Hostname: page-component-586b7cd67f-tf8b9 Total loading time: 0 Render date: 2024-11-27T14:42:28.507Z Has data issue: false hasContentIssue false
  • English
  • Français

The reconstruction of the abbey church at St-Denis (1231–81): the interplay of music and ceremony with architecture and politics*

Published online by Cambridge University Press:  05 December 2008

Anne Walters
Anne Walters
Affiliation:
University of Chicago
Get access Rights & Permissions [Opens in a new window]

Extract

During the Middle Ages, the construction or renovation of a great church was a vast undertaking in terms of time and of money. An array of cathedrals that still rise above the cities and towns of northern France bears witness to medieval ingenuity and industry. The first structure which embodies many of the elements that are now called Gothic was begun at St-Denis in 1140. The church as we know it, however, was not entirely the work of the twelfth century. Only much later, between 1231 and 1281, was St-Denis finally completed. Substantial evidence of the thirteenth-century rebuilding is found in the monument which stands just to the north of Paris. But the stones of St-Denis do not tell the entire story: a small handful of documents refer to stages of the fifty-year reconstruction of the abbey, and now, new witness exists in the form of the liturgical manuscripts which have survived from the thirteenth century.

Type
Research Article
Information
Early Music History , Volume 5 , October 1985 , pp. 187 - 238
Copyright
Copyright © Cambridge University Press 1985

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 The principal monographs on the architecture of St-Denis now available are Crosby, S. M., The Abbey of St.-Denis, 475–1122 (New Haven, 1942)Google Scholar; Formigé, J., L'abbaye royale de Saint-Denis: recherches nouvelles (Paris, 1960)Google Scholar; and Gerson, P., ‘The West Façade of St.-Denis: an Iconographic Study’ (Ph.D. dissertation, Columbia University, 1970).Google Scholar Other discussions of the church can be found in von Simson, O., The Gothic Cathedral: Origins of Gothic Architecture and the Medieval Concept of Order (New York, 1956), pp. 61–141Google Scholar; Bony, J., French Gothic Architecture of the 12th and 13th Centuries (Berkeley, 1983), pp. 61–3, 366–80Google Scholar; Frankl, P., The Gothic-literary Sources and Interpretations through Eight Centuries (Princeton, 1960), pp. 3–14Google Scholar; Sauerländer, W., Gothic Sculpture in France, 1140–1270, trans. Sondheimer, J. (New York, 1972), pp. 11, 17, 38Google Scholar and passim; and Stoddard, W. S., Monastery and Cathedral in France (Middletown, Connecticut, 1966), pp. 101–11.Google Scholar

2 A brief study of the thirteenth-century architecture is contained in Crosby, , L'abbaye royale de Saint-Denis (Paris, 1953), pp. 57–65.Google Scholar These pages will be superseded by Bruzelius, op. cit.

3 Three chronicles, all emanating from St-Denis in the thirteenth and fourteenth centuries, mention the beginning, the end and several intermediate steps of this process. Bruzelius (op. cit.) examines these records in detail, and they are treated below in the discussion of the liturgical changes.

4 These practices are the subject of Walters, A., ‘Music and Liturgy at the Abbey of Saint-Denis, 567–1567: a Survey of the Primary Sources’ (Ph.D. dissertation, Yale University, 1984).Google Scholar

5 Apart from the work just cited, the only writing that is devoted entirely to an aspect of music and ritual at St-Denis is Huglo, M., ‘Les chants de la missa greca de Saint-Denis’, Essays Presented to Egon Wellesz, ed. Westrup, J. (Oxford, 1966), pp. 74–83.Google Scholar Niels Rasmussen called for further investigation of these areas in ‘The Liturgy at Saint-Denis: a Preliminary Study’, a paper delivered at the International Symposium on St-Denis and Abbot Suger held at Columbia University, New York, in 04 1981.Google Scholar

6 The amount of literature in both fields is vast. The abbey is distinguished not only for its architecture, but also for its remarkable stained glass, sculpture, objets d'art and illuminated manuscripts. Access to the specialised studies in these areas is given in the bibliography of Crosby, S., Hayward, J., Little, C. T. and Wixom, W. D., The Royal Abbey of Saint-Denis in the Time oj Abbot Suger, 1122–1151 (New York, 1981), pp. 119–24.Google Scholar See also Avril, F., L'enluminure à la cour de France au XlVe siècle (New York, 1978), pp. 32, 80–4.Google Scholar The historiographic activity of the monks and its subsequent interpretation by later scholars are reviewed in Spiegel, G. M., The Chronicle Tradition of Saint-Denis: a Survey (Brookline, Massachusetts, 1978).Google Scholar Other recent works include E. Brown, A. R., ‘The Ceremonial of Royal Succession in Capetian France’, Speculum, 55 (1980), pp. 266–93Google Scholar; and her forthcoming book, Dynasticism and Death at St.-Denis: the Forging of the Dionysian Alliance 987–1328.

7 The comprehensive histories of the abbey are: Doublet, J., Histoire de l'abbaye de S. Denys en France (Paris, 1625)Google Scholar; Félibien, M., Histoire de l'abbaye royale de Saint-Denis en France (Paris, 1706)Google Scholar; and d'Ayzac, F., Histoire de l'abbaye de Saint-Denis en France, 2 vols. (Paris, 1860–1861).Google Scholar Of these three works. Felibien's eighteenth-century monograph has become standard because of its meticulous use of primary sources. See the appraisal of these and other peripheral studies in Crosby, , Abbey, pp. 17–19.Google Scholar

8 No better witness to the longevity of the regal–monastic relationship can be cited than the original charter, Paris, Archives Nationales (abbreviated F-Pan), k. 1, no. 7. This document from Clothair n to the abbey, dated 625, is the oldest royal act preserved in the Archives Nationales. It is published in Tardif, J., Monuments historiques: cartons des rois (Paris, 1866), p. 4.Google Scholar

9 The function of the abbey as royal necropolis is treated in Erlande-Brandenburg, A., Le roi est mort: étude sur les funérailles, les sépultures et les tombeaux des rois de France jusqu ' à la fin du XIIIe siècle (Geneva, 1975), pp. 68–78, 97–106Google Scholar; Giesey, R., The Royal Funeral Ceremony in the Renaissance, Travaux d'Humanisme et de Renaissance 37 (Geneva, 1960), pp. 29–34, 125–44Google Scholar; and in Brown, op. cit. On the ceremonies for the interments of kings, see Walters, ‘Music and Liturgy’, pp. 106–9.

