With The Romanesque as Relic: Architecture and Institutional Memory at the Collegiate Church of Saint-Omer, Michalis Olympios contributes to ongoing discussions about the architectural visualization of institutional history practiced by medieval religious foundations in Latin Europe. This article focuses on the collegiate church of Saint-Omer (Pas-de-Calais), a rare surviving example of a building from the region of French Flanders preserving architectural fabric fromthe eleventh to the sixteenth centuries. More specifically, Olympios examines the Romanesque apsidiole in the chapel of Notre-Dame-des-Cloches and its integration into the edifice's Gothic north transept, erected in the third quarter of the fifteenth century. A close reading of the architecture, the narrative and hagiographic sources, and unpublished archival documents demonstrates that, as in many other instances from across Europe, the retention of this earlier feature reflects the secular chapter's conscious decision to showcase the antiquity and prestige of the church by providing visual “evidence” of its foundational myth.
Ravaged by the Hundred Years’ War, the Wars of Religion, the French Revolution, and two world wars, the region of French Flanders has suffered a severe depletion of its medieval monumental heritage.1 Because of this sobering history, architectural historians have repeatedly hailed the former collegiate church at Saint-Omer (Pas-de-Calais) as a key monument for comprehending the development of Gothic architecture in northern France and the Low Countries, given that its fabric preserves evidence for a series of more or less well-defined building campaigns dating between the thirteenth and sixteenth centuries (Figure 1).2 Surprisingly, the church's Romanesque component has received scant mention in the literature, despite the acknowledged dearth of extant eleventh- and twelfth-century structures in that part of northern Europe. One would expect the main survivor from this period—the semicylindrical apsidiole in the chapel of Notre-Dame-des-Cloches, opening off the eastern wall of the church's north transept—to have attracted more scholarly interest, if only for the way in which it has been integrated into the later Gothic work (Figure 2). In the mid-fifteenth century, when the north transept was rebuilt in a Flamboyant style, the Romanesque apse was given a new molded frame and crowned with an ogee-shaped hood mold (Figure 3). This relatively elaborate framing, coupled with the apsidiole's off-center placement in its new bay, served to underscore the apsidiole's autonomy from its revised architectural context and, by implication, its deliberate retention within the new scheme. In other words, the incorporation of the Romanesque apsidiole into the late Gothic transept was not an accident of the chantier's progress; it was intentional and meant to be perceived as such.
The stark juxtaposition of Romanesque and Gothic, of old and new, at Saint-Omer is neither exceptional nor an isolated case. Related phenomena in fifteenth-century Flemish art have long incited art historical comment. Erwin Panofsky was among the first and most influential scholars to suggest that the “naturalistic” depiction of Romanesque-looking buildings and individual architectural members in early Netherlandish painting indicated keen interest in the architectural heritage of a bygone era.3 The Prado's Betrothal of the Virgin by the Master of Flémalle (Robert Campin?), commonly dated to the 1420s–30s, has often been the linchpin of arguments concerning the sensitivity of medieval audiences to past and present artistic styles (Figure 4). Panofsky and later authors contended that the lavish Romanesque rotunda representing the Jerusalem Temple, framing The Miracle of the Staff on the panel's left, alluded to Judaism and the Era of the Old Law (or the Old Testament), whereas the majestic Gothic portal under construction on the right, serving as the backdrop for The Betrothal of Mary and Joseph, signaled the advent of Christianity and the Era of Grace (New Testament). Thus, the older architectural style stood for the past and the contemporary one for the present and future, revealing the artist's understanding—and possibly that of his patrons and intended audience—of the historical progression of architectural forms.4
The way in which the unfinished Gothic edifice looks poised to engulf the old Romanesque building in the Prado Betrothal parallels the process by which the Saint-Omer apsidiole was grafted onto its present surroundings. The rationale behind the apsidiole's preservation has hitherto remained elusive; to my knowledge, the question has never been posed. Nevertheless, if the evidence of contemporary Netherlandish painting is any indication, the visual collation of Romanesque and Gothic forms within the same space must once have carried specific symbolic meaning. In this essay I will propose a motive for the apsidiole's preservation on the basis of the church's material fabric and the extant archival documents, both published and unpublished, as well as pertinent comparanda from throughout Europe. I will show that the patrons’ incentive for such a calculated formal mélange was aligned with analogous initiatives elsewhere, bringing into sharper focus a set of common intellectual responses to particular spiritual and practical issues and, ultimately, a fundamental aspect of the mental habits of premodern societies.
