4 The thirteenth century
Scope and context
Garry Trudeau begins my favourite Doonesbury cartoon with three frames each bearing a date: (1) a hooded figure tries to escape a stoning: ‘1205 – A plague-stricken calligrapher is driven from Gunbad-I-Qabus’; (2) a mendicant fanatic prepares a burning at the stake: ‘1233 – A heretic perishes in Castile’; (3) a severed head hangs by its hair from the branch of a tree: ‘1276 – On the Malaysian archipelago, a gruesome trophy swings in a lagoon.’ The fourth frame cuts to Trudeau's ex-Hollywood starlet, Barbara Ann Boopstein, on a beach with a soda in her hand as she muses ‘Sometimes it seemed the 13th century would never end!’ 1 To turn abruptly to the music of the thirteenth century – although calligraphers and mendicant fanatics have their place there – is to feel some sympathy with Boopsie's ruminations. Not only did the ‘long’ thirteenth century never quite seem to end, but its beginnings were not quite obvious either. And although no one ever seems to argue for a ‘short’ century, the claims for thinking about the music of the thirteenth century from around 1170 up to ca1320 are strong. The blurred distinction between versus and conductus is as much geographical as chronological (between north and south as much as early and late twelfth century), and the emergence of a consistent style of Parisian organa is usually assumed to fall somewhere in the last quarter of the twelfth century. Yet these are the staple genres of the thirteenth century, with works copied and recopied at least as late as 1300, and the motet – although probably not surfacing before 1200 – retains its links with the thirteenth century well into the fourteenth alongside works of what might be thought to represent the avant-garde.
The scope of this chapter is largely limited to the genres of conductus, organum and motet. The chapter touches on vernacular monophony (treated in Chapter 12) only tangentially, and it centres on the northern French experience (those of other centres are discussed in chapters 6–10). Although much of the context for organum, and a significant part of the background of the motet, is liturgical, and much important liturgical development took place in the thirteenth century, plainsong and liturgy are explained elsewhere in this volume (chapters 1 and 11). And in its focus on conductus, organum and motet, this chapter challenges one of the fundamental questions of medieval music: the problem of its context and function. Up to around 1300, we have a fairly clear idea of the contexts for the composition, cultivation and consumption of music that was written down: plainsong and its offshoots can be locked into a broadly liturgical environment, and vernacular monophony into those of the aristocratic courts. Even Aquitanian versus, although occasionally suggesting less formal contexts, had a significant liturgical dimension. The massive growth of the northern counterpart of versus – the conductus – exacerbated the range of possible contexts for the genre. And although organum, as a result of its origins and compositional dynamic, could never be shaken from its liturgical context, the emergence of the motet poses problems of function – who composed? who listened? and who appreciated? – at almost every stage of the century.
Conductus
The most striking continuity between the Aquitanian and related repertories and the Parisian music of the later twelfth and thirteenth centuries lies in the versus and conductus. 2 At some points, the two terms almost coexist, and if they are thought to refer to monophonic or polyphonic settings of rhythmic texts in musical styles that do not depend on the use of borrowed material, this brings both categories into a very close alignment. Indeed, monophonic conductus copied well into the thirteenth century could resemble versus of a hundred years earlier.
The conductus is the only genre found consistently in all the major manuscript sources of the thirteenth century, and it exists in great numbers. 3 The largest collections are found in I-Fl Plut. 29.1 where 83 monophonic conductus, 130 two-part, 59 three-part and 3 four-part conductus are preserved. 4 They constitute key parts of the organization of the volume, with separate fascicles dedicated to each type and a separate fascicle given over to 60 monophonic conductus in rondellus form. 5 Parisian conductus appear to have been composed as early as 1160 and continued to be copied throughout the thirteenth century. The latest datable event in the text of a conductus is from the 1230s which might suggest a tailing-off in composition towards the mid century; 6 although conductus continued to be copied in the second half of the century, however, sources from around 1300 subject the music to significant editorial change.
