Alma Redemptoris Mater, Gaude Maria, and the prioress's tale

Robert Boenig. Notes and Queries. London: Sep 1999.Vol. 46, Iss. 3;  pg. 321, 6 pgs

Abstract (Document Summary)

A comparison of the Antiphon's use of "Alma Redemptoris Mater" and "Gaule Maria" in Chaucer's "The Prioress's Tale" reveals a correlation with the surprisingly young age the author creates for his murdered protagonist.

Full Text (3582   words)

Copyright Oxford University Press(England) Sep 1999

DISCUSSION of Chaucer's use of the Marian Alma Redemptoris Mater usually centres around its words rather than its music, with much scholarly debate over the Church calendar and that Antiphon's prescribed seasonal use as a means of situating the events of the Prioress's Tale at a certain time of year.1 A comparison of that Antiphon's music, however, reveals something I feel is likewise important: a correlation with the surprisingly young age Chaucer alone among the sources and analogues creates for his murdered protagonist.

As is well known, there are many medieval versions of the boy who sings a Marian song while walking about and is murdered by the Jews and whose miraculous post-mortem singing solves the crime. It has been customary to divide the versions into groups A, B, and C as a means of determining affinity and lines of influence:2 Chaucer's is counted among those of Type C, chronologically the sixth item among its ten. C is moreover the smallest group, and it is the only one of the three to give Alma Redemptoris Mater as the piece the boy sings. Types A and B mention other Marian pieces, by far most frequently the Tract Gaude Maria. Of the five C versions earlier than Chaucer's two choose neither Gaude Maria nor Alma Redemptoris Mater: British Museum MS Additional 11579 (fo. 5^sup v^) has him sing the Antiphon Ave Regina, and Royal MS 12.E.1, article 19 (fo. 170) has him sing the Antiphon Sancta Maria.

There are thus three uses of Alma Redemptoris Mater among the versions prior to Chaucer's, so it appears safe to say that his intentionality was not behind the shift away from Gaude Maria. The first of the three is the thirteenth-century Corpus Christi College Oxford, MS 32, Article 19 (fo. 92), a piece in Latin prose which maintains: Puer uero clericus erat habens uocem clarem. In eundo autem & redeundo semper in omnium audiencia cantabat hanc antiphonam: Alma redemptoris mater etc [`Indeed there was a boy cleric who had a clear voice. In going and returning he used to sing in the hearing of all this Antiphon: Alma Redemptoris Mater, etc.'].3 The singer is identified as a 'boy' who is a 'cleric', and Alma Redemptoris Mater is identified as an 'Antiphon'.

The second prior text is contained in the manuscript of the Franciscan friar William Herebert's early fourteenth-century translations into English of various hymns and antiphons (Phillipps MS 8336, fo. 205^sup v^). After the English verse version of Alma Redemptoris Mater, a short note is given in Latin prose very briefly summarizing the story that Chaucer would later give to the Prioress. Half of it reads: Hic nota de filio vidue qui semper eundo ad scolas et redeundo de scolis consueuit istam antiphonam decantare . . . [`Note that the son of a widow, going to school and returning from school, used to sing this Antiphon . . .']. The boy is the `son of a widow', and the piece is again described as an 'Antiphon'.

About fifty years later (c. 1375), during Chaucer's active literary career but before he composed the Prioress's Tale, the Vernon Manuscript (Bodleian MS 3938, fo. 124) provided a rhymed English version of the whole study, including a translation of three of the Antiphon's lines into English. The relevant portion reads:

The child has both a father and a mother here, and we have no reference to his going to and from school. He is, moreover, old enough to earn his living as a kind of street musician and, again, the piece is identified as an Antiphon.