The Collegiate Church of Saint-Omer from Its Foundation to Its Romanesque Incarnation
The church of Notre-Dame at Saint-Omer was founded in the seventh century by Audomarus (d. after 667), a missionary bishop elevated to the see of Thérouanne by King Dagobert I (d. 639). It constituted one of three sanctuaries owned by the monastic community established at the site of Sithiu; this had been granted to the bishop by Adroald, a well-to-do Frankish convert, and it eventually served as the founder's last resting place. Audomarus's posthumous prodigies, a few of which occurred at his tomb in Notre-Dame, soon led to his canonization—an honor also accorded Bertinus, another contemporary luminary of Sithiu who served as the community's abbot and was laid to rest in its church of Saint-Martin. The two closely associated yet distinct communities that developed around the churches housing the bodies of these revered founders were separated by a series of ecclesiastical reforms implemented in the ninth and tenth centuries. Consequently, the church of Notre-Dame came to be served by a college of secular canons, while that of Saint-Martin was connected to a community of Benedictine monks. The success of the cults of Audomarus and Bertinus, as fostered by their respective communities, eventually resulted in their names’ superseding the old church dedications to the Virgin and Saint Martin of Tours and paralleled the emergence of a strong sense of institutional identity, which conditioned the often tense and antagonistic relations between the communities of Saint-Omer and Saint-Bertin into the modern era.5
Little is known of the Merovingian church of Notre-Dame beyond a handful of brief mentions in the hagiographic literature. It appears to have been gravely damaged by fire in 1033, and to have been rebuilt from the ground up between the mid-eleventh and mid-twelfth centuries.6 Given the absence of any documented archaeological investigations inside the present building, the original form of this Romanesque edifice is a matter of conjecture. However, a small number of architectural elements dating from the Romanesque building's construction are preserved in the church's current configuration.
The most conspicuous of these survivals, and the only one previously identified, is the apsidiole in the chapel of Notre-Dame-des-Cloches, in the second bay of the north transept's eastern aisle (counting from the north; Figure 5). Semicylindrical on both exterior and interior and crowned by a semidome, the apse is set in front of a short barrel-vaulted straight bay, which opens directly into the transept's eastern aisle (see Figures 2 and 3). Even though the interior of the apse appears at a glance to be continuous with the masonry of the present transept, closer inspection reveals sutures between the edges of the barrel-vaulted bay and the late Gothic molded frame. The apse's exterior was articulated by a series of elegant engaged shafts (of which two remain in situ despite later reconstruction) banded at two levels by stringcourses exhibiting carved and molded ornament (Figure 6). The wall surface in the center of the structure, between the two stringcourses, is pierced by a single round-headed window, the frame of which is splayed on the interior.
The formal vocabulary employed here allows the identification of two further Romanesque fragments, namely, the round staircase turrets attached to the west of the portals in both transepts. The turret of the south transept façade, in much better condition than its northern counterpart, rises to most of its original height (Figure 7). Encroached upon by the later structures of the Gothic transept, it still retains an engaged shaft of the same type as those in the north transept apsidiole. The same is true of the north transept staircase, now drastically truncated in both its upper and lower parts and surviving as a narrow ring sandwiched between later elements (Figure 8). The positioning of the two turrets and the apsidiole indicates the general dimensions and plan of the Romanesque transepts, which did not differ markedly from those of their Gothic successors.
The Romanesque Transplanted: The Rebuilding of the North Transept and the Preservation of the Apsidiole of Notre-Dame-des-Cloches
Beginning in the early thirteenth century, the Romanesque church at Saint-Omer was replaced by a new Gothic building. The design of the chevet's interior elevation, conceived in the first decades of the century, dictated the form of all the later construction campaigns and contributed to the building's overall visual coherence. The Rayonnant-style south transept, with its imposing façade and portal, was erected in the third quarter of the thirteenth century, followed by the nave between the late fourteenth and mid-fifteenth centuries, and the Flamboyant-style north transept in the third quarter of the fifteenth century. These works were bookended by the consolidation and completion of the west belfry tower and front, which occupied the masons well into the sixteenth century.7 From ca. 1380 onward, the work's progress can be traced through the financial accounts of the church's fabric agency (fabrica, fabrique), among other sources. In the 1890s, Louis Deschamps de Pas studied and selectively published these financial accounts, and his two articles on the topic have shaped all later discussions of the church's architectural history and interior furnishings. Nevertheless, Deschamps de Pas glossed over several of the narrative's more intriguing details, concerned as he was with sketching the broad outline of the chantier's development over five centuries.8 Here, a fresh reading of the fabric accounts, chapter meeting records, and other unpublished archival material casts new light on the Gothic revamping of the north transept and the sparing of the Romanesque apsidiole.
The early thirteenth-century building campaign that gave rise to the Gothic chevet appears also to have produced the crossing, up to and including its western piers. It was here, stretching between the crossing's western piers, that the choir screen (doxal) would have been located, delimiting the church's liturgical choir at this end (see Figure 5).9 The two bays of either transept adjacent to the crossing were also built, at least partly, at this time. In both the south and north transepts, the entire elevation of these two bays dated to this period, especially in the eastern aisle, where the vaults are contemporary with the rest of the structure; conversely, in the western aisle, the vaults had to be reconstructed when the old Romanesque western wall in both transepts was removed in the fifteenth century. Later work on the transepts was aimed mainly at removing the remaining Romanesque masonry, patching up any scars caused by this procedure, filling in the last two bays, and fashioning new terminal walls to the north and south.
