THE CHAPEL OF ST CATHERINE AT THE CISTERCIAN ABBEY OF SAVIGNY: ‘UNEARTHING’ AN ARCHITECTURAL ENIGMA

Abstract

This article explores the history of St Catherine’s chapel at the abbey of Savigny, head of Normandy’s only monastic congregation. Built in the twelfth century, the chapel was, at the time of its demolition in 1705, the oldest remaining part of the medieval monastic complex. It therefore appears fairly regularly in the written record and has attracted not an insignificant amount of attention as a result. That said, the near total destruction after 1789 of Savigny’s buildings, and the often contradictory nature of those written sources by which antiquarians and academics have attempted, in the absence of sustained archaeological work, to reconstruct their medieval layout, mean that a great deal remains uncertain. St Catherine’s is no exception to this rule. Its precise location and design have to date been matters of conjecture, while a great deal of what has been written about it is either inaccurate or inconsistent (or both). This article brings together for the first time all the available references to (and scholarly discussions of) the building. It combines the findings of recent archaeological work with a reassessment of the written sources to argue that the chapel’s location within Savigny’s monastic precinct was almost unique in the Cistercian world, with its closest parallels being found instead in the Cluniac one. These circumstances were born more of accident than design, but they nevertheless presented challenges for Savigny’s medieval community, the consequences of which help shed light on wider issues relating to the use and reuse of Cistercian monastic spaces.

INTRODUCTION

The chapel of St Catherine at the Cistercian abbey of Savigny, ¹ head of Normandy’s only native monastic congregation, is an edifice that is at once well and little known. Built in the final quarter of the twelfth century, and named in honour of one of the most popular saints in medieval Europe, ² the chapel was, at the time of its demolition in 1705, the oldest remaining part of the medieval monastic complex. Prior to its destruction, the space had served various important functions, especially during the central Middle Ages, when it was home, among other things, to the tombs of various benefactors and, until 1243, the relics of Savigny’s ‘saints’. ³ As such, it appears fairly regularly in the written record and has attracted not an insignificant amount of academic attention as a result. That said, the near total destruction after 1789 of Savigny’s buildings, and the often contradictory nature of those written sources by which antiquarians and academics have previously attempted, in the absence of sustained archaeological work, to reconstruct their medieval layout, mean that a great deal remains uncertain. The chapel of St Catherine is no exception to this rule. Its precise location and design have to date been matters of conjecture, while a great deal of what has been written about it over the last three centuries is either inaccurate or inconsistent (or both).

The aim of this article, written in the context of preparing a critical edition and translation of Savigny’s book of miracles, ⁴ in which the chapel features centrally, is therefore to bring together for the first time all the available references to (and scholarly discussions of) the building. In doing so, it will combine the findings of recent archaeological work with a reassessment of the written sources to argue that the chapel’s probable place within Savigny’s monastic precinct was almost unique in the Cistercian world, with its closest parallels being found instead in the Cluniac one. These circumstances were born more of accident than design, but they nevertheless presented certain challenges for Savigny’s community, the consequences of which help shed light on wider issues relating to the use and reuse of Cistercian monastic spaces.

A BRIEF HISTORY OF SAVIGNY, 1112–1243

Before turning to such matters, however, it seems prudent to begin with a brief history of the house in which the chapel of St Catherine was located. Savigny’s early years have been the focus of scholarly attention since the early eighteenth century, when Dom Claude Auvry, prior of the abbey from 1698 to 1712, penned his Histoire de la congrégation de Savigny. ⁵ Established in 1112 by the hermit Vitalis (d. 1122), a contemporary of Robert of Arbrissel (d. 1116), the early Savigny community received its principal lay support from Stephen, count of Mortain (1112–35), whom Vitalis had served as chaplain, and Ralph i of Fougères (d. 1124), who granted the forest in which the abbey was located, thereby earning his family the rank of ‘founders’. In ecclesiastical terms, Savigny’s foundation took place within the wider context of the monastic revival of the late eleventh and early twelfth centuries, which saw the creation of various new orders, including, of course, that of Cîteaux. Like monastic reformers elsewhere, Vitalis had embraced an itinerant existence, and Savigny’s early history is characterised more by his prowess as a wandering preacher than its development as a formal institution.

It was only with the arrival of Vitalis’s successor, Abbot Geoffrey of Bayeux (1122–38/9), that both the abbey and the congregation it founded began to take shape. According to contemporaries, it was Geoffrey who imposed strict discipline upon the monks, for whom he helped acquire or create various texts to regiment their existence. ⁶ It was also under Geoffrey that the Savigniac filiation began rapidly to expand, especially in England, where as many as fourteen daughter houses had been established by the middle of the twelfth century. ⁷ Although impressive, such expansion was by no means without complications, and Geoffrey’s successor, Abbot Serlo (1140–53), found himself confronted with trying to manage restive daughter houses that were themselves struggling to negotiate the conflict between King Stephen (1135–54) and the Empress Mathilda (1102–67). The debate surrounding when and why Serlo decided to merge his congregation with that of Cîteaux has been both long and contentious, but it is now generally accepted that the incorporation itself took place in 1147, with Savigny becoming a daughter house of Clairvaux. ⁸ As for its consequences, even if it has been argued that Serlo did not obtain for the abbot of Savigny the exalted position once thought, ⁹ the abbey did retain for itself peculiar practices, particularly in the realm of finance, which have since been identified as a root cause of the later corruption of early Cistercian ideals. ¹⁰ In terms of Savigny itself, by the beginning of the thirteenth century it had seen its possessions and buildings greatly enlarged and improved by a succession of capable abbots, two of whom had risen to become abbots of Cîteaux. Much like Abbot Geoffrey in the second quarter of the twelfth century, the reign of Abbot Stephen of Lexington (1229–43) some one hundred years later was characterised by an energetic programme of reform and improvement. ¹¹ This culminated in the formal translation of the relics of Savigny’s ‘saints’ on 1 May 1243 to the main abbey church from the chapel of St Catherine, an event central in its history and one we shall return to frequently throughout this article.

