How choristers and organists improvised on chant during the Tudor period in England
Author: Jane Flynn
I'm a music historian and performer (organ, harpsichord, piano) whose research interests focus on the Tudor period: 1485-1603. I'm especially interested in the evidence for the the training of choristers and organists during that period that allowed them to produce quite complex multi-voice (polyphonic) music using only chant books or other sources containing only a single notated line of music.
This is the fourteenth and last of this section of my multi-part post on ‘How to play divers wayes upon the Plainsong Miserere’. (Please see previous posts below for earlier installments.)
Example 15 is Christe qui lux es et dies by Heathe, who may have been an organist at Exeter Cathedral. In this hymn verset the simple plainsong is in the lowest of three parts. For plainsong notes 3-15 the top part has a descant consisting of the rhythmic motif ‘dotted crotchet—quaver—crotchet’ that creates cross-rhythms against the semibreves of the plainsong, while the middle descant has 4:1 against the plainsong. For plainsong notes 16-27 the rhythmic devices used in the two descants are exchanged: the top descant is 4:1 while the middle part has the dotted motif. In the last section, plainsong notes 28-34, the top descant is 8:1 and the middle part is 4:1.
Example 15 [Thomas?] Heathe’s Christe qui lux es et dies (30513, fols 102v-103r):
Next week I’ll begin a new series of posts, the first giving background information on English descant sources, such as Leonel Power’s treatise (‘This tretis is contrivid upon the Gamme for hem that wil be syngers or makers or techers,’ GB-Lbl Lans. MS 763, fols 105v-113) and other early sources that refer to descant being learned on the keyboard as well as by singing. The (vocal) exercises on ‘plainsongs’ comprising solfa syllables in Power’s treatise will be presented in posts over a number of subsequent weeks as exercises for organists and singers (and anyone else interested in learning counterpoint!) to practice note-against-note descant.
This is the thirteenth of a multi-part post. (Please see previous posts below for earlier installments.)
Example 14, Preston’s Stirpe (from the sequence, Fulgens preclara of his Missa in die Paschae, or Mass for Easter day) is in three parts. The simple plainsong is notated with white semibreves in the middle part. The bass descant is notated in black semibreves, labelled with the proportion, 4:3 (sesquitertia). As discussed in the previous post (Ex. 13, Shelbye’s Miserere), Morley, in his Introduction, pp. 91-92, explains that ‘sesqui altra [3:2] and sesquitertia [4:3] they [the old descanters] denominated after the number of blacke semibriefes set for one note of the plainchant’. The treble descant in Preston’s Stirpe is comprised of short motifs, some of which correspond rhythmically with the plainsong (such as the first four notes: crotchet, two quavers, crotchet, a-g-a-c-b) and some of which correspond with the bass descant (such as repetitions of ‘dotted crotchet—quaver—crotchet’ at plainsong notes 10-12 and 25-26). The unpredictable occurrence of the motifs contributes to the cross-rhythm effects of the verset. For example, the pairs of quavers do not always fall as the second element as they do in the opening motif (see plainsong note no. 4, where the quavers comprise the fourth element); the duple rhythm ‘dotted crotchet—quaver, dotted crotchet—quaver’ at notes 26-28 (following the triple-rhythm ‘dotted crotchet—quaver—crotchet’) does not occur against one white semibreve of the plainsong: it straddles two of them.
Preston’s Missa in die Pasche includes other examples of proportions. Towards the middle of Confitemini (in the Gradual, Haec dies) the plainsong is in the lower of two parts: from plainsong notes 61-75 the descant is in the proportion ‘subsesquialtera’ or two notes to three of the plainsong (2:3); from 76-85 it is in ‘sequitertia’ (4:3); and from 86-117 it is in ‘dupla superbipartiens tertias’ (8:3), almost a lesson in how to break relatively easy proportions into progressively more difficult ones. By Morley’s day such proportions sounded old-fashioned and were open to ridicule:
those superparticulars and superpartients carry great difficultie, and haue crept into musicke I know not how, but it shold seeme, that it was by meanes of the Descanters, who striuing to sing harder waies vpon a plainsong then their fellowes, brought in that which neither could please the eares of other men, nor could by themselues be defended by reason (Introduction, Annotations to the First Part).
This is the twelfth of a multi-part post. (Please see previous posts below for earlier installments.)
