Guillaume-Gabriel Nivers and the Quest for Consistency in Counter-Reformation Chant

The Premonstratensians—also known as the Norbertines—are members of an order of canons regular that flourished in premodern Europe from the time of their founding in 1120.¹³ The order was created by St. Norbert of Xanten (ca. 1075–1134), a Benedictine charged by Pope Calixtus II to form a new community of priests within the Diocese of Laon. The site chosen for their mother abbey lay twelve miles from Laon, in the village of Prémontré from which the Latinized name of the order derives.¹⁴ The order spread beyond France and across the Christian West before diminishing in size and influence after the Reformation.¹⁵ Chant revision is of particular relevance to the Premonstratensian liturgy since its rite underwent a significant series of changes in the early modern period, a process implemented initially in France, and which then spread to the order’s monasteries across the rest of Europe. Successive modifications to the sung liturgy generated inconsistency, which in turn prompted Nivers’s more comprehensive reform of 1680. While analyzing sources from the whole of Europe would be beyond the scope of this article, the inconsistencies among Low Countries breviaries and antiphoners paint a clear enough picture to understand the Premonstratensians’ need for new chant books in the seventeenth century.

The sources discussed may appear to embody a stable chant repertory, delineated permanently on the page in solid black ink. On closer study, however, they reveal a living, evolving practice of sung liturgy, and more recent scholarship has begun to explore this diversity in detail. In her analysis of the Premonstratensian hymnary GB-Lbl Add. 15426, Sarah A. Long has observed the use of contrafaction between hymns for St. Augustine and St. Nicholas.¹⁶ Long’s analysis suggests that hymn texts with broad currency were tailored to specific Premonstratensian houses by incorporating them into a network of localized chant texts and melodies. These traits in GB-Lbl Add. 15426, compiled just prior to the Reformation in 1522, point to a practice of devotion tailored to local saints, which later Counter-Reformation revisionists attempted to correct. Alterations also arise in chant texts and melodies in the Office of St. Augustine, which Pieter Mannaerts has observed in B-Gu BKT.006.¹⁷ Emendations made to this manuscript focus on improving clarity in performance and include extensive erasures of both music and text. Accompanying these erasures are textual and melodic insertions in new hands and marginal notes that point to adjusted performance circumstances.¹⁸ These inconsistencies seem to have been relevant across the order’s history, as Cécile Davy-Rigaux argues from the context of Nivers, and Martin J. M. Hoondert from later notated graduals after the early modern era.¹⁹ Together, these studies point to a musical repertory, preserved in notated antiphoners, that was subject to change in the midst of Counter-Reformation ideals. The present article takes these debates a step further, establishing the differences between older sources and later revisions, and their connections to the 1680 reworking of the Premonstratensian antiphoner.

Much of the contextual background to these reforms has been addressed in Davy-Rigaux’s rich study of Niversian reform.²⁰ The need to revise the Premonstratensian rite had been expressed as early as 1660, when the Chapter-General expressed a wish to “remove the useless protractions of chant, correct their accents, and eliminate all forms of dissonances.”²¹ This frustration with the order’s liturgy followed a series of reforms, most of which were taken up partially. The first of these were printed editions of the breviary, missal, and processional in 1574 and 1578.²² Later editions emerged in 1608 and 1618, which attempted to adhere Premonstratensian chant more strictly to the reforms of the revised Roman Breviary.²³ There followed Pierre Gosset’s new publications of the breviary (1621), missal (1622), and Liber ordinarius (1628), followed by further reissues with corrections throughout the seventeenth century.²⁴ Gosset’s reworking found a balance between the older liturgy and the revised Roman rite.²⁵ Yet as not every newly reworked version of the liturgy was available to all the Premonstratensian houses, changes were not implemented nor were older sources corrected in a consistent manner. Such a multitude of newer printed breviaries seems not to have offered viable solutions for the order, and they may in fact have created further problems. Older antiphoners, which included the portions of the Divine Office that were sung, contained a liturgy that by the seventeenth century was increasingly out of date. Changes in the text and its precise placement—however small and cosmetic—led to confusion in the performance of antiphons, responsories, and verses, which formed a major feature of the canonical hours celebrated throughout the day. If older antiphoners were being sung from literally, those who celebrated the Divine Office—many of whom would have been reading from newly updated breviaries—would have been aware that their own sources were inconsistent with the chant that they heard around them. Hence, it was necessary to publish a more definitive version of the gradual and antiphoner, which could be distributed more widely across the order.²⁶

The Premonstratensians established a committee of senior canons to oversee the revision of their chants, and to “correct, abridge, adapt, and put chant into an appropriate form from the books of the Abbaye de Prémontré.”²⁷ Since he had both experience and connections, Nivers was a strong candidate to undertake this work. He lived close to the Latin Quarter on the Left Bank in Paris, near a Premonstratensian house just off the rue de Sèvres.²⁸ He was in contact with the committee from at least 1677, when abbot-general Michel Colbert’s mandatum described the order’s collaboration with Nivers as a chance to improve the quality and speed of the revision:

