Dunkeld Music Book: Chapter 5

Chapter 5
The supposed link between the Masses Fera pessima and Cantate Domino

A performance of ‘Carver’s Mass Cantate Domino’ (at Greyfriar’s Kirk on 19th August 1997) was announced as part of the Edinburgh Festival and reviewed two days later by Malcolm Hayes in The Scotsman.

This ascription has come about largely through the claims of Kenneth Elliott (1966, pp. x-xi, 297) who says that Cantate Domino

seems to be directly related to the five-part Mass Fera pessima [and] is a reworking of about 1525 for six voices, possibly by Carver himself, of the earlier five-part composition: some thematic material (especially the motive marked in the cantus firmus [see below, motive 'a'] is common to both works, though the six-part shows a more assured technical command…. It is an impressive work, and if by Carver – it is certainly very good Carver.

His arguments are taken up by James Ross (1993, p. 54) who writes that Cantate Domino

seems to be a fundamental reworking of melodic material used in the Mass Fera pessima, but the Phrygian modality has been abandoned and another cantus firmus, which has so far defied attempts at identification, seems to have been adopted.

It seems that we are skating on thin ice if we are depending simply on some ‘fundamental reworking of melodic material’ – not to mention the different mode and cantus firmus – in order to link these two works and, consequently, to ascribe Cantate Domino to Carver. Of course, if the melodic material is fundamentally linked, then it would be fair to say that the Masses are too. But is this motivic link both fundamental and significant, and even if it is, why should this necessarily be an example of self-parody? Supporting stylistic material would need to be provided in order to prove this hypothesis.

The four-note ‘motive’ called ‘a’ (in Elliott, 1997, pages unnumbered)

d f g bb
with its transpositions
e g a c and a c d f
– an ascending phrase, consisting of two minor thirds separated by a tone – appears in tens of works that the present writer has seen, ranging from chant and polyphony to organ masses.

Appendix 7, an extract from John Bryden and David Hughes' An Index of Gregorian Chant (1969, p. 288f.), lists 51 examples of Gregorian Chant beginning with these four notes in one transposition or another. Additionally there are many examples of chants in which these four notes appear, including the following, which are among the most common and well-known of all chants.

Ave maris stellaLU 1261
A solis ortus cardineLU 400
Salve Regina (simple tone): 'gementes et [flentes]'LU 279
Te DeumSarum chant
Polyphonic works include:
FayrfaxMass O quam glorificaCMM 17 p. 64
SheppardMass CantateEECM 18 p. 10
SheppardMass Be not afraidep. 126
Organ works in EECM 10 alone include:
AnonFelix namquep. 54
RedfordAgnus Deip. 18
PrestonFelix namque IIp. 69
PrestonFelix namque IVp. 69

Among the polyphonic compositions which use this ‘motive’ is the Mass Felix namque, in his commentary on which Elliott (1960b, p. 45) describes ‘motive a’ (there entitled ‘theme d’) as a subsidiary motive, e.g. in the Credo, ‘Et in unum Dominum’ (MB xv, p. 63). The offertory chant Felix namque (LU 1271) does not include the complete ‘4-note motive’ – only the first three notes are used – so it might be interesting to discover why it occurs in works based on the chant. Perhaps this is the key to the whole issue here – that a common formula is simply extended in different ways.

Additionally ‘motive a1’ (the upper three notes):

f g bb
and its transpositions occur even more frequently, including the opening notes of Lupi’s Salve celeberrima virgo occurring earlier in the manuscript. Felix namque has many examples of ‘a1’ in addition to ‘a’. Unless we are going to ascribe all these works to Carver we probably have to accept that these ‘motives’ are the common property of many writers at that time.

Additionally Cantate Domino differs from Fera pessima in a number of highly significant ways. Both the mode and the cantus firmus are different, as is the number of voices. The mode of Fera pessima is Phrygian, that of Cantate Domino is hypolydian/hypoionian. Of course this does not by itself ‘write-off’ the possibility that Carver also wrote Cantate Domino; it is possible that he used common ideas in freshly composing the six-voice Mass. It is also possible, though, that the composer of Felix namque was also that of Cantate Domino, or that someone else consciously or unconsciously used these common materials in homage or emulation, a practice common in the Masses of the late 15th and early 16th centuries (Mayer Brown 1982, p. 46), so much so that it was often used in order to learn the art of composition.

