¹ See Wegman, R. C., ‘New Data Concerning the Origins and Chronology of Brussels, Koninklijke Bibliotheek, Manuscript 5557’, Tijdschrift van de Vereniging voor Nederlandse Muziekgeschiedenis, 36 (1986), pp. 5–21CrossRefGoogle Scholar.

² Trent, Museo Diocesano, MS BL (Trent 93) and Trent, Castello del Buonconsiglio, MS 90. For a detailed inventory of the Brussels manuscript, see Curtis, G. R. K., ‘The English Masses of Brussels MS. 5557’ (Ph.D. dissertation, University of Manchester, 1979), pp. 6–7Google Scholar. For a more cursory inventory and details of manuscript structure, see Wegman, ‘New Data’, p. 20.

³ Similarities between this work and the anonymous mass have led Rob Wegman and Gareth Curtis to speculate independently that the mass may be by Frye. See, most recently, Wegman's review of Curtis's edition, The Brussels Masses, Early English Church Music 34 (London, 1989) in Early Music, 17 (1989), pp. 584–7Google Scholar.

⁴ These occur in the usual positions as follows: Sanctus: ‘Pleni sunt celi ∥ et terra ∥ gloria tua’ ∥, ‘Benedictus [rests] qui venit ∥ in nomine domini’ ∥; Agnus II: ‘Agnus dei [rests] qui tollis peccata mundi ∥ miserere nobis’∥. Though this practice is not unheard of outside English masses, its occurrence elsewhere is restricted to early Continental cycles showing clear English influence as, for instance, Domarto's Missa Spiritus almus, Simon de Insula's Missa O admirabile commercium (Trent, Castello del Buonconsiglio, MS 88 (hereafter Trent 88), fols. 304^v–;311^r) and Ockeghem's Missa Caput. Even in these works, however, its use is not as consistent and regular as in most English cycles.

⁵ This practice, though sometimes without the final return to tempus perfectum, is almost ubiquitous in English Agnus settings. Occasionally, as in the anonymous Missa Quem malignus spiritus and in Bedyngham's Missa Deul angouisseux, the first mensural change occurs at the second setting of ‘miserere’, a procedure also followed in a few Continental cycles as, for instance, in the masses O admirabile commercium and Grüne Linden (Trent 88, fols. 375^v–383^v). However, the usual Continental practice of changing mensuration at Agnus II is almost unheard of in English settings.

⁶ While full texting in lower voices, except in apparently faithful copies of English works, is rare in Continental manuscripts, English practice generally favours texting of contratenors and fluid tenor lines. This tendency has been noted by Planchart, Alejandro, ‘Fifteenth-Century Masses: Notes on Performance and Chronology’, Studi Musicali, 10 (1981), p. 29Google Scholar, and Curtis, , The Brussels Masses, p. xivGoogle Scholar.

⁷ This can be seen in most insular copies of English Sanctus settings, and in Continental ones which give reason to suspect closeness to English models. Many English copies, and some Continental ones, supply an incipit, though this is usually omitted in Continental copies, and the extra invocation is more often than not underlaid to the polyphony. A clear illustration of this type of discrepancy can be seen in the various copies of the Sanctus from the Missa sine nomine ascribed to Bedyngham: while the readings in Trent 93 and Trent 90 carry three invocations of the opening word, that in Oxford, Additional MS C87 (hereafter Add. C87) has only two invocations and a plainsong incipit.

⁸ The same situation obtains here as in the case of English Sanctus settings: most insular copies, and Continental readings which seem close to English antecedents, begin the polyphonic setting at ‘qui tollis’, while most Continental copies incorporate the opening words into the polyphony.

⁹ See, for instance, Kenney, S. W., Walter Frye and the Contenance Angloise (New Haven, CT, and London, 1964), p. 54Google Scholar; Wegman, R. C., ‘Concerning Tempo in the English Polyphonic Mass, c. 1420–70’, Acta Musicologica, 61 (1989), p. 64CrossRefGoogle Scholar, and Wegman's review cited above, note 3.

¹⁰ All musical references are to Curtis's, Gareth recent edition, The Brussels Masses, Early English Church Music 34 (London, 1989)Google Scholar. The masses are also published in Kenney, S. W., ed., Walter Frye: Collected Works, Corpus Mensurabilis Musicae 19 (n.p., 1960)Google Scholar.

¹¹ Kenney, , Walter Frye and the Contenance Angloise, p. 135Google Scholar.

¹² Ibid., p. 130.

¹³ Quoted in ibid., p. 134. Kenney's barring is 71f.

¹⁴ Ibid., p. 131. Kenney's barring is 153f.

