Studying ritual has never been an easy affair for historians, and even arriving at an understanding of how to define what the term encompasses and how best to incorporate it has proved challenging.Footnote ¹ If true of all periods, in medieval history ritual has become particularly controversial.Footnote ² Although no one in the field now wishes to abandon the concept of ritual entirely,Footnote ³ the best approach for investigating it remains elusive. Some researchers continue to seek inspiration in classic anthropological investigations of ritual, an approach often referred to as ‘thick description’, even though the goal was not merely to describe but to set rituals in the broader context of the societies that produced them, and to analyse through them how participants thought about their world.Footnote ⁴ Of course, medieval scholars do not have the luxury of doing fieldwork: written accounts and artistic representations are all the scholar has to rely on to interpret ceremonial. Yet medieval narratives have often been viewed with thoroughgoing scepticism, moving to bracket off descriptions of ritual actions and to reduce their discussion to suggesting their aims and outcomes. This tendency reflects a fundamental distrust of the sources to communicate the events in the past, but it would be puzzling to anthropologists and those influenced by them, who viewed creating a flowchart of the form of the ritual as a precursor to any further examination. Mediating this dispute ought to be of particular concern to those who study ecclesiastical history. Although the development of theology has traditionally been at its centre, it is impossible to study medieval religion without investigating the worship practices that played a prominent role within it.Footnote ⁵ This is no less true when examining ritual that had political, rather than purely religious, weight.

The many faces of ritual were pervasive throughout the medieval period, but one strain has generated considerable interest. As one of the primary intersections between the secular and sacred spheres in the ancien régime, royal coronations have been a subject of continual fascination both for scholarly and popular audiences.Footnote ⁶ The attraction to the theme has been particularly marked for medieval historians, given the ritual basis for many claims of power, but also because of the formative influence of the Middle Ages on the shape that the ritual assumed. It may be that the very ubiquity of coronations has meant that scholars have yet to discover what many sources related to them may reveal. One area of recent research infused with vitality has been the self-coronations of medieval monarchs, which can no longer be dismissed as anomalous events but rather represented a significant tradition with their own history.Footnote ⁷ Here the goal is not only to examine a coronation for its intrinsic interest, but rather as a case study to suggest how to analyse one overlooked coronation ceremony. While it will be clear that my approach is informed by anthropologically minded studies, and so, in my view, the operation of rituals can be described, this does not mean that students of ceremony should abandon in-depth source criticism and take narrative sources at their word. It argues instead that it is worth offering a plausible reconstruction of rituals and their contexts, upon which an understanding of the motivations behind them or the results of them hinges.

Neither rituals as a whole nor coronations in particular can be studied in the abstract. It makes sense then to turn our attention to the papal coronation of Louis the Pious (r. 813–40) in 816, to be distinguished from his earlier coronation at the hands of his father Charlemagne (r. 768–814) in 813.Footnote ⁸ In choosing this event, my inspiration is Janet Nelson, who saw rituals as invaluable, if methodologically challenging, sources to puzzle out early medieval politics.Footnote ⁹ Thomas Noble signalled the value of discussing the various descriptions of this coronation in tandem.Footnote ¹⁰ In comparison with Charlemagne’s own papal coronation,Footnote ¹¹ Louis’s papal coronation may seem at first blush peripheral and of limited interest, but a careful study of its components suggests that it has much to tell us about the broader environment in which it was created. What is more, an analysis of its sources demonstrates that, while this coronation was formulated at least in part with a political rationale, as is normally held, this by no means exhausts its significance. A closer look demonstrates that there was still considerable deliberation involved in formulating a coronation, and deciding which previous elements of coronations should be recycled or discarded. The decision to set it outside a mass, consideration of the queen’s role and incorporation of the symbolism of ancient Rome were all at stake.

