If we look at the structures of aristocratic dependence, these same processes can be seen again, from a different standpoint. Major aristocrats needed an armed entourage, of fideles or, to use the new terminology of the late eighth century and onwards, vassals: men who had sworn oaths of loyalty to them, and who had often, probably, gone through some form of ceremony representing dependence. As public power became weaker in many places, that ceremony became increasingly elaborate and ritualized, for personal bonds of this type became ever more clearly the key to effective political power. This was also increasingly linked to military status itself. Under Charlemagne, military service was still the theoretical obligation of all free men, but even then, in practice, warfare was carried out by professionalized soldiers, milites, most of them in the entourage of their sworn lords. From the ninth century onwards, military status was increasingly seen as the prerogative of an élite, and entry into it was also associated with a ceremony, increasingly often an ecclesiastical one. This network of rituals underpinned what historians after 1000 call ‘knighthood’, and one translation of miles is by now not just ‘soldier’, but ‘knight’.
This knightly imagery really belongs to a later period than this book covers. All the same, to call oneself a miles was by the tenth century in some places a claim to status. Not yet in Saxony; milites are generally (even if not always) second-order figures in Thietmar. But, once again, in West Francia and secondarily Italy, by the later tenth century a miles was a significant player, and milites were establishing themselves as the lowest rung of the aristocracy, rather closer to counts than they were to the upper strata of the peasantry. This time, England goes in part with West Francia, for miles there, although still often representing quite humble soldiers, was also one of the standard Latin translations of thegn, the basic stratum of the late Anglo-Saxon aristocracy, and fairly comfortably off (every thegn was supposed to hold five hides of land, roughly 2 square kilometres, not a small amount of full landholding; Ælfhelm Polga, a rather more substantial landowner, seems to have been a king’s thegn). The lords of Uxelles, who were rather richer than this, were milites (and also, significantly, nobiles) around 1000; in Italy, a famous law of 1037 of Conrad II conceded to all milites the right to inherit benefices, given not only by kings but also by counts and bishops; although they could still lose them if they committed certain offences, the gap between full property and benefices was receding at the level of legislation too. In many parts of Italy, indeed, milites themselves came in two levels, capitanei and valvassores; even the latter could be socially prominent, and would form the ruling class of twelfth-century cities, but the former were certainly, by 1000 at the latest, political leaders by contemporary Italian standards.
What these processes mean is that, in practice, more people could by now be counted as what we call ‘the aristocracy’. By the Carolingian period, the word nobilis can effectively be translated ‘aristocrat’, its bearers marked out by wealth and lifestyle. It was by no means a legally defined category, but it denoted a rather restricted and special group, those with a great deal of property, those with Königsnähe, those who might expect counties. This was changing by the later tenth century, and milites who had rather localized lands, like the lords of Uxelles, by now could be called nobilis, could behave like richer aristocrats, and, increasingly, could be treated as near-equals by counts. This stratum of the lesser aristocracy was all the same closer to the peasantry, of course, simply because it was less rich than the great ‘imperial aristocratic’ families. ‘Military’ families might be the lesser branches of great aristocratic clans, or the descendants of vassals of Carolingian counts and bishops, but they might well also be descended from the medium landowners of the eighth century, locally prominent families with close connections to their peasant neighbours, who had stayed in the professional military arena. Milites were therefore also much more likely to be interested in local domination, for the local level was the one they were closest to. Many of the more detailed aspects of the seigneurie banale were pioneered by milites. This was reinforced by the emergence of a sharp division between the aristocratic/military class and the peasant majority, theorized already by King Alfred in the late ninth century, and extended considerably in early eleventh-century political writings in West Francia, as the difference between ‘those who fight’ and ‘those who work’. That sharp division marks the defining of an aristocratic stratum fundamentally distinct from the peasantry, which legitimated the local dominance of even quite small castle-holding lords. But all this also means that the local, castle-holding, seigneurie-building lords of the Mâconnais and other parts of West Francia and Italy in 1000, however aristocratic they by now saw themselves as being - and were seen by others as being - would have been regarded as of no account at all by a Merovingian vir inluster or a Carolingian ‘imperial aristocrat’. Not only membership of ‘the aristocracy’, but also the right to an independent political protagonism, was now extended to far more people, even if, still, only to a small proportion of the population at large.