¹⁵ The same idea also occurs in the contratenor, for instance in the Agnus, bar 3 Curtis's view that Frye's ‘interest in the c f was not primarily connected with any distinctive melodic character it may have had in its normal form’ should perhaps be questioned in the light of such structural exploitation of the contour of a borrowed melody. (See ‘The English Masses, p. 71.) Closely similar figures occur elsewhere in Frye's work: see, for instance, Sospituli dedit, bars 26–8 and 49–50, and Nobilis et pulchra, Credo, bars 43–4 (see Example 21).

¹⁶ Bars 43–6 of the Credo of Nobilis et pulchra (see Example 21) provide a similar example. This passage was cited in the same connection by Kenney, , Walter Frye and the Contenance Angloise, p. 134Google Scholar.

¹⁷ Ibid., p. 134. The passage is quoted on p. 133.

¹⁸ A possibility first broached by Trowell, B. L. (‘Frye, Walter’, The New Grove Dictionary of Music and Musicians, ed. Sadie, S., 20 vols. (London, 1980), vi, p. 877)Google Scholar. The length of the work presupposes the original setting of a prosula Kyrie. Trowell suggested the prosula Conditor Kyrie, the text prescribed by the Sarum use for Trinity Sunday, the feast from which the cantus firmus of the mass is taken. The issue was explored in greater depth by Curtis, ‘The English Masses’, i, pp. 96–106. Curtis also attempted a reconstruction using the prosula Omnipotens pater (ibid., ii, pp. 26–9). More recently (The Brussels Masses, pp. 188–9), Curtis adopted a more cautious line, noting that while the poor relation between text and music may suggest that the work as it now stands is a contrafact, this need not necessarily imply that it was originally a Kyrie. As he observes, the motets in some so-called mass-motet cycles survive in different sources with distinct texts. Still, it is difficult to see how a mass-motet cycle would fit into the English mass tradition, especially since, as Reinhard Strohm has pointed out, all the other cases are largely confined to the related, Germanic, Trent and Strahov (Prague, Strahov Monastery, MS DG. IV. 47) manuscripts (‘Messzyklen über deutsche Lieder in den Trienter Codices’, Liedstudien: Wolfgang Osthoff zum 60. Geburtstag, ed. Just, M. and Wiesend, R. (Tutzing, 1989), pp. 88–9)Google Scholar. Strohm's view of two separate traditions involving, on the one hand, five-movement cycles including contrafact Kyries and, on the other, bona fide six-movement mass-motet cycles, seems to make best sense of the surviving repertory, but even this is not without its problems. The most significant of these centres on the still little-understood role of the prosula Kyrie outside England: if, as Strohm suggests, the masses on Meditatio cordis and Hilf und gib rat share with the Missa Summe trinitati the status of decapitated masses with contrafact Kyries, it is difficult to explain the lengths of these movements otherwise than as erstwhile prosula settings.

¹⁹ As in the Credo of Summe trinitati, see below.

²⁰ See also, for instance, Nobilis et pulchra, Kyrie, bars 44–7 (two parts). Upward-moving sequences do occur occasionally, as at bars 47–52 of the contratenor in the Credo of Nobilis et pulchra.

²¹ See bars 31–2 and 80–1.

²² Walter Frye and the Contenance Angloise, p. 144.

²³ As Kenney points out, the potential for reduced trio sections is almost never exploited (ibid., p. 145). However, this, as David Fallows pointed out to me, was standard practice in the mid-late fifteenth century.

²⁴ See note 1.

²⁵ See Saunders, S. E., ‘The Dating of the Trent Codices from their Watermarks, with a Study of the Local Liturgy of Trent in the Fifteenth Century’ (Ph.D. dissertation, University of London, King's College, 1985), pp. 87–8Google Scholar. Saunders notes, however, that the motet was added to the source by a hand not present elsewhere in the manuscript. This view is supported by Rebecca Gerber, who adds, though, that Johannes Wiser, the main scribe of the source, added the clefs on the recto of the opening which contains the copy (‘The Manuscript Trent, Castello del Buonconsiglio, 88: A Study of Fifteenth-Century Manuscript Transmission and Repertory’ (Ph.D. dissertation, University of California at Santa Barbara, 1984), pp. 19, 23, 25)Google Scholar. She infers from this that Wiser must have been close at hand when the motet was copied, though it seems more likely that his small additions came about through post facto editing due to negligence on the part of the main scribe of the piece. Whether this means that the work was copied significantly later than the main body of the manuscript, however, remains open to question.