Louis the Pious’s reign has been undergoing scholarly revision: in opposition to the portrait of tragedy and incompetence that had once been painted, recent evaluations have been more nuanced and more positively inclined toward Louis.Footnote ¹² Ritual has played a significant part in these discussions. These gravitate toward Louis’s undertaking of public penance, which was once seen as a negative turning point but has more recently emerged as an ambiguous, neutral or relatively insignificant event.Footnote ¹³ In Stuart Airlie’s masterful survey of Carolingian history, ritual makes several appearances in his handling of Louis’s reign, but most references to it are brief; he avoided investigating the internal workings of ceremonial.Footnote ¹⁴ Inasmuch as this coronation has received recent attention, it came from Philippe Buc, who, as part of his larger project, dismissed this coronation as an example of the insufficiency of narratives to reconstruct medieval rituals.Footnote ¹⁵ Buc’s position has come under considerable criticism.Footnote ¹⁶ Few researchers would now accept his blunt dismissal of the evidence of this or other rituals he handled, but his suspicion of the sources for ritual none the less became pervasive. Rutger Kramer studied the Moissac Chronicle that contains a version of this coronation, but his focus was on Frankish identity, not on describing any ritual; he raised but did not resolve provocative questions about the coronation of 816.Footnote ¹⁷

In my discussion of this coronation, an attempt will be made to demonstrate that it is possible to sketch out its four major elements (blessing, anointing, crowning, gift-giving), even though not all of the sources reliably list each of them. To do so, it is necessary not only to read one or a small sampling of different accounts of the coronation, but all of them together. This allows one to discern patterns that recur and further allows one to sidestep the criticism that an author may have omitted a detail in one version due to, for example, inattention or poor informants. Reading through the different versions reinforces the notion that there were different emphases in what the authors incorporate, but one can still use them to describe the events and to suggest in what ways authors might be manipulating them.

A few words are necessary to set the stage for the coronation. The traditional date on which it is said to have occurred is 5 October 816. A near contemporary annalist insisted on October of that year, and it was standard practice to hold significant events like coronations on a Sunday because of its elevated sanctity; 5 October happened to be the first Sunday available after the pope’s arrival.Footnote ¹⁸ It is not inconceivable, even if less likely, that the coronation could have been on 12 October. The journey for the pope was a rushed affair: the Royal Frankish Annals stated accurately that Stephen iv (r. 816–17) had not even been pope for two months before setting out, even though the annalist was operating with a mistaken idea of when he had been consecrated as pope.Footnote ¹⁹ Stephen iv could not have been pope for more than six weeks when he started travelling to Reims, since the journey probably took at least two months.Footnote ²⁰ Even if the pope immediately turned around and started back to Rome after the coronation was completed, a similar itinerary would put him back in Rome in December, in the middle of the Advent season. The danger then would be that the pope would have had to miss some of the cycle of stational masses said during Advent, at a time when it was still anticipated that he attend these in person and when it might anger the clergy whose church he had failed to visit.Footnote ²¹ That the pope was willing to travel to Reims in the face of this possibility indicates the deeper significance that this mission held for the papacy.

Louis arrived at the city chosen for the consecration, Reims, before the pope,Footnote ²² although this could hardly have been unanticipated given how far he had to travel in comparison to the pontiff. The greatest likelihood is that Louis had previously been in Aachen,Footnote ²³ and using the most efficient Roman roads, the distance he had to travel was about 285 km. This was an option given that Roman roads were preserved and maintained in the Carolingian period.Footnote ²⁴ On the other hand, the likely itinerary for the pope involved about 1,685 km, which meant that the brunt of the long and strenuous travel fell to him. This is in part a reflection of the more sedentary kingship that Louis enjoyed in comparison to Charlemagne,Footnote ²⁵ but it is a striking commentary on the imbalance of power between Louis and Stephen iv: even the pope had little choice but to pay deference to the Frankish ruler.Footnote ²⁶ The one benefit of Reims from the perspective of the pope was that he could stay on the Via Francigena for the entire trip. It may have been too much trouble for Louis to travel all the way to Rome, but by choosing Reims he was deprived of the symbolic value of the city of Rome, including its churches and relics, which loomed large in the Carolingian mind.Footnote ²⁷