Cantus firmus of Cantate Domino

Ross (1993, p. 54) informs us that ‘another cantus firmus, which has so far defied attempts at identification, seems to have been adopted’ for Cantate Domino. Hofman & Morehen (1987, 102) have cautiously identified this cantus firmus as an Advent antiphon ‘Cantate Domino et benedicite’ which appears in the Salisbury Antiphoner (Frere 1924, p. 121). My inspection of both the Antiphoner with its index and the expanded version of the index by Collamore and Metzinger (Frere 1990) show that this antiphon is the 3rd chant for the 2nd nocturn for matins on Fridays ‘Feria vi’ per annum, and that it is has been confused with the Advent antiphon 'Cantate Dominum canticum novum' for Fridays in Advent. Collamore and Metzinger have somehow omitted the first (of four) volumes of the Salisbury Antiphoner, beginning catalogue entries at p. 101, also omitting pp. a-z, A-Z, α-δ. Unfortunately, this antiphon bears only passing resemblance to what Elliott (1996, p. 297) has identified as the outline cantus firmus; there is certainly no appearance of the ‘4-note motive’ in that chant.

Bryden’s Index of Gregorian Chant has two entries with the melodic incipit beginning x24545797: Hodie vas electionis, (C24545797E) and antiphon in the fourth mode which appears in the 13th-century Worcester Antiphoner, and Pastor bonus animam suam, (G24545797E) to be found in the 12th-century Antiphonaire Monastique de Lucques (a variant reading) as well as the more accessible Antiphonale Monasticum, and Antiphonale Romanum. Perhaps these chants might shed some more light on this question.

Carver’s style

If the ‘4-note motive’ is insufficient in itself to determine common authorship of Cantate Domino and Fera pessima, we have to ask how stylistic evidence can help. Ross (1993, p. 7) claims that

Franco-Flemish polyphony is the basic stock of Carver’s style, but the spices which he blends into it are varied beyond expectation.

Although it is true that Scottish music was strongly influenced by Continental compositional techniques, it seems likely that some of the ‘spices’ are in fact English. Continental influence in England was different to that in Scotland. Tinctoris, in the 1470s and still true some 60 years later, wrote that

the French contrive music in the newest manner [i.e. imitation] for the new times, while the English continue to use one and the same style of composition, which shows a wretched poverty of invention (Strunk, 1981, p. 5).

Certainly, the flourishing state of art in England meant that composers were largely impervious to Continental styles. Scotland, in contrast, was strongly influenced, especially by compositional techniques. Jenkins, (1988, p. 97f.) asserts that during the reign of James IV (1473-1513), most of the influences were from England, largely due to the marriage of Margaret Tudor to James in 1503 when a cultural exchange – or at least a cultural import – is almost certain to have taken place.

The marriage of James V (1512-42) to two French princesses, Madeleine and, following her death, Mary of Guise, caused a more widespread Continental result, especially from Paris. As Jenkins (1988, p. 99) reminds us, however, we should not pretend that there are watertight categories about these two reigns, since Dufay’s Mass L’homme armé appears in the Carver Choirbook and Ashewell’s Mass Jhesu Christe and Vidi civitatem sanctam by van Wilder (a Netherlands composer active mainly in England) both appear in Eu 64, compiled during the reigns of James IV and V respectively.

We know from the Wode Partbooks (TWC1, p. 176f.) that Robert Fayrfax, who was associated with the English royal family, was highly regarded north of the border: he is represented in both the Carver Choirbook and the later Art of Music treatise as one of the great masters of polyphonic composition. Jamie Reid Baxter (1997, p. 15) considers that he might have travelled with the royal wedding party. The Carver Choirbook, which bears some resemblance to the Eton Choirbook, contains five English compositions by Fayrfax, Lambe, Cornysh and Nesbett.