²⁶ The other indisputably English example to have survived complete is the anonymous Missa Salve sancta parens in Trent 93/90, though Margaret Bent has proposed that one of the voices in this mass may in fact be a later addition to a three-voice work (Four Anonymous Masses, Early English Church Music 22 (London, 1979), p. 181)Google Scholar. Possible further candidates among the anonymous repertory include the cycle on folios 65^v-75^r of Rome, Cappella Sistina 14 (with a fragmentary concordance on fol. 30bis of Lucca, Archivio di Stato, Biblioteca Manoscritti MS 238 (the Lucca Choirbook) and the Missa Meditatio cordis, a ‘mass-motet cycle’ whose motet, like that of Summe trinitati, may also be a contrafact prosula Kyrie (Strahov, fols. 85^v-92^r; motet in Trent 88, fols. 284^v-286^r). The various fragmentary cycles include Plummer's Missa Nesciens mater, partly preserved in London, British Library Add. MS 54324, the Missa Requiem aeternam in Add. C 87, and a Missa Alma redemptoris mater of which parts survive in London, British Library Add. MS 54324 (Add. 54324) and in the Lucca Choirbook.

²⁷ See Wegman's review of Curtis, The Brussels Masses, p. 587.

²⁸ See Table 1. For outlines of the structures of Frye's masses, see Curtis, ‘The English Masses’, pp. 66–95.

²⁹ Examples of such use of repetition to create a sense of arrival could be multiplied many times. The conclusion of Agnus III, before the final ‘Dona nobis pacem’ is similarly climactic: here both discantus and tenor play with repeated ideas, in one case involving imitation between them (see, for example, Agnus Dei, bars 83–99). This passage gives a further indication of the subtle range of possibilities available within the basic principle of repetition: by turn, it can be exact or varied, immediate or separated by other material.

³⁰ A similar passage occurs in close imitation between discantus and contratenor in bars 11–12 of the Credo of Summe trinitati.

³¹ Seay, Albert, ed., Johannis Tinctoris: Opera theoretica, iia: Proportionale musices, Corpus Scriptorum de Musica 22 (n.p., 1978), p. 10Google Scholar.

³² Saunders gives paper dates of 1450–2 for the earliest copies of movements from the Missa sine nomine in Trent 93, and 1454–8 for the earliest copies of movements from Deul angouisseux in Trent 93 (see ‘The Dating of Trent 93 and Trent 90’, I codici musicali trentini, ed. Pirrotta, N. and Curti, D. (Trent, 1985), pp. 70, 73)Google Scholar.

³³ Strahov, fols. 53^r–53^v, 164^v–165^r, 167^v–171^r.

³⁴ For an examination of the structure of this mass, see Curtis, ‘The English Masses’, pp. 110–15.

³⁵ Rob Wegman, who has made a study of the paper which comprises this source, informs me that the closest paper he has been able to trace to that on which the Plummer mass is copied can be dated 1456–8. See Piccard, G., Die Ochsenkopf-Wasserzeichen (Stuttgart, 1966), nos. 385 and 391, pp. 427–8Google Scholar.

³⁶ For a discussion and dating of Lucca, see Strohm, R., Music in Late Medieval Bruges (Oxford, 1985), pp. 120–36Google Scholar.

³⁷ Parts of the Kyrie are found in Add. 54324 (see note 26). Strohm, p. 127, contains a description of the mass and (pp. 220–5) a transcription of the Sanctus up to the end of Osanna I. I am grateful to Professor Strohm for drawing my attention to the stylistic similarities between this mass and those under discussion.

³⁸ See ‘Frye, Walter’, p. 877. See also Strohm, pp. 125, 173.

³⁹ I know of only one cycle, which is otherwise totally unrelated to the two under discussion, with a Kyrie containing two distinct invocations of the Christe. This is the Missa Regina celi on folios 25^r-33^r of Trent, Castello del Buonconsiglio, MS 91. Two other cycles include ninefold sectional Kyries: Dufay's Missa Ave regina celorum, and the mass by Pullois, both of whose structures involve internal repetition.

⁴⁰ For editions, see Kenney, , ed., Walter Frye: Collected Works, pp. 17–19Google Scholar, and Charles, S. R., ed., The Music of the Pepys MS 1236, Corpus Mensurabilis Musicae 40 (n.p., 1960), pp. 131–3Google Scholar.

⁴¹ It may be worth noting in view of the cantus firmus of this one surviving motet by Frye that the composer was himself a member of the Confraternity of St Nicholas in London (see Curtis, , The Brussels Masses, p. ix)Google Scholar.

⁴² For an analysis of the cantus firmus treatment, see Sparks, E. H., Cantus Firmus in Mass and Motet (Berkeley and Los Angeles, 1963), pp. 75–9Google Scholar.

⁴³ Strohm, , Music in Late Medieval Bruges, p. 125Google Scholar.

⁴⁴ Nos. 30, 33, 34 and 43 in Charles, The Music of the Pepys MS 1236. Two further works, nos. 66 and 101, have the analogous combination c2c4f4.

⁴⁵ My thanks to Rob Wegman for this observation.

⁴⁶ See Wegman, ‘New Data’, p. 13.