In memoriam Björn Weiler
Introduction
Imperial legacies are very much ‘in’. One cannot walk into a bookshop without being confronted by multiple titles on the subject. And in an age of culture wars, they regularly grace the pages of our broadsheets, often in polemical terms. To date, the Middle Ages have contributed – and been asked to contribute – little to these debates, which tend to focus quite tightly (for reasons as understandable as they are problematic) on modern European empires. Yet there can be no doubt that processes of conquest, colonisation and exploitation have a longer history. Over three decades ago, Robert Bartlett persuasively argued that we cannot understand the colonial practices of modern Europe without appreciating their medieval origins. As he put it, ‘[t]he European Christians who sailed to the coasts of the Americas, Asia and Africa in the fifteenth and sixteenth centuries came from a society that was already a colonizing society’.Footnote 1 It is not my purpose to retrace Bartlett’s arguments, nor to survey how they might be adjusted in light of recent approaches to race and ethnicity in the Middle Ages.Footnote 2 Rather, I wish to explore some of the ways in which medievalists may learn from our modernist colleagues when it comes to tracing the afterlives of empire.
One of the signal lessons of work on modern empires is how long, complex and contested their legacies frequently are. In strictly political and territorial terms, most European empires have long since ceased to be; socially, culturally and economically, they nevertheless live on.Footnote 3 The aim of the present study is to explore how far this was true of medieval Europe’s largest and most successful empire, that of the Carolingian Franks. In order to make a massive subject manageable, it focuses on imperial legacies as refracted through the prism of the royal charter (or diploma). This is a type of document recording grants and confirmations of legal rights, which can be found across all regions of the former Carolingian empire. Because of its nature as a sovereign instrument, the diploma had been central to the Carolingian project from the start, articulating a vision for the correct order of society and offering a mechanism for monarchs to insert themselves into local society and politics.Footnote 4 Research over the past three decades has emphasised the symbolic significance of these documents, which were not simply legal instruments (though they were this too), but also highly public demonstrations of favour. To receive a royal charter was not simply to come into possession of a set of valuable rights; it was to establish oneself in a position of power and influence on the highest political stage.Footnote 5 Precisely on this account, such documents were as important to the beneficiaries as the issuers – and indeed, their production must be understood in terms of dynamic interaction between court and locality, often extending to considerable recipient influence on their drafting and copying.Footnote 6
In temporal terms, my focus is on the years between 887/8 and the 1020s. These have been chosen to encompass the period between the effective end of the Carolingian empire, marked by the deposition and death of Charles the Fat (the last monarch to rule these domains in a unified manner) in 887–8, and the consolidation of the resulting kingdoms of France and Germany, a process largely complete by the reigns of Robert the Pious (996–1031) and Henry II (1002–24).Footnote 7 Neither date, however, is a hard one, for my aim is to identify a set of developments, most of which have their origins before 887 and many of which can be traced into the later eleventh and twelfth centuries. In geographical terms, my focus is on what Heinrich Fichtenau memorably called ‘the sometime Carolingian empire’ (das einstige Karolingerreich), with particular attention to the emergent kingdoms of France (West Francia), Germany (East Francia), Upper/Transjurane Burgundy and Italy.Footnote 8 Developments in the charter traditions of each of these regions have been studied in detail, but typically in isolation, reflecting the magnetic pull of national historiographical traditions, which becomes almost inexorable as we edge towards the central Middle Ages. By tracing developments across the frontiers of medieval kingdoms and modern scholarly traditions, I hope to show that there is a bigger story to be told here, one that demonstrates that, much like later imperial projects, the empire of the Carolingian Franks cast a long shadow. The basic point I wish to make is that diplomas across the former Carolingian empire underwent a series of shared changes in the century and a half or so after 888. The precise timing of these shifts varied, as did their meaning. But by 1025 royal charters across all these regions shared more with one another than they did with documents of the 880s. This was not inevitable, has been insufficiently appreciated, and demands explanation.
The Carolingian diploma
As in many other domains, the advent of the Carolingian dynasty in 751 occasioned significant changes in the form and layout of the royal charter. While visibly heirs to the documentary traditions of the Merovingians – their predecessors on the Frankish throne – the diplomas of the early Carolingians tread their own path. These were written on large rectangular sheets of parchment, with special elongated letters employed for the first line of text (protocol) and final subscriptions towards the foot of the document. The most significant innovations are to be found in these latter closing elements (known as the eschatocol). By the reign of Charlemagne (768–814), a monogram has been introduced into the royal subscription here, replacing the cross employed by his father Pippin (thus restoring the earlier practices of the Merovingians); the subscription itself is no longer autograph (i.e. written by the ruler himself), as it had been under the Merovingians; and the closing valediction (bene valete) has been dropped entirely. Other changes are slighter, but no less noteworthy. By Charlemagne’s later years, Merovingian chancery script has stabilised into what may be called a distinctive Carolingian variety of half-cursive. Elongated letters are also now no longer reserved for the royal style/superscription within the protocol, but rather are employed for whatever text naturally fits on the entire first line of the document (typically the verbal invocation, superscription and the start of the publication formula). Another shift lies in who was responsible for these texts: whereas the notaries of the Merovingian period were largely (perhaps exclusively) laymen, now clerical scribes are the norm.Footnote 9
In the early years of the ninth century, this new matrix established itself. The highwater mark is reached under Charlemagne’s son Louis the Pious (814–40), during whose reign a degree of consistency is achieved which rivals that of the bureaucratic writing offices of the twelfth and thirteenth centuries. The resulting documents are typically laid out horizontally (i.e. in landscape format) on sheets of parchment roughly 400–550 mm tall and 500–700 mm wide. The essential elements include an opening staff-shaped symbolic invocation (or chrismon) representing Christ, followed by a first line in elongated letters (litterae elongatae); a generously spaced main body, written in a consistent half-cursive hand distinct from the Caroline minuscule bookhand increasingly used for other purposes; a royal/imperial subscription immediately below this, written in elongatae and incorporating the ruler’s own monogram; offset to the right of this (typically somewhat lower) and introduced by another symbolic invocation, a notarial recognition/subscription, also in elongatae, in the name of the scribe responsible for checking and approving the final text (typically a different individual from the notary of the main text), accompanied by a recognition/subscription sign signalling this assent graphically (and sometimes giving further details regarding the charter’s production in Tironian notes, an antique shorthand); the ruler’s seal to the immediate right of the recognition sign; and finally a dating clause along the foot of the parchment, detailing when and where the document was issued (Figure 1). There is some variation in execution, and different types of privilege might take different forms – the most exalted ones are sealed with gold or lead bulls and might bear a red legimus mark, while simple precepts could dispense with the royal/imperial subscription – but what impresses above all is the consistency achieved.Footnote 10 Despite the large number of notaries involved in this work – at least sixty-two under Louis the Pious, according to the recent critical editionFootnote 11 – there was a clear sense of the rules and how to execute these. In this respect, the most distinctive elements of the charter are the symbolic invocation and first line in elongated script (protocol) and the closing elements (eschatocol). Like the former, the latter are written largely in elongated letters, with the exception of the dating clause.
