeon died in Bulgaria. The story probably reflects official propaganda based upon popular superstition, and testifies to the impression which Symeon made on his contemporaries.
After Symeon’s death a striking change takes place in Bulgaro-Byzantine relations. Thirty-three years of almost unbroken war are succeeded by forty years of peace. From challenging Byzantine power in the Balkans and seeking to take over the empire’s role as the one fully legitimate Christian state, Bulgaria becomes a docile dependant of Byzantium.
At the same time the Slavonic Christian culture initiated under Boris and encouraged by Symeon in spite of his overriding military preoccupations continues to develop and flourish. Bulgaria does not become a cultural province of Byzantium. The reasons for this sudden change are not easy to discern with precision. Personalities play some role. Symeon’s successor, his second son Peter, was a colourless character unable to inspire either loyalty in his fellow-countrymen or fear in his enemies. The exhaustion caused by long years of war had certainly weakened Bulgaria militarily and economically. The losses in men of military age cannot be estimated, but they were certainly very high. Apart from the annual campaigns against Byzantium, Bulgaria had been repeatedly involved in hostilities in Serbia. There was also a running war with the Magyars after they settled in the Danubian plain; we have scarcely any information on its progress, but it doubtless contributed to the steady drain on Bulgarian manpower. Bulgaria proper had not been invaded since the Magyar invasions of the ’90s, so there would have been little recent damage to the economic infrastructure. There must however have been considerable interruption to the international trade to and from Constantinople passing through Bulgaria. The Byzantines would suffer less from this interruption than the Bulgarians, since this trade formed a small part of their total international trade, and in any case some of it could be diverted to other routes, particularly sea routes. A factor often forgotten was the loss to the Magyars during Symeon’s reign of Transylvania which Krum had won from the Avars in 805, leaving only the plain of Wallachia as a survival of the once vast Transdanubian Bulgaria. This loss must have reduced both the man-power available to the Bulgarian army and the revenues in kind accruing to the government from direct taxation of free peasants. We do not know whether the mines of Transylvania were exploited during the period of Bulgarian rule. If they were, this would aggravate the effects of the loss of most of Transdanubia.
Probably the most important factor of all in Bulgaria’s decline was the growing power of the feudal aristocracy, who diverted into their own pockets resources previously available to the community. It is difficult to trace this growth, whose effects become obvious in the mid and later tenth century. The evidence, such as it is, is presented elsewhere. If this growing feudalisation took place at a time when Symeon’s wars causing a temporary shortage of able-bodied men, landowners would do all in their power to keep the peasants tied to their estates. Furthermore the whole system of recruitment of the Bulgarian army,
which was essentially based on tribal levies, must have been affected by this radical and rapid change in the relation between man and man, and man and the land.
All these, and possibly other, factors underlie the sudden defusing of the Bulgarian situation after Symeon’s death in 927. His second son Peter succeeded him because his eldest son Michael was confined to a monastery for reasons which we cannot discern. This break with the long-standing Bulgar tradition of inheritance of the Khanate by primogeniture led to dissatisfaction among the boyars and magnates, culminating in the escape of Michael from his monastery in 930 to head a revolt in western Bulgaria.  For years he maintained his control of the western Macedonian mountains. Meanwhile Peter was married, within a few months of his father’s death, to Maria Lecapena, daughter of Romanus Lecapenus’ eldest son Christopher. His title of Basileus Boulgarōn was recognised, as was the independence of the Bulgarian patriarchate. The empire could afford concessions on points of protocol now that Bulgaria was no longer dangerous. An annual subsidy was promised to Peter, under the guise of maintenance for his Byzantine wife. The Bulgarian government was headed by George Sursubul, brother of Symeon’s second wife, who assumed the guardianship of Symeon’s younger sons, Ivan and Benjamin. Maria and her Byzantine entourage seem to have dominated the court, causing it to be distrusted by the mass of the people and by the territorial magnates.
There is no need to chronicle in detail the various exchanges of embassies between Preslav and Constantinople. They took place against a background of military defeat and political disintegration of the Bulgarian state. In 934 there was an invasion by the Magyars, which seems to have reached Develtos. They were probably aiming at Constantinople, but it was Bulgaria that was devastated. They repeated their raids in 943, 958 and 962. In 941 Prince Igor of Kiev led one of the great Russian attacks by sea against Constantinople. It was defeated. Igor then succeeded in negotiating an agreement with the Pechenegs, whose territory lay between Kievan Russia and Bulgaria, and whose friendship the Byzantines sought to maintain. The Pechenegs were to break their alliance with Byzantium and support a Russian land attack on Constantinople. The terrified King Peter of Bulgaria turned to Romanus Lecapenus for help. The emperor, by a lavish and judicious distribution of gifts, succeeded in dissuading Igor from pursuing his attack. The Pechenegs, disappointed of their expected booty, crossed the Danube and pillaged much of Bulgaria, while the Byzantines did nothing to help.
In Byzantine eyes Bulgarian losses did not count. It was not until 965 that Peter broke sufficiently free of Byzantine tutelage to negotiate an alliance with the German emperor Otto, directed particularly against the Magyars. But it seems to have been of little avail. In the face of this defencelessness local magnates became more and more independent of the government in Preslav and followed the example set by Prince Michael.
There was a wave of asceticism, as men sought refuge in monasteries and hermitages from a world which could not provide the minimal conditions of decent life. Peter himself was a devotee of holy men. The greatest of these was St John of Rila, a herdsman turned anchorite who lived for many years in a hollow oak, winning a reputation for sanctity that spread throughout Bulgaria. The monastery which he founded in the Rhodope mountains after his oak blew down flourishes to this day. He died in 946. Of more fundamental importance was the development among the mass of peasantry of a counter-church and counter-society, dualist in belief and irreconcilably hostile to Church, state and the established order of things. This drew substantial parts of the population into non-cooperation and civil disobedience against a society towards which they no longer felt any loyalty. These developments are analysed elsewhere in this book.
In 965 Maria Lecapena, the Byzantine queen of Bulgaria, died, and Byzantine influence at the court of Preslav diminished. The more warlike among the boyars called for an independent stand. However Peter continued to send envoys to Constantinople to collect the subsidy to which he had been accustomed. By this time Romanus Lecapenus had long ago been ousted from power by a palace revolution. His son-in-law and erstwhile protégé, the scholarly and cautious Constantine VII, had at last enjoyed the imperial authority which he inherited and had died in his turn. His son Romanus II had followed him to the grave after a short reign, leaving two infant sons, Basil and Constantine. Their mother, the dowager empress Theophano, had taken as her second husband Nicephorus Phocas, the commander-in-chief of the army, a member of a great family of magnates, and nephew of Leo Phocas whom Symeon had defeated at the Achelous. Nicephorus soon had himself proclaimed co-emperor with his young stepsons, and was the real ruler of the Byzantine empire in 965. An able general, a representative of the new class of feudal lords, and a convinced believer in the special status of the Byzantine empire, whose territory he was augmenting by his victories in Crete and on the eastern frontier, he was not a man to be trifled with.
Well aware that a party hostile to Byzantine interests was now in power in Preslav, he chose to strike a pre-emptive blow. Peter’s ambassadors were greeted with gross insults, the King called a prince clad in skins and their fellow-countrymen filthy beggars, and the request for subsidy—or tribute, as it had been called—was unconditionally rejected.
Realising that Nicephorus was bent on war, the faint-hearted Peter withdrew his request for money and apologised humbly for ever having made it. Nicephorus was unmoved, and sent his forces to reconnoitre the Bulgarian frontier. He wanted not merely to humiliate Bulgaria, but to destroy it and incorporate its territory in the Roman empire to which it properly belonged. But he was not anxious to fight on Bulgarian soil, where the many defiles in the mountains lent themselves to ambush and where so many Byzantine armies had met with disaster in the recent past. Like most good soldiers he did not like battles; they were too chancy. Others would fight for the empire’s interests. The patrician Kalokyras, a native of the Byzantine outpost of Cherson in the Crimea, was sent to the court of the pagan Prince Svjatoslav of Kiev with 1500 pounds of gold. Svjatoslav was easily persuaded to make war on Bulgaria though he had, as it turned out, other aims in mind than earning his bribe from the Byzantines.
In summer 967 Nicephorus accused Peter of having let the Magyars pass through his country to attack the empire—a transparent attempt to find a casus belli— and Svjatoslav crossed the Danube into Bulgaria with 16,000 men. Bulgarian resistance was crushed and the Russians swept through the country between the Danube and the Balkan mountains. During the winter Peter—or his boyars, since the old King had suffered a stroke at the news of the Russian invasion—made the only move open to them and called upon the Pechenegs, traditionally allies of Byzantium, to attack the Russians in the rear. This they did to such effect that Svjatoslav had to return in haste to save Kiev from capture. Then a dramatic change took place. Svjatoslav, incited by the treacherous patrician Kalokyras, conceived the project of attacking the empire itself through Bulgaria. He talked of establishing his capital at Little Preslav, the Bulgarian market town on the Danube, and extending his power as far south as he could. Much of this was idle dreaming. But Svjatoslav had immense resources of man power and was a formidable enemy. Nicephorus realised at once that things were not going as he had planned. He hastily sent a high officer of state to Preslav to sign an alliance with Bulgaria against the Russians and to make arrangements for the common defence of the Balkans.
While these negotiations were taking place King Peter died, on 30 January 969. He was succeeded by his elder son Boris, who had been retained long in Constantinople, partly as honoured guest, partly as hostage. In autumn 969 Svjatoslav crossed the Danube again with a Russian army supported by Pecheneg and Magyar mercenaries. The hastily-made defensive arrangements were of no avail. The Russian army swept through the northern provinces, took Preslav and captured the King and all his family, crossed the Balkan range into Thrace and took Philippopolis, where 20,000 citizens were impaled as a punishment for having defended their city.
In Constantinople consternation reigned. On 10 December 969 Theophano had Nicephorus Phocas murdered and her lover, the general John Tzimisces, proclaimed co-emperor and guardian of the young princes. Tzimisces, an experienced soldier, immediately took control of the situation. His first step was to exile the treacherous and impulsive dowager empress Theophano, who had raised him to the purple. Next he opened negotiations with Svjatoslav, offering him large subsidies if he would withdraw from what was rightfully the territory of the empire. The very existence of Bulgaria had been conveniently forgotten. Svjatoslav ’s reply was to order Tzimisces to withdraw into Asia, as he considered the whole European territory of the empire to be his. There was nothing for the Byzantines to do but settle down to along war.
It began in 970 with the defeat at Arcadiopolis of a Russian invading force by Bardas Sclerus, the new emperor’s brother-in-law, and its withdrawal into eastern Bulgaria. Tzimisces did not venture to pursue it, but waited until 971. In that year operations were further delayed by a revolt in Asia headed by Bardas Phocas, a kinsman of the murdered Nicephorus. Not until 972 did a well-prepared Byzantine army march into Bulgaria, while the fleet cruised along the Black Sea coast and up the Danube to cut off the Russians’ retreat. The Russians had left the mountain passes insufficiently guarded. Probably their army was largely engaged in police operations against the Bulgarian population. At any rate Tzimisces reached Preslav without a major engagement. The city was strongly defended, and it took several days of bitter fighting before it finally fell. By this time the Bulgarian capital had been reduced to a heap of ruins, among which lay the corpses of many of its inhabitants. It was never rebuilt properly, though its fortifications were patched up and it was rechristened Ioannopolis, after the emperor who had destroyed it.
A further series of battles and sieges, in which the fire-shooting ships of the fleet played a large part, forced the Russians to capitulate. Svjatoslav handed over all his prisoners, promised to leave Bulgaria for ever, and asked for the former treaties between Kiev and Constantinople to be brought into force again. Tzimisces, who had lost many men in the hard-fought battles, agreed to the terms proposed. The Bulgarian King Boris II was in the Byzantine camp. But neither he nor any other Bulgarian was consulted on the peace terms. The Bulgarian state was no longer recognised, and its former lands were treated as imperial territory. The Bulgarian royal treasure, including Symeon’s crown, was taken to Constantinople, where Boris II formally abdicated. His abdication put an end to the treaty obligations between Byzantium and Bulgaria. The Bulgarian patriarchate was allowed to lapse, and the church hierarchy brought under the control of Constantinople. The territory of eastern Bulgaria, devastated and depopulated, was incorporated into the Byzantine administrative system.
This was not quite the end of the story. Western Bulgaria and Macedonia had been untouched by the Russian invasion and were not under Byzantine occupation. The sons of a boyar, governor of one of the Bulgarian provinces, proclaimed themselves the successors of the dynasty of Krum and Boris and Symeon, and established an independent Bulgarian state in the west. The details of its origin are extremely obscure.
The western Bulgarian kingdom resisted Byzantine pressure for another half century. The organisation and administration of the country followed the pattern established by Boris and Symeon. But everything seems to have been on a smaller, more provincial scale. And though many churches and other buildings were constructed, Samuel’s court at Sofia, Vodena, Prespa or Ohrid was not a centre of Slavonic culture comparable to Preslav. For a time Samuel was able even to extend his territory at the expense of Byzantium. John Tzimisces made no further moves against Bulgaria before his death in 976. The army was probably too busy establishing and maintaining law and order in the newly-conquered provinces. His successor, the young emperor Basil II, now of age, was occupied for the first few years of his independent reign in dealing with the dangerous rebellion of Bardas Sclerus in Asia Minor. Samuel profited by making raids in all directions, and from 980 onwards drove southwards into the plain of Thessaly and besieged Larissa. In the end he captured it, helped by sympathisers within the city.