Conductus, like versus, are settings of Latin texts with an immense range of subject matter: poems alluding to political events rub shoulders with homiletic verse, and references to classical antiquity sit alongside settings of liturgical texts. 7 This wide range, while testifying to the far-reaching intellectual ambitions of their authors, creates significant difficulties in assigning any function to the genre of the conductus. It seems reasonable to assume that the range of textual reference implies a similar range of function, from a substitute for the Benedicamus Domino or Sursum corda to largely recreational contexts that would admit the satirical and the hortatory. There is, furthermore, no apparent mapping of number of voices onto particular types of text. 8
Polyphonic conductus exist in one of two forms described by contemporary theory: cum and sine caudis . 9 The presence or absence of such ‘tails’ (caudae) provides one of the clearest generic markers in the repertory. The conductus sine caudis is the simpler of the two types, and consists of a neumatic (note against note or neume against neume) setting of the poetic text. The notation of this music cum littera (texted) is invariably unmeasured in the earliest sources, and its rhythm is the subject of debate, with some advocating a metrical approach based on the framework of the rhythmic modes, and others proffering non-metrical rhythms, as here. Virtus moritur is a good example of a two-part conductus sine caudis ( Example 4.1 ). 10 Much of the piece is purely syllabic, with occasional moments where a ligature of two notes sets the syllable ( mo ritur and vi tium ), and there are occasional instances where three notes (but always in ligature) set a single syllable. 11
Example 4.1 Two-part conductus sine caudis, Virtus moritur , 1–20. (I-Fl Plut. 29.1, fols. 322r–322v)
Virtus moritur is a simple work, and rather rare in I-Fl Plut. 29.1. The conductus cum caudis is a more common but more complex proposition. Much of the setting of the poem is in the same cum littera style as the conductus sine caudis just described, but the composition is embellished by caudae; these are untexted (sine littera ) melismas that take the note-against-note melismas of Aquitanian versus and subject them to the discipline of the rhythmic modes. Conductus cum caudis therefore play musica cum littera and musica sine littera off against each other in the same composition. The beginning of Luget Rachel iterum , in two parts (duplum and tenor), demonstrates how the two discursive modes (musica cum and sine littera ) interact ( Example 4.2 ). 12
Example 4.2 Two-part conductus cum caudis, Luget Rachel iterum , 1–68 (I-Fl Plut. 29.1, fols. 359v–360)
The work opens with a cauda in the first rhythmic mode that uses both fractio (duplum 3 and 7; tenor 3) and reductio modi (tenor 1, 2, and 5–8). The first three lines of the poem are then declaimed in a simple neumatic style (although note the rhythmic complications in 28–31 that betray the genre's origins in Aquitanian versus), and are followed by another cauda (32–9) that introduces the cum littera setting of lines 4–6 of the poem; this closes with a third cauda (61–8). Example 4.2 gives slightly less than half the work, but the structure of the entire composition can be seen in an annotated edition and translation of the poetry ( Example 4.3 ). The three caudae begin the first and fourth lines with similar morphemes, lu get and lap so, and close the sixth (civitas ) (the caudae are shown in italics in the example); the example also shows how the seventh line begins with a cauda and how the entire composition ends similarly.
Example 4.3 Luget Rachel iterum . Text, translation and analysis
Whatever disagreement there might be concerning the rhythm of the cum littera sections of the conductus, there is no doubt that the caudae (sine littera ) are notated using the ligature patterns of the rhythmic modes and are therefore measurable. With this in mind, it is striking that the first three caudae of Luget Rachel iterum are all of exactly the same length (eight perfections), and this is the case no matter how the notation is transcribed, and no matter where one sets the boundaries of music cum and sine littera . These three caudae of identical length stand in contrast to the two others, one of six perfections (on Lan guent) and the other of no less than 26 perfections (on vic timam). At one level, the emphasis given these particular words – ‘weep’, ‘fall’, ‘city’, ‘mourn’ and ‘victim’ – picks out the key elements in a text that exploits a wide range of biblical and patristic imagery and that may relate to the fall of Jerusalem in 1187 (although the text is unlikely to predate that event, both the poetry and the music could have been composed later). But at another, more purely musical, level, the particular compositional organization of the caudae generates a set of symmetries in the first half of the piece that are deliberately denied in the second.