The reconstruction of the north transept arm depended on the completion of the rebuilding works in the adjacent cloister, carried out in the late 1440s. By 1448, the canons of Saint-Omer were contemplating the project's next step, which entailed inviting Jean Sterbecque, the master mason of the nearby abbey of Saint-Bertin, and masons from Béthune and Méréville to perform an inspection of the “old work of the church,” its Romanesque masonry, and to submit their opinions in writing.10 Following this, preparations for building could begin in earnest with the partial demolition of the Old Tower toward the cloister in 1449–50.11 Meanwhile, the carpenters put up covered walkways offering unimpeded access from the church to the cloister (to the north) and the chapter house (to the west), made a “tabernacle” for a crane used for lifting stonework, and cut the masons’ templates.12 At this stage, the chapel of Notre-Dame-des-Cloches was sealed off from the rest of the building by means of a timber partition.13
From that point onward, the fabric accounts contain several entries concerned with the purchase and carving of stone to produce ashlar, stringcourses, moldings, piers, bases, capitals, and other architectural members, while scaffolding was set up for the masons and hoisting machinery was put to intensive use; at the same time, the stonework retrieved from the demolition of the old walls was sold off, to be reused elsewhere. Construction began with the transept's north front, from the portal to the rose window and the upper gable; the main arcade piers soon followed. By 1458–59, the aisles were being vaulted and the chapels of Notre-Dame-des-Cloches, Saint Anthony (located in the bay in the transept's northwest angle), and Saint Andrew the Less (immediately to the south of Notre-Dame-des-Cloches) were undergoing reconstruction (see Figure 5). In the following year, the walls and vaults of these three chapels were completed, and flying buttresses were installed along the wall of the clerestory. The high vaults of the transept were put in place between 1460 and 1464, and the pavement below them laid immediately thereafter. The last major elements of the north transept to be finished were the tracery of the rose window (1467–70) and the sculpture of the façade gable (1470–72).14
Special status seems to have been accorded to the chapel of Notre-Dame-des-Cloches during the north transept's rebuilding. Not only was its apse partitioned off from the rest of the Romanesque transept (later pulled down) at the project's launch, but a temporary, tile-roofed wood shed was erected over it in 1457–58 to protect both the structure and the builders during the chapel's makeover in 1458–61.15 Furthermore, the chapel's altar might well have remained untouched throughout the “Gothicization” of its immediate surroundings. First, the appointment of chaplains continued unabated at least into the early 1450s; second, while a new altar mensa and piscina were procured for the chapel of Saint Anthony in 1458–59, and the altar in the chapel of Saint Andrew was dismantled by 1461–62 (remaining out of commission until 1466–67), the altar of Notre-Dame-des-Cloches received a new piscina and was being served by a chaplain in 1459–60, when major structural work concluded there.16 What is more, available textual evidence indicates that the chapel was never completely out-of-bounds to the faithful during the transept reconstruction campaign. In 1442–43, Notre-Dame-des-Cloches was provided with a donations chest for the benefit of the fabric agency, which also drew revenue from the chest placed next to the cenotaph of Saint Audomarus (located in the nave). While building was in full swing, between 1449 and 1472, money was drawn from the donations chest annually.17 The exact location of the chest within the chapel is unknown—as is the possibility of its temporary relocation to a more accessible spot near Notre-Dame-des-Cloches—but it appears that, despite the disruption wrought by the chantier, the chapel remained popular and reasonably well attended. The same cannot be said for the other north transept chapels in this period.
The Romanesque Eliminated: Notre-Dame-des-Cloches, the Tomb of Saint Erkembodo, and the Altar of the Virgin
To gain a better understanding of the attitudes of fifteenth-century builders and their patrons toward the chapel of Notre-Dame-des-Cloches, we now turn from the chapel's formal attributes to its furnishings, ornaments, and function within the church's sacred space. Admittedly, little is known about the chapel prior to its partial remodeling in the late 1450s. Its foundation date remains uncertain, as does the origin of its name. Emmanuel Wallet contends that the name originated in the use of the chapel as a space for baptizing the church's bells, while references to the chapel of the Blessed Mary “ad Campanas” or “iuxta Campanas” in medieval documents hint at its former proximity to a belfry. Whether the structure implied here was the belfry in the Old Tower—situated somewhere in the vicinity of the north transept, close to the cloister, before it was brought down in 1449–50—or a tiny belfry atop the chapel itself (first mentioned in the late fifteenth century) is a matter for speculation.18 The other name by which the chapel was commonly known in the Middle Ages, “of Matringhem,” recalls the bequest of rents and the foundation of Masses for the sake of Agnès de Boubers, Lady of Matringhem (as per her will of 1423, not executed until 1455); her relief effigial slab may have been displaced later, but it was still to be found near the chapel in the nineteenth century.19 An “ancient lamp” is the only item that can be assigned to the chapel's interior with any certainty at this stage.20 Other than that and the donations chest already discussed, a few brief mentions of repairs to the masonry of the chapel, the timber roof above it, and the lockable doorway leading into it round up the available early evidence.21
Much more is known about the chapel's interior arrangement and use following the alterations of ca. 1460. The completion of its structural envelope was immediately followed by the insertion of a piscina, the arched head of which (complete with miniature rib vaulting) was the work of the mason Pierre le Wart; the placement of two images, one of Saint Audomarus and one of Saint Erkembodo, executed on lead by the painter Pierre Pol; the affixture of a grille; and the making of a replacement wooden pulley for the suspension mechanism of the “ancient” lamp.22 In 1460–61, the window of the chapel was graced with a two-foot-high stone column carved with an image of the Virgin, and a wooden casket (of unknown purpose) was also made.23 In the following year, a small casket was commissioned for storing the vestments and headgear of the chapel's images, and a small wooden reliquary was made, adorned, and painted to enshrine a tooth of Saint Audomarus, which was thereafter permanently displayed on the altar and became the object of pious offerings.24
