It is against this background that one should view the various manuals of governance and law-collections dating from Leo’s reign. They evince his enthusiasm for order, godliness and good learning. Besides commissioning, compiling or interpolating these works he wrote numerous sermons. He aspired to be acknowledged as the fount of wisdom and pious enlightenment, judging by the description of his bathhouse near the palace complex. Leo’s sobriquet, ‘theWise’, implied in the bathhouse imagery, acclaimed by contemporary courtiers and derided by Symeon of Bulgaria, was not wholly undeserved. Like his father Basil I, he wished his rule to be associated with illustrious figures of the Christian empire’s acknowledged heyday, notably Constantine and Justinian. At the same time he propagated the idea of renewal in, for example, his highly euphemistic version of Basil’s accession: the former state of affairs had been removed together with Basil’s senior coemperor, Michael III, ‘for the purpose of fresh and well-ordered change’.10 The concept of ‘cleansing’ government and society of the corrupt and the obsolete is threaded through the novellae of Leo VI, an assemblage of 113 ordinances, mostly dating from the earlier years of his reign. They are largely concerned with morality and church discipline, and envisage a welltempered society whose laws apply to all men save the emperor; he has been granted ‘discretionary powers’ (oikonomia) over earthly affairs by God. The laws, it is repeatedly asserted, are to help men, bringing benefits to their souls as well as to their bodies.11 How far Leo’s novellae were practicable administrative instruments and how far they were enforced is, however, uncertain. The Book of the eparch was issued in 911 or 912 in the name of Leo VI. Its preface invokes by way of analogy the tablets upon which the Law was disclosed by God’s ‘own finger’ for all mankind,12 but its scope is confined to Constantinople, whose administration was supervised by the eparch. It regulates the conduct of nineteen guilds, and lays down harsh penalties for those who breach the regulations. General professions of concern for the welfare of the emperor’s subjects are here juxtaposed with detailed administrative procedures. The Book of the eparch reveals something of the government’s assumptions and priorities. It is particularly concerned with top-quality products such as silks, purple dyes, silver- or goldwork and spices. Five guilds connected with the silk industry receive detailed attention, whereas tanners and leather-softeners get cursory treatment and numerous other known guilds are not mentioned at all. The monopolisation and rationing out of luxury goods was the stock-in-trade of imperial statecraft, at home and abroad. Great efforts were made to ensure that the various stages of production and retail of silk remained in the hands of different professions, and dealers in less valuable goods such as groceries, meat and soap were also not to merge their enterprises. Small-scale units could safely be allowed to monitor their own operations and their own tax assessments and collections to a large extent; fewer officials were thus required for them. The Book of the eparch essentially envisaged self-regulation by craftsmen and traders in conjunction with the City authorities. A still more urgent priority for the government was provisioning at affordable prices. The heads of the fishmongers’ guild were to report to the eparch at dawn on the night’s catch, whereupon he set a price. The prices of meat and bread were likewise set by him; rigorous inspection of all weights and measures was enjoined. The drafters or revisers of the Book of the eparch assumed that residence in Constantinople was a privilege, and ‘exile’ was a harsh penalty in itself. No clear distinction was drawn between provincials and foreigners: for example, anyone ‘from outside’ bringing any kind of merchandise ‘into the God-protected city’ was to be closely supervised by the eparch’s deputy; a list of their purchases was to be made at the end of their stay, ‘so that nothing forbidden should leave the reigning city’.13 The sale of pigs and sheep was regulated in detail; the express aim was cheaper food for the populace, and the interests of provincial producers were secondary. All this probably had a positive effect on the citizens’ well-being, but it also publicised the emperor’s solicitousness. An emperor enjoying the citizens’ goodwill was screened against would-be usurpers. Leo broadcast his piety and accentuated the mystique of emperors born in the Porphyra (himself and his son Constantine). He maintained the festival celebrating the consecration of the Nea Ekklesia built by Basil I. A dirge composed soon after Leo’s death linked Constantinople and the reigning family thus: O City, sing, intone the praise of Basil’s noble offspring, For they impart a deeper hue To thy imperial purple.14 Bread-and-butter issues were at least as important as pomp in winning the sympathies of the populace. Leo seems to have realised this. Concentration on the emperor’s home town rather than the provinces is not particularly surprising.More striking is Leo’s assumption, in compiling his Tactica in (for the most part) the 890s, that the provinces are vulnerable to enemy attack and that this will continue indefinitely. He states that the work is for fighting the Saracens, who harass his subjects ‘day by day’.15Warfare is essentially defensive, and commanders must ensure that all necessities are removed from areas under attack to safe places, livestock dispersed and the population evacuated. The Arab raiders should be attacked only when returning, weary and preoccupied with booty. Here, at least, the emperor was