Organum
Continuities in liturgical music with the previous generations are less easy to identify. Parisian organum embodies the types of embellishment of plainchant that are strikingly absent from the Aquitanian repertory. However, to glance at the Compostelan chant settings is to witness some strong continuities with the Parisian repertory in terms of liturgical type selected for polyphonic treatment. Both concentrate on responsorial chants: responsories, graduals, alleluias and Benedicamus Domino. The only type found in Compostela and not in the Parisian repertory is the troped Kyrie. This preference explodes into a two-part repertory (as preserved in its most extensive source, I-Fl Plut. 29.1) of 34 items for the office (almost all responsories), 10 Benedicamus Domino settings, and 59 mass items (18 graduals; 41 alleluias). 13 The repertory stands in a tradition of composition for the liturgical year that encompasses the Choralis Constantinus and Bach's Church Cantatas. In addition to this two-part music is a smaller repertory of three- and four-part compositions (organa tripla and organa quadrupla ); of the latter, the graduals for Christmas and St Stephen, Viderunt omnes and Sederunt principes , have achieved celebrity through modern recording. 14
The chronology of Parisian organum is problematic. If we can be reasonably certain that the Compostelan repertory was complete by 1173, the beginnings of the Parisian repertory are shrouded in mystery. The theorist known as Anonymous IV, writing around a century after the event, identified the four-part Viderunt omnes and Sederunt principes as works by Perotinus Magnus, and pointed to a predecessor of Perotinus, Leoninus, whom Anonymous IV credited with the creation of the ‘magnu[s] lib[er] organi de gradali et antifonario pro servitio divino multiplicando’ (‘the great book of organum from the gradual and antiphonary to elaborate the divine service’). Perotinus was also credited with having ‘abbreviavit ’ the Magnus liber organi (‘abbreviavit’ could have meant anything from ‘shortened’, to ‘edited’, or even just ‘written down’). 15 Perotinus's two four-part organa were cited in Parisian episcopal edicts in 1198 and 1199 in a context that meant that the works were known in the city at those dates. While the documents provide a fixed chronological point, a blizzard of scholarly debate has interpreted and reinterpreted their implications: either the two organa quadrupla represented the final point of ‘development’ of the work of both Leoninus and Perotinus (in which case the entire project of Parisian organum was complete before the end of the twelfth century) or there are identifiable aspects of the ‘Perotinian’ repertory that could not have preceded the composition of the two datable works. 16 Arguments on both sides are fraught with teleological difficulties, and no clear answer can be given for the opening and closing dates for the composition of Parisian organum. The earliest manuscript of this repertory is D-W 628 which dates from no earlier than the 1230s, with other sources spanning the next half-century. 17 Attempts to delineate continuity and change in musical practice in this repertory, and to identify how long a period of time such a delineation covers, with the present state of knowledge, seem problematic. It almost goes without saying that the music preserved in the surviving sources, its notational style and what may be inferred from it may bear little relation to what had been composed perhaps sixty years previously.
In contrast to its chronology, the function of organa is clear: as partial settings of plainsongs, Parisian organa simply replace those sections of plainsong that they set. Settings survive for most of the major feasts in the liturgical year. Annuale feasts were provided with organa for the gradual and alleluia for the day itself and for the days of the octave (Christmas, Easter, Pentecost and the Assumption of the Blessed Virgin Mary); lesser feasts, of duplex rank, were provided with organa for the gradual and alleluia of the feast day itself only (for example, St Stephen, St John the Baptist and the Ascension); and semiduplex feasts were provided with an alleluia only (Finding of the Holy Cross and St Michael, for example). In addition to music directed towards one specific occasion, and partly to provide compositions for some feasts of duplex and semiduplex rank, the Magnus liber organi contains a substantial number of compositions for the Common. 18
Parisian organum is consistent in its selection of the solo sections of plainsong for treatment in polyphony. In terms of a work, then, a Parisian organum consists of sections newly composed by Leoninus, Perotinus or their contemporaries, and of sections in plainsong filtered through twelfth-century Parisian chant traditions. Example 4.4 gives the text of the entire gradual for the mass at the feast of Sts Peter and Paul (29 June), ‘Constitues eos V. Pro patribus’. 19 This shows clearly how the division between the sections of the chant for soloists and the rest of the choir (schola ) results in different structures for the respond (where only the first two words of the plainsong are set) and the verse (where all but the last two words are set); plainsong is in roman type, polyphony in italics ( Example 4.4 ).
Example 4.4 Text of gradual Constitues eos. V. Pro patribus
Different balances between solo and choral portions of the chant result in different structures in alleluias, responsories and Benedicamus Domino settings.