10 Suger's biography was first composed by his kinsman Guillelmus, a monk of St-Denis (see Guizot, F. P., ed., Collection des mémoires relatifs à l'histoirede France, VIII, Paris, 1825, pp. 161–205).Google Scholar Later authors to treat of Suger include Combes, F., L 'Abbé Suger: histoire de son ministère et de sa régence (Paris, 1853)Google Scholar; Cartellieri, O., Abt Suger von Saint-Denis (Berlin, 1898)Google Scholar; Aubert, M., Suger (Paris, 1950).Google Scholar Although the abbot did not institute the composition of chronicles at St-Denis, it was during his abbacy that these histories took on a royalist viewpoint (see Spiegel, , Chronicle Tradition, pp. 44–7).Google Scholar

11 Sumner Crosby's study of the twelfth-century church will soon be published posthumously. The details of the construction are recorded by Abbot Suger himself (see Panofsky, E., trans, and ed., Abbot Suger on the Abbey Church of St.-Denis and its Art Treasures, 2nd edn by Panofsky-Soergel, G., Princeton, 1979Google Scholar; and de la Marche, A. Lecoy, Oeuvres completes de Suger, Paris, 1867).Google Scholar

12 See the drawings of the building in Panofsky, Abbot Suger, first Illustration (pages unnumbered), and in Crosby, Abbey (coloured plans at end of text).

13 The Merovingian edifice at St-Denis was rebuilt under the powerful Abbot Fulrad (see Crosby, , Abbey, pp. 87–164Google Scholar). King Charlemagne honoured the convent with his presence at the dedication of the Carolingian church on 24 February 775 (see Walters, , ‘Music and Liturgy’, pp. 61–4Google Scholar).

14 See Crosby, , Abbaye, pp. 57–8.Google Scholar

15 The thirteenth-century historian Guillaume of Nangis states that the convent had not dared to tamper with the building ‘propter dedicationis mysterium quod eadem ecclesia noscitur a Domino suscepisse’ (Geraud, H., ed., Chronique latine de Guillaume de Nangis de 1113 à 1300, 2 vols., Paris, 1843, I, p. 183).Google Scholar

16 Ibid.; see also Felibien, , Histoire, p. 227.Google Scholar

17 The architectural details in this paragraph are cited from Bruzelius, op. cit. See her Figure no. 26; also Figure 1 of the present article.

18 These manuscripts are treated in depth in Walters, , ‘Music and Liturgy’, pp. 232–344.Google Scholar

19 The dates of these sources are discussed below.

20 The Chronicon is found in Vatican City, Biblioteca Apostolica Vaticana (abbreviated I-Rvat), MS lat. 309, fols. 17–58; also published in Berger, E., ‘Annales de Saint-Denis généralement connues sous le titre de Chronicon Sancti Dionysii ad Cyclos Paschales’, Biblio-theque de l' École des Charles, 40 (1879), pp. 261–95.CrossRefGoogle Scholar Two versions of the chronicle appear successively in the manuscript. In the first, the entries are made by different scribes, beginning in the ninth century. This document is considered to be particularly trustworthy because the scribal hands are contemporaneous with the events they record. The redaction of the second version dates from the thirteenth century and often elaborates on information presented in the earlier series of notes. The details of the translation of Hippolytus are found with identical wording in both sources: ‘Hoc anno in Purificatione Beate Marie translatum fuit cum magna sollempnitate corpus Sancti Ypoliti ab oratorio quod diu fuerat in media navi ecclesie in novum oratorium in sinistra parte novi operis’; I-Rvat lat. 309, fols. 32, 53v; Berger, , ‘Annales’, pp. 281, 290.Google Scholar Jean Vézin has recently enumerated the palaeographical features which, in addition to the St-Denis-orientated content of the manuscript, point to the abbey as the place of copying of the ninth-century portion of I–Rvat 309 (Vézin, , ‘Le point d'interrogation, un élément de datation et de localisation des manuscrits: 1'exemple de St. Denis au IXe siècle’, Scriptorium, 34. 1980, pp. 181–96).CrossRefGoogle Scholar

21 F–Pm 526, fols. 113v, 127, 157r-v; F–Pn lat. 976, fols. 92, 116.

22 Fol. 68.

23 Fol. 39v.

24 Fol. 112v.

25 Fol. 80v.

26 The dates of F–Pm 526 had recently been placed between 1234 and 1247. Auguste Molinier first assigned the year 1234 (Molinier, , Catalogue des manuscrits de la Bibliothèque Mazarine, 4 vols., Paris, 1885–1890, I, p. 211Google Scholar) on the basis of the obit of Philippe Hurepel, Count of Boulogne (d. 1234), which appears in the original hand of both calendar and manuscript. Charles Samaran adopted this date and proposed the terminus post quern non 1247, based on the addition of the obit of Abbot Eudes Clément. See Samaran, C. and Marichal, R., Catalogue des manuscrits en écriture latine portant des indications de date, de lieu ou de copiste, 6 vols. (Paris, 1959–1974), I, p. 245.Google Scholar

27 ‘ordinamus et statuimus ut festum ipsius pii confessoris et doctoris egregii [Bernardi] in crastino sancti Audoeni singulis annis a nobis sollempniter celebretur’; F-Pan ll 1159 (a thirteenth-century cartulary from the abbey), pp. 1–2; also reprinted in Félibien, , Histoire, Pièces justificatives i, no. clxx.Google Scholar At Eudes’ urging in 1232, the Cistercian monasteries had elevated the feast of St Denis to a higher rank in their liturgy (Ibid., no. clxvii). It is possible that this abbot felt somewhat obliged ten years later to respond in kind to the Cistercians, although there is no record of such motivation for his statute of 1241. Eudes here claims simply to honour St Bernard who, in the time of Suger, had come to the aid of the abbey on many occasions (‘qui in multis necessitatibus eidem Abbati [Suggerio] et ecclesiae nostrae subvenit’; Ibid., no. clxx). The Sandionysians did not, however, follow Eudes’ instructions to the letter. Although the abbot placed the feast on 25 August, the calendars from the abbey transmit 23 August.

28 The saint's name is none the less added later to the calendar.

29 Fol. 120.

30 The dean of the chapter of Meaux granted some relics of these saints to Abbot Matthew of Vendôme of St-Denis. His letter of 1259 is preserved in Félibien, , Histoire, Pieces justificatives i, no. clxxvi.Google Scholar From this, we can assume that the celebration was instituted at Si-Denis in this year. F–Pm 526 omits the service entirely, F–Pn lat. 976 adds it in a later hand (fol. 138), and F–Pn lat. 1107 includes the ceremony in the original hand of both calendar and text.

31 Until now the date of F–Pn lat. 976 had been only roughly estimated as 'end of the thirteenth, beginning of the fourteenth century’ (Bibliothèque Nationale, Catalogue général des manuscrits latins, 6 vols., Paris, 1939–1975, I, p. 348Google Scholar). Samaran omitted the manuscript altogether from his Catalogue (op. cit.), and A. G. Martimort accepted the date of the catalogue of the Bibliothèque Nationale in La documentation liturgique de Dom Edmond Martène: étude codicologique, Studi e Testi 279 (Vatican City, 1978), p. 543.Google Scholar

32 Earlier authors had suggested a date around 1254, based on the obit of Abbot William of Macouris (d. 1254) in the original hand of the calendar (see Leroquais, V., Les sacramentaires et les missels manuscrits des bibliothèques publiques de France, 4 vols., Paris, 1924, II, p. 140Google Scholar; also Bibliothèque Nationale, Catalogue, i, p. 404Google Scholar). Robert Branner places F–Pn lat. 1107 into a group of manuscripts whose characteristic features of illumination appear to have developed in the 1260s (Branner, , The Manuscript Painting in Paris during the Reign of St. Louis, Berkeley, 1977, pp. 130–2Google Scholar). In terms ofits liturgical presentation, the missal seems slightly earlier than I–Rc 603, which was copied after 1271 (see Walters, , ‘Music and Liturgy’, pp. 283–4Google Scholar). It is likely that F–Pn lat. 1107 was written between 1259 and about 1275.

