If the idea of late antique prosperity in the east has become generally accepted, so too has that of the seventh century marking the onset of a ‘dark age’, at least in the Byzantine world. Those areas fortunate enough to have been conquered by the Muslims prospered. In Syria, Palestine and Egypt, the picture appears very much one of business as usual. Not just for the inhabitants of Damascus which became the new Islamic capital (see above, p. 221), but for many provincial cities too the seventh century was a ‘good’ period. The buildings, public and private, the coinage, the pottery and the small finds from such well-excavated sites as Pella and Gerasa (Jerash) in Jordan, or Scythopolis (Bet Shean/Baysan) and Caesarea Maritima in Israel tell a clear story. For the villagers of the limestone massif too, recent work has done much to show that life carried on much as before, and field survey projects throughout the region give the same picture.12 But for the rump of the Roman east, those territories that stayed under the rule of Constantinople and made up the world of Byzantium, this period was grim. Anatolia and the remaining territories in Greece and the Balkans experienced recession on a scale that justifies the description ‘the collapse of the ancient economy’. The evidence is seemingly incontrovertible and rests on the same sorts of indices that have been used to show the dynamism and vitality of the late antique economy: buildings, pottery and coinage. In a series of seminal publications Clive Foss looked at the evidence coming from a number of excavated ancient cities inTurkey.At Pergamon, Ephesos, Sardis,Magnesiaon- the-Maeander, Priene, Miletos and Aphrodisias on the west coast, at Side on the south coast and from Ankyra in the interior of Anatolia the same picture emerged. A centuries-old urban economy came to a halt. The construction of public buildings, which in late antiquity may have shifted from theatres and baths to churches, was now limited to defensive walls and minimal repairs. The amphorae and red-slip pottery characterising late antique sites disappear – to be replaced by local, hand-made products. Copper coins, the loose change of late antique urban life, similarly vanish. The graphs of stray coin finds on these sites make the message plain. Once common, their number falls to almost nothing for the seventh and eighth centuries. Ephesos, one of the great cities of the late antique east, contracted to a walled settlement around the harbour. The decline of Sardis was starker still. The seventh century saw it transformed into a hilltop fortress, with a few knots of primitive dwellings on the plain beneath (fig. 31a). More or less the same can be said for the other sites too.Hierapolis, the city that had recovered robustly from a fourth-century earthquake, was struck again in the seventh century. There was no rebuilding this time.Hierapolis survived as a scatter of very basic farmhouses.13 Even Constantinople appears not to have been immune. There was certainly more coinage available in the imperial capital.Agraph of stray coin finds from the city-centre excavation of the church of St Polyeuktos shows a rate of coin loss that continued at a steady level right through these centuries. Similarly, although in Constantinople as elsewhere late antique pottery types had disappeared by the end of the seventh century, what replaced them is much clearer here. Constantinopolitan white ware, a type of glazed pottery made from local clay, appears in the seventh century to fill the gap left by the end of the red-slip tradition.14 Yet, even in the capital there is no doubting the basic trend. Late antique Constantinople was a boom town with a major building industry, capable of carrying out such huge projects as the long-distance water supply of the city, six kilometres of the Theodosian land walls, and 45 kilometres of the Anastasian long walls of Thrace, as well as erecting Justinian’s massive church of St Sophia in under five years. Projects like this, which in turn are only a fraction of the total number of houses, palaces, churches, cisterns, colonnades and monuments put up during these years, required brick production on an industrial scale. Recent work on the stamp markings of Constantinopolitan bricks has shown as much.15 All this came to a halt in the seventh century. For the following two centuries one has the impression of a huge salvage site. The paltry new building-work done in these years employed reclaimed bricks and marble and was decorated with reused mosaic cubes. Medieval Constantinople, like medieval Rome, would seem to have been living amidst the ruins left by giants.16 Recent work at Sagalassos tells the same story. The prosperous city of late antiquity came to an end in the seventh century. The tradition of public building and the Sagalassan pottery industry stopped. Hannelore Vanhaverbeke’s survey shows that neither the city nor the surrounding countryside were deserted; rather, life continued, but with fewer people and on a simpler scale. Pollen analysis suggests the disappearance of olives as a major crop, a decline in cereal production, the spread of woodland and perhaps a shift to pastoralism. What we seem to see is a society and economy top-sliced: deserted by its elites, leaving a poorer, tattier, rural world.17 By the tenth century at latest it is clear that theMediterranean economy was reviving and, equally, that the Byzantine world shared in this process.Using the same indices that plotted the decline of the late antique economy – buildings, pottery, coinage, settlement surveys and pollen analysis – a new prosperity can be seen emerging. One rough-and-ready index is provided by church-building. While there is very little to show for the seventh and eighth centuries, from the ninth century onwards the numbers of urban and rural churches rise and continue to go up through the tenth, eleventh and twelfth centuries. Byzantine church architecture achieves its classical form during these years, characterised by small domed buildings, their interiors covered in wall-paintings or mosaic, their exteriors enlivened by decorative brickwork.18 Similarly Byzantine glazed pottery, jars and amphorae become much more common. Likewise with coinage: the graphs of stray coin finds that ‘flatline’ for the seventh and eighth centuries take off again from the ninth century onwards, slowly at first, but on a steepening gradient. By the eleventh century, stray finds of copper coins are commonplace.19 Pollen analysis and patterns of alluviation can also be brought into play. The clearest pictures to date come from Greece and Crete, where from the ninth century onwards tree pollens decline and those from plants