Carolingian lords, just as in the period before 750, rewarded their military clienteles in different ways: by outright gifts of land, by hereditary leases, by revocable benefices. The difference between these latter was not always huge; unlike at the high aristocratic level, small-scale fideles and vassals might not be able to make their outrage felt if even their full property was confiscated by a count or bishop or abbot. As lay aristocrats, and also bishops and abbots, increased their land, they increased their entourages - their armies - by granting out more of it. In the tenth century, they would put their most prominent milites in charge of their castles and the local political powers associated with castle-holding. This would have been safe in the ninth century, because no miles could go it alone without facing ruin. In the later tenth century, however, when in some parts of Europe the ‘military’ stratum was gaining aristocratic identity and a sense of political protagonism, it was more risky. If counts could go it alone with respect to kings, castellans could also go it alone with respect to counts, as with the lords of Uxelles in relation to the counts of Mâcon. If a count or bishop lost control of his castellans, the whole framework of his power could unravel, and often did. Here, the ‘politics of land’ led firmly to political fragmentation of a most extreme kind - seldom before 1000, but often by 1050. The whole shape of politics could potentially change; the public world of the Carolingians might vanish, with nothing remaining in some areas except tiny private lordships.
This process has been called the ‘feudal revolution’ (or ‘mutation’) by many historians in recent years, and the issue has been sharply debated. Indeed, the ‘feudal revolution’ has become for some historians (particularly in France) shorthand for epochal change, the end of the ancient world itself in the most extreme formulations of the idea. The debate cannot be reprised here (it mostly has an eleventh-century focus), but some points can be made about it. One is that the catastrophist tone of many historians is out of place; the new ‘feudal’ world of the eleventh century may have been marked by more violence, for example, than its predecessor, but the difference was only in degree, not kind, as any reader of Flodoard’s Annals or Odo’s Life of Gerald (or, for that matter, the Annals of Saint-Bertin) will realize; military aristocrats of all types are always violent, and this did not change now that lesser milites were counted in. Another is, however, that there were real changes in some places, some of them very fast, as the Carolingian order was replaced by seigneuries; public assemblies finally vanished, relations of dependence became more prominent, power became more personal, even when it was in the hands of the same people. Comital power in a tenth-century autonomous county tended to have a very Carolingian format; but attempts to see a local seigneurie banale as simply the Carolingian political system writ small have not succeeded. As argued earlier, these shifts make the eleventh-century political world structurally different from the tenth, at least in the parts of Europe where they occurred.
Conv
ersely, this was not the case everywhere. Such shifts certainly did not occur in ‘outer Europe’, where no aristocrats were as yet sufficiently powerful, except in León-Castile. There is no sign that they were about to occur in 1000 in Anglo-Saxon England and Ottonian East Francia, and indeed, in the former, they never did (seigneurial-type powers in England were in the next centuries only held inside a lord’s own lands, and over unfree dependants). In most of what one can now call Germany, analogous processes were hardly beginning before 1100, and never had the form they took in what one can now call France. Even in Italy, where seigneurial fragmentation was often extreme, the continuing centrality of cities in most of the peninsula meant that there was always an alternative locus of political order, however informal, to that of the local lordship. In cities such as Milan or Lucca, the ‘military’ strata largely remained city-dwelling, even when, in the eleventh century, their rural lands acquired castles and seigneurial territories; this thus perpetuated a political community covering the whole city territory. We are left with France as the fulcrum of these ‘revolutionary’ changes. And not everywhere in France either, for, as we shall see in a moment, in Flanders, Normandy, Anjou and Toulouse, counts kept control of their castellans and of substantial elements of the Carolingian political pattern, into the twelfth century and beyond. The ‘feudal revolution’, particularly in its most dramatic form, as in the Mâconnais or, as authoritatively argued, in Catalonia, cannot be extended as a model to more than a minority of Europe, and not to large parts even of France.
