Notes on Alleluias in Modes 8 & 2

Tracks 114: Alleluias in mode 8
Tracks 1028: Alleluias in mode 2


The melodies of Introits can be found in the “Vatican Graduale,” Graduale sacrosanctae romanae ecclesiae de tempore et de sanctis SS. D. N. Pii X. pontificis maximi jussu restitutum et editum (Romae: Typis vaticanis 1908); and in the Graduale triplex, seu Graduale romanum Pauli PP.VI cura recognitum & rhythmicis signis a Solesmensibus monachis ornatum neumis Laudunensibus (cod. 239) et Sangallensibus (codicum San Gallensis 359 et Einsidelensis 121) nunc auctum (Solesmes 1979).

The “signs of nuance” can be studied in the facsimiles of the chant books “Laon 239” (“L”), published in Paléographie musicale X (1909); “St. Gall 359” (“C”) in Paléographie musicale IIe serie, II (1924); and most conveniently in the Graduale triplex, where the notation from L and C (and also “E,” Einsiedeln Cod. 121) has been entered by hand by Marie-Claire Billecocq and Rupert Fischer. Here, references are given to the Graduale triplex (“GT”).

Liturgical assignments are as in Rene Jean Hesbert, Antiphonale missarum sextuplex (Brussels 1935); reference is to the number of the Mass formulary, preceded by M. 

The very extensive post-archetypal repertory of Mass Proper Alleluias is published by Karlheinz Schlager as Alleluia-Melodien I, bis 1100, and Alleluia-Melodien II, ab 1100, Monumenta monodica medii aevi 7 (1968) and        8 (1987). 

The standard work on the Gregorian chant repertory, including relevant bibliography, is by David Hiley, Western Plainchant, A Handbook (Oxford: Clarendon Press 1993).  

The Alleluias are discussed on pp. 130–139. See also The Early Middle Ages to 1300, in The New Oxford History of Music, Vol.II (new edition, Oxford 1990), ed. R. Crocker and D. Hiley, pp. 214–222. 

The Alleluias are discussed by Willi Apel in Gregorian Chant (Indiana University Press, 1958), pp. 375–92. 

In James McKinnon, The Advent Project. The later seventh-century creation of the Roman Mass Proper (University of California Press 2000), the Alleluias are discussed on pp. 249–279.

The nuance notation is discussed in Eugene Cardine, Gregorian Semiology (Solesmes 1982), trans. by R. Fowels from “Serniologie grégorienne,” Etudes grégoriennes 11 (1970), pp. 1–158.

Modal finals and terminal descents are discussed by Jean Claire in “Les Répertoires liturgiques latins avant l'octoechos.  L'Office ferial Romano-Franc, “Etudes grégoriennes 15 (1975), pp. 5–192.

See also Richard L. Crocker, Introduction to Gregorian Chant (Yale University Press 2001).

For further resources, see the About the Archive.

Alleluia with Verses for the Mass

The acclamation “Alleluia” is sung in many different ways in a variety of liturgical contexts. As a Proper chant of the Roman Mass, it is sung with its own special melismatic extension, traditionally called “iubilus,” followed by a verse, usually in words drawn from the Psalter; the verse may include its own melisma, and may end with a repetition of the iubilus. Each individual Alleluia plus verse is assigned to a liturgical occasion.

The Sextuplex provides for an even 100 Alleluias with verses for Mass formularies. Of these, 68 can be found in the Graduale triplex; for 63, the Triplex includes notation (including nuance) from the chant books Laon MS 239 (L), St. Gall MS 359 (C), or Einsiedeln MS 121 (E). These 63 are recorded in the Archive.

 Of the 68 whose words are included in the Triplex, 4 lack the nuance notation (GT= Graduale triplex).

Cantate Domino: GT 330 (=Qui timent, GT 352)
Confitebuntur caeli: GT 478
Crastina die: GT 39 (=Benedictus es GT 375)
Regnavit Dominus: GT 242 

Of these, Cantate Domino uses the seventh-mode melody of Qui timent, included in the Archive. Confitebuntur caeli and Regnavit Dominus lack the notations from L, C, or E. Crastina die uses the eighth-mode melody of Benedictus es, for which the Triplex provides the notation from St. Gall 359. Benedictus es, however, has been omitted from the Archive, perhaps arbitrarily; I have viewed it as a Carolingian addition to the archetype, to be compared to but not included with the archetypal Alleluias. 