Figure 1. A diploma of Louis the Pious (21 April 839): D L Fr 397, Karlsruhe, GLA, A Nr. 6.
For many years, this was the visible face of royal authority. Significant shifts can be observed starting in the 850s, particularly within the East Frankish/German realm, where a new form of chrismon, new approach to the notarial recognition (now not necessarily autograph and sometimes immediately below the royal subscription), and new type of script (diplomatic minuscule) were introduced by the notaries known as Hadabert and (especially) Hebarhard.Footnote 12 But for all the novelty of these forms, the overwhelming impression created across the middle years of the century is of a gentle evolution and adaptation of earlier norms, rather than an attempt to invert these at a fundamental level. In order to trace the further development of these forms, I will now focus in turn on developments in West Francia/France, Burgundy, Italy and East Francia/Germany. Because my focus is on the visual and symbolic elements of these documents, only those which survive in their original format are relevant for these purposes. Modern facsimiles and digital reproductions, supplemented by my own (admittedly unsystematic) archival forays, suffice to establish broad trends, but further work will be needed to establish precise regional contours.
West Francia/France
The West Frankish realm (which for convenience I will often simply call France) is traditionally seen as the most innovative and dynamic region in these years. In part, the changes here are a function of a shift in how diplomas were produced. While we know less about charter production under Charlemagne and Louis the Pious than we would like, a clear centralising element is present. Whether the sixty-two distinct hands identified among the latter’s charters were all those of members of a formal writing office (viz. ‘the chancery’) may be doubted, but they were evidently well aware of one another’s work and capable of producing documents of a remarkably high consistency. Beneficiary production – the drafting and copying of charters by the institutions (most often religious houses) receiving them – can be identified in isolated cases, but this represents the exception, not the rule.Footnote 13 As we move into the middle years of the century, however, recipients start to figure more prominently in the record; and by the reign of Charles the Fat (876–87/8), who added France to his German and Italian domains in late 884, beneficiary production becomes the norm here.Footnote 14 This should not be seen as a failing. For a region in which the ruler was rarely present, but whose religious houses could produce documents of the requisite standard (such as in France under Charles), it made more sense to let the recipients take the lead than to try to maintain a cadre of perennially underemployed royal scribes. The effect, however, is that the charters start to show a marked regional quality, albeit with certain overarching trends.
The key development in this respect is a move from the traditional horizontal (landscape) layout towards a vertical (portrait) one, bringing the diploma more closely in line with the ‘private’ charters issued by individuals and institutions beyond the monarch. Vertical documents had been issued by Charlemagne and Louis the Pious upon occasion, but were extremely unusual before the middle years of the century. Under Charles the Bald (840–77), their numbers had been slowly growing; and there are other signs that the ‘canon’ established by Charlemagne and Louis was starting to loosen. These trends become more marked under Louis the Stammerer (877–9) and his sons, when horizontal orientation starts to win out; and they are maintained under Charles the Fat (884–7/8 in West Francia), well over half of whose surviving originals for French recipients are so-called cartae transversae (vertically oriented charters).Footnote 15 Another important novelty is the introduction of a form of script hierarchy. This may at first sound surprising, since the elongated letters employed for the protocol and eschatocol already offered a hierarchy of sorts, drawing attention to the opening and closing elements of the privilege. Hitherto, however, the same script had always been used for all details within any given section. What we now start to see, by contrast, is the occasional employment of majuscule forms within the dating clause to draw attention to the ruler’s name here. A good example is offered by Charles’s diploma of 28 August 885 for Langres, which was produced by a local notary operating under the auspices of the well-connected Bishop Geilo (Figure 2).Footnote 16 The inspiration for this almost certainly came from the tradition of writing the final amen of the apprecatio (a phrase such as in dei nomine feliciter amen at the end of the dating clause) in majuscule forms, which can be observed from the middle years of the century.Footnote 17 The effect, however, is a new emphasis on royal majesty. This is not the only sign of change within the eschatocol. In this diploma, as in Charles’s other West Frankish privileges, the symbolic invocation has been abandoned before the notarial recognition, suggesting that this was no longer considered a distinct element from the preceding royal/imperial subscription.
Figure 2. The eschatocol of Charles the Fat’s diploma for Langres (28 August 885): D K iii 129, DDK vii, pl. 4.
Following Charles’s deposition and death in 887–8, the West Frankish magnates chose the well-regarded count of Paris, Odo (888–98), as his successor. Odo was the first non-Carolingian to rule the region in over a century. Precisely because his accession represented a dynastic break, his court consciously sought to assert continuity.Footnote 18 In terms of the documents issued in his name, this meant a continuation and intensification of existing trends. Already under Charles, the simple precept (a streamlined version of the diploma, issued without royal subscription) had started falling out of use; now it becomes all but obsolete. The overall rate of diploma production also decreases, continuing trends visible since the death of Charles the Bald in 877.Footnote 19 In terms of format, horizontal documents continue to be issued, but vertical orientation predominates. Partly as a consequence, we often see the notarial recognition placed below the royal subscription, on an entirely different line. The seal is then placed atop the recognition sign (rather than to its right), a practice visible since the 870s; and the notarial recognition itself is often not autograph (i.e. undertaken by the notary named there). This is not the only sign of loosening conventions governing the eschatocol. In one of Odo’s earliest diplomas, elongated letters have been dropped for the recognition, which is simply written in the same half-cursive as the main text. From the mid-ninth century, we sometimes see the notarial recognition written in smaller elongatae than the royal subscription; this seems to represent a similar attempt to distinguish the two elements graphically. At least in this case, there was still a desire to underline the notary’s role, since his name is written in rustic capitals (IRANNUS NOTARIUS).Footnote 20 In two other cases, by contrast, the entire dating clause has been written in majuscules (for one of these, see Figure 3).Footnote 21 The latter practice presumably derives from a desire to draw attention to the dating clause in a similar manner to the subscription and recognition; and in an earlier diploma of 30 December 889, the dating clause is rendered in elongated letters.Footnote 22 In a diploma of October 893, meanwhile, Odo’s own name is highlighted within the dating clause by the use of rustic capitals, much as we have seen in Charles’s Langres privilege.Footnote 23 We also sometimes start to see the dating clause migrate up the parchment, following immediately on from the subscription and recognition, rather than being placed at the foot of the document.
Figure 3. The dating clause of Odo’s diploma for Count Richard (21 October 897): Recueil Eudes, no. 42, DDK vii, pl. 17.