attuned to life as it was lived in the eastern provinces. Much the same tactics are advocated in Skirmishing, which drew on first-hand experience of the ninth- and earlier tenth-century borderlands and was composed in the milieu of the Phokas family; it presupposes that humans as well as livestock will be amongst the raiders’ encumbrances, and the strat¯egos is to assume that his troops will be numerically inferior to the raiders.16 The subterranean settlements of Cappadocia provide material evidence for the insecurity of the south-eastern provinces. Some predate the Arab invasions, but others, such as Salanda, 80 kilometres west of Caesarea, were created then. Several of the millstones which closed its numerous entrances are still extant, though such ingenuity did not prevent this redoubt from being captured in 898 and again in 906–7. Skirmishing sets notably less store by man-made fortifications than by familiarity with mountain heights and natural defences from which observers can gauge enemy numbers and movements.17 Rapid movement was here at a premium, thus limiting what the mounted raiders could take back. Their numbers seldom exceeded 10,000, and were often far smaller. The brunt of the seasonal land raiding was borne in the south-east borderlands. Nonetheless, Skirmishing’s preoccupation with finding out the raiders’ targets betrays the difficulty of keeping track of them, let alone of mustering soldiers from widely scattered agricultural holdings. Its detailed provisions for coping with major invasions, replete with siege equipment, bespeak a state of alert and uncertainty as to where the next blow would fall. No less uncertainty overhung the southern and western coastal districts of Asia Minor. The amir of Tarsus despatched or led naval razzias, and these, like the piratical fleets operating from north Syrian ports, enjoyed a safe haven in Crete, if needed. It was there that the pirate chief Leo of Tripoli withdrew after sacking Thessaloniki, the empire’s second city, in 904, and there 22,000 prisoners were counted before being auctioned to the Cretans. For a while Leo’s fleet was expected to attempt an attack on Constantinople; it was probably this, rather than just the humiliation at Thessaloniki, that spurred Leo VI into large-scale countermeasures. But a combined land-and-sea operation soon collapsed. The commander of land forces, Andronikos Doukas, had recently led a successful incursion into Cilicia. He now fell under suspicion of rebellion and fled to Baghdad after holding out in the fortress of Kabala for six months in 905. A later, massive task force under the command of a trusted civil servant and relative-by-marriage of Leo VI, Himerios, was directed at least partly against Crete, from which the Byzantines had vainly tried to dislodge the Arabs in the ninth century. Himerios was no more successful in 911–12, even though he seems to have followed the precepts of Leo’s Tactica, and Leo of Tripoli remained at large in the Aegean for ten more years.18 Arab raids are quite commonplace in tenth-century hagiography; the tales may be fabulous, but their setting has substance. The sermons of Peter, bishop of Argos (c. 852–c. 922), and his Life concur in suggesting that the locals looked to the saints and to Peter himself, rather than to the emperor, for protection.19 Peter regularly ransomed captives from pirates who put in at Nauplion; and, reportedly through the miraculous production of flour, he acted to relieve a famine. Peter’s ransomings were not far removed from tribute, and it seems that a regular form of tribute was exacted from the inhabitants of southern Aegean isles such as Naxos. At one level these facts of provincial life make a mockery of the bien pensant Leo VI’s public pronouncements. Yet the raiding fleets were normally modest, and the boats in everyday piratical use needed to be small and light, to facilitate swift concealment in Aegean coves. So their carrying capacity was restricted. In any case, not even Byzantine orMuslim authorities could achieve high standards of seaworthiness; naval technology did not allow either side to dominate the seas, and vessels of any bulk tended to ply a limited – and predictable – range of routes. The Muslim fleets seldom liaised with one another, being intent on plunder, not conquest. The account of one of Leo of Tripoli’s captives of 904 suggests there was more or less covert trafficking between the Muslim and Christian zones, involving redeemable prisoners and other commodities.20 The smattering of copper coins belonging to Cretan amirs found on the Greek mainland may hint at commercial exchanges. In the border regions, local self-reliance and deals with the men of violence were unavoidable. Some of the areas most exposed to enemy raids actually showed signs of increasing economic activity and wealth. In Sparta and Corinth the coin sequences which had begun in the mid-ninth century continue uninterrupted through the first half of the tenth. Still more suggestive is the proliferation of painted chapels and churches in the rocks of Cappadocia. Some formed part of monasteries, but most were lay foundations, serving as shrines, marks of piety and oratories. Similar monuments may well have been raised above ground in other provinces, particularly those in northwest AsiaMinor, long secure from Arab raids.On the fertile southern shore of the Sea ofMarmara lay several large wealthy monasteries, and ports such as Kyzikos, Pylai and Trigleia offered outlets to convey produce and livestock to the megalopolis. Under intensive police and customs scrutiny, the Sea of Marmara was the inner sanctum of the empire, prosperous and secure. There are signs of economic dynamism at Constantinople itself in