1 It is a pleasure to record here my thanks to the governing bodies of the several institutions mentioned in the footnotes for their permission to consult and quote from archival and musical manuscripts in their possession, and to their archivists for their unfailing cooperation in making them available for study. The present article is a revised version of a paper first delivered at the Conference on Performing Pitch in Renaissance Music held at the London Early Music Centre in May 1981. Earlier periods are considered in ‘The Performing Ensemble for English Church Polyphony c 1320–90’, Studies in the Performance of Late Mediaeval Music, ed Stanley Boorman (Cambridge, 1983), 161–92, and – as a preliminary study of fuller research – ‘The Performing Pitch of English Fifteenth-century Church Polyphony’, Early Music, 8 (1980), 21–8, with ensuing correspondence in Early Music, 9 (1981), 71–5Google Scholar

2 Some of this evidence has recently been reviewed elsewhere for very similar purposes; see especially Wulstan, David, ‘Vocal Colour in English Sixteenth-century Polyphony’, Journal of the Plainsong and Mediaeval Music Society, 2 (1979), 19–60; idem, Tudor Music (London, 1985), 192–249, and Roger Bray, ‘More Light on Early Tudor Pitch’, Early Music, 8 (1980), 35–42, For a particular response to Bray, see Bowers, Roger, ‘Further Thought on Early Tudor Pitch’, Early Music, 8 (1980), 368–74. Wulstan and Bray have reached conclusions about vocal scoring, transposition and performing pitch that differ radically from those proposed here, I apologize to readers in advance for the frequency with which the ensuing footnotes will inevitably record both the fact of, and my grounds for, seeking to disagree with their previous scholarship in this held, while acknowledging the value of their contributions in focusing attention on the issue in the first placeGoogle Scholar

3 The fullest account currently available of the nature and composition of the liturgical choirs of pre-Reformation England appears in Frank LI Harrison, Music in Medieval Britain, 2nd edn (London, 1963), chapters 1 and 4Google Scholar

4 From the period c 1500–58 there are a few instances in which the appearance of certain very low notes in the bass voice extends the overall compass to as much as 26 notes, see, for example, William Cornyshe's Magnificat (Early Tudor Magnificats, l, ed Paul Doe, Early English Church Music, 4 (1965), 49, at bars 27 and 212) However, these are invariably found to occur as final notes sounding the root of the chord at principal cadences, they are essential neither to the harmony (the same note sounding an octave higher invariably serves equally well as the bass) nor to the progress of the polyphony As such they seem to be no more than notated concessions to the (normally unauthorized) bass habit of octave-doubling the root at main cadences, they are certainly extraneous to the limits of vocal scoring otherwise adopted by the composer, and have therefore been excluded from consideration in the following discussions of vocal ranges and overall compass A single a” in the top voice at the climax of Tallis's Gaude gloriosa dei mater and another in an exactly analogous place in his Ave rosa sine spinis (Thomas Tallis, ed Percy C Buck et al., Tudor Church Music, 6 (London, 1926), vi, 142, 178) extend the overall compass of these works exceptionally to 24 notes For a comparable phenomenon in some of the music of John Sheppard, see below, note 26Google Scholar

5 To minimize the risk of confusion, Latin terms conveying part-names are printed in italic (e.g. triplex, contratenor, tenor), while English vernacular terms denoting the names of particular vocal timbres are printed in roman (e.g treble, countertenor, tenor)Google Scholar

6 The Eton Choirbook, ed. Frank LI Harrison, Musica Briannica, 10–12, 2nd edn (London, 1967–73), n, 32; also Robert Fayrfax Complete Works, ed Edwin Warren, Corpus mensurabilis musicae, 17 (Rome, 1959–66), li, 26 ⁷ Warren, Robert Fayrfax, n, 59Google Scholar

8 John Taverner, ed Percy C Buck et al., Tudor Church Music, 1, 3; this edition is in process of being superseded by John Taverner Complete Works, ed Hugh Benham, Early English Church Music, 20, 25, 30 (London, 1978-)Google Scholar

9 See, for example, Gaude plurimum, Sospitati dedit egros and both settings of Dum transisset sabbatum (Benham, John Taverner, li, 32, and in, 75, 109, 127), the same divergence of the voice next above the bottom into voices of full range and voices of low range occurs also in the Mass ‘Corona spinea’ (voices III and IV) and the Meane Mass (voices II/III and IV) (Benham, John Taverner, l, 75, Buck et al, John Taverner, 1, 50). In fact, many of the parts written for this low-range voice occur in settings of Office responsories intended for performance by the total choir, and observe its restricted range simply because they consist of no more than the proper plainsong disposed as a monorhythmic cantus firmus, most probably intended to be sung by those members of the choir who were not skilled in singing from polyphonic notation, and who otherwise would not have been able to participate. However, this restricted range was sometimes applied equally strictly even to genuinely polyphonic parts appearing in the Tenor book, showing this divergence of the vocal timbre concerned to have been indeed an inherent feature of the development of the polyphony chorus at this periodGoogle Scholar