Davy-Rigaux has shown how the Premonstratensian authorities deemed Nivers to be a highly capable candidate, ideally suited for the ambitious task of updating the order’s chant books. As well as highlighting his experience with monastic orders, Colbert also drew attention to Nivers’s connections with the Faculté de Paris.³⁰ Davy-Rigaux concludes that Nivers must have visited Prémontré to consult the mother abbey’s original chant sources; while this theory is likely correct, there is no definitive evidence that he ever made this journey.³¹ It is known, however, that Nivers spent some years working on Premonstratensian sources. The four canons responsible for the revision committee explicitly state that Nivers worked on the books between September 14, 1676, and February 5, 1679.³²

Although he was not the only influence in the revision of Premonstratensian chant, it is useful to analyze the surviving sources from Nivers’s perspective. Not only was Nivers a leading reformer across lay and religious communities, but his surviving publications reveal much about his editorial policy. By the publication of his Dissertation, Nivers already had decades of experience in Catholic sung liturgy: he had been organist of Saint-Sulpice from the 1650s and was recently appointed as one of the four organistes du Roi in 1678.³³ As a musical liturgist, he was also responsible for the revision of chant in both parish and monastic contexts, where he expressed a wish to save Gregorian repertory from modern abuses.³⁴ As will be made clear below, Nivers can be understood as a reformer who aimed to recalibrate chant’s inner logic—including its melodic and syllabic structure—with the ultimate aim of better enunciating liturgical text for its listeners. Yet his image is not only that of a musical reformer but also of an editor who embraced contemporary social and cultural aesthetics. In the preface to the Dissertation, Nivers admires the “fathers” of the art—Ambrose, Augustine, Gregory—under whom Gregorian chant reached its apogee of sophistication and quality.

Nivers’s slant—one that invokes the glories of an older repertory—was by no means unique. Similar arguments emerge from the writings of other contemporary reformists, who deployed chant’s ancient origins as a tool to promote its musical value over measured music.³⁷ Given the number of changes to the Premonstratensian breviary over the course of the seventeenth century alone, it is tempting simply to read Nivers’s statements as complaints against more recent reforms; however, a further aspect of the contemporary situation also informed Nivers’s approach. It was owing (he claimed) to foreign interventions to the liturgy from the Middle Ages onward—including those enacted at Rome—that the art of singing chant as perfected by Gregory fell into disrepute in French eyes:

These views—which surely betray Nivers’s frustration with the most recent publication of the Breviarium Romanum—may be better understood in light of the political context in France. Throughout his reign, Louis XIV’s religious agenda fostered Gallican interests and rejected the prerogative of the papacy, which culminated in the adoption of the four Gallic Liberties and the confirmation of the droit de régale over France in 1682.³⁹ Louis’s relations with the papacy had been tested as early as 1673, when the king sought to extend absolute power over several provinces of southern France in opposition to Rome.⁴⁰ Nivers’s condemnation of Roman reform is consistent with sentiments observed by theorists of the day, and mirrors contemporary French attitudes toward Rome and the papacy.⁴¹ Having its mother house in France provided a significant background to Premonstratensian reform, since it placed the order under the influence of an increasingly strident Gallic agenda, one that promoted the French monarch’s role in governing the church within France, while diminishing papal influence.⁴² These multiple threads—musical, religious, political—thereby informed the revisions of Premonstratensian chant in the years leading up to Nivers’s influential reworking. These dimensions in turn affected the makeup of Premonstratensian chant to the north of the French border, as will be seen from surviving sources.⁴³

The fruits of Nivers’s collaborations with the order were concluded in two volumes: a new edition of the antiphoner (Antiphonarium Praemonstratense), and the collection of Mass chants for the gradual (Graduale Praemonstratense), both of which were published in 1680.⁴⁴ The Premonstratensian liturgy continued to evolve, however, with the most recent revisions occurring as late as the twentieth century. Given these subsequent adaptations of the Premonstratensian rite, Nivers and his contemporaries were for a long time forgotten in chant scholarship. More recently, however, their contributions have been recognized by both Premonstratensians and chant scholars alike.⁴⁵ Jean-Placide Lefèvre, in his wide-ranging survey of the Premonstratensian liturgy, argues that Nivers’s interventions exacerbated preexisting issues without restoring appropriate levels of solemnity to the chants themselves.⁴⁶ Davy-Rigaux has offered a more nuanced stance, observing that Nivers is representative of the editorial and political contexts of his time, and that his interventions in the Premonstratensian order must be considered alongside the wider revisionist policy in seventeenth-century France. Echoing Davy-Rigaux, Theodore Karp has shown that Nivers’s contributions are part of a wider history of Gallican interests that extended into the eighteenth century.⁴⁷ No scholarship, however, has considered this revision from the Premonstratensian perspective. Although Dietmar von Huebner has thoroughly analyzed earlier Premonstratensian antiphoners, he does not consider their role in the Divine Office alongside unnotated office books such as breviaries, which contain musical texts.⁴⁸ This article therefore examines Nivers’s contributions to the Antiphonarium Praemonstratense, while evaluating the state of the Divine Office and its need for revision. A further issue probed here is the extent to which Nivers’s editorial policy, honed outside Premonstratensian contexts, aligned with the needs and ambitions of the Premonstratensians. Through analyzing Premonstratensian sources from the Low Countries that underwent adaptation, it is possible to determine how Nivers’s revisions were received and understood within the confines of one specific geographical area, and whether they can be deemed successful.