Within the city of Reims, there was no question about where the coronation was to take place: in the cathedral dedicated to Mary.Footnote ²⁸ The state of this church at the beginning of the ninth century should not be confused with the soaring and lavishly decorated thirteenth-century cathedral that stands there today. While the architecture of this fifth-century church requires further study, from the state of our current knowledge it seems to have been a simple early Christian church built on the foundation of the tepidarium of old Roman baths. It had a narthex, a nave, two side aisles and a semicircular apse with a crypt that housed the relics of St. Remigius, but it did not feature a transept. The dimensions were approximately 50 m by 23 m, which means that only a modest crowd could have been accommodated within it. That the state of the church was unimpressive and even ramshackle when the ceremony took place is suggested by its demolition and the construction from scratch after the coronation had been completed.Footnote ²⁹ This was sponsored by Louis, likely out of gratitude for its having hosted the event, but had it already been in good condition, the emperor could merely have donated money for its beautification, rather than rebuilding. Within the church, the coronation took place in front of the altar dedicated to Mary,Footnote ³⁰ which is located in the middle of the late medieval church’s sanctuary. The church was built before the trend had started of placing the altar directly over the main relics of the saint, and so although Remigius’ body was in the crypt beneath the church, it was not located directly under the altar as Peter’s body would later be in Old St Peter’s, starting in the pontificate of Gregory i (r. 590–604).Footnote ³¹ One of the attractions of Reims as a location for the coronation is that it was the site of the baptism of Clovis (c. 466–511), Louis’s namesake and imagined ancestor.Footnote ³² While Clovis had indeed been baptised in Reims, the baptistery was a separate building constructed over the caldarium of the Roman baths.

The major players in the coronation all had good reasons for wanting to participate. The driving force for the ceremony was clearly Louis, who, with little advance notice, summoned the new pope soon after his consecration but before he had had much time to establish himself in the Eternal City.Footnote ³³ The claim of the Royal Frankish Annals that the emperor wished to establish a firm friendship with this new pope is plausible, and in earlier entries the annalist had made it clear why that might be advantageous.Footnote ³⁴ Leo iii (r. 795–816), whose pontificate had only been salvaged by Charlemagne’s earlier intervention, had executed the Roman ringleaders of a plot against him. The Annals reported bluntly, ‘The emperor was annoyed with these events.’Footnote ³⁵ Then, while Leo was lying on his deathbed, Romans plundered his goods and caused destruction. These episodes were obvious distractions for Louis when he was involved in the Eastern war with the Saxons and Obodrites, but they also provided a rationale for Louis’s nephew Bernard (r. 810–18), the king of Italy, to get involved in Rome and by doing so to make a case for his holding on to power. This came at a time when Louis was intentionally trying to sideline Bernard. Perhaps a closer personal relationship between the pope and Louis would allow a more pacified situation in Rome and less need for his or Bernard’s repeated intervention. The coronation was not constitutive for Louis, inasmuch as he had already been crowned and named emperor in 813,Footnote ³⁶ but one should hesitate before thinking of the coronation as smoke and mirrors. The Carolingian royal house held the pope in high regard, and previous monarchs had already entered into a spiritual alliance with him, in part by his standing as godfather in baptism.Footnote ³⁷ What is more, Carolingian elites attached significance to religious symbolism, often associated with the liturgy, and especially with reference to the city of Rome and the pope.Footnote ³⁸

Perhaps even more than for the emperor, another beneficiary of the coronation was his wife, Queen Irmingard (r. 778–818). Unlike her husband, she had not previously been crowned or been referred to as an empress (augusta); as a result of this ceremony, her status stood to be transformed. It may be that the decision on the queen’s involvement was not intended primarily to confer honour on her personally, but rather to ensure that the offspring of her union with Louis would be better positioned for the dynastic conflicts that were sure to follow.Footnote ³⁹ Irmingard had already given birth to Lothar i, Pippin i of Aquitaine and Louis the German, all of whom would be privileged as having issued from an emperor and empress over the offspring of Louis’s nephew Bernard. The joint crowning of the imperial couple may have been intentionally patterned on Pope Stephen ii’s blessing and anointing of Pippin the Short and his wife Bertrada in 754, which similarly secured the hold on power for the ruling couple and their offspring. Irmingard survived only around two years after the coronation, dying on 3 October 818. At least some of her Carolingian critics attributed her demise to her supposed role in the blinding and subsequent death of Bernard. The ninth-century ‘Vision of the Poor Woman of Laon’ had her tortured miserably by God in the afterlife for her evil deeds, with three rocks pressing down upon her.Footnote ⁴⁰ According to this source, poor Irmingard did not even get to retain the title from the coronation after her death, being referred to as queen (regina) rather than empress (augusta).