10 On 18 June 1551 the vicars choral of Salisbury Cathedral reported to the chapter that ‘sunt destituti de uno vicario cantante ad partem vocatam le Cownter tenor’, on 10 October following they presented to the chapter for admission as a lay vicar one John Seywarde, ‘cantantem ad partem contratenoris’ (Salisbury, Diocesan Record Office, Archives of the Dean and Chapter Reg Holte, ff 29^r, 29^v) These are the earliest instances known to me of English provenance in which the term ‘countertenor’ appears to be used to designate a particular vocal timbre, it suggests that by 1551 the divergence of the erstwhile tenor voice had progressed sufficiently far for the terms ‘cownter tenor’ and, presumably, ‘tenor’ to be understood to designate two distinguishable kinds of tenor voice, one having full range, the other concentrating on just its lower range.Google Scholar

11 For the sake of completeness, it may be mentioned here that in four- and five-part writing for lower voices only, and in six-part writing for full choir, the voice next above the bottom might sometimes be an intruded voice pitched midway between the normal pitches of the two lowest voices For some consideration of this phenomenon, see below, p 49 and n 25.Google Scholar

12 In what is in essence an identical table (Wulstan, ‘Vocal Colour’, Figure 8, p 40, and Tudor Music, 212), the third and fourth voices are marked, respectively, with the timbre designations ‘alto’ and ‘tenor’. In view of the fact that their bottom pitches are identical, and their overall pitches virtually so, this interpretation seems unacceptableGoogle Scholar

13 Wulstan (‘Vocal Colour’, 38–43, 47–50, and Tudor Music, 247) makes the assumption that five-part music of the period c. 1470–1520 required five different timbres of voice for its performance, however, in view of the evident congruence and identity of the requirements made by the third and fourth voices on their respective singers, this contention cannot be sustained It appears to derive from a misunderstanding of the nature of the relationship between contratenor and tenor parts at that period, and also of the five-way division of the singers of the choir of the household chapel of Henry Percy, Earl of Northumberland, recorded for the period c 1505–22, see below, pp 60–64 and AppendixGoogle Scholar

14 In practice, the six combinations afforded only five choices of ‘key’, since the A-lattice without signature disposed the semitone intervals in exactly the same places on the staves as the D-lattice with signatureGoogle Scholar

15 The epithets ‘high’, ‘medium’ and ‘low’ are deliberately avoided here, as bearing connotations of pitch entirely foreign to the role and meaning of the clef at the period to which this discussion relates.Google Scholar

16 For instance, the tenor of Taverner's second setting of Dum transissel sabbatum is a monorhythmic utterance of the proper plainsong, covering a compass of only a seventh, c–bb It appears in the C5 clef, which strictly does not belong in the F-lattice, but it could equally well have been notated in the C4 clef, which does so belong, and where the identical music is indeed placed in the exactly corresponding circumstances of Taverner's first setting of this text Benham, John Taverner, iii, 127, 109. In the few instances in which this phenomenon applies to the bass part, identification of the true lattice of clefs in operation needs special care, and has to be determined from the remaining voices rather than the bassGoogle Scholar

17 As for instance in Taverner's Kyrie Leroy (ibid., 136), in which the second voice, using only the ninth g–a’ of its usual eleventh g-c”, is nonetheless pitched a fifth above the third voice singing its usual c–f”, and a fifth below the top voice singing its usual d’–f” or g”Google Scholar

18 The clef, that is, performed a function analogous less to that of the modern clef that to that of the modern key signatureGoogle Scholar

19 John Sheppard, i, Responsorial Music, ed David Chadd, and John Sheppard, ii, Masses, ed Nicholas Sandon, Early English Church Music, 17–18 (1977, 1976), a third volume is awaited It must be stressed that the choir for which most of this music was composed, that of Mary I's Chapel Royal, was for its period uniquely large, staffed by 30 lay clerks and 12 boys (London, Public Record Office [henceforth P RO], LC 2/4 (1), ff 174^r-v; E 101 427/5, ff 10^r, 27^r-v; E 101 427/6, f 28^r), among composers of the 1550s, Sheppard (with his Chapel Royal colleagues and successors) was virtually alone in having so many men available that he could compose in as many as five real parts for adult voices, and one for boys, and still expect a satisfactory vocal balance The vocal scoring of his music is thus decidedly untypical of its time, and cannot legitimately be used as an exemplar against which to make comparisons with late sixteenth- and early seventeenth-century music It is instructive to compare the ‘bottom-heavy’ scoring of Sheppard's music written for the Chapel Royal with the ‘top-heavy’ scoring of the music whose texts indicate that he wrote it for the choir of Magdalen College, Oxford, where the superabundance of boys invited such scoring as the TrTrAATTB encountered in his two settings of Libera nos John Sheppard, i, Office Responds and Varia, ed David Wulstan (Oxford, 1979), 83, 86, 108Google Scholar