The ‘politics of land’ is, it must be stressed, hard for kings and other lords to keep on top of. There is a potential zero-sum game, in which the more a king or lord grants out, the less he has to give, and the less attractive his patronage seems. Marc Bloch in 1940 called this ‘the fragmentation of powers’, and his phrase still works as an image. There is an underlying tendency to the break-up of larger political systems in favour of smaller ones, at least at the edges of the systems, and in extreme cases (as with tenth-century West Francia) even in the centre. But an underlying tendency is not an inescapable one. Merovingian and Carolingian - and Lombard, Visigothic and post-750 Anglo-Saxon - royal courts were unavoidable points of reference for all political power. Those who failed to get there or did not try were failures, those who went it alone without them seldom survived. Similarly, some counts in tenth-century West Francia could ride the tiger of fragmentation into smaller units still, by avoiding civil war, by policing their castellans tightly, by fighting successfully on frontiers and thus having booty and sometimes extra land to give to their milites, by keeping control of justice, by tying their military dependants to them with as ceremonious a set of ties as possible, and (perhaps above all) by using force as violently and as ruthlessly as they could against anyone who tried to defy them. At the very end of the tenth century, Fulk Nerra managed this in Anjou, Richard II in Normandy, Baldwin IV in Flanders; they successfully kept the balance of power firmly on the comital side, even as some of their neighbours failed to do the same. The Ottonians and the West Saxon kings found the same task rather easier. There was nothing inevitable about the ‘feudal revolution’.
Aristocratic status derived from a variety of elements: high birth, land, office, royal favour, lifestyle, the respect of one’s peers. No one theorized the relative importance of these elements; people ‘just knew’ how they balanced out, and different people had different views about their importance, or their applicability to individuals. When Thegan denounced Ebbo, archbishop of Reims, saying that Louis the Pious ‘made you free, not noble, which is impossible’ (cf. above, Chapter 17), he invoked an absolute criterion that was seldom put so sharply in this period. Ebbo may have been of servile origin; but other political players criticized for their ‘low birth’, like Hagano, Charles the Simple’s counsellor, in Richer’s words, or Willigis archbishop of Mainz (975-1011), Otto I’s former chancellor, in those of Thietmar, seem to have come from lesser aristocratic families, who may well have seen themselves as nobiles. There was not a noble ‘caste’, marked out by unbreakable rules of blood-line, as emerged in some parts of later medieval Europe; there was a grey area of negotiation, marked out by the snobbery of social superiors at every level. It was inside this grey area that milites in some parts of Europe began to take on aristocratic trappings, and to make claims to a status hitherto unavailable to them, which many by 1000 were prepared to grant them. But in order to do so they had to behave like their richer and better-established peers.
Aristocratic behaviour had in many respects not changed greatly from the period before 750, discussed in Chapter 8. Silk clothes with gold and silver decoration, military expertise, hunting, remained basic aristocratic markers, as did a ready use of violence - the markers that were implicit in Odo of Cluny’s characterization of Gerald of Aurillac. Odo refers to Gerald’s education in ‘the worldly exercises customary for noble boys’ - hunting, archery, falconry - but only enough literacy to read the Psalter (though an extreme bout of acne persuaded his parents in Gerald’s case to give him a fuller literary education, in case they had to make him a priest). The Carolingian educational programme seems to have already become rather weaker, if this story relates to Gerald’s youth in the 860s rather than to Odo’s own day, although Gerald was at best on the fringes of the Carolingian aristocracy, and also living in a remote area. All the same, Gerald’s Psalter reminds us that the aristocratic sense of innate virtue - a feature of this period, as earlier - was not just expressed through military valour and the like, but also through an (at least imagined) sense of a special religious charisma and commitment, as we shall see in a moment. Aristocrats were also supposed to be welcoming and generous, at least to their equals; Henry I before he became king of East Francia invited his neighbours to a wedding feast in Merseburg, according to Thietmar, and ‘treated them with such familiarity that they loved him as a friend and honoured him as a lord’. Whether the hilaritas, ‘jolliness’, praised in some narrative sources, was the same emotion as the drunken overbearingness to social inferiors criticized in the Life of Gerald is not clear, though it is likely often enough to have been the case. One of the key elements in the lifestyle of the aristocracy was indeed the potential violence to social inferiors that our sources constantly stress. This was taken for granted when one was dealing with the highest aristocrats; if it often seems that the increasing local power of the military strata in West Francia is associated with more complaints of violence than had been the case for the Carolingian ‘imperial aristocracy’, it is likely that this is not just because milites were establishing seigneuries by a liberal use of force, but also because they were not yet (particularly by vocal ecclesiastical victims) regarded as having a legitimate claim to the violent behaviour of more ‘noble’ figures. If so, however, they soon would.