ORIGINS

For now, having set something of the wider context, let us turn to try and establish the chapel’s origins within the chronological and architectural framework of Savigny’s early years. According to an interpolation in a Savigny copy of the chronicle of Sigebert of Gembloux (c 1030–1112), the first abbey church, begun by Vitalis, was completed by his successor, Geoffrey of Bayeux, and dedicated in the presence of five bishops on 1 June 1124. ¹² It is often said that this structure was of wood, ¹³ although no evidence survives to corroborate this, with a later miracle account suggesting that it may have in fact been of stone. ¹⁴ Whatever the case may be, it is clear that the primitive abbey was soon deemed unsuitable for the community’s needs (the aforementioned miracle says as much), such that work on a new church was begun under Abbot Jocelin (1165–78) in either 1173 or 1174. ¹⁵ It is also from Jocelin’s reign that we find the earliest references to the chapel of St Catherine, with at least one source, discussed in more detail below, suggesting the edifice already existed in some form by 1172/3. What this was is unknown, but a charter of Andrew ii of Vitré (1173–1211), issued after the death of his father, Robert iii, in late 1173, ¹⁶ suggests it was substantial enough to be desirable as a place of burial, since Andrew granted revenue to the monks initially ‘towards the work on their new church’, and, ‘when this is finished, to lighting the chapel of St Catherine in which the tomb of lord Robert of Vitré lies’. ¹⁷ What is more, the chapel was soon ready to be dedicated, with this service being performed by Richard iii l’Évêque, bishop of Avranches (1170–82), on 27 May 1181. ¹⁸ Ralph ii of Fougères (d. 1191/4) attended, making a generous donation to mark the occasion. ¹⁹

According to the preface to the aforementioned Savigny miracles, written in 1243/4, Ralph ii was also at hand when the relics of the abbey’s five principal saints (Vitalis, Geoffrey of Bayeux, Peter of Avranches, Hamo of Landecot, William Niobé) were translated ‘from the old church, which at that time had been demolished’ (a veteri ecclesia, que diruta fuit tunc temporis) to the chapel of St Catherine, where they were placed ‘in a single stone tomb, fittingly carved’ (in uno sepulcro lapideo, decenter exciso). ²⁰ Performed by Abbot Simon (1179–84/5) and Peter, abbot of Clairvaux (1179–86), the translation is an event around which there has been much confusion, primarily with regard to the date at which it took place. Unfortunately, the preface to the miracles gives no chronological information outside that which we now know to be the reigns of abbots Simon and Peter. According to the editors of the Gallia Christiana, however, the translation was held on 30 March 1182 (30 Martii feria iii Paschæ 1182), although the source on which they based this claim is not identified. ²¹ Claude Auvry, on the other hand, argued it took place in 1181, claiming (again, without citing any sources) that Abbot Peter had visited Savigny that year ‘according to his custom’ (selon sa coutume), while noting that 30 March was also the date traditionally associated with the pilgrimage to venerate the abbey’s saints. ²² Subsequent scholars have tended to prefer the 1181 date, ²³ although not without sometimes muddying the waters further. Thus, Hipployte Sauvage (1823–1914) considered the translation of 1181 to be the second such event in Savigny’s history, the first having occurred when Vitalis’s relics were brought into the primitive abbey church in the presence of the bishops of Avranches, Le Mans and Rennes. ²⁴ The source on which he based this claim, however, namely Nicolas-Hugues Ménard’s (1585–1644) Martyrologium sanctorum, associates the event it describes with the later translation, orchestrated by Abbot Stephen of Lexington, which was commemorated on 1 May. ²⁵ Elsewhere, Louis Raison (1885–1943) believed Ménard’s description applied to the translation performed under Abbot Simon, arguing that the presence of the unnamed bishop of Rennes, whom he identified as Philip (1178–81), confirmed the 1181 date, since Philip died shortly after the translation at which he had supposedly been present. ²⁶

If the assertions of Sauvage and Raison can be reasonably called into doubt, it remains unclear as to the translation with which we should associate Ménard’s description. Despite Ménard’s own claims, it would seem not to be the later translation organised by Stephen of Lexington, since the Savigny miracles, written in the immediate aftermath of the event itself, name only Geoffrey, bishop of Sées (1240–58), as being present. ²⁷ Émile-Auber Pigeon (1829–1902) believed the three bishops named by Ménard were present at the translation held under Abbot Simon, which he dated 30 March 1182, a claim repeated, somewhat confusingly, by Hipployte Sauvage, ²⁸ and one also accepted by the most recent work to deal with Savigny’s relics. ²⁹ Sadly, nothing allows us to corroborate this, and the only medieval description of the event, found in the Savigny miracles, is notable for the fact that it mentions no bishop. ³⁰ It does record, however, that Simon and Peter were accompanied by ‘certain of their co-abbots’ (quibusdam coabbatibus suis), ³¹ something which may indicate that the translation was staged in the wake of a large meeting similar to that held a few years earlier at which Alexander, abbot of Cîteaux (1168–78), had been present. It has been suggested that this gathering was a reunion of the Savigny general chapter, ³² a meeting that, had it been held in 1181, would have been convened just five days after the dedication of the chapel of St Catherine on 27 May. ³³ Whatever the case may be, there is enough uncertainty for us to be able to say only that the translation most likely took place at some point after the chapel’s formal dedication on 27 May 1181, and certainly before the end of Simon’s abbacy in 1184/5. As noted above, however, it seems that the chapel was in use well before this date. The sole source for this is the vita of Peter of Avranches, which records that the monk Hamo of Landecot saw Peter in a vision when he was ‘in devoted and faithful prayer in the chapel of Blessed Catherine’ (in capella B. Catharinæ orationi devote ac ferventer incumberet). ³⁴ Hamo died on 30 April 1173, ³⁵ and it has been argued that Peter’s vita was probably written by someone familiar with both men’s experiences and regimens. ³⁶ That said, if the chapel seems to have existed from 1172/3, whether it was dedicated to St Catherine from the outset remains unknown.