33 F–Pm 526 is the only extant manuscript from the abbey which includes the ceremony In invencione sacri clavi (fols. 60–66v). This celebration on Friday of Easter week verifies that the recovery of the sacred relic of the Passion was the object of a short-lived service. The history of the event, which took place at St-Denis in 1233, is preserved in F–Pn n.a.1. 1509, and published in Aubry, P., ‘Comment fut perdu et retrouvé le saint clou de l'abbaye de Saint-Denys’, Revue Mabillon, 2 (1906), pp. 185–92, 286–300; 3 (1907), pp. 43–50, 147–82.Google Scholar Felibien mentions that Philip, chancellor of Paris, wrote a contemporary account of the Inventio (Felibien, , Histoire, pp. 228–32Google Scholar). It has been suggested that Philip was also responsible for the monophonic conductus Clavus clavo retunditur, which commemorates the finding of the nail (see Falck, R., The Notre Dame Conductus: a Study of the Repertory, Musicological Studies 33, Henryville, Pennsylvania, 1981, p. 111Google Scholar; and Aubry, , ‘Un chant historique latin du X111e siècle: le saint clou de Saint-Denys [1233]’, Bulletin François de la S.I.M., 1, 1905, pp. 430–4.Google Scholar Ruth Steiner discusses the rhythm of the piece in Some Monophonic Latin Songs from the Tenth Fascicle of the Manuscript Florence, Biblioteca Laurenziana, Pluteus 29.1, Studies in Music 16, Washington, D.C., 1963, pp. 53–6.Google Scholar) It should be borne in mind, however, that the monks of St-Denis contrafacted a chant for the occasion, the responsory Clavus refulgens (see Walters, , ‘Music and Liturgy’, pp. 328–9Google Scholar). In addition, the activities of a monk-trouvère as well as a noted cantor in the abbey during the 1230s offer other possible candidates for the authorship of Clavus clavo retunditur (Ibid., pp. 537–8).

34 The feast of Cosmas and Damian (27 September) appears at the rank of three lessons in the original hand of F–Pm 526 (fol. 175). A later scribe wrote ‘fiant xii lectiones de pluribus martyribus’ in the margin and replaced the number ‘tres’ with ‘xii’ in the text. F–Pn lat. 976 includes the rank of ‘xii lectiones’ in the original text (fol. 129v).

35 St Firminus was the principal honouree of the northernmost altar in the chevet (Figure l, no. 1; also Bruzelius, op. cit., Figure no. 26). In addition, however, the oratorium was dedicated to the angels, Matthew, Urban, Alexander, Eventius and Theodore, Clarus, Thomas, Dionysius of Corinth, Eligius, Lupus, Ouen, Anianus, Evortius, Columba, Faith, Thecla and Anastasia (see F–Pn lat. 976, fol. 157; Delisle, L., ‘Notice sur un livre à peintures exécuté en 1250 dans l'abbaye de Saint-Denis’, Bibliothèque de l'École des Charles, 38, 1877, p. 464Google Scholar; also Walters, , ‘Music and Liturgy’, Table vi-3, pp. 396–402Google Scholar).

36 A twelfth-century pontifical of the usage of Rouen, F-Pan n.a.l. 306, was adapted for processional use on the major feasts at St-Denis (see Leroquais, , Les pontificaux manuscrits des bibliothèques publiques de France, 3 vols., Paris, 1937, II, pp. 220–9Google Scholar).

37 ‘Si xii lectiones contingerint erit processio in capicio de festo contingente si altare consecratum fuerit in honore sancti’; F-Pm 526, fol. 78v; F–Pn lat. 976, fol. 48.

38 On the feast of St Augustine (28 August), the ordinaries state: ‘post magnificat fiat processio ad oratorium sancti Peregrini et post matutinas similiter cum responsorio Justum deduxit vel quolibet alio’; F–Pm 526, fol. 163; F–Pn lat. 976, fol. 121.

39 This increase is evident from the comparison of the processions of F–Pm 526 with those of F–Pn lat. 976 (see Walters, , ‘Music and Liturgy’, Tables vi-2 and vi-4, pp. 386–95, 403–8Google Scholar).

40 On the feast of St Germanus of Paris (28 May), a later hand writes: ‘fiat processio ad oratorium sancti mauricii’; F–Pm 526, fol. 129v.

41 Since F–Pm 526 was written at the outset of the reconstruction, it is possible that some of the processions which existed before the work began were temporarily suspended and then resumed in the 1240s.

42 Ibid., fol. 175.

43 Ibid., fol. 96v.

44 See Panofsky, , Abbot Suger, p. 118.Google Scholar

45 F–Pn lat. 976, fols. 157–8.

46 The processions in F–Pm 526 show that the altars of Benedict and Egidius in the crypt were still given some use (on 10 February, 11 July and 1 September). Only the latter was employed by the time F-Pn lat. 976 was written, however. The major problem encountered in comparing Suger's report with the one in F–Pn lat. 976 is that it is impossible to determine which of the secondary saints listed in F–Pn lat. 976 were also honoured at the altars of the chevet in Suger's time. In his account of the dedication of three chapels of the west façade in 1140, the abbot verifies that ‘subdedicatees’ were present at the altars of Mary, St Romanus and St Bartholomew (see Panofsky, , Abbot Suger, pp. 96–8Google Scholar). For the consecration of 1144, however, he provides only the name of the principal saint to be reverenced at each altar in the chevet, except in the case of St Walaburga (Ibid., p. 118). This he probably does simply in the interest of brevity, since it is very likely that more than one saint was honoured at each oratorium.

47 Histoire, pp. 535–6. The actual inscriptions have not survived, probably, as Delisle believed, because they were written on parchment (see Delisle, , ‘Notice’, p. 448Google Scholar). There is no way of determining their age.

48 Delisle did not place much confidence in the dates given on Felibien's inscriptions, which he considered to have been added after the thirteenth century (Ibid.). However, his preference for the list in F–Pn lat. 976 is not entirely reasonable, for in some cases it is clearly not contemporaneous with the rest of the manuscript. To cite an example: while the text of F–Pn lat. 976 calls for processions to the chapel of St Osmanna for the feasts of St Agnes (21 January) and the Ten Thousand Martyrs (22 June), the names of these saints do not figure in the list of consecrationes at the end of the ordinary. They do, however, appear in Félibien's inscriptions.