Herebert's language reveals some similarities with the anonymous thirteenth-century version (eundo . . . redeundo), suggesting perhaps reliance. But the Vernon MS's version omits this daily going to and from school, a detail Chaucer, of course, includes. The poet of the version included in the Vernon MS could have gotten the song's name from either of the two earlier versions (or, of course, elsewhere or independently) and then embellished and changed things according to his artistic inclinations. That Chaucer did not pick up on the pathos of the poor boy helping to support a father and a mother by street musicianship suggests his distance from the Vernon MS. The relevant portion of his version is as follows: the child is a 'wydwes sone' (line 502);5 he heard `Alma redemptoris' (518) in 'scole' (517) while learning an 'antiphoner' (519). He does not know what the words mean (523), `For he so yong and tendre was of age' (524). He asks an older school-fellow to expound its meaning (525-33), who gives him a limited answer, enough to convince the `little clergeon' (503) that it is in honour of the Virgin Mary (53743). He sings it `to scoleward and homward' (549), occasioning his murder.

Chaucer's language about the boy vigorously asserts his youth, his vulnerability, his inability to do things.6 He is but `seven yeer of age' (503) - the youngest of all the murdered boys among the versions of the story.' He is 'tendre' (524), 'sely' (512), 'innocent' (538, 566), 'yong' (515, 524, 644), and, above all, 'litel' (503, 509, 516, 552, 587, 596, 667, 682). He learns from a 'litel' (516) book but cannot understand Latin (523). Even his actions are cast in the language of debility rather than ability, for he `kan nate stynte of syngyng by the weye' (557, my emphasis).

This pathos is clearly something Chaucer brought to the story independently of earlier versions. In fact, Chaucer goes to some rather difficult lengths to generate this pathos, and it constitutes the major difference - alteration, perhaps of Chaucer's from all the other versions. As Carleton Brown comments:

In stating that the clergeon was seven years old, Chaucer, it would seem, deliberately altered his source in order to heighten the pathos of the story. Seven was the age at which boys ordinarily began school attendance; and, as the clergeon resolved to learn the anthem 'er Cristemasse is went,' it is clear . . . that the clergeon was still in his first term at school.8

As Brown further notes, this causes Chaucer's creation of an older school fellow, who is studying the antiphoner, not the first term's 'prymer' (517), who must explain the meaning of the piece the clergeon overhears, not reads.9

The piece that most of the murdered boys sing in Types A and B is the Tract Gaude Maria, and it is simply too difficult musically for a seven year old to sing. First, it is a Tract and not, as with the other pieces mentioned, an Antiphon. An Antiphon is a rather short piece used in the liturgy immediately before and immediately after the recitation of a psalm. The Tract's main function in the liturgy is to substitute for the Alleluia during Lent (when the utterance of 'alleluia' is liturgically inappropriate) and also at other times of a penitential nature.10 Alleluias are the most florid of all chants, difficult to learn and to sing, and the Tracts to a great measure retain the melismatic exuberance of the alleluias they replace. Gaude Maria is the Tract used during feasts of the Blessed Virgin during such times, most notably on the Feast of the Annunciation (15 March), which always falls during Lent. Tracts are mostly Psalm verses used without an Antiphon - a provocative reversal of the type of chant Chaucer's and the others among the C-group replace Gaude Maria with. Gaude Maria, though, as well as a few other tracts like Ave Maria and Tu es Petrus, are non-psalmodic tracts - that is, ones not containing phrases or sentences from the psalms. The musicologist Willi Apel describes the difficult musical style of the Tracts:

. . . [it] is considerably more ornate than that of the Introits. Every Tract includes a number of fairly extended melismas, and occasionally one comes upon melismas which are among the longest to be found in the entire repertory. . . . Between . . . extremes of syllabic recitation and long melismas the melodies move in a richly neumatic style, frequently employing groups of four, five, or six notes for one syllable.11

