Dunkeld Music Book: Chapter 4

Chapter 4
Text truncations

The three mass settings in Eu 64 all display examples of text truncation. As these are all somewhat unusual in their characteristics a wider study has been undertaken in order to place them in context and to identify the differences and reasons for these cuts.

In masses from these islands of the period (i.e. Old Hall Manuscript to before Byrd) it was usual for the section 'Et in Spiritum Sanctum' to 'et expecto resurrectionem mortuorum' (and often more) to be cut, resulting in a loss of around one-third of the text. These deletions occur in most of the Tudor English settings and in all but one of the Scottish Masses in the Carver Choirbook and the Dunkeld Music Book. The Credo truncations are generally of the section concerning the Holy Spirit and the Church which might possibly indicate doctrinal differences within the local Church – lex orandi, lex credendi: what we believe affects how we worship.

Ruth Hannas (1952) has made a study of deletions in the Credo of polyphonic Masses. Within a valuable survey of credal formulae, 'the sacred pivot of the service' (1952, p. 155), from the early Church fathers to the time of the Reformation, she makes the controversial argument that such cuts were 'a faithful, datable, reflection of political, sociological, and religious struggles (1952, p. 155). She refers briefly (1952, p. 157) to the practice of 14th-century deletions in the plainchant Credo, ascribed by Max Sigl to the 'misadventure of scribes'. She considers the practice to be a reflection of the differing creeds formulated at various times by, and in response to, different factions within the Christian Churches (1952, p. 157, 159, 180-2). She finally attributes 'such deletions in the polyphonic masses to insecurity caused by vacillating policies during the reigns of Henry VIII and Edward VI (as paraphrased in Turbet, 1994, p. 25). Indeed, Elliott (1996. p. ix), writes that

as the mass [sic] became an increasingly uneasy instrument of liturgical observance, it is little wonder that large sections of the text were omitted from Carver's polyphonic setting of 1546 [Pater Creator omnium]. (There may of course be a simpler explanation for these omissions – purely musical reasons.)
Elliott has briefly alluded to what is presumably the liturgical composer's intention, viz. to create a musical work based on a given text. This view is taken by, amongst others, Hugh Benham (1977, p. 12). and E. Moohan. Moohan has 'examined some 250 works of the fifteenth and early sixteenth centuries and found no reasons for these omissions except a purely musical one' (cf. Elliott, 1996, p. xii).

Although this is likely to be true, Hannas' study may be of more value than first thought in attempting to establish the origins of text deletions, even if the musical intentions are clear. It is possible that her article has been misrepresented. When she writes 'deletions in the Credo of Masses composed during the reigns of Henry VIII and Edward VI may safely be attributed to the insecurity caused by vacillating policies', this statement means what it reads, and only that. It would be unjust to suggest, as does Turbet (1994, p. 25, perhaps following Benham, 1977, p. 225), that she is claiming that this is the cause of all text truncations, as her article makes clear.

We should now establish the musical practice of text truncation which, as we have seen, goes back to the Use of Salisbury. Benham (1977, p.13) and Wulstan (1985, p. 259) suggest that the practice of telescoping of the texts, i.e. singing two parts of the text simultaneously, probably arose from a continental manuscript in which a scribe failed to fill in the words of the second part, only to be followed by an imitator's slavish representation of his copy (Trent Codices 90 and 93); the omission is not in the English source. This tradition of Credo omissions from Dufay onwards may then have been re-imported into England. If, as Wulstan (1985, p. 284) states, composers 'often took their words from earlier settings…rather than seeking their text in the Service books' then these musical origins for the cuts could be accounted for.

Liturgical musicians know the common texts of the Mass and other services so well that they can recite or sing them from memory with little thought. As with all well-known texts, however, whilst an addition would stand out 'like a sore thumb', it is possible to miss a line without realising the omission. Hannas (1952, p. 169) sees in the Old Hall MS. 'a picture of pre-Reformation and Reformation insecurity … [in which] there is at times an obscurity of words which may have been intentional'. Perhaps, after all, as Benham (1977, p. 225) says, Hannas is being a little over-imaginative.

No matter the reasons for the deletions, the Council of Basle in 1431 passed resolutions requesting the clear and complete singing of the Credo, without any textual omissions, in church services (Strohm, 1993, p. 251).

Benham (1977, p. 12) identifies four standard patterns in Credo cuts in English Mass settings (none are to be found in Continental settings) which may be due to local preference or current fashion. The first three of these cut from 'Et in Spiritum Sanctum Domium' returning at 'Et expecto', 'Et vitam venturi saeculi' and 'Amen'; the fourth is longer, beginning at 'Et iterum venturus est cum gloria'. Occasionally there is also a section in the first half of the Credo omitted: 'Deum de Deo … de Deo vero'.

The one Scottish Mass in which the Credo is set complete is the Mass Pater Creator omnium in which Kenneth Elliott (1996, p. 54ff) has reconstructed the music, based on the chant but set in faburden according to the rules set out in a later manuscript, the Airt of Music Collecit out of all Ancient Doctouris of Music (GB-Lbl Add. 4911) known as 'Scottish Anonymous' of uncertain date between 1558 and c. 1580. The Mass Pater Creator omnium is complete precisely because it is a harmonisation of the chant (US p. 66, Credo IV in the current Graduale Romanum). In the Gloria only the beginning and ending are set, the rest is sung to chant.

What is highly unusual about Eu 64, however, is that text omissions occur not only in the Credo but also in the Gloria. The reader is referred to Appendix 6, a table showing truncated and troped texts in the Carver Choirbook, the Dunkeld partbooks and in a number of contemporary English Masses.