Pope Stephen iv had previously been a deacon active during the reign of Leo iii, who had promoted him to the diaconate.Footnote ⁴¹ Deacons in Rome were generally gifted administrators and were intimately linked to the planning and performance of the liturgy.Footnote ⁴² It is at least possible that he had personally witnessed the coronation of Charlemagne, although our sources are silent on this. Whatever talents the pope brought to the occasion, the main reason for his involvement was the remuneration the pope could hope to receive. Popes had looked to the Carolingians for protection since the mid-eighth century. Stephen iv could also restore to the city Romans who had been exiled after Charlemagne’s intervention with Leo iii, clearing up what had been a drawn-out contretemps. The pope, in addition, received portable gifts from Louis to bring back to Rome. The greatest offering Louis provided, however, was to turn over to the pope some of the lands of central Italy in what would later become the Papal States; this donation was not yet officially announced, but it was already agreed upon in Reims.Footnote ⁴³ In his brief life in the Liber pontificalis, this complex of gifts and favours secured from Louis is thought of as Stephen’s sole accomplishment as pope, and the rest of his pontificate was so unremarkable that his biographer did not bother to complete the life.Footnote ⁴⁴ Stephen had also strengthened the position of the pope as kingmaker, which was not immediately clear although it would be in the next generation of feuding Carolingian kings. Stephen died in Rome on 24 January 817, less than a month after his return; he had spent most of his time as pope in transit. Ultimately, whatever his plans, Louis received little benefit from his newfound bond with this pope.

Before proceeding to reconstruct the coronation, it is first important to bring together the primary sources that make such a reconstruction possible.Footnote ⁴⁵ They have not previously been assembled and discussed as a complete set. Nine primary sources of relevance in this case have been identified, and each is listed in rough chronological order, providing English translations of excerpts of the most significant sections involving the ceremonial:

What is one to make of the variations of the coronation preserved here? Although it may be tempting to abandon any attempt at synthesising these diverse sources, their evidence allows for a fuller discussion of the service, both its form and a more general characterisation of it. After all, even in a more complicated example like the later actions of Charles the Fat (r. 881–7) at the royal assembly at Frankfurt in 873, medieval narratives that contain it all share a consensus on the same basic events, although they diverge on how to make sense of what happened.Footnote ⁵⁹

The most immediate finding is that this coronation was unlikely to have occurred during mass, which would make it distinct from Charlemagne’s imperial coronation.Footnote ⁶⁰ Most of the sources make no reference to a mass at all, which would be strange for authors who wished to stress the sacrality of the scene. The Royal Frankish Annals state that it happened after a mass, although it is unclear if they refer to masses celebrated on the Sunday or in the days leading up to the coronation,Footnote ⁶¹ whereas Thegan stated that it happened before a mass. This is the single best piece of evidence that the coronation took place apart from a mass, since the entire source stresses the sacred nature of Louis and his status as king and priest; the author was looking for opportunities to exalt rather than diminish Louis’s holiness. It may well be that this was part of a campaign to stress Louis as the legitimate emperor after a period of crisis.Footnote ⁶² The one source that does insist that the coronation took part during a mass was the Astronomer, but this should be read as part of a general trend in his work to stress Louis’s enthusiasm for worship and to claim that important events in Louis’s life were connected with the mass, leading the author to add the coronation to that pattern. The Astronomer, for instance, twice stressed Louis’s passion for the liturgy, which supposedly inspired him to call a synod at Aachen in August 816 to seek its reform.Footnote ⁶³ Louis decided to deal with his rebellious son Lothar leniently during the mass for the fourth Sunday in Lent, because of the joyous atmosphere created by the entrance antiphon.Footnote ⁶⁴ He was fully restored to the Christian community after the Bernard affair by a mass celebrated at Metz in 835, which featured seven archbishops saying prayers of reconciliation for Louis.Footnote ⁶⁵ Louis’s inability to attend mass during Lent showed his weakness and inability to curb the feuding of his children in his final days, and a mass was held for him immediately before his death, at which he received the viaticum.Footnote ⁶⁶