LOCATION

If establishing something of its origins is not without its difficulties, then much the same is true when it comes to determining the place occupied by the chapel of St Catherine within Savigny’s monastic complex. As with the dedication and translation, this is a topic that has attracted a good deal of antiquarian and academic attention, although much of it contradictory in nature. The chapel’s destruction in 1705 means that it does not appear in any of the later descriptions or plans of the monastic complex, drawn up in the second half of the eighteenth century and first half of the nineteenth, details from which have recently been brought together and analysed by Jean-Baptiste Vincent and Adrien Dubois. ³⁷ The near total destruction of Savigny’s buildings, and the lack of any sustained archaeological work on the site in the modern era, mean that those attempting to reconstruct the organisation of the monastic complex have nevertheless relied on these early-modern documents in their efforts to reconstruct Savigny’s medieval layout. This is also true with regard to what few early-modern references we have to St Catherine’s chapel, with these descriptions being at the origin of certain assumptions and assertions, some of which have acquired the status of established but unverified truths, repeated as much in antiquarian works as academic ones.

The inherent difficulties of working with these references can be illustrated by two examples from the seventeenth and early eighteenth centuries. The first is found in the history of Savigny written by Claude Auvry. In a sentence immediately following an allusion to the dedication of the chapel by the bishop of Avranches, the former prior of Savigny goes on to say that:

Il y avoit déjà quelques années qu’on avait ruiné et démoli la première église de Savigny, dont une partie étoit tombée elle-même; ³⁸ mais afin que les religieux ne fussent pas sans avoir un lieu convenable pour y faire le service divin, l’abbé Joscelin, qui gouvernoit en ce temps-là, avoit fait bâtir un oratoire près du cimetière de l’abbaye, où l’on faisoit les exercices de piété qu’on fait ordinairement dans les autres églises. Ce fut dans cette chapelle que le bienheureux Hamon de Landacop, religieux de Savigny, étant en prières, vit au nombre des saints S. Pierre d’Avranches, son confrère, tout eclatant de lumière. ³⁹

In this, we would seem to have evidence, written by someone who knew the chapel of St Catherine before its destruction in 1705, ⁴⁰ suggesting that the building was ‘near the abbey’s cemetery’, which, according to a plan of 1795, was located on the north side of the abbey’s east end (fig 1). ⁴¹ Sadly, whatever certainty we might think to find here soon turns illusory upon closer inspection. In the first instance, the cemetery was located quite some distance from where we would expect to find a structure we know later formed an integral part of the monastic complex, both medieval and early modern. What is more, it is difficult to reconcile Auvry’s assertion with an earlier seventeenth-century reference to St Catherine’s. This can be found in Arthur du Monstier’s (1586–1662) Neustria pia, which, in describing the tomb of Abbot John le Verrier (1390–1405), notes that it was located ‘in Capella S. Catharinæ, quæ est post dormitorium’. ⁴² It is this assertion that was repeated without attribution by Jean-Jacques Desroches (1797–1862), ⁴³ whose own citation was taken up by Hippolyte Sauvage (again, without express attribution). He then seems to have combined this with Auvry’s claim, which, in turn, led him to argue that the chapel ‘était au sud des dortoirs et attenante au cimetière’, ⁴⁴ an idea that has been repeated by various modern scholars. ⁴⁵ As for du Monstier’s claim, which is not without its own complications, we shall return in due course to look at its implications for what we can say about the chapel.

Fig 1. 1795 map of the abbey of Savigny. BnF, Cartes et plans, H 189097. Image: reproduced with permission from the Bibliothèque nationale de France.

For now, let us turn to examine the medieval references both to the building itself and to what has been traditionally referred to as the ‘cloister’ of St Catherine, a space as enigmatic as the edifice after which it was named. Besides mentions in charters of the sort noted above, the chapel of St Catherine appears in two main medieval texts: the book of Savigny miracles, written in 1243/4, and the so-called Chronicon Savigniacense, long-known by an extremely poor seventeenth-century printing but now recently edited to critical standards. ⁴⁶ Like some of the diplomatic evidence, the latter of these deals exclusively with burials in and around the chapel, which, as home to the relics of Savigny’s saints until 1243, served as one of the abbey’s key intercessory spaces. ⁴⁷ The remains of Robert ii of Vitré were therefore soon joined by those of his sons, while the so-called Chronicon records that the chapel was the last resting place of Goranton v of Vitré, a member of an older, unrelated lineage, known today as the Goranton-Hervé de Vitré, who was buried there on 26 December 1241. ⁴⁸ Besides being interred in the chapel, lay benefactors were also buried in the aforementioned ‘cloister’ bearing its name. Of these interments, the most interesting is that of Nicholas Avenel, lord of Chalandrey, who, on 17 January 1242, was laid to rest ‘in claustro Sancte Katerine versus infirmitorium’. ⁴⁹ This is important for a number of reasons, not least because it is the only known medieval reference to the chapel of St Catherine that situates it in relation to another building within the monastic complex. It is also one that, to our knowledge, has never been fully taken into account in previous work on the chapel. ⁵⁰