Within the polyphonic sections of Parisian organum are two principal stylistic divisions with a third subsidiary one. Terminological issues are problematic here, but a distinction may be drawn between organum per se and discantus. Both are visible in the verse of Constitues eos ( Example 4.5 ). Organum per se takes the notes of the chant and disposes them in long values in the tenor, above which is composed a voice part that creates largely perfect and imperfect consonances. The notation of both parts is unmeasured, and – as in the case of the cum littera sections of conductus – there is significant debate about the extent to which metre governs the rhythm of this music. 20 Organum per se occupies 1–12, 19–29 and 62–5 of Example 4.5. These sections, where musical imagination is given a rhythmically free rein, are contrasted with sections in discantus known as clausulae. Here (30–61), the notes of the tenor are organized rhythmically according to one of the rhythmic modes (here mode 1), and may be repeated (as they are here; 45–61). This gives a series of repeating patterns – ordines – for one or more cursus. 21 The final note of the last ordo of the first cursus (45) is the first note of the second cursus, so even though the ordines of the second cursus are the same as the first, the musical results are different, with pitch and rhythm being displaced so that what fell on a longa in the first cursus falls on a brevis in the second, and vice versa. Above this tenor is composed a duplum that in this case follows the same rhythmic mode. Phrases may overlap or may cadence simultaneously, as is the case here. As in the case of organum per se , cadences are planned around perfect and imperfect consonances. The verse of Constitues eos also includes an instance of copula. Alongside organum per se and discantus, copula is a third discursive mode in organum; it seeks to impart the metre of discantus to the duplum only, and it may be identified by the fact that the highly varied ligature patterns that characterize organum per se in general give way to the patterns (here 3+2+2; 3+2+2) of modal rhythm. Copulae will also exhibit periodic phraseology, usually antecedent–consequent patterns, and also melodic sequence, as the copula in Example 4.5 suggests (13–17). 22
Example 4.5 Two-part organum, Constitues eos. V. Pro patribus , V. 1–65 (I-Fl Plut. 29.1, fols. 121v–122r)
With an understanding of the various compositional discourses in play, organa emerge as complex and unique structures that play off plainsong and polyphony in ways that are entirely dictated by the liturgical structure of the original plainsong, and that creatively juxtapose organum per se , discantus and copula within the polyphonic sections. Constitues eos again shows how this works. Example 4.4 shows how the polyphonic verse includes no less than six clausulae, alternating with seven passages in organum per se . As in the case of the musica sine littera in the conductus cum caudis , there is clearly some sense of balance in the composition and distribution of the clausulae: the second and fourth are of the same length, while the first is twice the length of the second and fourth; by contrast, the third, fifth and sixth are of eleven, ten and twelve perfections respectively.
A key feature of the repertory of organum duplum is the different selection of polyphonic material in different sources. In the case of Constitues eos , the polyphonic material for the respond is largely the same in the two sources (I-Fl Plut. 29.1 [also known as F ], which forms the basis of examples 4.4 and 4.5 , and D-W 1099 [also known as W2 ]) that preserve the piece. But in the case of the verse, there are three sections that are entirely different, and all three affect the clausulae on ‘ pat ribus ’, ‘ nati ’ and ‘ fili i ’. In the case of ‘nati’, the clausula is replaced in D-W 1099 by organum per se . Although the first clausula (32 perfections in I-Fl Plut. 29.1) is replaced with one of 28 perfections in D-W 1099, the clausula on ‘filii’ in both manuscripts is of identical length, thus preserving the symmetry with the clausula on ‘tuis’. 23
Three- and four-part organa adhere to most of the principles that obtain for organa dupla with the key exception that the upper voices in the sustained-tone sections are subject to modal rhythm and are notated in modal notation. The aural difference is striking, especially in the four-part works: in the sustained-tone style, three voices are in modal rhythm and one not, but in discantus all four voices are in modal rhythm; the shift in sonority is nothing like as great as from organum per se to discantus in two-part works, and the change from one musical discourse to another is not as sharply etched. Furthermore, the repertory of organa tripla and organa quadrupla is much smaller, and the opportunities for swapping clausulae around are that much more reduced.