By Odo’s later years, a faction led by Fulk of Reims was backing the claims of Charles the Simple (898–922) to the throne. The latter was a full-blooded Carolingian (a posthumously born son of Louis the Stammerer), as his name proudly proclaimed, and Odo was apparently only able to defuse the threat by promising Charles the succession upon his death (conveniently, Odo had no sons of his own). From a documentary standpoint, what is interesting is that Charles’s accession in 898 saw something of a reassertion of the traditional Carolingian charter matrix: horizontal layout once more becomes the norm and the treatment of the closing eschatocol starts to conform more closely to traditional practice (elongatae only for the subscription and recognition; the latter offset to the right of the former). Diplomas of the newer vertical type do not disappear overnight; and the result is not so much a sharp break, as a gentle change of course: while eleven of Odo’s seventeen surviving originals had been laid out vertically, under Charles this proportion reduces to seven out of twenty-seven. If we break the distribution down across Charles’s reign, however, the trend becomes more marked: three of Charles’s first six originals are laid out vertically, proportions broadly similar to those seen under Odo; thereafter, just two of twenty-one are.Footnote 24 Charles framed his reign as a reassertion of the best traditions of his Carolingian forebears; his charters helped proclaim this message.Footnote 25
Still, developments under Odo would have a future. Following the deposition of Charles and the brief reign of Odo’s younger brother Robert I (922–3), the latter’s son-in-law, Raoul of Burgundy (923–36), succeeded to the throne. Raoul’s reign saw vertical layout win out decisively. If these forms were in some sense associated with Odo and Robert, they also take forward trends already present under Charles the Fat. Raoul’s first surviving original, a confirmation for Langres of 30 May 927, is strongly vertical in layout, and the remaining two take similar forms.Footnote 26 Yet this is not the only change. In Raoul’s diplomas, we can start to identify the influence of the distinctive diplomatic minuscule script long favoured for charters in the east. The latter is in effect an adjusted Caroline bookhand, so the appeal of such forms to scribes better acquainted with book production is understandable.Footnote 27 We can also observe a further loosening of conventions governing the eschatocol. Thus, in Raoul’s first two (of three) surviving originals, both the royal subscription and notarial recognition are in half-cursive, rather than the elongatae we might expect. This cannot be a result of scribal incompetence, for elongated letters are still found in the opening protocol; rather, it reflects increasing uncertainty about the nature and purpose of the distinct elements of the eschatocol. Another significant change is the abandonment of the recognition sign, which had been a feature of diplomas since the Merovingian era. Already under Charles and Odo, this had started to lose its original meaning as a symbolic representation of the Latin subscripsit, ‘has subscribed’: it was often no more than an empty arch under which the seal was placed, and there is little evidence for the continuing use of Tironian notes here.Footnote 28 A final notable feature of Raoul’s diplomas is his monogram. This is formed in the same manner as those of Odo and Charles, the ultimate model being Charlemagne’s distinctive cross monogram. Yet the proportions are entirely different, with Raoul’s monogram often more than twice as large. This can be ascribed to local Burgundian influence, for we see something similar in the diplomas of Raoul’s neighbours in Upper Burgundy.
The reign of Louis IV (936–54), the exiled son of Charles the Simple who succeeded Raoul in 936, witnesses a partial return to the traditional forms. While vertical layout remains more common than it had been under Charles, there is an unmistakable effort to return to the classic matrix, with four of Louis’s seven surviving originals laid out horizontally. Yet if Charles’s charters had largely succeeded in restoring earlier conventions, Louis’s display a freer mix of the old and new. In all four of the horizontally oriented privileges, the notarial recognition is rendered in minuscule forms, leaving only the protocol and royal subscription in elongated letters.Footnote 29 As previously, scribal inability cannot be responsible for this; and, indeed, in two of these diplomas elongatae are still used for the name of the notary within the recognition. It is also in these years that we first start to see a pure diplomatic minuscule of the East Frankish (Hebarhardian) variety used to write these documents, reflecting wider East Frankish influence on Louis’s court (Louis was married to Gerberga, the sister of the East Frankish ruler Otto I).Footnote 30
Under Lothar (954–86) and Louis V (986–7), vertical orientation is firmly established as the norm for diplomas and diplomatic minuscule becomes increasingly common for the main text.Footnote 31 These years also see the rate of charter production reach its lowest point, reflecting the wider political travails of the last Carolingian rulers.Footnote 32 Further developments can be witnessed in the treatment of the final dating clause. Earlier diplomas had sometimes seen this wander up the parchment; and the occasional extension of elongatae to this element under Odo already heralded a degree of uncertainty as to its status. Now the dating clause is sometimes repositioned above the subscription and recognition; in one case, it is dropped in favour of a lapidary phrase (in elongated letters, immediately below the royal subscription and notarial recognition) announcing the place (but not date) of issue.Footnote 33 These years also manifest important changes to the royal seal. Hitherto, all French rulers had employed antique gem seals on the Carolingian model. Under Lothar, by contrast, we see the adoption of a majesty seal, presenting the monarch en face and with regalia. This is first attested in May 966 and its adoption was clearly a response to the introduction of a similar seal by Otto I upon his imperial consecration in February 962. As Otto’s nephew – the son of Gerberga with Louis IV – it is hardly surprising that Lothar should be among the first to have got wind of these developments. These forms would thereafter be employed by all of Lothar’s successors down to Robert the Pious.Footnote 34
Following the definitive eclipse of the Carolingian line in the west in 987, developments become harder to trace in the absence of critical editions of the relevant documents. The accession of Hugh (987–96) signalled a rupture here no more than that of his great-uncle Odo a century previously. Much like Odo’s diplomas, Hugh’s acts are at pains to present the king as the natural heir to his Carolingian forebears.Footnote 35 This is not, however, to say that they are identical to those of his predecessors. In particular, the form of the closing eschatocol continues to evolve in interesting and important ways. In a diploma for Fleury of 993, the dating clause has moved into the main text, preceding the royal subscription.Footnote 36 Of Hugh’s five remaining originals, one lacks a dating clause; two have the dating clause integrated into the subscriptions, just above the seal; and the final two have it in the traditional position at the foot of the act (in one case, in elongatae, in the other, in diplomatic minuscule).Footnote 37 Hugh’s first four privileges retain notarial subscriptions/recognitions, but these have disappeared in his final two. This latter shift is accompanied, but probably not occasioned, by the introduction of the subscription of Hugh’s son Robert. Though Robert was king in name only before his father’s death in late October 996, he had been consecrated on Christmas Day 987, in a move to secure the future of the fledgling dynasty. Such double subscriptions had previously been employed by Louis the Pious with his eldest son Lothar I; more recently, they had been used by the West Frankish Lothar, whose son and heir Louis V had been associated with his rule in a similar fashion.Footnote 38 Finally, there are signs that conventions governing the protocol were also now starting to loosen. One of Hugh’s originals dispenses entirely with elongatae; and it is perhaps no coincidence that this is also the only diploma to forego a dating clause.Footnote 39
Under Robert the Pious (996–1031), diplomas undergo more dramatic shifts. As Geoffrey Koziol notes, the most striking of these is the introduction of third-party subscriptions at the foot of the document; we also see a new form of enlarged royal monogram, in which the cross features prominently. These shifts do not take place upon Robert’s accession, but rather coincide with the conquest of Burgundy (initiated in 1003–5 and completed in 1016 and 1030–1), reflecting the fraught circumstances of these events.Footnote 40 Indeed, the obvious precedent for Robert’s new monogram is that of Raoul of Burgundy; and as in that case, local Burgundian influence was presumably at play. In part, these changes reflect a further borrowing of conventions from ‘private’ charters. It was customary to have the transactions of local magnates and institutions subscribed by third parties; and as the distinction between royal and princely authority diminished, it is hardly surprising that we should see a convergence between the two documentary traditions here.Footnote 41 Yet the circumstances under which these practices were introduced suggests that they are not a straightforward product of royal weakness (though this was, perhaps, the necessary precondition). Rather, these shifts seem to have been part of an effort to present a more consensual face to kingship, emphasising the active participation of the kingdom’s great and good in politics by means of their subscriptions to important acts of state. In strictly diplomatic terms, the ground had been prepared by the loosening of conventions governing the eschatocol over the preceding half-century. Moving the dating clause up had created space at the foot of the charter; and the integration of Robert’s subscriptions into his father’s acts (in which he often had only passing involvement) may already have suggested the potential for including individuals beyond the ruling monarch.