Before looking at the implications of this in more detail, however, it is necessary first to dwell on the so-called ‘cloister’ of St Catherine itself. As the above makes clear, this was a space large enough to house burials, but its form and location remain unknown. Much hangs, of course, on how we interpret the term claustrum, which was not always used to mean cloister in the strict sense. ⁵¹ As for the term as it is used here, Lindy Grant took this to mean that the chapel of St Catherine ‘had its own cloister’, ⁵² a suggestion repeated by Véronique Gazeau and Cécile Chapelain de Seréville-Niel. ⁵³ Julien Bachelier, on the other hand, has suggested that we should see in the term nothing more than an ‘enceinte fermée’. ⁵⁴ As we shall see, there is evidence from the Savigny book of miracles that can be interpreted to support the latter of these views. There is also evidence to suggest that the chapel was somehow linked with the chapter house, something that rather complicates the incorporation of any substantial architecturally cloistered space within any proposed reconstruction of the chapel of St Catherine itself.

As for how the chapel has come to be associated with the chapter house, there are two strains of evidence: one textual, the other archaeological. Of these, the textual is somewhat problematic. Besides a late (and previously unknown) reference to what is now likely a lost document of 7 August 1708, which apparently described St Catherine’s as being ‘située dans le petit jardin des religieux, derrière leur chapitre’, ⁵⁵ the idea that the chapel and chapter house shared some connection seems to have first been raised by Jean-Jacques Desroches, who, having noted that Abbot John le Verrier was buried in St Catherine’s, goes on immediately to say that ‘cette chapelle servait de salle du chapitre aux religieux’. ⁵⁶ It is a claim that has been repeated by antiquarians and academics alike. ⁵⁷ Desroches gives no source for this information, but its origins can be reasonably deduced. Indeed, as we have seen above, we know that Desroches based his claims as to the location of John le Verrier’s tomb on the (unattributed) work of Arthur du Monstier. We also know, thanks to citations elsewhere in the same article, that Desroches, like anyone working on the ecclesiastical history of France then and now, had consulted volumes of the Gallia Christiana. ⁵⁸ The first edition of this work, published in 1656, is silent as to the location in which John le Verrier was buried, ⁵⁹ but a later volume, revised and updated by the Maurists, and first published in 1759, notes that the abbot was interred ‘in capitulo’. ⁶⁰ Since there is no other known mention of the chapel of St Catherine being used for chapter meetings before Desroches, it would seem that he, aware of the two differing locations for le Verrier’s tomb, nevertheless sought to square this particular circle by either wilfully or mistakenly ignoring the chapel’s destruction in 1705 and conflating the two together.

Of course, if we accept this hypothesis, and thereby dismiss the idea that the chapel of St Catherine was ever used to hold chapter meetings, the question remains as to whether either Arthur du Monstier or the editors of the Gallia Christiana are mistaken in their claims. To help answer this, we can turn to a number of eighteenth-century drawings, today conserved at the Bibliothèque nationale de France in ms français 4901. This manuscript contains sketches of four tombs once found at Savigny. On fol 190r are drawings of the tombs of the aforementioned John le Verrier and of a certain Jean de Landevic, the inscription on which records that he died on 23 March 1414. The other two drawings are found on fol 192r. They show the tomb of the lords of Vitré and Dinan, and that of William of Saint-Brice, who died in 1318. Beneath these drawings, we find the following text:

Ces deux tombeaux sont dans le mur, du costé de l’epitre, d’une ancienne chapelle qu’on pretent avoir esté la premiere eglise de l’abbaye de Savigny, et les deux autres sont dans une petite chapelle, de l’autre costé.

Since we know that the lords of Vitré were buried in the chapel of St Catherine, and that their remains were interred in a single tomb, ⁶¹ it would seem that these drawings capture what was in the space before 1705. They therefore help confirm du Monstier in his assertion concerning the tomb of John le Verrier. What is more, the above description also shows that, by the beginning of the fifteenth century, the chapel of St Catherine was home to a small side chapel, located on its Gospel (right-hand) side, and founded, according to the inscription on his tomb, by Jean de Landevic, ⁶² with the Vitré/Dinan and St Brice tombs apparently in niches in the wall on the Épître (Epistle, or left-hand) side.

This is important not only because it allows us to verify du Monstier but because it gives a sense of the scale of the chapel of St Catherine, which was large enough to accommodate not just tombs of fairly substantial proportions, ⁶³ but also a separate side chapel built within its footprint. By the early seventeenth century, the chapel was also home to the tomb of Abbot Claude du Bellay (1588–1609), a fact noted by Arthur du Monstier, who chose in this instance to describe St Catherine’s as ‘the small church’ (minor ecclesia). ⁶⁴ As for the Gallia Christiana and John le Verrier, we know that St Catherine’s was either destroyed as a result of the fire that swept through the abbey on 12 August 1705 or was knocked down in order to rebuild the dormitory. ⁶⁵ If the latter were the case, thereby leaving the interior undamaged but nevertheless at threat, then it is likely that a tomb like le Verrier’s was relocated from the chapel into the chapter house. ⁶⁶ This, after all, was the traditional resting place of Cistercian abbots, and the move would have posed little difficulty had the two structures been somehow connected.