This article considers three case studies of chant revisions witnessed in Premonstratensian breviaries and antiphoners. While they are by no means representative of the entire chant repertory, the examples have been chosen owing to their frequent appearance in older manuscripts of the order, their current availability in archives and digitized scans, and the fact that their offices are both unique and elaborate.

The sources relevant to this study can be divided according to their type, their date, and the location of their manufacture or use. Appendix 1 provides a list of all sources consulted.⁴⁹ This investigation limits itself to sources pertaining to the celebration of the Divine Office that were owned by (and most likely also used at) Premonstratensian houses around the Low Countries: namely, antiphoners that include the sung texts of the whole Divine Office, and breviaries that contain the texts recited or sung by the celebrant. Since the Premonstratensian revisions took place alongside Counter-Reformation reforms that encompassed other communities, however, appendix 1 also includes significant non-Premonstratensian sources for comparison: namely, publications of the Antiphonarium and Breviarium Romanum.⁵⁰ The study does not include missals or graduals used for the celebration of the Mass. It also omits sources that focus on specific services or functions in the Divine Office—such as the vesperale for vespers chants, or processionals. Books of hours, intended mostly for private devotion, are also beyond the scope of this study, although their features are of interest and are discussed extensively in scholarship elsewhere.⁵¹

Of the sources used by Premonstratensians in their liturgy, the earliest are manuscripts. Those consulted for this study—primarily antiphoners—can also be categorized by the nature of their contents. Since antiphoners often have larger physical dimensions to ensure that they are legible to the multiple singers often grouped around them, most contain just half of the liturgical year’s chants. Winter volumes contain the repertory from the beginning of Advent up until Easter, whereas summer volumes encompass chants from Easter until Advent. The manuscript sources for this study include several winter antiphoners, such as the three manuscripts made for Grimbergen Abbey situated just to the north of Brussels: B-Br 217, 5642, and 5643.⁵² The revisions in these sources coincide with the newer editorial standard of Nivers’s Antiphonarium Praemonstratense, indicating that someone updated them after 1680. Three further sources of interest include antiphoners that partially or entirely escaped later attempts at chant reform. The first of these is B-AVna IV.412, made for the abbey at Averbode, which shows no trace of erasures or correction.⁵³ The remaining two are the more complex: NL-OHnp 77 and 93 from the priory of Sint-Catharinadal, both of which bear witness to partial reform, although conducted in an inconsistent manner. The summer sources include the aforementioned B-Gu BKT.006, also known as the Tsgrooten Antiphoner.⁵⁴ This elaborately decorated manuscript sits alongside the winter antiphoner GB-Lbl Add. 15427—not examined in the study—both made under Antonius Tsgrooten’s tenure (1504–30) as Abbot of Tongerlo Abbey.⁵⁵ Three other summer sources were revised to reflect later Niversian interventions: Grimbergen Abbey’s B-Br 210 and Sint-Catharinadal’s NL-OHnp 76 and 76a. One further source from Averbode, B-AVna IV.413, escaped revision like B-AVna IV.412, but is not believed to have been compiled as a companion to this similar source.⁵⁶

Manuscript antiphoners are not the only sources relevant here. Although most breviaries do not contain musical notation, they are of interest to any study that involves texted liturgical chant. While antiphoners are intended for the use of those who sang the antiphons, responsories, and other chants, breviaries provide the texts, rubrics, and other indications for the celebrant.⁵⁷ Given that breviaries need only be used by one individual and are generally unnotated, they are smaller in size and often include the entire liturgical year in a single volume; hence, they require fewer resources to manufacture and are cheaper to produce.⁵⁸ Given their size and portability, breviaries were particularly well suited to newer printing technologies deployed during the Protestant Reformation.⁵⁹ The large number of printed breviaries that survive from the sixteenth and seventeenth centuries attest to this newfound enthusiasm. Based upon surviving records, it is likely that breviaries would have far outnumbered the antiphoners available within any institution’s holdings.⁶⁰ But while the availability of newly printed breviaries can be explained by their ease of production, there was also a practical need for them as a tool to introduce the latest religious reforms at minimal cost. Read in tandem, antiphoners and breviaries point toward a complex revision history, in which the chant in older manuscript antiphoners was updated to varying degrees of consistency to accord with the newest printed breviaries. And although these breviaries indicate textual revisions, they do not normally account for variations in a chant’s melody, which are instead dependent upon local and institutional practices.