The chapel, its relics, and its images often prompted gifts of money, precious metals, cloth, animals, grain, and produce, while the selling of badges to devout visitors generated additional revenue.25 Moreover, Masses and rents founded in the 1470s accounted for the chapel's artificial lighting, from the seven candles burning before the image of Notre-Dame-des-Cloches on the six feasts of the Virgin, established by Canon Jean Magistri, to the maintenance of the hanging lamp at the expense of Canon Jean Dessinges.26 These and other bequests of the collegiate church's canons indicate sustained support for the chapel, its contents, and its functions, as do the frequent requests for burial in or in front of the chapel.27 Its liturgical vessels, textiles, reliquaries, and images were the objects of almost constant update and repair, as were its masonry and carpentry, to the end of the Middle Ages.28
Judging by the chapel's dedication and textual evidence regarding the existence there of a venerated stone sculpture of the Virgin, the main cult enacted in this space was surely that of the Mother of God. The permanent placement of the reliquary of the tooth of Saint Audomarus on the altar, coupled with the commission of a painted representation of this saintly figure at the time of the chapel's renovation, demonstrates that the memory of the collegiate church's founder was also celebrated here. Nevertheless, it is the cult of the relatively obscure Saint Erkembodo, whose image in the chapel formed a pendant to that of Audomarus, that is instrumental in unlocking the secrets of Notre-Dame-des-Cloches. Erkembodo was an eighth-century abbot of Sithiu and bishop of Thérouanne, whose life and deeds are known almost exclusively through the mid-tenth-century Gesta abbatum Sithiensium of Folcuin, a monk at Saint-Bertin, and a short and derivative vita composed by Abbot Jean III of Saint-Bertin (1186–1230). On the one hand, Folcuin's account consists of little more than a collection of four charters issued during Erkembodo's abbacy by the Merovingian kings Chilperic II and Theuderic IV, and by a certain Rigobert, to protect and augment the possessions of the Sithiu community. On the other, Jean III's life of the saint offers but a retelling of Folcuin's text, updated with a few brief passages inspired by oral tradition. According to the author, the factual penury of his narrative was due to the lack of written evidence regarding the saint's deeds, including his miracles, which were allegedly numerous but appeared not to have been recorded out of negligence.29
We are better informed about the development of Erkembodo's cult. The relics of Saint Erkembodo, “the fourth bishop of Thérouanne after Audomarus,” were elevated from his tomb in 1085, according to an inscription “in ancient characters” (antiquis litteris) on both sides of a bronze cross found in 1466 inside the reliquary containing his body.30 This elevatio could well have occurred while construction of the Romanesque church was under way; after all, in the recollection of the elders, as mediated through the saint's life in ca. 1200, this building campaign was supposed to have been funded largely by donations made to Erkembodo's tomb in return for the many great miracles being performed there.31 Once removed from the tomb, the saint's remains were encased in an oak vessel bound shut with iron bands within a silver-gilt reliquary shrine, which was normally situated in an elevated place in the choir, alongside the shrines of Saints Audomarus, Austrebertha, and Folcuin.32 In 1466, the shrine was opened, the relics of Saint Erkembodo were displayed to the faithful, and the saint's head was translated into the old reliquary of the head of Saint Audomarus, Audomarus's head being moved to a new, custom-made, sumptuous receptacle on the same day.33 From then on, on the days of great feasts and other momentous occasions, both the body and the head of Saint Erkembodo were brought down from their repositories, placed in the middle of the choir, on the high altar, or elsewhere, and carried on procession to counter adverse weather conditions, to supplicate for peace, to safeguard the sovereign's prosperity, to commemorate the latter's military triumphs, and for various other reasons.34
The tomb or, more accurately, the cenotaph of Saint Erkembodo is still found at the west end of the church's ambulatory, attached to the wall between the northeastern crossing pier and the choir main arcade pier directly to the east (Figure 9, see Figure 5). Before it was transferred to this location in 1835, the monument lay adjacent to the terminal wall of the eastern aisle in the north transept.35 It consists of a rustic-looking lidded sarcophagus of reddish sandstone, raised on the back of two severely abraded lions carved in black marble (Figure 10). The monolithic rectangular tomb-chest and its lid are both frugally adorned with incised parallel striations in various straight and curvilinear patterns. It is typically dated to the Merovingian period, even though its form does not fully match the products of any of the established regional traditions of sarcophagus production in Frankish Gaul.36 In formal terms, the closest comparanda may be a series of extremely plain Rhenish late Roman sarcophagi akin to those discovered in the course of recent excavations in Trier and Cologne.37 Given that the monastery of Sithiu is known to have held possessions in that region since before the ninth century, inherited and jointly maintained by its succeeding institutions until the mid-sixteenth century, it is logical to assume that the tomb-chest was a German import. However, although the reuse of Roman sarcophagi for early medieval burials was common, it is unclear whether the tomb-chest housed Erkembodo's body between his interment in the eighth century and his elevation in the eleventh, or if it was adopted later as a suitably old-fashioned cenotaph, when the two lion supports would presumably have been added.38
Most important here is the original location of the saint's tomb. Folcuin states that Erkembodo was buried by the people in the church of Saint-Omer, in front of the altar of the Holy Mother of God.39 The thirteenth-century life specifies that the burial took place “in front of the high altar dedicated to the Holy Mother of God Mary, in whose honor the church of Saint-Omer had originally been built and dedicated by the holy prelate Audomarus.”40 What is more, in the version of Folcuin's Gesta copied by the Bertinian monk Alard Tassart at the beginning of the sixteenth century, which brims with scribal interpolations, the altar of the Virgin is identified as that in the chapel of Notre-Dame-des-Cloches, where Erkembodo's stone tomb was still found in the copyist's day.41 It is not known whether the cenotaph, once at the north end of the north transept's eastern aisle, next door to Notre-Dame-des-Cloches, occupied the same location in the early nineteenth century as it did in the early sixteenth, yet by the late Middle Ages the topographical link between tomb and chapel was indisputable. If anything, the commission of a painting of Saint Erkembodo for Notre-Dame-des-Cloches at the time of its refurbishment in 1459–60 implies that this connection would have been on the canons’ minds at a crucial juncture in the project's development.