The shock produced by the fall of Larissa roused Basil II to action.
For the rest of his life he dedicated himself with almost paranoic single-mindedness to the conquest of the Bulgarian successor-state and the re-establishment of Byzantine power—and his own personal power— throughout the Balkan peninsula. There may have been deep psychological reasons for his sudden abandonment of the pleasures of Byzantine life to become a warrior-monk wholly devoted to the pursuit of power, but they were not the determining factor in the outcome. The Byzantine empire, prosperous and expanding, could not after all that had happened tolerate a Bulgarian state on its frontier. And a Bulgarian state based on the mountainous west of the country, far from the populous and fertile eastern provinces, a state which could only attack Constantinople and its neighbourhood by dangerously lengthening its lines of communication, could not hold out indefinitely against the full concentration of Byzantine military strength.
The war need not be recounted in all its picturesque detail. In any case gaps in our sources make it impossible to reconstruct all the campaigns. The Bulgarians had their victories. They caught Basil’s army in a defile in 986 and nearly destroyed it. They even recaptured the old capitals of Preslav and Pliska during the following years. They took Dyrrhachium on the Adriatic coast and gained an outlet to the west. These were the result of skilful use of the formidable peasant army which the Bulgarians could put in the field, and of the sympathy which they enjoyed from the Slavonic inhabitants of many of the regions which they conquered. They were also attributable to Basil’s preoccupation first with rebellion in the east, then with a threatened Russian attack. The former was crushed in the field of battle, the last evaded by diplomatic means, when Prince Vladimir was baptised and received as bride the emperor’s sister Anna. By 990 Basil was ready to deal with Bulgaria. Slowly but surely during the years the Bulgarians were driven back towards the centre of their power in the high mountains of western Macedonia. Basil struck now from this direction, now from that. And he was successful in detaching from their allegiance several of the Bulgarian commanders opposing him. These he rewarded with high office and rich estates. Among them was the eunuch governor of Skopje, Romanus, son of King Peter, the last living descendant of the dynasty of Krum and Boris. He ended his days as a Byzantine patrician in command of the fortress of Abydos on the Dardanelles. These desertions became more and more frequent as Samuel lost control of strong point after strongpoint. Even his own daughter Miroslava went over to Byzantium with her Armenian husband Ashot of Taron.
Side by side with this welcome for high-born deserters went increasing Byzantine brutality towards rank and file prisoners, military or civilian. In 1014 some 15,000 Bulgarian soldiers were captured after a desperate engagement in the upper Struma valley. Basil blinded 99 out of every 100 and left the hundredth man with one eye to guide his companions back to their master. When this ghastly cortege reached Ohrid the shock killed King Samuel. In 1016 during a campaign in Macedonia Basil put out the eyes of every Bulgarian he found, soldier or civilian. Medieval warfare was a bloody business. But Byzantine rulers could take the long view and generally conducted their wars with an eye to the peace which would follow, when their erstwhile foes might become their friends. The ‘frightfulness’ of the last years of the Bulgarian war is an indication that no settlement was hoped for or wanted. Unconditional submission to the empire was to be the only outcome.
After Samuel’s death the Bulgarian Kingdom began to disintegrate as its parts were cut off from one another by Byzantine advance and as various members of the royal family set themselves up as ephemeral local rulers. One after another they surrendered to Basil, to be rewarded with high office and grants of land. The last effective King of Bulgaria, John Vladislav, who proudly entitled himself ‘Emperor of the Bulgarians’, fought desperately until he was murdered by an unknown assailant before the walls of Dyrrachium early in 1018. By later in the same year all was over. Basil received the surrender of Ohrid from the dowager queen Maria, widow of King Samuel, and the last Bulgarian fortress of Pernik, west of Mount Vitosha, was delivered over by its courageous defender, the Bulgarian general Krakra. Isolated pockets of resistance probably held out in the mountains, necessitating operations by Basil in northern Greece, but the Bulgarian state no longer existed. Its territories had become Byzantine provinces, its citizens subjects of the emperor in Constantinople.
Yet this was not a return to the age of Justinian, before the Slav invasions of the Balkans. During the centuries of Bulgarian rule important changes had taken place. The Slavs in peninsular Greece lost their language and their ethnic patterns of behaviour. The period of most rapid Hellenisation was in the ninth century. After that we hear only of relatively small pockets of unhellenised Slavs in inaccessible areas such as Mount Taygetus in the Peloponnese. The same is true a fortiori of the large Slav communities settled in Asia Minor, which entirely lost their ethnic identity and became absorbed into the local population.
This was what had been happening for centuries in the Roman empire and in its Byzantine successor, as tribal communities became absorbed into a common society of Hellenic culture.
In what had been Bulgarian territory, however, this process did not take place. For practical purposes we can take the southern frontier of Symeon’s Bulgaria as the northern frontier of Hellenic speech and national consciousness. Though a Byzantine province from the end of the tenth or early eleventh century until the end of the twelfth, Bulgaria maintained its Slavonic speech, its literary culture based upon that speech, its peculiar stock of folk tales and legend, its poetry and song, and doubtless its dance, its mode of dress, its way of life and its consciousness of its own history. There was much inter penetration of course. Many of the Bulgarian aristocratic families inter married with similar Byzantine families and became virtually Hellenised. Greek was evidently widely known in the towns of Bulgaria during the two centuries of Byzantine rule. But no Byzantine Greek who went to Bulgaria was in any doubt that he was in a foreign land. As the Byzantine military and political power declined catastrophically in the late twelfth century one of the many Bulgarian revolts was successful and Bulgaria was restored to independent statehood. The administration and titulature in this second Bulgarian Kingdom owed much to Byzantine models. But it also perpetuated much that went back to the days of Boris and Symeon, whose heirs its rulers felt themselves to be. It was a land of Slavonic culture, though of course many of its churchmen and other intellectuals were equally at home in Greek. Tŭrnovo, the capital, was the centre of a vigorous school of writers and translators whose medium was Slavonic. Once again an independent church, using Slavonic liturgy and under its own patriarch, was set up. The second Bulgarian Kingdom maintained itself until the end of the fourteenth century, when it vanished as a political entity in the general conquest of the Balkan peninsula by the Ottoman Turks.
What had taken place was the formation of the Bulgarian nationality. Not of the Bulgarian nation, for this is a further development which requires many conditions not existing in the Middle Ages, such as a single economy, a common market, a single, sovereign state, and soon. Nations are largely the product of the modern industrial world. By nationality is meant an important and often long-lasting intermediate stage between the fragile unity of tribe or clan, based on real or imagined kinship and liable to be ruptured by membership of some other community, such as a city, a religious group, etc., and the unity of the modern nation-state.
The formation of a nationality is a complex process and its details vary from case to case. What seem to be necessary—though not always sufficient—conditions are a common language, a common literature employing that language, a common territory of significant extent, a common historical tradition (of which we find a trace in the so-called List of Princes, probably a Slavonic translation of a Greek text, carved on stone at Pliska or Madara ), and no doubt common legends, songs, rituals, games, food habits, patterns of dress, and so on. There is frequently, but not always, a common religion. There is usually, but not always, a common state structure arising out of the pre-existing tribal and clan structures, sometimes under the catalyst of invasion or penetration from outside, but not imposed from outside as an act of policy by another state. It was on such a basis and in such ways that the English or French nationality developed out of the welter of Germanic and Romano Celtic communities in the one case, and out of the semi-tribal Frankish monarchy and the surviving Gallo-Roman communities in the other. Similarly in Bulgaria during the existence of the Bulgarian state a common nationality was evolved out of the semi-pastoral Turkic Bulgar horde, the various tribal principalities and ephemeral unions of the Slavs, the Romanised or Hellenised Thracians and Daco-Moesians and doubtless other elements too. The leaders of the revolt which led to the revival of an independent Bulgarian state at the end of the twelfth century were probably of Vlach origin. Once formed, Bulgarian nationality was immensely tougher and more durable than the pre-existing communities. It is significant that John Vladislav, the last ruler of independent Bulgaria before the Byzantine conquest, emphasises in an inscription that he is ‘Bulgarian by birth.’  It was resistant to Byzantine absorption and later to Turkish conquest, and in the nineteenth century formed the foundation upon which the modern Bulgarian nation-state was built.
The Bulgarian nationality is one of the earliest discernable in medieval Europe, and certainly the earliest within the range of influence of the Byzantine empire. The empire was ordered upon quite different principles. It was certainly not based on ethnic unity, whether real or fictitious. It did not have a common language in the sense of a common mother tongue for its citizens. They included not only Greek speakers, but Slavs, Armenians, Georgians, Abkhazians, Arabs, Syrians, Latins, Turks and many others, who lived intermingled with one another and were not territorially separated. Greek was the language of administration and culture.
Use of it had no ethnic implications at all. Gregory Pakourianos, commander-in-chief of the army in the early years of Alexios I Comnenas, was a Georgian. No doubt he spoke Greek perfectly and transacted all the important business of his life in it. But in the monastery which he founded in 1083 at Bachkovo in former Bulgaria he insisted that Georgian be the language of liturgy and of everyday usage, and that no Greek be admitted as a member of the community.  The number of non-Greeks in the empire increased rapidly in the tenth century, as regions on the eastern frontier, where the mass of the population were Armenians, Kurds or Syrians, were subjected to Byzantine rule. The empire owed its unity—which was very real—to its obedience to a single ruler appointed by God, to its common administrative and legal system, to the ease of movement within it, and to many other actors. It had a common past to look back on, but much more important was the common future to which it looked forward, when God’s design would be fulfilled. The whole structure and functioning of Byzantine society were unfavourable to the development of nationalities in the sense here discussed. It would certainly be wrong to speak of a Greek nationality in the ninth and tenth centuries, though Greek was doubtless the native tongue of the majority of the inhabitants of the empire. But those Greek speakers called themselves either Romans—a concept which included also non Greek speakers— or Cretans, Peloponnesians, Bithynians etc.  In the course of time a Greek nationality did emerge. Perhaps the beginnings of it can be discerned in the twelfth century, and there is no doubt of its existence in the last centuries of the Byzantine empire. But it long post-dated the formation of Bulgarian nationality. And it did not until relatively recent times shake itself free from the ecumenicity and supra-national nuances which it had inherited from Byzantine political thought and practice. How far the existence of Bulgaria on the very doorstep, as it were, of regions with a fairly homogeneous Greek population contributed to the formation of Greek nationality by its example and its polarising effect is an interesting question which cannot be discussed here. What we are concerned with is Byzantine-Bulgarian relations in the ninth and tenth centuries. And we have seen that they are not a mere fortuitous succession of hostility and friendship, but that they embody a recognisable historical process, to which parallels can be found in western Europe at the same time and in other parts of the world in later ages.
4. The Land
Both Bulgaria and the Byzantine empire were agrarian countries, in the sense that by far the largest part of the population of both was directly engaged in the production of food and raw materials through agriculture and stock breeding. This is true of all medieval societies. There were however significant differences between the agrarian processes of the two regions, which were partly the result of permanent geographical differences, and partly determined by the history of the societies which inhabited them.
The territory of the Byzantine empire in the ninth and tenth centuries belonged in the main to the Mediterranean world, characterised by mild, wet winters, long, hot, dry summers, the proximity of sea and mountains, rapid loss of surface water and absence of perennial rivers. Even the plateau of Asia Minor has Mediterranean features, and is sharply distinguished from the much more arid plateau of Iran, which favours nomadism. Anatolian agriculture was based upon the use of rain water and cultivation of fruit trees together with cereals, in the words of Xavier de Planhol ‘une polyculture sèche méditerranéenne’.  It is probable that the warmer, moister conditions of antiquity had favoured the introduction into high Anatolia of crops which can no longer be cultivated there. For instance the extensive olive plantations at Synnada in Phrygia  have long ago disappeared, and may already have vanished by the tenth century.  At any rate the Mediterranean pattern of agriculture had imposed itself throughout the empire so far as natural conditions allowed. This essentially involved extensive olive cultivation, viniculture, winter-sown wheat or barley cultivated often between rows of olive trees, much miscellaneous fruit cultivation, including figs, cherries, plums, almonds, walnuts etc., quick-ripening crops of green vegetables harvested before the summer drought begins, and in particular pulses. Intensive cultivation, without fallow, was practised in a few favoured locations. Fodder crops, such as lucerne, formed a significant proportion of agricultural production. There was little artificial irrigation, other than the digging of wells. Even the terracing of hill sides was probably not much practised in the ninth and tenth centuries.
The history of Byzantine or Bulgarian agriculture remains still to be written. However we do possess two interesting texts which throw light on Byzantine agriculture, the Farmer’s Law, which most but not all authorities now date about 700,
and the Geoponica, an agricultural encyclopedia compiled in the middle of the tenth century but embodying much earlier material. The differences between the pictures which they paint are striking. They are only partly to be accounted for by the likelihood that the Farmer’s Law deals with a commune settled on frontier or abandoned land, like the western mark, and practising extensive subsistence agriculture, while the Geoponica is written for the growing class of large landowners cultivating for the market. They also reflect the steady improvement in the techniques of Byzantine agriculture in the intervening centuries. Of particular note is the much greater variety of crop plants mentioned in the later text, and the frequent reference to the cultivation of leguminous crops to be ploughed in to increase the nitrogen content of the soil. 
