The cycle Hodie nobis de virgine, consisting of eight motets for four voices, belongs to the ‘core’ group of the so-called motetti missales.^[1] The unique source that contains the cycle, Librone 1 (the first of the Gaffurius Codices), bears an attribution to ‘Loyset’ [Compère] both in the index and over the first motet. The cycle undoubtedly dates from the years of Compère’s service at the Milanese court of Galeazzo Maria Sforza (1474–77).^[2]

Franchinus Gaffurius included the cycle among the ‘motetti missales consequentes’ in the autograph index of Librone 1;^[3] its belonging to this special category is further revealed by the presence of loco rubrics and of two motets (nos. 5 and 6) forming a Sanctus–Elevation complex (see below). In the Gaffurius Codices, only two other cycles have the loco rubrics: the Galeazescha (in Librone 3), again by Compère, and Ave domine Iesu Christe (in Librone 1), whose attribution to the same composer is debated but, in my view, tenable.^[4] Outside the Libroni, the rubrics appear in the two anonymous cycles of Munich 3154,^[5] but for some motets only.^[6] In Hodie nobis de virgine the rubrics are arranged as follows:

The lack of a rubric for no. 5 is explained by its nature as a ‘hybrid Sanctus’.^[7] The first part of the motet sings the initial portion of the Sanctus from the Mass Ordinary (‘Sanctus … gloria tua’), whereas the text of the second part consists of a verse praising Christ’s incarnation (‘Verbum caro factum est et habitavit in nobis et vidimus gloriam eius’).^[8] Technically, then, item no. 5, being a Sanctus, does not require a loco rubric. Its second part is mostly set in successions of fermata chords, just as the ‘ad Elevationem’ motets in Ave domine Iesu Christe and Galeazescha: the theological parallel between Christ’s incarnation and the transubstantiation of the consecrated Host is, after all, obvious. The indication ‘Verte folium’ (‘Turn the page’) at the end of item no. 5 (bottom of f. 176r) links it to the following motet (no. 6), forming a Sanctus–Elevation complex:^[9] as Agnese Pavanello has observed, in this repertoire ‘polyphonic music was performed continuously at least from the Sanctus to the canon missae, including the consecration of the host and the following prayers’.^[10] For the rest, the loco rubrics are identical to those of Ave domine Iesu Christe and Galeazescha.^[11]

Similarly to those of Ave domine Iesu Christe and Galeazescha, the text of the present cycle is a compilation from different sources; differently from the other two cycles, however, in Hodie nobis all the texts converge on one liturgical feast, Christmas. Most textual segments seem, in fact, to derive from Advent and Christmastide offices.^[12] Sections in prose and in verse frequently alternate: the hymn stanzas are easily recognizable,^[13] whereas it is not always possible to find an exact correspondence between each prose segment and a single liturgical item, given both the patchwork-like nature of the cycle text and the inherent repetitiveness of the exuberant Christmas liturgy. Besides striving to individuate all the source texts, we may wonder about the logic and ultimate goal of the compilation. What at first sight might seem a haphazard collage of more or less familiar fragments, at closer inspection reveals, if not a clearly discernible narrative or development, at least an accumulation of insistently reiterated and interwoven motives – motives that of course illuminate the theological mystery of Christmas, as well as, appropriately for the motetti missales, its liturgical re-enactment during Mass.^[14]

Immediately noticeable is what we may call the motive of the ‘liturgical present tense’:^[15] the word hodie (today) recurs eight times throughout the cycle.^[16] Furthermore, there is an insistence on the salvation brought to ‘us the faithful’ by the Holy Child, Emanuel.^[17] Notice, on the one hand, the recurrence of nobis (again eight times) and, on the other, the multiple nomina Christi, many of which emphasize his role as savior: mundi salus, salvator saeculi (twice), salvator noster, salvator dominus, salutis auctor, redemptor omnium.^[18] Obviously central is the motive of incarnation, with forms of the verbs nascor (to be born) and gignere (to beget) occurring, with various repetitions, in virtually all motets: the only exception is the Sanctus–Elevation motet (no. 5), in which, however, the same concept is even more emphatically expressed by means of the phrase ‘Verbum caro factum est’. If the laus angelorum is another recurring motive,^[19] considerably more pervasive and theologically crucial is the Marian theme, present in most motets, prominent in motet 7, and completely predominant in motet 2. It can be said that the first section of the initial motet already presents a programmatic summary of the whole cycle text. ‘Hodie nobis de virgine Christus nasci dignatus est’: in the liturgical present, Christ, incarnated in the womb of the Virgin (and transubstantiated in gremio ecclesiae), brings salvation to men according to the divine plan of Redemption.