associated with agriculture and deforestation go up; a cycle of erosion and the silting-up of coastal plains appears to have been triggered by growing exploitation of fragile hill-slopes.20 Survey projects in central and southern Greece, on Cyprus and Crete and now in Turkey too, all point towards middle Byzantine rural growth. Everywhere there are more sites with more things; and the more we can identify middle Byzantine pottery types, the more obvious does this trend appear.21 Cappadocia is perhaps the most striking example, since there the rock-cut churches, houses, store-rooms and water-systems actually survive as visible structures and the churches can be dated by their wallpaintings. New caves were being dug in the later ninth century. Numbers increase through the tenth, and the region was clearly thriving in the eleventh.22 Towns follow the same pattern. The excavation of middle Byzantine towns may not have revealed anything very photogenic, but the houses, churches, workshops, alleys, yards, cisterns and rubbish dumps of Athens, Corinth and Sparta in Greece, Amorion, Aphrodisias, Pergamon and Sardis in Turkey show much more happening than had been the case in the seventh century. The story of Byzantine Constantinople is obscured by the fact that modern Istanbul lies on top of it, and the record of urban archaeology in the city is patchy at best; but the basic trend shows through.23 Written sources confirm this picture.Merchants, craftsmen and markets are a commonplace in middle Byzantine texts. The Book of the eparch, dating from the late ninth or early tenth century, regulates the commercial world of Constantinople. We hear about guilds and money-dealers, merchants coming with silk fromSyria and the trade that brought sheep, pigs and cattle to feed the city. Much of this trade in livestock had its roots in Anatolia. In the eleventh century, John Mauropous described his see of Euchaita in Anatolia as the centre for an important annual livestock market. One of the termini for the drove-roads heading to the imperial city was Pylai on the Gulf of Nikomedeia. Leo, metropolitan of Synada, banned from the capital in the late tenth century, could grumble that there was an express service for animals to the imperial city, but not for him. Boats appear in documents from the monasteries of Mount Athos, as do shops, acquired as investments. Jewish merchant communities are known, not just from Constantinople, but from Mastaura, a small town in western Turkey, 100 kilometres from the sea. The Life of Metrios describes the saint picking up a bag of gold coins dropped by a merchant, heading for market in the next small Paphlagonian town. Agricultural expansion, too, is well attested. The Athos archives talk of the opening up of new land and the planting of new vineyards and olive groves. Michael Psellos talks about doing much the same for the monastery of Medikion which he had acquired on the shore of the Sea of Marmara. Eustathios Boilas’ will shows him opening up extensive new farmland near the empire’s eleventh-century borders.24 These references could easily be multiplied, but to count them up would have no statistical value. They are cited as examples of the ordinary, and to illustrate the fact that in town and country, commerce, manufacture and agriculture, the middle Byzantine empire appears to have been a prosperous world. During the later decades of the eleventh century the empire suffered a profound and prolonged crisis. By the later 1080s most of Anatolia had slipped from imperial control, the south Italian provinces were lost, and it appeared likely that the Balkan provinces would go the same way. Under Alexios I Komnenos (1081–1118), his son and grandson, the empire pulled back from the brink, but much was lost for good (see below, pp. 610–12, 629–46). The crisis’ economic consequences are palpable. The Cappadocian rockcut building boom came to an end; in the central Anatolian theme of the Anatolikoi, the bustling – if not very smart – eleventh-century town of Amorion survived only as a small village, the former cathedral becoming a storehouse and stable.25 Cappadocia and the Anatolikoi were lost to the Turks, but the effects were no less profound in what remained of imperial territory. Constantinople as a centre of demand and population inevitably dominated the Byzantine economy more fully than it had done before; at the same time, even if not severed by a strict frontier, the imperial city was now separated from much of its natural hinterland. The Anatolian droveroads, the livestock markets, the new farms of Eustathios Boilas’ eastern estates would never be Byzantine territory again. The loss of territory entailed loss of tax revenue. The empire had until now paid its servants in gold coin, in an annual ceremony that saw the highest earners drag bags of gold coin across the palace floor. This cash fuelled demand for land, services and goods. The system was showing signs of strain under Isaac I Komnenos (1057–9), but under Nikephoros III Botaneiates (1078–81) it went bankrupt and under Alexios I Komnenos was permanently pruned. The Byzantine state still raised taxes and still paid some of its servants in gold coin; but it could no longer do so for them all. The twelfth-century empire saw a great deal more tax farming, in cash and kind; and many more of the state’s servants were now rewarded with land not gold.26 Yet, although in some respects these years of crisis were a watershed, in others theywere clearly not. AsMichaelHendy pointed out,27 in those areas remaining under Byzantine rule (in other words Greece, the Balkans and, after the reconquests of the early twelfth century, most of the coastlands of Asia Minor) the tenth- and eleventh-century trend seems to have continued, if anything on an increased scale. Hierapolis has been cited twice already, as an example of late antique provincial prosperity and as one of subsequent urban collapse. Growing again from the ninth century on, it was in the twelfth century very much in the empire’s ‘wild east’, a frontier territory where a deacon could be described as an expert sheep-rustler and Turkish raiders could be seen from the city walls. Yet even here the ruins of the ancient baths were reoccupied as workshops for potters and blacksmiths, and excavation has uncovered quantities of good-quality glazed pottery for which there was clearly a market.28 Similar evidence could be cited from town and country, from all parts of what was left of Byzantine territory. Despite the changes brought about by the late eleventh-century crisis, Byzantium did not enter a renewed ‘dark age’. Rather, the evidence points to the twelfth century as actually the peak of the middle Byzantine revival.