Related sources

Other sources

by first line
   by composer



Papers and notes

General Index of music editions
by first line
   by composer


Editions and papers
on this site:

Complete Works of Gilles Mureau

Amiens MS 162 D

Sacred music of the 15th century

Uppsala MS 76a

Peter Woetmann Christoffersen

Papers on

Basiron’s chansons
Busnoys & scibes PDF
Chansons in Fa-clefs
Chansoner på nettet
Fede, Works
Dulot’s Ave Maria
Open access 15th c.
MS Florence 2794


Ave regina celorum 3v · Frye, Walter

Appearance in the group of related chansonniers:

*Laborde ff. 8-9 » Ave regina celorum« 3v (beginning of S is missing) PDF · Facsimile

*Leuven ff. 1v-2 »Ave Regina celorum« 3v PDF · Facsimile

*Wolfenbüttel ff. 1-2 »Ave regina celorum« 3v ((beginning of S is missing) PDF · Facsimile

Other sources:

Berlin 78.C.28 ff. 47v-49 »A« 3v
Bratislava 33 »Ave regina celorum« 3v
Florence 112bis ff. 29v-30 »Ave regina celorum« 3v
Florence 2794 ff. 15v-16 »Ave regina celorum« 3v
Hradec Králové II A 7 pp. 408-409 »Ave regina celorum« 4v (contra (I) added)
Kraków 40098 ff. G3v/G9v/G11v »Ave regina celorum« 3v · Facsimile
Munich 5023 ff. 12v-13 »Ave regina celorum« 3v (parts of S and T)
Munich 810 ff. 37v-39 »Ave regina celorum« 3v wal frey · Facsimile
Perugia 431 ff. 82v-83 »Ave regina celorum« 3v · Facsimile
Sevilla 5-1-43 ff. 37v-38v »Ave regina celorum« 3v (lacking end of T and C) Frye
Trento 90 ff. 298v-299 »Ave regina celorum« 4v (high superius added) · Facsimile
Trento 90 ff. 371v-372 »Ave regina celorum« 3v · Facsimile
Verona 757 ff. 53v-55 [Without text] 4v (contra (II) added)

Intabulations, citations, appearance in paintings etc., see Fallows 1999, pp. 572-574.

Edition: Gutiérrez-Denhoff 1988 no. 1 (Wolfenbüttel).

Text: Latin antiphon (14th century); LU p. 1864; RH 2072; Mone 1855, vol. II, p. 202.

After Laborde and Leuven:

Ave Regina celorum, 1)
mater regis angelorum,
O Maria flos virginum
velud rosa vel lilium.

Funde preces ad dominum
pro salute fidelium,
O Maria flos virginum 2)
velud rosa vel lilium.

Hail, queen of heavens,
mother of the king of angels.
O Mary, flower of virgins
like the rose or the lily.

Carry prayers to the Lord
for the salvation of the faithful.
O Mary, flower of virgins
like the rose or the lily.

1) Laborde, lines 1-4 are missing.
2) Leuven, lines 7-8 are missing.

Evaluation of the sources:

The main scribes of the Laborde, Leuven and Wolfenbüttel chansonniers used different exemplars when they entered this Latin song. It was the opening song in all three manuscripts. The exemplars used for Laborde and Wolfenbüttel were clearly related, while the one for Leuven belonged to a different tradition.

Beautiful painted initials surely adorned the first pages with music in Laborde and Wolfenbüttel. These pages have been cut out by later owners. Also the following folio in Wolfenbüttel was mutilated, when the initial for the upper voice was torn out. The missing sections of the superius parts are easy to reconstruct by consulting Leuven and the repeat of the music in the song’s second section (bb. 13-22 = 37-46).

Laborde and Wolfenbüttel exhibit the usual variance in their use of ligatures and coloration, and they have small variants in the music (S bb. 26 and 44, T b. 6, and C b. 3). Moreover, Laborde probably had the full Latin text in the superius, while Wolfenbüttel does not repeat the two last lines “O Maria ...”. Instead lines 5-6 are distributed to cover the whole second section. Wolfenbüttel is without any key signatures, Laborde likewise, but a very discreet b-rotundum at the start of the tenor voice points out that a key signature of one flat must be assumed in the lower voices.

The Leuven version has a one flat key signature in all three voices, continually in tenor and contratenor, but on and off in the highest voice. This produces a performance very similar to the one suggested by the notation in the Laborde chansonnier (cf. the editions). Its music shows up some traits that are unique for this source, for example in the superius bb. 30.3-31.1, which eliminates the dissonant f’ found in other sources, and in b.34.3 the dissonant g’ is avoided. In the contratenor there are rhythmical changes in bb. 3-4, and bb. 17.2-18.1 and 41.2-42.1, which are identical in most other sources, are here changed in different ways.  