Although a wide-ranging survey of compiled texts for Christmas worship and devotion, cutting across various musical and literary genres (and cultural layers), is still wanting, recent research has spotlighted a tradition of Franco-Flemish Christmas motets with hyper-composite texts. The best known are three five-voice motets with cantus firmus in the tenor: O admirabile commercium by Johannes Regis (ca. 1425–ca. 1496), Factor orbis by Jacob Obrecht (1457/8–1505), and Nato canunt omnia by Antoine Brumel (ca. 1460–?1512/13). Although based on a different compositional concept than the cantus-firmus-free four-voice Hodie nobis, they share a few traits whose contemplation can help elucidate Compère’s cycle.^[20]

To begin with, all three motets have long texts that draw, as said, from disparate liturgical, and sometimes para- or extraliturgical, sources. Obrecht’s Factor orbis, for instance, has been defined as a ‘grand celebratory Christmas concoction of twenty discrete texts of which six retain their original melodies’.^[21] Even though its text, unlike that of Hodie nobis, does not contain sections in verse, the compilation, combining excerpts from Advent and Christmas liturgical items (interspersed with ‘noe noe’ passages), is in principle similar.^[22] The textual sources for Regis’s O admirabile commercium include not only liturgical items, but also metrical cantiones or carols.^[23] Brumel’s text adds a sequence to the mix, and is at times modelled on Regis’s one.^[24]

Turning to the music, previous studies have demonstrated how in these motets a composite musical structure responds to the composite structure of the text,^[25] if not always with the transparent sectionalization of Compère’s cycle, surely with some remarkable changes of texture and of the multiple parameters that in a different context I have gathered under the label of ‘sonic style’.^[26]

The final relevant point is that in setting these composite texts, musicians not only incorporated cantus prius facti (whether or not as cantus firmi properly speaking), but in certain cases reacted to the styles traditionally associated with those texts. This is especially noteworthy in the case of songs belonging to traditions other than highbrow polyphonic culture. In the case of Regis, for instance, the quotes from the cantio Magnum nomen domini (mm. 12–16 and 32–41)^[27] prompt the adoption not only of the associated melody, but also of a ternary metrical matrix that we find in such settings of the cantio as that of the songbook Berlin 190,^[28] ff. 31r–32r.^[29] The same applies to the other cantio quoted in the motet, Universalis ecclesia: the text quotes are accompanied (mm. 67–91) by the borrowing of the associated melody and of its syllabic character and metrical matrix,^[30] as again we can notice by comparing the version in Berlin 190, ff. 27v–28v.

Considered in this light, certain aspects of Hodie nobis and of other Milanese motetti missales cycles seem less idiosyncratic than they are normally judged: at least in certain situations, textual collages seem naturally conducive to a sectionalized musical treatment, and the incorporation of pre-existing elements and textures does not necessarily and exclusively relate to ‘Italian’ or lauda-related materials.^[31] For Hodie nobis it is not yet possible, pending further research, to point to precise references or borrowings, but certain passages do raise suspicions. The most prominent example regards motet 3, whose tenor part may go unnoticed within the visual context of the modern score, but immediately manifests its distinctive nature if contemplated in the notation of Librone 1 (f. 173v). On the one hand, the fact that the Tenor is the only voice to sing the entire text of the motet certifies that it constitutes the primary layer of the compositional project; on the other, its formulaic, patterned shape is as far as it can be from that of a freely-invented tenor part. Ludwig Finscher has already proposed to look for a pre-existing model in the cantiones repertory,^[32] but no likely candidate has emerged so far. In other motets of the cycle there are sharply-profiled melodic segments that stand out remarkably from the surrounding ones, suggesting a derivation from pre-existing tunes (e.g. the phrase ‘In choro angelorum’ in motet 7, mm. 18–21).

As mentioned above, prose and verse alternate in the cycle. Even though there is no rigid correspondence between textual and musical metre, the preference for setting hymn stanzas syllabically in a quick ternary metre seems evident. In motet 4 the passage from q to q3 (m. 20) coincides with the transition from a prose verse to a pair of hymn stanzas. Motet 6, the only one consisting entirely of two hymn stanzas, is also unique in adopting the u3 measure, without the midway mensural change that characterizes motets 1–4 and 7–8 (by contrast, motet 5 is entirely set in u2).^[33] In motet 8, the metre q3 and the syllabic declamation are already established when the hymn text starts, but the delivery of the hymn stanza – at the conclusion both of the motet and of the cycle – obtains further emphasis by means of prevailingly homorhythmic textures. For the hymn stanza in motet 1 (mm. 10–18) the quick ternary meter is at first implicit, and then achieved through coloration.