Other than the Mass à3 sine nomine (and Carver's Pater Creator omnium, already mentioned) in the Carver Choirbook, the only masses with any Gloria cuts seem to be those in Eu 64, which might indicate a Scottish practice, or perhaps a local custom in part of Scotland. However, the English Mass Jhesu Christe by Ashewell also has Gloria cuts. It could be asked whether this is coincidence or whether there was some link between the English composer, cantor at Durham Cathedral in 1513, and the anonymous composers of the Eu 64 Masses.

We might also ask how and why Jhesu Christe was copied into a Scottish manuscript. Ross (1993, p. 6) says that English settings of the Magnificat and Salve Regina were probably copied into the Carver Choirbook around the time of Margaret Tudor's marriage to James IV in 1503 and, as we saw earlier earlier, it is likely that other music was brought north in the wedding party. Although this date is too early for Jhesu Christe (Ashewell was only about 15 years old in 1503), it does indicate that there were cultural links between the 'auld inemeis' even if there were real political difficulties.

In addition to the usual English practice of omitting a large section of the Credo, 'Et in Spiritum Sanctum … in remissionem peccatorum' there are other cuts which are unique to Jhesu Christe. In the Gloria, Ashewell omits 'Domine Fili unigenite', in the Credo, 'et invisibilium' and 'qui propter nos homines et propter nostram salutem descendit de caelis'. The reason for omitting 'et invisibilium' in the Credo seems to be simply a mistake. There would appear to be no doctrinal reason for missing out 'and [maker of all] invisible [things]'.

The only other complete surviving work by Ashewell is the Mass Ave Maria (also for six voices) which appears, with the full text of Jesu Christe, in the Forrest-Heyther partbooks copied by or for Taverner in 1526. In Ave Maria the only cut is in the Credo, at 'Et in Spiritum Sanctum … Et expecto'.

Anonymous: Mass à3

The anonymous Mass à3 from the Carver Choirbook (in MB xv, pp. 1-8 and MS, pp. 240-254) omits most of the text of the Gloria and the Credo.

Anon: Felix namque

Highly unusual for a British Mass of this period is the setting of the Kyrie, although it is by no means unique; it was more common in 15th-century England than was once thought. When it is set, it is usually troped following the Use of Sarum. Here it is set in the non-troped, Continental fashion.

The Gloria cut is from 'qui tollis peccata mundi, suscipe…' to 'tu solus altissimus, Jesu Christe'. The Credo cut begins not at 'Et in Spiritum Sanctum' but at 'Et resurrexit' (these two words appear in only one voice) thus omitting about 50% of the text. More significantly, almost all reference to the glorification of Christ, his resurrection and ascension, is cut. There is an example of parallel texts at

et homo factus est. Crucifixus etiam pro nobis
perhaps showing English influence. The Benedictus trope 'Marie filius' appears in this setting, as in Carver's Pater Creator omnium and in the Sarum Masses on Saturday of the Blessed Virgin Mary (see forthcoming publication in EECM 58)

Anon: Cantate Domino

As mentioned earlier, this Mass setting has common elements with two other Scottish Masses, with similar cuts in the Gloria and Credo. In the Gloria the truncation begins a line earlier than in Felix namque, at 'qui tollis peccata mundi, miserere nobis', whilst the Credo is even more brief, omitting the entire central section concerning the humanity of Christ: incarnation, birth, death, resurrection and ascension.

Sheppard: Be not afraide and Carver: Fera pessima

These two Masses both have cuts in the same place, and none other. In the Credo the cut begins at 'Et iterum venturus est cum gloria' instead of 'Et in Spiritum Sanctum'. One other Mass, Taverner's O Michaell, begins the truncation similarly but sets 'cujus regni non erit finis'. The question then arises as to whether there might be any link between the Masses Be not afraide and Fera pessima on the basis of identical text omissions; a further search of the repertoire reveals, however, that Tye's four Masses The Westron Wynde, Mean, Peterhouse, and Euge bone have identical cuts.

In addition to the question of text truncation in Scottish Masses, one could reasonably ask about the setting of the Kyrie in some Masses. In light of the prediliction for text truncation, together with the wide variety of forms manifested, perhaps the question we should be asking is why some texts were left in, rather than cut.


35 Whether the music settings were cut or not, presumably, and we have no evidence to suggest otherwise, the texts were recited completely by the celebrant at the altar whilst the music was sung by the choir.
36 A notable exception to the norm was Nicholas Ludford's 'demonstrated prediliction for setting the Credo complete' (Bergsagel, 1962, p. 41)
37 Whilst it is true that traditional theological understanding of Church authority was rigourously attacked during the reign of Henry VIII, the English king himself remained a Catholic all his life, and little could be done to change the liturgy (Harrison, 1968, p. viii), a situation which was also beginning to emerge north of the border, (cf. Wormald, 1981, p. 99). It was only with his son's succession that the theological and liturgical reformation took hold, resulting in the two prayerbooks of 1549 and 1552 (Harrison, 1968, p. vii) which incorporated Cranmer's re-working of existing materials, including the daily office provision based on the Quiñones Breviary. The second prayerbook was a far more drastic revision in which the severity of Calvinist theology is considerably more apparent.
38 It is all too easy to pick up a reference in a secondary source and to allow one’s argument to 'colour' what one reads; the necessity of looking to primary sources whenever possible is paramount in research.
39 cf. Maynard (1961, vol I, p. 6-15), Elliott & Rimmer (1973, p.26f.) Woods (1988, p. 37f.)
40 GB-Ob Mus.Sch.E.376-81.


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