Authorship of Felix namque and Cantate Domino

Interestingly, and perhaps wisely, Ross (1987, p. 22 and 1993, p. 54) is ambivalent concerning the authorship of Felix namque. He generally agrees with Elliott’s view that Carver was probably the composer of Cantate Domino:

it now seems likely that the greatest triumph of Carver’s later creative period is not to be found in the Carver Choirbook but in [the Dunkeld partbooks]. Such are the similarities in style and even in material between Carver’s Mass Fera pessima and the anonymous Mass Cantate Domino…that there seems little doubt that it too is the work of Robert Carver.

At the same time (1993, p. 77) he suggests

it is easy to picture the Mass Felix namque as the work of the mature David Peebles, who certainly had the necessary grasp of High Renaissance style to produce a piece of such consistently high standard. If this is the case, the Mass would fill the long gap between the two motets [Si quis diligit me and Quam multis, Domine, both in the Wode Partbooks] of 1530 and 1576 respectively… his only other surviving extended compositions.

Ross adds that there are some ‘curious points of contact’ (1993, p. 7) between Cantate Domino and Felix namque. We might well ask about these ‘similarities of style’ and ‘curious points of contact’, together with which ‘High Renaissance style’ he is referring to. Besides, a ‘consistently high standard’ is hardly an argument for particular composer-identification.

To return to the newspaper announcement of ‘Carver’s Mass Cantate Domino’ at Greyfriars Kirk. Neither newspaper editors nor the general public are too worried about academic debate as to whether X or Y composed a work, especially if Y is unknown, since Carver’s name is more likely to sell tickets than his poor relation Anonymous. Hayes (1997, page unknown), reviewing the concert, writes

With a starting time of 10:45pm and a ticket price of £11 for less than 45 minutes of music, this concert looked like a suspiciously nocturnal and expensive showcase for Carver’s music. Festival-goers thought otherwise. Greyfriars was packed for this performance of the Mass Cantate Domino – an arrangement, probably but not absolutely certainly by Carver, of an earlier mass [sic.]. …Carver’s setting (come on, on this evidence it must be his) took wing like an eagle in flight.

Indeed Baxter (1997, p. 1) writes of a general Scottish musical philistinism:

most Scottish concert-goers and music lovers, let alone anyone else, would be hard pressed to name a Scottish composer other than Robert Carver or James Macmillan.

And even if it is established convincingly that Z wrote the piece and that it should be called ‘Canticum novum’ rather than ‘Cantate Domino’ it may still take some considerable time for the average concert-goer to accept this, as is the case with Crux fidelis which is certainly not by John IV of Portugal, and Clark’s Prince of Denmark’s March, known as ‘Purcell’s Trumpet Voluntary’ by most wedding couples and clergy.


41 The name given to Carver’s Mass à5 sine nomine has altered a number of times. Since the relevant page of the Carver Choirbook has been trimmed at this point, the page now reads ‘Tenor a / pess / a’. In his dissertation (1960, p. 12) Elliott refers to it as ‘Carver’s large-scale festival mass à5’, and some thirty years later (1996, p. vii) as Fera pessima. Woods (1984, p. 225) calls it a pestilentia. Ross (1987, p. 24) gives it the title a pessinuntia, promising to explain the links between his title and the phrygian mode of the work in his forthcoming book (1993). There is no mention of a pessinuntia in the book, presumably because he had changed his mind in the intervening six years.
42 Its opening is based on what we would now call an inversion of this figure: cega(g)
43 Although both Sheppard Masses mentioned only have brief mention of the ‘4-note motive’ they both have variants, including e.g. with a passing note between the lower two notes, repeated or missing notes, and descending motives. Both use a similar Cantus tessitura to Cantate Domino which encourages the present author’s argument.
44 cf. Appendix 7 for explanation.
45 Tournai, 1922.
46 Tournai, 1906, c. 1934 (Desclée no. 818) and 1949 (Desclée no. 820) respectively.


You can contact Alistair Warwick at:
Ardarroch, Glen Road, Dunblane, FK15 0GY, Scotland
+44 (0)1786 823000 | +44 (0)7792 566349