On Bulgarian agriculture we are less well informed. Most of the territory of Bulgaria belongs to the world of central Europe, with its extensive plains and wide valleys among the mountains, its heavily wooded mountain-sides which helped to retain the ground water, its heavier soils, its perennial rivers, its less hot, dry summers and its cold, though fairly short winters. It is reasonable to suppose that regions which had for two centuries been outside the Byzantine empire were less affected by the technical improvements just mentioned than those regions which were all the time the object of Byzantine administration. This supposition is borne out by the statement in the Life of Clement of Ohrid that the saint brought to the land of the Bulgarians all kinds of fruit trees from the land of the Greeks in order to improve local cultivation.  We should therefore expect a less varied as well as a different agriculture in Bulgaria. The scanty historical sources reveal wheat and millet as the principal cereal crops, though barley and oats were known.  At the time of the Byzantine conquest of Bulgaria, at the beginning of the eleventh century, the land tax was paid in wheat, millet and wine, and there is no reason to suppose that this was not the practice in the tenth and probably in the ninth century too. Viniculture was general. The Slavs would not be familiar with it in their homeland on the Vistula, but may have come into contact with it on their way southwards through Transylvania. The Bulgars, a nomadic pastoral people, certainly did not cultivate the vine before their arrival in the Balkan peninsula. It seems most likely that the newcomers learned the complex and sophisticated technique from the earlier Romanised or Hellenised inhabitants. This is further evidence for some degree of continuity of settlement and cultivation in the northern Balkans.
The invaders did not find a tabula rasa. Fruit trees are often mentioned, but with no details. The olive does not seem to have been generally cultivated — as indeed it could not be in many parts of Bulgaria for climatic reasons. Where we do hear of it in the period after the Byzantine conquest, it is always in southern Macedonia or on the Black Sea coast, regions which both climatically and culturally had more in common with the southern Balkans than with their immediate hinterland. Flax was produced not only for local use, but also for export to Constantinople in the tenth century.  Among vegetables we hear of beans, cabbage, garlic and melons. Stock-breeding played a greater role in Bulgaria than in the European provinces of the Byzantine empire. In the kitchen middens of Pliska pig bones predominate. At Popina near Silistra, dating from the eleventh and twelfth centuries, those of large cattle are much more numerous. This may be an accident. But Bulgarian archaeologists in recent years have adduced further evidence for the growing importance of large cattle in the Bulgarian economy between the eighth and the twelfth century.  This would indicate a growth of cleared pasture land at the expense of forest, and suggests an increase in the rural population, and a growth in the area brought under cultivation, since much of the pasture would be fallow land. Sheep were bred everywhere in the mountains, and sheepskin rugs from Bulgaria were much appreciated in Constantinople. Horses are rarely mentioned, and horse bones are infrequent at Pliska and other sites. Some scholars argued that the horse was scarcely used, in spite of the suitability of certain regions of Bulgaria for horse-breeding and the presence of imperial stud-farms there in the twelfth century. This view is perhaps exaggerated. After all the Proto Bulgars were horsemen, and in their assimilation by their Slav subjects they are unlikely to have lost all their inherited skill and values. Yet Bulgaria in the ninth and tenth centuries was not a land of horsemen. The Bulgarian army was mainly an infantry force, while the Byzantine army depended upon a core of cavalry partly maintained by land grants. Bee-keeping was extensively practised in Bulgaria, and we hear of honey from the Struma valley being sold in the market at Constantinople in the tenth century. In a world where sugar cane is unknown—except in remote areas where it had been introduced by the Arabs —honey is the main source of sugar. But it may be that bees were kept in Bulgaria partly for their wax, which for certain purposes, such as lamp fuel, replaced the olive oil used in Byzantine territory.
Iron agricultural implements were widely used by the Byzantine peasantry, and scythes, iron axes and the like are mentioned both in the Farmer’s Law and in hagiographical texts. The use of iron in the countryside was probably more widespread than in the contemporary west.  In Bulgaria village smiths were common. One is mentioned in the Miracle of St George,  and John the Exarch includes smiths in his list of artisans. 
Iron plough-shares, sickles and hoes have been found at Pliska and other Bulgarian sites. There is no doubt that much skilled iron-work was carried out in Bulgaria, though the difficulty of dating finds makes it hard to distinguish the period of independence from that of Byzantine occupation. It is likely, however, that the wooden plough was commoner than the iron-shod plough, as it was in Bulgaria at later periods. Recent study has suggested that different types of plough were used in different regions of Bulgaria, perhaps reflecting the different ethnic origins of the population.  The variety of plough patterns is clear; the explanation remains uncertain. They are all essentially scratch-ploughs. The plough was normally drawn by a pair of oxen. Indeed the pair of oxen forms the unit of tax-assessment at the end of Bulgarian independence, and no doubt much earlier too. There is no sign of the type of heavy wheeled plough with a mould-board, which requires a team of four or more oxen. There is little evidence for more sophisticated tools such as scythes.
The ninth and tenth centuries were a period of rapid and decisive transformation in the relations between men and the land in the Byzantine empire. The seventh and early eighth centuries had seen the development of a fairly uniform early Byzantine agrarian society. The great estates of late antiquity had largely vanished during the long years of Persian, Arab, Slav, Avar and Lombard invasions. The pandemic of bubonic plague in the middle of the sixth century may have set off a demographic chain reaction resulting in a marked fall in the population. At any rate there is plenty of evidence for the abandonment of cultivable land. And large numbers of ‘barbarians ’ were settled, mostly in Asia Minor, as cultivators and sometimes as soldiers, presumably on land which had been neglected.
At the same time the loss of the corn supply from Egypt, first temporarily interrupted in 608, then again in 619 after Egypt fell to the Persians, and finally terminated in 641, led first to a chronic shortage of cereals in Constantinople and in army establishments, and then to the development of cereal production in regions near the capital where more lucrative cash crops had been produced before, or which had lain uncultivated.
The great estate, with its slaves and coloni, vanished, and its place was taken by communes of free peasants—free in the sense that they owed no dues or services except to the central government. In such a village commune the cultivated land was individually owned, the often extensive wasteland communally owned, and user rights regularly redistributed. There were various possible vestiges of an earlier communal ownership of the cultivated land, such as the right to pick fruit to eat from another man’s trees, but in general this was a society of individual landowners. What made it a commune was not so much its internal organisation as its relation to the state. For the members were jointly responsible for the payment of taxes to the imperial government. As a consequence they had a prior right of purchase should a fellow-member wish to sell his holding, and the right to occupy and cultivate land which for any reason was left uncultivated. This is the agrarian society revealed by the Farmer’s Law. Its resemblances to the western mark are striking, but its differences are also worthy of note.
Side by side with these peasant communes there were probably larger holdings of a different character, held by soldiers. Unfortunately the origin of these stratiōtika ktēmata is a matter of some obscurity. It seems most likely that it is connected with the origin of the theme system, whereby the ancient provinces were replaced as administrative units by military areas, each with its own army formation permanently based in its territory. But the matter cannot be proved. And the origin and development of the theme system is the subject of debate, which need not concern us here. The stratiōtika ktēmata are first encountered unequivocally in the legislation of the Macedonian emperors of the tenth century. They certainly existed much earlier. There is a clear reference to a military holding in the Life of St Philaretos. It seems most likely that during the seventh century, and probably as early as Heraclius, the custom grew up of distributing land holdings conditional upon military service. These were heritable, and their owners enjoyed immunity from certain taxes and all corvées due to the state. Their obligation was to provide a soldier, properly equipped. He might or might not be the owner of the holding ; often he was the owner’s son. The holding of a cavalryman, for example, would be considerable; it had to provide two horses, and no doubt a groom, as well as the soldier himself. Such land-holders were, in terms of a later social system, small gentry, not peasants.
Infantry holdings, of which we hear much less in the sources, would be substantially smaller, but still bigger than the average holding of a member of a peasant commune.
By the second half of the eighth century, then, the Byzantine empire consisted largely of a mosaic of free peasant communes and military estates. There were some surviving larger estates, worked by tenants, slaves or wage-labourers, and of course there were extensive monastic estates, though their extent had been reduced by the Iconoclast emperors. Early in the ninth century a process begins which does not reach its climax until the eleventh. Successful military leaders and civil functionaries invest their gains in land, establish large estates, and gradually oust the peasant communes and the military proprietors. The factors favouring this were several: the relatively advanced state of Byzantine agriculture, which permitted large profits on the investment of capital, the ease with which individual holdings could be established by enterprising—and lucky—peasants on former wasteland or land not belonging to a commune, the relative freedom from invasion and war in most regions of the empire, the absence of any other possibility of placing capital. The process in the end went far beyond questions of land-tenure and led to the development of a hereditary feudal aristocracy, creaming off the surplus production of the primary producers and eventually taking over the central state apparatus. In this context ‘feudalism’ does not imply a complex system of mutual obligations between superior and inferior, as in the classical feudalism of parts of western Europe. It is used in a much more general sense of a society in which the land belongs to a class of magnates and they extract part of the farmer’s surplus production from him in the form of rents in money, kind and services. The legal form in which this basic economic relation manifests itself depends upon the structure and history of the society in question, the strength or weakness—or even total absence—of a central government, and so on.
In the sphere which we are immediately considering, we find a concentration of land in the hands of certain wealthy families and monasteries acquired in a variety of ways, legal and illegal. The new landowner collects the taxes due to the state on these lands, but does not necessarily pay them over in their entirety. In course of time he acquires in various ways fiscal immunity, and also steps up the amount extracted from his tenants. In addition he imposes on them various kinds of corvées and sometimes exercises minor judicial rights over them.
Though, owing to the strength of the central state apparatus, the kind of elaborate hierarchical structure characteristic of the feudal West never developed in Byzantium, those historians are right who describe the pattern evolving in the ninth and tenth centuries as Byzantine feudalism. But the dependent peasantry of the later Byzantine period are not the direct continuation of the coloni of the later Roman empire. There was a long intervening period when free peasant proprietors prevailed. 
The successive stages of ‘Byzantine feudalism’ are difficult to trace. No doubt their rapidity and their form varied much from region to region. Yet on the main line of the process there is no longer any dispute. In the European provinces there are probably three zones to be distinguished: Eastern Thrace, the hinterland of the capital: Western Thrace, Macedonia and Epirus; and peninsular Greece. The first was a zone of relatively intensive cultivation, in which peasant communes, if they ever existed, were early ousted by civil and military officials, whose quite small estates— proasteiá—formed a tight mosaic. These landowners were, at this stage, generally able to protect themselves from the rapacity of the nascent feudal magnates. The second was an area only gradually brought under firm Byzantine control after being overrun by the Slavs in the sixth and early seventh centuries, and largely lost once again to the Bulgars in the reign of Krum. It was essentially a region of free peasant proprietors, with strong traces of an earlier tribal organisation. The third, always something of a backwater in Byzantine life, offered a mosaic of partly or wholly Hellenised Slav communes; surviving medium estates of the ancient type, monastic estates, and large feudal estates of the new type, particularly on land recently regained from the Slavs. We need not believe everything which the Continuator of Theophanes writes about the widow Danielis (Theoph. Cont. 226 ff., 316 ff.). But she is clearly a type-specimen of the new variety of landowner, except that she probably made more use of slave labour than would have been possible, say, in Asia Minor. By the middle of the eleventh century the Theban tax-roll —whatever the precise status of the document published by Svoronos—shows a large part of central Greece in the hands of absentee landlords, many of whom hold posts in the provincial administration.  The situation in Asia Minor at the end of the eighth century is depicted in the Life of St Philaretos;  Philaretos was probably an enterprising peasant — a kulak — rather than a dignitary.
The problem of the concentration of property in the hands of the dynatoi attracted the attention of the Macedonian emperors, who saw in it a threat both to the revenues of the state and to its military power. In a series of legislative acts they, as well as the usurper Romanos Lekapenos, sought to combat the acquisition of village commune lands and military holdings by grandees, whether by purchase or by fictitious adoption, testamentary disposition or other means. Their measures seem to have been of little avail. In the first place Leo VI may have already revoked the right of pre-emption in a Novel,  if indeed it is genuine. Secondly, the execution of these measures was in the hands of the very class who stood to profit most by the acquisition of peasant holdings, the high military and civil officials. 
We are less well informed on Bulgarian agrarian relations. Some probable hypotheses can be formed on the basis of the Responsa of Pope Nicholas I to Boris and of the Zakon Sudnyj Ljud’ĭm. The Proto-Bulgars did not at first cultivate the land, but they did practise stockbreeding. Whether flocks were individually owned, or belonged to clans or tribes, we do not know. But it is likely that the great Bulgarian officers of state—the boyars—who were the descendants of the former tribal nobility, possessed large flocks with grazing rights over great areas. However they readily took to agriculture of the feudal type, though probably without the technical skill and scientific interest which many Byzantine landlords displayed. By the early ninth century they had ceased to be primarily pastoralists.
The main agricultural labour force was provided by the Slavs. Originally organised in communes based on kinship, with a fairly egalitarian system, they were early forced to pay a part of their surplus to the Bulgar state, which also transported and settled them, in tribal units, in new areas. Over and above the tribute which they paid to the state, many of them were quasi-feudal dependants of members of the nobility. The stages by which they reached this state of dependence are not recorded, but can well be imagined: the tribal leaders become owners of the tribal land, and then become themselves fused with the Proto-Bulgar aristocracy. Ninth-and tenth-century sources show peasants falling into groups: free peasants, epoikoi, Slavonic epigi, owing dues to the state; peasants dependent on a lord, paroikoi, Slavonic paritsi; a third class, in some way less free than the paritsi were the otrotsi, whose status resembled that of a slave. At the lowest level of the nobility we find the Kmet, probably in origin the head of a free commune, now turned into an instrument of state or landlord.