As the above suggests, however, the reliability of what textual evidence there is for any connection between the chapel of St Catherine and the chapter house can be called into question. Fortunately, recent archaeological work on the site helps lend significant support to the idea, with recent geophysical surveys (the first ever performed) revealing a rectangular anomaly measuring around 10m in width and 6m in length, protruding beyond what would have been, after August 1705, the eastern wall of the chapter house (fig 2). Due to difficulties with the current terrain, the results of the survey do not capture the entirety of whatever structure was once there. Nevertheless, its size and position has led Jean-Baptiste Vincent to suggest that it is most likely the foundations of St Catherine’s chapel, which he argued would have formed an annexe to the chapter house. ⁶⁷

Fig 2. Geophysical survey of Savigny site (2016). Image: reproduced from Vincent 2019.

As we shall see, this hypothesis is the most convincing to date as to the chapel’s former location. It is not without its problems, however, both in relation to the written evidence and Vincent’s wider reconstruction of the medieval Savigny site (fig 3). In the first instance, unlike another recent suggestion as to the building’s location, which placed it in the south range between the refectory and the hostelry, ⁶⁸ the idea that the chapel of St Catherine extended to the rear of the chapter house corresponds with Arthur du Monstier’s description of it being ‘behind the dormitory’, since this was itself located in the northern part of Savigny’s huge east range (105.5m long and 12.2m wide). ⁶⁹ Siting the chapel there would also bring it closer to the monks’ cemetery, which Claude Auvry claimed was nearby. As noted, the 1795 plan shows this confined to the northern side of the abbey’s east end, and thus still some distance away, although this document is not always an accurate guide. ⁷⁰ As such, if the cemetery extended around to the southern side of the east end, then it would have been just to the north of St Catherine’s chapel. What is more, adjoining the chapel to the rear of the chapter house even allows us to see why the idea that it was also used for chapter meetings gained currency.

Fig 3. Proposed reconstruction of medieval Savigny site (twelfth to thirteenth centuries). Image: reproduced from Vincent 2019.

Locating the chapel here is not without its problems, however, in particular when one remembers the reference to the abovementioned burial ‘in claustro Sancte Katerine versus infirmitorium’. First of all, in Vincent’s reconstruction of the medieval complex, the infirmary is located to the south of the refectory and thus over 100m away from the proposed chapel site. As such, either this proposal is wrong or we must look to situate the medieval infirmary elsewhere. Rather confusingly, the second of these ideas is entertained by Vincent himself, who suggests (once as a possibility and a second time as a certainty) that the southern part of the east range contained an infirmary. ⁷¹ If this were the case, then it is easy to square the mention of a burial in the ‘cloister’ of St Catherine ‘towards the infirmary’ with the proposal that the chapel formed an extension of the chapter house. That said, choosing to locate the chapel here means we must return to the ‘cloister’ itself and to the question of what sort of structure this was. One possible solution is that the ‘cloister’ of St Catherine was not a separate edifice or space but simply the name given to the main cloister’s eastern walk, off which the entrance to the chapter house and, by extension, the chapel itself, was located. The problem with this idea is that we also have medieval references to burials ‘in claustro ante capitulum’ that otherwise make no mention of St Catherine’s. ⁷² With this in mind, we might therefore wish to see the ‘cloister’ as a separate structure in the traditional sense of an arcaded quadrangle. The columns of three such arcades from Savigny are to be found today at the château of Les Louvellières (fig 4). ⁷³ Traditionally thought to have come from the abbey’s main cloister, ⁷⁴ they probably belonged to St Catherine’s chapel, ⁷⁵ although most likely to its interior rather than any cloistered quadrangle, since no trace of such a structure was revealed by the geophysical surveys noted above.

Fig 4. Savigny columns at the château of Les Louvellières. Image: reproduced with permission from the Archives départementales de la Manche.

In order to arrive at something of a conclusion, it is necessary to turn to an account in the Savigny book of miracles. Although this has not escaped notice, it has never been discussed in any great detail and is typically cited via a highly abbreviated nineteenth-century printing. Given such circumstances, and given its importance not just in this context but in relation to a range of other issues discussed below, it is worth reproducing here in full:

Die lune ante Ascensionem, hora tertia, armiger quidam, Johannes nomine, frater domini Guillelmi de Exclusa, dum esset in claustro Sancte Katerine, prospexit in pratellum, et vidit de sub tumba in qua sanctorum corpora jacuerant fumum igneum in maxima quantitate exeuntem de terra et ascendentem per vitream fenestram, ut sibi videbatur, et dixit Garino, fratri suo: ‘Videsne fumum quem video?’ Et dictus Garinus videre non potuit, et statim fumus disparuit. Et intravit dictus Johannes majorem ecclesiam. Paulopost rediit in claustrum et vidit iterum fumum, facientem reflexum ad locum unde prodierat. Eundem vero fumum vidit frater Rad(ulfus) de Corce, monachus Savign(eii), qui ibidem presens erat. Qui vero astabant intra capellam Sancte Katerine ea hora qua fumus visus est, mira odoris fragrancia respersi fuerunt. Et notandum est quod dictus Johannes ante hujus visionem quasi incredulis extiterat eorum que dicebantur de sanctis, qui postea valde edificatus recessit, et dixit dum egrederetur portam, quod pro equo suo vel multo majori precio non vellet, quin ea die ad Savign(iacum) venisset. ⁷⁶

[On the Monday before Ascension Day, ⁷⁷ at the third hour, a certain armour bearer, named John, brother of lord William of L’Écluse, ⁷⁸ while he was in the cloister of St Catherine, looked upon the pratellum, and saw that under the tomb in which the saints’ bodies had lain fiery smoke was emerging in large quantities from the ground and was ascending through the glass window, or so it seemed to him. And he said to Warin, his brother: ‘Do you not see the smoke I see?’ And the aforesaid Warin could not, and at once the smoke disappeared. And the aforesaid John entered the great church. A little later, he returned to the cloister and saw the same smoke, making so as to return to the place from which it had emerged. Ralph of Courcy, ⁷⁹ monk of Savigny, who was there present, saw the same smoke. Those who were standing within the chapel of St Catherine at the hour when the smoke was seen were sprinkled with a wonderful fragrant odour. And it is to be noted that the aforesaid John, who before his vision had been sceptical of those speaking of the saints, left afterwards greatly edified, and said while going out of the gate that he would not wish for his horse or any greater price so as not to come to Savigny that day.]