20 It is in so far as they rely on the extension to the pre-Reformation period of conclusions derived from sources (both musical and literary) dating from later than c 1600, that the opinions concerning performing pitch and vocal scoring expressed by Wulstan (‘Vocal Colour’ and Tudor Music) and Bray (‘More Light on Early Tudor Pitch’) appear to be unsustainable and invalidGoogle Scholar

21 Bowers, ‘The Performing Pitch of English Fifteenth-century Church Polyphony’, and, more fully still, in Roger Bowers, ‘Choral Institutions within the English Church Their Constitution and Development, 1340–1500’ (Ph D dissertation, University of East Anglia, 1975), 6001–25Google Scholar

22 Ibid, 6026–90 passimGoogle Scholar

23 Wulstan (‘Vocal Colour’, 25–6, 43–7, 56, and Tudor Music, 224–5, 240–1) suggests that the voices of boy choristers in the early Tudor period did not break until around the age of 18, and that in consequence of the advanced physical size and maturity of such boys, and of their resulting increased power and capacity of lung, they were able to sustain an extraordinarily high tessitura, extending from f’ to b” b or even c”‘ This idea seems to be misconceived on numerous grounds Firstly, of course, the breaking of the boy's voice cannot be separated from the general progress of growth and maturity, any 18-year-old with a pre-adolescent voice would perforce have not an adult, but still only a pre-adolescent physique (Tanner, James M., Growth at Adolescence, 2nd edn (Oxford, 1962), 78 and passim; David Sinclair, Human Growth after Birth, 3rd edn (Oxford, 1978), 115–16, and 21–35, 89–116 passim) But anyway, it is entirely clear that in the late fifteenth and early sixteenth centunes the trained singing-boy's voice broke not at 18 but at about 14 or 15 This is shown by analysis of the careers of almost 100 boys who served as chonsters of St George's Chapel, Windsor between 1461 and 1499 (data from account rolls of Treasurer, Chorustis subsection of Stipendia section of Expense. Windsor, St George's Chapel, the Aerary Archives of the Dean and Chapter, xv 34 49–71); confirmation is provided by the Chantry Certificates of 1545 and 1547/8, which list by both name and age the choristers of a number of collegiate churches, none of whom exceeded 15 years of age, even despite the likely presence among them of boys whose voices had in fact broken, but whose services were being nominally retained in order to qualify them for the usual ex gratia payment at the dissolution as compensation for loss of employment (a few extracts are printed in Arthur F Leach, English Schools at the Reformation (Winchester, 1898), part n, passim, eg pp 31, 219) It may be noted that university colleges such as King's Hall, Cambridge, whither suitable ex-choristers were sent from the Chapel Royal for the furtherance of their education, effectively applied for admission a minimum age qualification of about 17 or 18, once their voices had broken, former choristers would still have to spend some three or four years attending the grammar school of the royal household perfecting their knowledge of Latin grammar before they were able to meet the entry qualifications for colleges which admitted undergraduates (see Alfred C Cobban, The King's Hall in the University of Cambridge (Cambridge, 1969), 58–62) This explains the regulation from the ‘Black Book of Edward IV’, quoted by Wulstan, ‘Vocal Colour’, 46 and Tudor Music, 240–1, giving 18 as the age at which choristers with broken voices might be sent from court to university. Neither physically nor vocally, therefore, can the early Tudor chorister have differed very much from his present-day successorGoogle Scholar

24 In this discussion, the presumption will be made, in cases where there is no evidence to the contrary, that while methods of tone-production, articulation, intonation and the like may well have varied from period to period, yet the actual range of pitch realizable by the generality of human voices 400–500 years ago cannot be significantly different from that manifested today, the progress of human evolution is too ponderous and slow for any discernible modification to the larynx and other vocal apparatus to have occurred in so short a time Further, during discussion of this paper during the London Early Music Centre Conference on Pitch, it was pointed out from the floor that studies conducted by physiologists had shown that the pitch of any human voice bears no correlation to the height or weight of its owner, changing patterns of nutrition have made twentieth-century men and boys taller and heavier than their sixteenth-century ancestors, but this process will not have affected the prevailing pitches of their respective voicesGoogle Scholar