Dom Hesbert’s index to the Sextuplex lists with extreme care the details of how each item of chant actually appears in each source, or how its presence might possibly be implied by the configuration of the source; and this index should be consulted for precise information. The Archive, however, refers to the Sextuplex only to establish a given item in liturgical practice circa AD 850 (rather than to establish the precise role in that practice). Hence, while a reference to the Sextuplex (as an M-number) will be given for every Alleluia sung in the Archive, for an Alleluia that appears in two or three sources but only once in each and with a different assignment, that Alleluia will be given only one reference, usually to the most characteristic assignment. And for Alleluias that appear only in the three different summary lists at the ends of B, C, and K, the reference will be to M199, the number under which these lists appear in the Sextuplex.

Working from Dom Hesbert’s analytic index of the Sextuplex , which shows the distribution of each item among the six sources, I find that less than twenty-five Alleluias of the one hundred listed appear either in five of the sources, or in all six. Many of the other seventy-five Alleluias appear in only one source, for instance Rheinau (R) or Corbie (K); a number appear only in the pair of sources Senlis (S) and Compiègne (C); some appear only in the list “De circulo anni” at the end of C. The practice of singing Alleluia with verse at Mass, then, used a pool of less than twenty-five items that were widely shared, plus many more each of which was sung in only a few places, or even only one place; or were used in several places assigned in each place to a different occasion, as to individual saints’ days. In this respect the Alleluia repertory differs sharply from the other repertories of Mass Propers in the relatively low proportion of the repertory that is widely shared.

The use of model melodies, whereby a single melody is provided with several alternate sets of words for use on several different occasions, is greater in the Alleluia repertory than in the other Propers. Multiple setting of a model melody is to be distinguished from shared use of idiom (“formula”). Among the Graduals, only those in second mode show a comparable use of a model melody, while those in fifth mode, for instance, show instead a heavy use of shared idiom.

In the Alleluia repertory, there are three model melodies, one in second mode, one in fourth mode, one in eighth mode. Settings of these three model melodies account for twice as many items than idiosyncratic melodies in those modes, and for almost half of the repertory. The melodies of the word “Alleluia” (including the iubilus) have a certain similarity in the three models, especially compared to the variety among the idiosyncratic melodies. The verse melodies in the three models, however, differ greatly among themselves: “Ostende” is remarkably long and discursive; “Dies sanctificatus” is compact and repetitive; “Excita” seems closer to the norm compared to idiosyncratic verses. And only “Excita” shows the rounding of the verse with the melody of the iubilus. This rounding is the first of three important stylistic features that have been identified in the Alleluia; the second is the expansion of the Alleluia incipit and iubilus to make the verse melody; the third is the presence (or absence) in the verse of a single long melisma, not related to the Alleluia incipit or iubilus.

The use of model melodies has been taken as the basis for far-reaching ideas about the generation and development of the Gregorian repertory as well as of other, very different, repertories of music. Here I want only to notice two aspects of the way model melodies are used in the Alleluia repertory as reported in the archetype. First, setting a model melody to new words (especially when drawn from Scripture, usually from the Psalter for Alleluia verses) can certainly be said to be a function of cantorial activity, of singing. It is doubtful that it can be done well without musical ability. Nevertheless, that ability is not one of singing or composing a new and different melody. And while we need to take note, in appreciating a given melody, that it may have several alternate sets of words, appreciation is based primarily on the melody as it is sung with only one of those sets.

The second aspect concerns the presumption of historical development of the repertory; specifically, Did the repertory begin with model melodies, then proceed later to idiosyncratic ones? That presumption (which is not for discussion here) might well turn out to be true of the Alleluia repertory as found in the archetype; but its truth would have to be demonstrated by induction from traces left in the documentary record, not by deduction from a principle. Various documentary records show that the re-use of model melodies (along with the parallel phenomenon of recombination of shorter or longer idioms) can occur just as well at the end of (or in the middle of) a stylistic development. Model melodies from the archetype can be set with new words in the twelfth century, or in the twentieth—and, if done with skill, are undetectable as later artifacts.

Re-use of model melodies is not well correlated with breadth of distribution in all six sources of the Sextuplex. Of the twenty-three Alleluias with the widest distribution (in five or six sources), fourteen are idiosyncratic. It is true that out of seven Alleluias for the five feasts (Nativity, Epiphany, Easter, Ascension, Pentecost), five (or six, if we count Spiritus Domini, which is archetypal for Pentecost), use model melodies. One Alleluia, however, does not. Any attempt to draw firm conclusions by presuming relationships among the ideas of model melodies, calendric prominence or antiquity, and breadth of usage must pause in contemplating Alleluia Pascha nostrum, for here the most idiosyncratic item of the most idiosyncratic group (seventh-mode Alleluias, which use no model melody) has the broadest possible distribution (all six sources) and is assigned to Easter, the prime celebration.