As the English text above suggests, the key to determining what was understood as the ‘cloister’ of St Catherine hinges on how we choose to translate the word pratellum. Like claustrum, this had various meanings in the Middle Ages, including the generic ‘small meadow’. ⁸⁰ In this context, however, one possible interpretation is ‘garth’. If correct, this brings to mind a scene in which John of L’Écluse looked upon an open, grass-covered court, itself surrounded by a covered walk in which he was standing. Alternatively, pratellum could simply mean ‘grassy area’, a translation that would be appropriate had the ‘cloister’ of St Catherine been nothing more than the ‘enceinte fermée’ suggested by Julien Bachelier.

Somewhat frustratingly, the miracle story in itself does not allow us to settle the matter beyond doubt, although its narrative does help further contextualise things. In the first instance, John is said to have entered the abbey church from the ‘claustrum’, only to have returned there ‘a little later’ (paulopost). Had he been standing in the main cloister, of which part was perhaps named after the chapel to the rear of the chapter house, then he would have been easily able to access the main church (and return to where he had been) by the door leading into the nave from the cloister’s north walk. But the issues noted above make this unlikely. So does the statement that John saw the miraculous smoke ascend ‘through the glass window’ (per vitream fenestram). Unfortunately, the precise location of this window is not stated, but, given that those in St Catherine’s are said to have experienced ‘a wonderful fragrant odour’ (mira odoris fragrancia), presumably understood to have been caused by the smoke, then it seems reasonable to assume that the window in question belonged to the chapel. ⁸¹ Assuming this lay to the rear of the chapter house, then the window in question must have formed part of its eastern façade. This, in turn, means that John was most probably standing to the chapel’s east in its associated claustrum, which was itself nothing more than a plot of land somehow delineated, although not by any kind of monumental architectural structure (at least not one large enough to register geophysically), the pratellum of which we should most probably understand to be nothing more than a ‘grassy area’ and its associated claustrum as a ‘close’. Here, John would have found not only the stone tomb used previously to house the relics of Savigny’s saints (a point we shall return to below), but also various lay burials, including that of Nicholas Avenel, who had been laid to rest just over a year earlier in the southern part of the close, towards the infirmary. From here, John would have faced quite some walk to gain access to the abbey church (another point we will look at below), but his perambulation around the site speaks to a larger issue of lay access to the claustral complex and the role played by the chapel of St Catherine in it, one to which we shall now turn.

USE, REUSE AND CONTEXTS

At the time John of L’Écluse experienced his miraculous vision, the chapel of St Catherine had been part of Savigny’s claustral complex for over seventy years. By the early 1240s, this complex had reached the form it would essentially retain for the next four-and-a-half centuries. The journey up to this point had not always been an easy one. Within a decade of the work begun by Abbot Jocelin on a new church, Savigny was apparently in some financial difficulty, presumably as a result of costs associated with construction. ⁸² The monks were eventually able to enter their new abbey on 15 August 1200, but it would not be officially dedicated until two decades later on 10 May 1220. ⁸³ Once completed, it was easily the largest church in Normandy, dwarfing neighbouring abbeys and cathedrals alike, with the nine radiating chapels of its east end clearly inspired by those of its mother house at Clairvaux. ⁸⁴ Despite these developments, the chapel of St Catherine remained an integral and focal part of the Savigny site. At the time of the abbey’s dedication, it was not only then the oldest part of the monastic precinct, with its story harking back to what must have been seen, even in the first quarter of the thirteenth century, as Savigny’s golden years, during which many of its holy men had lived, but the chapel was still home to their relics, which had been joined at some point by those of a certain Adeline, a woman with a reputation for sanctity, who had died before 1181/6. ⁸⁵ What is more, its position as an annexe to the chapter house meant that it was located next to one of the most important buildings in the claustral complex. As noted above, this made it one of Savigny’s key intercessory spaces and thus an extremely desirable location in which lay benefactors sought burial, with these still taking place in the chapel and its adjoining close in 1241/2, as we have seen.

The chapel’s site and status must nevertheless have posed something of a dilemma for the Savigny community. In the first instance, its relationship to the chapter house was extremely unusual in the Cistercian world. Indeed, while it was not unknown for adjoining chapels to be added to Cistercian chapter houses, examples of such things are very rare indeed, ⁸⁶ and are limited to abbeys in what is today Germany and to periods later than what was to be found at Savigny. ⁸⁷ In fact, the closest parallel to Savigny’s chapel of St Catherine, which extended on an east–west axis off the north-eastern corner of the abbey’s north–south aligned chapter house, itself measuring at least 20m in length and divided into bays by six central columns, ⁸⁸ comes not from the Cistercian world but the Cluniac one. Adjoining chapels dedicated to the Virgin Mary (known as ‘Lady Chapels’) were a feature of Cluniac chapter houses, with the example at Cluny (fig 5) itself inspiring imitators at houses either within its filiation (eg Charlieu) or its zone of influence (eg Hirsau). ⁸⁹ Admittedly, these were often much larger structures than the chapel of St Catherine, but they nevertheless introduced into the lives of the chapter houses to which they were adjoined the same elements as we find at Savigny.