These were, however, not the only changes. Rather, we see a complete rethinking of the matrix of the royal charter, with new diplomas laid out in ways that barely nod to earlier traditions. In a number of documents, all pretence of diplomatic minuscule or half-cursive is abandoned in the main text in favour of a simple Caroline bookhand. In these texts, elongated script is generally dropped for protocol and eschatocol. In some cases, this is replaced by simple majuscules; in others, all effort to emphasise these elements is given up. In one case, the monogram itself has been omitted within the royal subscription in favour of a simple (apparently autograph) cross.Footnote 42 Even when present in the opening protocol, elongatae are frequently absent from the royal subscription and notarial recognition, whose forms are increasingly assimilated to those of other third-party subscriptions – indeed, the former often just takes the form of monogram and seal.Footnote 43 In a number of cases, the notarial recognition disappears entirely, as we saw under Hugh. Robert is also the first French monarch to make a habit of appending his monogram and seal to the acts of other individuals and institutions as a sign of his assent (previously this would have required a separate charter of confirmation), a practice which becomes more widespread under his son and successor Henry I (1031–61).Footnote 44 Aligned with these moves is the introduction of a script hierarchy into the body of the diploma: now for the first time we start to see the name of the monarch (Robert) and other important players (the recipient, patron saints) highlighted here by the use of display script.Footnote 45 Previously, this had only been seen in passing in the dating clauses of Charles the Fat and Odo, though trends in this direction can be observed in other regions. The result of these shifts is not so much a reduction in royal majesty, as its recalibration: the monarch remains centre stage, but within a more open and fluid documentary tradition.
Upper/Transjurane Burgundy
Despite the relatively small corpus of originals, France thus presents us with a picture of dynamic change. Having established the contours of developments there, we may make quicker work of Burgundy, Italy and Germany. For the first of these regions, we possess a similarly good run of facsimiles, albeit covering an even smaller corpus and geographical area. As we might expect from its location, Upper (or Transjurane) Burgundy stands somewhere between France, Germany and Italy in terms of developments: it shares some of the conservativism characteristic of the latter regions, while still participating in many of the trends visible in the former. The essentials are already clear in the earliest surviving diploma of Rodulf I (888–912), issued on 10 June 888 in favour of his sister Adelheid (alias Adelaide).Footnote 46 This diploma is laid out in traditional horizontal fashion, with elongatae for protocol and royal subscription. The notarial recognition is not preceded by a symbolic invocation and is rendered in the same minuscule forms as the main text; and the final dating clause is placed notably above the foot of the parchment – both developments we have seen in French documents of these years. The script of the main text, recognition and dating clause retains half-cursive features, but is moving towards what can better be termed diplomatic minuscule. Most later documents of the tenth century retain elongated script for the notarial recognition, but otherwise follow the matrix established here. The one significant development is that the recognition sign, still present in this document (beneath the seal), disappears soon thereafter. Otherwise, the most distinctive feature of the Burgundian diploma tradition – beyond the modest number of documents issued – is the use of enlarged monograms by rulers. The basis of these is the cross-monogram developed under Charlemagne and used by all French rulers from Charles the Bald on. However, the dimensions are now stretched to produce signs two to three times as large, a feature we have met in passing in the diplomas of Raoul (who hailed from the neighbouring duchy of Burgundy) and Robert the Pious. There are some signs of a move towards vertical orientation, as a few tenth-century diplomas are almost square; and the vertically arranged imitative copy of a privilege of 9 March 968 may reflect the dimensions of the lost original.Footnote 47 Compared with what we see in France, however, these documents remain remarkably conservative. A striking contrast is presented by the diplomas of neighbouring Provence (itself an independent kingdom until incorporated into Burgundy by Rudolf II in 933), which typically take the vertical forms seen in France.Footnote 48
By the reign of Rudolf III (993–1032), however, notable shifts become evident within the Burgundian charter tradition. Already in an unsealed privilege of March 994, elongatae have given way to simple majuscules in the protocol and notarial recognition (though they are retained for the royal subscription); these majuscules have also been extended to the dating clause, which follows immediately on from the recognition (rather than being placed at the foot of the parchment). The main text itself is written in a standard Caroline hand, rather than diplomatic minuscule, a feature found in another privilege of two years later.Footnote 49 Much as in France, the distinction between charter and book script seems to have been falling out of use (or fashion) and many other notaries merely give a nod to the tradition of diplomatic minuscule by extending the ascenders and descenders of what remains otherwise a simple bookhand. Even diplomas true to the traditional matrix betray significant shifts. In Rudolf’s privilege of 1011 in favour of his betrothed Irmengard, the elegant diplomatic minuscule of the main text is also employed for the notarial recognition, where we would normally expect elongatae. The parchment itself is almost square, signalling a move towards verticality; and within the dating clause, Rudolf’s own name is rendered in majuscules, reflecting new experiments with script hierarchy.Footnote 50 Another noteworthy feature is the diploma’s seal. Whereas Rudolf I had employed a gem seal of antique vintage, in keeping with Carolingian tradition, and Rudolf II’s seal is unknown, Rudolf III is presented en face with regalia. The ultimate model is clearly the imperial seal of Otto I, which had found such swift imitation in France; and in fact, an antiquarian description of the (otherwise unattested) second seal of Rudolf III’s father, Conrad (937–93) – who had been fostered at Otto I’s court – reveals that the move to this iconography had already taken place by September 967, around the same time as in France.Footnote 51 The diploma in question is in all probability the work of the ‘Pandulf the chancellor/notary’ (Pandulfus cancellarius) named in the notarial recognition, whose hand is found in three other charters of these years.Footnote 52 This was, therefore, not an isolated or eccentric performance. And though none of Pandulf’s other surviving originals share these features, they are found in more developed form in a diploma of 1029 in favour of Cluny (Figure 4). This privilege is strongly vertical in orientation and employs a consistent script hierarchy, with elongatae for the protocol and royal subscription, diplomatic minuscule for the main text and recognition, and rustic capitals within the former for the name of Queen Irmingard, to whom the 1011 diploma had been issued. The hand itself reveals close similarities with that of Pandulf and is almost certainly to be ascribed to the ‘unworthy priest Albker’ (Albker indignus presbyter) named in the recognition, who has been identified as one of Pandulf’s most able and influential students.Footnote 53 These trends are also visible in the work of Hubert, another of Pandulf’s students, whose hand is attested in documents of 1025 and 1029: the first of these is laid out in traditional landscape fashion, but the second is portrait; and in both, majuscules are employed for Rudolf’s name within the final dating clause.Footnote 54 One further feature is worthy of note: in the work of Hubert, the dating clause has started to merge with the notarial recognition, which now frequently follows the date in a manner similar to that seen in France.