Those who analyse chant always divide it into various types - syllabic, neumatic, and melismatic.12 Syllabic is the style characterized by what we would now term one 'note' per syllable, while neumatic, which takes its name from the symbols known as 'neums' that represent two, three, or four notes, is characterized by the text moving from syllable to syllable by neums rather than single notes. Melismatic combines several to many neums to create richly ornate passages gracing one syllable of text. The various liturgical functions often pre-determine the style: alleluias, for instance, are always melismatic, for from at least St Augustine on, the Alleluia was mimetic of the jubilus, the pure singing of the saved in Heaven, where words dissolve into pure music.13 The singing of melismas or many notes per syllable - was so difficult that it gave rise to a separate musical genre to compensate for the difficulty in learning them - the Sequence. The Carolingian Notker Balbulus recounts how a monk visiting his monastery of St Gall from the Norman abbe of Jumieges taught him how to remember the long passages by mnemonically adding mental words to them14 - an interesting musical instance of the medieval art of memory so eloquently described by Mary Carruthers.15 Notker was so taken by the words he used for aid in memory that he began to write them down and sing them independently, thus starting the fashion for the chant known as the Sequence.

The identification of Gaude Maria with a melismatic chant thus would alone indicate that Chaucer's seven-year old student who has yet to learn Latin would not be able to master it. But the case is even more decided than that, for in many ways Gaude Maria stands out as one of the most difficult of the four thousand or so chants contained in the Liber usualis.16

First, it has an exceptionally extended range. The Tracts in general, being quite florid, often extend downwards to the G below low A - that is, to the note below the one that normally began the count in the early Middle Ages. Guido of Arezzo had to invent a new name for it, terming it 'gamma', the Greek letter for 'G' - the G, in other words, an octave below the lowest G the earlier theorists would allow. This occurs in the Tracts Emitte Spiritum" and Tu es vas18 as well as in several responsories, a couple of offertories, and the famous Alleluia Stabat mater.19 It occurs in Gaude Maria on the -e- of the first word, set by short descending and ascending melismas. Gaude Maria, in other words, simply goes too low for the immature voice of a seven year old to sing.

The low note appears in the first melisma of one of what we would now term fifteen 'notes' (the repeated 'notes' on the same line or space representing lengthening rather than separate notes). The words of the first section of the Tract are as follows:

Gaude Maria Virgo, cunctas haereses sola interemisti. Quae Gabrielis Archangeli dictis credidisti. Dum virgo Deum et hominem genuisti: et post partum, Virgo inviolata permansisti. Dei Genitrix, intercede pro nobis.20

[Rejoice, Virgin Mary: you alone have destroyed all heresies; you that believed in the words of the archangel Gabriel bore God and man while a virgin, and after the birth you remained a virgin.21

Melismas fall on the -o- of Virgo (9 notes), the -e- of Gabrielis (21 notes); the -an- of Archangeli (25 notes), the -o- of the next Virgo (29 notes), the -i- of hominem (8 notes), the -i- of genuisti (11 notes), the -sti- of that word (9 notes), the -o- of the last Virgo (7 notes), the -iof Dei (7 notes), the -o- of nobis (13 notes), and the -is- of that word (18 notes). If the adult and literate Notker had trouble with melismas, so much more would a seven year old, first-term student still working on his primer. Chaucer's insistence on a first-term boy would also preclude Gaude Maria, since it would not be sung until the following Lent.

The responsory to Gaude Maria, moreover, reveals it to be among the most difficult of all Gregorian chants to sing. Its words reveal why it was useful for the story of a boy murdered by the Jews:

Gabrielem archangelum scimus divinitus te esse affatum. Uterum tuum de Spiritu Sancto credimjus impraegnatum; erubescat Judaeus infelix, qui dicit Christum Joseph semine esse natum.

[We know the archangel Gabriel to have addressed you divinely; we believe your womb to have been made pregnant by the Holy Spirit; let the miserable Jew be ashamed who says that Christ was born of the seed of Joseph.]22

The musical setting to the words erubescat Judaeus extends to the c three notes beyond the g which is a full two octaves higher than the gamma at the chant's beginning. Gaude Maria is, in short, a virtuoso piece.