25 In compositions using four or more men's voices, the occasional use of this baritone voice, pitched exactly between bass and tenor, can be traced back to the Eton Choirbook repertory see, for example, John Browne, Stabat mater dolorosa, Stabat iuxta Christi crucem and O regina mundi clara in Harrison, The Eton Choirbook, i, 43, 64, 72 This appears to be the only instance in which composers ever scored for a non-standard voice pitched in the space between two of the standard timbresGoogle Scholar

26 John Sheppard occasionally adopted a 24-note compass, extending the standard compass by one semitone through writing the bass down to E, see, for example, Reges Tharsis et insule, Gaude, gaude, gaude, Maria virgo, Christi virgo dilectissima, Filie Ierusalem (Chadd, John Sheppard, i, 23, 99, 118, 138) See also note 4 aboveGoogle Scholar

27 ‘Statuimus igitur et ordinamus cum consensu decani ecclesie nostre cathedralis prediete ad ipsius cultus divini pleniorem augmentationem et ut omnis spintus in ecclesia dei laudet dominum, quod sint ibidem perpetuo quatuor clerici laici concinuas voces habentes et musice docti, quorum unus ad minus semper sit basse naturalis et audibilis vocis; aliorum vero trium voces sint suaves et canore, ita quod a communi vocum succentu possint naturaliter et libere ascendere ad quindecim vel sexdecim notas’ Oxford, New College, Archives of the Warden and Fellows, 9432, ff 21^v–22^r A sealed copy of this ordinance survives as London, British Library, Cotton Charter xii 60Google Scholar

28 While demanding nothing different in terms of simple vocal range, it is certainly true that a polyphonic bass line regularly descending to F and G requires in this lowest register a more flexible voice than does a purely harmonic bass line exploring similar depths It was, perhaps, the particular ability of English basses to negotiate these lower reaches of their range that so struck an Italian visitor that he described the basses of Henry VIII's Chapel Royal in 1515 by his vernacular term contrabassi: ‘Ditta messa fu cantata per la capella de questa Maestà [Henry VIII], qual veramente è più presto divina che humana, non cantavano ma jubilavano, et maxime de contrabassi, che non credo al mondo sieno li pari’ I Diarii di Marino Sanuto, ed Federico Stefani, Guglielmo Berchet and Nicolò Barozzi (Venice, 1879–1903), xx, col 266 Although this document emanated from diplomatic circles, it can by no means be dismissed as mere propaganda, diplomatic hyperbole or obsequious exaggeration, since it occurs not in official correspondence, but in an informal letter sent by Nicolo Sagudino (Secretary to Sebastiano Giustinian, Ambassador in England of the Republic of Venice) to a friend at home, Alvise Foscari This, and Sagudino's other letters to Foscari, show him to have been a shrewd and well-informed observer of musical matters (and also himself a keyboard-player), and his report must be accorded considerable weight, as a reflection of a trustworthy witness's genuine and enthusiastic admiration for what he had heard.Google Scholar

29 Wulstan (‘Vocal Colour’, 51) suggests that the specified compass of 15–16 notes was intended to comprehend all four lay clerks, and not just the upper three of their voices The Latin is not totally unambiguous, nevertheless, in the absence of some phrase such as omnium eorum from the phrase a communi vocum succentu there are no grounds for departing from the plainest sense of the Latin original, especially in view of the degree to which the compiler of this very long document normally lost no chance to extend his verbiage whenever opportunity arose.Google Scholar

30 London, British Library, Add. MSS 17802–5, for a description and inventory see Bray, Roger, ‘British Museum, Add MSS 17802–5 (The Gyflard Part-Books) An Index and Commentary’, Royal Musical Association Research Chronicle, 7 (1969), 31–50 A tentative suggestion of Westminster Abbey as the provenance of these books has been made see Benham, John Taverner, iii, p xvii.Google Scholar

31 British Library, Add. MSS 17802–5, ff 111^r, 107^r, 109^r, 102^r, Benham, John Taverner, iii, 100.Google Scholar

32 Cambridge, University Library, MSS Peterhouse 485, 487, 490, 488, ff 03, 02 and 05, G3, and M4 respectively, see Hughes, Anselm, Catalogue of Musical Manuscripts at Peterhouse, Cambridge (Cambridge, 1953), 39 Annotations in the music indicate that this was a scribe's crude preparatory draft, made prior to the production of neat copy, the original provenance of these particular fascicles is not known, but it has been observed that their watermark occurs also in the ‘Henrician Set’ of Peterhouse partbooks of c 1539–40. see Sandon, Nicholas, ‘The Henrician Part-books at Peterhouse, Cambridge’, Proceedings of the Royal Musical Association, 103 (1976/7), 110Google Scholar