Motet
The interchangeability of sections of organum, especially in the two-part repertory, led to an autonomous existence for the clausula, and collections of short fragments of discantus were features of some of the manuscripts preserving the repertory as a whole. The clausula began to take on a life of its own, a life that was responsible for the creation of the motet. 24
The emergence of the motet, which has been dated anywhere between 1200 and 1220, problematizes the relationship between words, notes and notation. 25 Put very simply, the motet was created by adding words to a free-standing clausula as examples 4.6a and 4.6b show.
Example 4.6a Two-part clausula [Immo]latus est , 1–39 (I-Fl Plut. 29.1, fol. 158r)
Example 4.6b Two-part motet Immolata paschali victima / Latus , 1–39 (I-Fl Plut. 29.1, fols. 411r–411v)
Both the motet Immolata paschali victima / Latus and its source clausula share a tenor that is identical in almost all respects. 26 In addition, it is easy to see how each of the notes of the ligatures in the clausula now carries a single syllable of the motet's text; this, however, does some violence to the melodic integrity of the clausula's duplum. Although the first two perfections text the clausula exactly, the third and fourth not only carry two syllables but they also modify the clausula's original rhythm. Similar modifications may be found at perfections 31 and 35.
To a degree, the structure of the motet poem is determined by the phrase lengths of the clausula, but there is significant licence: over the course of two perfections, either two, three or four syllables may be deployed, and several lines of poetry are set to a single musical phrase ( Example 4.7 ).
Example 4.7 Immolata paschali victima / Latus . Analysis of motetus text
The text exhibits a wide range of line lengths, ranging from three to eight syllables, and there is no correlation between phrase ends and paroxytones or proparoxytones; nor does there appear to be any logical relationship between musical phrase length and poetic line length. In short, there seems to be a lack of interest in what are conventionally considered text–music relations. What is, however, largely consistent is rhyme. A single rhyme dominates most of the motet, and changes only for the last 4 lines of the 23-line lyric. This is a critical point in the poem and in the motet, since the word that triggers the change of rhyme, ‘latus’ (the side [of Christ]), assonates with the word of the plainsong from which the tenor is taken: ‘immolatus’.
Such correspondences suggest a relationship between the text of the poem and its host plainsong. The plainsong from which the clausula is ultimately taken is ‘Alleluia. V. Pascha nostrum immolatus est Christus’, and this furnishes much of the text of the first few lines of the poem. Going one step further to the rest of the Easter Day liturgy, from which Alleluia. V. Pascha nostrum is taken, shows that ‘victima’ is the first word of the sequence for the mass, Victimae paschali laudes , and that the references to ‘azima’ derive from the communion chant for the mass on Easter Day, Pascha nostrum immolatus , which includes the line ‘itaque epulemur in azimis’. The rest of the poem depends on Old Testament analogies – Jonah and the whale, Joseph and the Fiery Furnace – that point to other incarcerations followed by resurrections, which are then made to return to the side (‘latus’) of Christ. 27
The close relationships between the text of the motet, its host plainsong and the background liturgy or the mass of Easter Day strongly suggest some sort of liturgical or paraliturgical context for the motet. One possibility is that the clausula should be texted when it forms part of a performance of the entire organum in which it is found; in other words, motets found in manuscript collections should be reinstated in their parent organa. 28 While such a suggestion is entirely plausible, it does not account for motets whose texts are polemical, hortatory or otherwise unrelated to the liturgy in the way that Immolata paschali victima is. The problems of function are here analogous to those of much of the conductus repertory, and may be just as varied. 29
The relationship between clausula and the early motet brought into play two related phenomena: (1) the idea that musicians might compose multiple works (clausulae or motets) for the same context and expect them to be swapped around over time by those who sang them; and (2) the idea that texts could be added to melismatic music. It was a small step to the further practice that characterizes the motet in the thirteenth century: the addition of new voice-parts – a triplum to a two-part work for example – and the replacement of one text with another. Coupled to the possibility of using vernacular texts, these ideas created a potential kaleidoscope of musical and literary practices that were exploited with enthusiasm throughout the thirteenth century and into the fourteenth. With the simple two-part motet (motetus and tenor) being possibly the oldest product of these practices, almost every combination of text and music was exploited during the course of the thirteenth century: voice-parts could be piled up so that not only did a triplum appear but so too could a quadruplum be added to a three-part work (sometimes confusingly called a ‘double motet’) to create a motet in four parts. French and Latin texts could coexist in the same work (known as a ‘bilingual motet’). 30