Alleluias in the Eighth Mode

Of the sixty-three archetypal Alleluias provided with nuance notation in the Triplex, fourteen have melodies in the eighth mode. These fourteen are recorded and listed here on tracks 1–14.

Of these fourteen verses, ten are sung to a single model melody. In two of the remaining four, one and the same set of words is sung to two similar but not identical melodies. The other two have each their own melody. Thinking of the sets of words in the verses, then, there are thirteen items for fourteen liturgical occasions.

Thinking from a purely melodic point of view, however, there are but three distinct melodies, and two more that are closely related to each other. Of the three distinct melodies, one is the model melody used for ten verses. Hence, if the two related melodies are considered different enough to be distinct (that is, one is not a contrafact of the other), then there are five eighth-mode melodies used in the archetype.

The ten Alleluias that use the model melody are assigned to occasions from every part of the year, without showing a particular principle or program of assignment. There is one festal assignment (Ascension), and two assignments to the Nativity, Mass I and Mass II (but only Mass III counts as festal). Eighth-mode Alleluias are provided for the First Sunday of Advent, the First Sunday after Easter (that is, the Octave, and also the preceding Saturday), and the Sunday after Ascension. For saints’ days, there are two for Virgins, one for Apostles, one for Pontiffs. Some of those for the temporale are prescribed by only one or two of the six sources of the Sextuplex; and those for saints’ days are assigned singly by each individual source to various different occasions. It seems difficult to determine a principle governing the use of this model melody, or for which set of words the melody was originally used. It is commonly known as “Ostende,” but that is only because that verse, being for the First Sunday in Advent, was the first of the ten to appear in any given Graduale.

The two Alleluias with similar melodies and the same words (Confitemini Domino quoniam bonus) seem intended primarily for the Holy Saturday Vigil Mass, and used also for two or three other occasions in Paschaltide (Dom Hesbert points out that since the Sextuplex has no melodies, it is impossible to tell which melody is intended by the references in that source.) The other two melodies that do not use the model melody are for a Sunday in Paschaltide and a Sunday after Pentecost.

Only one verse out of the fourteen takes its words from a Gospel—from Luke, for a Sunday in Paschaltide.. The other thirteen verses are all from the Psalter; but these seem to have been selected with close attention to the propriety of the words to the intended occasion, with no indication of any psalmic cursus. Alleluia verses in general share with Communion antiphons a tendency to echo other Proper texts of the formulary to which they are assigned.

Throughout the Notes, I use the medieval chant scale of pitch letters:
            A B C D E F G a b c [“middle C”] d e f g
Hence, the four finals of the modes are
            D (modes 1, 2), E (3, 4), F (5, 6) G (7, 8).


The pitch set for the eighth-mode model melody (Alleluia “Ostende”) extends from F below the final G to e a major sixth above; but the e is used only twice (and not prominently) in the verse and not at all in the Alleluia, which remains in the lower part of the set with the F prominent, even functioning like a lower reference pitch. The locus F a c seems to dominate until the cadence.

While the Alleluia is relatively short, the verse is extended by the longest melismas among the three model melodies used for Alleluias (those in modes two, four, and eight). The verse shows a clear upper reference pitch c, framing with the cadential pitch G the locus G a b c, the most common one for eighth mode generally. But towards the end of its closing melisma the verse makes a prominent if brief move to F, returning via a-c; and before that there is a cadence on F approached through b-flat. And b- flat itself is prominent in the verse, being used in an approach to a cadence on a and even on G; and throughout the verse there is an unusually frequent alternation of b with b-flat in melismatic passages. This alternation has a unique effect; but its occurrence in the close of the verse is part of a cadential idiom familiar from melismatic passages in other kinds of eighth-mode chant. Nevertheless, this Alleluia and verse are distinctive, not just among the Alleluias but also the other proper chants at Mass.

The two Alleluias with verse Confitemini use the same pitch set, but show none of the distinctive features of the model melody, making use instead of idioms more common in responsorial chant, and not just of the eighth mode. But one of these Alleluia melodies (track 12), the one traditionally intoned at the Holy Saturday Vigil Mass to signal the return of the Alleluia after Passiontide, seems to be unique.