49 F–Pan ll 1157, pp. 85–7, 91–3; ll 1159, pp. 54–6, 56–8. These anniversaries honoured former abbots of St-Denis.

50 The data that follow are derived from Walters, , ‘Music and Liturgy’, Tables vi-2 and vi-4 (pp. 386–95, 403–8).Google Scholar

51 See also Bruzelius, op. cit., Figure no. 26.

52 Altar no. 3 (see Figure 1) was dedicated to St Maurice at some time in the thirteenth century. Under Suger's abbacy, however, the chapel had been consecrated to St Eustace in 1144. It is likely then that the diverted processions in F–Pm 526 refer to what was later the altar of St Maurice (Figure 1, no. 3). Prior to Suger, the oratorium of Eustace stood in the southern portion of the twelfth-century transept (see Panofsky, , Abbot Suger, pp. 155. 157–8).Google Scholar

53 See Bruzelius, op. cit.

54 ‘Post terciam fiat processio in cappis ad oratorium sancti Johannis … Descendendo de capicio et non eant in navim sed ante passum per rubeas portas in choro intrent.’ F–Pm 526, fol. 137v; F–Pn lat. 976, fol. 99v. No doubt the altar of St John is the one which was consecrated in his honour by Suger in 1144 and rededicated to Romanus of Blaye in the thirteenth century (see Figure 1, no. 9; also Walters, , ‘Music and Liturgy’, pp. 390, 401).Google Scholar

55 ‘ante missam fiat processio … descendendo de capicio … cum venerint ad passum eant per rubeas portas in choro’; F–Pm 526, fol. 17v.

56 If the procession for John the Baptist occurred on Sunday, then both the cloister and the nave were employed: ‘Et si dominica fuerit fiat processio per claustrum … et redeant in capicio per portam eburneam … Et descendentes de capicio eant in navim … et sic redeant in choro’; F–Pm 526, fol. 137v; F–Pn lat. 976, fols. 99v–100. The ‘ivory door’ (eburnea porta) which is mentioned in this quotation is probably the one which Suger names in his De consecratione (Panofsky, , Abbot Suger, pp. 116, 117Google Scholar). Panofsky speculates on the two possible locations of this door: it led either from the southern transept wing into the cloister, or from the ‘confessio’ into the crossing; Ibid., p. 248. The references to the door in the ordinaries would seem to support Panofsky's first hypothesis. If, however, the eburnea porta was an entrance into the crossing, it would probably have been from the chevet instead of from the ‘confessio’.

57 The later ordinary in fact suggests that the non-dominical procession for John the Baptist entered the choir from the north side: ‘fiat processio ad oratorium sancti Romani … Descendendo de capicio et non eant in navim, sed dum pertransierint ad oratorium sancti Hippoliti [located presumably in the first chapel to the west of the north transept] intrent in chorum’; F-Pan l 863, no. 10, fol. 106.

58 Louis worshipped in the chapel frequently during his visits to St-Denis, and the care of its liturgical books is attested in records of the abbey (see Ayzac, , Histoire, II, pp. 189–94Google Scholar).

59 Fol. 36.

60 F–Pn lat. 976, fol. 12; F-Pan l 863, no. 10, p. 32. The latter manuscript is described in Table 6.

61 See Walters, ‘Music and Liturgy’, chapter 1, ‘The Merovingian Period’.

62 Ibid., pp. 50–7.

63 Several sequences appear to have been contrafacted at St-Denis for use on feasts of saints whose relics were present in the church (Walters, , ‘Music and Liturgy’, pp. 505–7Google Scholar). See the discussion of the melismatic Benedicamus Domino and Ite missa est, whose melodies were drawn from responsories, antiphons, kyrie tropes, alleluia verses and sequences. Many of these chants were proper to the Sandionysian liturgy (Ibid., pp. 507–19). Frank Harrison has noted a similar series of Benedicamus Domino settings in a mid-thirteenth-century Sarum missal (see Harrison, , ‘Benedicamus, Conductus, Carol: a Newly-discovered Source’, Acta Musicologica, 37, 1965, pp. 35–48CrossRefGoogle Scholar; idem, Music in Medieval Britain, London, 1958, pp. 74–5; and Walters, , ‘Music and Liturgy’, pp. 507–19).Google Scholar

64 Ibid., pp. 95–7. The rationale behind this apparent conservatism will be treated below. Despite the monks' refusal to adopt polyphony, their liturgy contained analogous musical features. While the cathedral of Notre Dame employed two- and three-voice settings of the Benedicamus Domino, for example, St-Denis used the elaborate monophonic settings mentioned above.

65 Ibid., pp. 164–76, 524–31.

66 A cleric named Vandelmarus, who instructed the members of the royal chapel during the ninth century, received his musical training at St-Denis from a magister Teutgarius. In 1239 the cantor of St-Denis, Guillelmus, figured prominently in the musical portion of the ceremonies that accompanied the arrival of the Crown of Thorns from Constantinople. Around the same time, an anonymous monk-trouvère from the abbey composed three songs which survive in the chansonniers from the period. One of his works, En non Dieu, c'est la rage, is also found as the superius of a motet with the same title in the Notre Dame polyphonic sources. The study of music theory at St-Denis in the late thirteenth and early fourteenth centuries is attested in a plainchant treatise by ‘frater Guido monachus monasterii sancti dionysii in francia’ (treated below), and possibly in a mensural treatise by ‘Petrus de sancto Dionysio’ (ed. Michels, U., Johannis de Muris Notitia artis musicae, Compendium musicae practical, Petrus de Sancto Dionysio tractatus De musica, Rome, 1972Google Scholar). On all of these musicians, see Walters, , ‘Music and Liturgy’, pp. 531–42.Google Scholar The ceremonial modifications of the thirteenth-century abbots Eudes Clément and Matthew of Vendome have been mentioned. Suger, too, was active as a liturgist prior to and during the twelfth-century reconstruction of the church (Ibid., pp. 100–3).

67 The Abbey of St-Corneille-de-Compiègne was reformed by Abbot Suger around 1150. Extant liturgical books from this establishment demonstrate conclusively that Suger imposed Sandionysian musical and ceremonial customs there (Ibid., pp. 103, 305–6, 312– 13, 515). Furthermore, the monks of Solesmes have noted an affinity between the thirteenth-century Worcester Antiphonal (Paléographie Musicale, ser. i; 12, Tournai, 1922) and the graduals and notated missals from St-Denis with respect to melodic variants in chants of the Mass (Le graduel romain, 4 vols., Solesmes, 1957–1960, iv/1, pp. 245, 261Google Scholar and passim). Hesbert likewise demonstrates that the ordering of representative chants for the office in the two houses is quite similar (see Hesbert, R.-J., Corpus antiphonalium officii, 6 vols., Rerum Ecclesiasticarum Documenta, series major, Rome, 1963–1979, v, pp. 431–2, 440Google Scholar). The level at which these institutions may have influenced one another has not yet been determined, but it is noteworthy that St-Denis numbered among its foreign possessions the priory of Deerhurst near Worcester from the eleventh century to the mid-fifteenth (see Knowles, D. and Hadcock, R. N., Medieval Religious Houses: England and Wales, London, 1971, p. 64Google Scholar). A trafficking of liturgical books over a long period of time could have been occasioned by this association. The editors of the Worcester Antiphonal determine that the manuscript preserves a pre-Conquest chant tradition which may be close to that of Corbie (Knowles, , The Monastic Order in England: a History of its Development from the Time of St. Dunstan to the Fourth Lateran Council, 943–1216, Cambridge, 1950, pp. 552–4Google Scholar). St-Denis, too, had close ties with Corbie in the early centuries.