Fig 5. Plan of Cluny ii (c 1050) showing the Lady Chapel. Image: reproduced from Conant 1968.

The most important of these elements, from our perspective, were those of liturgy and access. At Cluny, therefore, the Lady Chapel played a central role in rituals associated with illness and death, such that its location next to the infirmary, from which the sick and dying were brought, earned it the moniker ‘Notre-Dame de l’Infirmerie’, ⁹⁰ a name also given to the Lady Chapel adjoining the chapter house at the Cluniac priory of Souvigny. ⁹¹ Since chapter house and chapel shared a direct physical connection, this ensured that the former partook of the sacredness of the latter. ⁹² We sadly have no evidence of similar practices in St Catherine’s chapel, but this was nevertheless a space home to sepulchral monuments and their associated rituals. It was also in the vicinity of Savigny’s infirmary and a place to which, as we shall see below, the sick came to receive thaumaturgic healing through Savigny’s saints. The chapel was no doubt accessible from the chapter house by a doorway, as was the case in the Cluniac examples, and it is not impossible that another entrance opened onto the close at its eastern end, just as a doorway communicated with the infirmary-cemetery precinct to the east of the Lady Chapel at Cluny. ⁹³ If this were the case, it would certainly have allowed a pilgrim like John of L’Écluse a much more efficient means by which to access the main church than by otherwise walking all the way around the outside of the abbey to its western door.

But, whatever the similarities, the impact that the one space could have on the other was something that must have been difficult to ignore at Savigny. The issue of who could be interred at the abbey, and where within it, was therefore one in which the monks were ostensibly bound by strict Cistercian statutes, which placed an emphasis on restraint. ⁹⁴ The same regulations sought to impose a clear boundary between the lay and monastic more generally, something the Cistercians took great pains to achieve through their architecture. ⁹⁵ The reality, of course, was often not as rigid as the ideal. Cistercian houses were therefore not only home to lay burials but they sometimes fought to defend their rights in this regard, as Savigny did itself at the beginning of the thirteenth century. ⁹⁶ What is more, while the Cistercians did not encourage public veneration of their relics to the same extent as their Benedictine counterparts, they were not against relics per se, with houses either seeking permission from the general chapter to venerate a local holy figure or writing hagiographical texts in their honour. ⁹⁷ Savigny did both. ⁹⁸

But encouraging such activities and preventing them from infringing on monastic observance was always a delicate balancing act. Given its location, this act must have been particularly hard to achieve in relation to St Catherine’s chapel. The importance of the monastic chapter house, second only to that of the abbey church itself, has been noted above. It was of significance not just to the community in general but specifically to the abbot, since it was from his stall that he there exercised both spiritual and civil authority over his brethren. ⁹⁹ As his own letters and visitation accounts make clear, Stephen of Lexington was committed to the aesthetic austerity espoused by the Cistercians. ¹⁰⁰ He would have therefore been keenly aware that the presence in St Catherine’s of both lay tombs and locally venerated relics risked blurring the boundary between the secular and monastic worlds, just as had happened at Cluny, where Abbot Peter the Venerable (1122–56) deplored how laymen and servants loitering in the cloister had turned it almost into a public street. ¹⁰¹ That this blurring was already taking place is recorded by the Savigny book of miracles, which shows pilgrims spending considerable and intimate time in St Catherine’s, and, by extension, the chapter house. Thus, on the eve of their translation to the main abbey church, a certain James, son of Geoffrey Bacon, was said to have lain there in prayer overnight beneath the stone tomb in which Savigny’s relics had been kept. Earlier that same day, the bishop of Sées had seen celestial fire descending from the chapel’s ‘highest vault’ (a superiori testudine), a phenomenon he witnessed along with the ‘many people who were there praying’ (populo multo ibidem orante), among them a boy recently healed. ¹⁰²

Of course, the translation of the relics to the main abbey church, whose ambulatory and radial chapels offered a space far more conducive to the installation and worship of Savigny’s saints, ¹⁰³ was no doubt designed in part to alleviate this issue, one that had perhaps been further complicated by the addition of two more lay burials in and around St Catherine’s in 1241/2. ¹⁰⁴ Whatever the case may be, the Savigny miracles record that pilgrims continued to visit the chapel after the translation. These included, among others, the young son of a certain Durand Doe, who crawled between the supporting columns (intercolumnas) beneath the saints’ old tomb, and Ivo of Guingamp, who was said to have spent ‘a few days and nights’ (aliquot dies et noctes) in the chapel in vigil and prayers. ¹⁰⁵ What the miracles do not record is how these individuals accessed the chapel itself. Given that the existence of a door in its eastern end is hypothetical, it is possible that pilgrims were only able to enter St Catherine’s via the door off the main cloister to the chapter house itself. If so, this means they would have been regularly penetrating a physical and conceptual space at the heart of monastic observance, one that was supposed to be a haven of tranquillity cut off both physically and spiritually from the outside world. ¹⁰⁶ Even if a door allowed them to enter St Catherine’s chapel from its eastern end, they may have nevertheless proceeded to the abbey church, as suggested above, via the nearby doorway linking chapter house to transept (fig 6).

Fig 6. Proposed reconstruction of (a) the chapter house and (b) the chapel of St Catherine within the medieval Savigny site. Image: author.