Alma Redemptoris Mater, however, is not. It exists in three versions, each well within the range of the little clergeon. Like Gaude Maria, it praises the Virgin, but its use was prescribed not for the Feast of the Annunciation in Lent but for Vespers from the Saturday before the first Sunday in Advent until the Feast of the Purification (2 February), though it could be used at other times as well.23 The version prescribed by the Liber usualis24 has a moderate range from f to g' - a modest octave and a note. It admittedly starts with a challenging melisma of thirteen notes, but afterwards, though, it avoids melisma enitrely proceeding syllabically and occasionally neumatically until the end. Seemingly hesitant about the moderate difficulty of the first melisma, the compilers of the Liber usualis provide a version they term `simple tone'.25 Its range is the same octave and a note - d to d', but it proceeds entirely syllabically, except for the neumatic setting of its first syllable. It is of course a perfect first chant for a young child to learn. Audrey Davidson, though, argues that the Sarum, not the Gregorian version was the one Chaucer had in mind.26 This is the version Beverly Boyd includes in modern notation in the Variorum edition of the Prioress's Tale.2' In difficulty this version lies somewhere between the first version in the Liber usualis and the alternate simple version there. Its range is the same - an octave and a note.It has a thirteen-note melisma on the first syllable and fluctuates between syllabic and neumatic styles until the end, never offering a second melisma.

The task is not to decide which of these three versions Chaucer had in mind and the little clergeon in his mouth, for that is impossible, though the likely one is the Sarum chant, for Sarum Use was widespread in Chaucer's England. The point rather is that each is simple in both general and specific terms - general, for the range and ornateness of each is below average for medieval chant (with the `simple tone' version very near the bottom), and specific, for in contradistinction with the virtuoso Gaude Maria, each is child's play. With the setting of a 'simple' tone version, moreover, it is clear that the Church had some concern to keep Alma Redemptoris Mater as easy as possible.

The preliminary conclusion we may make is that Chaucer, desiring pathos in his creation of a boy younger and weaker than all the others among the sources and analogues, has chosen the appropriate chant. As J. C. Wenk suggests,28 Chaucer knew Gaude Maria, knew it moreover as associated with the story of the murdered boy, but chose Alma Redemptoris Mater instead as the more 'appropriate' for his purposes. Yet he was demonstrably not the first to associate this chant with the story. Is there anything, then, that we can say about Chaucer's intentionality?

The second conclusion that follows is intended to be suggestive. What strikes me as odd is how sketchy the connection of Alma Redemptoris Mater with the story of the murdered boy is before Chaucer. There are, as we have seen, three prior surviving texts. The thirteenth-century Latin version contained in Corpus Christi College Oxford MS 32 is, of course, in a language Chaucer knew - yet one he tended to avoid when possible: he seems to have used French 'trots' while translating Boece, Melibee, and the Griselda tale. Friar William Herebert's connection of the tale with Alma Redemptoris Mater is a mere marginal note in Latin next to his English rendering of it. We could argue that the connection had been established by the time the friar penned his note, yet his language (euntes . . . redeuntes) reveals a more or less direct dependence on the earlier manuscript. And the Vernon Manuscript's version seems far removed from Chaucer's: remember the boy street musician who has both a mother and father and does not go to school.

Yet I do not wish to go so far as to suggest a Chaucer reinventing the wheel that connected Alma Redemptoris Mater and the story of the murdered boy, though to my mind the thesis has some limited cogency. What makes more sense to me is a Chaucer who knew of the association of both chants with the story, one who, knowing both because of their use in the liturgy, saw the inherent possibilities in Alma Redemptoris Mater over Gaude Maria. If, in weighing his alternatives, Chaucer noticed the great reduction in difficulty from Gaude Maria to Alma Redemptoris Mater, this might in turn have been one of several things that suggested reducing the boy's age further than what the possible sources allowed.29 In other words, Alma Redemptoris Mater might have significance as a cause rather than a result of the very littleness of Chaucer's little clergeon.