68 Hilduin's history of Dionysius the Areopagite is published in Migne, J.-P., Patrologia latina, cvi (Paris, 1864), cols. 23–50.Google Scholar The basic studies of the profoundly important theory of Areopagitism at St-Denis include Théry, G., ‘Contribution à l'histoire de l'aréopagitisme au IXe siécle’, Le Moyen Age, 25 (1923), pp. 111–53Google Scholar; and his Études dionysiennes, i: Hilduin, traducteur de Denys and ii: Hilduin, traducteur de Denys: éedition de sa traduction, Études de Philosophic Médiévale 16 and 19 (Paris, 1932–1937).Google Scholar For a rapid summary of the history of the Areopagite controversy at St-Denis, see Spiegel, , Chronicle, pp. 23–4, footnote 43.Google Scholar

69 See the translation of this work in Campbell, T. L., Dionysius the Pseudo-Areopagite: the Ecclesiastical Hierarchy (Lanham, Maryland, 1981)Google Scholar; and commentaries in Rutledge, D., Cosmic Theology: the Ecclesiastical Hierarchy of Pseudo-Denys: an Introduction (London, 1964)Google Scholar; and Hathaway, R. F., Hierarchy and the Definition of Order in the Letters of Pseudo Dionysius: a Study in the Form and Meaning of the Pseudo-Dionysian Writings (The Hague, 1969).CrossRefGoogle Scholar The Byzantine emperor Michael the Stammerer had sent a Greek manuscript (now F–Pn grec. 437) of this treatise to Louis the Pious in 827.

70 The proper and ordinary chants which were executed in Greek, as well as the transformation of the service that occurred in the seventeenth century, are discussed in Huglo, ‘Chants’. For further details and some recent findings, see Walters, , ‘Music and Liturgy’, pp. 271, 274–6, 471–7.Google Scholar Portions of the ordinary in Greek were performed at St-Denis and throughout the West on high feasts as early as the eighth century (see Atkinson, C., ‘O Amnos Tu Theu: the Greek Agnus Dei in the Roman Liturgy from the Eighth to the Eleventh Century’, Kirchenmusikalisches Jahrbuch, 65, 1981, pp. 7–30Google Scholar; and idem, ‘Zur Entstehung und Uberlieferung der “Missa graeca” ’, Archiv für Musikwissenschaft, 39, 1982, pp. 113–45).

71 The order of feasts in Table 7 follows that given in F–Pn lat. 976, fol. 159.

72 The contents of this manuscript are highly abbreviated (see Walters, , ‘Music and Liturgy’, pp. 339–44).Google Scholar

73 Amédée Gastoué has noted the similarity between the rubrics ‘ante’ and ‘ad’ evangelium and the respective functions of these chants (Gastoué, , ‘Le chant gallican’, Revue du Chant Gregorien, 42, 1938, p. 11Google Scholar).

74 Hodie completi sunt, for example, serves as antiphon to the Magnificat at second Vespers in the monastic sources R, D and S in Hesbert's Corpus (ii, no. 95d). in the secular source B (Ibid., i, no. 95b), and as antiphon to the Benedictus at Lauds in the secular source E (Ibid.).

75 See the outlines for the Order of Mass in both rituals in Anglès, H., ‘Latin Chant Before St. Gregory’, Early Medieval Music up to 1300, ed. Hughes, Anselm, New Oxford History of Music, II (London, 1954), pp. 67, 76–7.Google Scholar Other readily accessible descriptions of the Gallican ordo include Stablein, B., ‘Gallikanische Liturgie’, Die Musik in Geschichte und Gegenwart (abbreviated MGG), ed. Blume, F., 16 vols. (Kassel, 1949–1979), IV, cols. 1299–1325Google Scholar; and Huglo, , ‘Gallican Rite, Music of the’, The New Grove Dictionary of Music and Musicians, ed. Sadie, S., 20 vols. (London, 1980), vii, pp. 113–25.Google Scholar

76 ‘The Antiphonae ante evangelium in North-Italian Sources’, paper read at the International Congress on Medieval Studies, Kalamazoo, Michigan (05 1982).Google Scholar

77 Hodie Maria virgo for Assumption and Verbum caro for Christmas were used in both rituals.

78 Kenneth Levy has recently reviewed these reforms in ‘Toledo, Rome and the Legacy of Gaul’, Early Music History, 4 (1984), pp. 49–99.Google Scholar (I am grateful to Professor Levy for making his work available to me prior to its publication.) The documents in which Charlemagne himself mentions the importation of Roman cantus are his Admonitio generalis of 23 March 789 (Boretius, A., ed., Capitularia regum francorum, Monumenta Germaniae Historica, Legum Sectio 2, Hanover, 1883, p. 61Google Scholar) and his Epistola generalis (786–800; Ibid., p. 80). Further details are provided by two ninth-century authors: John the Deacon (Migne, , Patrologia latina, lxxv, Paris, 1862, cols. 87ffGoogle Scholar) and Notker Balbulus (Monumenta German-iae Historica, Scriptores 1, Hanover, 1839, pp. 734ff). The discrepancies between these two later accounts are treated in Van Dijk, S.J. P., ‘Papal Schola versus Charlemagne’, Organicae voces: Festschrift Joseph Smits van Waesberghe (Amsterdam, 1963), pp. 21–30.Google Scholar

79 The letters of Pseudo-Germanus are found in Migne, , Patrologia latina, LXXII (Paris, 1878), cols. 89–98Google Scholar; and more recently in Gamber, K., ed., Ordo antiquusgallicanus, Textus Patristici 3 (Regensburg, 1965)Google Scholar; and Ratcliff, E. C., ed., Expositio antiquae liturgiae gallicanae (London, 1971).Google Scholar

80 ‘De Aius ante Evangelium. Tunc in adventum sancti evangelii claro modolamine denuo psallit clerus Aius in specie angelorum ante faciem christi ad portas inferni clamantium Tollite portas principes vestras et elevamini porte aeternales et introibit dominus virtutum rex gloriae. De Evangelic Egreditur igitur processio sancti evangelii velud potentia christi triumphantis de morte cum praedictis harmoniis’; Ibid., p. 7. Although Tollite portas is not rubricated, there can be little doubt that an antiphon is implied. Clearly, this was thought to have been the chant intended in the later Middle Ages.

81 It should be emphasised, however, that the criteria are hard to pinpoint and often seem contradictory. For this reason it is difficult to make a comprehensive statement about the presumed stylistic characteristics of this repertory. At the turn of the century, Louis Duchesne noted the elaborateness of Gallican chant-texts (Duchesne, , ‘Sur 1'origine de la liturgie gallicane’, Revue d'Histoire et de Littérature Religieuses, 5, 1900, p. 37Google Scholar), and Michel Huglo has recently summarised his own observations and those of others, affording a general impression of what this body of chant might have been like. The discussion in this paragraph is drawn largely from his ‘Gallican Rite’, pp. 114–17.