The other two Alleluias that do not use the model melody seem also unique, at least among the Alleluias of the archetype. Alleluia Deus iudex makes prominent use of the C a fifth below the final G; the whole pitch set then seems to be the octave C up to c, topped by an expressive d and e, thereby recalling the pitch set used in one or two Offertories presumed to be Gallican. Alleluia Surrexit Dominus approaches its final G, in the verse as well as the Alleluia, through b-flat supported by F, with several repetitions that emphasize the distinctive color of the cadence. These two melodies, out of the thirteen in eighth mode, are the only ones to show the rounding of the verse with the melody of the iubilus.

Alleluias in Second Mode 

Of the sixty-three archetypal Alleluias provided with nuance notation in the Triplex, fourteen have melodies in the second mode. These fourteen are recorded and listed here on tracks 15–28.

Of these fourteen items, ten use the model melody, traditionally known by the name “Dies sanctificatus” from its most prominent use at Nativity, Mass III. (For one of the ten, Redemptionem, the version notated in the Triplex (L, C) shows variants in the verse.) The remaining four have each their own individual melody.

Thinking of these fourteen items from a purely melodic point of view, there are five distinct second-mode melodies used in the archetype (one being used for ten different occasions).

Of the ten Alleluias that use the model melody, seven have liturgical assignments that show a clear plan, as here.

M11: Nativity, Mass III (25 December)            Dies sanctificatus
M12: St. Stephen (26 December)                       Video caelos
M14: St. John Evangelist (27 December)          Hic est discipulus
M16: St. Silvester (31 December)                       Inveni David
M18: Epiphany (6 January)                                Vidimus stellam
M20: St. Felix (14 January)                                  Disposui testamentum
(M22: St. Marcellus (16 January)                       (Inveni David)
M24: Sts. Fabian & Sebastian (20 January)       Sancti tui    

In this list, the second-mode model melody is used on feasts and saints’ days (those with even-numbered Mass formularies in the Sextuplex) from the Nativity through Sts. Fabian and Sebastian. The model melody is not used in the odd-numbered formularies for various obvious reasons, for instance because an Alleluia is not appropriate to the formulary for Holy Innocents (M15: 28 December). For the First Sunday after Nativity (the Sunday in the Octave, and the only Sunday before Epiphany), the Sextuplex provides Dominus regnavit, another second-mode Alleluia, one of the four that does not use the model melody and the only one of those that is used in the winter cycle.

This plan places the seven melodies in the winter cycle of feasts and saints’ days. The other two instances using the model melody are for saints’ days after Easter (Sts. Vitals et al., and St. Peter). Redemptionem, the model melody with variants in the verse, is assigned to Easter Tuesday.

The first five items listed, for occasions from Nativity through Epiphany (but excluding Inveni for                 St. Silvester, 31 December), appear across the board in the Sextuplex, represented in each of the six sources; they are basic to the archetype. The exception, Inveni, has a solid representation but not in every source; also it is used in some of the other sources of the Sextuplex for other saints’ days. The four instances using the model melody for saints’ days (two after Epiphany and two after Easter) have a much weaker representation in the Sextuplex, some of them appearing in only one of the six sources, or for a different saint in each different source.

The four items not included among the ten just listed use second-mode melodies that differ from the model melody, and indeed from each other. Like the items for saints’ days with or without the model melody, these four individual melodies show a representation either scattered among the six sources of the Sextuplex or else very little representation.

From this very disparate assignment of second-mode Alleluia melodies to liturgical occasions in the archetype it seems very difficult to conclude anything decisive or definitive about the history or development of the melodies themselves.

The second-mode Alleluia melodies, notated (in the Triplex) with one exception on a D final, use a pitch set extending from the A a fourth below the final to a a fifth above the final; but above that there is very infrequently a b-flat. In Confitemini, which is the exception notated on a final in order to provide a low        B-flat, there is a c at the top of the range. The model melody shows a gapped scale in its descent below the final: D, C, A.

The model melody moves in the same three overlapping loci as melodies in first mode (but of course without the upper register of first mode). There is a very characteristic restricted locus of D E F with F being a strong upper reference pitch. This F is also the upper reference pitch for the locus C D E F, in which the F is strongly supported by the C at the fourth below. There is also a less clear locus D E F G a, which in this melody makes use of G near the top, and C below. The model melody uses its D final and also its F as recitation-like tones.

The model melody has a short, modest melisma for the alleluia, and a different melisma, similarly short, that comes twice in the verse. This verse melisma, along with the close of the verse, uses idioms found occasionally in other kinds of second-mode melodies. Considered as a whole, however, it seems unique in the repertory of Gregorian Mass Propers; no other melody is quite like it. It is the other four melodies, those that do not use the model melody (while sharing the pitch set and some of the same features, but with idiosyncratic variation in each case), that seem more related to second-mode melodies in other kinds of chant.