One must engage in learned speculation to discern why Louis had not followed the example set by Charlemagne. Crowning Louis outside of the mass was an option. This is something that was clear to near-contemporary Byzantine emperors, whose coronations incorporated prayers and acclamations with religious themes but were separate from any mass.Footnote ⁶⁷ It may also have been that after hearing about Charlemagne’s crowning, Louis was not anxious to participate in a mass in which he would be relegated to an afterthought for most of the ceremony and appear subordinate to Stephen iv.Footnote ⁶⁸ The most compelling reason for setting the coronation apart from a mass was Irmingard’s presence. It was considered taboo at this time for women to approach the altar during a mass, since their impurity could pollute the pristine sacrifice of the mass.Footnote ⁶⁹ This was a long-standing norm, and surely one that the papacy embraced, but it is equally reflected in contemporary Carolingian ecclesiastical canons.Footnote ⁷⁰ The fear of women approaching the altar normally applied to women standing too close or serving the altar during a mass, whether to touch sacred vessels or to distribute the eucharist. Yet the ban would likely have applied by extension to any woman approaching the altar. From the perspective of the clergy, it may have been preferable to suppress a crowning during the mass and remove this danger entirely.

When it comes to the coronation, the sources make it possible to distinguish four major elements in the ritual involving Louis. These are listed in the order that they were held, with the numbers of the sources (see above) that mention each step and brief commentary:

Although the most likely series of movements has been reconstructed, the only source that features all four is Ermoldus. In seeking out the emperor’s mercy, he likely attempted to provide an account that conformed closely to ritual norms. Though Ermoldus is more verbose than the others, they do not contradict any of his most important features. This does not mean that all of the other sources were deceptive, only that they reported on either the information they had received or they simply chose to emphasise some elements over others. None of these documents are liturgical scripts, and they were not intended to stage a similar ritual again. Most are fairly brief and unadorned. It should also be said that biblical and classical references were inserted in speaking about the ritual, but this seems to be a stylistic consideration rather than anything more substantive about what happened. When more than one element appears in a source, they are listed in the same order. The outlier is the Astronomer, who placed the giving of gifts on the days before the coronation, and who had the crowning before the blessing – though in the latter case it might reflect precedence rather than which came first.

While no surviving ordo mentioned the steps involved in Louis’s coronation, the movements reconstructed from narrative sources find an echo in genuine ordines, albeit those that were much older or more recent. Louis’s coronation has at least a rough similarity with the early medieval consecration of a pope.Footnote ⁷⁵ With the pope in front of the confessio (the spot for relics located under the altar), two prayers were also said over the pope, and he was anointed. Instead of a crown, though, he received the pallium as a symbol of his reign; and instead of exchanging gifts, he exchanged the kiss of peace with other clergy in attendance. All of this occurred outside a mass. Stephen iv’s participation made this an obvious model for him to seize upon, since he was familiar with it and had undergone it himself recently. Three influential later ordines are also fertile for comparative data, albeit with the last being chronologically remote – the Erdmann ordo from c. 900, the Ratold ordo from c. 980, both known by the clergy in Reims; and the first written coronation from Reims from c. 1230.Footnote ⁷⁶ It is striking to note that, in these ordines, the basic elements of blessing, anointing and then the coronation all still were present and performed in front of the altar. Save for the last step, they were done before the celebration of the mass, which had become the norm. The giving of gifts was still a part of the ceremony in this later version, although it was displaced into the mass that was celebrated after the crowning.Footnote ⁷⁷ During the offertory in the first two ordines, the king provided the bread and wine for the sacrifice of the mass to the clergy. In the Reims ordo, the king offered the archbishop of Reims twenty gold coins. Inasmuch as the head clergyman still provided gifts, it was the insignia he extended to the king, if not the service he performed for the king in presiding over the rite.

