Hostname: page-component-76d6cb85b7-dqfph Total loading time: 0 Render date: 2026-07-14T07:15:05.884Z Has data issue: false hasContentIssue false
  • English
  • Français

Travelling Manuscripts and Liturgical Collaboration between England and the Continent in the Eighth Century: the ‘Sacramentary of Boniface’ and the Gelasian Sacramentary of Groß Sankt Martin in Cologne

Published online by Cambridge University Press:  19 February 2026

Arthur Westwell Open the ORCID record for Arthur Westwell [Opens in a new window]
Arthur Westwell*
Affiliation:
Fakultät für Katholische Theologie, Universität Regensburg, Germany
*
Email: arthur.westwell@theologie.uni-regensburg.de
Rights & Permissions [Opens in a new window]

Abstract

Among little-known fragments of sacramentaries, two examples in insular half uncial in German libraries offer us significant evidence of the activities of early English missionaries and scribes. The first has long been known in German as the Sacramentary of Boniface, and its unique and archaic content shows it is among the only surviving evidence of liturgical practice in England itself in the eighth century. The second, in scattered fragments from the monastery of Groß Sankt Martin in Cologne, offers an important witness of the Gelasian of the eighth century (as distinct from the Old Gelasian Sacramentary), a compilation exclusively known on the Continent. Liturgical evidence offers a framework to go beyond the uncertain attempts to date and localize the particularly conservative script of these fragments. Analysis of their content shows how English scribes made a decisive input to the transformations of the continent’s liturgy and the dissemination of new forms of mass book.

Information

Type
Research Article
Information
Early Medieval England and its Neighbours , Volume 52 , 2026 , e2
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - SA
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-ShareAlike licence (http://creativecommons.org/licenses/by-sa/4.0), which permits re-use, distribution, and reproduction in any medium, provided the same Creative Commons licence is used to distribute the re-used or adapted article and the original article is properly cited.
Copyright
© The Author(s), 2026. Published by Cambridge University Press

IntroductionFootnote 1

The remains of Anglo-Saxon liturgical books are far from abundant, but fragmentary witnesses found in libraries on the Continent have met with relatively little attention, even among scholars of early English manuscripts. Even though German scholars associated one such fragment directly with Boniface (c. 675–754) himself, this evidence has rarely been employed to assess the impacts of the Anglo-Saxon presence on the Continent.Footnote 2 In general, liturgical manuscripts have been a minor consideration or absent from view in illuminating the missions. But liturgical texts can be of great help in providing frameworks for palaeographical dating and localization. Equally, as the efforts to establish the families and types of the earliest sacramentaries have stalled since the end of the twentieth century, new discoveries have not been permitted to update established narratives.Footnote 3 In particular, the role of the Anglo-Saxons in contributing to and spreading Carolingian forms of the mass book in the late eighth and early ninth centuries has never been fully acknowledged. Two fragmentary manuscripts, both written in insular half uncial assigned to Northumbrian hands, bear witness to how the English missionaries initially brought books that were recognisably ancient out of England, but then rapidly began to adapt to the new liturgical forms around them, while also making a clear contribution to the compilation of new forms of the mass book on the Continent, which became exceedingly influential.

I: The ‘Bonifatius-Sakramentar’ in Regensburg

The first of the two books in Northumbrian hands exists today in three fragments, comprising six folios, which were all part of the collection of documents of the Cathedral Library of Regensburg, used to cover bills and receipts in the seventeenth century.Footnote 4 They have since been scattered. The three fragments are now:

  1. 1. The ‘Walderdorffer fragment’. This was found in the Cathedral Library and acquired by Hugo Graf von Walderdorff (1828–1918), an aristocratic local historian of Regensburg, and held in his castle at Schloss Hauzenstein outside of the city until the early twentieth century.Footnote 5 The three necrological notices found in it were published in MGH in 1905.Footnote 6 It then vanished from view, but was sold at Sotheby’s a century later in 2007, without any indication of the intervening owner.Footnote 7 It seems it was acquired there by the collector, Ernst Boehlen (1935–2022) of Bern, who gave it the signature MS 702. His collection, including the fragment as Lot 2, was offered again by Sotheby’s in 2024 and then again in 2025.Footnote 8 However, the rather extreme estimate was not met, and the fragment remains unsold at time of writing.

  2. 2. Regensburg, Bischöfliche Zentralbibliothek, Fragment III 1.1. Found by Klaus Gamber in 1974, it is the only piece of the Sacramentary that has remained in Regensburg.Footnote 9 In the older scholarship, it is discussed under the shelfmark Cim 1, which has since changed.

  3. 3. Berlin, Staatsbibliothek zu Berlin, Preußischer Kulturbesitz MS Lat. fol. 877. Siffrin reproduced the remarks of Hermann Degering (1866–1942), then director of the Berlin Staatsbibliothek, who noted that the fragment had been acquired by the Staatsbibliothek in 1920 from the estate of a book binder from Regensburg, who had apparently taken it from the binding of an old book.Footnote 10

On the basis of the Berlin and Walderdorffer fragments, Siffrin first published the suggestion that the lost sacramentary was a ‘Bonifatianum’, suggesting that Boniface himself could have brought the book to Regensburg, and gifted it to Gaubald (d. 761), whom he made Bishop of Regensburg in 739.Footnote 11 This identification was repeated in Gamber’s edition, and Gamber added the Regensburg fragment, undertaking a new edition of all six surviving folios under the title ‘Bonifatius-Sakramentar’.Footnote 12 A not trifling problem with the otherwise attractive idea is that Boniface did not come from Northumbria, though, of course, that does not absolutely exclude the possibility he ever possessed the book.Footnote 13 In the meantime, other suggestions for what Boniface’s sacramentary might have looked like have been made by Hohler and Hen, but a lack of manuscript evidence means the discussion remains rather speculative.Footnote 14 In any case, Siffrin and Gamber seem to have worked only from relatively poor black and white photos of the Berlin and Walderdorff fragments which are reproduced in Gamber’s edition, and they missed certain elements that a new investigation of the Berlin folios can enable. A discovery made here does offer a new connection of the fragments found in Regensburg to later manuscripts of the monastic foundation of Boniface in Fulda and to the insular community at Lorsch. Given this, a new treatment of these fragments is in order.

The Bonifatius-Sakramentar: a Description

The manuscript of the ‘Bonifatius Sacramentary’ is a relatively large book, and far from the archetypal ‘travel missal’ we might expect from the bags of a missionary, as it measures today around 328 × 243 mm, each folio larger than an A4 page, with a written area of c. 245 × 180 mm (see Fig. 1). In the Sacramentary portions, it is written over twenty-three lines and in two columns. The Walderdorff fragment and the Regensburg fragment both originate from the middle of a quire, while the Berlin fragment is the outer bifolium of a quire.Footnote 15 The script of the fragments had been characterised by Bischoff thus:

Calf parchment, ruled after folding in the insular manner … The script is a lovely, cultivated Anglo-Saxon [majuscule]Footnote 16 of Northumbrian type in the thin and terse form, which leads over to the minuscule … It uses [minuscule] d and uncial d; the minuscule forms of n and r, majuscule of s predominates … the red rubrics in the same writing style. Initials are in Anglo-Saxon style, surrounded by red points. The script and the type of material, as well as the preparation of it, are truly English.Footnote 17

Figure 1.

A portion of the Canon of the Mass in the ‘Sacramentary of Boniface’. Regensburg, Bischöfliche Zentralbibliothek, Fragment III 1, 1r (Northumbria, s. viiimed).

The particular script is now known in the English-speaking world as insular half-uncial.Footnote 18

Our fragment has the hallmarks of a Northumbrian production. Anglo-Saxon ‘diminuendo’ is practised at the beginning of the prayers after the initials, most consistently in the more elaborate Regensburg fragment. The initials are relatively simple, only drawn in black ink, but always surrounded by bright red points, while certain letters are also decorated with red dots making elaborate patterns within the belly of the letter (e.g. within the large P at the beginning of the mass for Sexagesima, which is the most elaborate letter of the Berlin fragment, there are five groups of red dots arranged in triangular forms or in figures of 8 in the H of Hanc Igitur in the Regensburg fragment). Rows of red dots and v shapes are also used between individual prayers of the Canon in the Regensburg fragment, to separate and distinguish them. The initial A has two forms, a more elaborate one in the Regensburg fragment, whose left vertical stroke descends significantly and whose strokes end in clubs, and a less elaborate one in the Berlin fragment, which ends in spirals. Such spiralling lines are often how the artist ends the final stroke of the letters P, C and D, leading into the belly of the letter, and likewise within an O in the Regensburg fragment. The initial VD for the prefaces in the Berlin fragment takes the form of a large U with a small d whose ascender is horizontal and crosses the stroke of the U. The Walderdorffer fragment has the most elaborate initials, the abbreviated KL for the word Kalendis at the beginning of the month, each one measuring about 12 lines of the manuscript. The same vocabulary of initials appear here, including the spirals, the bifurcated terminus, but added to them are lobes and filled in sections in colour, especially green, red and yellow, as well as the overlapping and looping forms that end in the head of a bird on the verso of the first folio. Red dots also form varying patterns around and within the KL initials. Finally, the Latin of the manuscript has a trait found in other insular manuscripts, the doubling of s, which can be seen in the common liturgical word ‘quaessumus’, a trait I have never seen in continental sacramentaries.Footnote 19 All three folios of the manuscript are markedly transparent, especially the Regensburg and Berlin fragments, which suggests the manuscript may have become wet at some point before it was unbound.

We know that the Boniface Sacramentary was on the Continent by at least the second half of the eighth century. Two mentions of Agilolfing family members were added, a notice for ‘theobaldi ducis’, the duke of Bavaria (d. 724), written by an insular hand dating to the second half of the eighth century, while the notice on the birth or death of Theoto (c. 770–after 793), the son of Duke Tassilo III (c. 741–c. 796) (‘natalis theotoni filio tassiloni duce’), in early Caroline minuscule is probably the latest addition.Footnote 20 These Agilolfing additions would likely pre-date the abdication of Tassilo in 788, but the calendar ceased to be updated beyond then. For the location specifically to Regensburg, we luckily also possess the addition of the commemoration of Saint Emmeram (‘et sancti emhramis’) in ‘an insular influenced pre-Caroline script’.Footnote 21

A date from the first half of the eighth century is often given to the manuscript, and has to be assumed for an attribution to Boniface to be correct; it was thus embraced wholeheartedly by Sotheby’s. But Bischoff’s later judgement, reproduced from a letter of the year 1954 in Mohlberg’s edition of the Berlin fragment, should be acknowledged before this date can be insisted upon:

One is not permitted to dare to be all too precise with the dating. I would not believe that one can go before the middle of the eighth century, not only because of the script, but also because of the initials (of the calendar); also the end of the century does not come, in my opinion, into consideration, but rather 750–784. The sentence about the origin, ‘written in England, in Northumbria’, I would rather, after more recent experiences, form thus: ‘written by an English scribe in the Northumbrian tradition’. Because, in principle, one must probably admit the possibility that a manuscript prepared by an English scribe and written in the native script, wherever it might be, does not differ tangibly from one created in England itself.Footnote 22

Nevertheless, in the case of our fragments, liturgical content permits a distinction to be made between truly English and English manufacture on the Continent.

The Liturgical Content of the ‘Boniface Sacramentary’

It is unfortunate that so little of the ‘Bonifatius-Sakramentar’ has survived. The surviving folios show it was a remarkable and singular testimony to the liturgy practised in Anglo-Saxon England before the periods of transformation of the ninth and tenth centuries, after which the surviving manuscripts of missals and sacramentaries from England show that the tradition it represents had been essentially extirpated by varied imports from the Continent.Footnote 23 Essentially, the contents of the three fragments are as follows.

  1. 1. Walderdorffer Fragment: a Calendar from June to October.Footnote 24

  2. 2. Regensburg Fragment: a section of the Canon Missae, the unchanging series of prayers that is said in every mass celebrated, from mid-way through the Communicantes prayer beginning with the list of Roman saints from ‘petri, pauli, andreae, iacobi, philippi … ’ to the end of the Libera Nos.

  3. 3. Berlin Fragment: proper masses for Christmas, Saint Stephen and for the season of Sexagesima, sixty days before Easter. Next to the mass offered for the Sundays of Sexagesima, faintly written in red in the margin is the numeral XVI, indicating it was the sixteenth mass of the Sacramentary. Such numbers may have appeared next to the Christmas and Stephen masses as well, but have been cut away.

The version of the Canon Missae written in the Regensburg fragment predates the seventh century, and shows some similarities to the equally archaic version transmitted in the Stowe Missal: Dublin, Royal Irish Academy, MS D II 3 (s.VIII/IX, Tallaght).Footnote 25 Also notable is the preponderance of central Italian saints in the calendar of the Walderdorffer fragment, especially among those feasts that are marked with the rubric ‘ORAT ET PR’, meaning ‘orationes et praeces’. As Siffrin and Gamber concluded, the rubric means that the original sacramentary had a mass given for that day in the sacramentary that followed. This is a practice unique to the Walderdorrfer fragment, but it strongly supports the original unity of the calendar with the sacramentary. In fact, the Stephen mass in the Berlin fragment is entitled ‘Orationes et praeces in natale sancti Stephani’, and the same title is also used repeatedly for masses in the fragment of a contemporary Anglo-Saxon sacramentary in Sankt Paul in Lavanttal, thus showing it was a common English trait.Footnote 26 Of the masses marked in this manner, ten are common to almost all sacramentary traditions, as they are well-known Roman feasts.Footnote 27 Eight are found in the tradition of the Old Gelasian Sacramentary, and these include characteristic central Italian saints of Lazio, Campania and Tuscany, who were never specifically celebrated in Rome, but, through the Gelasian tradition, were incorporated into many continental books.Footnote 28 But five of these saints, two from Naples and three from Capua, are not found in any surviving medieval sacramentary or mass book.Footnote 29 As Siffrin identified, many of the same saints also appear in the calendar of Willibrord, Paris, Bibliothèque nationale de France, Latin 10837, fol. 34–41, 44 (Echternach, s.viii1, before 728), though they are not there marked as ‘orationes et preces’, which is a feature unique to our calendar.Footnote 30 The Old English Martyrology, dating to the ninth century, shows that the ‘old sacramentaries’ of England had basically the same list of saints, while the ‘new sacramentaries’, Roman ones which we can identify with the Gregorian Sacramentary, did not.Footnote 31 This confirms a general veneration with proper masses of several Italian saints in Anglo-Saxon England, among them Neopolitan and Capuan saints who were never adopted in France. That our book ultimately had an archetype from Campania, and perhaps Capua itself, seems likely.Footnote 32

For an idea of how the sacramentary itself looked, we must turn to the Berlin fragment, the only piece that maintains the actual mass prayers that are the sacramentary’s main content. They comprise the remains of four masses: on the first folio, Christmas (which begins in the middle of the first of two prefaces), the mass of Saint Stephen (26 December), which ends in the post communion prayer, then two masses for the fasting time of Sexagesima (60 days before Easter). The second mass which ends after one line of the preface, is designated to be intended for all the Sundays of a Sexagesima season, and not just the one Sunday before Lent began, as is elsewhere usual (here: ‘Inter ieiunia sexagessimae diebus dominicis’). The other mass, which lacks its title, may have been intended for the single Sunday of Sexagesima itself (per Gamber), or possibly for celebration of the week days in the fasting season (according to Siffrin).Footnote 33 Because the mass provides for many Sundays of Sexagesima, the mass book seems to have intended to provide masses not for a forty-day Lenten fast, but rather a sixty-day fast that included Lent.Footnote 34 This custom, known early on in the Greek East, is suggested by witnesses like the pericope lists in the Lindisfarne gospels, and in the sixth-century Capuan Gospel Book in Fulda, the ‘Victor Codex’ (Fulda, Landesbibliothek, MS Bonifatianus 1), which was likely brought to the Continent by Saint Boniface.Footnote 35 It may therefore have been the practice in Campania, but no other surviving sacramentary provides a fasting period only for Sexagesima, undifferentiated from the rest of Lent, in this fashion.

Equally significant and rare in the text of the Berlin folios is the provision of multiple choices of each prayer in the mass sets of the Temporale. The Christmas mass set has two prefaces, two post communions, and a prayer AD POPULUM. One Sexagesima mass set has two secrets, two prefaces and an AD POPULUM. The other mass set intended for the Sexagesima Sundays has no fewer than three collects and three secrets, and may have had more than one preface originally. For this final mass, which is the only mass of the Temporale whose title has survived, it is made clear in this title that the mass set was intended to offer a choice of texts for multiple masses celebrated in the season, i.e. the priest might choose which sequence of prayers from the set he would celebrate on each Sunday in the Sexagesima season. We can assume the same was true of the Christmas mass set, whose choice of texts likely covered the Christmas season until Epiphany. In each case, the priest could assign a combination of prayers as he wished to each of the days in the period. This relative freedom in comparison to almost all surviving later sacramentaries is another sign of the relative antiquity of the model for the Boniface Sacramentary.Footnote 36 As the mass for Stephen could only fall on one day, 26 December, it does not offer options of prayers, but only one of each, and this was likely true of the other masses of the Sanctorale.

Of the surviving prayers, Gamber, Siffrin and Mohlberg offered the following conclusions about the parallels in other sacramentary traditions.

  1. 1) Nine texts found in the Old Gelasian Sacramentary. Bon 19, 20, 21, 23, 27, 28, 32, 33, 34.

  2. 2) Six texts appear in the Veronense. Bon 20, 23, 36, 37, 38, 39.

  3. 3) Five texts found in no established sacramentary tradition. Bon 17, 18, 24, 25, 26.

In addition, three prayers not found elsewhere are found in the Gelasian of the eighth century (Bon 22, 29, 30), and four in the Ambrosian sacramentaries of Milan (Bon 30, 31, 36, 38).