The three case studies offer a comparative analysis of Premonstratensian chant sources, dated between 1400 and 1680, from the Low Countries. They document the complexity of the prior art that Nivers and his contemporaries encountered. These studies also examine the imperfect way in which older sources were adapted to adhere to newer editorial standards. A key question here is the precise date from which the Premonstratensians began to revise their chant texts, and this article provides a putative terminus ante quem based upon the earliest surviving revised breviaries. Following this point of debate is a secondary issue: namely, how these changes were handled in light of Nivers’s 1680 revision. Further questions specific to the Low Countries concern how earlier sources were adapted to reflect Nivers’s reforms and whether they were done so in a consistent manner.

The case studies—the first concerning solely textual revision, the second and third featuring corrections to both music and text—come from three different offices, two from the summer half of the antiphoner and one from the winter. Appendices 2–4 list the complete contents for each office; the specific chants that form the focus here are as follows:

1. Chants for Ascension [summer]

   “Audistis quia [ego] dixi vobis vado” (first vespers, Magnificat antiphon)

2. Chants for Pentecost [summer]

   “Dum complerentur dies” (matins, first nocturn, first responsory)

   “Et apparuerunt illis” (matins, first nocturn, first verse)

3. Chants for St. Andrew’s Day [winter]

   “Ambulans Jesus iuxta mare” (first vespers, Magnificat antiphon)

   “Adoremus victoriosissimum regem Christum qui” (matins, invitatory)

   “Cumque carnifices ducerent” (matins, first nocturn, third verse)

“Audistis quia [ego] dixi vobis vado” is set for the feast of the Ascension (see appendix 2), which falls in the summer half of the antiphoner, forty days after Easter. Since a liturgical day begins on the preceding evening, the first office is usually vespers; however, important feasts often have a second vespers service at the end of the astronomical day.⁶¹ This chant, specified for first vespers, is the antiphon for the Magnificat, a canticle taken from the Gospel of Luke from the words of the Virgin Mary at the Visitation. The text is taken from John 14:28, in which Jesus tells his disciples that he will soon depart from them, prefiguring his death. This antiphon therefore links Christ’s death and resurrection with both his departure at the Ascension and his reunion with mankind at the Second Coming. The text is presented below in its original version (left) as found in sources that predate sixteenth-century textual revisions (for example, Kerver, 1515, 105r). Subtle changes appear in the revised version (center) attested in later breviaries and antiphoners. The approximate translation (right) applies to both versions of the text. Phrases that include points of revision are distinguished by bold type.

This example serves first to make a general point: that textual revisions to the Divine Office were being made by the Premonstratensian order far earlier than Nivers’s 1680 reworking of the antiphoner. This is apparent from the textual disparities between older versions of the chant in manuscripts, and the revised versions both in these sources and in printed breviaries. Observe, for instance, the consistent use of the earlier antiphon text in older manuscripts and breviaries, and the subtle insertion of “ego” and transposition of “ad patrem vado,” which only emerge in later sources. The 1513 publication of the Breviarium Praemonstratense, published in Paris by Petit, Marnef, and Regnault (fig. 1, top of second column), contains the earlier version.⁶²

In antiphoners such as B-Gu BKT.006 and B-Br 210, however, corrections are implemented to bring the chant in accord with the later modifications. In B-Br 210, an earlier source dating to approximately the 1480s, the original layer of text was erased (scraped off) and the new readings were inserted by a hand that tried to emulate the original (see fig. 2). This editorial intervention was handled with considerable skill and is in fact only detectable from the occasional trace of textual and notational erasure, as well as the subtly different hue of ink for later insertions.⁶³ The case is different for B-Gu BKT.006. Here, too, the editorial changes were achieved by erasure, but there was no attempt to hide the interventions: the revisions are in an entirely different hand and inserted interlineally by a later scribe with a thinner nib (see fig. 3). The distinctness of these later revisions clarifies the older layer, which is identical to the original reading of B-Br 210.

These textual alterations are by no means unique to the Premonstratensians: in the case of this chant, the revised text accords with the 1592 Clementine Vulgate. But while it may seem tempting to tie such liturgical revisions to the reform of their Biblical source, the 1592 Clementine revision is not consistently implemented in the remainder of the chants surveyed in this article. While changes such as these may seem minor and without consequence for either textual meaning or performance intelligibility, questions still linger concerning the chant’s consistency across Premonstratensian communities, and the circumstances of its performance. The later Tongerlo antiphoner, compiled almost a decade after Petit, Marnef, and Regnault’s issue of the breviary, retains the older wording. This preference for the older text suggests that alterations took some time to percolate beyond Paris and Prémontré. Evidently, even amidst sixteenth-century revisions that were disseminated by a burgeoning print industry, manuscript antiphoners continued to be produced at great expense with outdated chant texts that did not accord with the newest edits issued at Prémontré.⁶⁴ How, then, was this text sung once the later versions became more accepted across Premonstratensian communities? It is tempting to suppose that Nivers’s revisions indicate how the revised texts were sung; however, there is little evidence to suggest that his melodic reworking of Premonstratensian chant had any clear precedent.⁶⁵ It is possible that chants like “Audistis quia [ego] dixi vobis vado,” given the lack of sources before 1680 that present their texts with revised melodies, led to the frustration felt by Premonstratensians, which eventually led them to seek Nivers’s assistance.