Figure 4. Rudolf III’s diploma for Cluny (14 January 1029): D Rudolf. 120, DDK ix, pl. 24.
Kingdom of Italy
For Italy, the evidence is richer than for Burgundy, even accounting for the much patchier coverage of modern reproductions. The political scene in many respects resembles that in France, with multiple dynasties vying for power and influence. But while the situation stabilised in the latter region by the 940s, in Italy it would take external conquest by Otto I in 962 before a single family could establish itself securely on the throne. And while Italian documentary culture was very much sui generis, it shared one essential feature with its French and Burgundian counterparts: it boasted a good supply of recipient scribes capable of producing diplomas of a high standard.Footnote 55 From the beginning of Charles the Fat’s reign (879–87/8 in Italy), he relied heavily on beneficiary notaries from the region, typically men recruited from the entourages of the leading bishops. This reflects an intensification of trends already visible under Louis II (844–75), who had made considerable use of recipient and occasional scribes.Footnote 56 Significantly in this regard, we already encounter vertically oriented charters in Charles’s early years in Italy. Recipient production and vertical orientation are clearly connected here, much as they were in France, for portrait layout had long been standard for the local (‘private’) charters with which recipient scribes were better acquainted. It is, therefore, no coincidence that one of our earliest examples should be a beneficiary product of December 880 in favour of the monastery of Tolla in Emilia.Footnote 57 A similar picture is presented by two rough-and-ready privileges of July 883 concerning the bishopric of Bergamo, both of which take vertical forms and were produced by similar hands, apparently those of local clerics.Footnote 58 Other developments can be identified in these documents. In none is the traditional symbolic invocation found before the notarial recognition; and in the first (for Tolla) the recognition itself has moved from being offset to the right of the royal subscription to immediately below this on a different line. The Tolla privilege also boasts a two-line dating clause, of the kind we occasionally see in France and Burgundy, while in the second Bergamo privilege Charles’s name is rendered in majuscules here. These documents are all written in what may be termed a distinctive Italian take on diplomatic minuscule, rather than the older half-cursive forms; and in the Tolla diploma, the recognition sign is omitted in a manner familiar from France and Burgundy.
Yet if it superficially looks like Italian diplomas are moving in the same direction as their French counterparts, closer examination reveals greater adherence to the traditional matrix. Even under Charles the Fat, horizontally oriented privileges remain the norm; and they would continue to do so into the 970s. A sense of developments is offered by the privileges preserved in Bergamo: Charles the Fat’s two diplomas of July 893 may have been strongly vertical in layout, but that of his successor Arnulf of Carinthia for the bishopric (early 894) is effectively square; and Berengar I’s privilege of 898 for Ermenulf (apparently the count of Stazzona) takes traditional horizontal forms once more.Footnote 59 This return to the classic layout coincides in part with a change in the agency responsible for these acts, for the Arnulf privilege is modelled closely on the work of a court notary (Aspert C), while the Berengar diploma is the work of a court hand (Marziano A); and it is interesting to note that vertical orientation returns in a later privilege of 904 for the see, for which the recipients were responsible.Footnote 60 It would be wrong, however, simply to equate landscape layout with what we might once have called the ‘chancery’. The Arnulf diploma is the work of a local scribe imitating a court hand, while examples of vertical orientation can also be found in the work of court notaries (including a later privilege of Berengar for the see).Footnote 61 In this respect, developments in Italy stand somewhere between those in Burgundy and France: vertical orientation established itself early, as in the latter region, but horizontal forms remained the norm, as in the former. We also see a similar simplification of the old Carolingian system of privileges of differing degrees of formality, with simple precepts disappearing almost entirely, save for grants and confirmations of royal protection.Footnote 62 More specific trends can be identified, with portrait layout enjoying particular popularity under Guy of Spoleto (888–94) and his son Lambert (894–98), while being notably less favoured by their rival Berengar I (888–924): four of Guy’s eleven’s originals (a significant minority) are vertically oriented, whereas just four of Berengar’s sixty-nine take these forms.Footnote 63 The evidence does not permit as straightforward a political reading as it does in France, but there may be some significance to these trends. For in contrast with Guy and Lambert, Berengar enjoyed a direct link to the Carolingian family in the form of his mother Gisela (a daughter of Louis the Pious) and sought to emphasise this at every turn in his diplomas.Footnote 64
The main development over the first half of the tenth century is a progressive loosening of conventions governing the eschatocol. Already in Charles the Fat’s Tolla diploma, we saw that the recognition sign might be omitted; and this becomes increasingly common in the following years. Signs are still attested into the 950s, but are clearly starting to lose their function: they are sometimes displaced from their original position at the end of the recognition and often left simply as an unfilled arc, under which the seal might be placed (much as in France).Footnote 65 The Tironian notes which used to fill these are now rare: they are attested under Berengar I, Guy and Lambert, and Louis of Provence (900–5), but only in a limited range; and they disappear entirely by the second decade of the tenth century.Footnote 66 The recognition clause itself normally occupies the traditional position offset to the right of the royal subscription, though the kind of two-line subscription-recognition seen in Charles the Fat’s Tolla diploma is a common alternative. In the 880s and 890s, we still occasionally find symbolic invocations used to introduce the recognition, but these are progressively abandoned, much as they had been north of the Alps. Likewise, while the recognition itself may be autograph, it is often now completed by other (anonymous) scribes on behalf of the named notary. Elongated letters are still typically employed for both the subscription and recognition, but with the latter taking notably smaller forms. This practice of distinguishing the two elements graphically is attested across the Frankish domains from the mid-ninth century, but achieved particular popularity south of the Alps in these years. Here scribes go to such efforts that we sometimes encounter comically exaggerated forms for the royal subscription; in other cases, it can be difficult to discern whether the notarial recognition is in elongatae at all.Footnote 67 In a small number of cases, we see elongatae dropped for the recognition, in a manner similar to that seen in France and Burgundy.Footnote 68 Related to these shifts is a move towards reducing the proportions of the protocol, bringing the elongated letters there in line with the reduced dimensions of those in the recognition (in contrast to the longer forms reserved for the royal subscription). We can also discern changes in the opening symbolic invocation. While this typically continues to take the form of the tried and tested staff-shaped labarum, periodically the C-formed chrismon first developed by Hebarhard in East Francia is found as an alternative. Finally, similar variation is visible in the treatment of the final dating clause. From an early date, we find majuscule forms employed here, at least for the final N on amen within the apprecatio. In a distinctive charter of protection (mundiburdium) of 912, this is introduced by a decorative cross, apparently on the model of the recognition (which was traditionally preceded by a symbolic invocation).Footnote 69 This suggests a degree of uncertainty as to the distinction between these elements; and in two diplomas of Lothar II (947–50), the son and successor of Hugh of Provence (926–47), we find the entire dating clause in majuscules, much as we occasionally see in France (for one of these: Figure 5).Footnote 70
Figure 5. The dating clause of a diploma of Lothar of Italy (19 May 947): D Lo ii 2, API 31.