82 ‘Toledo’, pp. 51ff. Levy focuses on the non-psalmic texts of eighteen offertories, suggesting that these ‘librettos’ resemble texts which originated in certain areas of Gaul and were transported into other ancient rituals. He hypothesises that the Gallican music of these and other offertories may hence be represented entirely or in part in the Mozarabic, Milanese and even Gregorian melodies that have survived.

83 On the execution of annuale and semiannuale feasts in the monastery, see Walters, , ‘Music and Liturgy’, pp. 524–31.Google Scholar

84 The importance of Dagobert is explained below. The history of the funeral of Philip Augustus, recorded by the chronicler Guillaume Le Breton, is more detailed than that of any other king who was buried at St-Denis in the Middle Ages (see Delaborde, H.-F., Oeuvres de Rigordet de Guillaume le Breton, historiens de Philippe Auguste, 2 vols., Paris, 1882–1885, I, p. 325Google Scholar; also Walters, , ‘Music and Liturgy’, pp. 108–9Google Scholar).

85 Gastoué, , ‘Le chant gallican’, pp. 10–12.Google Scholar

86 Stablein, , ‘Gallikanische Liturgie’, cols. 1308–9, 1311, 1313Google Scholar; Hansen, F. E., The Grammar of Gregorian Tonality: an Investigation based on the Repertory in Codex H 159, Montpellier, 2 vols. (Copenhagen, 1979), I, pp. 198–202.Google Scholar

87 Gastoue alludes to the presence of antiphons in mass formularies of certain ‘anciens coûtumiers’, but he does not name specific manuscripts. Indeed, he states that these pieces doubled as antiphons to the Magnificat at the hour of first Vespers, which is almost never the case at St-Denis (Gastoué, , ‘Le chant gallican’, pp. 11–12Google Scholar; cf. Table 7 above).

88 F–Pm 526, fol. 140v; F-Pn lat. 976, fol. 102.

89 See Roederer, C., ‘Eleventh-century Aquitanian Chant - Studies relating to a Local Repertory of Processional Antiphons’, 2 vols. (Ph.D. dissertation, Yale University, 1971), I, pp. 92–3, and II, p. 152.Google Scholar The Aquitanian manuscripts are likely preservers of some Gallican material because they emanate from a region which was far removed from the centre of the eighth-century reforms. Stäblein speculates that much ‘gallikanisches Gut’ is to be found in processional antiphons (‘Gallikanische Liturgie’, col. 1307), and Huglo names the processional antiphon for Palm Sunday, Collegerunt V Unus autem, as a Gallican vestige (Huglo, , ‘Antiphon’, The New Grove Dictionary, I, p. 480Google Scholar).

90 Pseudo-Germanus mentions this procession in his description of the Gallican mass, cited above. The thirteenth-century ordinary from Bayeux makes it clear that the chant was processional in the later Middle Ages as well: ‘then let the deacon … intone the antiphon Christus natus est. When this has been initiated, let him immediately seek a blessing while taking up the gospel book, and let him go to the pulpit for reading. And when the antiphon has been sung to its end’ (‘tunc diaconus … incipiat Ant. Hodie Xpistus natus est que incepta, statim petat benedictionem sumens textum evangeliorum et eat in pulpitum ad legendum, antiphona finaliter decantata’; Chevalier, U., ed., Ordinaire et coutumier de l'eglise cathedrale de Bayeux, Bibliothèque Liturgique 8, Paris, 1902, pp. 62–3Google Scholar). The use of the antiphon before the gospel at Bayeux is treated below.

91 The relationship between Hodie illuxit and Salvator omnium had previously been noted by Gastoue, (‘Le chant gallican’, p. 12)Google Scholar, Stäblein (‘Gallikanische Liturgie’, cols. 1308–11) and Huglo, (‘Gallican Rite’, p. 122)Google Scholar, but without reference to the use of both melodies as gospel antiphons in the abbey. Stäblein is probably correct in opting for Salvator omnium as the original, and presumably Gallican, text because of the presence of the preces ‘dicamus omnes miserere’. These words are modified in Hodie illuxit to read ‘dicamus omnes alleluia’.

92 See Hesbert, , Corpus, III, p. 450 (no. 4689).Google Scholar

93 Paléeographie Musicale, ser. i, 16 (Solesmes, 1957), fols. 43, 108. See the discussion of the origin of this manuscript in Walters, ‘Music and Liturgy’, pp. 435–6.

94 F–Pn lai. 776, fol. 143v.

95 F–Pn lat. 903, fol. 145. (Fols. 1–133 are published as Paleographie Musicale, ser. i, 13, Tournai, 1925.)

96 F–Pn lat. 1120, fol. 176.

97 F–Pn lat. 1121, fol. 172.

98 F–Pm 526, fol. 137v; F–Pn lat. 976, fol. 100; F-Pan l 863, no. 10, fol. 106v.

99 The performance directions for this antiphon are stated below.

100 ‘Zwei Melodien der altirischen’, Musicae sciential collectanea: Festschrift Karl Gustav Fellerer zum siebzigsten Geburtstag am 7. Juli 1972, ed. Hüschen, H. (Cologne, 1973), pp. 595–7Google Scholar.

101 See Hesbert, , Corpus, III, p. 115 (no. 1951).Google Scholar

102 F–Pm 526, fol. 125.

103 See Walters, , ‘Music and Liturgy’, pp. 21–3.Google Scholar

104 The passion is found in Krusch, B., ed., Passio sanctorum martyrum Dionisii, Rustici et Eleutherii, Monumenta Germaniae Historica, Auctores Antiquissimi 4/ii (Berlin, 1885), pp. 101–5.Google Scholar Léon Levillain demonstrates that this text, composed in the fifth century, has come down in a version from the end of the eighth century (Levillain, , ‘études sur l'abbaye de Saint-Denis à l'époque mérovingienne’, Bibliothèque de l'École des Charles, 82, 1921, pp. 5–28).CrossRefGoogle Scholar

105 Epistolae Variorum, Monumenta Germaniae Historica, Epistolae 5/i: Karolini Aevi, iii (Berlin, 1898), pp. 330–1.Google Scholar Louis had previously asked Hilduin to compose a life of St Denis and to append some hymns and an office for the saint (‘una cum ymnis, quos de hoc gloriosissimo martire atque pontifice habes, et officium nocturnale subjugas’, Ibid., p. 327).

106 Migne, , Patrologia latina, cvi, cols. 23–50.Google Scholar

107 F–Pn lat. 17296, fols. 277v–229. See also Hesbert, Corpus, ii, no. 114a, b. Huglo notes that antiphons with verses also occur in the offices for the two feasts of St Paul and for St Lawrence. He suggests that the verses functioned like versus ad repetendum and compares them to the presumed Gallican antiphons with verses for Maundy Thursday (Huglo, , ‘Antiphon’, p. 478).Google Scholar

108 On the use of the neuma triplex in this responsory, see Steiner, R., ‘The Responsories and Prosa for St. Stephen's Day at Salisbury’, The Musical Quarterly, 56 (1970), p. 172Google Scholar; Handschin, J., ‘Trope, Sequence and Conductus’, New Oxford History of Music, II (1954), pp. 143–5Google Scholar; and Stablein, , ‘Tropus’, MGG, XIII, col. 812.Google Scholar

109 ‘Resp. Post passionem, Ver. Quicum imminere in cantu privato a duobus’; F–Pm 526, fols. 183r-v; F–Pn lat. 976, fol. 137.