The second set of chants comes from Acts 2:1–3 and appears in the office for Pentecost (see appendix 3). One of the most important feasts in the Christian calendar besides Easter itself, Pentecost held a prominent position in chant manuscripts, as can be seen from B-Gu BKT.006’s lavish miniature display (see fig. 4).⁶⁶

Two Pentecost chants, the early morning Matins responsory “Dum complerentur dies” and its associated verse “Et apparuerunt illis,” received significant editorial changes, in both text and music, at the hands of Nivers. The earlier (left) and revised (center) texts are presented here along with an approximate English translation (right), with points of revision highlighted in bold.

Premonstratensians implemented these changes in their manuscripts to bring the repertory into agreement with Nivers’s new edition and with post-Tridentine revisions to the Roman Breviary.⁶⁷ Specific points of textual divergence occur in two places in the responsory: Nivers supplies “in eodem loco” instead of “dicentes” in the first part, and “vehementis et replevit” instead of “torrens replevit” in the second part, both corresponding to revisions made in earlier Roman and Premonstratensian chant books printed from both the sixteenth and seventeenth centuries.⁶⁸ These textual changes, whereby scribes are forced to erase and then write over the original text, are the most obvious and easily recognizable indicators of manuscript revision. They are shown in figures 4 (above) and 5, which present points of erasure and insertion in B-Gu BKT.006 and B-Br 210.

B-AVna IV.413 (fig. 6), however, presents the older version of the chant. It is apparent that this source was never adapted to reflect Nivers’s revision. It can therefore be useful as a benchmark to assess the other sources. The manuscript is also of interest since it suggests that, despite his ambitions, Nivers’s revisions in the 1680 Antiphonarium Praemonstratense apparently did not reach all communities. That some sources were valuable enough to be preserved, yet never underwent revision, points to a less than cohesive process of chant revision.⁶⁹

Unlike the Averbode manuscript, two sources that survive from Sint-Catharinadal bear clear, and unusual, signs of revision: the staves have been painted over to erase earlier versions of the chant. Among the various tools available for the corrector looking to erase—scraping, rubricating, strikethroughs, or adding dots of deletion—this method is highly uncommon. Such an approach, not attested in any other Premonstratensian sources from the region, appears consistently in Sint-Catharinadal sources.⁷⁰ Yet there is nuance to this practice within Sint-Catharinadal: while both NL-OHnp 76 (fig. 7) and 76a (fig. 8) have been revised throughout the Pentecost office, the editors of NL-OHnp 76 neither added the new notation on the painted-over staves nor revised the text. Such a policy suggests that the priority was to eliminate outdated musical material. It diverges from NL-OHnp 76a’s policy of inserting the new repertory, both textual and musical. This priority is also distinct from the scribe of NL-OHnp 93 (see fig. 9, below)—a source discussed in the final case study—who revised older chants that were still relevant, while retaining unchanged (and unredacted) the older chants that had fallen out of use.⁷¹ This divergence points to two practices regarding musical objects: one in which sources hold current repertory alongside older material; and another in which current repertory must be the only material present in a source, with anything else that contradicts it removed.

“Dum complerentur dies” was revised both textually and musically in NL-OHnp 76a, B-Br 210, and B-Gu BKT.006. That is evidenced through the full erasure and new notation in the Sint-Catharinadal source, and the removal and overwriting of specific notes in the sources from Tongerlo and Grimbergen. While the scribes attempted, with varying degrees of success, to mimic the original textual hands, there are observable differences in both penmanship and the hue of the ink. And, as in the first case study, the editor of B-Br 210 was surprisingly effective in covering their tracks. These editors show aesthetic concern for and technical skill in emulating each manuscript’s original style, which is balanced alongside the practical need to respect textual and melodic revisions to the repertory.⁷²

While such textual and musical revisions confirm the later interventions, inconsistencies can also be observed in earlier versions of the chant. The opening of the respond in B-AVna IV.413 (fig. 6, above) has not been corrected to match Nivers’s 1680 revisions, yet the older chant is not entirely stable. The opening phrase “Dum complerentur” presents a clear case of neume erasure, with an original podatus figure on a and b erased, and replaced by a new podatus one tone lower in what is unmistakably a different hand with less precise penmanship. The difference is subtle, but nevertheless suggests that while there was relative uniformity in the “Dum complerentur dies” respond, occasional instances of variation existed across the sources and would have been observed in performance.⁷³