33 MS 487, f O3^vGoogle Scholar

34 For instance, in John Sheppard's setting of Hec dies (Chadd, John Sheppard, i, 39) the overall compass is 22 notes, G–g”, and the piece may be sung at this pitch in modern terms; however, all the voices have relatively narrow ranges, and all the criteria would still be met if the piece were performed at F–f”.Google Scholar

35 Normally, a compass of 18–19 notes predicates performance by lower voices alone, i e. ATB and various combinations thereof However, it would be admissible for pieces for Lady Mass, at which service in many institutions only the boys and a handful of men were required to be present, to be performed instead by TrAT, and much such music may well have been composed with such interchangeability of performing medium in mind. Pieces with a compass of 14–16 notes may be singable by treble and alto voices, alto and tenor, or tenor and bass, depending on the date of composition, manuscript context and liturgical destinationGoogle Scholar

36 The elaborate ‘clef-code’ proposed in Wulstan, ‘Vocal Colour’, 33–41, and Tudor Music, 203–14, by application of which it is said that correct sounding pitch may be deduced from ‘written pitch’ through precisely prescribed degrees of ‘transposition’, seems thus to be entirely imaginary, at least in respect of pre-Reformation music Ideas on these lines become applicable only in the second half of the sixteenth century, when the newly introduced practice of composing for mixed ensembles of voices with instrument(s) first began to cause the concept of singers’ looking to the clef to ascertain and fix performing pitch to crystallize ³⁷ In 1535 see Valor ecclcsiasticus, ed. John Caley and John Hunter (Record Commission, 1810–34), in, 6, 12–13Google Scholar

38 In 1548, at the dissolution of the college eleven vicars choral (two of the statutory vicarages choral apparently being vacant), four clerks and the Master of the Choristers (Nicholas Ludford, nominally occupying the office of Verger), and seven choristers London, P.R O, E 301 88.Google Scholar

39 16 vicars choral, the four lay clerks of Bishop Sherburn's foundation, and eight choristers. Oxford, New College, Archive 9432, f 33^rGoogle Scholar

40 By the early sixteenth century, it is clear that the holders of lay choir clerkships at any type of church were always skilled singers of polyphony However, in respect of choirmen in priest's orders, circumstances varied At those major collegiate chantries (e.g Tattershall, Fotheringhay) at which the chaplains of the choir doubled as the managing fellows of the college, probably administrative and business rather than musical ability was chiefly sought in the holders of chaplaincies, and the ability to sing polyphony was not a criterion of importance However, at the major collegiate churches of prebendal canons, where there was a residentiary chapter to manage the complexities of the business side of the church's affairs, it appears that the ability to sing polyphony was at least a high, and sometimes an essential priority among the criteria for the appointment of all the vicars choral, whose functions were almost exclusively liturgical and musical (see, for example, the regulations of 1503 for Ripon Minster and 1507 for York Minster quoted in Harrison, Music in Medieval Britain, 176, 181) At the academic colleges of Cambridge and Oxford the senior fellows managed the administrative and business sides of affairs, but thereafter were distracted by academic duties from fulfilling a full round of sacerdotal functions in the college chapel, for the chaplains of the choir, therefore, the principal qualification for appointment had to be suitability to serve as simple chapel priests and fit into an academic community, but so broad a criterion would still have left scope for adding the ability to contribute to the singing of polyphony, if men so qualified could be foundGoogle Scholar

41 In 1535 see Caley and Hunter, Valor ecclesiasticus, ii, 299Google Scholar

42 Ten lay clerks and ten choristers, at least from 1492/3 onwards. Receiver's Accounts 1492/3, 1495/6: Penshurst Place (Kent), MSS of Lord De Lisle and Dudley, U1475 Q 16/1–2.Google Scholar

43 The statutory complement c 1480 was four chaplains, eight clerks, sixteen choristers, optional Master of the Choristers (Her Majesty's Commissioners, Statutes of the Colleges of Oxford (Oxford and London, 1853), ii/1 Magdalen College, pp 6, 24) These numbers were mostly maintained between 1500 and 1547, with a full-time Master of the Choristers from 1510 onwards (Oxford, Magdalen College, Archives of the President and Fellows. Bursary Book 1501–7, Commune and Stipendia Capellanorum et Clericorum sections of Libri Computi 1490/1510–1543/59).Google Scholar

44 The statutory complement in 1453 was ten chaplains, six clerks, sixteen choristers (The Ancient Laws of the Fifteenth Century for King's College, Cambridge and for the Public School of Eton College, ed James Heywood and Thomas Wright (London, 1850), 20–1) These numbers were mostly maintained between 1503 and 1547 (Cambridge, King's College, Archives of the Provost and Fellows Libri Communarum 11–14, Conducticii et Ciena subsections of Pensiones sections, and Exhibucio Chorustarum sections, of Mundum Books 9–12)Google Scholar