However we choose to interpret matters, the very presence of pilgrims in the chapel would have brought them into close contact with the sanctity of the chapter house. Unease with pilgrim access to just such a space has been identified in miracles from the Cistercian house at Melrose, and may have been behind the transfer of a tomb from the chapter house at Newminster. ¹⁰⁷ This was a step Savigny itself appears to have eventually taken, as the miracle of John of L’Écluse suggests, by moving the tomb commissioned by Abbot Simon to the chapel’s close. ¹⁰⁸ Once here, pilgrims would have been able to access it by entering the abbey’s grounds by the eastern gate, without then needing to penetrate the buildings of the claustral complex, even though they might still sometimes walk around the site, interacting with monks as they went, as the case of John of L’Écluse illustrates.

CONCLUSIONS

Having played such a central and defining part in the life of Savigny up to the middle of the thirteenth century, the story of St Catherine’s chapel after this date is one of a gradual slide towards apparent obscurity. The reasons for this are manifold. In the first instance, the translation of Savigny’s relics to the main abbey church, combined with the removal to the chapel’s close of the tomb in which they had previously lain, no doubt achieved what we must presume was intended, and redirected the focus of both the monks and lay visitors to the east end of the abbey church. As the Savigny miracles make clear, this carried its own risks (in one instance, a ‘possessed’ man taken into the abbey’s east end grabbed Hamo’s relics and threw them against the wall), ¹⁰⁹ but pilgrim access to the monastic precinct could be controlled here in a way that must have proved difficult in St Catherine’s. The translation was, of course, also part of a larger effort to have Savigny’s holy figures, including Adeline, whose relics remained in the chapel of St Catherine, canonised. This effort was ultimately unsuccessful, and was no doubt badly undermined by Stephen of Lexington’s departure for Clairvaux in December 1243. Stephen’s successor, Abbot Stephen of Châteaudun (1243/4–55), seems to have had little appetite for pursuing what his predecessor had started, to the extent that the cause of the Savigny saints was taken up by Ralph iii of Fougères (1210–56), who wrote to the pope in 1244 pleading their case. ¹¹⁰ The papal response does not survive, ¹¹¹ and the issue seems never to have been taken up again with any vigour. The result was that the saints, along with the spaces with which they were associated, never attracted wider veneration or renown, such that even Savigny’s own annals, when recording the visit of the pious Louis ix (1226–70) around Easter 1256, note not that he was shown Savigny’s relics or the ancient chapel in which they had once lain, but only that he ‘ate with the convent in the refectory’. ¹¹² Somewhat ironically, it is the entrance to this building, now known as the ‘porte Saint-Louis’ after the abbey’s illustrious visitor, that is the best known remnant of Savigny’s otherwise lost buildings.

Of course, the destruction of the abbey’s buildings was also accompanied by the destruction of its documents. Thus, while we are fortunate to be able still to consult Savigny’s extensive collection of charters, conserved today largely at the Archives nationales de France, little remains of its library and nothing at all of the more modern records relating to the abbey held at Saint-Lô, which were uncatalogued at the time of their destruction on 6 June 1944. ¹¹³ What little we know of St Catherine’s in the late medieval and early-modern periods often comes from chance references to – or the happy pre-war publication of – documents from this collection, the full potential of which will now never be known. Any apparent silence, therefore, should not necessarily be interpreted as proof of a decline in the chapel’s status, especially since some of the evidence discussed above shows that it was a place in which benefactors still wished to be interred and to which, as the case of Jean de Landevic illustrates, they sought to contribute materially.

As to what this article has contributed to the history St Catherine’s, it is freely admitted that, in the absence of the sort of prolonged archaeological work that has been carried out on the monastic precincts of many of Savigny’s contemporaries, the proposal above as to the chapel’s location is only the most plausible interpretation of the evidence as it currently stands. As the work of Peter Fergusson on Rievaulx and Canterbury has shown, arrangements to the east of the east range could often be messy and ad hoc in nature, leaving archaeological traces that are difficult to untangle. ¹¹⁴ As such, sustained excavations in this part of the Savigny site would, in the first instance, help to reveal more clearly what has been suggested by the 2016 geophysical surveys, as well as perhaps unearth other as yet undetected buildings in the same area, such as the sixteenth-century chapel of Holy Sepulchre, located ‘next to the monks’ cemetery’ (prope cimiterium cœnobii), the existence of which has all but escaped academic notice. ¹¹⁵ Such work would also presumably allow for any unearthed structure thought to be St Catherine’s to be dated, something that could potentially shed light on Savigny’s early years beyond its architectural history. As we have seen, although the chapel’s origins are associated with the work on the new abbey church begun by Abbot Jocelin in 1173/4, there is evidence to suggest it existed in some substantial form before this date. The earlier that any structure to the east of the chapter house can be dated, the closer this would take us to the point before Savigny was subsumed by Cîteaux. The abbey’s constitutional framework before this event is unknown, but it has been suggested that the Savigniac constitution was, if anything, based upon the model of Cluny. ¹¹⁶ As such, if excavations were to confirm St Catherine’s proposed location, then what seems only to be coincidentally echoing Cluniac practices could instead be interpreted as having been deliberately built to mimic them. After all, the Savigny community in the early 1170s was no doubt home to those who remembered the abbey before 1147, while we have noted how it retained for itself certain peculiar practices after this date. Perhaps its architecture was among them. Whatever the case may be, if much about St Catherine’s chapel and the area within the monastic precinct in which it was located remains to be discovered, the above has hopefully shown that it was a structure at once important and unusual within the Cistercian world, and, if nothing else, that it, like the wider complex of which it formed a part, is worthy of further study, both historical and archaeological.