Such textual and notational minutiae paint a picture of inconsistency, one which must have frustrated Premonstratensians enough to call for Nivers’s wide-ranging reforms. The question remains, however, as to whether the 1680 reforms resulted in a more definitive version of the sung liturgy. While Nivers’s efforts did result in simplification and cohesion in general, closer scrutiny of these manuscripts suggests that consistency was not automatically guaranteed.⁷⁴ This fact is apparent from the inconsistent revisions to the notation of the verse. See, for instance, the syllabic displacement in example 1, which presents the opening phrase in four of the surviving sources. The minor, apparently superficial difference between Nivers, Tongerlo, and Sint-Catharinadal’s use of a punctum and virga on “apparuerunt” and a simple punctum in the Grimbergen and Averbode sources is followed by more obvious divergences at “singulos eorum.” Both B-AVna IV.413 and B-Br 210 agree consistently in preserving the most florid version, whereas NL-OHnp 76 and 76a and B-Gu BKT.006 replicate the Niversian interventions, e.g., beginning “singulos” on d rather than c and reducing the subsequent melisma on “singulos” to a single punctum. The Sint-Catharinadal sources likewise follow Nivers’s reduction of the melisma on “eorum” (by eschewing the final a–G–a figure), but B-Gu BKT.006 does not coincide here and instead retains the more florid melisma of B-AVna IV.413 and B-Br 210. The Sint-Catharinadal sources seem to have been edited with a more careful appreciation of the revisions made by Nivers, whereas the Grimbergen and Averbode manuscripts retain many of the notational features of the older manuscripts. It is the Tongerlo source B-Gu BKT.006, however, that demonstrates the greatest degree of flexibility in sometimes maintaining, sometimes discarding, older melodic features in response to the Niversian revision. In sum, while some of these discrepancies are minor, it is apparent that the scribes’ engagement with the revisions expected of them resulted in the very same sort of inconsistency that had compelled the Premonstratensians to consult Nivers in the first place.⁷⁵

The third case study is a selection of chants from various points of the office for the feast of St. Andrew the Apostle on November 30 (see appendix 4). This feast often coincides with the first Sunday of Advent, which represents the start of the liturgical year.⁷⁶ St. Andrew’s Day is therefore the first major feast in the sanctorale, the cycle of liturgical feasts—largely comprising days that commemorate saints—which fall on fixed dates, as opposed to the moveable feasts of the temporale.⁷⁷ The Magnificat antiphon “Ambulans Jesus iuxta mare” appears at the end of first vespers; “Adoremus victoriosissimum” is the invitatory that follows at the subsequent Matins service.

All chants in B-AVna IV.412 have been left unedited, consistent with other sources from Averbode. Of the six sources analyzed, five antiphoners—B-Br 217, 5642, and 5643, and NL-OHnp 77 and 93—have been fully edited in response to Nivers’s 1680 antiphoner. These edits only apply to “Ambulans Jesus iuxta mare,” since the invitatory “Adoremus victoriosissimum” was not included in the 1680 issue of the Antiphonarium Praemonstratense.⁷⁸ It is notable, however, that these sources keep the earlier invitatory, retained in its original state. These Premonstratensian manuscripts—like those of other orders—preserved traces of older repertory that sat alongside new material that was still sung. This coexistence of old and new repertory is particularly apparent in NL-OHnp 93, mentioned in the previous case study, in which the painted-over staves of “Ambulans Jesus iuxta mare” contrast with the unpainted, original notation of “Adoremus victoriosissimum” and subsequent chants (see fig. 9). Presenting new chant alongside obsolete material appears not to have been an issue to the scribes and editors of this manuscript.⁷⁹ The community of canonesses presumably would have been familiar enough with the revised liturgy to know which chants had fallen out of use and should thereby be skipped. Such revisions nevertheless point toward a diversity of editorial policies that existed at this priory, as the strategy witnessed in NL-OHnp 93 contrasts with the editor’s more thorough approach toward erasure in NL-OHnp 76 (fig. 7, above) for “Dum complerentur dies,” and the similarly consistent correction in NL-OHnp 76a (fig. 8, above).

The two Sint-Catharinadal sources that preserve “Ambulans Jesus iuxta mare” offer a useful indication of when the following invitatory was removed from the Premonstratensian liturgy. NL-OHnp 93, the earlier source, contains the earlier chant, like the Grimbergen and Averbode manuscripts. It dates to around 1618, based upon the records of Sint-Catharinadal and the Brabants Historisch Informatie Centrum.⁸⁰ NL-OHnp 77, which was compiled in 1676, four years prior to Nivers’s published revision, lacks the invitatory and proceeds straight to chants for Lauds (fig. 10). There is therefore a period of just under sixty years during which this transition occurred, leading to the omission of “Adoremus victoriosissimum” from newly compiled antiphoners.⁸¹ Given the proximity of NL-OHnp 77’s compilation to the publication of Nivers’s Antiphonarium Praemonstratense, it is possible that the source was assembled and inscribed based upon prior knowledge of the revised 1680 antiphoner’s future contents, although the “Adoremus victoriosissimum” invitatory is attested in breviaries without notation as late as 1675.⁸²

The original version of “Ambulans Jesus iuxta mare”—preserved entirely in only B-AVna IV.412 (fig. 11)—differs substantially from the Niversian revision in both its text and music. The Sint-Catharinadal and Grimbergen sources have been edited to adhere completely to the new Niversian standard. Upon closer scrutiny, however, it is apparent that each of these sources contained, prior to their revision, melodies that show remarkable conformity to B-AVna IV.412.