The Zhupan, originally a Slavonic tribal leader, had become absorbed into the Bulgar ruling class and probably retained only a tenuous connection with the kinship community which was his original raison d'etre.
The Zakon Sudnyj Ljud’ĭm ( the legal code promulgated under Boris or Symeon) rules that if a ljudin flees from his knez to another knez he is to be beaten and returned. This has been taken as an indication that some Bulgarian peasants were tied to the soil as early as the end of the ninth century. But knezhenie is a territorial division in Old Slavonic texts, and the knez may be some kind of regional governor. The prohibition may refer to soldiers leaving the military district in which they are enrolled. The Bulgarian archbishopric at the beginning of the eleventh century possessed estates comprising not only land but a fixed number of cultivators. Its ownership of these estates, together with the men on them, was recognised by Basil II after the destruction of the Bulgarian state.  These cultivators appear to have been attached to the soil, though when or how they became so is uncertain. About the same date we meet a certain Glad, governor of a region between the Danube and the Maros in south-eastern Hungary. He had been sent from Vidin by the Bulgarian king as governor —it is not clear by what king —and left undisturbed by King Stephen I of Hungary. He and his son Octhum had ‘a countless multitude of unbroken horses’ as well as 'pecora infinita’ landed estates (allodia) and manors (curiae), and the right to collect taxes on salt coming down the Maros. But these two picturesque characters are probably not typical Bulgarian aristocrats. One of them had seven wives, 'quod in religione Christiana non perfectus erat'. 
At first sight the situation in Bulgaria is similar to that in contemporary Byzantium. But there are differences. First, the formation of large estates begins later and advances more slowly in Bulgaria. There were certainly many free peasants still on the land when Basil II put an end to Bulgarian independence. Second, there seem to have been no military holdings in Bulgaria. The Bulgarian army was raised by a kind of levée en masse carried out by the local authorities, and was less of a professional élite than the army of Byzantium. Third, the effect of the destruction of free communes was probably more drastic in Bulgarian society. Men weeded out by one process or another from the wasteful subsistence agriculture could not so readily find other employment, thanks to the backwardness of Bulgarian industry, and they easily fell into debt bondage. The Zakon Sudnyj Ljud’ĭm constantly speaks of paupers, thieves and bandits. This is evidence of the impoverishment of the peasantry as men lost their original connection with their commune and their land.
There is no evidence for any attempt by the Bulgarian rulers to restrain the concentration of land in the hands of the boyars.
The same forces, then, were at work in Byzantium and Bulgaria, but they operated at different speeds, and their starting points were not the same.
The southern part of the Balkan peninsula was a region of ancient city life. In the northern part the cities were mostly established during the period of the Roman empire, though there were a number of earlier foundations along the west coast of the Black Sea. However, by the end of antiquity there was little difference between the cities of Greece and those of Illyricum, Thrace and Moesia, except that the former were more numerous and comprised practically the whole territory of the provinces, while the latter were fewer in number and were separated by imperial estates, tribal lands and other non-city territories.
The ancient city was in principle internally self-administering. Hellenistic or Roman governors were concerned with maintaining peace and if necessary arbitrating between cities. Roman procurators were concerned with fiscal relations between the imperial government and individual citizens or the civic collectivity. Otherwise the city ran its own affairs by means of an assembly of the citizens—which in most instances soon lost its legislative power, though it might long continue to function as an electoral body—a council, and magistrates. In practice this state of affairs was often modified. As early as the second century a. d. imperial officials took charge of the internal affairs of cities, first as an emergency measure, later on a more permanent basis. But the power of the curiales, those families whose wealth entitled them to membership of the council, remained real enough until the fourth century, especially in the larger cities. And in these there emerged within the curiales a smaller group—the principales—who formed a self-perpetuating oligarchy of local notables in whose hands lay whatever remained of the city’s autonomy, plus a number of fiscal functions delegated to the city authorities by the imperial government. It was not until the fifth century that the system of self-administration began to breakdown, being taken over in part by officials appointed by the central government, in part by the bishop and his clergy. The process of decline is admirably summarised by Justinian in the Proem to his Novel 38 (536 a. d.). By this time, and probably already a generation earlier, the city councils had ceased to exist, their functions being taken over by an imperial official, the vindex. But the ordo curialis remained both as a legal entity and as a social group, contriving to share power with the state and the church. They are found effectively running cities under various names—protiktores is one—as late as the end of the sixth century.
It is not until the reign of Leo VI (886-912) that city councils were legally abolished, since civil affairs have been transformed, and everything now depends on imperial providence and administration’ (Leo VI, Novel 46). 
This ruling oligarchy was made up of the owners of substantial estates within the city territory. For the cities of antiquity were not essentially centres of industry and trade, but agrarian centres, ‘Landstädte’. Each lived mainly from the produce of its own territory. Its governing class consisted of landlords, not of merchants or industrial ists. The city provided facilities for the distribution of agricultural produce and offered to the landlords and their families and dependants a style of life impossible in the surrounding countryside, symbolised by the baths and the theatre which, along with temples and other public buildings, were to be found in the humblest of cities. A few cities, Athens, Corinth and Syracuse in the fifth century b.c., Alexandria, Antioch and Thessalonika in Hellenistic times, developed further into centres of manufacture and long-distance trade, less dependent on their immediate surroundings than the majority of cities. But these great cities were always few in number, and when circumstances became unfavourable they regressed to the status of agrarian cities.
Such were the cities of antiquity. The Synecdemus of Hierocles, a sixth century document, lists 265 cities in the Balkan provinces and their offshore islands. Most of these were very small places indeed, but they had the attributes of a city, not a village. And most of them were in the southern part of the peninsula: Macedonia, Thessaly, Achaea, Epirus and Crete between them account for 179 of the 265.
In the series of invasions from the late fourth century onwards, many of the cities of the Balkan provinces were captured and sacked. Some were rebuilt and resettled at once, others left as a heap of ruins. By the mid sixth century peace appeared to have been long re-established in the Balkans. Many cities repaired their walls and there was much public building, but it was the result of imperial munificence, not of local initiative. Some cities had already ceased to exist. Most, in spite of their outward façade, had lost many of their wealthiest and most influential citizens, particularly in the regions near the Danube frontier. But the cities were still there, living off the peasants and slaves who cultivated their land. Few of them now had theatres, but they all had churches of greater or lesser magnificence. And they represented the ideal of civilised life. The inroads of the Slavs and Avars put an end to city life in the northern Balkans except along the sea coast.
Cities were systematically destroyed and their inhabitants either fled or were taken prisoner or killed. The Avars had some interest in the use of city amenities but the Slavonic peasants had no use for cities, which were for them mere encumbrances upon the soil.
The narrative historians tell us very little about the fate of the cities in the Dark Age—approximately 600-800 A.D. Here and there the archaeologist now begins to shed a little light on the problem, but his evidence is often extremely difficult to interpret. However, many of the great coastal cities of Greece present no problem: Thessalonika, Chalkis, Athens and Corinth continued not only to be inhabited, but to function as cities throughout the Dark Age, though they might be again and again besieged, and sometimes temporarily captured by the invaders. The same is true of Thebes, and very likely of other cities in east central Greece. Patras was in all probability also in continuous occupation, though there is a tantalising story about the evacuation of its inhabitants to southern Italy. Many of the smaller inland cities in Hierocles’ list were probably abandoned by their inhabitants and their walls allowed to fall into ruin. But others may well have continued as Greek islands in a Slavonic sea. Further north-east Constantinople remained the imperial capital throughout, though on occasion besieged by Persians, Avars and Arabs. The coastal cities of the Sea of Marmara and of the western shore of the Black Sea held out too, though some of them were temporarily captured by the Avars. The great fortress-cities protecting Thrace, Develtos, Adrianople and Philippopolis also remained in continuous Byzantine occupation, as did Serdica (with only a brief interruption) right up to its capture by the Bulgarian Krum in 809. Of the cities of Moesia, western Thrace and Macedonia we know little, and all the evidence is against continuity of occupation. The fact that some of them reappear as cities in the later Middle Ages is to be explained partly by the convenience of their situation, and most of all by the supply of building material which they provided ready to hand. We must not imagine the city notables and the bishop fleeing before an advancing horde of Slavs and Avars. Most of them had already left, when the security which permitted them to exploit the city lands could no longer be guaranteed. The cities of the north ceased to function as cities before they were captured; sometimes the Slavs even found the city sites abandoned.
The Slavs and the Bulgars in time, as their tribal structure broke down and as they became drawn into the network of long-distance trade which linked Constantinople with northern and western Europe, established their own urban settlements.
These were often in the vicinity of a Greco-Roman city for convenience in the supply of building material, but rarely on the same site. Thus Serdica was not destroyed, but was replaced by a Slav settlement a short distance away, while the old centre was not reoccupied until much later. Slavonic/Bulgar Dristŭr was built near ancient Durostorum (which may never have been sacked). As we shall see, these new cities were of a different type, fulfilling a different function in society. 
If the ancient city was the centre of an agrarian community, enjoying internal self-government and providing the essentials of the good life for its citizens, or at any rate for some of them, the early medieval city in the Byzantine empire was often primarily of military significance — a kastron rather than a polis. Small in size, militarily defensible, it housed a garrison and its dependants in time of peace, and served as a refuge for the surrounding population in time of war. It did not aim to provide the amenities of the ancient city—though it might well have had baths. And its affairs were controlled by the garrison commander, not by magistrates chosen from among the citizens. Such a kastron might occupy the site of an ancient city, if it had a steep acropolis or other easily defensible feature, as at Philippi, Amphipolis (renamed Chrysopolis), Neapolis (renamed Christopolis, now Kavalla), Demetrias, Lamia, Corinth, Argos, Naupaktos. In some cities a much lower and less fortress-like acropolis had to suffice, faute de mieux, as at Thebes, Athens, Chalkis, Adrianople. Sometimes only a part of the area within the oldcity walls was fortified, as at Theveste in Africa, where one corner of the old city was surrounded by a wall and ditch, or at Athens. Sometimes the kastron occupied, not the site of the classical city but a nearby hill top or projecting cliff, often the very spot from which the earlier inhabitants had descended to establish their city in the plain. Such was the case at Chonai in Phrygia, successor to Kolossai, and at a number of sites round the margin of the Thessalian plain and in Arcadia. This often involved a change of name.  Some kastra bear no relation to classical cities, but were built in response to Byzantine military problems. The best example is Monemvasia; others are Kastron Maina, Ioannina, and the old fort at Corfu. Many of the Dalmatian coastal cities are of this kind, for example Kotor, Dubrovnik, Split, Trogir. On the Black Sea coast, on the other hand, the major cities founded in antiquity remained firmly in Byzantine hands, serving as military bases and entrepôts, until the Bulgarian expansion under Krum early in the ninth century.
Only a few great cities retained the physical pattern of the ancient city, with a long perimeter-wall fully maintained. In the first place the capital, which was in any case exceptional. Then Thessalonika, the second city of the empire, which had a heavily fortified citadel within its perimeter-wall; Naples, which was similarly structured, with a citadel built by Belisarios within the ancient city walls; Nicaea, and a few other places.
In all these cities of the new type there was some population other than soldiers and their direct dependants. Peasants cultivated the fields within easy reach of the shelter of the walls. They no longer paid taxes through the city authorities, as members of a civic community, but directly to an imperial official, as individuals or as a village commune. In addition artisans, traders and others might gather round the fortress walls. Thus in favourable circumstances we find the medieval pattern of inner and outer city developing. In the heavily fortified inner city lived the ruling classes—army commanders, landowners who had migrated from the countryside, and so on; the outer city, lightly fortified if at all, was the dwelling place of craftsmen, merchants, and the peasants who cultivated the fields nearby. The inner city served as a refuge for the inhabitants of the outer city in time of war. The two formed a single social and economic complex.
In due course a new patriciate began to emerge in some of these double cities, especially those prominent in trade. Whether they are original inhabitants of the kastron, landed proprietors from the surrounding country, or successful citizens of the outer city is not clear. Thus we hear of prouchontes in Sparta in 998, of archontes in Athens in the twelfth century. A city nobility is found choosing magistrates at Edessa in 1098, and at Dubrovnik in 1192. There was a local nobility and local magistrates at Cherson, whose defection from allegiance to the imperial governor is envisaged by Constantine VII. It is unlikely that in any of these cases we have a survival of the civic institutions of late antiquity. They are rather the result of social differentiation within early medieval cities, in some cases combined with the relative weakness of the central government. This process was no doubt under way in the ninth and tenth centuries, though we do not generally see its results until much later.
Such was the general situation in the ninth and tenth centuries in the Balkan peninsula and nearby. It must be remembered that until the Seljuk conquests at the end of the eleventh century the main economic and demographic weight of the Byzantine empire lay in Asia Minor, and that, the capital and Thessalonika apart, the European provinces were relatively backward and neglected, with fewer and smaller cities.
It is significant that in the first half of the tenth century there were about 371 bishoprics in Asia Minor as against 99 in mainland Europe, 18 in the Aegean islands, and 16 in Byzantine south Italy.  The cities of Asia Minor were involved not only in production and commercial exchange with one another, but also in long-distance trade with Russia and the Crimea, the Caucasus, central Asia, China, the Caliphate and India. 