We saw in Chapter 18 that a sense of a more dynastic family identity was stronger in the tenth-century aristocracy than formerly. This should not be pushed too far. Growing family rights to offices, and the consequential more prominent role of women as intermediaries between generations, are visible in the tenth century, at least in the highest aristocratic strata. But families were still fairly flexible entities; kinship ties of all kinds had gained in strength by now. Men and women were not tied to a single male-line lineage for their identity; surnames rarely existed yet. Thietmar pays almost as much attention to his mother’s kin, the counts of Stade, as to the counts of Walbeck on his father’s side. Furthermore, if maternal ancestors had a higher status than paternal, or more political purchase, they were often stressed more by their descendants, as when around 1012 Constantine, biographer of Bishop Adalbero II of Metz (d. 1005), stressed his descent from Henry I of East Francia, his mother’s mother’s father, above all; his father, Duke Frederick of Upper Lotharingia, by contrast does not have his ancestors listed, presumably because they were less distinguished. All the same, paternal kin, other things being equal, already mattered most; it was from them that most land would be inherited,
and with them that it would be - sometimes acrimoniously - divided. This would simply be accentuated as office-holding became less vital an element of aristocratic identity and land became more important.
Families continued to feud with each other, too. The imagery of faida, ‘feud’, or, more generally, bellum, ‘war’, frequently appears in narratives, as in the ‘Babenberger’-‘Conradine’ bellum of the 900s in the Middle Rhine, in which the Babenberger Henry was killed in 902, then his brother Adalbert killed Conrad, father of the future king Conrad I, before King Louis the Child was able to intervene, executing Adalbert in 906. All the same, not all these feuds reinforced patrilineal families: the murder of the leading Lotharingian count Megingaud in 892 was avenged on his killer Alberic in 896 by his widow’s second husband’s uncle. These were regional alliances fighting for supremacy, more than kin-groups expressing identity through honour-killings, even if the imagery of revenge was there, and was powerful. At the highest level, indeed, political rivalry could break families up; we saw at the start of this chapter how Ottonian patronage had split the Billungs. On a smaller scale, Thietmar himself found his paternal uncle Marquis Liuthar extremely unwilling to let him take over the family church of Walbeck in 1002, until Thietmar gave Liuthar a large pay-off (and also paid off his predecessor, whom Liuthar had put in), though this may simply be the product of tensions implicit in all inheritance divisions, which have indeed broken up many tight lineages in history, and certainly did so in the ninth and tenth centuries. Which is to say: we must not overstate family solidarity. Families could break up, and be redefined; family links were in any case only one available social bond, alongside personal dependence on kings and other lords, and political/factional alliances of other kinds. All the same, the imagery of kinship was important to aristocrats and widely used; it was kin who could choose whether to accept compensation for killings or continue the faida (as we see, for example, in Charlemagne’s capitularies; he sought to make compensation compulsory); family and kin bonds underlie all inheritance, much political strategy, and an increasing proportion of aristocratic identity.
The Inheritance of Rome: Illuminating the Dark Ages, 400-1000 Page 67