Even considering the marked motivic unity and formulaic tendency of the entire cycle, it might not be mere coincidence that the same melodic phrase in the Tenor closes all the three stanzas derived from the hymn Christe redemptor omnium (motet 1, mm. 16–18; motet 4, mm. 27–29 and 43–45, 46–49). In the finale of motet 4, furthermore, the phrase features prominently within a special ‘sound object’ (mm. 40–49) in which Cantus and Tenor sing in a quasi-canon at the octave, the Altus intones a recitation-like series of semibreves on d', and the Bassus bounces between g and d – possibly a compositional allusion to improvisational practices. The same melodic phrase occurs at the very end of the cycle (motet 8, mm. 34–36), as if appended to the last hymn stanza.^[34]

All these clues suggest that Compère might have used pre-existing hymn melodies or formulas, even though we are no longer (or not yet) able to recognize them. Once again, the ‘coincidence’ with what happens in another polyphonic setting from the same environment reinforces the suspicion: the hymn-line formula in the Tenor of motet 1 (mm. 11–14) is unmistakably similar to the one used by Gaspar van Weerbeke for setting hymn lines in the sixth motet of his motetti missales cycle Ave mundi domina (Quem terra pontus aethera, T, mm. 15–19 and 37–41; see Fig. 1).^[35]

Considering the different details and the respective compositional contexts, I am not suggesting a direct imitatio/aemulatio relationship between the two passages: rather, it may be a common root in some for us still unknown (and possibly northern) model of hymn singing.^[36]

If the chase for the pre-existing materials and ideas that may have contributed to inspire certain features of Hodie nobis and of the other motetti missales cycles remains open,^[37] there is virtually no need to underscore the degree of unification and integration accomplished by Compère, already brilliantly elucidated by Finscher.^[38] To the unification of the cycle – achieved inter alia by means of a shared tonal type, recurring mensuration patterns, repetitive melodic and contrapuntal building blocks (sometimes in exposed positions, notably at the incipits), and intricate motivicity – corresponds an ingenious characterization of the individual motets, sometimes more subtle, sometimes (especially for the Sanctus–Elevation complex) more conspicuous. 

* * *

The cycle and all its component motets are unica in Librone 1. The version copied by a single hand (Scribe A)^[39] appears to be correct and reliable, also thanks to a few interventions and corrections by a different, unidentified hand (see the Critical apparatus).

The most apparent differences with the previous editions by Finscher and by Dino Faggion^[40] are that the present one provides a critical apparatus and a separate annotated edition of the texts, and that it retains the original mensuration signs and rhythmic values.

A few notes about text underlay, one of the notoriously problematic aspects of editing late fifteenth-century music, are in order. As stated in the MCE Editorial methods, the text underlay adopted in the score ‘should be regarded more as a proposal than as a prescriptive solution’. Whether due to the scribe or to the composer, the text underlay of Hodie nobis in Librone 1 is, for some motets, comparatively complete and accurate, to the point of looking, at times, genuinely ‘prescriptive’ (see e.g. motet 1). Unsurprisingly, however, it is impossible to slavishly follow its distribution for long without incurring awkward situations, for instance for such melismatic passages as those in motet 5, or in sections in which the syllables seem fairly recalcitrant in corresponding to the available notes, or vice versa (e.g. in motet 6).

Moreover, the text is rarely, if ever, complete in all voices. This is normally because of the dialogic character of Compère’s vocal orchestration, with frequent exchanges between voice pairs. In the special case of motet 3, however, the reason is that only the Tenor (syllabically) sings the entire text, whereas the other more melismatic voices sing a selection (the Cantus does have an indicative text underlay in the manuscript, but Altus and Bassus are provided with nothing more than an incipit and a midway cue). Given the frequency of the phenomenon, the lack of a textual portion in one or more voices is not indicated in the Critical apparatus.

I have tried to observe the ligatures, assigning them a single syllable whenever possible (for an exception, see motet 3, B, 33–34), and, conversely, to underlay different syllables to consecutive notes on the same pitch (for an exception, see motet 5, A, 71–72, where I have inserted a dashed tie between the two d's, instead of postulating a repetition that would have affected all voices). As to repetitions, they are sometimes explicitly indicated in the manuscript (e.g. motet 1, T, 19–22; motet 8, T, 30–34); I have inserted them, though sparingly, to fill the gaps between textual segments that are ‘spaced’ in the manuscript, and especially where clearly profiled melodic reiterations seem to require them. Added text is marked in italics in the score; only major interventions in the matter of shifting words and syllables are recorded in the Critical apparatus.

Milan, May 2020