Thus, the Boniface sacramentary uses prayers that are also found in the Old Gelasian Sacramentary, especially those in Sexagesima, but, however, they are not in the same place or the same order as the manuscript of the Old Gelasian itself.Footnote 37 The Old Gelasian is found complete only in one manuscript, Cittá del Vaticano, Biblioteca Apostolica Vaticana, Regin. Lat.316 (Chelles, s.viiimed.).Footnote 38 While an essentially Roman origin for the Old Gelasian was once assumed, as argued by Chavasse, it is more likely that the Old Gelasian was a compilation made from some Roman liturgical booklets, along with other Italian material, somewhere outside of Rome.Footnote 39 Among the best evidence for this is the many central Italian saints whom the Old Gelasian apparently shared with our Boniface Sacramentary, as witnessed in the Walderdorffer calendar fragment. We have no evidence, for example, that Rufus or Priscus of Capua, both of whom have masses in the Old Gelasian, were celebrated with proper masses in Rome itself, but masses for them were found in mass books in England.Footnote 40 Another fragment from an Anglo-Saxon centre with strong parallels to the Old Gelasian is the folios in uncial, London, British Library, Add. MS 37518, 116r–117v (England?, s.viii1, provenance is Belgium from s.xii), with matins and vespers prayers appearing largely as in the Old Gelasian.Footnote 41 Important is Baumstark’s observation that the quire signature twelve surviving on this London fragment meant it was significantly shorter than the Vatican Old Gelasian manuscript. Thus, the English books were not identical to the Vatican Old Gelasian as copied at one of the nunneries in the Paris basin, probably Chelles, and not only in the choice of some prayer texts that do not appear in the latter.Footnote 42 The copyists at Chelles must have substantially reworked any Anglo-Saxon exemplar they had before them, and the exemplar also may not have come from the same centre as our Boniface book at all. But the common material that the Vatican copy shares with these English fragments, and the central Italian mass sets of the Gelasian, clearly imply that significant material that marks out and distinguishes the Vatican Old Gelasian was imported to the Continent from the insular world, as Bullough, Hohler and Hen have all suggested.Footnote 43 In fact, the title ‘Orationes et Praeces’ is used consistently for masses in the Old Gelasian as it was used in our Boniface Sacramentary, which is probably also the influence of its English model, as the term is not commonly used to mean a mass anywhere else.Footnote 44 It is not known whether the Boniface sacramentary was divided in three ‘books’ like the Vatican Old Gelasian, which some have supposed was an English trait.Footnote 45

On the other hand, several of the prayers exist also in the even older traditions of the Veronense compilation, partially preserved only in Verona, Biblioteca Capitolare, cod.LXXXV (s.vii, Northern Italy, likely Verona).Footnote 46 While Veronense contains mass prayers written in Rome in the fourth and fifth centuries, it proves that such prayers were still circulating elsewhere in Italy until the seventh century or later, as the only surviving copy seems to come from Verona. In fact, Veronense also uses the term ‘Orationes et Praeces’ twice, though in a more restricted and specialised way.Footnote 47 The preface for the Sexagesima feast in the Boniface Sacramentary is, however, the only text which is only otherwise found in Veronense and nowhere else.Footnote 48 It is possible that unique elements of the Boniface Sacramentary, like the secret, preface and post-communion text of the Stephen mass, might have been originally preserved in similar collections, as the Veronense itself manifestly does not contain everything that was circulating in Italy.Footnote 49

However, various prayers in the surviving Berlin fragment of the Boniface Sacrametary belong neither to Gelasian nor Veronense, as they today survive. A beautiful preface for Christmas, for example, with a striking creedal statement, confirms that the Boniface Sacramentary must have taken some of its unique texts from an original central or southern Italian source, which no longer survives. In the Boniface Sacramentary, the text of the preface, with my translation following, runs:

Vere dignum. Qui in principio erat aput te deum patrem uerbum. Per quem facta sunt omnia. Qui coeternus tibi et in tua atque ex tua substantia manens homo factus est et coepit esse quod non erat. et quod tecum erat esse non desinit in saecula et ante saecula deus. In similitudinem passionum nostrarum secundum hominem nasceretur et peccati nescius habitum peccati carnis adsumeret. In substantiam diuinitatis tuae naturam generis mortalis accipiens ut quod nostrum est suscipiens commonicaret ille nobis quod suum est. Quem laudant.

(It is worthy and right. Who in the beginning was with you God the Father, the word though whom all things were made. Who is coeternal with you and in you, made from your substance. He was made man and began to be what he was not, but did not cease to be what he was with you, God, in the ages and before the ages. In the likeness of our sufferings he was born as a man, and he did not know sin, and did not assume the habit of the sins of the flesh. In the substance of your divinity, accepting the nature of the mortal race, that, receiving what is ours, he might give us what is his. Whom [angels] praise.)

The closest surviving parallel to this text is in Benevento, Archivio Archeviscovile, VI 33 (Benevento, x/xi), 3(bis)r, the earliest of the surviving Beneventan mass books, and which presents this same preface, but with a longer ending.Footnote 50 This indicates, once again, that the Boniface Sacramentary transmits to us Italian traditions from outside Rome, which had a certain afterlife in the Old Gelasian, but which mostly were replaced in areas of Northern Italy that accepted the Gregorian Sacramentary under Carolingian influence in the ninth and tenth centuries. Further south and not under Carolingian control, Benevento remained more untouched, and continued to transmit this text until the end of the tenth century, but it vanished in later Beneventan missals.

However, we actually have another earlier witness to a form of the same preface among the Anglo-Saxons on the Continent in a fragment found in Basel, Universitätsbibliothek, Fragment I 3a and I 3b (viii2, Anglo-Saxon centre on the Continent), this time one clearly written on the Continent, according to Bischoff, in which the preface begins similarly, but is then rather abruptly and clumsily adapted to the feast of Candlemas, and to the dedication of Jesus in the Temple.Footnote 51 That the Basel fragment was in some way associated with Fulda is strongly supported by the fact that its version of the preface for Candlemas, unknown to Dold or to Gamber, is also found in the curious and antiquated collection of prefaces in Merseburg, Domstiftsbibliothek, cod. 136, 75v–76r (ixmed., Fulda).Footnote 52 The later, well-known Sacramentary of Fulda in Göttingen also uses the Candlemas preface, albeit in a version that has been even more edited and only hints at the original text.Footnote 53 That the Boniface Sacramentary in Regensburg and the tradition of Fulda align in a partly shared text of a rare Beneventan preface suggests that Fulda had access to mass books resembling the Boniface Sacramentary at one point in its history. Indeed, a firm creedal statement like this one would have suited Boniface’s stringent concern for orthodoxy in his mission work.Footnote 54

A Trace of the Tradition at Fulda and Lorsch

There is one final element of the Boniface sacramentary, found in the Berlin fragment, that had escaped the liturgical scholars who examined only the black and white photographs of it. This is the inobtrusive rubricated title squeezed onto the line next to the beginning of the Communicantes prayer provided for the Christmas mass (Berlin, lat.877, 1r, second column, lines 9–10; see Fig. 2). This is invisible in the black and white photos.

Figure 2.

A folio containing a mass for Christmas in the ‘Boniface Sacramentary’. Berlin, Staatsbibliothek zu Berlin, Preußischer Kulturbesitz MS Lat. fol. 877, fol.1r (Northumbria, s. viiimed).

Here, the prayer is entitled ‘CONIUNCTIO’. Though the titles of mass prayers generally vary little, this title is very unusual. In almost every other surviving sacramentary, this kind of interjection, which replaced one of the fixed prayers of the Roman Canon on the most important days, is simply called ‘INFRA ACTIONEM’, meaning it is inserted into the ‘act’, which is to say, the Canon. This is true even in the Old Gelasian, in which the same prayer is given for the Christmas mass as in our text.Footnote 55 The term ‘CONIUNCTIO’ for this prayer appears otherwise in the sacramentaries from two distinctive places, the monasteries of Lorsch and Fulda.

  1. 1. Göttingen, Universitätsbibliothek, cod. theol. 231 (s.x, c. 975, Fulda), uses it seven times.Footnote 56

  2. 2. Merseburg, Domstiftsbibliothek, cod. 136, 45r, 47v, 65v, 67v (in several hands, but applied to the Hanc Igitur interjection, not the Communicantes) (s.ixmed., Fulda).

  3. 3. Augsburg, Universitätsbibliothek, Cod. I.2.4° 1 (fragment), at fol. 1v (s.ixex., Lorsch).Footnote 57

  4. 4. Cittá del Vaticano, Biblioteca Apostolica Vaticana, Palat. lat. 495, at 61v, 79v, 99r, 101r (s.x, Lorsch).Footnote 58

  5. 5. Chantilly, Bibliothèque du château, Ms. 40, at 12r (s.xex, Lorsch).Footnote 59

  6. 6. New York, Burke Library, Ms. 59 (fragment), at 2v (s.xi, Lorsch).Footnote 60

This term is thus characteristic of the later liturgical monuments of these two monasteries with a strong Anglo-Saxon heritage.Footnote 61 It continued to differentiate Lorsch and Fulda’s sacramentaries into the tenth and even the eleventh centuries, though it faded from use after that. But it had never been found before in an Anglo-Saxon book. Our ‘Boniface Sacramentary’ proves that the term was actually used in Anglo-Saxon books and that it was then passed to Fulda and Lorsch from the ancient Anglo-Saxon sacramentaries, which they must have, at one time, possessed.

Nevertheless, there are at least two other manuscripts that use the title ‘CONIUNCTIO’, and they are found in Italy, specifically Verona.

  1. 1. Verona, Biblioteca Capitolare, cod. XCI, 176v (Verona, ix1), which uses it for a Te igitur prayer.Footnote 62

  2. 2. Verona, Biblioteca Capitolare, cod. XCVII, 17r, 23v, 59r, 69v, 81v, 85v (Verona, c. 1000), which uses it exactly as our text does, for the Communicantes.

This secures a likely Italian origin for the term. Our oldest Roman (pre-)sacramentary, copied also likely in Verona, the so-called Veronese, uses ‘CONIUNCTIO’ once, in another similar context, for the special prayer ‘Hanc etiam oblationum domine tibi uirginum sacratarum …’ added during the Canon for the mass said at the consecration of Virgins.Footnote 63 As in the consistent use of ‘orationes et praeces’ to mean a mass, our ‘Boniface Sacramentary’ therefore shows an antiquated terminology, as it is clear the term is a very old one, even if the usage seems to have evolved with time. In our case, the term suggests a ‘joining’ (conjunction) of those celebrating the mass with the saints named in the Communicantes and thus the eternal church, or perhaps also an ‘interjection’ in the Canon, as in cases where it replaces other terms in the Canon. Notably, however, it is never found in the admittedly later Beneventan and central Italian books, where we might expect the strongest traces of the oldest layers of Anglo-Saxon liturgy to be, nor was it used by later books from Regensburg.Footnote 64 While Gamber wondered if it was Roman in origin, it might be that this usage was a distinctive old feature of certain liturgical books in Italy, and perhaps specifically in the North in Verona, or maybe Ravenna.Footnote 65 Consistently, the Italian features of the English mass books are brought into connection with the figure of Abbot Hadrian of Nisidia near Naples (d. 710), who came to Canterbury and not Northumbria, yet we should admit that Italian influence on the English sacramentary tradition could have proceeded through various actors and varied channels, including from other centres throughout the peninsula, and from Anglo-Saxons bringing books back from Italy too.Footnote 66 What is certain is that the term ‘CONIUNCTIO’ was known in our Anglo-Saxon book, and its lasting usage in Fulda and in Lorsch must come from the Anglo-Saxon heritage of the two monasteries.

Thus, while it is unlikely that the ‘Boniface Sacramentary’ truly belonged to Saint Boniface himself, it is clear that it represents the kind of book that he and his fellow missionaries must have also brought to Fulda and to Lorsch. Indeed, Frank was able to show that Boniface repeatedly drew on an Old Gelasian prayer in his letters, which at least proves he shared some material likely to also have been in a tradition that resembled our book.Footnote 67 The mass for the Vigil of Ascension entered into Willibrord’s calendar (Paris, Bibliothèque nationale de France, Latin 10837, 42v–43r), at a similar date to the writing of the Regensburg Sacramentary, is a similar mixture of Old Gelasian with additions from Veronense and also offers an alternative prayer each for three of the texts of the mass, indicating, it seems, that this was generally the kind of liturgy written down by the Anglo-Saxons in the (earlier) third quarter of the eighth century.Footnote 68 What is more, the ‘Boniface Sacramentary’ is also strikingly different from anything else present on the Continent in the eighth or the ninth centuries. The Vatican Old Gelasian (possibly c. 750) is its closest relation, but still with only around half the material of the Berlin fragment in common. If Bischoff’s suggestion that the initials of the ‘Boniface Sacramentary’ should be dated to after 750 is accepted, the Regensburg Anglo-Saxon book remains a monument of the missions that Boniface inspired and contacts he fostered, but a later date does invite us to adjust our focus and consider those lesser known or unknown figures who followed in his footsteps to Bavaria — one of whom carried from Northumbria to Regensburg a mass book of grand size and appearance but of distinct and antiquated character, which must have been as striking on the Continent then as it remains today.Footnote 69

II: The Gelasian Sacramentary from Groß Sankt Martin

The second of two apparently Northumbrian fragmentary sacramentaries on the Continent has not undergone the same level of analysis as the Bonifatius Sacramentary and has never been edited. Thus, I offer an edition of the text in the Appendix to this article. The orthography of the manuscript is here maintained, including where the scribe of the Münster fragment writes out the syllable ‘ae’ while the other scribes do not. It is today divided between three libraries.

  1. 1. Münster, Universitäts- und Landesbibliothek, Fragmente Kapsel IV, 8.Footnote 70 One and a half folios (half folio cut vertically). This fragment was first identified by Bernhard Bischoff.Footnote 71 The most extensive description so far was by Freise.Footnote 72 A Werden provenance is known thanks to the note written vertically across the now separated two fragments, in which the Werden Prior Johannes Hittorp (1517–1542) notes that he borrowed this book (‘hunc librum’), a copy of Smaragdus of Saint Mihiel’s Commentary on the Benedictine Rule, from the collection of Groß Sankt-Martin in Cologne, with the allowance of the latter’s abbot, Gerhard II von Loe (1507–1547) and asked that it be returned after his death.Footnote 73 It seems that this was not followed as, like some other Werden fragments, the fragment came to Münster after the secularization of Werden in 1803.Footnote 74 Freise made the most of the association with Werden, asserting that the sacramentary ‘was kept in the Werden Monastery Library since the ninth century’, and thereby offering the hypothesis that it could be linked with the Frisian Saint Liudger, the founder of Werden Abbey, who could have acquired it in York in the 760s or 770s, perhaps from Alcuin himself.Footnote 75 But Hittorp’s note actually proves that the book to which the fragment was bound was not at Werden at all before the sixteenth century, but rather was acquired by Hittorp at Groß Sankt Martin in Cologne. The original sacramentary had been destroyed there and used to cover a copy of Smaragdus in the collection of the Cologne monastery which Hittorp took back to Werden.

  2. 2. Cologne, Historisches Archiv, Best.7001, Fragmente B, Nr. 24, Nr. 123 and Nr. 124.Footnote 76 Nr. 24 comprises a double folio and two single folios. The third folio (a singleton) is essentially illegible. Nrs. 123 and 124 are two halves of a single folio, cut in half horizontally across the middle, and both clearly used for the fly leaves of the same book. The two fit perfectly together, without loss of script, but the inner sides of each folio have suffered. The name of Groß Sankt Martin occurs on one side of each leaf (‘Bibl. Mart.-maj’), indicating they all came from that monastery’s library. Of the three, Nr. 24 was first located by Cunibert Mohlberg, who shared photos of these folios with Bannister, who published them in 1911, but Nrs. 123 and 124 have never been studied.Footnote 77 A later study of Nr. 24 was undertaken by Frank.Footnote 78

  3. 3. Paris, Bibliothèque Nationale de France, Latin 9488, 3r–4v.Footnote 79 This fragment has traditionally been identified with Echternach, but this is not proven.Footnote 80 There is liturgical, palaeographical and decorative evidence that shows it is also the same original sacramentary as the other two fragments. As it is clear only one complete mass is missing between the two folios, we have here the second innermost two folios of a quire, with just one further bifolium originally being between them.

First, we have established that the Cologne and Münster fragments actually have the same provenance: the monastery of Groß Sankt-Martin in Cologne. Contact with Werden is only a product of a lax late medieval book return policy, and their possession by Werden earlier or by Liudger is therefore exceedingly unlikely.Footnote 81 The likelihood that both Cologne and Münster fragments belonged to the same original sacramentary, which has been mooted several times before, is, however, considerably enhanced.Footnote 82 For proof, it suffices to compare the physical state, palaeography and initials of the fragments, while the content, as we shall see, makes the assocation of the three fragments undeniable (see Figs. 3 and 4). With the almost complete height of the original book preserved, Cologne, Historisches Archiv, Fragm. Nrs. 123 and 124 together measure 320 × 215mm, giving us the best idea of how large the complete book was, and that it was originally comparable in size to the Boniface Sacramentary. The portions in Cologne, Fragm. Nr. 24 were more severely cut and measure 207–10 × 148–58 mm. The more complete fragment in Münster, Fragment IV, 8 fol. 2 measures 272 × 247 mm, giving us the best evidence of the original width, since its side margins are more intact. The Paris fragments are currently 280–3 × 205–20 mm, and thus also fit within the same range, though they have been cut at both sides and at the top and bottom.Footnote 83 The number of lines is similar in each case: originally 25 lines in the Paris fragment (24 with one missing in the second folio), 25–6 lines on the page of the Cologne fragment, 25 lines in the Münster fragment, though 27 in the last two folios of Fragm. Nr. 24 of Cologne. But it is likely the difference in the number of lines simply represents a departure of the later scribes towards the end of the sacramentary from an earlier practice, perhaps in order to finish the sacramentary on the available parchment. Both fragments present their text in two columns, which measure c.75 mm horizontally and c. 275 mm vertically each in the Münster and Cologne fragments. The condition of the Cologne leaves is markedly worse, but the Münster and Cologne fragments have both essentially lost their rubrication. In the Paris fragment, the rubrication remains, and the titles of masses and individual mass prayers are written in the same insular half uncial as the main text, only red coloured. This would accord with what can be glimpsed of the remains of the rubrication in the other two books, and was the same in the Boniface Sacramentary too, showing that these early Anglo-Saxon mass books did not tend to change scripts for the titles of masses, as many continental books of the same genre do.