In later years, we can observe further variation. In a confirmation of Berengar II (950–61) in favour of S. Michele in Barrea of October 953, diplomatic minuscule and elongatae are abandoned in favour of a Caroline bookhand, by what is apparently a recipient notary. Script is not the only oddity here: the parchment itself is irregular and contains two large holes, around which the scribe has had to write. A diploma of a few months earlier, produced by the hand which had supplied the protocol and eschatocol of the S. Michele privilege (apparently that of a court scribe), only makes a marginally better impression: the parchment is similarly irregular, displaying a number of small holes; and while the hand is more assured, it displays an uncertainty rarely seen in earlier years.Footnote 71 Berengar’s other diplomas conform more closely to the established matrix, so it is hard to imagine that such variations were intentional.Footnote 72 More notable, perhaps, are experiments with script hierarchy. In the last surviving original of Hugh and Lothar, a donation of February 946 for the canons of Piacenza, the name of the see’s patron saint (Anthony) is highlighted in majuscules (eclesia beati ANTONINI). Likewise, in one of Lothar’s diplomas of the following year for the nunnery of Senatore in Pavia the name of the house’s abbess (Irmengard) is written in Rustic capitals (ERMENGARDA venerabilis abbatissa) (Figure 6).Footnote 73
Figure 6. A section of Lothar’s diploma nunnery of Senatore (23 September 947): D Lo ii 4, API 23.
East Francia/Germany
Following Otto I’s conquest of Italy in 961–2, the Italian diplomatic tradition becomes part of the East Frankish/German one, so it is to the latter region that we must now turn. Superficially, Germany presents us with the most stable picture in these years. East Francia was the one region to elect a bona fide Carolingian in 887/8 (albeit an illegitimate one), in the form of Arnulf of Carinthia (887–99). And while the dynasty was eclipsed upon the death of Arnulf’s son Louis the Child in 911, the result was not the kind of fierce competition seen in France and Italy, but rather the establishment in relatively quick succession of two new dynasties, the Conradines and Liudolfings (or Ottonians). The retrospective nature of our narrative sources doubtless does much to obscure the challenges of these years. But it remains true that by the 940s royal authority had been reasserted more successfully in the east than in any of the other of the constituent parts of Charles the Fat’s empire, a fact reflected in the high (and growing) rate of diploma production.Footnote 74 That these documents should reveal a high degree of continuity in layout and appearance is therefore natural; and, indeed, they find their closest analogues in Burgundy, where similar political stability is visible. Still, it would be wrong to confuse continuity with stasis, and a number of subtle but important shifts can be observed in these years.
Unlike his privileges for France and Italy, Charles the Fat’s charters for the East Frankish realm are largely the work of court notaries. And even where the recipients take the lead (as at centres such as St Gallen), they tend to stick closely to the established East Frankish matrix: horizontal layout remains the norm and there is little experimentation with the subscription and recognition.Footnote 75 Yet Charles’s scribes not only inherited the forms created by Hadebert and Hebarhard; they also developed them further. One of the most important innovations in this respect is to be found in the dating clause. Alongside diplomatic minuscule, Hebarhard had introduced the use of majuscules for the day of the month and/or final amen of the apprecatio here, a practice maintained under Charles and his successors. What is new now, however, is that Charles’s own name starts to be rendered in majuscules here. These forms are not seen under his father or elder brothers and can be found in all three of Charles’s earliest originals; and it was presumably such documents which furnished the model for the scribes of the Langres and Bergamo diplomas of the 880s, who treated his name similarly, as we have seen.Footnote 76 More specific trends can also be identified, with the use of majuscules for the ruler’s name enjoying particular popularity in diplomas for St Gallen, a house with close ties to Charles’s Alemannian court.
Under Arnulf, we see the geographic focus of the court switch from Alemannia to Bavaria, but the face of the royal authority remains largely unaltered: horizontal (or at most square) orientation remains the norm for diplomas and the broad lineaments of the Hebarhardian matrix are maintained. Majuscules often continue to be employed for the ruler’s name within the dating clause and there are signs that this approach was starting to influence the treatment of personal names within the main text. Thus in a privilege of 21 November 889 for Würzburg, not only is Arnulf’s name rendered in rustic capitals in the dating clause, but the same treatment is extended to Bishop Arn in the main text; the same is true of Count Arnulf of Duria in a privilege of 15 May 898.Footnote 77 Thereafter, the use of display script in the main text would remain a rare but occasional variation on the established canon, appearing in at least one diploma of Louis the Child (899–911).Footnote 78 Under Conrad I (911–18), Henry I (919–36) and Otto I (936–73), this practice remains rare, with majuscules tending to appear for the ruler’s name in the dating clause, if at all. That experiments with script hierarchy had not been abandoned is revealed by Otto’s privilege of October 936 for Corvey, in which the names of the monastery’s patron saints, Stephen and Vitus, are rendered in elongatae in the main text.Footnote 79 In two diplomas of late 958, produced by the distinctive Lotharingian draftsman-scribe Otpert, we then encounter elegant rustic capitals for selected personal and place names within the body of the charter (for the second of these, see Figure 7).Footnote 80 The other section to witness significant evolution is the eschatocol. Here the recognition sign is progressively separated from the recognition clause and sometimes found to the right of the seal, completing the process of disassociation started by Herbarhard. As part of this, Tironian notes become increasingly rare: these are attested into the 940s, but in most cases they seem to have been copied from earlier documents and active composition in this demanding tachygraphic script can no longer be securely attested.Footnote 81
Figure 7. A section of one of Otpert’s diplomas of 958 (2 December 958): D O i 198, Magdeburg, LA Sachsen-Anhalt, U1, Ia Nr. 9.