This agreement suggests that prior to Nivers’s revision, there was in fact a general consistency, at least on a local level, in the St. Andrew office. Such cohesion is immediately visible from the Grimbergen sources, whose original, melismatic melody is still legible through the erasures (B-Br 5642 is given as an example, in fig. 12). It is likely that such melismatic writing also existed in the Sint-Catharinadal sources. Even though the overpainting has obscured any notational traces, the spacing of the original text reveals the melismatic nature of their original melodies (see, for instance, the space left for the six-note opening figure in NL-OHnp 93 in fig. 9, above). It should be noted, however, that while the new Niversian standard was widely adopted across corrected antiphoners, here too this process was not accomplished in a consistent manner. It is striking that after such a comprehensive endeavor to update the order’s older chant manuscripts there should remain differences between the revised sources, especially for a feast as important as St. Andrew’s. Example 2 compares the earlier version of the chant against Nivers’s revision and divergences in the sources. Of note is the figure on “vidit,” with three distinct interpretations. The earlier version in B-AVna IV.412 has a punctum (F) followed by a clivis (F–D) on the second syllable. Nivers’s revision simplifies this figure to two puncta (D and F) for each respective syllable. This version appears in all revised sources except B-Br 5642, which presents a pes (D–F) on the first syllable and a punctum (F) on the second. Such a discrepancy is unexpected in this source, which was constructed and presumably revised alongside its two companions from Grimbergen. The manuscript appears to have been constructed along very similar lines to B-Br 5643: the contents of their chants and even their layouts on successive pages are almost entirely identical throughout both manuscripts.⁸³ Since these are the smallest of discrepancies, such divergence may point to an editorial oversight. It suggests, nevertheless, that the procedure of updating the order’s chants did not produce uniform consistency.

Indeed, similar inconsistencies emerge throughout the rest of the office. Observe the chant “Cumque carnifices ducerent,” the third verse of the first nocturn of Matins. The verse is not included in Nivers’s 1680 publication. Most Low Countries sources consequently lack any editorial intervention for this chant, proving that these revisions were being taken from Nivers’s standard. The two Sint-Catharinadal sources, however, include a revised melody that is independent of the other Premonstratensian sources. While the earlier melody in B-Br 217, 5642, and 5643 (see B-Br 5643 in fig. 13) accords with NL-OHnp 77 and 93 (see NL-OHnp 77 in fig. 10, above) in its general outline, there are numerous moments in which underlay is displaced, and significant inconsistency occurs at cadential figures. A melismatic motif on “et dicentium” appears in B-AVna IV.412 (fig. 14) and the three Grimbergen sources; however, it has been erased from the Sint-Catharinadal revisions. That the Sint-Catharinadal sources have been edited to comply in part with the Niversian revisions, while also retaining their own distinct, revised melodic identity in this verse (which was omitted by Nivers), suggests that the revision process was more complex than a simple updating after the publication of the 1680 antiphoner. The revisions in the Sint-Catharinadal sources suggest that alternative practices survived, which remained in use at least up to the period of reform. Updating these parts of the liturgy may also have been unnecessary if such revised versions were available elsewhere.⁸⁴

How is it that a feast as important as St. Andrew’s was left in such a state of inconsistency? Andrew was recognized as a key apostle and martyr in early Christendom. His feast was of semiduplex rank in many institutions in the Low Countries, such as at the collegiate church (now cathedral) of St. Michael and St. Gudula in Brussels.⁸⁵ His feast day held yet greater distinction across Premonstratensian communities, recorded as duplex rank in all printed breviaries and in the calendar of B-Gu BKT.006’s sibling volume, the hymnary GB-Lbl Add. 15426.⁸⁶ That such a high-ranking sanctorale feast should show inconsistencies in both pre- and post-Niversian states suggests that while there was a clear need for Nivers’s 1680 revision, his Antiphonarium Praemonstratense did not achieve the consistency that was so desired.

During the early modern period, the sung liturgy of the Christian West underwent a period of significant reform. Advocates of such changes, inspired by the Counter-Reformation, aimed at correcting the textual and melodic inconsistencies that hindered the clarity of the Mass and Divine Office as it was sung in churches, monasteries, and convents throughout Christendom. Such concerns were germane to the Premonstratensians in 1677 when they commissioned Nivers to create a new version of the Antiphonarium Praemonstratense. The chants surveyed in this article help provide a richer context behind this commission, since they explain why such a definitive reworking of the order’s chants was necessary.