Figure 3.

Half of a folio, cut vertically, containing a blessing for the fifth Sunday after Epiphany in a Gelasian Sacramentary from Groß Sankt Martin. Münster, Universitäts- und Landesbibliothek, Fragmente Kapsel IV, 8, 1r (Anglo-Saxon centre on the Continent, (Cologne? Echternach?), s.viii2).

Figure 4.

A folio containing masses for Saint Juliana and the Cathedra of Saint Peter in a Gelasian Sacramentary from Groß Sankt Martin. Münster, Universitäts- und Landesbibliothek, Fragmente Kapsel IV, 8, 2v (Anglo-Saxon centre on the Continent (Cologne? Echternach?), s.viii2).

The use of a majuscule script, ‘decorative capitals’ in Bischoff’s words, with the letters filled in yellow and red, for the first word of the first prayer of a mass following the initial is an identical practice in both the Münster and Cologne fragments, and strikingly differentiates them from the Boniface Sacramentary. Alike between two of the fragments are ‘OMNIPO’ on Münster, Fragmente Kapsel, IV, 8, 2r at the beginning of the mass of Saint Juliana and ‘OMNIPS’ at the opening of the ‘Dominica uacat’ in Cologne, Fragm. Nr. 124. Smaller and more stylised capitals are used in Münster, Fragment IV 5, 1v for ‘FAMILIAM’ at the opening of the mass for the fifth Sunday after Epiphany. The Münster artist even re-draws around the inner portions of these letters in a further layer, now white, but possibly originally red lead, since degraded, then a second black line, which almost gives them the startling effect of 3D perspective. Though, as is well known, the practice of elaborate display script was used in insular gospel books at the opening of each of the four gospels, the highlighting of the first word or line of the individual mass sets of a sacramentary in display script, and thus a much more lavish use of display script, is not seen in the Sacramentary of Boniface or other known Anglo-Saxon sacramentaries of this time. But it can be seen in continental books of this genre, for example, in the Missale Gothicum, Cittá del Vaticano, Biblioteca Apostolica Vaticana, Regin. lat. 317 (s.viiiin, Autun or Faremoutiers), e.g., 89r, 93v, 133r.Footnote 84 However, it seems that this practice was only done in the first half of the sacramentary, which would be the Temporale and Sanctorale, represented by Cologne, Fragm. Nrs. 123 and 124, and the Münster fragments. In the common masses of the Paris fragment, and the votive masses of Cologne, Fragm. Nr. 24, this practice is no longer carried out, and it probably distinguished the parts of the sacramentary with the greatest dignity and importance, i.e. the cycle of the year.

Several different hands were at work in our sacramentary, and Bannister identified two in Cologne, Fragm. Nr. 24. But it is clear that the Münster fragment was the work of the master. He wrote more elegantly, fitting just around 15 letters on the line, while one of the Cologne scribes in Fragm. Nr. 24, fols. 3–4, very distinct from the others, can get up to 23 letters. The Münster fragments have an additional form of punctuation, the ‘point and long angular comma’, as the CLA Addenda calls it, at the end of the prayers, that is not present in the Cologne fragments or the Boniface sacramentary, standing for the ending of the prayers, and represent the words ‘per dominum nostrum’ that the scribes of the Cologne fragments write out in full. But the Paris fragment shows a similar punctuation practice, especially visible in Paris lat. 9488, 3r, where the endings of prayers also have a long angular comma with multiple points, or three commas. Particularly, in the right column of 3r or 4r, one can even see a nearly identical form of punctuation to that used in the Münster fragment, which is identified by Bannister as the Tironian Z standing for ‘per’.Footnote 85 Then, on 4v, the per with a p abbreviation is written out, as in the Cologne, Fragm. Nr. 24. Thus the Paris fragment is linked to both others. Other fundamental characteristics of the script are the same in all three fragments, including, especially, the ‘triangular finial at the left on the top stroke of g and t’ (CLA Addenda). The finial is also present in the Paris lat. 9488 fragment, as it is not in, for example, the Trier Gospels (s.viii1, Echternach), or the plainly earlier Boniface Sacramentary.Footnote 86

The initials of the Münster fragment are also considerably more accomplished, which suggests they too were the work of the Northumbrian master scribe. They clearly use the same basic vocabulary of forms as the Boniface Sacramentary, the IN on Münster, Fragment IV, 8, 2v can be compared to the IN of the Regensburg fragment of Boniface. But the initials are more elaborate in Münster than in Regensburg. The profuse spirals are drawn much thinner and longer, looping around many more times. The initials are also filled with colour. The D of ‘DEUS’ in the mass for the Cathedra of Peter, for example, is filled with repeating interlacing forms in figures of eight, and each compartment filled in red and yellow. Bird heads appear consistently between the fragments, on the DS on Münster, Fragment IV, 8, 1r (with a flowering plant in its open mouth), and on the DS of Cologne, Fragm. Nr. 24, 2r, which is basically identical to the more visible DS in Paris lat. 9488, 4r.Footnote 87 The bird is especially well formed on the IN on Münster, Fragment IV, 8, 2v (in yellow, red and white) and curling up inside the belly of H of ‘Haec’ on Cologne, Fragm. Nr. 123v, col. 2. But the Cologne and Paris fragments are significantly weaker in execution. Their spirals are heavier and thicker and the colours more thickly applied, without the delicacy of the Münster fragments (Figs. 5 and 6).

Figure 5.

A half folio cut horizontally, containing the end of the mass for the Ember Saturday and opening of the Dominica Vacat in a Gelasian Sacramentary from Groß Sankt Martin. Cologne, Historisches Archiv, HAStK-RBA Best. 7050B (Fragmente B), 124v (Anglo-Saxon centre on the Continent (Cologne? Echternach?, s.viii2).

Figure 6.

A folio from a fragmentary Sacramentary, containing the end of a Common Mass for a Virgin. Paris, Bibliothèque nationale de France, Latin 9488, 4r (Anglo-Saxon centre on the Continent (Cologne? Echternach?, s.viii2). Source gallica.bnf.fr/Bibliothèque nationale de France.

This might suggest we have another product of a Northumbrian scriptorium. In the examination of the Münster fragment for the Supplement to CLA, Bischoff stated baldly that the sacramentary was ‘written in Northumbria’, despite noting privately that this was very difficult to decide.Footnote 88 But the CLA entry on Cologne, Fragm. Nr. 24 admits more doubt: ‘Written probably in Northumbria, possibly in a Continental centre under Northumbrian influence’.Footnote 89 Before this, Bannister had admitted the same difficulty: ‘the same script was used in the houses founded by Northumbrian monks of that date on the banks or in the neighbourhood of the Rhine’.Footnote 90 Of the Paris fragment he supposed, ‘some monastery in the East of the Frankish kingdom which had inherited English MSS or where English scribes were employed’.Footnote 91 The dating of the Münster fragment by Bischoff to ‘saec. VIII1’ is also bolder than the CLA on Cologne, Fragm. Nr. 24 or on Paris lat. 9488, which has only ‘VIII’ in each case. Meanwhile, the Paris fragment was dated very boldly by Bannister to the first ten, if not the first five, years of the eighth century.Footnote 92 The Paris fragment was also elsewhere argued to be ‘roughly contemporary’ to the Trier Gospels (Trier, Domschatz, MS. 61 (VIII1, Echternach), and located to Echternach, described as written in a similar half-uncial of ‘Phase II’, dated by Netzer between 720 and 740.Footnote 93

Some of these conclusions might suggest the Sacramentary of Groß Sankt Martin was a contemporary Northumbrian production to the Boniface Sacramentary, and Mohlberg, for one, asserted that the script was very close, but a comparison of the two, especially on the basis of the initials of the sacramentary, suggests that the Groß Sankt-Martin Sacramentary should be regarded rather as a later book.Footnote 94 It also lacks conspicuous Anglo-Saxon forms of Latin, but rather has spellings that were described by the CLA as rather unusual for Anglo-Saxon manuscripts.Footnote 95 But the question of when and where it was made is significantly complicated by the liturgical content of the Sacramentary.

The Liturgical Content of the Groß Sankt Martin Sacramentary

The Münster fragment originally came first, as it offers us the end of Candlemas (2 February) (followed by a gap), then Agatha (5 February) and the beginning of the fifth Sunday after Epiphany, then there is another gap before fol. 2 begins, on which we find the feasts of the Roman martyrs Valentinus, Vitalis, Felicula and Zeno (14 February) on 2r, the Gelasian mass of Juliana of Nicomedia or Cumae in Campania (16 February), and the beginning of the mass for the Cathedra Sancti Petri (22 February). The single folio in Cologne, Fragm. Nrs. 123 and 124 gives us most of the mass for the Ember Saturday of September and the following ‘Dominica uacat’. The Paris fragment offers masses for the Common of Saints, offering on the first folio the end of the mass for a single martyr, then continuing on the next with the second half of the mass for a single virgin, and first half of a mass for many saints. Cologne, Fragm. Nr. 24 begins with the rites for the nuptial mass (fol. 1), then offers a series of votive masses, for finding serenity, for storms (fol. 2), for peace, for charity and almsgivers (fol. 3), then a mass for a new house, and a series of blessings of new fruits, including those for grapes or beans and apples at the end (fol. 4).

A complication to the hypothesis of a Northumbrian origin and an early date for the manuscript is the type of sacramentary it represents, which is entirely distinct from the Boniface Sacramentary. Here, the union of all of these fragments is again established. As Gamber identified, the Münster fragment contains texts that align with the Gelasian Sacramentary of the eighth century, especially the tradition he calls the ‘S-Typus’, which is best represented by Sankt Gallen, Stiftsbibliothek, cod. 348 (Chur-Rhaetia, s.viiiex.).Footnote 96 Some of the most characteristic masses of that tradition appear here, especially the Cathedra of Peter. This was an originally Roman feast but its mass form was not preserved in any surviving Roman sacramentary. It was only preserved by older liturgical currents preserved in books copied outside Rome, especially the Gelasian of the eighth century. This was noted in Willibrord’s calendar, though it is not in the Old Gelasian. There is also the celebration of Felicula, Vitalis and Zeno alongside Valentine on 14 February; while Willibrord’s calendar mentions only Valentine, and the Old Gelasian celebrates three of them, but not Zeno.Footnote 97 The ordering of the masses, in which a mass of the Temporale (the fifth Sunday after Epiphany), appears with the masses of the Sanctorale is that of the Gelasian of the eighth century, not the Old Gelasian, which holds these different masses in distinctive sections. All of this is true of the folio made up by Cologne, Fragm. Nrs. 123 and 124. Their texts for the Ember Saturday also follow the Sankt Gall manuscript exactly.Footnote 98 Equally, the Paris fragment also belongs to the same tradition, as Bannister identified long ago, and gives its common masses in exactly the same order as Sankt Gall, cod. 348.Footnote 99 Notable also is that the fragments also all show the text of Sankt Gall, cod. 348 before it was later corrected, thus the exact same version of the Gelasian text as well.Footnote 100

Because Bannister was working on his study of Cologne, Fragm. Nr. 24 before the edition of Sankt Gall, cod. 348 by Mohlberg in 1918, he chose to compare Fragm. Nr. 24 to the Old Gelasian, which is, one might say, a cousin of the Gelasians of the eighth century, but one with a very different order and many different prayer texts.Footnote 101 Sankt Gall cod. 348 has also lost the votive masses which would parallel Cologne, Fragm. Nr. 24. This disguised the fact that these fragments are just as much parallel to the Gelasian of the eighth century as the rest of the book, as was established by several scholars after Bannister.Footnote 102 Thus, in other examples of the Gelasian of the eighth century, we find all the prayers of Cologne, Fragm. Nr. 24, including those Bannister could not find in the Old Gelasian, and they are also in the exact same order in the Gelasian of the eighth century, which differs from the Old Gelasian. For example, a contemporary companion piece of Sankt Gall, cod. 348, the more fragmentary Sacramentary Sankt Gall, Stiftsbibliothek, cod. 350 (Chur/Rhaetia, s.viiiex.), has preserved its votive section better, which always parallels Cologne, Fragm. Nr. 24.Footnote 103 Thus, the fragments in Münster, Cologne and Paris belong to the same sacramentary tradition. This confirms that they were all from the same manuscript, a large and fine copy of the Gelasian of the eighth century in a style of Northumbrian half uncial. This book was likely at Groß Sankt Martin in Cologne, possibly where it was destroyed and used to bind several other books, one of which, a copy of Smaragdus, was lent to Werden, which is why we find fragments of two folios today in Münster. The provenance of the Paris fragment remains obscure, but certainly an original presence in the East of Germany is quite plausible, before it perhaps crossed into France.Footnote 104

But the possible dating of the manuscript to the first half of the eighth century is, as Gamber rightly emphasized, problematic for the established understanding of what the Gelasian of the eighth century is, and whence it comes.Footnote 105 The consensus among liturgists remains that the Gelasian of the eighth century was created at the monastery of Flavigny towards 770.Footnote 106 Nevertheless, some of the reasoning used to construct this date and location was rather tendentious, as Moreton and Gamber made clear.Footnote 107 The theory of Flavigny origin, which uses a mass for Saint Praiectus of Clermont in many, but not all, copies of the Gelasian of the eighth century as a base, was undergirded by the assumption that one must seek a powerful patron for a new mass book ‘type’, as it was simply assumed that a new book would never have found success without political force from the centre behind it.Footnote 108 In this case, it would be Pippin III (c. 714–768), whom liturgists sometimes claimed had founded Flavigny, though he did not.Footnote 109 But questioning the association with Pippin allows us to envision the sharing of the text among travelling ecclesiastics, and in communication through monastic networks, and perhaps even among the Anglo-Saxons on the Continent. Allowing a dating to the first half of the eighth century, the Groß Sankt Martin fragments would be the oldest known copy of the Gelasian of the eighth century by half a century, as Sankt Gall, cod. 348 is usually dated around 800, and the Sacramentary of Gellone to the last years of the eighth century. Yet it is striking that, while the existence of Cologne, Fragm. Nr. 24 was known to Bourque, he left this fragment, apparently the earliest surviving witness to the tradition on palaeographical grounds, entirely unmentioned in his synthetic discussion of the origins.Footnote 110

Frank, however, recognised the problem in his investigation of Cologne, Fragm. Nr. 24. He returned to Bischoff, from whom he received further palaeographical insight in a letter of 1952, which he reproduced. Bischoff was markedly less certain here:

I must emphasize again that it is difficult to date Anglo Saxon majuscule exactly. One could perhaps even speak of an artificial tradition of this type of writing for liturgical codices, which lasted into the late eighth century. North English character must be correct, but naturally that also applies for the continental colonies. In my opinion, the tendency to add rather weak ornament on and in the black initials witnesses more surely to their creation in the second half of the eighth century. With liturgical manuscripts even more than with other categories (e.g. legal books), it is to be expected that the absolute chronology of the manuscripts can by no means be harmonised with the appearance of the new recommended types. I am inclined to date Regin.316 [the Old Gelasian] to the second quarter of the eighth century. The chronological relationship of the text types must therefore be analysed primarily according to their internal developmental history.Footnote 111

This ‘tendency’ in the ornament, noted by Bischoff, was exactly what we previously suggested distinguished most obviously the Groß Sankt Martin book from the Boniface Sacramentary. All of Bischoff’s reservations apply as well to the Münster fragment, and to the further pieces in Cologne and Paris. With this uncertainty, we must agree with Bischoff that the liturgical tradition of the text type should take primacy over the pure palaeographical verdict, leaving us only the strong probability that the text is later than the Boniface Sacramentary. Other manuscripts of the Gelasian of the eighth century are dated by palaeography to the late eighth century, and none can be securely dated to any time before 800. The probability, therefore, would be that our sacramentary also significantly postdates 750. Renewed caution about the dating of Northumbrian manuscripts, especially as expressed in CLA, seems a prudent broader lesson. Still, the version of the Gelasian of the eighth century that we find in in our book, the version shared with Sankt Gall, cod. 348, is the earliest form of this tradition.Footnote 112 The Groß Sankt Martin Sacramentary could comfortably be twenty years older than Sankt Gall, cod. 348, whose only secure terminus ante quem is the year of the death of Bishop Remedius of Chur, whose memento was given in a later addition, that is 806. The Sacramentary of Gellone, which is already a later development of the type than our book of Groß Sankt Martin, dates to c. 790, according to Deshusses, but could be earlier.Footnote 113

Equally, Bischoff admits uncertainty on the localization of the sacramentary. Given that no Gelasian of the eighth century can be located to England itself, and there is no firm evidence of knowledge of this type of sacramentary in England, our book is much more likely to have been made on the Continent than in the insular world.Footnote 114 A plausible scenario would see a Northumbrian master, who wrote the Münster fragment and created its finer initials, overseeing students on the Continent, who wrote and decorated the Cologne and Paris fragments, especially Cologne, Fragm. Nr. 24, whose initials bear less resemblance to any insular precedents. Bearing in mind that the Gelasian of the eighth century was principally distributed and copied by monasteries, the likelihood is that the book was copied at an Anglo-Saxon monastery. One possibility is in Cologne, and possibly at Groß Sankt Martin itself.Footnote 115 Although there is no surviving evidence from the Middle Ages, the historian Aegidius Gelenius (1595–1656) may have taken from local tradition his claim of an original foundation for Groß Sankt Martin in the eighth century by the Northumbrian missionaries, Otger (d. c. 713), Wiro (d. c. 753) and Plechelm (d. c. 732), with support of Pippin of Herstal (c. 635–714) and his wife Plechtrudis (d. c. 725).Footnote 116 Perhaps the sacramentary indicates there is a kernel of truth in this. Echternach would also remain a possibility, if the pieces of the sacramentary came to Groß Sankt Martin only later, as Paris, lat. 9433 has often been brought into connection with Echternach.