One of the striking features of the ceremony here was the presence of Irmingard. It is simply not the case that she was incorporated into all of the accounts of the coronation,Footnote ⁷⁸ even though there is no doubt that she was a part of the ritual itself. She only appears in (2), (5) and (6), in which it is clear that, like Louis, she too was blessed and crowned. In neither of the two sources that speak about Louis’s anointing does it mention that Irmingard was anointed, and this in sources that are favourable to both members of the imperial couple. This means simply that Irmingard was not anointed, a prerogative reserved for the king alone and expressing his superiority. It is likely that the Astronomer ignored Irmingard’s presence entirely because he insisted that the coronation had taken place during the mass, and he would have had to confront her proximity to the altar during a mass. The omission of Irmingard does not seem to be caused by her death and any bad repute that lingered after it; (1), (3) and (4) were all written before she died, as were likely the references in (9), and these too omitted her.

On the whole, it is easy to see this ritual as a pale reflection of Charlemagne’s papal coronation, much as generations of scholars had seen Louis’s reign in comparison to Charlemagne’s. Reims was no Rome, and its cathedral was no Old St Peter’s. The size of the church limited how many people could possibly have been in attendance. They would have lacked the large number of clergy in attendance or the professional schola cantorum of the city of Rome. No matter how drawn out the steps were, this was still likely a fairly short ritual. Louis already had the title of emperor and an imperial crown before it began. Yet in spite of this, it was also a ceremony that showed that the pope and emperor were capable of improvisation and innovation, perhaps borrowing selectively from the papal consecration. As opposed to Charlemagne’s crowning, this one was intentionally staged outside a mass, incorporated an anointing for the emperor and was directed to the emperor and also to the queen. In making Irmingard a part of the ceremony, the clergy had to distance the ceremony from a mass, and they chose to deny her anointing, but they had found a way to accommodate the emperor’s wish to include her. The full implications of this decision were not evident to all contemporaries until the imperial couple moved against Bernard, but the ritual effectively broadcast that they and their line were the only true rulers.

The sources that report on the coronation are positively inclined toward Louis and the ritual. This is hardly surprising, given that the authors either were at his court or in well-connected ecclesiastical institutions and stood to benefit from praising the ruler. One can only imagine the negative reception if the coronation had been evaluated by a partisan of Bernard. If genuine, two diplomas from Louis demonstrated that he personally wished to spread the good word of the coronation. Ebbo further created propaganda, although it was intended primarily to draw on the coronation to promote his cathedral rather than Louis’s reign. All of these sources anticipated that a ritual conducted with the pope would reflect well on the king, and they sought to advertise that fact. The sole exception is Stephen’s biography in the Liber pontificalis, which is also self-promoting, and along these lines reported only what the papacy had gained by its role in the crowning – not the coronation itself. The best propaganda is not created from whole cloth, but instead draws on broadly accepted facts; and so it is still inadvisable to dismiss the ritual or its components as a fabrication. Years removed from the events in question, the author of the Annals of St Bertin was even able to draw a comparison between when Louis was rightly given the crown by the pope and when he was restored after his deposition; both of these events were mediated through saintly defenders.Footnote ⁷⁹ This entry went more to burnishing Louis’s legacy rather than to help him during his reign.

Narrative sources are often incomplete, only preserving what an individual author considered worthy of note or emphasis; they were not intended to perform a ritual. As with any source, manipulation of the events does occur, but with an attentive eye, one can discern and account for it, such as the Astronomer’s references to the role of the mass leading him to insert the coronation inside a mass. Within sources, all kinds of allusions, whether biblical or classical, are recurrent, but this does not necessarily touch on the factual basis of the events described, only how they are framed. And within literary creations like Ermoldus’ poetry one expects to find stylised elements like elaborate prayers and a crown attributed to Constantine. But even here precious material appears that is missing from the other sources. His prayers are the only hard evidence that the coronation was intended to extend a special blessing on Irmingard’s offspring even as Ermoldus took liberties with the prayer; scholars must tacitly accept it for this interpretation to hold. In lining up the accounts chronologically, no clear progression over time can be discerned as to how the coronation is described. So, for example, Thegan was just as capable of writing a propagandistic account before the civil war had erupted in the 830s as those who wrote in its wake.