Further impetus for change came from Otto I’s conquest of Italy, after which, as noted, the Italian and East Frankish diplomatic traditions merge. The challenge for the modern historian lies in balancing the richer evidence north of the Alps, where survival rates of originals are higher and modern reproductions more readily accessible, with the more fragmentary picture in the south. Already in his abortive first invasion of Italy in 951–2, Otto had issued diplomas for local recipients with the assistance of a new (apparently local) draftsman-scribe called Wigfrid. Like earlier rulers of the region, Otto also made use of recipient notaries, as in a privilege for Archpriest Eistulf of Vercelli.Footnote 82 Following Otto’s conquest of the region and imperial consecration in 961–2, we see more of this interaction.Footnote 83 A two-line subscription-recognition now becomes standard for both regions, developing trends seen in Italy since the 880s. Within the latter region, the recognition is often still rendered in smaller elongatae than those used for the royal/imperial subscription; however, by the reign of Otto III, they have started to take similar dimensions, aligning with East Frankish norms. Likewise, the traditional staff-shaped labarum is initially retained as symbolic invocation within the region, but already under Otto II this starts to cede to the Hebarhardian C-formed chrismon favoured in the north. The conquest of Italy also accelerated the abandonment of the recognition sign north of the Alps. Though in both regions this element had lost its original function, in East Francia it had been retained as a symbol through the 940s and 950s – indeed, we see a number of playful experiments with its composition.Footnote 84 Following 962, however, it becomes progressively rarer, disappearing almost entirely by the later years of Otto II (973–83).
We see a similar confluence in approaches to diploma orientation. Under Otto I, landscape is the norm in both regions. Portrait orientation represents a frequent alternative in Italy, where it can be found in just over a third of his diplomas. By contrast, such cartae transversae are vanishingly rare north of the Alps: in over 200 original charters of Otto I, there are just three certainly authentic cases.Footnote 85 By the reign of Otto III (983–1002), however, vertical orientation had won out decisively south of the Alps; and under Henry II (1002–24), it starts to gain ground in the north, where it would establish itself as the norm by the end of the eleventh century.Footnote 86 Henry’s prize foundation at Bamberg represents a frontrunner in this regard: of the impressive run of twenty-seven privileges issued on the occasion of its foundation (11 November 1007), eight are roughly square and another three clearly (if in some cases only slightly) vertical in orientation, proportions considerably higher than those of Henry’s other privileges of these years.Footnote 87 These documents were largely the work of two Bamberg notaries, the most active of whom seems have hailed from France or Lotharingia (or perhaps Italy).Footnote 88
In both regions, this shift in orientation is accompanied by the more frequent use of display script. Again, the starting point seems to be the reign of Otto III, when scribes began exploring the possibilities of script hierarchy more systematically. Display script is now found not only in the dating clause and (increasingly) main text of charters, but sometimes also in the opening protocol, which itself is occasionally rendered in majuscules instead of elongatae. The latter development bears a resemblance to what we have seen under Robert the Pious in France, whose diplomas also sometimes ditch traditional elongatae for majuscules here. The difference, however, is that under Otto III it is square capitals – the most painstaking and high grade of the antique majuscule scripts – which are substituted, suggesting that the specific inspiration came from the epigraphic models the emperor and his court had encountered south of the Alps.Footnote 89 Indeed, the scribes responsible for these diplomas were all capable of producing traditional elongatae, so this does not reflect a retreat from older traditions. The earliest documents bearing these forms were all produced in Italy and there are signs of earlier experimentation in the region under Otto I and Otto II, whose Italian notaries had occasionally written the ruler’s name in larger or more generously spaced elongatae within the protocol.Footnote 90 As we have observed in Otpert’s diplomas of 958, there had been a similar interest in the potential of display script in the north; and in practice, the two traditions seem to be converging here. As with layout, what had begun as an experiment under Otto III starts to become a regular part of the repertoire under Henry II, a significant minority of whose originals display a consistent script hierarchy.
The most important and immediate effect of Otto I’s conquest of Italy, however, was the development of a new seal to proclaim his imperial dignity. Hitherto, Otto and his predecessors had maintained the militant East Frankish tradition of depicting themselves in profile, armed with shield and spear. Following Otto I’s imperial consecration on the feast of the Purification of Mary (2 February) 962, this cedes to presentation en face and crowned in majesty, holding the symbols of office (staff and sceptre), setting a trend that would soon find imitation in France and Burgundy.Footnote 91 Otto was partly building on Italian traditions, since Hugh and Lothar’s seals had presented them with crown and sceptre (albeit in profile). This new iconography swiftly established itself as the norm – indeed, Otto II would simply inherit his father’s fifth (and final) imperial seal upon his accession in summer 973 – and we next see significant changes under Otto III, upon whose imperial consecration (21 May 996) a new iconography is introduced, presenting the emperor standing in majesty. The following year, this is replaced with another form, with Otto seated enthroned in majesty. And although the latter seal was soon replaced with series of lead bulls, it would furnish the model for all of Otto’s successors, many of whom alternated the use of a wax majesty seal with that of a lead bull.Footnote 92
Conclusions
So what do these changes signify? The degree of regional variation is worth noting from the outset. The Hebarhardian C-formed chrismon is a distinctly East Frankish development, which only gradually finds imitation in Italy and never gains much traction in Burgundy and France. Likewise, the loosening of conventions governing the eschatocol is most visible in France, Burgundy and Italy: only in these regions are minuscules frequently found for the notarial recognition; and it is only here that majuscules or elongatae are ever used for the full dating clause. Finally, the wholesale abandonment of diplomatic minuscule and elongated script is unique to France and Burgundy in the eleventh century (a few anomalous Italian examples notwithstanding), where it constitutes part of a wider reorientation of the diploma tradition. Yet these differences obscure a much larger set of shared changes across all regions. These include: a transition from half-cursive to diplomatic minuscule; a move from landscape (horizontal) to portrait (vertical) layout; the evolution of a strictly two-line subscription-recognition; the dropping of the recognition sign; the introduction of a script hierarchy, first in the dating clause and then within the main text (and sometimes also protocol); and the introduction of a majesty seal. Some of these developments, such as the move towards vertical orientation, are most visible in France, which is often seen as the most innovative region in these years. Many others, however, seem to have taken place independently across multiple regions. Portrait layout itself is attested in Italy from an early date; and the shifts identified in the treatment of the recognition and dating clause are found in France, Burgundy and Italy at around the same time. Other changes clearly have their origin in Germany and Italy: the former is where diplomatic minuscule was first developed in the 850s and 860s, slowly finding emulation in France and Italy over the course of the following century; and both seem to have been at the forefront of introducing a more consistent script hierarchy.