‘Benedictiones quas faciunt Galli’

There is, however, one final liturgical complication. This is the presence in the Münster fragment of a blessing as part of the mass for Candlemas, appearing at the end of the mass itself, after the post communion (Münster, Fragmente Kapsel, IV, 8 fol.1r), the text ‘Deus qui cunctae non capiant …’ (Fig. 7). A second blessing is found in Paris, lat. 9488, 3v appended to the mass for a martyr: ‘Benedicat uos dominus deus noster etipse in cor uestrum …’. Blessings of the people before communion like this were a feature of the ‘Gallican’ liturgy practiced in Merovingian France and North Italy.Footnote 117 The testimony of the Boniface sacramentary, which does not have one even at Christmas, suggests that they were not incorporated in the earliest English books. Boniface himself was unfamiliar with them when he encountered them on the Continent, and hence he seems to have asked Pope Zacharias I (Pope 741–752) if they were permissible, which the Pope denied in his surviving response of 751, which clearly describes them as a practice of the ‘Galli’.Footnote 118 But this did nothing to stem their popularity. The famous late Anglo-Saxon benedictionals show this type of blessing’s enthusiastic reception in England more than a century after our book, where they inspired composers to write several new ones, but they are originally continental imports to the island, probably during the ninth century.Footnote 119

Figure 7.

A folio containg parts of blessings of a house and of new fruits, in a Gelasian Sacramentary from Groß Sankt Martin. Cologne, Historisches Archiv, HAStK-RBA Best. 7050B (Fragmente B), 24, 4r (Anglo-Saxon centre on the Continent. (Cologne? Echternach?), s.viii2).

In all the other Gelasians of the eighth century, these blessings are either not given at all (as in Sankt Gall, cod. 348), or they appear in a separate collection, placed apart from the masses to which they belonged, a text type also known as a ‘benedictionale’, as in the sacramentaries of Gellone, Angoulême, or Phillipps.Footnote 120 The blessing used for Candlemas in the Münster fragment appears, for example, in the Sacramentary of Gellone’s blessing collection, used for the Nativity of Mary.Footnote 121 The one in Paris, lat. 9488, is only otherwise found in the second Freising Benedictional, a complete Gallican collection in an independent book.Footnote 122 The direct attachment of the blessings to the masses is such a rare trait, especially in the eighth century, that it offers a final and clear support that the Paris fragment also belonged to the same original manuscript as the Münster and Cologne fragments. Among varied forms of personalization of the Gelasians, it seems to be a manipulation of the tradition by the Anglo-Saxon scribes, likely in Germany, who gave us the Sacramentary of Groß Sankt Martin. This represents their enthusiastic adoption of a practice that was continental or ‘Gallican’, and a clear departure from what Boniface would have likely endorsed, given expressed papal scepticism.

The Gelasian of the Eighth Century among the Anglo-Saxons

Using our Paris fragment, Baumstark constructed a theory that the Gelasian of the eighth century originated in England itself, and that it was the book used by Boniface and brought by him to the Continent.Footnote 123 This ran into difficulties regarding the continental witness of the tradition and the lateness of surviving manuscripts, and it was never widely accepted. Equally, Baumstark wrote before the discovery of our Boniface Sacramentary, which showed that the initial and earliest waves of Anglo-Saxon missionaries at the time of Boniface possessed books that were very different from the Gelasian of the eighth century. It also seems perilous to build too much upon such small fragments, as, for example, Gamber did, when he also used the Paris fragment as proof that the Gelasian of the eighth century was already in England very early, which, as we have seen, is actually rather unlikely.Footnote 124

Nevertheless, the Groß Sankt Martin Sacramentary adds to other evidence that the Anglo-Saxons made some contribution to the spread of the Gelasian of the eighth century on the Continent. For example, they seem to have popularised it in Bavaria, particularly in the Main valley, at the end of the eighth century. However, this was in a second wave of liturgical influence that geographically and chronologically distinguishes itself from the Boniface Sacramentary’s presence further south, on the Danube in Regensburg. There are two other fragmentary Gelasians of the eighth century written by Anglo-Saxons or in insular hands. These are:

  1. 1. Würzburg, Universitätsbibliothek, M.p.th.f.42, 103r–v.Footnote 125 Written in Anglo-Saxon minuscule on the Continent in the late eighth century, and used to bind a manuscript in Würzburg Cathedral Library. Beautiful colourful initials at the opening of each mass, two of which survive.

  2. 2. München, Bayerische Staatsbibliothek, Clm 29300 (4 + Wertheim am Main, Staatsarchiv, Rosenbergisches Archiv Lehen- und Besitzrechte, Fragm. 1 (R-Lit. A Nr. 2101 Nr. 1) + St. Petersburg, Academy of Sciences, Institute of History West-European Section 3/625 + Würzburg, Staatsarchiv HV, Fragm. 1 + Würzburg, Universitätsbibliothek, M.p.th.f. 176.Footnote 126 The book was written in Anglo-Saxon minuscule on the Continent, dating to the end of the eighth century or beginning of the ninth. The provenance of the fragments is the monastery of Neustadt am Main, for whose accounts they were employed as binding material in the seventeenth century. The initials are simply drawn in black, but the O is often filled with whimsical faces. The uncial script used for the headings of prayers and masses distinguish the text from the other Anglo-Saxon books, and is rather continental in style. Liturgically, this has clearly developed away from the ‘S-Typus’, especially in its incorporation of an elaborate baptismal ordo into Lent.

Among the earliest copies of the Gelasian of the eighth century is a third fragmentary example that was written in Northern Italy, but also under a certain insular influence.

  1. 3. Budapest, Országos Széchényi Könyvtár, Cod. lat.med. aevi. 441.Footnote 127 4 folios, containing masses for Lent. Written in Italian uncial script, dating to the second half of the eighth century. Lehmann identified a certain insular influence on the script, in particular the use of the ‘quam’ abbreviation sign and the ‘per’ abbreviation.Footnote 128 Bobbio, Pavia or Verona have all been suggested.

Gelasian Origins in Light of the Groß Sankt Martin Sacramentary

Rejecting the theory of Flavigny, Moreton argued that the origin of the Gelasian of the eighth century would likely be a monastery in the Rhaetian Alps in the third quarter of the eighth century, because the manuscripts that have the least additional material added to the original core originate in that region, specifically, the two Sankt Gallen codices from Chur, cod. 348 and cod. 350.Footnote 129 The Anglo-Saxon fragments from Groß Sankt Martin are examples of the same type, and seem likely to be earlier than the Chur examples, but appear less ‘pure’ in their incorporation of the blessings, so they may nuance, rather than question, Moreton’s conclusions. If the Gelasian originated in Rhaetia or Alemannia c. 770–780, Anglo-Saxons must have taken it very quickly up the Rhine, and, within a few decades by the end of the century, they had brought it to Würzburg as well. Their mobility and networks must be part of the reason for its success.

But their involvement must have gone deeper. The comparison to the ‘Boniface Sacramentary’ offers the probability that some Anglo-Saxon material was used in the original creation of the Gelasian of the eighth century. The two prefaces for the first Sexagesima mass in the ‘Boniface Sacramentary’ (Bon 29 and 30 in Gamber’s edition), are found not in the Old Gelasian, nor in Veronense, but only later in manuscripts of the Gelasian of the eighth century, including in Sankt Gallen, cod. 348, for a Thursday in Lent and for a Sunday after Pentecost.Footnote 130 We can therefore be fairly certain these two prefaces were in the complete Groß Sankt Martin Sacramentary, which follows Sankt Gallen, cod. 348 so closely in all surviving content. But these appear as part of the independent content associated with the tradition of the Gelasian of the eighth century, and are of otherwise entirely unknown origin. The complex compilation of the Gelasian of the eighth century, more than a simple fusion of the Gregorian and Old Gelasian that Vogel and Bourque presented, must have had access to a version of, or similar sources to, the ‘Boniface Sacramentary’ as well.Footnote 131 In fact, the second of these two prefaces also appears in the Basel fragment, associated with Fulda, confirming that precisely the kind of texts that made up the Gelasian of the eighth century were circulating among Anglo-Saxons on the continent.Footnote 132 Such sources could also have transmitted some of the material the Gelasian of the eighth century has in common with the Old Gelasian, as Moreton made plain that the Vatican Old Gelasian was not used in this compliation process, but rather that both shared common sources. The workshop that made the archetype of the Gelasian would very likely have to be somewhere with Anglo-Saxon connections, and likely somewhere where Anglo-Saxons could work alongside continental partners in the compilation of a new form of mass book. This could indeed have been one of the famed monasteries of Alemannia, Sankt Gallen or Reichenau, and further research on surviving fragments from this region may help to pinpoint it.Footnote 133

Conclusion

Our two fragments, which certain palaeographical verdicts have presented as Northumbrian twins, have revealed themselves to be a far more distinctive duo, giving distinct evidence of the processes of exchange of liturgical texts over the Channel. Yet there is one thing the two manuscripts share, and this is their format. Both are large and handsome, and our estimation of the resources available to the Anglo-Saxons on the Continent should rise accordingly. They are also both written in two columns. Two-column sacramentaries are far from the norm, especially in France and Germany, where large square-shaped folios written in a single column predominate. But all the Anglo-Saxon fragments mentioned in this article are written in two columns, including those likely written in Würzburg, and excepting only the Basel and Budapest fragments, neither of which was written by an English scribe, but only under certain Anglo-Saxon influence. This layout was something to which Anglo-Saxon scribes held very firmly, no matter the type of sacramentary, and it would have distinguished sacramentaries written by the Anglo-Saxons from those written by nearly anyone else. There is one other place where mass books were regularly written in two columns, and this is Southern and Central Italy, in the later surviving witnesses, such as, for example, the Beneventan missal mentioned above.Footnote 134 We can therefore imagine that the Italian books which arrived in England before the eighth century, and brought with them the exotic saints of Naples and Capua, whose names remain in the Walderdorffer calendar, were also written in two columns, and that this old practice became normal in England, surviving stubbornly to the end of the eighth century even among English scribes on the Continent, until French practice and exemplars overwhelmed it during the course of the ninth century. Later Anglo-Saxon mass books all adopt French conventions, and are written in a single column.

The ‘Boniface Sacramentary’ was an old-fashioned type of book even when it was written. In almost every trait, in its organization with the term ‘orationes et praeces’, the title ‘CONIUNCTIO’ for the Communicantes interjection, its sixty day Lenten fast, its variable options of prayers for each mass, and in the choice of prayers it offers, particularly for Stephen and Christmas, it shows traits rarely seen in surviving liturgical books, and that suggest sources that likely pre-date the eighth-century date of the manuscript itself by at least a century. Only a few other tiny fragments may offer more of the same kind of text, but the freedom of assigning of mass prayers to individual feasts leaves open the possibility of significant internal variation between these Anglo-Saxon books. The Vatican Old Gelasian is certainly a relative, built on some of the same material, but differently organised, much more extensive in its number of mass sets, more restrictive in the options of prayers for each mass, and lacking the Berlin fragment’s most ancient traits. This witnesses the extensive processing that any Anglo-Saxon material present in the Old Gelasian underwent at Chelles, and also the tendencies that continental books were undergoing as we come towards the middle of the eighth century. Quite rightly, therefore, did the Old English martyrology, from the vantage point of the ninth century, call a tradition similar to the Boniface Sacramentary the ‘older sacramentaries’. But the Martyrology contrasted the older books not with the Frankish Gelasians of the eighth century, but with the Roman Gregorian sacramentaries as the ‘newer sacramentaries’.Footnote 135 The Old English martyrology therefore shows absolutely no knowledge of the Gelasian of the eighth century. The Gelasian of the eighth century was a continental phenomenon, and, according to our current knowledge, it seems never to have crossed the Channel into England itself.

Although the calendar of the Walderdorffer Fragment was updated on the Continent for perhaps a few decades, and the manuscript of the ‘Boniface Sacramentary’ seems to have survived in Regensburg until the sixteenth century, when it was unfortunately dismembered, there exists no known copy of its tradition made on the Continent, and its unique mass texts were not copied. A similar tradition has only left certain small traces in the books of Fulda and Lorsch.Footnote 136 The disappearance of this ‘Boniface’ tradition on the Continent is a sign of the success of the Gelasian of the eighth century, and, in its wake, of the Roman Gregorian Sacramentary, during the Carolingian period. There was a clear acculturation among the Anglo-Saxons on the Continent to the Gelasian of the eighth century, represented by the copying of the Groß Sankt Martin Sacramentary and the fragments from Bavaria.Footnote 137 They were therefore able to enter decisively into the vibrant liturgical conversation happening on the Continent. But that might have been made easier by the fact that all the manuscripts of the Gelasian types, both the Vatican Old Gelasian and the many copies of the Gelasian of the eighth century, show a certain English imprint, in different ways. This included the masses for Campanian saints which the English would have known in their ‘older sacramentaries’. But a number of individual prayer texts found in these books, like the prefaces of the Gelasian of the eighth century which we found in the Anglo-Saxon fragments, would also have been familiar, whether these were old Roman, Italian or even of English composition.

The Merovingian precedents for Carolingian liturgical achievements have been re-emphasized, and the old picture of a hopelessly decadent continental church saved by the infusion of rigor from the insular world is rightly discarded.Footnote 138 Nevertheless, the extremely decisive but still poorly understood transformation of the continental liturgy during the eighth century that saw a complete departure from the ‘Gallican’ forms of the mass, and the adoption of sacramentaries that were more ‘romanised’, if not always truly Roman, certainly drew on Anglo-Saxon precedents and sources, along with the initiative, resources and expertise of continental centres like Chelles and Rhaetia or Alemannia. Such transformations did not proceed, and could not have proceeded, simply from the edict of any monarch, or the importation of a single authoritative Roman exemplar, as the older scholarship had once assumed. They relied upon the mobility of clergy carrying books and libelli and sharing interesting new discoveries and compositions, on a continual collaboration of enthusiasts across cultures and traditions, on active and intricate work of comparison and compilation from diverse sources in engaged scriptoria, and, above all, on liturgical practitioners doing their best to make the liturgy comprehensible and meaningful in ways that made sense to them.

Acknowledgements

I would like to thank the editors, including Emily Thornberry, for their work on editing, my reviewers, as well as Jo Story and David Ganz for discussions on the fragments.

Appendix: an Edition of the Groß Sankt Martin Sacramentary

Münster, Landes- und Universitätsbibliothek, Fragmente Kapsel IV, 8

  1. 1 [fol.1r] …esse facias a fut(u)rum

  2. 2 (…) Deus qui cum te no(n) capiant caeli d(ig)natus es in templo utero uirginali include. Ut mater integra haberet a fructum de spiritu et incorruptionem de parte. Da hui(c) plebi angelum custodem qui filium mariae fidem concipiente praedixit. Sanctificet gregem tuum illa benediction que(m) sine semine humano redemptorem uirginis firmauit…

  3. 3 [fol.1v] …et de percepto pane iustitiae et de tuae domine festiuitate martyre agathe quia interuentionibus tibi placentium confidimus nobis ad perpetuam uitam profutura quae- sumpsimus.

  4. 4 (DOMINICA V POST THEOPHANIA) FAMILIAM tuam (quaesumus domine confide) custod(i ut quia in) sola spe gratiae celestis innititur tua semper protection munia(tur. Per.)

    [one column missing]

  5. 5 [fol.2r, col.1] …muniamur auxilio et magnifico profitiamur exemplo.

  6. 6 (SUPER OBLATA) Ad martyrum tuorum ualentini uitalis feliculae et zenonis domine festa uenientis cum muneribus nomine tuo dicatas offerimus ut illis reuerentiam differentes nobis ueniam impetremus.

  7. 7 P(OST COMMUNIONEM) PRotege domine plebeum tuam a festiuitatem martyrum tuorum ualentini uitalis et (feliculae et zeno)nis quam nobis [col.2] tradis assidue debita tibi persolui precibus concedes sanctorum.

  8. 8 (…) OMNIPOTENS sempiterne deus qui elegis infirma mundi ut fortia quaeque confundas. Dá nobis in festiuitate sanctae martyrae tuae iulianae congrua deuotione gaudere ut et potentiam tuam in eius passione laudamus et prouisium nobis percipiamus auxilium.

  9. 9 [fol.2v, col.1] IN sanctae martyrae tuae iulianae passionem praetiosa te domine mirabilem praedicantes munera uotiua deferimus praesta quaesumus ut sicut eius tibi gratia sint mereta sic nostrae seruitutis accepta reddantur officia.

  10. 10 POST COMMUNIONEM. Libantes domine mensae tuae beata mysteria quaesumus ut sanctae iulianae martyrae tuae interuentionibus temporalem et prae(sentem) nobis miseri(cordiam ferant. Per)

  11. 11 [col.2] (…) DEUS QUI beato (a)postolo tuo petro conlatis clauibus regni caelestis animas legandi atque soluendi pontificium tradedisti concede ut intercessionibus eius auxilio a peccatum nostrorum nexibus liberemus.

  12. 12 (…) Ecclesiae tuae quaesumus domine praeces et hostias beati petri apostoli commendet ora(tio ut quod pro) illius gloria…

Paris, Bibliothèque nationale de France, MS latin 9488, fols. 3–4

  1. 13 [fol. 3r] (…pecca)ti ut in te eticia [sic. etiam] exultemus in misericordia in quo ill letatur in gloria.

  2. 14 POST COMMUNIONEM. Sumpsimpus domine sancti .il. martyris solemnitate celestia sacramenta cuius suffragiis quaesumus largiaris ut quod temporaliter gerimus, eternis gaudiis consequamur.

  3. 15 SUPER POPULUM. Pleps tua domine sancti matyris (tui .il. te glorificante magnificet et eodem semper pre)cante te mereatur habere rectorem.