Example 13, Miserere by William Shelbye (d. 1584, of Canterbury Cathedral) is in three parts. It has the plainsong in the top part with each note repeated (a kind of plainsong figuration). In the first half of the verset the descant in the lowest part is labelled with the proportion 9:2 and the descant in the middle part is labelled 3:2. In the second half, the proportions of the two descants are exchanged (a process also used in Ex. 15).
One kind of plainsong figuration has been discussed in earlier posts (see Ex. 5b, Burton’s Tibi omnes, Ex. 7c, Redford’s Te prophetarum, and Ex. 11, Blitheman’s Eterne rerum I), involving breaking a simple note into two or more notes of shorter duration that equal its note-value. This can be done in a basic way (two minims replacing a semibreve, for example), or elaborately, to the extent that the plainsong is made into a melody like the descant, as described by Morley (Introduction, p. 90): ‘making your plainesong as [i.e. like] your descant notes, and so making [counterpoint] vpon it’. Another kind of plainsong figuration is ‘making it [each note of the plainsong] two long, three long, et cetera. Or three minimes, fiue minimes, or so forth, two minimes and a crotchet, three minimes and a crotchet, fiue minimes and a crotchet, et cetera’ (Morley, Introduction, p. 90).
In Shelbye’s Miserere each note of the plainsong is made two semibreves. Without the repetition of each note of plainsong the proportions against the two descants below would be 3:1 and 9:1. However, Shelbye, like other Tudor composers, enjoyed syncopation, and by repeating each note he creates the effect of the proportion ‘sesquialtera perfect’ (2:3) against the middle part, and ‘quadrupla sesquialtera’ (9:2) against the lowest part.
According to Morley (Introduction, pp. 91-92) ‘sesqui altra [3:2] and sesquitertia [4:3] they [the old descanters] denominated after the number of blacke semibriefes set for one note of the plainchant’. His example of ‘sesquialtera perfect’ is shown in Fig. 4 (bars 1-2); ‘in the third barre you have broken sesquialtra, & the rest to the end is Quadrupla sesquialtra, or as they termed it, nine to two’. A contrapunctus of Morley’s example, in which bars 3-7 are reduced to the basic 3:2 (‘sesquialtera perfect’), is shown in Fig. 5.
Figure 4 Morley, ‘sesquialtra’: three notes of descant (three black semibreves, in the top staff) to two semibreves of the plainsong (lower staff):
Figure 5 A contrapunctus of Morley’s example of sesquialtera (Fig. 4):
Note that the only dissonance in Fig. 5 is a 7-6 suspension in bar 2. In bars 4-6 of Fig. 4 (9:2) the descant comprises stepwise motion (creating several passing-tones) and consonant leaps.
Fig. 6 gives an example of 2:3 (subsesquialtera), Burton’s Te ergo, which has two descant notes for every three of the plainsong (a faburden), which is in the lower part. The dissonances are as follows: at plainsong notes 5, 17, 26, a 4-3 suspension; at note 8, a 7-8 bass suspension; at 29 a 7-6 retardation (a bass suspension resolving upwards).
Figure 6 Avery Burton, Te ergo (Te Deum, 29996, fol. 24r):
Example 13 William Shelbye, Miserere (30513, fols 47v-48v):
This is the eleventh of a multi-part post. (Please see previous posts below for earlier installments.)
Example 12, Blitheman’s Gloria tibi trinitas II (the second of his six settings of the antiphon in the Mulliner Book) is mostly in two parts with a final section in three parts. It has a simple plainsong in the top part. For plainsong notes 1-15 there is one descant below with 8:1; for plainsong notes 16-26 there is a change of proportion to 6:1; and for several notes (17, 19, 21-26) the descant consists entirely of consonances. At plainsong note 27 a second descant is added that is also 6:1 against the plainsong, i.e. the two descants are 1:1 in relation to each other; they frequently move in parallel 3rds or 6ths (as in Ex. 4b).
As discussed in Part 1 of this blog (Fig. 2), Tudor choristers began their study of descant by singing consonances (3rd, 5th, 6th and octave) above a plainsong. Lionel Power’s treatise ‘contriued vpon the gamme [gamut]’ for teaching children deals only with this stage in their education. Singing descant below plainsong is significantly more difficult to do. The consonances are the same, of course: as Morley explains (Introduction, p. 86), in ‘bass descant’, either you may reckon your cordes [i.e., intervals] from your base vpwardes, or from the plainesong downewarde, which you list. For as it is twentie miles by account from London to ware, so is it twenty from Ware to London’. But as Thomas Campion points out (A new way of making fowre parts in counter-point, London, 1610 [C2r-v]) when a consonance is inverted it can turn into a dissonance (such as a consonant perfect 5th inverting to a dissonant perfect 4th):
Such Cords as stand aboue the Notes of the Base [i.e. plainsong] are easily knowne, but such as in sight are found vnder it, trouble the young beginner; let him therefore know that a third vnder the Base, is a sixt aboue it, and if it be a greater [major] third, it yeelds the lesser [minor] sixt aboue; if the lesser third, the greater sixt. A fourth vnderneath the base is a fift aboue, and a fift vnder the Base is a fourth aboue it.