the early tenth century. The size of the population remains uncertain, but the number of buildings was apparently increasing. Leo’s novellae regulate building land and the spaces to be preserved between buildings, in ways not found in the Justinianic planning legislation, and this hints at greater building density.21 Yet even in the megalopolis, driver of the Byzantine economy, the scale of activity and growth was modest. The citizens’ needs could apparently be met by twenty-four notaries. Five of the nine owners of the shops listed in a mid-tenth-century rental note were officials or title-holders, and only one is identifiable by his trade. The richest pickings came from supplying the state or holding office, and the government was by far the largest employer in Constantinople. The palace complex required many hundreds of servants; eunuchs, pages and foreign bodyguards were reportedly numbered in their thousands. Most of those attending banquets or other ceremonies were holders of offices, heads of guilds or other such city worthies, but persons who held titles yet lacked a state function could attend. A text deriving from Leo VI’s reign specifies the sums payable for certain court titles and offices, and indicates the roga payable annually by the treasury to titleholders according to their rank. Provided that the purchaser lived on for several years, he could make a profit, but the advantage lay mainly in the conspicuous connection with the imperial court, invaluable given the multifarious dealings which any man of property would have with tax inspectors and other officials.22 The purpose of the unremitting palace ceremonial was set out by Constantine VII Porphyrogenitus (945–59) in the preface to the handbook on ceremonies he commissioned: ‘may it be an image of the harmony of movement which the creator gives to all creation, and be regarded by our subjects as more worthy of reverence and therefore more agreeable and marvellous.’23 The establishment over which the emperor presided was as just and as immutable as God’s, and attempting to overturn it was tantamount to challenging God’s order of things – and no less wicked or futile. The ceremonies also dramatised the emperor’s role as the sole source of legitimate authority, and of serious money. Leo VI recommends the appointment as general of a ‘good, well-born and rich [man]’ even while piously urging a more meritocratic approach.24 Leo probably appreciated how much the running of his army in the provinces depended on officers’ local connections and resources. The rank and file did not receive substantial regular cash wages, and Leo’s Tactica discusses the problem of ensuring a high turn-out of well-drilled soldiery after a call to arms. His solution is a combination of fiscal privileges for the soldiers with the arousal of religious fervour throughout provincial society, so that non-combatants would be predisposed to contribute unstintingly to the war effort. In this respect, at least, the Muslims’ mobilisation of their society to participate in the jihad appeared to Leo a shining example.25 The reforms would have to be carried through by one of the army’s few full-time components, officers above the rank of droungarios. These were appointed directly by the emperor and drew their salaries from him, but their effectiveness would not be the less for their being gentlemen of private means. The strat¯egos who commanded them had to cope with enemy incursions. He had to take major decisions, and possessed sweeping powers to requisition and to evacuate civilians.He was left largely to his own devices, but the term of office was short and he was forbidden from owning land in the theme he governed, a provision evidently designed to prevent close ties growing up between the governor and local society. It could not always be enforced, especially in the distant south-eastern borderlands. Yet on the whole a balance was struck between affiliations, imperial and local. Imperial propaganda did not merely proclaim an ideal of good order from the palace. The palace rites nearly all involved prayer or the veneration of the sainted.Many involved liturgical celebrations in St Sophia or churches outside the palace complex. The emperor constantly led his entourage in prayers for the welfare of his subjects, acting together with the patriarch and fortified by the concentration in his palace of Christendom’s finest relics, the Instruments of Christ’s Passion among them. The rhythmic intercession gained in significance from the disorder which many provincials endured, constituting both an oasis and a clarion call for supernatural aid. Such a combination of imprecation and material splendour amidst all-enveloping turbulence could be found in the west, in Cluny, and the spell which Cluny’s sumptuous liturgies cast on the propertied classes of Francia was perhaps akin to that of the basileus’ festive prayers in Byzantium. His ritual displays of intimacy with God and philanthr¯opia for his subjects were the visible accompaniment of works of legislation and tabulations of good administrative practice. Those who did not view their interests or spiritual salvation as best served by the imperial establishment were too poor, localised and ill-equipped to take concerted action; the nearest they came was to respond tardily, if at all, to the general call-to-arms which the authorities periodically issued. Widespread if unchronicled apathy meant that strat¯egoi had little hope of turning their forces against the government successfully. Their regular soldiers were too few and often too dispersed, and their principal mode of guerrilla warfare was ill-suited for an assault on Constantinople’s formidable walls, which were ringed by water.