  4. 16 ALIAS ORACIONES. Sancti .il. martyris tui domine nos oracio sancta concilet. Que sacris uirtutibus uenerande refulge(t).

  5. 17 Beati martyris tui.il. nos quaesumus domine precibus adiuuemur eteius digna solemnia celebrantes tuo nomine fac nos semper esse deuotus. Per.

  6. 18 Beati martyris tui .il. nos (quaesumus domine patrociniis collatis non deserat, qui fragil)[fol.3v]itatem nostram et meritis tueatur et precibu

  7. 19 Benedicat uos dominus deus noster et ipse in cor uestrum influat ipse loquatur in uobis et ipse operetur in uobis. Ipse graciarum suarum ymbribus cordium uestrorum arida inriget uacua repleat inculta fecundet ATque isita intra uos (uirtutum semina, in fructus iustitiae per bonorum) operum incrementa perducat. Tribue domine intercedente beato martyro tuo .il. cupiditatum laqueos euitare presencium passionum certamina superare spiritalium nequiciarum tela contemnere. UT rore super eos sue benediccionis infusum, ita eis iugem tribuat in(crementum. Ut eos supplices liberes in futurum. Amen.)

  8. 20 [fol.4r] (…sa)cratissime uirginis martyrisque tuae .il. festiuitate laudare et benedicere debemus per christum domine nostrum, pro cuius castitates ardore ista et omnes sanctae uirginis a beata maria exemplum uirginitatis accipientes praesentis seculi uoluptatis omnes hac dilicia neglexerunt, ut ipsum filium tuum inuiolabilem sponsam cum ornatis lampadi(bus ei obuian)tes meruiussent abire. In cuius regni gloria cum coronis uirginitatis et palmas florentibus sicut sol sine fine fulgebunt. per.

  9. 21 POST COMMUNIONEM. Adiuuent nos quaesumus domine et hec misteria sanc[ta] que sumpsimus et beatae .il. intercessio ueneranda. per.

  10. 22 ALIA ORACCIO. Deus qui inter cete(ra) potenciae tuae miracul(a) etiam in fragil(i) sexu uictoria castitatis et martyris contulisti, da quaesumus ut (beate et sancte et uirginis) martyrisque tuae .il. adiuuemur meretis cuius beatitudinis inradiamur exemplis.

  11. 23 IN NATALE PLURIMORUM SANCTORUM. Presta domine quaesumus ut sicut sanctorum tuorum nos natalicia celebranda non deserunt, ita iugiter suffragiis comitentur. Per.

  12. 24 Sancti tui quaesumus domine iugiter nobis a te et ueniam postulent et perfectum. Per.

  13. 25 SUPER OBLATA. (Mu)nera ple(bis tuae) domine quaesumus beatorum sanctorum tuorum illorum fiant grata suffragiis et pro quorum triumphis tuo nomine offeruntur ipsorum digna perficiantur et meretis. Per.

  14. 26 UD Et te in tuorum honore sanctorum illorum glorificare qui et illis pro certaminis constanciam beatitudinem tribuisti sempiternam et infirmitati nostre talia prestetiti suf(fragia…)

Cologne, Historisches Archiv, Fragm. Nr. 124 and Fragm. Nr. 123

  1. 27 [Fragm. Nr. 124r, col. 1] sacrata deferim(mus ut dum) grati per(ce)p(tis existem)us (effi)ciamur percip(iendibus) fructibus g(ratiores. Per)

  2. 28 (Ut nos domine tribuis solem)ne tibi deferre [Fragm. Nr. 123r col. 1] ieiunium sic nobis (quaesumus indulgentiae praesta subsidium. per)

  3. 29 (…) DEUS CUIus ad(oranda poten)tia maiestatis flammae seruientis incendium sanctis tribus pueris in splendore (demutatum) est anima(rum) ecclesiae tuae (…) [Fragm. Nr. 124r, col. 2] (…propitiatione caelesti popu)lus tu(us ereptus) exultet. Per.

  4. 30 Haec (hostia domine quaesumus) ut uincula nos(trae) iniqu(itatis absoluunt et tuae nobis mise)ricordiae dona con[Fragm. Nr. 123r, col. 2]ciliet. Per dominum nostrum.

  5. 31 (UD et ti)bi (sanctificare) ieiunium qu(od) nos (ad) aedficati(onem anima)rum et (casti)gationem corporum (ser)uare docuisti. Quia strictis (cor)poribus (anim)ae sag(in)atur (et) in quo ex(terior) ho(mo) noster affligitur (…) [Fragm. Nr. 124v, col. 1](…misericor)diar(um tuarum) quas peccatoribus pie semper (…ut non solum) a (cibis) sed a peccatis omnibus menti abstibus deuotionis tibi ieuinio placeamus. Per (christum)

  6. 32 [Fragm. Nr. 123v col. 1] Perficiant in nobis domine quaesumus (tua) sa(cramenta quod continent…) g(eri)mus rerum (uer)itate campiamus. Per.

  7. 33 (Suscipe domine) preces (popu)li suppli(cantis) et nostra (uota ieiunii)[Fragm. Nr. 124v, col. 2] …perfice (sacramentum).

  8. 34 (DOMINICA UACAT) Omnipotens sempiterne deus misericordiae (tuam) ostende (supplicibus) precibus ut qui (de meritorum qualitate) diffidimus (non iu)dicium t(uum) sed indulgentium (senti)amus. PER.

  9. 35 [Fragm. Nr. 123v, col. 2] Huius te domine muneribus (placemus) obla(tione) et perpetuae (uitae par)ticipes huius (o)per(a)tione reddamur. Per

  10. 36 UD deus. Qui uicit diabolum (ut) paradysum restituit (ut uitae) ianuas p(atefecit)…

Cologne, Historisches Archiv, Fragm. Nr. 24

  1. 37 (fol. 1r, col. 1] …ge(neris ordina)sti be(nignus) ad(siste ut) quod te (auctore iun)gi(ter te auxi)ilian(te seruetur) per.

  2. 38 Q(uaesumus omni)p(oten)s (deus institu)ta (prouidentiae tuae) pio fa(uo)re comi(tare) et quos l(egitimam…). Per.

  3. 39 A(desto domine…) famularum tuarum illarum quam (tibi) offer(ant) p(ro famula tua) [col. 2] ill(a…) adsume ut qu(od tua) dispositione expeditur tu(a) grat(ia con)pletur. Per.

  4. 40 UD (ut quod nu)ptiaru(m) blando (concordie) iug(o) et (insolubili) pacis (uinculo) ne(xisti ut multi)plicandis ad(optionem filii) sancto(rum conu)bi(o)rum f(ecunditas) p(udica) ser(ui)r(et tua enim domine) p(rouidentia) [fol.1v, col.1] (t)ua gratia inef(f)abilis modis (u)trumque dis(pensat) ut quod (g)eneratio ad mundi (et) det (orna)tum (re)genera(tio) ad ae(clesi)ae perducat (a)ugmentum. per christum.

  5. 41 (…Hanc i)gitur oblatio famularum tuarum illarum quam (t)ibi offerunt. pro famula tua i(l)la quaesumus domine placatus accipias (p)ro qua maies(t)atem sup(p)lices exoramus (ut sicut eam) ad e(t)atem nup(tiis co)ngruentum (peruenire tribuisti sic) eam [col.2] consortio maritali (tuo) munere (co)polatam desiderata sobole gaudere perficias adque oblatam seriam cum suo coniuge probeas benignus annorum. Per.

  6. 42 INFRA (ACTIONEM…) H(anc igitur) oblationem famulorum tuorum illorum a(dque illarum) quam (tibi) offerunt ob (d)iem trecisimum con(iuncti)onis su(ae uel annu)alem qu(o die eos iu)gili (uinculo sotiare diga)tus es pla(catus) sus(ci)pias dep(raeca)mur…

  7. 43 [fol. 2r, col. 1] …serenitatem (nobis) tribue supplicant(es ut qui) pro p(eccat)is nost(ris iuste ad)fli(gimur) misericordia(e tuae) pre(ueniente) clemen(tiam sentia)mus. Per.

  8. 44 DEUS (qui omnium) rerum (tibi ser)uie(ntum na)tura p(er ipsos motus) ae(ris ad cultum) tu(ae maiestatis) ins(titues) tranquili(tatem) nobis misericordiae tuae remotis largire terroribus (ut cu)ius iram (e)xpa(ui)mus cle(men)t(iam sentiamus) per dominum nostrum.

  9. 45 [col. 2] P(raueniat nos quaesumus) domine g(ratia tua sem)per et (subsequatur) et (has oblationes) quas pr(o peccatis…)ad(sume ut per intercessionem sanctorum) tu(orum cunctis) nobis (proficient ad salutem. Per)

  10. 46 Ud (per christum dominum nostrum….) contremiscit qui pertingit a fine usque (ad finem fortiter) et dispon(it)[fol. 2v, col. 1](…su)pp(lices te rogamus…) ator (…) pro(pitius ut o)mnia (que imb)rium (densitate) iac(tantur…)[col. 2] (subleuent)ur (atque) pro eccle(siae tuae unitat)em (iubeas) amouere (de)curus. Per quem.

  11. 47 (…) P(leb)s tua domine (cap)iat sacr(e benedictionis…) depr(cationibus (adiuuatur). Per.

  12. 48 (…o)mnipotens sempipiter(ne deus par)ce (metuentibus…n)oxios…

  13. 49 [fol. 3r, col. 1] …defen(sione tua fidimus) null(is arma timiamus…)

  14. 50 DEUS q(ui misericoride tuae…)

  15. 51 [fol. 3v col. 1] (…su)is (uictum indigentib)us sub(ministrat Quatenus haec deuotio…) semp(er accepta) per (dominum nostrum)

  16. 52 [fol. 3v, col. 2] (…indi)gen(tibus) administrat (ut dominus caelestis sua misericordi)a (terrenam elemosinam) con(piset et spiritales) diui(tias largiatur. Tribuat) ei mag(na pro paruis) pro terr(enis caelestia) pro tempo(ralibus sempiterna) per (dominum)

  17. 53 ORA(TIONES AD MISSA) D(eus qui post baptismi sacramentum secundum a)bo(litionis peccatorum) ely(mosinis indidisti) Respice p(ropitius su)per (famul)um (tu)um (illum cuius operis tibi gracie referuntur fac eum praemio) beatum (quem fecisti pietate deuo)tum. per dominum (nostrum)

  18. 54 Deus (qui homini ad tuam imaginem facto) etiam spi(iritalem alimonium) prep(erasti. concede filio) nostro f(amulo tuo illo) qui in paup(eres tuos tuo seminat) do(na…)

  19. 55 [fol. 4r, col. 1] …domus suae placatus suscipias deprecamur. pro quo in hac abitatione auxilium tue magestatis deposco ut mittere ei digneris angelum tuum sanctum ad custodiendos omnes in hac habitatione consistentibus. Per Christum dominum nostrum.

  20. 56 (…) Omnipotens sempiterne deus qui (f)acis mirabilia magna (s)olus pretende super has famulus elegentes in hac domo spiritum gratiae salutaris et ut conplaceant tibi deus in ueritate tua perpetuum eis rore tue benediccionis effunde. Per dominum nostrum

  21. 57 (…) TE Domine sancta pater omnipotens aeterne deus supplices deprecamur ut misericordiam tuam iugiter nobis conced(as) suffitienter mensium (cupias) et fructum omnium (largiter tribuas uineam) [col. 2] quoque substantiam abu(n)dantem arborum fetus prouentus omnium rerum adque ab his omnibus pestatis uniuersas procellas grandinis amouere digneris. Per.

  22. 58 (…) Misericordiam pietat(is) tue supplices deprecamus omnipotens eterne deus ut oblationes populi tui quas tibi de suis primicii(s) offerunt benignignus [sic] suscipere digneris trib(ue) eis domine in hoc seculo habundantiam tritici uini et olei in future autem uitam eternam comemo(ratio)nem quoque facientibus nobis beatissimorum martyrum et confessorum uen(iam pec)catorum largire dignare. Per.

  23. 59 (…) Oremus pietatem tuam omnipotens deus ut has pri(mitias creature tuae)[fol. 4v col. 1] quas aeris et pl(uuiam tempera)to nutrire dign(ati sun)t benedicci(oni)s tuae (i)mbre (per)fundas tribuasque populo tuo de tuis (mun)eribus tibi semper gratias agere ut a fertilitate terris esurientum (ani)mas bonis affluentibus repleas et egenus et pauper laudent nomine glorie tue. Per.

  24. 60 Domine sancte (pa)ter omnipotens eterne deus qui celum et terram mare et omnia creasti te supplices quaesumus (ut hunc) nouum fruc(tum benedicere) et sanctificare digneris et (m)ultiplicare (habun)danter offerentium tibi (ut) repleas eorum cellaria (cum for)titudine fruc(menti et uini) ut laetantes in eis referant tibi deo. omnipotenti laudis et gratias. Per.

  25. 61 (…) B(en)edic d(omine h)o(s) fructos (nouos uue siue fabe) quos tu domine per r(orem cael)i et (inun)dantia (pluuiarum) et tempora serena atque tranquilla ad matu(utira)tem perducere dignatus es ad percipiendum nobis cum gratia(rum) accione in nomine domini iesu Christi. per quem haec omnia domine (semper bona creas).

  26. 62 (…) T(e deprecamur o)mnipotens (aeterne deus) ut benedicas (h)unc (fructum nouum) pomorum (ut) qui esu inter(dicte arbo)ris laetalis p(oma in) prot(oparente…su)mus per in(lustrationem) unici filii tu(i redemptioris) dei ac (domini) nostri iesu (christI) bene(dictionem sanctificata) omnia adque benedic(ta de)pulsis adque (abiectis uet)ust(i host)tis adque (primi facinoris) incent(oris insidiis) salu(bri)ter ex huius diei ann(iuersari)a so(lempnitate de uniuersis…).

1 Sg 188, Aug 214, Gel 199, Eng 207; 2 Gel 2077; 3 Sg 192, Aug 219, Gel 205, Eng 212; 4 Sg 193, Aug 220, Eng 213, Eng 231; 5 Sg 211, Aug 238, Gel 225, Eng 231; 6 Sg 212, Aug 239, Gel 227, Eng 232; 7 Sg 213, Aug 240, Gel 229, Eng 233; 8 Sg 214, Aug 241, Gel 230, Eng 234; 9 Sg 215, Aug 242, Gel 231, Eng 235; 10 Sg 216, Aug 243, Gel 233, Eng 237; 11 Sg 217, Aug 244, Gel 234, Eng 238; 12 Sg 218, Aug 245, Gel 235, Eng 239; 13 Sg 1469, Gel 1773; 14 Sg 1470, Gel 1774; 15 Sg 1471, Gel 1775; 16 Sg 1472, Gel 1777; 17 Sg 1473, Gel 1778; 18 Sg 1474, Gel 1779; 19 Freising II 395-400; 20 Sg 1486, Gel 1836; 21 Sg 1487, Gel 1837; 22 Sg 1488, Gel 1833; 23 Sg 1489, Gel 1789; 24 Sg 1490, Gel 1790; 25 Sg 1491, Gel 1791; 26 Sg 1492, Gel 1792; 27 Sg 1226, Aug 894, Gel 1501, Eng 1371; 28 Sg 1227, Aug 895, Gel 1502, Eng 1372; 29 Sg 1228, Aug 896, Gel 1503, Eng 1373; 30 Sg 1229, Aug 897, Gel 1504, Eng 1374; 31 Sg 1230, Aug 898, Eng 1375; 32 Sg 1231, Aug 899, Gel 1506, Eng 1376; 33 Sg 1232, Aug 900, Gel 1507, Eng 1377; 34 Sg 1233, Aug 901, Gel 1508, Eng 1378; 35 Sg 1235, Aug 903, Gel 1510, Eng 1380; 36 Sg 1236, Aug 904, Gel 1511, Eng 1381; 37 Aug 1644, Gel 2629; 38 Aug 1645, Gel 2630; 39 Aug 1646, Gel 2631; 40 Aug 1647 Gel 2632; 41 Aug 1648; 42 Aug 1649; Gel 2629; 43 Aug 1683, Gel 2669; 44 Aug 1684; Gel 2670; 45 Aug 1685, Gel 2671; 46 Aug 1686, Gel 2672; 47 Aug 1687, Gel 2673; 48 Aug 1688, Gel 2674; 49 Aug 1791, Gel 2767, San 164 50 Aug 1792, Gel 2768, San 165 51 Aug 1799, Gel 2781, San 172; 52 Aug 1800, Gel 2782, San 173; 53 Aug 1801, Gel 2784, San 174; 54 Aug 1802, Gel 2785, San 175; 55 Aug 1851, Gel 2825, San 215: 56 Aug 1852, Gel 2826, San 216; 57 Aug 1854, Gel 2828; 58 Aug 1855, Gel 2829; 59 Aug 1856, Gel 2831; 60 Aug 1857; Gel 2832; 61 Aug 1858, Gel 2833; 62 Aug 1859, Gel 2834

References

1 In addition to standard abbreviations, the article uses additional abbreviations within the conventions of liturgical scholarship. The numbers following apply not to pages, but to individual formulae of the edition. They are also used in the edition in the appendix. Bon = Das Bonifatius-Sakramentar und weitere frühere Liturgiebücher aus Regensburg, ed. K. Gamber (Regensburg, 1975); Sg = Das fränkische Sacramentarum Gelasianum in alamannischer Überlieferung, ed. by L. K. Mohlberg (Münster, 1918); Gel = Liber sacramentorum gellonensis, ed. A. Dumas, 2 vols., CSL 159–159A (Turnhout, 1981); Aug = Liber sacramentorum augustodunensis, ed. O. Heiming, CSL 159B (Turnhout, 1984); Eng = Liber sacramentorum engolismensis, ed. P. Saint-Roch, CSL 159C (Turnhout, 1987); San = Ein Sankt Galler Sakramentar Fragment, ed. G. Manz (Münster, 1939); GeV = Liber sacramentorum romanae aeclesiae ordinis anni circuli, ed. L. Mohlberg (Rome, 1960); CLLA = K. Gamber, Codices liturgici latini antiquiores, 2 vols., 2nd ed. (Freiburg, 1968) + Supplementum. Ergänzungs- und Registerband, with B. Baroffio, F. Dell’Oro, A. Hänggi, J. Janini and A. M. Triacca (Freiburg, 1988). CP = Corpus Praefationum, ed. E. Moeller, 5 vols., CCSL 161–161D (Turnhout, 1980); CBP = Corpus Benedictionum Pontificalium, ed. Edmond Moeller, 4 vols., CCSL 162–162C (Turnhout, 1971–1979).