Figure 3 Intervals and their inversions (M = major; m = minor; P = perfect):
The process of learning bass descant is made much easier by physically playing intervals and their inversions on a keyboard instrument: a perfect 5th played on a keyboard not only fits neatly under the hand, it also looks similar to the lines of a staff turned sideways to the right, and therefore makes ‘sighting’ intervals concrete. Ex. 1 in Part 1 of the blog, Blitheman’s Dignare, which is 3:1, all consonances, could be used as a lesson in how to practice playing only consonances below a plainsong.
(Note: Ex. 12 is recorded up 4th, since the lowest note on the Wingfield organ is F.)
Example 12 Blitheman’s Gloria tibi trinitas II (antiphon, 30513, fols 90r-91v):
This is the tenth of a multi-part post. (Please see previous posts below for earlier installments.)
Example 11 is Blitheman’s Eterne rerum I, the first of four versets of this hymn in the Mulliner Book, is in two parts. The lower part (based on the plainsong) is in broken semibreves, down an octave from the plainsong as it was notated in chant books, given here on the third staff. (See Parts 5 and 6 on plainsong figuration.) The figuration gives the impression of an elaborate piece with imitation between the parts (for example between the simple plainsong notes 3-9, and between 25-29) but the simple (i.e. unornamented) plainsong played down an octave fits contrapuntally with the descant with only five places that are problematic.
For learning how to descant on plainsong figuration, a small amount breaking can be used to solve these five problematic places, so that the descant may be played in its original octave with a much simplified plainsong, and more breaking added gradually over time (using Blitheman’s verset as a guide):
At plainsong note no. 13 there is a leap in the descant from a crotchet c which is a 4th against the plainsong, and although escape tones were allowed in Tudor style in short note-values, the progression leading to the 4th is in similar motion, which gives it too much emphasis. One solution is to play minims C-G in the plainsong instead of the semibreve G (i.e. similar to the breaking in the verset, a C held over from the previous note followed by minim G).
Between plainsong notes 15-16 there are parallel 7ths—again, ‘two discords together’ were allowed in short note-values, but the second note is a minim. One solution is to play minims Bflat-A in the plainsong instead of the semibreve A (cf. the verset, which has four crotchets, Bflat-C-A-D).
At note no. 19 there is another crotchet escape tone, but the leap is unusually large–an 11th. One solution is to play minims Bflat-A in the plainsong instead of a semibreve A (cf. the verset, which has two crotchets C-Bflat followed by a minim A).
Between notes 20 and 21 there are parallel octaves, which are forbidden. One solution is to play minims G-Bflat intead of semibreve G at note 20 (cf. the verset which also breaks note 21 into minims).
Between notes 26 and 27 the descant has two minims, bflat’-a’, the second of which is dissonant above the G in the plainsong, too long a note-value to be acceptable as a passing-note. One solution is to play minims G-A in the plainsong instead of a semibreve G (cf. the verset, which has four crotchets, G-D-C-A).
In the middle of the verset the descant includes a few repetitions of an offbeat rhythmic motif that was a favourite with Tudor organists (dotted crotchet—quaver—crotchet); see Redford’s Dignare (Ex. 9), Preston’s Stirpe (Ex. 14), and Heath’s Christe qui lux (Ex. 15). Towards the end of Eterne rerum, the plainsong figuration includes syncopated patterns while the descant has a run of 8:1.
Note: In the recording, the left hand part is played up an octave towards the end, when the music goes below F, the lowest note on the Wingfield organ.
Example 11 Blitheman’s Eterne rerum I (30513, fols 55v-56r):
This is the ninth of a multi-part post. (Please see previous posts below for earlier installments.)