Simply describing the complexities of the relationships between surviving thirteenth-century motets is a serious barrier to understanding; trying to explain how musicians acted as they reworked musical and poetic material is even more challenging. 31 A sense of how the repertory worked in the second half of the thirteenth century may be gained from looking at a single group of compositions that shares the same musical material:
There is no clausula on which this collection of works is based, and it might be reasonably argued that the two-part Latin motet – with a motetus text that cites the text of the plainsong and is therefore closely allied to a liturgical context in the same way as Immolata paschali victima – is the first link in this complex chain. But deciphering compositional priority beyond that point becomes very difficult. From the sources in which they are found, it could be argued that the two-part French motet is the second stage in the process – a simple contrafactum of the Latin motet – and that the three-part ‘French double motet’ arises out of adding a newly composed triplum and poem. But the dates of the manuscripts for both compositions are not sufficiently differentiated to be certain, and it is entirely possible that a musician could have taken Hec dies leticie / Hec dies – added a triplum directly and two new French texts, and that the two-part French motet is simply a reduction created by dropping the triplum and its text. What does seem more likely is that the three-part Latin motet Salve, virgo Katherina / Sicut solis radium / Hec dies is a further contrafactum of the French double motet. 36
One way of making sense of the wide range of motet types found in the thirteenth century is to look at the way in which thirteenth-century musicians tried to organize their understanding: to look at the way in which manuscripts were organized. The best-known source for the thirteenth-century motet is the so-called Montpellier Codex, F-MOf H 196. In its earliest form (it was added to at least twice later) from around 1270, it divided its contents into four-part motets, three-part bilingual motets (texts in Latin and French), three-part Latin motets, three-part French motets, and two-part French motets. The critical principles of organization and identification for whoever planned the structure of this book were the number of voice parts and the language of the poetry.
For the repertory of French motets, in two and three parts, the position is even more complex as they create intertextual links between the repertory of secular monophonic song, medieval romance and other literature via the sharing of refrains . 37 A refrain is a short phrase of poetry that reappears in more than one literary or musical context; the intertextuality may relate to the poetry or to the poetry and music together. A simple example is the motet Amis, vostre demoree / Pro patribus , which is found in the sixth fascicle of F-MOf H 196. 38 The end of the motet shares its text and music with a secular monophonic song and a treatise on love by Gerard of Liège, the Quinque incitamenta ad deum amandum ardenter ( examples 4.8a and 4.8b ). 39
Example 4.8a Two-part motet Amis vostre demoree / Pro patribus , 26–46 (F-MOf H. 196, fol. 249r)
Example 4.8b Monophonic song: Moniot d'Arras, Amours me fait renvoisier et chanter, refrain (F-Pn fonds fr. 844, fol. 118v)
The song by Moniot d'Arras (fl. 1213–39) is a chanson à refrain , where the refrain appears at the end of each stanza; apart from a few ornamental melodic variants and some slight lexical changes the refrain is identical to the end of the motetus of Amis, vostre demoree . 40 The notation in the two sources for the song, F-Pn Fr. 844 and 12615, is unmeasured, whereas – as is always the case in F-MOf H 196 – the notation of the motet is measured, and this difference is retained in the example. 41 Such simple cases are outnumbered by far more complex intertextualities where more than one refrain is in play in a single motet, where the refrain is broken up, where the music of the refrain is retexted within the motet within which it is found, etc. 42
Throughout the creation of all these complex musical and literary networks, some things remained constant: plainsongs were still the source for tenors, and – with the new notation that differentiated graphically between longae and breves – modal rhythms could be expressed, and the system largely continued to dictate the rhythmic structure of the music and the declamation of the poetry. Towards the end of the thirteenth century, two of these constants were subject to change. The first innovation was the inclusion of vernacular songs as sources for tenors, and the second was a key change to the rhythmic profile of the music. While French tenors form a small part of the overall repertory of motets, in the manuscripts in which they are found they have a proportionally higher profile (seventh and eighth fascicles of F-MOf H 196 and I-Tr vari 42). 43 Secular songs have a much higher incidence of repetition than plainsong, and in some cases composers experimented with matching the repetitions in the upper voices with results that – musically at least – came to look like the polyphonic songs that would become so important in the fourteenth century. 44
The second major change to the profile of the motet at the end of the thirteenth century was the end of the relationship between rhythmic structure and declamation. Examples 4.7 and 4.8 show compositions where the declamation of the text largely follows the rhythmic mode of the piece. Towards 1300, not only did the brevis – which up till now had never been divided into more than three semibreves – become divided into much larger numbers, but syllables of the poetry were declaimed at a similar rate ( Example 4.9 ).