Further evidence demonstrates that the site of Saint Erkembodo's tomb was pivotal in shaping the final plan for the rebuilding of the north transept and the retention of the Romanesque apsidiole. It comes in the form of a bull issued by Eugene IV in 1441 in favor of the reconsecration of the collegiate church of Saint-Omer following the completion of a substantial part of the reconstruction, which had been ongoing for some time. This document states:
Bishop Eugene, servant of the servants of God, to his beloved sons, the provost, dean, and chapter of the church of Saint Audomarus of Saint-Omer, in the diocese of Thérouanne, greetings and apostolic blessings.…The petition shown to us recently on your behalf stated that, from a saint's life and calendars and certain other ancient texts and documents of your church, it appears that this same church had been dedicated in the name of the Blessed Virgin Mary by Saint Audomarus, at the time bishop of Thérouanne, and that the feast of that dedication day had been celebrated of old on the eve of the feast of the Blessed Luke the Evangelist. However, of that church, which later, on account of its great age, partly collapsed and was partly consumed by fire, only the portion where its high altar was located remained enclosed within the church that was later built to a much larger size and on a grander scale; because of this, some are concerned about whether, after the church has thus been constructed, it is still considered consecrated, or whether it will need to be reconsecrated. For this reason, it has been humbly requested of us, on your behalf, that we deign, in our apostolic benignity, to suitably provide on these matters.42
Eugene's letter fills in some gaps in the story presented thus far. According to the petition addressed to the pope, the chapter had drawn information from old documents to compose an account of the origins of the church at the time of Saint Audomarus. The original building, impaired by age and fire, had been substituted by a new, larger church, except for the area of the high altar, which had been preserved within the later structure. From the vantage point of 1441, the only part of the church that had not been given a Gothic facelift was the north transept, which would have retained its distinctive Romanesque appearance until the masons began work a few years later. Thus, the site of the high altar of Audomarus's old church (as distinct from the site of the Gothic building's high altar) was already identified with the area where the chapel of Notre-Dame-des-Cloches lay. Such an identification probably emerged through a reading of the life of Saint Erkembodo by Jean III of Saint-Bertin, which, as shown, alleged that the altar of the Virgin (located in front of the prelate's tomb) had served as the church's high altar. The likeliest scenario, then, is that the relics of Erkembodo were discovered before 1085 in or next to what was later known as the chapel of Notre-Dame-des-Cloches—or, at any rate, his tomb would have been sited there by the fifteenth century. Around 1200 came the first claims that the tomb lay before the old high altar of Audomarus's church. By ca. 1440, while the chapter was investigating the chronology of its church, the presence of Erkembodo's tomb in the north transept must have led to the identification of the altar of Notre-Dame-des-Cloches as the old high altar. From the 1440s to the 1460s, the Romanesque apsidiole, likely with the altar of the Virgin housed within, was preserved and reintegrated into the new Gothic transept.
It becomes evident that by the second quarter of the fifteenth century the site's eleventh- to twelfth-century construction phase had largely been “forgotten” by the chapter—a paradoxical development, given the pains taken to preserve the Romanesque architectural tissue soon after. The last prominent material vestige of the Romanesque building, the north transept, was linked to the original, seventh-century church founded by Audomarus. The altar of Notre-Dame-des-Cloches assumed the role of the Merovingian church's high altar, in front of which the founder's saintly successor, Erkembodo, had been laid to rest. This attempt to bridge the chronological distance between the collegiate church's founding and its present, by “folding” time so as to excise any intermediate episodes that might complicate the simplicity of the narrative, was a purposeful process of mythification meant to evoke the institution's long and glorious history, one inscribed within a vibrant intellectual tradition cultivated across medieval Europe.