One omission from the discussion of sources for Louis’s papal coronation has been iconographical representations, which have featured prominently in discussions of late medieval crownings.Footnote ⁸⁰ Although there are later imaginative representations of Louis having been crowned, no image chronologically proximate to the coronation exists. Two relevant pictorial sources do, however, merit attention. A gold solidus was issued by Louis in 816–18 and likely minted at Aachen in commemoration of his papal crowning.Footnote ⁸¹ On the obverse of the coin appeared a profile bust of Louis with a Roman-style laurel wreath (another form of a corona) on his head; and on the reverse was a cross encircled with a wreath with the inscription ‘munus divinum’ (‘divine gift’). The minters of the coin appear to be in conversation with (5), in part stressing the same ancient Roman roots for Louis’s reign that had been evoked in reference to Constantine’s crown. It is also possible that the message on the reverse implied a rejection of the munera Petri (‘gifts of Peter’) from the same source, with instead the argument being asserted that Louis’s crown was given to him directly from God rather than through the intermediary of the pope.Footnote ⁸²

The other source that bears mention is a frequently recopied manuscript illumination of Louis as a soldier of Christ, which appeared in the context of Hrabanus Maurus’ early ninth-century De laudibus sanctae crucis, which was to become a well-attested image in copies starting from the ninth century.Footnote ⁸³ No evidence links it directly with Louis’s papal coronation, and it more generally relates to his public presentation. While the imagery here is complex, Louis is depicted as a Roman military officer, and aspects of the illustration borrow from late antique imperial portraiture. It was not until the fifteenth century that any variation of this theme first incorporated a crown,Footnote ⁸⁴ but it is plausible that the nimbus that surrounded Louis’s head may serve as a substitute crown in earlier versions. The inscription that appears in the nimbus is TV HLVDOVVICUM CRISTE CORONA (‘You, Christ, crown Louis’). Much like Louis’s coinage from a similar period, Louis projected authority visually by drawing upon ancient Roman models and by suggesting that the ultimate source of his crown was from a divine rather than papal donor.

The last point lays bare the ambiguity at the heart of the enterprise of the pope’s crowning the king. Louis likely wanted to exalt Irmingard and his children with her, but our sources hint at several problematic issues that simmered under the surface. Louis wished to incorporate sacred elements that could only be dispensed by a religious authority like the pope, such as anointing and blessings, but he did not want it within a mass. Iconography produced afterwards stressed the divine rather than papal source of the crown. It was perhaps because of Stephen iv‘s concern to receive a donation of land that he was willing to allow some distancing from the mass, but without recognising that this would set a precedent. What is more, Louis wanted to have his queen receive a special blessing, but her gender raised concerns about her approaching the altar in the course of a mass. Louis employed Roman symbolism, but the church that rested on Roman baths, the crown that allegedly came from Constantine and the iconography all related to ancient rather than papal Rome, and he had not found it compulsory to travel to the Eternal City himself to be crowned. Louis’s decision to receive coronation from the pope followed in the footsteps of his father Charlemagne, but the ceremony departed from that famous event by being performed outside the mass and displacing it from Old St Peter’s, and he accepted anointing at the hands of the pope. All of these considerations bespeak an atmosphere of negotiation and experimentation that still governed what would later be essential components of coronations.

This discussion also points to some final reflections on analysing a medieval coronation or any medieval ritual. Even at its best, this kind of reconstruction will lack the colourful detail contained in classical anthropological reports and will strike some as schematic. And, naturally, as with any retelling of past events, there is no way to verify every detail. In spite of these limitations, though, if scholars wish to continue to engage with ritual, there seems to me no other option than to work with imperfect sources to describe the form of ritual as well as possible. This approach takes seriously the experiences of contemporaries and the weight they attached to the action of ceremonial, rather than allowing source criticism to become an end unto itself. As part of this process, one must remain sensitive to the lessons that anthropologists long ago taught historians to guide our analysis. Some of these are evident in this discussion, such as the tension between purity and pollution, especially prominent in negotiating how women were involved in the ceremony; the mediation of the boundaries between the sacred and secular; the development, modification and eventual crystallisation of ceremonial forms; and the production of other media responding to claims made initially through ritual. These concepts are useful to better understand one particular coronation, but also to provide comparative data to investigate if the same or a different complex of concerns surfaced in other coronations or entirely different rituals. This engagement is preferable to operating with the assumption that the sources disqualify any reconstruction, a position that runs the risk of obscuring how ritual operated, what it meant and why it mattered.