45 The statutory complement in 1400 was ten chaplains, three clerks and sixteen choristers (Statutes of the Colleges of Oxford, i/5 New College, p 4) These numbers were mostly maintained between 1500 and 1547, with addition to clerks of one Master of the Choristers. Oxford, New College, Archives 5530 (Hall Books, 1500/1–1542/3), 7462–7511 (Commune and Custus Capelle sections of Bursars’ Accounts, 1500/01–1546/7)Google Scholar

46 In 1499 there were 16 vicars choral, 13 clerks and 13 choristers Vicariis, Clericis and Chorustis sections of Treasurer's Account 1498/9 Windsor, St George's Chapel, Archives of the Dean and Chapter xv 34. 71Google Scholar

47 In 1535 See Caley and Hunter, Valor ecclesiasticus, n, 85 (noting that £54 17s 6d = 36s 7d × 30)Google Scholar

48 In the late fifteenth century see Bowers, ‘Choral Institutions within the English Church’, 6047–57Google Scholar

49 Graham, Rose, ‘The Priory of Lanthony by Gloucester’, Victoria County History of Gloucestershire, ed William Page et al. (London, 1907-), n, 87Google Scholar

50 Certified copy of indenture of employment, dated 29 March 24 Henry VIII: London, PRO., E 315 93, f 231^vGoogle Scholar

51 Although recently described as ‘no longer extant’ (Wulstan, ‘Vocal Colour’, 42, and Tudor Music, 234), the manuscript does in fact survive, and is now preserved at Alnwick Castle (Northumberland), Estates Office, Archives of the Duke of Northumberland, MS 98A. A transcription of the whole book has been published The Regulations and Establishment of the Houshold [sic] of Henry Algernon Percy, the Fifth Earl of Northumberland, ed Thomas Percy (London, 1770) However, this cannot be used safely for modem study purposes, it contains no critical apparatus, and unfortunately, of its commendably few mistakes and omissions, two occur in passages relating to the earl's household chapelGoogle Scholar

52 Oxford, Bodleian Library, MS Eng misc. b 208 I am most grateful to Mr Andrew Wathey for drawing my attention to the existence of this sourceGoogle Scholar

53 Edmund Turges's Gaude flore virginale, discussed below, p. 66 and n 75, Robert Wilkinson's nine-part Salve regina and Nicholas Ludford's six-part Mass ‘Videte miraculum’ appear to be the only substantial exceptions to this ruleGoogle Scholar

54 In similar vein, a late fifteenth-century set of instructions for tuning a five-course lute used the terms ‘trebill’ and ‘secunde trebill’ to denote the two highest courses, which, despite the terminology, were tuned not in unison but with the ‘secunde trebill’ a fourth below the ‘trebill’ See Page, Christopher, ‘The 15th-century Lute New and Neglected Sources’, Early Music, 9 (1981), 13–14Google Scholar

55 List A of c. 1505, being a kalendar or list of all authorized members of the earl's household, would be used as a recruiting list to fill vacancies, for this purpose, it would be important to know the approved numbers of trebles and meanes composing the authorized complement of singing-boysGoogle Scholar

56 Wulstan (‘Vocal Colour’, 43) claims that in five-part polyphony of the period c 1500–1625, it is the part written for alto voices (third vocal timbre reading upwards) that is normally doubled to create the fifth voice This may be true of the music of the last 50 years or so of the sixteenth century, but in music of its first half it is consistently the line written for tenor voices (second vocal timbre reading upwards) that is doubledGoogle Scholar