Folding Time: The Visualization of Institutional History at Saint-Omer and Its Broader European Context
The Romanesque apsidiole of the chapel of Notre-Dame-des-Cloches in the north transept of the collegiate church of Saint-Omer stands at the nodal point of various institutional traditions linking the Gothic church to the memory of its past. The images of the two saints—the church's founder, Audomarus, and his successor to the see of Thérouanne, Erkembodo—that were made to adorn the chapel after its refurbishment in the early 1460s served as perennial reminders of the chapter's pre-Carolingian beginnings in the monastery of Sithiu. Apart from these painted representations, the relic of the tooth of Audomarus, permanently situated on the altar of Notre-Dame-des-Cloches, and the cenotaph of Erkembodo were meant to concretize such associations. As we have seen, Erkembodo's sarcophagus was believed to indicate the site of the Merovingian church's high altar and likely played a decisive part in funding its Romanesque reconstruction. On the other hand, Audomarus's tooth was one of a group of devotional foci inside the church relating to its founding prelate, the others being the shrines of the saint's body and head in the choir, his cenotaph in the nave, and his chalice and crosier (which also contained one of Audomarus's teeth). All of these relics were on occasion brought out together to honor the saint—for instance, during Easter services and the feast of the Deposition of Saint Audomarus.43
The altar where the tooth reliquary stood was thought to have been constructed on Audomarus's orders; this may have been the reason it was presumably spared during the Gothic rebuilding of the north transept. Further, the preservation of the apsidiole in which this altar was housed was perhaps due to the direct contact between the individual stones and Audomarus's saintly remains. According to the Jesuit Jacques Malbrancq (ca. 1579–1653), when the relics of the saint were displayed at a public ostension in 1052, they were deposited for a time on stones intended to be used in building the Romanesque church, construction of which was in full swing at that time.44 Provided that Malbrancq is to be trusted in his handling of lost documentary sources, the relics’ thaumaturgic aura might thus have been expected to rub off on the stones, creating a series of contact relics that would endow the Romanesque edifice, viewed as the founder's original building by the late Middle Ages, with considerable prestige. Not only would Audomarus have been responsible for the building of the old church and its altar, but he would also have sanctified its interior space through his relics, both physical and otherwise.45
The Saint-Omer canons’ desire to make explicit their institution's early history by retaining physical referents that could be seen and touched (if the faithful were encouraged to touch the stones exposed to Audomarus's relics) corresponds with analogous strategies observed at other medieval and Renaissance religious institutions. It also points toward what the early twentieth-century literary critic Van Wyck Brooks would call a “usable past.”46 Indeed, recent art historical work has highlighted the “afterlife” and referential potential of medieval art and architecture. In Anachronic Renaissance (2010), Alexander Nagel and Christopher Wood use the term “substitution” in discussing artistic and architectural production in the premodern world. Until and during the Renaissance (when the competing principle of “authorial performance” ultimately prevailed), they write, works of art and architecture were seen as connected to earlier originals through a chain of intermediary replicas, irrespective of their actual date. In other words, such objects were not tethered to a linear concept of time; rather, they partook of a kind of “double historicity,” understood as both the ur-objects and their later substitutes. In instances like this, time was manipulated so that a medieval or Renaissance object could effectively stand in for its earlier predecessors, reaching back to the perceived original, without courting incongruity or disbelief.47 Critics of Nagel and Wood's interpretive model have questioned the inner workings of their binary “substitution”/“performance” model, its chronological limits, and its applicability to all medieval and Renaissance works of art.48 Although “substitution” does seem applicable to the case discussed here, it sidelines the more subtle processes of patronage and intentionality in deliberately evoking a sense of the distant past.
In Die Inszenierung der Vergangenheit im Mittelalter (2003), Stephan Albrecht attempts to delve deeper into the thought processes, aims, and ambitions underlying the retrospective architectural mise-en-scènes commonly practiced by medieval religious institutions wishing to promote carefully edited versions of their own history.49 Through detailed analyses of buildings and objects as well as relevant textual testimonies, Albrecht aims to reveal the strategies by which two Benedictine communities, Saint-Denis and Glastonbury, staged fictional instantiations of their pasts, repurposing and reinterpreting particular items of material culture. Such “memorabilia” (Erinnerungsstücke) gained memorializing value due to their age and real or imagined association with important historical figures—rulers, founders, and saints (ideally, founder-saints). Once recognized, these objects—including parts of buildings—would often be divorced from their original contexts and given new lives in new environments, serving as immutable symbols (statische Zeichen) contrived to trigger memory recall. A conceptual link to the founder and a community's original building was of utmost importance in shaping and consolidating a given institution's collective memory, thus legitimating its continued existence and bolstering its competitiveness vis-à-vis rival institutions. Although Albrecht focuses on just two case studies, he demonstrates that such commemorative strategies were legion in the religious landscape of medieval and early modern Europe.50
The example of Saint-Denis near Paris is enlightening in this regard (Figure 11). A number of recent studies have convincingly shown that Abbot Suger's reconstruction of the westwork and chevet of the royal abbey's church in the 1130s and 1140s was conceived less as a bold gesture steeped in Gothic avant-gardism and more as an attempt to recast the venerable Carolingian building in a splendid new guise, adopting some of its dimensions and proportions, recycling some of the old architectural members, and imitating their early medieval forms in order to produce a harmonious and distinctly retrospective whole. Although some Merovingian marble columns and capitals had been reused as spolia in the Carolingian church, Suger attributed most of the then-extant edifice to the abbey's alleged founder, Dagobert I, disregarding evidence of the building's pre-seventh-century origins and the eighth-century reconstruction campaign. Furthermore, the legend of the miraculous consecration of the Carolingian church's nave by Christ himself, fully developed by the early twelfth century and emphatically referenced by Suger, provoked some reluctance on his part to tear down this portion of the building, which had acquired relic status due to direct physical contact with the divine. Suger's editing of his church's history around two main poles—the foundation by Dagobert and the nave's glorious consecration by Christ—exercised a powerful pull on later medieval Dionysian authors. Taking a page from the enterprising abbot's book, these authors claimed that the rebuilding of the nave in the thirteenth century, by then considered inevitable due to the building's great age, had been the structure's first alteration since its Dagobertian foundation—a contention resulting in the removal of Suger himself from the official narrative.51