2 Among the vast bibliography on Boniface, recent treatments are found in: A Companion to Boniface, ed. M. Aaij and S. Godlove (Cambridge 2020); L. von Padberg, Bonifatius. Missionar und Reformer (Munich, 2003); Bonifatius. Vom Angelsächsischen Missionar zum Apostel der Deutschen, ed. M. Imhof and G. Stasch (Petersberg, 2004); Otherwise, on the missions, see I. N. Wood, ‘The Missionary Life’, The Cult of Saints in Late Antiquity and the Early Middle Ages, ed. J. Howard-Johnston and P. A. Hayward (Oxford, 1999), pp. 167–93; R. McKitterick, Anglo-Saxon Missionaries in Germany: Personal Connections and Local Influences, Vaughan Paper 36 (Leicester, 1991), repr. as article I in her Frankish Kings and Culture in the Early Middle Ages (Aldershot, 1995); J. Palmer, Anglo-Saxons in a Frankish World, 690–900, Stud. in the Early Middle Ages 19 (Turnhout, 2009).

3 For new turns in liturgical studies (but not sacramentaries), Understanding Medieval Liturgy, ed. H. Gittos and S. Hamilton (Ashgate, 2016); neither of our fragments is fully discussed in R. Pfaff, The Liturgy in Medieval England: a History (Cambridge, 2009).

4 H. Gneuss and M. Lapidge, Anglo-Saxon Manuscripts: a Bibliographical Handlist of Manuscripts and Manuscript Fragments Written or Owned in England up to 1100 (Toronto, 2014) [hereafter Gn-L, ASMss], 791.

5 H. Walderdorff, Regensburg in seiner Vergangenhert und Gegenwart (Regensburg, 1896) pp. 20–1, 297, n. 1. The fragment had, according to him served as the cover of an act in the Regensburg Archive ‘ein ursprünglich aus dem bischöflichen oder Domkapitelischen Archive herrührender Aktendeckel überzogen’.

6 MGH Necrologia Germaniae III: Dioceses Brixinensis Frisingensis Ratisbonensis, ed. L. Baumann (Berlin, 1905), 369.

7 Sotheby’s Western Manuscripts and Miniatures (London, 4 December 2007), no. 44.

8 Description and photos at: https://www.sothebys.com/en/buy/auction/2024/important-medieval-manuscripts-from-the-collection-of-the-late-ernst-boehlen/sacramentary-of-st-boniface-a-fragmentary-bifolium. The fragment was also offered, again unsuccesfully, on 10 July 2025 (https://www.sothebys.com/en/buy/auction/2025/books-manuscripts-and-music-from-medieval-to-modern-l25404/fragment-c8th-sacramentary-of-st-boniface-in-latin).

9 K. Gamber, ‘Das Regensburger Fragment eines Bonifatius-Sakramentars. Eine neuer Zeuge des vorgregorianischen Meßkanons’, RB 85 (1975), 266–302. B. Bischoff and V. Brown, ‘Addenda to Codices Latini Antiquiores (I)’, MS 47 (1985), 320–66 at 357. The fragment has been digitised: https://nbn-resolving.org/urn:nbn:de:bvb:re5-fragment.iii.1.1-6.

10 P. Siffrin, ‘Zwei Blätter eines Sakramentars in irischer Schrift’, Jahrbuch der Liturgiewissenschaft 10 (1930), 1–39 at 1–2. The text is also edited in Missale Francorum, ed. L. K. Mohlberg (Rome, 1957), pp. 71–9.

11 P. Siffrin, ‘Das Walderdorffer Kalendarfragment saec.VIII und die Berliner Blätter eines Sakramentars aus Regensburg’, Ephemerides Liturgicae (1933), 201–24. The connection had apparently been first made by Walderdorff himself, in unpublished remarks; for Boniface’s activity in Bavaria: L. Good, ‘Boniface in Bavaria’, A Companion to Boniface, ed. Aaij and Godlove, pp. 299–326.

12 Gamber, Das Bonifatius-Sakramentar.

13 D. Bullough, ‘Alcuin and the Kingdom of Heaven’, Carolingian Renewal: Sources and Heritage (Manchester, 1991), pp. 161–240 at 168 denies any connection to Boniface and points out that Gamber rather unfortunately placed Boniface’s cloister of Nursling in Northumbria, when it was, of course, in Wessex, see Gamber, ‘Das Regensburger Fragment’, p. 267. Nevertheless, Boniface and his fellow scholar from Wessex, Lull of Mainz (c. 710–786), did ask for, and receive, books from Northumbria, see W. Levison, England and the Continent in the Eighth Century (Oxford, 1946), pp. 139–40.

14 C. Hohler, ‘The Type of Sacramentary used by St Boniface’, Sankt Bonifatius. Gedenkgabe zum zwölfhundertsten Todestag (Fulda, 1954), pp. 89–93 argued without reference to these or any other early English fragments that Boniface used the type on which the ‘Sarum’ missals are based, which he terms the ‘Family of St. Amand’, but does not go into the necessary depth to prove this. A continental origin of the ‘family’ he identified and more than a century after Boniface, in line with the manuscript evidence, is clear, see A. Westwell, The Carolingian Sacramentaries of Saint-Amand: Art, Script and Liturgical Creativity in an Early Medieval Monastery (Berlin, 2024), pp. 363–7; While the suggestion of Y. Hen, ‘Rome, Anglo-Saxon England and the Formation of the Frankish Liturgy’, RB 112 (2002), 301–22, at 316, n.78, that Wilibrord and Boniface were each connected to distinct liturgical traditions is interesting, the assertion that Boniface would have used a Gallican sacramentary like the Missale Gothicum remains very difficult to substantiate. Boniface himself only ever quoted Old Gelasian prayers (see below, p. 18), never Gallican ones, nor is there any known presence of Gallican prayers in Fulda, Würzburg or Mainz, or even a single recognisably Gallican prayer copied in an Anglo-Saxon script that would prove that he or his circle had used them. Thus, there remains no manuscript evidence of a Gallican rite ever being practiced in England or by English clergy, or any indirect discussion of the liturgy by the Anglo-Saxons that would support it either. Notably, a possible Northumbrian origin posed by Hen for the Gallican palimpsest (München, Bayerische Staatsbibliothek, Clm 14429) in an entirely Irish script remains speculative. Even if there are some Northumbrian influences in the decoration of this palimpsest, that does not rule out Northumbrian influence on an Irish production, let alone a possible continental origin in an Irish centre. Even though Hen (at 312, n.5) cites B. Bischoff, Die südostdeutschen Schreibschulen und Bibliotheken der Karolingerzeit, I (Wiesbaden, 1940; repr. 1990), 243–4 and II (Wiesbaden, 1980), 242–3, in which no Northumbrian origin is actually suggested, Bischoff was against an origin of the palimpsest in England, as is clear in the edition of the palimpsest by Das irische Palimpsestsakramentar im Clm 14429 der Staatsbibliothek München, ed. A. Dold and L. Eizenhöfer (Beuron, 1964), pp. 30*–31*: ‘Für die Herkunft kann m.E. nur Irland in Frage kommen’. At Hen, ‘Rome, Anglo-Saxon England’, p. 315, the presence of the earliest masses for Saints Matthew and Luke in the Irish palimpsest is indeed interesting, but, given the sparsity of any surviving Gallican witnesses comparable to the Irish palimpsest, remains difficult to interpret as indisputable sign of English influence, especially as the Irish palimpsest contains no English saints, nor any of the roster of Italian saints we know were widely celebrated in England (see below), nor do we know anything near enough about Irish liturgy to rule out that Irish liturgical interest in the same saints explains this. The apostles were being increasingly commemorated across the Continent too, see E. Rose, Ritual Memory: the Apocryphal Acts and Liturgical Commemoration in the Early Medieval West (Leiden, 2009). The idea that the preponderance of Roman saints in the Missale Gothicum (Hen, ‘Rome, Anglo-Saxon England’, pp. 318–21) should come from Anglo-Saxon influence is again difficult to substantiate, and Merovingian interest in Rome was certainly sufficient to explain them. The diversity of the Merovingian liturgy means that processes of Romanization could easily have proceeded at different paces in different centres, so that the Missale Gothicum could well be a precocious forerunner for future trends. Furthermore, even if the celebration of such Roman saints may have been imported into Merovingian Francia, the mass sets composed for them in the Missale Gothicum are still Gallican (and, thus, in our current state of knowledge, not English), including the mass prayers like the collectio, post nomina, ad pacem and contestatio or immolatio, terminology also never found in Anglo-Saxon evidence. Thus, I fully accept that the Anglo-Saxons played a role in the importing of a Roman(ised) liturgy into Francia, in the form of the Old Gelasian, as the Boniface Sacramentary strongly supports. But until we find a Gallican mass in Anglo-Saxon script, or even one cited by an Anglo-Saxon, it is preferable to assume the Gallican liturgy did not cross the Channel in the other direction in any sustained way, or was ever used by Boniface. Hen’s argument is reiterated in Y. Hen, ‘Missionaries and Liturgy’, Bonifatius. Leben und Nachwirken. Die Gestaltung der christlichen Europa im Frühmittelalter, ed. F. Felten, J. Jarnut and L. von Padberg (Mainz, 2007), pp. 341–52 at 351–2. Further on the Missale Gothicum, see M. Smyth, La liturgie oubliée. La prière eucharistique en Gaule antique et dans l’Occident non romain (Paris, 2003), pp. 71–81, who underlines additional Irish (in his words Celtic), i.e. not Anglo-Saxon, influences on Gothicum, at 79–81, which would bear on this question. This would rather suggest an Irish Gallican liturgy, exactly as in the Munich Palimpsest, was used as a source for the Missale Gothicum, rather than a hypothetical Anglo-Saxon one. Indeed, the Irish peregrini would be the perfect intermediaries, able to bring together possible Anglo-Saxon and Gallican liturgical currents.

15 Gamber, Das Bonifatius-Sakramentar, pp. 45–6.

16 Nb. This is corrected from the misprint in Bischoff’s text, which mistakenly uses minuscule.

17 B. Bischoff, Die südostdeutschen Schreibschulen I, 183–4: ‘Kalbpergament, in insularer Weise nach dem Falten liniiert … Die Schrift ist schöne, zuchtvolle angelsächsische Minuskel [sic.] northumbrischen Typs in der schlanken und gedrängten Form, die zur Minuskel überleitet … Sie gebraucht d und unziales d; die Minuskelformen von n und r, die Majuskelform von s überwiegen … Die roten Rubriken in der gleichen Schriftart. Initialen im angelsächsischen Stil, rot umpunktet. Die Schrift und die Art des Materials und dessen Zubereitung sind echt englisch’. Bischoff offers as comparison CLA II 138 (fragmentary gospels in Cambridge, University Library, MS Kk.1.24 and the British Library) and 194b (London, British Library, Egerton MS 1046), both judged as Northumbrian (Gn-L, ASMss 21 and 410).

18 See the notice on our fragment R. Gameson, ‘Northumbrian Books in the Seventh and Eighth Centuries’, The Lindisfarne Gospels: New Perspectives, ed. R. Gameson (Leiden, 2017), 43–83 at 76.

19 Siffrin, ‘Zwei Blätter’, pp. 4–5.

20 The addition of Theobald’s name was compared by Bischoff, Die südostdeutschen Schreibschulen I, 184 with the Anglo-Saxon hand in two manuscripts from Sankt Emmeram: München, Bayerische Staatsbibliothek, Clm 14080 and Clm 14653, both dated to the second half of the eighth century.

21 This does not ipso facto link the entry to the reburial of the relics of Emmeram in the monastery by Bishop Gaubald, as Gamber, ‘Das Regensburg Fragment’, p. 267 suggested.

22 Missale Francorum, ed. Mohlberg, pp. 72–3; for problems with CLA on this question, see also R. McKitterick, ‘Anglo-Saxon Missionaries in Germany: Reflections on the Manuscript Evidence’, Trans. of the Cambridge Bibliographical Soc. 9 (1989), 291–329.

23 Conclusions of Siffrin and Gamber are summarised in C. Cubitt, Anglo-Saxon Church Councils c. 650–680 (London, 1995), pp. 137–40; on the later transformations, see Pfaff, The Liturgy in Medieval England, pp. 88–90, 110–18, 124–6; R. Pfaff, ‘Massbooks: Sacramentaries and Missals’, The Liturgical Books of Anglo-Saxon England, ed. R. Pfaff (Kalamazoo, MI, 1995), pp. 7–34 at 11 ff.; Hohler, ‘Some Service Books’; D. Dumville, Liturgy and the Ecclesiastical History of Late Anglo-Saxon England. Four Studies (Woodbridge, 1992). Later Anglo-Saxon books contain masses for local saints that often seem to be very old, but these are clearly the local additions to books that are entirely continental in their organization, and in much of their content. They are, in the lion’s share, the kind of Gregorian Sacramentary with Carolingian Supplements that dominates France as well.

24 Also edited in Missale Francorum, ed. L. C. Mohlberg (Rome, 1957), pp. 79–85.

25 Essentially, the Host is broken (‘Sacramenta confracta’), before the saying of the Lord’s prayer, which Gregory the Great (d. 604) decreed should come after it, see Gamber, ‘Das Regensburger Fragment’, p. 293. and elsewhere for variants in common with the Stowe Missal.

26 For the Sankt Paul im Lavanttal fragment, see CLLA, Supplementum 419*; CLA X 1459; Gn-L, ASMss 933.5, edited at Gamber, Das Bonifatius-Sakramentar, pp. 77–9.

27 Cosmas and Damian (27 September), Archangel Michael (29 September), the octave of Peter and Paul (6 July) Assumption of Mary (25 July), Sixtus (6 August), Laurence, with a vigil (here ‘ieiunio’, also the Old Gelasian terminology) (9–10 August), Hippolytus (13 August), Nativity of Mary (16 August), Cyprian (14 September). For the list of saints, see further K. Gamber, ‘Das kampanische Messbuch als Vorläufer des Gelasianum’, Sacris Erudiri 12 (1961), 5–111, at 43–51.

28 Priscus of Capua (1 September), Marcellus and Apuleius of Capua (7 October), the Maccabees 1 August), Donatus of Arezzo (7 August), Magnus of Fabrateria (Anagni) (19 August), Rufus of Capua (27 August). Also the Passion of John the Baptist (29 August), celebrated in the Gelasian, and in Central Italy (see Die ältesten Kalendarien aus Monte Cassino, ed. E. A. Lowe (Munich, 1908), pp. 26–7), but not part of Roman mass books.

29 Quintus of Capua (5 September), Sinotus of Capua (7 September), Januarius of Naples (19 September), Sosius of Naples (23 September) and Lupulus of Capua (15 October).

30 Siffrin, ‘Das Walderdorffer Fragment’, pp. 203–12; The Calendar of St. Willibrord from MS. Paris lat. 10837, ed. H. Wilson (London, 1918) at 9–12, CLA V, 606a; CLLA 414; Gn-L, ASMss 897; Y. Hen, Culture and Religion in Merovingian Gaul, AD 481–751 (Leiden, 1999), pp. 103–6; F. Lifschitz, The Name of the Saint: the Martyrology of Jerome and Access to the Sacred in Francia, 627–827 (Notre Dame, IN, 2006). However, two characteristic Gelasian saints, Rufus and Magnus, are not in Willibrord’s calendar.

31 Not merely the Gelasian Magnus, Rufus, and Priscus, but also Januarius and Sosius of Naples, and Quintus, Sinotus and Lupulus of Capua were found in these old sacramentaries (‘ealdan sacramentorium’), whose masses survive in no extant book. See Das altenglische Martyrologium, ed. G. Kotzor, 2 vols. (Munich, 1981) II, 196, 200, 201, 211, 215, 229. As June is lost from the Walderdorffer fragment, we cannot see if the Boniface Sacramentary also had a second feast of Priscus of Capua on 1 June, possibly a dedication of the now destroyed church in Capua, and Nicander of Venafro (17 June), both known in the martyrology (Das altenglische Martyrologium II, 112, 122), but the analogy to the Willibrord calendar would suggest they were in a lost folio and also provided with masses in the original sacramentary.

32 J. Chapman, ‘The Capuan Mass-Books of Northumbria’, Notes on the Early History of the Vulgate Gospels (Oxford, 1908), pp. 144–61.

33 Gamber, Das Bonifatius-Sakramentar, p. 65; Siffrin, ‘Zwei Blätter eines Sakramentars’, p. 31.

34 Siffrin, ‘Zwei Blätter eines Sakramentars’, pp. 19–29.

35 Ibid. pp. 23–4.

36 On this development see A. Bouley, From Freedom to Formula: the Evolution of the Eucharistic Prayer from Oral Improvisation to Written Texts (Washington, DC, 1981), pp. 209–15.

37 Gamber, ‘Das kampanische Meßbuch als Vorläufer’, pp. 75–6. The nine are GeV 91, 111, 88, 63, 1223, 164, 126, 165, 170 (mostly scattered through Lent).

38 CLLA 610; Liber sacramentorum romanae aeclesiae ordinis anni circuli, ed. L. Mohlberg (Rome, 1960); Chelles is likely, as in Bernhard Bischoff’s ‘Die Kölner Nonnenhandschriften und das Scriptorium von Chelles’, Mittelalterlichen Studien. Ausgewählte Aufsätze zur Schriftkunde und Literaturgeschichte, 1 (Stuttgart, 1996), 395–411; A facsimile is Sacramentarium Gelasianum e codice Vaticano Reginensi latino 316 (Vatican City, 1975).