It is time to examine more closely the differences between the Byzantine provinces and Bulgaria. First, cities were far more numerous in Byzantine territory. Not all the civic communities listed by Hierocles survived, but a surprising number did, and one or two new ones had been added. Apart from the special cases of Constantinople and Thessalonika, there were flourishing cities on the coasts of the Black Sea, the Sea of Marmara and the Aegean, and in peninsular Greece. Inland Thrace seems to have been fairly thinly populated, and to have had few cities other than the chain of frontier fortresses. Macedonia was largely outside Byzantine control. It is sometimes very difficult to distinguish between genuine medieval cities and mere fortresses—episcopal lists are not always as helpful as they should be, However, the following is a list of cities in the Byzantine provincesofthe Balkans which are described by twelfth-century sources (generally Arab geographers) as ‘populous’, ‘wealthy’, ‘prosperous’, ‘megalopolis’ and similar terms: Corinth, Athens, Thebes, Larissa, Kitros, Ioannina, Kastoria, Thessalonika, Serres, Zichna, Philippi, Melnik, Rodosto, Nauplia, Mossynopolis. Demotika, Adrianople, Philippopolis, Sparta, Levadeia, Demetrias, Armyros, Karystos, Christopolis (Kavalla), Drama, Selymbria, Herakleia, Kallipolis, Panados.  Not all these cities were flourishing two centuries earlier. The building record in Athens, for instance, suggests that the eleventh and twelfth centuries were a period of enrichment and expansion. Yet many clearly were cities in the medieval sense as early as the ninth century. And the list given above is not exhaustive.
In Bulgaria the picture is very different. The coastal cities of Mesembria, Develtos and Anchialos were captured from the Byzantines by Krum in the second decade of the ninth century, and later returned to the Byzantines. There is no reason to suppose that their predominantly Greek population withdrew entirely, or that they lost their importance as entrepots. Odessos (=Varna) had fallen to the Bulgarians earlier, we do not know when.
Serdica was captured by Krum in 809: a place of some importance in late antiquity and the seat of an archbishop, it had its water supply and aqueducts restored by Tiberius II at the end of the sixth century. But it seems to have dwindled in size and importance and was little more than a garrison town at the beginning of the ninth century. The Byzantine soldiers were no doubt taken prisoner or sent home. But the Bulgarians themselves do not seem to have settled in the city at once.
There were the two successive Bulgarian royal capitals of Pliska and Preslav, both in north-eastern Bulgaria, in the region where the Bulgar state was first established on former Byzantine territory. Pliska has generally been identified with the extensive ruins at Aboba, though there are still scholars who argue against the identification.  Accepting provisionally that Aboba is indeed Pliska, two points must be made at the outset: (i) The site was that of a late Roman city, though which it was we do not know; (ii) The excavations of Škorpil and Uspensky in 1899-1900 were not carried out in such a way as to enable relative dates, let alone absolute dates, to be established with any certainty.
It is therefore difficult to distinguish between what the Bulgars found there and what they may have added later. There is an inner city, enclosed by a rectangular stone wall, of about 0.5 square kilometres. Surrounding this is a vast earth wall, in places still 10 metres high, enclosing an area generally rectangular in pattern of 23 square kilometres, most of which does not seem to have been built over. Neither the stone inner wall nor the earthen outer wall has been dated, so any hypothesis concerning their relation is possible. In the inner city are baths, large buildings heated by hypocausts etc., which may well belong to the original Roman city, as well as several monumental buildings of massive construction thought to have been palaces. Much of the space within the stone wall is not built over. Beneath the massive stone buildings archaeologists have traced the outlines of an immense building of square or rectangular form, apparently mainly constructed of wood. The outer city contains a number of churches, including a large Basilica 99 by 29.5 metres, with three aisles and a single apse. This has been taken to be a court church built by Boris immediately after his conversion. But it remains to be shown conclusively that it is not a church of late antiquity. There are also a number of substantial dwellings, a large building used as a pottery workshop, many small, square, half-subterranean dwelling houses, and several burial mounds, including one with a cremation, burial and remains of horses, doubtless the tomb of a pagan Khan or tribal leader.
A number of inscriptions in Greek were found in the city, recording victories, deaths of boyars, treaties, construction of buildings, inventories of military stores, etc.
The simplest set of hypotheses to explain all this is that the Bulgar Khan, established his fixed residence on the site of a small late Roman city, which supplied convenient building material, and surrounded it with an immense earthwork rampart within which he, his men and his flocks could shelter in time of danger. His great throne room, not unlike the banqueting hall of Attila described by the historian Priscus, was of wood, as were the dwellings of his boyars and their retinues. Some of the installations of the late Roman city, such as the baths, may have been put into working order by captive or fugitive Byzantine craftsmen. These wooden buildings were destroyed by fire when Nicephorus captured and burned Pliska in 811.  The capital was rebuilt partly in stone by Krum and Omurtag and perhaps above all by Malamir, who was a great builder to judge by his inscriptions. The Bulgarians by this time had many Byzantine artisans in their service from the newly-conquered territories. Masonry from the pre-existing late Roman city was used, and probably from other sites as well, and decorative tiles may have been brought from all over Moesia, particularly from the ruined Danube forts. The architects who constructed these rather heavy, massive buildings were not master-craftsmen from Constantinople but builders from provincial cities now in Bulgar territory, and the models they followed were surviving late Roman buildings in the northern Balkans rather than the latest fashions in metropolitan architecture.
Preslav, some distance south-west of Pliska, had an outer wall of stone, following the contours of the irregular terrain, and enclosing an area of 3.5 square kilometres, and a rectangular inner wall, also of stone, enclosing an area of about 0.25 square kilometres. It too was built on the site of a small late Roman city, but the orientation and outline of its buildings is independent of any earlier construction. Fortunately it was not excavated with the techniques of the late nineteenth century; indeed relatively little excavation has yet been done there. The excavations at present being carried out by the Bulgarian Academy of Science, under Professor S. Stanchev and others, should in due course provide solutions to many problems, and give us much information on life in ninth- and tenth-century Bulgaria. In the outer city is a square building with a stone pylon, later converted by the addition of an apse.
It has been suggested that this is a pagan cult site adapted to Christian use. There is also a well-preserved round church with a rectangular atrium containing a font. Its sophisticated construction, the finely-carved capitals, the niches with ceramic plaques and other features make it unique in old Bulgarian architecture. Opinion is divided as to when and by whom it was built, some holding it to be a late Roman church. This and other problems connected with the round church will be solved only by further excavation at Preslav. Six or seven probable monasteries have been wholly or partially excavated, including one with a wooden church on stone foundations. Unlike Pliska, Preslav has much decorative sculpture, evidently carried out by local craftsmen and incorporating animal-motifs unusual in Byzantine art. Within the outer wall are remains of smithies, and goldsmith’s and jewellers’ workshops. It was long believed, on the basis of an inscription, that Khan Omurtag built a palace at Preslav, but this view is now generally rejected.  It did not become the royal capital until the beginning of the reign of Symeon in 893, and most surviving structures, apart from the possible converted pagan site and the enigmatic round church, date from Symeon’s time or later.
Pliska and Preslav are the two most prominent Bulgarian cities. Both are in origin and function quite unlike medieval Byzantine cities. Pliska was essentially a military camp which only gradually took on some of the aspects of a city, as artisans settled there—whether willingly or not we can scarcely guess —to serve the needs of army and court. After the conversion of the country to Christianity churches were built, doubtless on the initiative of Boris or some of his boyars. It does not seem to have been a significant centre of trade. The principal industries were those connected with the army and the court — iron-working and fine metal work—and those who carried them on are likely to have been Greeks from the cities captured by the Bulgarians. As well as a few Greek inscriptions there are many Slavonic inscriptions and graffiti in the Cyrillic alphabet and one brief inscription in Glagolitic.
John the Exarch in the Preface to his Shestodnev describes the magnificence of the Palace at Preslav.  And indeed art and technology are far more developed at Preslav than at Pliska. Archaeological evidence and that of early Bulgarian literature point to varied and developed industry, particularly metallurgy, ceramics and glass, and textiles.
But Preslav was none the less a royal foundation and a royal residence, and the activities of most of its citizens were connected directly or indirectly with the court and the church. It does not seem to have been a centre of long-distance trade. Many of its craftsmen were in some kind of dependence upon the king or boyars. It was in a sense an artificial creation, rendered possible and desirable by the concentration in the hands of the ruler and his boyars of much of the surplus produced by the Bulgarian peasants. Though an administrative and military centre as was Constantinople, Preslav did not play the major role which Constantinople played in industry and trade. The absence of coined money restricted internal trade between city and countryside. We must not imagine at Preslav the complex and sophisticated control of supply and prices of essentials and key luxuries for which the Book of the Prefect is evidence at Constantinople.
So much for the two principal cities of Bulgaria, which had an unusual character of their own. There is less to be said concerning the other cities of ninth- and tenth-century Bulgaria. The Black Sea cities, Mesembria, Develtos and Anchialos, although captured by Krum, were only intermittently in Bulgarian hands. The Thirty Years Peace signed between Leo V and Khan Omurtag gave these former naval and military bases to the Byzantines,  and in Byzantine hands they remained until Symeon took them at the end of the ninth century. After Symeon’s death they appear to have been returned to the Byzantines. Though there was some church building at Mesembria in the tenth century, and it served on occasion as a meeting point for Bulgarian and Byzantine diplomats, it seems to have lost much of its earlier importance. The series of seals of Kommerkiarioi of Mesembria in the seventh and eighth centuries does not continue. Trade between Bulgaria and the empire seems to have been mainly routed through Constantinople or Thessalonika. Odessos (Varna) may have remained for some time in Byzantine hands after the foundation of the Bulgarian state, but by the middle of the eighth century it was a Bulgarian possession. It seems to have been a place of little account, in spite of its excellent harbour. The same is true of Dionysiopolis (Balčik) to the north. The Bulgarians never built a fleet and were not interested in naval bases.
Of the Danube cities Singidunum (Belgrade) may have been considerable —it was the seat of a bishop in the ninth century —but as a frontier post it was mainly of military importance. Bononia (Vidin) had its fortifications repaired, probably after the loss to Bulgaria of the territories north of the Danube in the reign of Symeon.
Destroyed by the Magyars, it was rebuilt by Symeon. Like Singidunum it was primarily a military base. Červen was also the seat of a bishop and appears to have been the principal Bulgarian city on the lower Danube, overshadowing the Slavo-Bulgar settlement near the former Roman town of Sexanta Prista, out of which the city of Ruse later grew. Silistra, a major city in late Roman times, whose fourth-century bishop Auxentius was a pupil of Ulfila and a writer, declined during the sixth-century invasions. Its last recorded bishop, Dulcissimus, fled before the Avars to the safety of Odessos. Neglected for a time, the fortifications were apparently repaired by the Bulgarians at the end of the seventh century. The city held out against the Magyars in the reign of Symeon, but was captured by Svyatoslav in 969. Several Proto-Bulgarian inscriptions testify to its importance. 
Cities in the interior of Bui aria were few and of little importance. In the far west Ohrid owed its importance to its choice by Clement as his missionary centre. Church building began in the lifetime of Clement, and by the end of the tenth century Ohrid must have had a considerable population. The great fortress which dominates the lower town was built by King Samuel at the end of the tenth century. We know virtually nothing of the economic and social organisation of the city. Skopje is scarcely mentioned until the final battles between Basil II and King Samuel of Bulgaria. It was then a fortress of great strategic importance, whose governor and commander of troops was no less a person than Roman-Symeon, son of King Peter and Maria Lekapena. He eventually betrayed the stronghold to the Byzantines. There is no evidence that Skopje was a significant centre of trade or industry. Serdica was another matter. There a Slavonic settlement grew up beside the late Roman fortress, which was long abandoned after its capture by Krum in 809. It must have been a considerable place, as it became the seat of a bishop in 864, when a church hierarchy was established in Bulgaria. Later it was able to hold out against Basil II himself, and was not finally captured until 1016, after a siege which lasted 88 days. Through these bare facts we dimly see a city of some size, both a fortress and an industrial, commercial and ecclesiastical centre. Philippopolis was also the centre of a bishopric. We know very little of its size or importance in the ninth and tenth centuries, and it would be unjustified to suppose that the large and flourishing city described by Idrisi and the historians of the Crusades in the twelfth century existed two and a half centuries earlier.
The original inhabitants were taken prisoner when the city was captured by Malamir, and there is no evidence of their ever returning. Finally little Preslav’, Preslavets, situated somewhere in the Danube delta, probably near Černavoda, and possibly the site of the Bulgar court before Asparuch entered Roman territory. When Svyatoslav captured it in 968 he was overwhelmed by its riches, reports the Russian Primary Chronicle. The passage is worth quoting: — Svyatoslav said to his mother and his boyars:
It is not agreeable to me to dwell in Kiev. I wish to live in Preslavets on the Danube, for there is the centre of my lands, for thither all good things converge—from the Greeks gold and textiles, wines and a variety of fruits, from the Czechs and the Hungarians silver and horses, from Russia hides and wax and honey and slaves.
Preslavets was evidently a centre of international trade between Byzantium, Russia and Central Europe, to which merchants from many lands resorted. But in the absence of other evidence it would be unwise to credit Preslavets with a large settled population on the basis of a single anecdote.