The ideologically charged juxtaposition of old and new in Saint-Denis, at the dawn of the Gothic, was paralleled by analogous phenomena elsewhere. The decades around the middle of the twelfth century saw a deliberate mixing of early medieval spolia, Romanesque, and early Gothic architectural parts in several chantiers belonging to religious institutions in Paris and across northern France. Such mixing was geared toward celebrating these foundations’ antiquity and royal patronage and tapping into the considerable prestige that came with such associations.52
At Canterbury Cathedral, the restoration of the east end (damaged by the great fire of 1174) under the successive master masons William of Sens and William the Englishman entailed the retention of the crypt and outer walls of the “Glorious Choir” begun under Archbishop Anselm in 1096 and completed by 1130. The new Gothic central vessel was inserted into the shell of the Romanesque structure, while an effort was made to merge formal vocabularies and achieve some degree of visual continuity between old parts and new. Aesthetic homogeneity was not an end in itself, however; it appears that there was also concern for maintaining as much as possible of the building's prefire sacred topography. According to Gervase's treatise on the events surrounding the fire and the subsequent repairs, the cult of the Anglo-Saxon (and later) saintly archbishops whose tombs were located in the transepts and near the high altar had to continue unabated at the sites prescribed by ancient custom. Thus, even though the older, Romanesque edifice here was not explicitly tied to a single institutional founder at the time of its rebuilding, it was nevertheless bound to a corporate body of local holy figures and the site-specific rituals honoring their memory.53
In the Holy Roman Empire, the east end of Magdeburg Cathedral was rebuilt from the late 1200s with a French-inspired chevet plan (featuring an ambulatory and radiating chapels) and a more traditional Romanesque elevation. This appears to have been one of the earliest sustained Gothic interventions in the design of a great church in the German lands, and the novelty was compounded by the slight change in orientation between the old and the new structure and the suppression of the crypt. The late Roman granite and porphyry columns inserted in the angles between the bays of the hemicycle, retrieved from the cathedral's Ottonian predecessor following its destruction by fire, and other distinctive design elements retained from the old building all conjured the memory of the old Reichskathedrale and its founder, Emperor Otto I (d. 973), who had been buried in the east end, as had his spouse, Editha.54
The development of the Gothic style north of the Alps, especially in its Rayonnant variant, from the thirteenth century onward promulgated a brittle, linear aesthetic predicated on relentlessly modular and unified design principles that were not conducive to the incorporation of earlier architectural materials.55 Nevertheless, there were other ways to visualize an institution's past, ways that did not always rely on the use or imitation of actual spolia. The conception and execution of the Saint Wenceslas Chapel in the south transept of Prague Cathedral under Emperor Charles IV (ca. 1358–67, with additions in the early 1370s) is a case in point (Figure 12). The cathedral's second master mason, Peter Parler, designed the oratory in honor of Bohemia's major patron saint as a self-contained unit not conforming to the regular geometrical grid that defines the rest of the church's plan. Essentially a church within a church, the Wenceslas Chapel was built above the tomb of the canonized Duke of Bohemia in a retrospective style quite unlike the virtuoso, up-to-date Gothic that Parler employed in the rest of the edifice. The combination of its square plan, dark interior, round arches, and eccentric features, such as the dado ornament of shimmering semiprecious stones and its figured vault, defied contemporary design orthodoxies and, in some respects, approximated Romanesque models. Consequently, even though no part of the tenth-century rotunda founded by Wenceslas (and subsequently housing his mortal remains), or even the Romanesque basilica that succeeded it, was preserved in the fourteenth-century structure, the memory of the original building lived on in the historicist architecture of the centrally planned, virtually autonomous chapel. This space's “otherness” and simulated antiquity fit in well with its ceremonial role during royal coronations, when new Bohemian rulers were associated with their predecessors through the ritual animating the cathedral's sacred topography.56
Prague Cathedral's stylistic pluralism, usually attributed to the remarkable invention of Parler, acting on the wishes of Charles IV, may be contextualized by a series of comparable cases of architectural historicism observed throughout Latin Christendom from the latter half of the fourteenth century onward. The revival of local Romanesque and early Gothic models in Ireland, Scotland, the Holy Roman Empire, and Tuscany in this period—to mention but a few of the more extensively studied examples—has occasionally been interpreted as emanating from swelling “nationalist” feeling that made local claims on earlier artistic styles thought to represent the halcyon days of particular polities, cultures, or ethnic and social groups.57 In these circumstances, architectural retrospection was often less an ad hoc solution, tailored to the commemorative needs of any given religious institution, and more a cohesive visual language employed to underscore larger shifts in identity and self-perception. In its embodiment of a generic sense of “antiquity” (perceived as temporally distant from the Gothic present) and its diverse localized variations, the Romanesque became the “go-to style” for nostalgic exaltation of the romanticized local past, whether institutional, ethnic, or otherwise. This is the broader context in which the suggestive juxtaposition of old and new, Romanesque and Gothic, at Saint-Omer should be seen.