39 A. Chavasse, Le sacramentaire gélasien (Paris, 1958); C. Coebergh, ‘Le sacramentaire gélasien ancien. Une compilation des clercs romanisants du VIIe siècle’, Archiv für Liturgiewissenschaft 7.1 (1961), 45–88; J. D. Thompson, ‘The Contribution of Vaticanus Reginensis 316 to the History of Western Service Books’, Studia Patristica 13 (1975), 425–9.

40 As in the Old English martyrology, Das altenglische Martyrologium I, 258*–259*.

41 CLA II 176; CLLA 411; Gn-L, ASMss 292. Edited by A. Baumstark, ‘Ein altgelasianisches Sakramentarbruchstück insularer Herkunft’, Jahrbuch für Liturgiewissenschaft 7 (1927), 130–6. Important intervention in H. Frank, ‘Die Briefe des Heiligen Bonifatius und das von ihm benutzte Sakramentar’, Sankt Bonifatius. Gedenkgabe zum zwölfhundertsten Todestag (Fulda, 1954), pp. 58–88, at 75, n. 69.

42 Bullough, ‘Alcuin and the Kingdom of Heaven’, p. 212: ‘it seems to me perverse to doubt that the sacramentaries used in English churches in the late seventh and for much of the eighth century were of a type close to although by no means identical with the ‘Gelasian’ sacramentary in Vat.Reg.lat.316’.

43 C. Hohler, ‘Theodore and the Liturgy’, Archbishop Theodore. Commemorative Studies on his Life and Influence, ed. M. Lapidge (Cambridge, 1995), pp. 222–35, at 226–8; C. Cubitt, ‘Unity and Diversity in the Anglo-Saxon Liturgy’, Stud. in Church Hist. 32 (1996), 45–57 at 53–5; Y. Hen, ‘The Liturgy of St. Willibrord’, ASE 26 (1997), 41–62, at 49–52; see also K. Gamber, ‘Das altkampanische Sakramentar. Neue Fragmente in angelsächsischer Überlieferung’, RB 79 (1969), 329–42.

44 Liber sacramentorum romanae aecclesiae, ed. Mohlberg, pp. 7, 17, 20, 28, 77, 81, 82, 88, 90, 100, 104, 110, 114 and more; Siffrin, ‘Das Walderdorffer Kalendarfragment’, p. 221.

45 C. Hohler, ‘Some Service-Books of the Later Saxon Church’, at 61–2; Y. Hen, ‘The Liturgy of St. Willibrord’, p. 50, but it is not true that the Old Gelasian and the Sacramentary of Echternach are the only continental sacramentaries to divide masses this way, as Hen suggested. There is, for example, the ninth-century Sacramentary of Saint-Amand in Stockholm, for which no English influence can be detected at all, see Westwell, The Carolingian Sacramentaries of Saint-Amand, pp. 235–6, 403–6.

46 CLLA 601; Sacramentarium veronense, ed. C. Mohlberg (Rome, 1956).

47 Ibid. pp. 56, 108.

48 Veronense, ed. Mohlberg, n. 39.

49 Siffrin, ‘Zwei Blätter’, pp. 13–4 ‘ganz die Färbung leonianischer Orationen’ (i.e. Veronense).

50 CLLA 430; Facsimile in Le manuscrit VI-33. Archivio Arcivescovile Benevento. Missel de Bénévent, ed. J. Hourlier and J. Froger (Bern, 1983); edition by S. Rehle, ‘Missale Beneventanum (Codex VI 33 des Erzbischöflichen Archivs von Benevent’, Sacris Erudiri 21 (1972–1973), 323–405 at 10; Gamber had edited the preface in ‘Das kampanische Messbuch’, pp. 36–7.

51 A. Dold, ‘Ein kostbares Sakramentarfragment der Basler Universitätbibliothek’, Scriptorium 6 (1952), 260–73: ‘(UD qui i)n prin(cipio er)at aput te deum patrum uerbum per quem facta sunt (omnia) qui coaeternus tibi et in tua (atque e)x tua substantia manens (in nostrae naturae) similitudine … sacrificum (u)ix offe(re sufficit celi et) terrae possessor h(odie gra)nddeui symionis inua(lidis ges)tatur manibus acce(dit confitentis) et testificantis oracul(um uiduae) quo decebat ut ab utro (que annun)tiaretur sexu utrius(que saluator)’.

52 Merseburg, Domstiftsbibliothek, cod. 136, 75v–76r: ‘UD qui in principio erat apud te deum patrem uerbum. per quem facta sunt omnia. qui coaeternus tibi et in tua atque ex tua substantia manens. homo natus est in similitudinem passionum nostrarum. diues in suo pauper in nostro. par turturum aut duos pullos columbarum. sacrificio uix sufficit caeli terreque possessor hodie grandeui symeonis in ualidis gestatur manibus. A quo mundi rector et dominus predicatur. Accidit etiam testificantis oraculum uiduę quoniam decebat ut ab utroque annuntiaretur sexu utriusque saluator’. It would be too involved to go into this complex text in depth, but the argument of K. Gamber, ‘Das Basler Fragment. Eine weitere Studie zum altkampanischen Sakramentar und zu dessen Präfationen’, RB 81 (1971), 14–29, that the Basel fragment had been falsely reconstructed by Dold, and that instead we have two distinctive prefaces, one for the Circumcision and one for Candlemas, while superficially reasonable, is contradicted by the Merseburg text, which has the same text as the Basel fragment with the same abrupt transfer, via the contrast of richness and poverty, from the creedal statement to the narration of the Presentation. It is perhaps still likely that both texts show an accidental conflation of two originally distinct prefaces, but this occurred before the Basel fragment was written. For dating of the Merseburg manuscript, see B. Bischoff, Katalog der festländischen Handschriften des neunten Jahrhunderts, II: Laon – Paderborn (Wiesbaden, 2004), 184: ‘Reine Fuldaer Min … Fulda, IX Jh., ca. Mitte’.

53 Sacramentarium Fuldense saeculi x, ed. G. Richter and A. Schönfelder (Fulda, 1912, repr. Farnborough, 1980), p. 25: ‘UD. aeterne deus. In exultatione precipuae sollemnitatis hodiernae, in qua coaeternus tibi filius tuus unigenitus in nostra natus substantia a parentibus legali traditione deportatur in templum, idem legislator et custos, praecipiens et oboediens, dives in suo, pauper in nostro: par turturum uel duos pullos columbarum sacrificium uix offerre sufficit caeli terraeque possessor. Hodie grandeui Symeonis iuualidis gestatur in manibus, a quo mundi rector et dominus praedicatur. Accedit etiam oraculum uiduae testificantis, quoniam decebat, ut ab utroque annuntiaretur sexu utriusque saluator. Quem laudant’.

54 Siffrin, ‘Zwei Blätter’, p. 11, n. 1 places the composition of the preface in the context of the Council of Chalcedon (451).

55 Liber sacramentorum, ed. Mohlberg, p. 9, no. 21.

56 C. Coeburgh, ‘“Coniunctio” dans le sacramentaire de Fulda’, Ephemerides Liturgica 71 (1957), 428–30; Sacramentarium Fuldense, ed. Richter and Schönfelder, pp. 80 (likely mistakenly at the Hanc Igitur), 87, 88, 110, 112, 113, 343; more on the sacramentaries of Fulda, see E. Palazzo, Les sacramentaires de Fulda. Étude sur l’iconographie et la liturgie à l’époque ottonienne (Münster, 1994); C. Winterer, Das Fuldaer Sakramentar in Göttingen: Benediktinische Observanz und römische Liturgie (Petersberg, 2009).

57 CLLA 775 (with old signature); B. Bischoff, Die Abtei Lorsch im Spiegel seiner Handschriften (Lorsch, 1989), pp. 54, 102.

58 Bischoff, Die Abtei Lorsch, pp. 85, 124; L. Eizenhöfer, ‘Das Lorscher Sakramentar im Cod. Vat. Pal. Lat. 495’, Die Reichsabtei Lorsch. Festschrift zum Gedenken an ihre Stiftung, 764, ed. F. Knöpp, 2 vols. (Darmstadt, 1977) II, 129–69.

59 Bischoff, Die Abtei Lorsch, pp. 67, 93, 102.

60 T. Licht, ‘Beobachtungen zum Lorscher Skriptorium in karolingischer Zeit’, Karolingische Klöster. Wissenstransfer und kulturelle Innovation, ed. J. Becker, T. Licht and S. Weinfurter, Materiale Textkulturen 4 (Berlin, 2015), 145–62 at 159.

61 H. Spilling, ‘Angelsächsische Schrift in Fulda’, Von der Klosterbibliothek zur Landesbibliothek, ed. A. Brail (Stuttgart, 1978), pp. 47–98; Bischoff, Die Abtei Lorsch, pp. 22, 48–9; for Fulda, see also J. Raaijmakers. The Making of the Monastic Community of Fulda, c. 744–c. 900 (Cambridge, 2012).

62 K. Gamber, Sakramentartypen. Versuch einer Gruppierung der Handschriften und Fragmente bis zur Jahrtausendwende (Beuron, 1958), pp. 38, 91.

63 A. Dold, ‘Bedeutsame Reste dreier gregorianischer Sakramentar’, Ephemerides Liturgica 50 (1936), 359–74, at 369–73. See Veronese, ed. Mohlberg, at 37, 56.

64 A. Irving, ‘Mass Books in the Beneventan Zone’, On the Typology of Liturgical Books from the Western Middle Ages, ed. H. Buchinger and A. Irving (Münster, 2023), pp. 43–113; on Regensburg, K. Gamber, ‘Liturgiebücher der Regensburger Kirche aus der Agilolfinger- und Karolingerzeit’, Scriptorium 30 (1976), 3–25.

65 K. Gamber, ‘Ein römisches Eucharistiegebet aus dem 4.-5- Jahrhundert’, Ephemerides Liturgica 74 (1960), 103–14 at 111.

66 See B. Bischoff, ‘Theodore and Hadrian in England’, Biblical Commentaries from the Canterbury School of Theodore and Hadrian, ed. B. Bischoff and M. Lapidge (Cambridge, 1994), pp. 133–89, at 164–7; other options in Hen, ‘Rome, Anglo-Saxon England and the Formation of the Frankish Liturgy’, at 308–9; D. Pelteret, ‘Travel between England and Italy in the Early Middle Ages’, Anglo-Saxon England and the Continent, ed. G. Waxenberger, H. Sauer and J. Story (Tempe, 2011), pp. 245–73.

67 Frank, ‘Die Briefe des Heiligen Bonifatius’, at pp. 60–83.

68 Hen, ‘The Liturgy of St Willibrord’, pp. 57–9; L. Eizenhöfer, ‘Zu Bannisters Echternacher Meßformulare für Vigil von Christi Himmelfahrt’, Colligere Fragmenta. Festschrift für Alban Dold zum 70. Geburtstag, ed. B. Fischer and V. Fiala (Beuron, 1952), pp. 166–72.

69 Re-balancing mission narratives: L. van Padberg, Mission und Christianisierung: Formen und Folgen bei Angelsachsen und Franken im 7. Und 8. Jahrhundert (Stuttgart, 1995); I. Wood, The Missionary Life: Saints and the Evangelization of Europe 400–1500 (Harlow, 2001), pp. 57–141.

70 Gn-L, ASMss 856.3.

71 B. Bischoff, V. Brown and J. John ‘Addenda to Codices Latini Antiquiores (II)’, MS 54 (1992), 286–307 at 298.

72 E. Freise, ‘VII.39. Sakramentar “Gelasianum” mixtum’, 799. Kunst und Kultur der Karolingerzeit. Karl der Große und Papst Leo III in Paderborn. Katalog der Ausstellung Paderborn 1999, ed. C. Stiegemann and M. Wemhoff, 2 vols. (Mainz, 1999) II, 485–7.

73 Münster, Fragmente Kapsel, IV, 8, vertically written note across two folios, now separate: ‘Ego frater Iobannes Assindie dictus prior in Werden profiteor manu propria quod hunc librum qui dicitur Smagaddus super Regulam sanctissimi Patris nostri Benedicti accepi accommodatu a domino venerando Gerbardo abbate sanctissimi Martini in Colonia, quae quicumque hora post obitum meum, quem contedat altissimus beatum iuxta solitam clementiam suam, experietur sine mora testitetur’. For Abbot Gerhard, see P. Opladen, Gross St. Martin. Geschichte einer stadtkölnischen Abtei (Dusseldorf, 1954), p. 173.

74 Most well known are the fragments of the Werden Glossary, destroyed by the bombing of Münster in 1945 (Münster, Universitätsbibliothek, Paulinus 271). Also fragments in Münster Nordhrhein-Westfälisches Staatsarchiv, recorded by B, Bischoff, Katalog II, 303 or the early fragment of a copy of Bede in the Universitätsbibliothek described in E. Freise ‘VII.41. Beda Venerablis. Historia ecclesiastica gentis Anglorum’, 799. Kunst und Kultur der Karolingerzeit, ed. Stiegemann and Wemhoff, II, 489–90 and in CLA Supplement 1848.

75 Freise, ‘Sakramentar “Gelasianum” mixtum’, p. 485: ‘seit dem 9. Jahrhundert in der Werdener Klosterbibliothek aufbewahrt worden ist’.

76 CLLA 415; Gn-L, ASMss 836.5. These fragments have been digitised and can be viewed online (https://historischesarchivkoeln.de/archive.xhtml?id=Vz%20%20%20%20%20%2010000385b%F6#Vz______10000385b%F6 ).

77 H. M. Bannister, ‘Fragments of an Anglo-Saxon Sacramentary’, JTS 12 (1911), 451–4. The two new fragments resurfaced, presumably thanks to Bischoff, sometime between the publication of E. Bourque, Études sur les sacramentaires romains, II: Les Textes Remainiés, pt.1: Le Gélasien du VIIIe siècle (Quebec, 1952), where only fragm. 24 is noted at pp. 27–8, and the publication of the eighth volume of CLA in 1959, in which all three fragments are noted and discussed. Gamber, CLLA, p. 235, listed the three signatures, but actually only discussed Nr. 24. This has perhaps led some to assume Fragm. Nrs. 123 and 124 had already been analysed. For example, Pfaff, ‘Massbooks’, p. 9, lists all three shelfmarks, but only describes the content as ‘eight votive masses’, i.e. no mention of Fragm. Nrs. 123 and 124.

78 Frank, ‘Die Briefe des Heiligen Bonifatius’.

79 CLA V 581 ‘Origin probably Ireland or Northern England’. CLLA 803; edited by H. M. Bannister, ‘Liturgical Fragments’, JTS 9 (1908), at 398–404: ‘provenance … some monastery in the East of the Frankish kingdom which had inherited English MSS or where English scribes had been employed’; Pfaff, The Liturgy in Medieval England, pp. 42–3; Gn-L ASMss 893.5.

80 Hen, ‘The Liturgy of St. Willibrord’, pp. 55–7; N. Netzer, Cultural Interplay in the Eighth Century: the Trier Gospels and the Making of a Scriptorium at Echternach (Cambridge, 1994), pp. 113, 115.

81 It has continued to be repeated: e.g. Gameson, ‘Northumbrian Books’, p. 80: ‘Werden (s.IX)’; R. Bremmer, ‘The Anglo-Saxon Continental Mission and the Transfer of Encylopaedic Knowledge’, Foundations of Learning: the Transfer of Encyclopaedic Knowledge in the Early Middle Ages, ed. R. Bremmer and K. Dekker (Paris, 2007), pp. 19–50 at 25, 47, who places it in Liudger’s library.

82 Freise, ‘Sakramentar Gelasianum Mixtum’, at p. 487; also a possibility in Gn-L, ASMss at p. 601; Gamber also discussed with this Bischoff in a letter of 21 October 1986 in the Nachlass of Bischoff, now München, Bayerische Staatsbibliothek, ANA 553 C,I GAMBER, Klaus: ‘Die Schrift scheint mir überaus ähnlich zu sein. Unterschiedlich ist nur die Zeilenzahl 24 statt 27. Auch sind dort (Cologne) die Initialien nicht künstlerisch so wertvoll wie im neuen Fragment (Münster)’.

83 Bannister, ‘Liturgical Fragments’, p. 398.

84 CLLA 210.

85 Cf. Bannister, ‘Liturgical Fragments’, p. 402, n. 5.

86 Nor can it be seen in the Durham Cassidorus or the Egerton Wisdom, but it can be seen in Cambridge, MS Kk.1.24. But none of these manuscripts have initials that parallel the Münster/Cologne sacramentary.

87 I thank Teresa Porcini and Jo Story for noting and communicating this latter parallel, which makes it further clear that the Paris fragments should also belong to the same manuscript.

88 Bischoff, Brown and John, ‘Addenda (II)’, p. 298.

89 CLA VIII, p. 40, n. 1165.

90 Bannister, ‘Fragments of an Anglo-Saxon Sacramentary’, p. 451.

91 Bannister, ‘Liturgical Fragments’, p. 399.

92 Ibid. p. 398.

93 Netzer, Cultural Interplay, pp. 112–16; Hen, ‘The Liturgy of St. Willibrord’ at 55–7 also suggests Paris, lat. 9488 is ‘early eighth century’.

94 Paraphrased from a letter in Siffrin, ‘Das Walderdorffer Kalendarfragment’, pp. 202–3: ‘unserem Fragment gleiche Schrift der Berliner Blätter nahe an die Schrift der von ihm [Mohlberg] im Kölner Stadtarchiv gefundenen Sakramentarblätter, die von einer charakteristisch nordenglischen Hand der ersten Hälfte des 8.Jh. geschrieben sind [i.e. Fragm.24]’.

95 CLA V, p. 19, speaking of Paris, lat. 9488, fols. 3–4, states ‘frequent confusion of ci and ti (is) unusual in insular MSS’.