Although the repertory maintained a reasonable degree of cohesion in the fifteenth to eighteenth centuries, the three office chants considered above show how problematic moments of divergence could be, even within the relatively small geographical area of the Low Countries. Most of these inconsistencies are textual and are mirrored in sources from other monastic and lay communities that were revised in the wake of the Council of Trent.⁸⁷ While the divergences in the texts are often minor—making very little substantive impact in the exegetical message of the chant—their frequency compounded their cumulative effect. And such textual inconsistencies would have been a source of confusion when combined with further melodic disparities. These inconsistencies would have been noticed particularly by those who read the updated liturgy from newer, printed breviaries while listening to the chant as it was sung from older antiphoners. In a performance context, these textual and melodic differences could have led to the frustrations expressed by the Abbot-General in the order’s acts of 1660. Nivers’s publication of the 1680 Antiphonarium Praemonstratense can be understood, then, as an attempt on the part of Premonstratensians to revise such irregularities. Printing technology, now in its heyday, resulted in both faster and wider dissemination throughout houses across Europe, leading to what the Premonstratensian authorities hoped would become a much more cohesive chant repertory through the order’s first printed edition of the antiphoner.

It would not, however, be entirely true to say that Nivers’s contributions brought about greater consistency in the order. Some of his revisions, albeit motivated by worthy ideals, were not readily understood or uniformly implemented by scribes tasked with the updating of older manuscripts. That can be explained by the diversity of revisions in some of the chants, particularly in the “Et apparuerunt illis” verse for Pentecost, and the “Cumque carnifices ducerent” antiphon for the feast of St. Andrew. The diversity of these revisions suggests that the manuscript updates may have been implemented in a rushed manner, perhaps during a temporary period of access to the new issue of the 1680 antiphoner. Regardless of the circumstances of these revisions, it is apparent that the attempt to reverse the inconsistency of the order’s chants was not fully achieved by the 1680 version. It is arguable, instead, that the changes it wrought led to occasional confusion and, in the case of Sint-Catharinadal, unique interventions by individual scribes.

Despite their imperfect states, surviving manuscripts that retained older chants still held considerable value after Nivers’s publication, particularly as not all Premonstratensian houses may have had access to the 1680 print.⁸⁸ These manuscripts’ continued value may explain their retention and reuse. Scribes demonstrated varied priorities in updating older sources, which only increased the degree of diversity in their makeup and contents. Some, such as the scribes of NL-OHnp 76, were ruthless in both revising and excising that which was out of date, whereas others, like the scribes of NL-OHnp 93, showed little compunction at leaving older, obsolete repertories alongside newer, more relevant chants. Other scribes, such as those of the four Grimbergen sources, were more diligent in their updating of chants: their priorities extended to both erasing old text and notation, and inserting newer revisions in as close a hand to the original. This more labor-intensive approach ensured that each manuscript appeared to contain a stable repertory that had never needed revision in the first place.

These diversities point toward the varied functions of textual and musical sources. Some were maintained as a standardized repository that accurately recorded the order’s chant texts and melodies so that any user could accurately sing the Divine Office from a faithful copy. Others may have been more practical, representing an assemblage of old and new repertories that a knowledgeable user—familiar with both the current repertory and the history of the specific source—would have known how to navigate.⁸⁹ They would have recognized which chants could be read literally, which required intelligent reinterpretation, and which needed to be ignored entirely. Chant manuscripts can be seen, then, more as a collection of prompts in the hands of an experienced singer. Even as late as the seventeenth century, they functioned as an aide-mémoire to an ancient but living chant tradition. They testify to an oral practice in which those skilled enough in the art would have only required momentary access to a source—regardless of how accurate it was—to recreate a more precise ideal contained in their memories. For all the attention given, therefore, to the departure of the art of memory in the early modern era, its influence was still felt.⁹⁰ Nivers’s antecedents were intended to be superseded, perhaps forgotten; however, their traces were still very much part of the Premonstratensians’ experience of the liturgy as they negotiated the new 1680 antiphoner.

In appendix 1 shaded rows indicate manuscript sources, unshaded rows printed sources. Appendices 2–4 use the following formatting conventions and abbreviations. Shading indicates the inclusion of the specified chant. An asterisk (*) indicates an unnotated chant. Strikethrough indicates a chant that has been crossed out in the source. Rubrics are italicized. Chants in boldface are discussed in the article. Cantus ID numbers are included in parentheses. Genres are indicated by standard abbreviations: A = antiphon, A[b] = Benedictus antiphon, A[m] = Magnificat antiphon, A[n] = Nunc dimittis antiphon, Ca = canticle, Cap = capitulum, Co = collect, H = hymn, I = invitatory antiphon, Ps = Psalm, R = responsory, V = responsory verse, W = versicle.