Conclusions: The Romanesque and the Staging of the Past at Saint-Omer—a Road Map to the Future?
In a recent essay, Bernd Schneidmüller calls attention to the role of medieval historiography in promoting visions of the past that serve the needs of the present, in part by undermining the strict distinction between historical reality and fiction long espoused by positivist scholarship.58 Art and architectural historians challenged by the composite, multilayered fabric and multivalent spaces of medieval religious buildings have begun to realize the importance of addressing issues of stylistic and formal choices through the lens of local, primarily institutional history—however partial, contrived, or mythologizing—and its topical concerns for legitimation or primacy. In cases such as those at Saint-Denis and Canterbury, interpretation of the material evidence has relied on a corpus of well-known texts detailing and often justifying the choices made by patrons and master masons in terms of architectural preservation or renewal. In other instances, such as that of the collegiate church of Saint-Omer, the relevant information lies scattered across archival documents (frequently unpublished) and tends to be more oblique—yet it is no less relevant to a wider evaluation of how medieval religious institutions saw themselves and their pasts at different times.
The preservation and recontextualization of the Romanesque apsidiole of the chapel of Notre-Dame-des-Cloches at Saint-Omer complies with the overall pattern of “creative” history writing observed elsewhere. By the first half of the fifteenth century, the still-lingering parts of the Romanesque church were being “read” as the last vestiges of the original, seventh-century building founded by Saint Audomarus himself, in the manner already discussed for Saint-Denis and Magdeburg; the elision of the intermediary eleventh- to twelfth-century phase, to which the remains actually belonged, recalls the consignment to oblivion of the Carolingian church at Saint-Denis in favor of that of Dagobert, considered by the time of Suger to have been the abbey's founder. The apsidiole of Notre-Dame-des-Cloches was thought to shelter the Merovingian high altar, before which Saint Erkembodo, Audomarus's successor, had been buried; furthermore, from at least the early 1460s, the reliquary of the founder's tooth was permanently sited on that altar. The presence of the cenotaph of Saint Erkembodo and the tooth of Saint Audomarus—commemorated in the painted figures of the two prelates displayed in the chapel after its refurbishment—and, perhaps, the sanctification of the apsidiole's masonry through contact with the founder's relics sealed the structure's bond with local sainthood and showcased divine intervention, much as at Saint-Denis, Canterbury, and Prague. Here, as in several of the examples surveyed above, the Romanesque was charged with portraying the distant past in contradistinction to the Gothic, a trend now more readily recognizable in the Burgundian Netherlands in the work of early fifteenth-century painters, such as the Master of Flémalle and Jan van Eyck.
Even if the strategies employed to render institutional history visually explicit at the collegiate church of Saint-Omer were common elsewhere at that time, the specific reasons behind Saint-Omer's staging of the past remain unclear. The chapter's bitter rivalry with the monks of neighboring Saint-Bertin—over which of the two institutions possessed the “true” relics of Saint Audomarus—certainly played a far greater part in the canons’ attempts to amplify the founder's presence in the church's sacred space than any contemporary political or other event. Nevertheless, in spite of this dispute's longevity, from the mid-eleventh to the late fifteenth century and beyond, few traces of it appear in the documentary record between the 1320s and the 1450s—that is, the years in which one would expect the Romanesque church to have been identified with its Merovingian forebear.59 It is possible that opposition between these sibling institutions reinvigorated the founder's cult and intensified efforts to secure his presence at the collegiate church, culminating in the revival of the seventh-century church within its new Gothic shell.
Ultimately, the case of Saint-Omer is but one among many instances of the architectural representation of institutional history in the Middle Ages. As Albrecht has demonstrated, the examples mentioned above could well be multiplied. With Saint-Omer as a springboard, the way forward will involve the concurrent examination of architectural fabric and any pertinent textual evidence, some of which may still lie virtually unexplored in account books, charters, and other manuscripts filed away in local archives.
The possibility of gaining a fuller picture of medieval religious institutional life—its collective concerns, aspirations, and ambitions for the present and the future—through textual and architectural investigations shows considerable promise, especially when these investigations are conducted on a pan-European scale. Perhaps more important, such research efforts offer fertile ground for collaboration among art, architectural, and documentary historians seeking release from the constraints of traditional disciplinary boundaries and movement toward a more holistic appreciation and reconstruction of the past—an endeavor that would have struck a resonant chord with their medieval predecessors.