In our period, then, the European provinces of the Byzantine empire, like the rest of the empire, were a land of cities, most of which were continuations of ancient cities, though their structure and function had often been considerably modified. Bulgaria was a land from which most ancient cities had vanished, and where the largest and most flourishing urban communities were those gathered round a royal court—military-administrative cities. This, however, is too static a view of the matter. The relative stability which Bulgarian rule gave to a region where for two centuries no state had been able to exercise effective power favoured urban development. Archaeological evidence from medieval Bulgaria is usually difficult to interpret. But it would be abundantly clear, even without John the Exarch’s account, that Preslav was far more of a city than Pliska. And the emergence in the period of Byzantine rule in Bulgaria in the eleventh and twelfth centuries of a number of substantial cities, described in glowing terms by Idrisi or by the Crusader historians, is evidence for the formation in the ninth and tenth centuries of urban nuclei where before there had only been settlements of Slav peasants in the shadow of the crumbling ruins of a late Roman city or fortress.
We cannot follow this process in detail, but there can be no doubt of its reality. The heritage of classical urban life had been more definitively broken in Bulgaria than in Thrace and peninsular Greece, to say nothing of Asia Minor. The new cities were less linked geographically and historically with those of late antiquity than was the case in Byzantium, and they began their development in a centralised territorial state which had no ancient tradition of civic autonomy. In this way the Bulgarian cities differed from those of the Byzantine empire, though both belonged to a common genus. The massive fortifications still surviving not only at Pliska and Preslav, but at Vidin, Ohrid and elsewhere, underline the military character of the early Bulgarian cities, with their heavily defended inner citadels, surrounded by or adjacent to an outer city inhabited by farmers and artisans. These latter primarily served the needs of the feudal upper classes of the inner city, and were in the period we are studying only beginning to develop an economic life of their own.
6. Industry and Trade
The Byzantine empire was in general an area of advanced technology, with a stable monetary system, a large internal market in medieval terms, and extensive trade with distant lands. The eastern provinces of the Roman empire had been the most industrially developed. In spite of the loss of Syria, Palestine and Egypt in the seventh century this superior technology was maintained, and its level even raised - by the new contacts with the Fertile Crescent, the Iranian world, and regions further east which developed through the spread of Islam.
One must beware in this connection of being misled by modern analogies. Industrial society of today makes most of its ultimate products available to all its members or most of them: this is a necessary condition of its existence. In the medieval world this is not so. A very high degree of sophistication and specialisation in the production of articles consumed by a small minority of the population may coincide with primitive methods in the production of articles for mass use. This was certainly the case in the Byzantine empire. Manufacture of silk textiles, jewellery, perfume etc. involved complex division of labour, though these products were not widely consumed. At the same time such articles as shoes or agricultural implements were produced by simple methods involving little division of labour.
There were certain industrial differences between the Byzantine empire and the rest of Europe which immediately affected the life of the ordinary man. First, iron was much more widely used and smiths were commoner. Second, because of more developed trade a variety of raw materials was available which was not found in the rest of Europe. Third, the existence of a stable monetary system and of regular exchange between country and town based upon it meant that manufactured articles were more widely used in Byzantine society than in most regions of Europe.
This brings us to the question of trade, which cannot in the Byzantine context be separated from industry. Hagiographical and other texts make it clear that merchants who sold goods which they did not manufacture were an everyday feature of Byzantine society, at any rate in the towns.  Such merchants often bought up agricultural produce from the peasants in the country and brought it to the towns to be sold.
Fairs and markets were found everywhere; the smaller ones served as points of exchange between local producers and consumers, and in particular between urban craftsmen and the surrounding peasants. An example is the Paphlagonian fair described in the Synaxary of Constantinople pp. 721-2 (ed. Delehaye). Larger fairs attracted merchants from all over the Byzantine empire and beyond it. The best-known example is the fair at Thessalonika described in the twelfth-century Timarion, but certainly originating several centuries earlier, at which merchants selling goods from all over the Mediterranean world and as far afield as Germany, Portugal and Georgia gathered together annually on the feast of St Demetrius in October. An intermediate case is the annual fair at Ephesus at which, we are told, a hundred pounds of gold were collected in taxes (Theoph. p. 728).
Lastly there was long-distance trade. Merchants specialised in importing goods from the Moslem world, from Iberia, from the Khazars, from Russia, from central Europe, and so on. Many of these were Armenians, Jews, and Syrians who, being to some extent outsiders everywhere, could easily move between different societies. Much was brought to Constantinople by foreign merchants, whose rights and privileges were minutely regulated in commercial treaties. This extensive trade, in which Byzantine manufactured goods were usually exchanged for raw materials from lands outside the empire, was controlled by government officials—Kommerkiarioi—at points on the main trade routes, who both collected taxes on goods passing through and prevented the export of forbidden goods. This regulated system of trade was rendered possible by the stable conditions over a large area which the empire guaranteed, and the special terms of trade which it was able to negotiate with foreign governments, thus creating an area of trade far more extensive and homogeneous than any existing elsewhere in Europe. 
It is against this background of widespread commercial exchange, local, regional and international, that the state of Byzantine industry and trade in the ninth and tenth centuries can be understood. Evidence on the goods actually produced depends mainly on the work of the archaeologist. Until recently Byzantine archaeology was treated in a somewhat stepmotherly fashion by classical archaeologists. There are however published reports from the Great Palace at Constantinople, from Athens, from Corinth, from Cherson, from Pergamum, from Sardis and from Ephesus which provide much information on Byzantine artefacts.
As can be seen, we are desperately short of material from Constantinople outside the Great Palace area, and from most regions of Asia Minor. Any conclusions reached must therefore be provisional. One of the best surveys of the archaeological evidence for Byzantine industrial production during the ninth and tenth centuries is to be found in A. P. Kazhdan, Derevnja i gorod v Vizantii, IX-XV vv., 1960, 190-249.
The ninth century seems to have been a period of rapid growth and improvement in Byzantine handicrafts, after a century and a half of stagnation. This applies not only to the luxury crafts, producing jewellery, high quality silk garments, enamels, icons, reliquaries and other ecclesiastical furniture, but also to those catering for the modest demands of the ordinary citizen. In pottery, for instance, we find side by side with traditional red clay-ware, undecorated or decorated by stamp or applique, a variety of new types—polychrome-ware decorated in slip or underglaze painting, white clay-ware with yellow-green slip, red clay-ware covered with slip and decorated with incised designs, and so on. There is evidence for the large-scale importation of certain wares from other parts of the empire, for example from Constantinople to Cherson, from Thessalonika to Corinth, etc. And the quality of the best pottery steadily improves, the walls become thinner and more uniform, the firing more regular. An Italian observer, whose date is unfortunately uncertain—he may have written as late as the twelfth century—comments on the novelty of Byzantine polychrome pottery. 
Byzantine glass-ware is well represented in the Corinth finds. The glass workshop excavated there, however, belongs to the eleventh to thirteenth century and is of doubtful relevance for the period under study. Techniques combining blowing and moulding were in use, and the better wares were decorated with zoomorphic and anthropomorphic motifs in gold, colour, or by incision. Coloured as well as plain glass was produced. Byzantine glass-ware was highly prized abroad, and glass vessels were among the gifts sent by emperors to foreign rulers. Theophilus Presbyter was struck by the technical process employed by the Greeks in the manufacture of coloured glass. 
Textiles in the ninth and tenth centuries were woven from wool, flax and silk. There was much home production, particularly of woollen garments. But articles of everyday use such as linen sheets were also produced as commodities for the market in many cities of the empire.
Very fine linen sheets were among the articles brought by the widow Danielis from her workshops at Patrae to Constantinople in the middle of the ninth century, and there was clearly much importation and exportation of the finer grades of linen. Silk spinning and weaving and the making up of silk garments and other objects was a virtual Byzantine monopoly in the early middle ages. It was based on local silk production in areas where the mulberry was cultivated as well as on imported raw silk, and on a succession of specialist skills—spinning, dyeing, weaving etc. —which required long study and practice. The highest quality silk of all was not produced for sale by commercial entrepreneurs. It was made in the imperial factories situated within the perimeter of the Great Palace at Constantinople by specialist workers whose posts were hereditary, and was for distribution by the emperor to Byzantine dignitaries and foreign monarchs. Great care was taken to prevent these products coming on to the market or the secrets of the technique by which they were made being divulged. There was a much higher degree of division of labour in these imperial factories than in ordinary commercial workshops, because large numbers of people worked together under the same roof and were subject to a common direction. In spite of the precautions taken there must have been some spill-over of technique from the imperial silk factories to private workshops.
Private silk production was probably confined to Constantinople until the ninth century, and then slowly spread to Corinth, Thebes, Thessalonika and other cities suitably placed. We hear of Jewish silk-weavers in Sparta in the tenth century. All its products were relative luxuries, and were widely exported.
Leather-working is often mentioned. A quarter of Constantinople was known as ‘The Shoemakers’. Metal-working of all kinds, woodwork, ship building, candle-making, soap-making, baking, perfumery etc. are mentioned frequently in sources of the period, though few of their products survive. The products of another group of crafts do survive in the form of Byzantine buildings. New styles, new solutions to technical problems, and new aesthetic ideals marked the renaissance which began in the ninth century. Few datable buildings other than churches remain, but they are numerous—the Church of the Virgin of the Pharos, the Attik-Camii, the Kalender-Hane, the Fenari-Issa in Constantinople, the Škripu church in Euboea, St Nicholas in Sparta, the earliest buildings erected by St Athanasius on Mount Athos, the Nea Mone in Chios and many others.
Compared with western Europe at this time, the Byzantine empire was a region of technological innovation, of increasing division of labour, and of commodity production of luxury goods for an extensive market, which might extend far beyond the confines of the empire.
Broadly speaking, three types of organisation of production and sale can be distinguished. First, that of imperial factories. Here there is no market; producer and consumer are in a sense the same. The craftsman is protected against economic disaster, and is usually in return restricted in his freedom to some extent. Apart from the manufacture of the highest quality of silk goods, which has been discussed, imperial factories produced jewellery and cult objects of various kinds, gold and silver plate and other luxury items for use or distribution by the imperial household. But the sphere par excellence of the imperial factories was the production of armaments. There was a great arsenal in the Mangana quarter of Constantinople, where military equipment of all kinds was made and stored in conditions of tight security. Some products, such as Greek fire and the ‘siphons’ through which it was discharged, were made only in the capital. Other articles such as arrow heads, swords and so on were manufactured elsewhere. At Corinth, for instance there seems to have been a workshop producing such articles. It is not clear whether this was under imperial or private control. The workers in imperial factories might be free or slaves, men or women. Women were employed particularly in textile production, and slaves probably performed tasks calling for less skill. The most skilled workers were mainly freemen, but their occupation was often hereditary, and they were forbidden to leave it. This is a pattern of 'palace industry’ which can be traced back through the Hellenistic world and ultimately to the great Bronze Age civilisations of the Near East. It permitted far-reaching division of labour and the development and conservation of very great skill. But concentration of production in large units of this kind was not generally economically viable in a society in which transport costs were relatively high. So imperial factories did not normally compete in an external market with smaller local producers, They were limited to the production of goods for which the state was the only or the principal customer.
Private producers and distributors of a great many commodities were organised at Constantinople in guilds—somata—whose activities are minutely regulated by the provisions of the Book of the Prefect issued by Leo VI. The literature on this text is immense, and the present writer has no desire to add to it.
It appears that the many regulations of the Book of the Prefect are best explained by a compromise between two purposes. The emperor and the government wanted to keep supplies of essential food, clothing and other commodities flowing regularly and at stable prices. Their motive was primarily political, to prevent mass disaffection among the citizens of the capital, which could weaken the position of the ruler and might be exploited by a rival. This is a continuation of a policy observed by rulers of great states and by the ruling oligarchies of cities in the ancient world, which found its expression in the distribution of food free or at reduced price to citizens, and in other measures of social security. The aim was to smooth out the inevitable fluctuations in the price of basic foodstuffs and articles of mass consumption arising out of variations in supply, and above all to prevent price rises stimulated or aggravated by hoarding. There can often be a conflict of interest between the ruling class as a whole and individual members of it in this connection. One of the purposes of such regulations both in the ancient world and in the Byzantine empire was to ensure that the interest of the ruling class as a whole prevailed.
The second purpose discernable behind the regulations is to safeguard the interests of the producers themselves—to distribute work between all members of the guild, to prevent competition in prices, to prevent concentration of production, and to prevent the growth of vertical monopoly. In these ways the steady livelihood of each member of the guild would be best assured. These were on the whole the purposes which medieval craft guilds in the West pursued. The difference between the situation in western Europe and in the Byzantine empire lay in the fact that the Byzantine guilds developed in the shadow of a great centralised state which, following tradition dating from antiquity, controlled production and distribution for its own political purposes, and would only tolerate the existence and activity of craft guilds if they did not frustrate its political purpose.
Whether there were such guilds outside Constantinople at this time we do not know. They certainly did exist later, and it is probable that they date back to the ninth century if not earlier. They may have developed more freely and represented more closely the interests of their own members, since they did not directly affect the politically sensitive capital city. They are likely to have been much more limited in number than at Constantinople and confined to a few major cities such as Thessalonika.
Where we have no evidence, however, speculation is unwise.