57 In seeking the resolution of these phenomena it is of course entirely inappropriate to turn to evidence deriving from some 120–30 years later It is indeed true that (1) two early seventeenth-century sources (quoted in Wulstan, ‘Vocal Colour’, 42, and Tudor Music, 233–4) list the five separate timbres of voice then in use in apparently exactly the same terms – treble, meane, countertenor, tenor, bass – as in the Northumberland Household Book (see Butler, Charles, The Principles of Music (London, 1636), 40–2, London, British Library, MS Royal 18 B. xix, f 8^V) It is no less true that (2) most late sixteenth-century and early seventeenth-century pieces employing the special high treble voice then cultivated occupy an overall compass of 23 notes, written F–g” (choir pitch; ‘transpose up’ about a minor third for modern pitch), ostensibly identical to that of much early sixteenth-century music However, these fortuitous correspondences, though beguiling, are entirely misleading if they be thought to suggest that the degree of ‘transposition’ which is certainly appropriate to the later music should be considered equally applicable to the earlier music as well. In the first place, the history of church music over the long intervening period of 120 years is simply too turbulent and disrupted to allow any such uncritical correspondence to be drawn between the two, in the absence of evidence demonstrating the existence of continuous traditions of performing practice through that period. (It may be noted that whereas Wulstan, ‘Vocal Colour’, 42, describes MS Royal 18. B xix as ‘late 16th-cent’, its references to ‘the late Queene’ and ‘the kings servants of his household Chappell’ (ff 5^V, 15^r) make clear that it postdates 1603 In fact it more probably dates from c. 1620–30, and therefore cannot be considered as independent evidence deriving from a date lying usefully between c 1505 and 1636) Secondly, the clef-configuration pertaining to the later (i.e. early seventeenth-century) body of music using the high treble voice is commonly G2–C1–C3–C4–F4, which certainly shows five distinct vocal timbres packed into the 23-note overall compass, by contrast, the clef-configuration prevalent in five-part music of the period of the Northumberland Household Book (first quarter of the sixteenth century) is the significantly different G2–C2–C4– C4–F4, showing only four distinct timbres, one being duplicated This disparity renders untenable any attempt to draw direct parallels between the early sixteenth-century and the early seventeenth-century material, no matter how beguiling the fortuitous similarities of terminology may beGoogle Scholar

58 Modern words such as ‘counterfeit’ still retain the original connotation of adjacency and (in this case, spurious) identicalityGoogle Scholar

59 See above, note 10 After all, not until the contratenor had evolved into a part consistently written in both a tessitura and a clef substantively higher than the tenor part could the term ‘countertenor’ come to designate a vocal timbre higher than tenor, and this did not occur before the middle of the sixteenth century at the earliestGoogle Scholar

60 Alnwick Castle, MS 98A, p 55 (Percy, The Regulations and Establishment of the Houshold of Henry Algernon Percy, 47–8, Percy omits ‘oone at iij marc’ between ‘ij at v marc a pece’ and ‘oone at xl s’)·Google Scholar

61 At institutions like King's College, Cambridge, where likewise the singing-men's salaries were not laid down by statute but were left to negotiation between the Provost and each individual, wages similarly varied widely in recognition of age, experience and responsibility (see The Ancient Laws of the Fifteenth Century for King's College, Cambridge, and for the Public School of Eton College (London, 1850), 120; for the sums agreed during the period c 1500–20 see Conducticii and Ciena subsections of Pensiones section of annual accounts of Bursars. Cambridge, King's College, Archives of the Provost and Fellows, Mundum Books, vol 9 and parts 1 and 2 of vol. 10). At the permanent collegiate churches, established singing-men might settle for a minimum of £2 0s 0d p a, but this sum allowed for the provision of free and permanent accommodation, in the case of a peripatetic household chapel choir this could not be offered, raising the minimum cash salary acceptable to about 50s or four marks (£2 13s 4d) p.aGoogle Scholar

62 Oxford, Bodleian Library, MS Eng misc b 208, f 50^r. The two similar lists in Alnwick Castle, MS 98A, pp. 320, 321 (Percy, The Regulations and Establishment of the Houshold of Henry Algernon Percy, 369, 374), do include the tenors among the organ-players, but both these latter lists are generally of doubtful reliability, see Appendix, pp. 73–5.Google Scholar

63 In any event it may be recalled that this terminology occurs not in a musical context but in an administrative one, and may have emerged not from an informed chapel source, but may have been dreamt up merely for accounting purposes by musically uninformed clerks in the Counting HouseGoogle Scholar

64 In continental church polyphony, the development of vocal scoring during the second half of the fifteenth century followed lines quite distinct from those in England, producing the very different performing media analysed by Fallows, David, ‘The Performing Ensemble in Josquin's Sacred Music’, Tijdschrift van de Vereniging voor Nederlandse Muziekgeschiedenis, 35 (1985), 32–66Google Scholar

65 Bray (‘More Light on Early Tudor Pitch’, 41–2) argues that the Northumberland household chapel choir was an institution of major significance, and that far from being peripheral and untypical, it may be taken as representing in its constitution the mainstream of all the major choirs of its period The then Earl of Northumberland was indeed one of the more influential noblemen of his time, and the importance, and perhaps even prominence, of his household chapel is not at all in doubt, but in pre-Reformation terms, given the wide variety of emphases exhibited by the various choral constitutions then in vogue, there was in reality no one single mainstream of which the earl's chapel might be considered typical It cannot therefore be Argued that every other major choir of the period necessarily shared the same constitution as that of the Earl of Northumberland, the division of the boys’ voices practised there need not be typical of the others at allGoogle Scholar