96 CLLA Supplementum, p. 50. On Sankt Gall 348, CLLA 830.

97 The Calendar of St Willibrord, ed. Wilson, p. 4; Juliana of Cumae is also present in our fragment, and in Willibrord’s calendar. She has an Old Gelasian mass (GeV 841–842), one that was straightforwardly taken into the Gelasian of the eighth century.

98 Compare Das fränkische Sacramentarium Gelasianum, ed. Mohlberg, pp. 180–1.

99 Bannister, ‘Liturgical Fragments’, p. 399: ‘it agrees with the later “Gelasian” or mixed Sacramentary S (MS St Gall cod. 348), usally ascribed to about the year 800 … and from the kindred MS R (Rheinau No 30 at Zurich) … basically identical with S and with R’.

100 In the edition given in the Appendix below, 17 ‘abstibus’ not ‘abstinentes’ and ‘strictis’ not ‘restrictis’, compare Sg 1230. In 12 ‘tibi’ missing, added to Sg 218, 10 ‘gloriosis’ missing, added in Sg 216; 7 is missing the added clause at the end ‘eorum nostrarum absolutionem culparum’ in Sg 213; 24 the uncorrected ‘a te’ instead of the corrected ‘apud te’, compare Sg 1429 etc.

101 Bannister, ‘Fragments of an Anglo-Saxon Sacramentary’.

102 Identified first by A. Baumstark, ‘Untersuchungen’, Die älteste erreichbare Gestalt des Liber Sacramentorum anni circuli der römischen Kirche ed. K. Mohlberg and A. Baumstark (Münster, 1927), p. 27*: ‘die von einer charakteristischen nordenglischen Hand der ersten Hälfte des 8. Jahrhunderts geschriebene Sakramentarbruchstücke des Kolner Stadt-Archivs … gehören der Sphäre der Votivmessen und Segnungsgebete an, und es sind G[ellone], Ph[ilipps] und Sangall. 350, wo sie in gleicher Reihenfolge erscheinen’; Bourque, Étude II, pt.1, p. 27: ‘Six de ces formules se retrouvent dans le Saint-Gall 350 aux nn.164, 167, 174, 175, 215 et 216’; Frank, ‘Die Briefe vom Heiligen Bonifatius’, pp. 83–8; Subsequently in Gamber, CLLA, p. 235; Elsewhere Gamber wondered if the Cologne Fragm. Nr. 24 transmitted an original Northumbrian type from which the Gelasian of the eighth century took its votive masses, but the proof from the Münster, Paris and additional Cologne fragments that the Temporale and Sanctorale were identical to cod.348 shows this was not the case, see Gamber, ‘Das altkampanische Sakramentar’, at pp. 338–9.

103 Ein Sankt Galler Sakramentar Fragment, ed. Manz, pp. 16–7; 27–8. To note that this is also true of the two blessings of fruits that Bannister said were not found in any printed edition, Bannister, ‘Fragments of an Anglo-Saxon Sacramentary’, p. 454. This conclusion has led to errors in Gamber and in F. Spiegel, ‘The “tabernacula” of Gregory the Great and the Conversion of Anglo-Saxon England’, ASE 36 (2007), 1–13 at 9–10. The mass for blessing of new fruits is not ‘a service without precedent in the Gelasian Sacramentary … or in any other foreign liturgical book’ and cannot be seen as evidence of an insular harvest festival, as Spiegel claims; the two prayers are actually found, as below, exactly as in our book in Augustodunensis, ed. Heiming, pp. 230–1 (Ph 1854 and 1855), and in Gellonensis, p. 445 (Gel 2828 and 2829), in the latter with slightly different incipits and an erasure of the ending of the second, plus addition of a new post communion to make them a complete mass, when originally they were simply two blessings.

104 CLA V 581: ‘an entry in Latin in sixteenth-century cursive (fol. 4v) has an umlaut-like accent above u which suggests a Germanic region; a sixteenth-century entry in French occurs on fol. 3’.

105 CLLA Supplementum, p. 50.

106 Dating based on T. Klauser, ‘Die liturgischen Austauchbeziehungen zwischen der römischen und der fränkish-deutschen Kirche vom achten bis zum elften Jahrhundert’, Historisches Jahrbuch 53 (1933), 169–89 at pp. 173–5, repr. in Gesammelte Arbeiten zur Liturgiegeschichte, Kirchengeschichte und Christlichen Archäologie (Münster, 1977), pp. 139–54; Vogel, Medieval Liturgy, pp. 70–8.

107 B. Moreton, The Eighth-century Gelasian Sacramentary: a Study in Tradition (Oxford, 1979); K. Gamber, ‘Die Heimat und Ausbildung des Gelasiana saec.VIII’, Sacris Erudiri 14 (1963), 99–129.

108 On Praiectus, see B. Moreton, ‘A Patronal Festival? St. Praiectus and the Eighth-Century Gelasian’, JTS 27 (1976), 369–80; Bourque, Étude sur les sacramentaire romains II, pt.1, 229: ‘sans l’intervention personell du roi france, il est moralement impossible d’expliquer la diffusion rapide et l’adoption jusque dans les plus hautes sphéres du type nouvellement formé, si divers d’apparence du précédent Gélasien et conséquemment peu susceptible de plaire aux moins et clercs fort conservateurs en matiére liturgique’ (‘were it not for the personal intervention of the French king, it would be morally impossible to explain the rapid spread and adoption in the highest circles of the newly formed type, so different in appearance from the previous Gelasian [the Old Gelasian] and consequently unlikely to please the clergy, who were very conservative in liturgical matters’).

109 Association with Pippin in E. Bishop, Liturgica Historica (Oxford, 1918), p. 152, n. 1, p. 183; Vogel, Medieval Liturgy, p. 73: ‘In 742 Pippin III founded an abbey at Flavigny-sur-Ozerain’; in fact, Pippin merely gave Flavigny fishing rights, and asked them to pray for him, see C. Bouchard, The Cartulary of Flavigny, 717–1113 (Cambridge, MA, 1992), pp. 33–4; More cautious is Y. Hen, The Royal Patronage of Liturgy (London, 2001), pp. 57–61, but he argues that Flavigny’s prestige due to the royal connections had implicit influence on the Gelasian’s dissemination.

110 Bourque, Étude sur les sacramentaire romains II, pt.1, 222–3 acknowledges the presence of the eighth-century Gelasian in England (or rather in Anglo-Saxon manuscripts), but makes nothing of this.

111 Quoted in Frank, ‘Die Briefe’, pp. 86–7: ‘Ich muß wiederum gestehen, es ist schwer, angelsächsische Majuskeln genauer zu datieren, man kann vielleicht sogar von einer künstlerischen Tradition dieser Schriftart für liturgische Codices sprechen, bis in das späte VIII. Jahrhundert. Nordenglischer Charakter dürfte zutreffen, was natürlich auch für die festländischen Pflanzstätten gilt. Sicherer zeugt m. E. für Enstehung in der II. Hälfte des VIII. Jahrhunderts die Neigung an und in den schwarzen Initialien (schwächliche) Ornamente anzubringen. Es ist wohl überhaupt bei liturgischen Hss. noch mehr als bei anderen Kategorien (z. B. Rechtsdenkmälern) damit zu rechnen, daß die absolute Chronologie der Hss. sich keineswegs mit dem Auftreten der neuen empfohlenen Typen in Einklang bringen läßt. Ich bin geneigt, den Regin.316 erst in das III. Viertel des VIII. Jhs. zu setzen. So ist also das zeitliche Verhältnis der Texttypen vornehmlich nach der inneren Entwicklungsgechichte aufzuhellen’.

112 Convincingly established in Moreton, The Eighth-Century Gelasian, though his ignoring of Cologne Fragm. Nr. 24 and most other fragments is unfortunate; also argued by Gamber, ‘Heimat und Ausbildung’, pp. 101–2; our fragment also goes some way to proving that the idea that Gellone represents the earliest redaction, closest to the supposed Flavigny archetype, as in Liber sacramentorum Gellonensis, ed. Dumas, I, xxiii–xxvi, or in Vogel, Medieval Liturgy, p. 76, cannot be true. For example, Gellone’s manipulation of the blessings of fruit as we find them in our book (see above, n. 75), are clearly posterior. The additions that the compilers of Gellone have made to the votive masses like Gel 2679 and 2770, 2783, 2827 are missing in Cologne Fragm. Nr. 24.

113 J. Deshusses, ‘Le Sacramentaire de Gellone dans son contexte historique’, Ephemerides Liturgicae 75 (1961), 193–210.

114 This was, in the end, also Frank’s conclusion about Fragm. Nr. 24: ‘Die Briefe’, pp. 87–8: ‘Die Kölner Blätter sind erst in der zweiten Hälfte des 8. Jahrhunderts von einem Angelsachsen in einer angelsächsischen Pflanzstätte des Festlandes geschrieben worden’.

115 On the monastic character of the Gelasian of the eighth century, see Vogel, Medieval Liturgy, p. 73; Moreton, The Eighth-Century Gelasian, p. 15.

116 Aegidius Gelenius, De admiranda sacra et civili magnitudine Coloniae (Cologne, 1645), pp. 375, 721; Nevertheless, the existence of a monastery of Groß Sankt Martin can only really be confirmed from a (re-) foundation in the tenth century, which placed it under ‘Schotten’. Opladen, Gross St. Martin, pp. 13–24.

117 J. Deshusses, ‘Le bénédictionnaire gallican du VIIIe siècle’, Ephemerides liturgicae 77 (1963), 169–87.

118 Pope Zacharias I, Letter to Boniface, ed. M. Tangl, MGH, Epist. Select. 1 (Berlin, 1916), 194–201 at 198.

119 A. Prescott, ‘The Text of the Benedictional of St Aethelwold’, Bishop Aethelwold: His Career and Influence, ed. B. Yorke (Woodbridge, 1988), pp. 119–47. The palimpsest in Cambridge, Trinity College Library, MS 24, fol. 78–83 (s.viii/ix, England) is sometimes described as a ‘Benedictional’ (see Gn-L, ASMss 202), but the indications of M. R. James, A Descriptive Catalogue of the Manuscripts in the Library of Trinity Hall (Cambridge, 1907), p. 40 show it did not contain this kind of blessing of the people for communion, but rather the common and not specifically Gallican blessings for objects and places, with the surviving title being ‘benedictio domus novae’; H. Mayr-Harting, The Coming of Christianity to Anglo-Saxon England (Pennsylvania, 1991), pp. 178–80 supposed that the Anglo-Saxons used the Gallican blessings before this, but his only tangible evidence is Latin 9488, which he gives as ‘early eighth-century’.

120 Liber sacramentorum gellonensis I, ed. Dumas, 262–300; Liber sacramentorum augustodunensis, ed. Heiming, pp. 151–5.

121 Gel 2077.

122 CBP 249. Specifically, in the second Benedictional of Freising in R. Amiet, The Benedictionals of Freising (London, 1974), p. 97 (II 395–400), which is related to the Gelasian Benedictionals elsewhere (see pp. 49–62).

123 Baumstark, ‘Untersuchungen’, pp. 134*–41*; also A. Baumstark, Missale Romanum (Eindhoven, 1929), pp. 29–99.

124 Gamber, CLLA, p. 373: ‘Die Bedeutung dieses kleinen Fragments liegt darin, daß es zeigt, wie früh bereits Sacramentaria Gelasiana mixta den Weg nach England gefunden haben’ [‘The significance of this small fragment (latin 9488, fols. 3–4) lies in the fact that it shows how early the Gelasian of the eighth century had already come to England’].

125 CLA IX, 1409: ‘Written presumably in an Anglo-Saxon centre in Germany’; CLLA 804 suggests Würzburg itself; digitized at http://vb.uni-wuerzburg.de/ub/mpthf42/pages/mpthf42/205.html. Not in Gn-L, ASMss. Never edited but contains texts for the Sunday after Christmas, and Epiphany: Sg 88, 80, 90, 91, 92, 93, 94 and 95. It is impossible to say why, but the post communion for the Sunday mass has been moved from a previous Sunday compared to other Gelasians.

126 CLA IX 1372: ‘Written presumably in an Anglo-Saxon centre in Germany’; Bischoff, Brown and John, ‘Addenda (II)’, pp. 303–4; B. Bischoff, Katalog II, 75, n. 2296, ‘Deutsch-insulares Gebiet, VIII/IX Jh.’; CLLA 851. The best account is J. Hoffmann, ‘Verstreute Blätter eines deutsch-insularen Sakramentars aus Neustadt am Main’, Mainfränkisches Jahrbuch für Geschichte und Kunst 9 (1957), 133–41, but the fragments in Munich and in the Würzburg Staatsarchiv have still never been individually analysed. Not in Gn-L, ASMss.

127 CLLA 832. CLA XI, 1590; C. Mohlberg, ‘Note su alcuni sacramentarii’, Reconditi della Pontificia Accademia Romana di Archeologia 16 (1940), 131–70.

128 P. Lehmann, ‘Mitteilungen aus Handschriften V’, Sitzungsberichte der Bayerischen Akademie der Wissenschaften. Phil.-hist. Abteilung 4 (1938), 3–93, fragment at 7–19, at p. 11: ‘Beinflussung durch ein nicht fernes insulares Zentrum und seine Schreibgewohnheiten ist wahrscheinlich’.

129 Moreton, The Eighth-century Gelasian, p. 173; further on the origins of the Gelasian, see Westwell, The Carolingian Sacramentaries of Saint-Amand, pp. 53–65.

130 See CP 450, 1290 and Das sacramentarium gelasianum, p. 138 (Sg 898) and pp. 45–6 (Sg 300).

131 Vogel, Medieval Liturgy, p. 78.

132 Dold, ‘Ein kostbares Fragment’, p. 265: ‘… tur animarum laudes que tuas domine fidenter intendes coniungere uocibus angelorum qui gloriam tuam concinnunt sine fine dicentes’, though Dold could not identify it.

133 For Anglo-Saxons at these monasteries, see Levison, England and the Continent, p. 136, n. 1. My forthcoming article, A. Westwell, ‘Die Dynamik und Vielfalt des liturgischen Wandels in frühmittelalterlichen Klöstern am Beispiel der frühen und fragmentarischen Sakramentare der Reichenau und des Bodenseeraums’, Vorträge und Forschungen. Konstanzer Arbeitskreis für Mittelalterliche Geschichte (forthcoming), demonstrates on the basis of some little known fragments that Reichenau is a definite possibility.

134 For example, CLLA 1412, CLLA 1413. The earliest example is a palimpsest CLLA 1401, from the eighth century.

135 For example, the Old English Martyrology suggests that neither older nor newer sacramentaries had a mass for the conversion of Saint Paul (25 January), nor the masses of Saint Benedict or the Maccabees, which are all entirely characteristic of all Gelasians of the eighth century.

136 Hohler, ‘The Sacramentary Used by Saint Boniface’, p. 92 notes the presence of a mass of Saint Cuthbert in the Sacramentary of Fulda (Fuldense, ed. Schönfelder, p. 13), also noted by Levison, England and the Continent in the Tenth Century, p. 146, n. 3. This mass of a Northumbrian saint was likely also from an English book, but is also likely to have been transmitted by English hands other than the Wessex man Boniface himself. The Lorsch manuscripts are, apart from the use of CONIUNCTIO, essentially Gregorian sacramentaries with little to distinguish them, and thus show no other traits that link them to our books.

137 Bannister assumed that there would be no real difference between the books in England and those copied on the Continent, because there was no real liturgical tradition in the areas evangelised, see Bannister ‘Liturgical Fragments’, p. 399: ‘where there were no previous local uses which they might have desired to adopt, they would in all probability have continued the use to which they had been accustomed’.

138 For the Merovingians, Hen, Culture and Religion in Merovingian Gaul; a good historiographical summary in Palmer, Anglo-Saxons in a Frankish World, pp. 9–21.

Figure 0

Figure 1. A portion of the Canon of the Mass in the ‘Sacramentary of Boniface’. Regensburg, Bischöfliche Zentralbibliothek, Fragment III 1, 1r (Northumbria, s. viiimed).

Figure 1

Figure 2. A folio containing a mass for Christmas in the ‘Boniface Sacramentary’. Berlin, Staatsbibliothek zu Berlin, Preußischer Kulturbesitz MS Lat. fol. 877, fol.1r (Northumbria, s. viiimed).

Figure 2

Figure 3. Half of a folio, cut vertically, containing a blessing for the fifth Sunday after Epiphany in a Gelasian Sacramentary from Groß Sankt Martin. Münster, Universitäts- und Landesbibliothek, Fragmente Kapsel IV, 8, 1r (Anglo-Saxon centre on the Continent, (Cologne? Echternach?), s.viii2).

Figure 3

Figure 4. A folio containing masses for Saint Juliana and the Cathedra of Saint Peter in a Gelasian Sacramentary from Groß Sankt Martin. Münster, Universitäts- und Landesbibliothek, Fragmente Kapsel IV, 8, 2v (Anglo-Saxon centre on the Continent (Cologne? Echternach?), s.viii2).

Figure 4

Figure 5. A half folio cut horizontally, containing the end of the mass for the Ember Saturday and opening of the Dominica Vacat in a Gelasian Sacramentary from Groß Sankt Martin. Cologne, Historisches Archiv, HAStK-RBA Best. 7050B (Fragmente B), 124v (Anglo-Saxon centre on the Continent (Cologne? Echternach?, s.viii2).

Figure 5

Figure 6. A folio from a fragmentary Sacramentary, containing the end of a Common Mass for a Virgin. Paris, Bibliothèque nationale de France, Latin 9488, 4r (Anglo-Saxon centre on the Continent (Cologne? Echternach?, s.viii2). Source gallica.bnf.fr/Bibliothèque nationale de France.

Figure 6

Figure 7. A folio containg parts of blessings of a house and of new fruits, in a Gelasian Sacramentary from Groß Sankt Martin. Cologne, Historisches Archiv, HAStK-RBA Best. 7050B (Fragmente B), 24, 4r (Anglo-Saxon centre on the Continent. (Cologne? Echternach?), s.viii2).