110 Anton Baumstark notes that new liturgical features do not immediately supplant old ones; rather, they are first notated alongside the former usages (Baumstark, , Liturgie comparée: principes et méthodes pour l'étude historique des liturgies chrétiennes, 3rd edn, Chevetogne, 1953, p. 26).Google Scholar

111 Example 3 demonstrates that the text of O beate Dyonisi varied even within the sources from St-Denis. The word caritate appears only in the version from the office of the saint (F–Pn lat. 17296, fol. 232v), while qualitate is found in the feast of All Saints (fol. 240) and in the antiphona ante evangelium in GB-Lva 1346–1891 (see Table 7). Other antiphonals from the tenth to thirteenth centuries likewise record qualitate: Mont-Renaud (Paléographie Musi-cale, ser. i, 11, fol. 106v); F–Pn lat. 10244 (fol. 199) and lat. 12584 (fol. 333v) from St-Maur-des-Fosses; F–Pn n.a.l. 1535 (fol. 113) from Sens, and Paris, Bibliothèque Ste Geneviève, MS 1270 (fol. 275v) from Chelles. One wonders if caritate might be a trace of the Gallican text, preserved, like the second verse Quicum imminere, only in the Sandionysian office for the patron (F–Pn lat. 17296, fol. 232v).

112 F–Pn lat. 12044, fol. 154. This manuscript and another St-Maur antiphonal, F–Pn lat. 12584, are examined in Renaudin, A., ‘Deux antiphonaires de Saint-Maur’, Etudes Gregoriennes, 13 (1972), pp. 53–150.Google Scholar

113 Fol. 348v.

114 Guy's treatise, as yet unedited, is contained in London, British Library, Harleian MS 281, fols. 55v–96. The work comprises two parts, theoretical and practical, and dates from between about 1299 and 1318. It is a disquisition on plainchant, drawing on Guido of Arezzo, Johannes de Garlandia and Petrus de Cruce. The author shows his familiarity with Parisian practice by distinguishing the usage of the abbey (‘secundum usum nostrum’) from the customs of the cathedral. He also describes the performance of textless melismas, or neumae, in sequences (see Huglo, , ‘Guy de Saint-Denis’, The New Grove Dictionary, VII, p. 859Google Scholar; and Walters, , ‘Music and Liturgy’, pp. 338–40).Google Scholar

115 Fol. 81v.

116 Ibid.

117 Ibid., fol. 82.

118 Hucke, , ‘Toward a New Historical View of Gregorian Chant’, Journal of the American Musicological Society. 33 (1980), p. 442.CrossRefGoogle Scholar The author first enunciated his ideas in detail in ‘Die Einfuhrung des Gregorianischen Gesangs im Frankenreich’, Römische Quartalschrift, 49 (1954), pp. 172–85.Google Scholar

119 See Marossezéki, S. R., Les origines du chant cistercien: recherches sur les réformes du plain-chant cislercien au XIIe siècle, Analecta Sacri Ordinis Cisterciensis 8/I–II (Rome, 1952), especially pp. 70–3Google Scholar, where the amputation of long vocalises is discussed; also Waddell, C., ‘The Origin and Early Evolution of the Cistercian Antiphonary: Reflections on Two Cistercian Chant Reforms’, The Cistercian Spirit: a Symposium, ed. Pennington, M. B. (Spencer, Massachusetts, 1970), pp. 190–223.Google Scholar

120 John states that Charlemagne assigned two clerics to Metz. These instructors raised the level of singing there, and by their example, all ofGaul was corrected (‘totam Galliam suam [Carolus] correxit’); Migne, , Patrologia latina, lxxv, col. 91.Google Scholar A later chronicler, Ademar de Chabannes, records that Gallican antiphonals were corrected as well: ‘Correcti sunt ergo antiphonarii Francorum quos unusquisque pro arbitrio suo viciaverat vel addens vel minuens’; Chavanon, J., ed., Adémar de Chabannes: Chronique (Paris, 1897), p. 82.Google Scholar Van Dijk dismisses the latter report as unwarranted embellishment of John's text, especially since it is unlikely that the antiphonals of the late eighth century were notated (Van Dijk, , ‘Papal Schola’, p. 28Google Scholar). The idea of ‘correction’ is common to both texts, however, even if Adémar did not realise that the revisions were of oral and not of written practice.

121 Huglo, , Les tonaires: inventaire, analyse, comparaison (Paris, 1971).Google Scholar See also Hucke, , ‘Toward a New Historical View’, pp. 443, 465.Google Scholar

122 This theory would be the Frankish corollary to Levy's suggestion that the urban Roman offertories based on psalmic texts ‘would have gone north during the seventh to ninth centuries, there to be tailored into transalpine musical style’ (Levy, , ‘Toledo’, p. 96).Google Scholar

123 Even before the practice of perpetual psalmody was instituted in the seventh century, the monks appear to have adopted the ritual which the Council of Tours handed down in 567 (see Walters, , ‘Music and Liturgy’, pp. 32–4Google Scholar). This Gallican ordo for the arrangement of psalms and antiphons in the office hours is published in Maassen, F., ed., Concilia aevi merovingici, Monumenta Germaniae Historica, Legum Sectio 3: Concilia, i (Hanover, 1893), p. 127Google Scholar; and in De Clercq, C., ed., Concilia Galliae A. 511–A. 695, Corpus Christianorum, Series Latina 148A (Turnhout, 1963), pp. 182–3.Google Scholar

124 The library was already famous in the ninth century and maintained its high standing until 1567, when Huguenot intruders dispersed many of its holdings. Despite this catastrophe, the abbey still possessed many precious books at the outset of the French Revolution. A thorough examination of this celebrated establishment is forthcoming in D. Nebbiai-Dalla Guarda, La bibliotheque de l'abbaye de Saint-Denis-en-France du IXe au XVIIle siècle. See also the descriptions of manuscripts in É. Lesne, , Histoire de la propriété écclésiastique en France, iv: Les livres, scriptoria et bibliothèques du commencement du VIIIe á la fin du Xle siècle (Lille, 1938)Google Scholar; and in de Montesquiou-Fezensac, B., Le trésor de Saint-Denis: inventaire de 1634, 3 vols. (Paris, 1973–1977).Google Scholar The destruction wrought by the Huguenots is treated in Carolus-Barré, L., ‘Pillage et dispersion de la bibliotheque de l'abbaye de Saint-Denis, ler octobre–10 novembre 1567’, Bibliothèque de l'École des Chartes, 138 (1980), pp. 97–101.CrossRefGoogle Scholar

125 1–Rvat 257. This early eighth-century mass book bears two shelfmarks which demonstrate that it was housed in the St-Denis library in the thirteenth and fifteenth centuries (see Wilmart, A., Codices reginenses latini, 2 vols., Vatican City, 1945, II, p. 23Google Scholar). Although the later provenance of the codex is St-Denis, there is no direct evidence to suggest the abbey as scriptorium or original destination (see Walters, , ‘Music and Liturgy’, p. 65Google Scholar). The manuscript may none the less have been consulted by the monks during the later Middle Ages.