The third pattern of production and sale is that of the private producer working on his own with no more aid than that of his family or possibly one or two wage labourers or slaves, subject neither to regulation nor to protection by others. This was probably the commonest pattern in the production of articles for everyday consumption. Such producers usually sold directly to the ultimate consumer. Middlemen were not prominent. And such enterprises were always small, catering for a limited local clientele. There were no economic advantages to be gained by concentration of production in these cases.
It should however be borne in mind that far more was produced by the user for himself than in the modern world. Thus spinning, weaving, tailoring or dressmaking were regular domestic activities, and much agricultural equipment was made by peasants themselves, supplemented by the occasional work of the travelling smith— kōmodromos—and the wheelwright. In addition monasteries often produced almost all their own requirements, not only for the monks but also for the numerous laymen employed by a rich monastery. The trades represented in the great monastery of St John of Studios at Constantinople are mentioned elsewhere in this book. The signatories of an Athonite document of 1154, monks of the relatively small monastery of St Philotheos, include a carpenter, a shoemaker, a cooper, a tailor, a weaver, as well as a fisherman and a boatman.  A list of monks at Lavra of approximately the same date includes shoemakers, carpenters, a cook, a boatman, a builder, a shipwright and a rope-maker. The situation was broadly similar two or three centuries earlier. In the same way owners of large estates tended to produce what they needed from their own resources and with their own labour.
This tendency towards autarky, together with the low standard of life and the high cost of transport were factors limiting the development of commodity production for an extensive market in the Byzantine empire. In fact the whole economy was organised for consumption by a limited group—the court, the church, the state dignitaries, the wealthy landowners—not for mass consumption. The Byzantine city drew food, raw materials and semi-manufactured goods from the surrounding countryside, but it did not in return supply the countryside with the products of its industry on any scale.
The peasants were paid in money, most of which they used to pay their taxes.
Byzantine international trade was in a somewhat different position. The technical superiority of Byzantine methods of production enabled the empire to export luxury goods such as gold and jewellery, enamels, textiles, carved ivory etc. for consumption by the wealthier classes of the western and northern world. Much of this export was non. commercial in character, taking the form of gifts to foreign potentates. Indeed many luxury goods were not allowed to be exported by private individuals at all. Travellers’ luggage was examined by customs officers—kommerkiarioi—at frontier posts and forbidden goods were confiscated, as Liutprand of Cremona discovered to his cost in 968. Another export was technical skill. Byzantine architectural expertise contributed to the building and decoration of churches all over northern and western Europe, from St Sophia in Kiev to Charlemagne’s basilica in Aachen. No doubt Byzantine skill made its contribution to the production of less durable artefacts of all kinds. We know little about the way in which this export of know-how was organised. In many cases craftsmen would be sent directly from Constantinople to a foreign capital by the emperor to carry out a particular task. In other cases the export may have taken place from a Byzantine provincial town and had a more commercial character.
Foodstuffs and bulk goods of Byzantine origin were not normally exported in the middle Byzantine period. On the contrary food was imported to the capital from foreign territory and in particular from Bulgaria. A considerable part of Byzantine trade consisted in the re-export of goods imported from southern and eastern Asia—spices, silk, incense and other luxury goods of small bulk. The rate of commercial profit on these re-exports was probably high, as the Byzantines still had a virtual monopoly of the supply of these essential commodities to western and northern Europe.
Byzantine imports from the west and north fall into two main categories—raw materials and labour. To the former belong the hides, furs, dried and salted fish, amber and honey imported from Russia; the furs, linen, honey and metals from the northern Balkans, etc. The latter comprise the numerous slaves imported from central and northern Europe through Venice, Thessalonika and the capital itself, to work as domestic servants, industrial labourers—often mere sources of energy—and land workers.
Foreign trade was fairly rigidly controlled, particularly as regards exports.
There was no attempt at an export drive, no desire to conquer new markets, for there were probably none to conquer. Byzantine merchants did not themselves travel far afield to sell their goods. They waited in Constantinople and Thessalonika for foreigners to come to them, driven by the economic forces generated by a sharp difference in productivity. Much foreign trade was governmental in character, and not necessarily commercially profitable. The merchant occupied a humble and undistinguished station in Byzantine society, though he could raise capital fairly easily. Legislation of Leo VI permitted loans of money at 4 per cent per annum (Novel 83).
Nevertheless, the empire’s balance of trade was a positive one, such was the demand in the developing countries of northern and western Europe for Byzantine luxury manufactured goods. The disparity was made good by the import of gold, which flowed from all over Europe to the treasuries of Constantinople.
In spite of apparent restrictions upon commercial enterprise, it is clear that in the period under discussion Byzantine industry and trade was immensely more developed than in other parts of Europe, and comparable to those of the Moslem world. Commodity production for an extensive market obviously played a large role, its importance being indicated by the amount of money in circulation. This kind of evidence is extremely difficult to use, and expert opinions vary on the interpretation of numismatic evidence. For what it is worth, however, the number of coin finds, both in hoards and in single coins, outside the confines of the empire and within its bounds, begins to rise again sharply early in the ninth century after a sudden general drop about the middle of the seventh.
It has been argued that this is evidence of the growing prevalence of a natural economy during the Dark Age of the seventh and eighth centuries, and the resurgence of a market economy based on money from the beginning of the ninth century. The argument is put most forcibly by A. P. Kazhdan, Derevnja i gorod v Vizantii, IX-XVV., (1960), 260-72 and ‘Vizantijskie goroda v VII-XI vv.’, Sovetskaja Arkheologija 21 (1954). More cautious, perhaps sometimes too cautious, attitudes are expressed by P. Grierson, ‘Coinage and Money in the Byzantine Empire 498-c. 1090’. Settimane di Studio del Centro Italiano di Studi sull'Alto Medioevo 8 (1961), 411-53; R.S. Lopez, ‘La crise du besant au IXe siècle’, Mélanges H. Grégoire II, (1950), 412; P. Charanis, ‘The significance of coins as evidence for the history of Athens and Corinth in the seventh and eighth centuries, Historia 4 (1955), 169ff; D. M. Metcalf, Coinage in the Balkans 820-1355, (1965) 1-47.
The general conclusion, supported by narrative texts and by other archaeological evidence, that a resurgence of the Byzantine market economy and a movement away from natural economy began in the early ninth century, is generally accepted.
If Byzantium was at this time a land of money economy, Bulgaria was one of natural economy. The Bulgarian state issued no coinage of its own. Byzantine coins are certainly found in Bulgarian sites, but hardly in any numbers until the middle of the tenth century, and then only in urban sites. There is nothing to suggest that they were used to a significant extent in internal trade. They were rather used by the rich as a means of hoarding wealth in a portable form or for purchasing foreign luxury goods. The almost total absence of money in Bulgarian society is strikingly brought out by contemporary evidence. Land taxation was entirely in kind, and the Byzantines were forced to maintain such a taxation in kind after the conquest. When in 1040 they tried to convert the land tax to a money payment, which had long been the practice in other regions of the empire, they found themselves faced with a revolt on a national scale headed by Peter Deljan. The treaty signed by Michael II and Omurtag in 815 allowed the ransoming of Byzantine prisoners held by the Bulgarians at the rate of two buffaloes for each man.  The Arab geographer al-Masudi, writing in the tenth century but reflecting conditions somewhat earlier, says that the Bulgarians paid for goods in oxen and sheep.
These facts indicate an economy very different in its total structure from that of the Byzantine empire, in spite of the many detailed similarities between them. With this in mind we can begin to examine the scanty data furnished by written sources and by archaeology. Bulgarian society started from a position of relative technological backwardness. The cities, centres of industry in the late Roman period, had been largely destroyed and the craftsmen in them killed or dispersed. The Slav communes were practically self-sufficient, producing food, clothing, houses and such tools as they needed themselves. The two possible exceptions were probably pottery and iron goods. A common type of wheel-turned pottery with incised designs and out-turned lip is found in Slav sites all over central and eastern Europe. They were probably produced by village potters who served the needs of several surrounding villages. Similarly the iron tools which the Slavs certainly used were the product of village smiths or travelling smiths.
Bulgar society was technologically sophisticated in the sense that its ruling class used, either by purchase or by forced labour, the products of numerous artisans, who may have been largely non-Bulgar. The silver-ware found in a probable Proto-Bulgar site at Nagy Szent Mjklós in Hungary is more indicative of Bulgar taste than of Bulgar technology. Nevertheless there must have been many craftsmen working around the courts of the Khan and his principal boyars.
We read in the works of John the Exarch and others of numerous craftsmen—shipwrights, coppersmiths, goldsmiths, weavers, tanners, blacksmiths etc. Excavation at Pliska, Preslav and elsewhere reveals much sophisticated metal-work—including a dentist’s probe—besides well-turned pottery, including much polychrome-ware resembling, but not copying, contemporary Constantinopolitan ware, jewellery and other luxury articles.  This was probably largely produced for the court and the upper classes of the capital. Surviving buildings or those clearly defined by existing ruins testify to the skill of the architects and builders but also to their unfamiliarity with contemporary Byzantine building practice. Sculptors decorated these buildings with reliefs in which Byzantine and oriental motifs are fused. The latter may be due to the tastes and traditions of the Proto-Bulgars.
Side by side with this industrial production within the cities for a limited urban market, the economic life of the countryside went on with scarcely any change—autarkic, without money, and largely without a market. This last point needs some qualification. Cosmas the Presbyter in the mid tenth century speaks of peasants coming to the city for trade. But in the absence of money this was probably a matter of exchanging occasional surpluses for such products of city industry as might be readily available. The cities of Bulgaria were essentially fed from two sources, the cultivation of the land immediately surrounding their walls, and the product of the tax in kind redistributed by officials and magnates to their dependents. Trade was a secondary source. It is noteworthy that there is no word for ‘price’ in early Slavonic texts: cěna, which later has this meaning, means ‘compensation’ or ‘fine’ in the Bulgarian legal code.
It is this relative absence of internal trade which explains the dependent situation in which many of the craftsmen found themselves. Cosmas the Presbyter writes of craftsmen working for their lord during the day and praying by night. The craftsman had only a few customers, or possibly only one, for whom he worked to order, probably with materials provided by the customer.
He was dependent on his customer or customers for the necessities of life, and indeed was, mutatis mutandis, in the same position as those soldiers who are described in early Bulgarian inscriptions as ‘kept men’ of the Khan. He was essentially unfree. There were, however, also free workers negotiating contracts with their customers.
At the same time Bulgarian industry did not remain static. There was some overspill of technique from the royal capital to smaller cities, where it was adapted to local needs. At Oescus (Gigen) in the tenth century much domestic pottery of a traditional type was produced. But a curious closed vessel was also found, with three vent pipes uniting at the top, which has been variously interpreted as a perfumery still or a ritual object (the last resort of the archaeologist who does not know!).  And one industry which processed Bulgarian raw materials seems to have made the crucial step from domestic to craft industry in the period under study. The Book of the Prefect speaks of Bulgarians bringing linen to Constantinople, which they cannot sell directly in exchange for such Constantinopolitan goods as they need.  This passage has been variously interpreted. It may be that these Bulgarians are peasants from Bulgarian Thrace, part-time weavers rather than merchants. Earlier in the same chapter of the Book of the Prefect (9.1) we hear of linen ‘from beyond the Struma’ which is paid for in money by the drapers. This linen is presumably brought to Constantinople by merchants operating under the terms of the commercial treaty between the two countries. At any rate these passages point to the existence of an exportable surplus of textiles. There are indications elsewhere that dressed skins, sheepskin rugs and the like were imported from Bulgaria to Constantinople. So here again we may have a case of the change from subsistence production to market production.
Why did this development of productivity not lead to a more generalised market economy and the use of coined money in Bulgaria? The answer, apart from the general economic backwardness of the country, may lie in the fact that the Bulgarian proprietors, who controlled not only the agricultural surplus, but also such industrial production as existed, found it easier to dispose of their surplus in the nearby Byzantine empire in exchange for Byzantine luxury goods than to encourage the development of home markets, as sometimes happened in western Europe. Church property would be particularly important in this respect. It is perhaps significant that Cosmas the Presbyter condemns monks who ‘engage in buying and selling, sending letters and instructions everywhere, and seeking to become known even in other lands.’
Thus the proximity of the empire with its—in medieval terms—developed market economy enabled the Bulgarian feudal magnates to avoid the dilemma created by their own growing wealth and the poverty of the agricultural world in which they lived; it prevented the Bulgarian economy developing in the directions taken in the contemporary Frankish Kingdom, for example.
Another factor limiting the free development of the Bulgarian economy lay in the needs of the army. Krum put in the field an army of 30,000 men in defensive armour—holosidéroi—with 5,000 wagons with iron tyres, and a variety of siege engines.  We have no comparable information regarding Symeon’s armies, but there is no reason to regard them as smaller or less well armed. Rather the contrary. The equipment and maintenance of such an army must have involved virtually all the metal workers, wheelwrights, harness-makers, carpenters etc. of the country. How this was organised we do not know. There is no sign of long-established government arms factories or arsenals. Presumably it was done by direct personal mobilisation of such craftsmen as were not already working under the direction of the ruler or of one of his boyars. Armaments production, especially under Krum or Symeon, clearly represented a much larger part of the total industrial production of the country in Bulgaria than it did in Byzantium. To that extent industry was isolated from the market and encouraged to specialise in certain limited types of product for which there was little or no outlet on the market.