The real challenge lies, however, not in identifying, but in explaining, these shifts. There is a perennial danger of assuming that what seems significant to the modern scholar was necessarily so to a medieval ruler and his entourage. In his survey of similar developments in the ninth century, Mark Mersiowsky concludes that certain rulers and courts were more interested in the appearance of their official documents than others.Footnote 93 This is almost certainly true and holds in equal measure for these years. Here Charles the Simple, Berengar I and Otto I stand out as particularly image-conscious where their diplomas are concerned. Others, such as Louis the Child and Berengar II, were apparently less troubled by such matters (for obvious reasons, in the first of these cases). Of the shifts we have observed, some evidently stem from a desire to find new ways of representing and underlining royal majesty: in the newly competitive world of post-Carolingian politics, monarchs had to find fresh ways to present their authority and distinguish this from that of their leading magnates.Footnote 94 This is most clearly the case with the introduction of the majesty seal, pioneered by Otto I and soon copied by his Carolingian and Rudolfing counterparts; it may also be true of the use of display script in the dating clause and (latterly) main text of diplomas, where it is often employed for the ruler’s name.Footnote 95 Other developments reflect the working out of trends already inherent in the Carolingian charter tradition, occasioned in part by cross-fertilisation with local ‘private’ charters. It is into this category that shifts in layout and loosening conventions governing the eschatocol most naturally fit. Others still must be understood in terms of the wider textual world of our scribes. Important here is the innovative potential of diplomatic minuscule, which has been largely overlooked by previous scholarship. This was in essence an adjusted bookhand; and it is notable that the first experiments with script hierarchy follow fast on the heels of its introduction, with East Francia leading the way. Script hierarchy was a well-established feature of book production, so it is hardly surprising that notaries employing what was effectively a bookhand should be more likely to apply practices from book production to their work – as we also see in the growing use of word-spacing and punctuation in these years.Footnote 96 Finally, it is worth noting that in most cases the developments we have been tracing have their origin not in the 880s, but the 850s and 860s. Under Charles the Fat, we can already observe distinct local documentary traditions, each moving in subtly different (if recognisably similar) directions. This does not mean we need to return to the old fascination with the Treaty of Verdun (843) as the beginning of the end of the Carolingians, but it does mean we should take the reigns of Charles the Bald, Lothar I and Louis the German seriously as the starting point of something distinctive and new.Footnote 97
Above all, these are European trends, which demand European explanations. Carolingian unity may have been on the wane since the 840s and decisively shattered in 887/8, but rulers and notaries of the following century and a half remained in regular contact with one another. As late as 1025, the northern Italian magnates could conceive of inviting Robert the Pious south to rule them.Footnote 98 This was a world in which the Carolingians and their empire remained present, even when – especially when – they were absent; in which Hugh Capet chose to be consecrated at Noyon in 987, the site of Charlemagne’s coronation over two centuries earlier; and in which Otto III sought out the grave of his great Carolingian forebear at the chapel of St Mary (Marienkapelle) in Aachen at Pentecost 1000, apparently in a botched attempt at establishing a cult of his sanctity.Footnote 99 Yet what the Carolingian tradition meant varied considerably across time and space. It is this dynamism that we see reflected in the charter record. We are observing variations on a theme, the same distinctive repertoire being employed in subtly different manners across the post-Carolingian west. These variations are necessarily local and must be understood within the venerable (typically national) scholarly traditions in which they have hitherto been studied. This local flavour can, however, only be fully understood and appreciated within the bigger picture of a shared documentary world, moving in a recognisably similar direction.
Whereas Robert Bartlett once spoke of the ‘Making of Europe’ in the post-Carolingian centuries, Michael Borgolte has written, with equal justification, of Europe ‘discovering its diversity’.Footnote 100 These positions may seem opposed, but actually constitute either side of the same coin: as Europe became more interconnected, local differences came to matter more, not less.Footnote 101 In the charter record of these years, we witness neither a monochrome and unchanging Carolingian order, presaging the advent of the Single Market and European Union, nor a hyper-localised world of national kingdoms, laying the foundations for the nation-states of the nineteenth century. Post-Carolingian Europe was intensely competitive, yet deeply interconnected, the very crucible in which the Ancien Régime was forged. To appreciate its documentary traditions is to take a decisive step towards understanding both the making and diversification of Europe. Let us hope that where nineteenth-century scholars feared to tread, a younger generation, trained in the increasingly interconnected Europe of the twenty-first century, will be ready to forge a path.
Acknowledgements
I would like to thank Sarah Hamilton, Simon MacLean, Jason Glenn and the two anonymous peer reviewers for their helpful comments on draft versions of this article; Jan Machielsen for his editorial support and guidance; Giacomo Vignodelli for access to his important study of the diplomas of Hugh and Lothar in advance of publication; and the Monumenta Germaniae Historica (Munich) for high-quality scans of the DDK volumes. This article was originally given as a lecture to the Royal Historical Society on 1 February 2024. It was written at a happier time when my long-time friend and interlocutor Björn Weiler was alive and well. I dedicate it to his memory, in the hope that it may inspire the kind of comparative historical work so close to his heart. For ease of reference, the following abbreviations are employed: AC = Archivio capitolare; AD = Archives départmentales; AN = Archives nationales; API = Archivio paleografico italiano, IX (Rome, 1920–36); ARTEM = Chartes originales antérieures à 1121 conservées en France, ed. C. Giraud, J.-P. Reanult and B.-M. Tock (Nancy, 2010), available online at http://telma.irht.cnrs.fr//outils/originaux/index/; AS = Archivio di Stato; BayHStA = Bayerisches Hauptstaatsarchiv; BnF = Bibliothèque nationale de France; DDK = Diplomata Karolinorum. Recueil de reproductions en fac-similé des actes originaux des souverains carolingiens conservés dans les archives et bibliothèques de France, ed. F. Lot and P. Lauer (9 vols.; Paris, 1936–49); DDKR = Diplomata Karolinorum. Faksimile-Ausgabe der in der Schweiz liegenden originalen Karolinger- und Rudolfinger-Diplome, ed. A. Bruckner (4 vols.; Basel, 1969–74); FSI = Fonti per la storia d’Italia; GLA = Generallandesarchiv; LA = Landesarchiv; LBA = Lichtbildarchiv älterer Originalurkunden, available online at http://lba.hist.uni-marburg.de/; MGH = Monumenta Germaniae Historica; Newman (with number of document) = W. M. Newman, Catalogue des actes de Robert II, roi de France (Paris, 1937); StA = Staatsarchiv.
Open access