126 On Suger's Vita Ludovici Grossi and his histories of Louis vii, see Spiegel, , Chronicle Tradition, pp. 45–52.Google Scholar

127 Three early twelfth-century works, the Gesta gentis Francorum, the Nova gesta Francorum and the Abbreviatio gestorum Franciae regum, are written in the form of universal chronicle. They stress the supposed origins of the French nation in the race of Trojan heroes (Spiegel, , Chronicle Tradition, pp. 40–4Google Scholar). This interest in ancient learning which arose in the twelfth century is treated in Haskins's, C. H. classic study, The Renaissance of the Twelfth Century (New York, 1960).Google Scholar

128 St-Denis is considered as one of the possible places of origin of Le pèlerinage de Charlemagne (see Bates, R. C., ‘Le pèlerinage de Charlemagne: a Baroque Epic’, Yale Romanic Studies, 18, 1941, p. 21Google Scholar; and Adler, A., ‘Pelerinage de Charlemagne in a New Light on Saint-Denis’, Speculum, 22, 1947, pp. 550–61CrossRefGoogle Scholar). The word geste, apart from signifying ‘deeds’, conveys the notions of family and of lineage, as is evident in the chanson de geste, of Jourdain of Blaye: ‘Huimais orréz avant de lor lingnie/ Et de la geste qui des barons issirent’ (11. 8, 9); and ‘Je voz affi qu'il est de Tranche geste’ (1. 1525); see Dembowski, P., ed., Jourdain de Blaye: Chanson de geste (Chicago, 1969).Google Scholar (I am grateful to Professor Dembowski for his thoughts on this work.) R. H. Bloch demonstrates the ways in which the idea of lineage serves as the structural underpinning for the chanson de geste (Bloch, , Etymologies and Genealogies: a Literary Anthropology of the French Middle Ages, Chicago, 1983, pp. 92–4Google Scholar).

129 See Walters, ‘Music and Liturgy’, chapter 3, ’The Calendar of Feasts and Anniversaries’.

130 Ibid., pp. 116–17, 164–78. See also Barroux, R., ‘L’anniversaire de la mort de Dagobert à Saint-Denis au XIIe siècle: charte inéditéde l'abbé Adam’, Bulletin Philologique et Historique (1942–1943), pp. 131–51.Google Scholar

131 Hesbert points out that the use of the word natale in this rubric reveals the monks’ would-be portrayal of Dagobert as a saint (Hesbert, , Corpus, II, p. xivGoogle Scholar).

132 ‘Deus omnipotens qui electis tuis in agone certantibus gloriae coronam largiris et sortite [sic] premia regni celestis tradidisti inter quos athletam tuum dyonisium atque eleutherium seu rusticum in tua confessione palmam [sic] martirii consecrasti, presta, redemptor mundi, ut quod meritis non valemus per eorum suffragia veniam consequamur.’ The gradual has recently been published in fascimile edition in Hesbert, , Le graduel de St. Denis, Monumenta Musicae Sacrae 5 (Paris, 1981)Google Scholar. See pp. xxiv, 234–5 (F–Pm 384, fol. 134).

133 See Hesbert, , Corpus, I, no. 114b; III, p. 145 (no. 2179).Google Scholar The antiphon O beate Dyonisi is also found in this series.

134 Hesbert, , Graduel, p. 233Google Scholar; F–Pm 384, fol. 134.

135 See Levillain, , ‘Études’, pp. 22–3.Google Scholar The author cites documents from the Merovingian era to demonstrate this point: an original charter from Clovis ii, dated 22 June 653 (F-Pan K. 2, no. 3; published in Tardif, , Monuments, p. 10Google Scholar); and the martyrology of St Jerome (see Delehaye, H., ed., Martyrologium Hieronymianum, Acta Sanctorum: Novembris, ii/2, Brussels, 1931, p. 547Google Scholar).

136 As yet I have been unable to locate a sacramentary source for this prayer-antiphon, hence it is possible that the text could have been written in the ninth century. If this is the case, I would suggest that it was composed from a Merovingian model, in which the names of the saints appear in the older order.

137 Bibliothèque Nationale, Catalogue, II, pp. 105–6.Google Scholar On the St-Cyprien text, see Petit-mengin, P., ‘Cinq manuscrits de Saint Cyprien et leur ancêtre’, Revue d'Histoire des Textes, 2 (1972), pp. 198–201Google Scholar; and idem, ‘Un monument controversé, le “Saint Cyprien” de Baluze et Dom Moran’, Ibid., 5 (1975), pp. 99, 101, 126.

138 Fol. 262.

139 The neumes in fact help establish the probable location of the book at St-Denis in the eleventh century, for the shelfmark of the library of the abbey (fol. 2) dates from the fifteenth century. Jean Vézin believes that the codex was copied in the late ninth or early tenth century, probably in Tours, and that it was at St-Denis by the eleventh century. (I am grateful for Professor Vézin's remarks to me on the provenance of this manuscript.)

140 The selection of tropes and sequences, on the other hand, fluctuated from one book to another at St-Denis.

141 On O virgo virginum, see Constantinescu, R., ‘Alcuin et les “Libelli Precum” de l'époque carolingienne’, Revue d'Histoire de la Spiritualité, 50 (1974), pp. 40–1.Google Scholar The Hodie-antiphons and their possible Byzantine connections are treated in Baumstark, , ‘Die Hodie-Anti-phonen des römischen Breviers und der Kreis ihrer griechischen Parallelen’, Die Kirchen-musik, 10 (Paderborn, 1909), pp. 157, 158Google Scholar; and in Huglo, , ‘Relations musicales entre Byzance et l'occident’, Proceedings of the XIIIth International Congress of Byzantine Studies, Oxford, 1966, ed. Hussey, J. M., Obolensky, D. and Runciman, S. (London, 1967), p. 273.Google Scholar

142 St Cuthbert, for example, mentions that Bede sang the antiphon O rex gloriae from his deathbed on the feast of the Ascension, 26 May 735 (see Arnold, T., ed., Symeonis monachi opera omnia, Historia Dunhelmensis Ecclesiae, I, London, 1882, p. 44Google Scholar; also Constantinescu, op.cit., p. 48).

143 Chevalier, op. cit., pp. 62, 69, 71, etc.; and Delaporte, Y., L'ordinaire chartrain du XIIIe siède, Societe Archéologique d'Eure-et-Loir, Mémoires 19 (Chartres, 1953), pp. 86, 89, 113, etc.Google Scholar

144 My article on the antiphons of these and several other establishments is forthcoming.

145 On Easter, the ordinaries state: ‘Ad finem cuius [sequentiae] transmittal cantor quem-dam bene cantantem in capicio ante martyres qui honeste ante evangelium intonet antiphonam Crucem sanctam subiit’; F–Pm 526, fol. 57v; F–Pn lat. 976, fol. 30v.