Other differences compared to the Laborde-Wolfenbüttel version seem somewhat to set apart traditions of circulation. Bar 9.3 in the Laborde contratenor has a highly dissonant minima a (superius and tenor have b’ and g respectively), which is also found in Wolfenbüttel, in the three-part version of Trento 90, the slightly younger French chansonnier Florence 2794 and in the Italian MSS, Sevilla 5-I-43 and Perugia 461. Leuven and the four-part version in Trento 90 along with the embellished version in Glogauer Liederbuch (Kraków 40098) remove the dissonance by singing b. In Laborde the contratenor starts bar 13 with a semibrevis c followed by a minima f, and we find the same in Wolfenbüttel and Florence 2794. Leuven fills the space with the figure minima a - semibrevis f, which also is in most other sources. These two points alone seem to indicate that Laborde, Wolfenbüttel and Florence 2794 belongs to a French tradition from the 1460s and 1470s, while Leuven along with Trento 90 (4v) and Kraków 40098 represent a different tradition, and that the Italian sources like Trento 90 (3v) mix the traditions.

The misreading in the contratenor’s bar 41.2 of a punctus prolonging a semibrevis as a minima rest crops up in both versions of the song in Trento 90, and is found in the Italian MSS Berlin 78.C.28 and Florence 112bis (cf. Gutiérrez-Denhoff 1988, p. 109) – and it was in the exemplar used for the Leuven version. The Leuven scribe made a different interpretation of the next few notes, just like he did in the parallel passage in bars 17.2-18.1.

The majority of sources have no key signatures. Both versions in Trento 90 and the MS Perugia 461 have a key signature of one flat in the tenor, while the scribe of Sevilla 5-I-43 has put in some accidentals in the tenor having a similar effect. According to Gutiérrez-Denhoff 1988, p. 108, the Florentine MS Berlin 78.C.28 has one-flat key signatures in the lower voices. It is obvious that the song has to be performed with an almost constant inflection of the F-mode’s fourth scale degree. The scribes are not in complete agreement on how much guidance they have to give the singers. The Leuven version’s explicit use of key signatures in all three voices matches its unique musical variants. In most cases the scribe or his exemplar has sought for a sound more consonant than what we find in other sources: The Leuven version stands as a special development of the tradition it belongs with.

The picture drawn by the sources for this widely circulated song is blurry and confusing. It sets out that Leuven, which in some ways stands alone, and Laborde-Wolfenbüttel belong to two different, but vaguely defined traditions. However, taking the words in account, the three “Loire Valley” chansonniers agree on using the usual antiphon text in line 5 “Funde preces ad dominum”, where all other texted sources have “Funde preces ad filium”.

Comments on text and music:

The very popular prayer to the Virgin appears in many sources originating from widely dispersed locations in France, Central Europe and Italy. Moreover, it appears in intabulations, its tenor is used in other compositions, and its notation is as an easy recognizable symbol for the prayer in three paintings (see the list in Fallows 1999). It is song-like in character, composed for two voices an octave apart, which for long stretches move in parallel tenths, thirds and sixths, enlivened by short snatches of imitation, with the contratenor supporting the structural duet around or below the tenor.

It has been proposed and to some degree accepted that the song was a contrafactum, that Walter Frye composed the song as an English ballade, the original of which has not survived. Its unusual form (ABCB) with a two-line refrain could certainly be explained as a simplification of its secular formal layout, when it was recreated as a prayer song with the well-known antiphon text (cf. Sylvia W. Kenney, ‘Contrafacta in the Works of Walter Frye’, Journal of the American Musicological Society 8 (1955), pp. 182-202). However, the close relationship between the Latin words and musical lines of the upper voices might be an argument that the song was composed for the prayer. It opens with melismatic “Ave” followed by a homorhythmic recitation of “regina”, also the words “mater regis” stand out; and in the repeated refrain the tenor is organized in four-note segments exactly fitting “O Maria / flos virginum”, which the superius plays with in free imitation. Thus the song might be an original musical prayer created by Frye, which inspired other small Latin songs appearing primarily in secular collections.

The euphonious character of its two-part structure with the tenor clearly in the primary and its unusual form may be inspired by contemporary simple polyphony sung in monasteries and confraternities. Here we find stanzaic Latin songs with a variety of refrains, often starting with “O ...”; mostly in unmeasured simple notation, but coloured by strings of thirds or sixths, see further the introduction to my edition, Songs for funerals and intercession. A collection of polyphony for the confraternity of St Barbara at the Corbie Abbey. Amiens, Bibliothèque Centrale Louis Aragon, MS 162 D, especially part II ‘The simple polyphony in Amiens 162’ [] or pp. 44 ff in the PDF-version []. “Ave  regina celorum” could be a stylized version of this tradition made by a professional English musician, which made its way into the Continental repertory around the middle of the century. This explains the uncertainties shown by the different scribes of the sources and its unusual formal layout just as well as the theory that it was a contrafactum.

PWCH November 2017