66 Cambridge, University Library, MSS Peterhouse 471–4 See Sandon, ‘The Henrician Part-books’, 106–40Google Scholar

67 Ff l^r of MSS 472 and 473 (the index to MS 471 is lost) In the indexes to MSS 472–4, John Mason's O rex gloriose (no 3 in Sandon's inventory: ‘The Henrician Part-books’, 134) is also marked ‘men’; unfortunately, owing to the loss of its highest voice from MS 471, it is not now possible to determine the overall compass of this pieceGoogle Scholar

68 Sandon, ‘The Henrician Part-books’, 112–13, Roger Bowers, entry relating to Peterhouse MSS 471–4 in Cambridge Musical Manuscripts, 900–1700, ed Iain A Fenlon (Cambridge, 1982), 132–5Google Scholar

69 Oxford, Magdalen College, Archives of the President and Fellows roll without reference, in bundle labelled ‘Sacrist's inventories’, m 6r, partially printed in William D Macray, A Register of the Members of St Mary Magdalen College, Oxford (Oxford, 1894–1915), ii, 210 The book concerned appears to have contained polyphony for Lady Mass, for the interchangeability of vocal scoring at this particular service, see above, note 35Google Scholar

70 See above, p 55 and note 43Google Scholar

71 See above, p. 55 and note 45.Google Scholar

72 Oxford, Bodleian Library, MS New College 368, fragment no 38 (provisional numbering) For the curious circumstances of the discovery of these fragments see Richard W Hunt, ‘The Medieval Library’, New College Oxford, 1379–1979, ed John Buxton and Penry Williams (Oxford, 1979), 339.Google Scholar

73 Harrison, The Eton Choirbook, iii, 1–13.Google Scholar

74 Hall Book 1507/8, sub Capellani. Oxford, New College, Archive no 5530 (Hall Books 1501–43), fourth gathering (unfoliated). There survive compositions from this period attributed to both Turges and Sturges, it seems likely (though not certain) that these are but variant spellings of the same nameGoogle Scholar

75 Of the original contents of the Eton Choirbook, no other has two equal boys’ voices at the top of its scheme of vocal scoring, and no piece with full three-octave compass divides the alto voice. However, the nine voices involved in Robert Wilkinson's Salve regina, incorporated into the Choirbook as a late addition after the composer's death (i e c 1515–20), include two trebles and one alto (with three tenors, one baritone and two basses). Harrison, The Eton Choirbook, i, 90 See also note 53 aboveGoogle Scholar

76 The ‘John Broune that sang the mean’ in the choir maintained in the parish church of St Laurence, Reading, in 1534/5, on whom the churchwardens spent 4s 4d that year to provide him with a new coat, was probably one of the boy choristers, since it was normally the boys of the choir who were thus rewarded in kind as well as in cash (Reading, Berkshire Record Office MS D/P97 5/2, p 198) Unfortunately, it has not yet proved possible to determine the relative numbers of men and boys in this choir, however, a small parish church choir such as this, unendowed and supported largely on the proceeds of annual collections organized by the churchwardens and the gilds, seems likely – on grounds of expense in this instance – to have been among those that would opt for the cheaper expedient of preferring boys to men for the alto partGoogle Scholar

77 Although little used in modern English cathedral choirs (Coventry and Westminster being conspicuous exceptions), the boy's alto voice is in fact by no means rare, and is indeed much cultivated in European liturgical choirs as the normal exponent of the alto partGoogle Scholar

78 A mid-fifteenth-century handbook of instructions for the singing of descant recorded that descant in the ‘meane’ sight was sung by a man's voice ‘[The quatreble sight] of descant longip to a childe to syng, for pe sight of pe mene and of pe quatreble bepe bope one sight in degre Therfor pe mene muste be sunge of a man and pe quatreble of a childe’. London, British Library, MS Lansdowne 763, f. 113^v, printed, not wholly accurately, in Meech, Sanford B, ‘Three Musical Treatises in English in a Fifteenth-century Manuscript’, Speculum, 10 (1935), 260. The recent notification by Wulstan (‘Vocal Colour’, 47, and Tudor Music, 242) that it is ‘a persistent misapprehension’ to consider that the term ‘meane’ denoted a man's voice is true enough for the post-Reformation era, but clearly requires revision in respect of the earlier periodsGoogle Scholar

79 It may be noted that the adoption of the boys’ meane voice in the performance of composed polyphonic music seems at present not to have been coeval with that of the treble, but to have been a departure inaugurated later, the earliest music involving the boys’ treble appears to have been written in the 1450s and 1460s, but as yet the earliest evidence suggesting that any choir used boys’ alto voices dates only from c 1505.Google Scholar