Bulgaria of course lay on important international trade routes. Some of the trade between the Byzantine empire and the west—some of which was, in its turn, an extension of Byzantine trade with the Moslem world and regions further east—must have passed through Bulgaria. The Byzantines succeeded in profiting by this long-distance trade but the Bulgarians seem not to have done so, or at any rate not to a significant extent. In Byzantium this was done by an elaborate system of customs barriers, both at the frontiers with foreign states and at points within the empire itself, at each of which a kommerkiarios levied tolls on goods in transit in accordance with a complex and sophisticated scheme.  No such system seems to have existed in Bulgaria. The frontiers were constantly guarded in the time of Boris, and no doubt later, but the prime purpose was military security rather than fiscal advantage. The picture which John Cameniates paints of Byzantine-Bulgarian relations in the neighbourhood of Thessalonika —
incidentally at a time when the two countries were at war!—does not suggest the existence of any serious customs barrier on the Bulgarian side. No doubt the Bulgarian rulers levied duty on foreign goods in transit. But they had not the reserve of literate minor functionaries needed to make such a control effective.
Thus in a variety of ways the Bulgarian economy, which was at the start relatively underdeveloped, was distorted by the closeness of the Byzantine empire, by its domination of international trade, and by the demands provoked by the frequent military conflicts between the two countries. Regions with an economy equally well developed, such as the Arab world, derived advantage from the proximity of the empire. Those which were significantly weaker saw their subsequent economic development hindered rather than helped by this proximity.
2. The Balkans in Late Antiquity and the Origin of Bulgaria
1. Cf. E. Gren, Kleinasien und der Ostbalkan in der wirtschaftlichen Entwicklung der römischen Kaiserzeit, 1941.
2. Cf. H. Thompson, ‘Athenian Twilight: a.d. 267-600’, Journal of Roman Studies 49 (1959), 61-72.
3. Cf. V. Georgiev, ‘The Genesis of the Balkan Peoples’, Slavonic and East European Review 44 (1965-6), 285-97.
4. Olympiodorus frg. 27.
5. Jerome ep. 60.4.
6. Cyril of Scythopolis, ed. E. Schwartz, 193-24ff. ; John Moschus, Migne PG 87.3 3025B; Theodore of Petra, Life of Theodosius, Migne PG 114.495B; Antonini Placentini Itinerarium 183.
7. Expositio totius mundi et gentium 50; Justinian Novel 26 pr.
· C. Weigand, Die Aromunen, 2 vols., Leipzig 1894-5;
· M. Gyóni, ‘La transhumance des Valaques balcaniques au Moyen Age’, Byzantinoslavica 12 ( 1950), 29ff.;
· C. Daicovici, E. Petrovici, G. Stefan, Die Entstehung des rumänischen Volkes und der rumänischen Sprache, Bucarest, 1964;
· H. Mihăescu, ‘Die lateinische Sprache in Sudosteuropa’, Zeitschr. f. Balkanologie 6 (1968), 128-36.
· C. Höeg, Les Saracatsans, 2 vols., Paris-Copenhagen, 1925-6;
· V. Marinov, Prinos kŭm izuchavaneto na proizkhoda, bita i kulturata na Karakachanite v Bŭlgarija, Sofia, 1964;
· id., ‘Podvizhno pastirstvo v Bŭlgarija i na balkanskija poluostrov’, Izvestija na Etnografskija Institut i Muzej 8 (1965), 79ff.;
· J.K. Campbell, Honour, family and patronage: a study of institutions and moral values in a Gieek mountain community, Oxford, 1964.
10. On the fate of the cities of the Balkans, and in particular of Bulgaria, during the Avar and Slav invasions cf. V. I. Velkov, Gradŭt v Trakija i Dakija Prez Kŭsnata antichnost, Sofia, 1959 — an admirable work which unfortunately does not go beyond 600; V. Beševliev, ‘Les cités antiques en Mésie et en Thrace et leur sort à l’époque du haut moyen âge’, Etudes Balkaniques 5 (1966), 207-20—where füll references to the literature will be found; and the opening chapters of S. I. Lishev, Bŭlgarskijat srednovekoven grad, Sofia, 1970.
11. Cf. V. Beševliev, ‘Aus der Geschichte der Protobulgaren’, Etudes Balkaniques 6 (1970), 39-56.
12. Cf. V. N. Zlatarski, ‘Izvestieto na Mikhail Sirijski za preselenieto na Bŭlgarite’, Izvestija na Bŭlgarskoto Istorichesko Druzhestvo 4 (1915), 38-52.
13. On the relations oř the Proto-Bulgars with the various Slav tribes of Moesia and on the frontiers on the earliest Bulgarian state cf. the discussion by I. Dujčev, ‘Obedinenieto na slavjanskite plemena v Mizija prez VII v.’, Izsledvanija v chest na Marin Drinov, Sofia, 1960, 417-28.
14. Cf. Zh. N. Vŭzharova, ‘Slavjani i Prabŭlgari (Tjurko-bulgari) v svetlinata na arkheologicheskite danni’, Arkheologija 13 (1971 ), 1-21; id., ‘Pamjatniki Bolgarii kontsa VI-IX u, i ikh etnicheskaja prinadlezhnosť, Sovetskaja Arkheologija, 1968, 3, 148-59; id., Slavjanski i Slavjanobŭlgarski selishta v bŭlgarskite zemi ot Kraja na Vl-XI v., Sofia, 1965.
15. V. Beševliev, Die protobulgarischen Inschriften, Berlin, 1963, No. 4.
3. Bulgaro-Byzantine Relations in the Ninth- and Tenth Centuries
1. Cf. the most recent examination of the evidence by Alkmene Stavridou-Zaphraka, Hē synantēsē Symeon kai Nikolaou Mystikou (Augoustos 913) sta plaisia tou Byzantinoboulgarikou antagōnismou, Thessaloniki, 1972.
2. Migne PG 111.449 ff.
3. On Bulgar primogeniture cf. most recently G. Tsankova-Petkova, ‘Vlijanija na vizantijskite politicheski institutsii u Bŭlgarite prez XI vek’, Studia Balcanica 2 (1970), 98-9.
4. Cf. E. Georgiev, ‘Prabŭlgarskoto letopisanie’, Izsledvanija v chest na Marin Drinov, Sofia, 1960, 369-80.
5. Cf. I. Zaimov, Bitolski nadpis na Ivan Vladislav, samodŭrzhets Bŭlgarski, Sofia, 1970, 79 ff.
6. Cf. L. Petit, Typicon de Grégoire Pacourianos pour le monastère de Petritzos: Supplement to Vizantijskij Vremennik 11 (1904), xi ff.
7. Cf. I. Dujčev, ‘Tsentry vizantijsko-slavjarskogo obshchenija i sotrudnichestva’, Trudy Otdela Drevnerusskoj Literatury 19 (1963), 107-29.
4. The Land
1. Les fondements géographiques de l'histoire de l'Islam Paris, 1968, 204.
2. Strabo 12.577.
3. Cf. L. Robert, Journal des Savants, 1961, 147.
4. On the Geoponica, of which we do not yet possess a satisfactory critical edition cf. J. L Teall, ‘The Byzantine Agricultural Tradition’, Dumbarton Oaks Papers 25 (1971), 33-59, esp. 40-44. Translations into Pehlevi, Arabic, Syriac, Armenian and Latin of the work or of its principal source bear witness to the wide influence of Byzantine agricultural theory.
5. Vita Clementis 23.68.
6. On the importance of millet among the Slavs of the north-east Balkans cf. Zh. N. Vŭzharova, O proiskhozhdenij bolgarskikh pakhotnykh orudij, Moscow, 1956, 42.
7. Book of the prefect 9.6.
8. Cf. Zh. N. Vŭzharova, Slavjano-bŭlgarskoto selishte kraj selo Popino, Silistrensko, Sofia, 1956, 92-4.
9. Cf. J. L. Teall, ‘The Byzantine Agricultural Tradition’, Dumbarton Oaks Papers 25 ( 1971 ), 51; G. Duby, L'économie rurale et la vie des campagnes dans l'occident médiéval, 1962, 71-9.
10. I. Dujčev, Iz starata bŭlgarska knizhnina I, 73.
11. Ibid., 80.
12. Cf. Zh. N. Vŭzharova, O proiskhozhdenii bolgarskikh pakhotnykh orudij, Moscow, 1956, passim.
13. The literature on Byzantine land-tenure is immense. Cf. exempli gratia :
· G. Ostrogorsky, Pour l’histoire de la féodalité byzantine, Brussels, 1954;
· P. Lemerle, ‘Esquisse pour une histoire agraire de Byzance’, Revue Historique, 219 (1958), 32-74, 254-84 and 220 (1958), 43-94 (these two studies are fundamental to all modern discussions of the problems);
· P. Kazhdan, Derevnja i gorod v Vizantii IX-X vv, Moscow, 1960;
· Germaine Rouillard, La vie rurale dans l’empire byzantin, Paris, 1953;
· S. Vryonis, ‘Byzantium: the social basis of the decline in the eleventh Century’, Greek, Roman and Byzantine Studies 2 (1959) 159-175;
· K. I. Watanabe, ‘Problèmes de féodalité byzantine. Une mise au point sur les diverses discussions’, Hitotsubashi journal of Arts and Sciences 5/1 (Jan. 1965), 31-40 and 6/1 (Sept. 1965), 7-24.
14. N. G. Svoronos, ‘Recherches sur le cadastre byzantin: le cadastre de Thèbes’, Bulletin de Correspondance Hellénique 83 (1959), 1-145.
15. Ed. M. H. Fourmy and M. Leroy, Byzantion 9 ( 1934), 113ff.
16. P. Noailles and A. Dain, Les Nouvelles de Léon VI le Saga, Paris, 1944, 376.
17. A convenient summary of the laws in question can be found in D. A. Zakythinos, Hé Byzantinē autokratoria 324-1071, Athens, 1969, 252-64.
18. J. Ivanov, Bŭlgarski starini iz Makedonija, Sofia, 1931, 547-62.
19. Anonymi (P. Magutri) Gesta Hungarorum, ed, Aem. Jakulovich, (Scriptores Rerum Hungaricarum 10), Budapest, 1937, 49-50; Legenda S. Gerbardi episcopi, ed. E. M. Madzsar (Scriptores Rerum Hungaricarum 11) Budapest, 1938, 489-90.
1. Novel 46, Noailles-Dain, op. cit., 185.
2. A summary of the evidence for and against continuity of occupation of various city-sites will be found S. N. Lishev, Bŭlgarskijat srednovekoven grad, Sofia, 1970, 5-26.
3. Cf. E. Kirsten, Die byzantinische Stadt, Berichte zum XI Internationalen Byzantinisten-Kongress, München, 1959, V. 3, Anm. II, 69-72, where a survey of the evidence is given.
4. Nova Tactica, ed. H. Gelzer in Georgii Cyprii descriptio orbis Romani, Leipzig, 1890, 61-83.
5. For a survey of the economic and demographic situation of Asia Minor in the eleventh Century cf. S. Vryonis, The Decline of Medieval Hellenism in Asia Minor, Berkeley-Los Angeles-London, 1971, 1-68. There is no comparable survey for an earlier period.
6. References to sources in P. Tivčev, ‘Sur les cités byzantines aux XIe-XIIe siècles', Byzantinobulgarica 1 ( 1962 ), 145-82 and in P. Charanis, ‘Observations on the Demography of the Byzantine Empire’, Proceedings of the XIIIth International Congress of Byzantine Studies, Oxford, 1967, 460.
7. Cf. D. Krŭndžalov, ‘Is the fortress at Aboba identical with Pliska, the oldest capital of Bulgaria?’, Slavia Antiqua 13 (1966), 429-49, for a careful statement of the case against the identification.
8. Theophanes 490.26; George the Monk, II, 774.19.
9. Cf. V. Beševliev, Die protobulgarischen Inschriften, Berlin, 1963, 276.
10. German translation in V. Beševliev and J. Irmscher, Antike und Mittelalter in Bulgarien, Berlin, 1960, 235.
11. V. Beševliev, Die protobulgarischen Inschriften, Berlin, 1963, 190-206.
12. V. Beševliev, Die protobulgarischen Inschriften, Berlin, 1963, Nos. 3, 81, 85.
6. Industry and Trade
1. Cf. A. Rudakov, Ocherki vizantijskoj kul’tury po dannym grecheskoj agiografii, Moscow, 1971,154-74.
2. Cf. R. H. Bautier, The Economic Development of Medieval Europe, 1971, 64.
3. Theophilus Presbyter, De diversis artibus 2.16, p, 47, Dodwell.
4. Theophilus Presbyter, De diversis artibus 2.13; 2.15, pp. 45, 46, Dodwell.
5. P. Lemerle, A. Guillou, N. Svoronos, Actes de Laura, I. Des origines à 1204 Paris, 1970, 324.
6. V. Beševliev, Die protobulgarischen Inschriften, Berlin, 1963, 202.
7. Cf. I. Changova, ‘Kŭm prouchvaneto na preslavskata risuvana keramika', Arkheologija 1972, 33-9.
8. Cf. M. Stancheva and L. Doncheva-Petkova, ‘Srednovekovna bitova keramika ot Eskus pri s. Gigen’, Arkheologija 1972, 22-32.
9. Book of the Prefect 9.6.
10. Their nature is discussed by V. M. Zlatarski, Istorija na bŭlgarskala dŭrzhava prez srednite vekove 1, 1918, 415-20.
11. Cf. H. Antoniadis-Bibicou, Recherches sur les douanes a Byzance, Paris, 1963.
[Back to Index]