Example 10, Redford’s Salvum fac is from the same Te Deum as Ex. 9, and, like all of its versets is also ‘on the faburden’. Here the faburden melody is in the middle part, but as it is doubled in parallel 6ths above, the plainsong itself sounds in the top part. The faburden melody is extended towards the end of the verset: instead of ending with the repeated note g (corresponding with the repeated e” of the plainsong notes sounding above) the notes c’-b’-a’ are inserted between them, to help make the cadence or ‘close’ (the term used by Tudor musicians). Such extensions of the close are typical of Tudor organ versets: the last note, the penultimate note, or the antipenultimate note is often lengthened and/or decorated melodically or rhythmically.
In Salvum fac the descant in the lowest part begins on its own with ‘fore-imitation’ (E-F-G-a) of the first three semibreves of the faburden (e-f-a’). For plainsong notes 1-17 the lowest part has a descant that is mostly 4:1 before the proportion changes to 3:1 against the faburden/plainsong. In the manuscript the change of proportion is indicated by the figures 3:2 and the use of coloration: three black minims take the place of two white minims. At the end of the verset, the proportion is changed to 2:1, and a fourth note is added to each of the last two sonorities, making a fuller texture.
Example 10 Redford’s Salvum fac (Te Deum, 30513, fol. 63r):
This is the eighth of a multi-part post. (Please see previous posts below for earlier installments.)
Example 9, Redford’s Dignare is from his Te Deum (30513, fol. 63r), in which all the organ versets are based ‘on the faburden’ of the plainsong, rather than on the plainsong notes themselves. In a basic ‘text-book’ faburden melody each phrase usually begins and ends on the same notes as in the plainsong, and otherwise the notes are usually put a 3rd up from the notated chant, although a unison with the chant may be used at any time. Faburden melodies are therefore variable, depending on what the performer/composer wishes, and sometimes even the notes at the beginning and ends of phrases are put up a 3rd. The example given by Morley (Introduction, Annotations ‘Upon the second Part’ for page 70; several copies are available at ISMLP) is of the plainsong Conditor alme siderum with ‘The faburden of this hymne’, in which the last note of each phrase (not the first) is the same as the chant; Morley also breaks the last few notes of the faburden into smaller note-values. Since Morley is describing the ‘old’ three-part improvised faburden, he comments that ‘And though this [the faburden] be prikt [i.e. notated] a third aboue the plainsong, yet was it alwaies sung vnder the plainsong’. Using this method, the faburdener would ‘sight’ in unison with, or a 3rd above, the notated plainsong and then ‘in voice’ transpose them all down an octave; alternatively the plainsong itself would be sung up an octave. Whichever way the transposition was done, the resulting sonority consisted primarily of parallel 6ths. Between these two parts another voice sang parallel 4ths against the plainsong.
When the faburden melody was to be used instead of the plainsong as the basis for improvised descant it was sometimes written out, making it easier to sight consonances from; see Lbl Harley MS 2945. Another method of making improvising on faburden easier is shown in a printed Sarum hymnal (Ruremond, 1528), in which the faburden is indicated by dots written above the plainsong (cf. Morley’s example).
Redford’s Dignare has a 1:1 descant below, and the top part has an ostinato point consisting of six repetitions of the rhythmic motif ‘dotted crotchet—quaver—crotchet’ (followed by a minim), which creates cross-rhythms against the plainsong minims. (On ‘points’ see Part 4.) Melodically this point consists of three repetitions of a six-note motif: a consonance, an upper neighbour, a return to the consonance; a repeat of the same consonance which then becomes a dissonance (a suspension) that is resolved with a ‘jumped resolution’ (i.e., a step up and then a leap down of a 3rd). Bathe gives examples of two suspensions (‘binding descant’), the second of which has a jumped resolution: 
William Bathe, True Art of Musicke,‘somtymes for the [part] of a not[e] a discord is vsed: commonly either binding, or with a prick [dotted note] as for example:
In Redford’s point these six notes are repeated twice, ending with a minim c’ (i.e. decorating the note c’); note the ‘barline’ placed after the point. The point then occurs on the note a, followed by another ‘barline’. For the third occurrence of the point it is shortened and altered to form the close; Morley, Introduction, p. 95, uses the term ‘dissolved’ for this procedure.
Example 9 Redford’s Dignare (Te Deum, GB-Lbl Add. MS 15233, fol. 6r):
 See Jane Flynn, ‘Tudor organ versets: echoes of an improvised tradition’, Journal of the Royal College of Organists n.s. 3 (2009), 5-23 (at 21-22).
 See Bathe, True Art of Musicke, ed. Karnes, p. 125.