Example 4.9 Three-part motet Aucun ont trouvé chant par usage / Lonctans me sui tenu / Annun[tiantes] , 1–16 (F-MOf H. 196, fols. 273r–274r)
The two lower parts of Aucun ont trouvé chant par usage / Lonctans me sui tenu / Annun[tiantes] behave very much like any mid-thirteenth-century motet, but the triplum exhibits divisions of the brevis into three semibreves (which, although found in the motetus, never carry more than one syllable), five, six and seven breves in the space of a dozen perfections. 45 More than anything, this particular change in the motet repertory marked an audible change to the texture of the work, and a shift in aesthetic focus. 46
If the function of non-liturgical Latin motets is unclear, the environment in which motets with French texts were cultivated is even more opaque. Two pieces of evidence may be brought to bear on the question. The theorist Johannes de Grocheo wrote enticingly about how the motet should ‘not…be propagated among the vulgar, since they do not understand its subtlety nor do they delight in hearing it, but it should be performed for the learned and those who seek after the subtleties of the arts. And it is normally performed in their feasts for their beautification…’ 47 Johannes's comments are in the context of what he calls musica canonica , in other words the music of clerics, and raise as many questions as they provide answers. They do however clearly locate the motet within an educated domain in which the motet's complexity is valued as much as its sonority or style. The second piece of evidence concerns the career paths of the singers at Notre Dame and some of the criticisms levelled at them: many of the key musicians at Notre Dame were forced to resign each year; they may or may not have been re-employed. Moves from an ecclesiastical position to others were therefore more than possible, and these might well have included more courtly environments in which vernacular poetry was more the norm. The collision, then, between musicians trained to sing and perhaps compose (or at least modify) organa and clausulae and a vernacular culture may well have triggered the earliest motets with French texts. Coupled to comments such as those of Robert de Courson, who criticized those who employed the magistri organici for their ‘scurrilous and effeminate things’, these observations might lead one to the same courts that gave room to trouvères and their musicians in a search for the origins of the French motet. 48
Towards the fourteenth century
Older histories of thirteenth-century music suggest that organum ceased around 1220, conductus around 1240 and that the motet occupied most of the rest of the century until it itself was supplanted by what used to be called the isorhythmic motet around 1300. This seems far too simple a view when the longevity of both organum and conductus is considered: both genres seem to have coexisted with the motet – and were performed throughout the century and beyond. But such coexistence came at a price. The rhythmic and notational changes that were triggered by the changes in motet composition in the middle of the century had an immense impact on both organum and conductus. In both cases, it was the sections that were preserved in an unmeasured notation – organum per se and musica cum littera respectively – that were subject to change. Theoretical writings from the later thirteenth century try to explain organum and conductus in terms not only of mensural notation but also of a sophisticated mensural notation as found in such treatises as Ars cantus mensurabilis or the treatise of Anonymous IV. 49
Musical sources for both conductus and organum from around 1300 exhibit the same sorts of trends: recasting the notation of organum per se and musica sine littera in mensural notation. There used to be a time when scholars would extrapolate backwards from these sources to argue that their notation represented the original rhythmic delivery of these sections, but viewing these sources as products of changing views of notation and rhythm – as part of the reception of organum and conductus – probably does greater justice to the complexity of thirteenth-century musical history. 50 The consequences were far-reaching for polyphonic music in general: whereas both organum and conductus (especially the conductus cum caudis ) exploited contrasting musical discourses that were characterized by the relationship between words and notes and by their rhythmic profile, the motet exploited a single musical discourse – preferring to exploit the vertical complexity of polytextuality and multiple voices. Around 1300, this difference was not quite entirely effaced but significantly weakened: in organum, the difference between organum per se and discantus remained in terms of the relationship between voices and in the conductus between syllabic and melismatic settings of the text. But the rhythmic differences that underpinned these differences were completely eradicated in the sources that were copied around 1300 and which presumably bear witness to a style of performance then current.