The examples cited by polyphonic theorists up to Guido are drawn from the Te Deum and also from the sequence, hymn and antiphon. Scholars have often concluded from this that early medieval polyphony was a practice that ‘stood outside the official Roman liturgy and its Gregorian chant’ (Stäblein, IMSCR IX: Salzburg 1964, ii, p.72). At first, supposedly in recognition of the ‘inviolability of liturgical chant’, it was performed only ‘where the liturgy allowed a certain freedom’ (Waesberghe, AMw, xxvi, 1969, p.264). Also its terminology is said not to include ‘the names of Gregorian chant types’, but rather ‘unaccustomed terms’ such as ‘canticum’, ‘carmen’ and ‘cantio’.
In reality, nowhere in theoretical literature, or indeed in any other relevant writings of the time, is an express distinction to be found between more and less ‘inviolable’ chants. The most that occurs is reprimand for deviations from liturgical conformity. Even this occurs primarily on political grounds. What is more, it is very difficult to conclude from the terminology of early medieval polyphonic theory that ‘official’ plainchant was excluded from polyphonic treatment. If theorists were to deal not with individual chant types but with the various chants of the liturgy in general (‘ecclesiastica cantica’, ‘sacra cantica’, ‘ecclesiastica carmina’, etc.), it was only logical that they should use correspondingly general terms. For Gregorian chant itself, for instance, ‘carmen gregorianum’ was a current overall expression (see MGG1, Stäblein: ‘Choral’, §1).
More conclusively still, as early as the first half of the 11th century, in the very earliest surviving practical sources of polyphony, categories of Gregorian chant are included that can in no sense be called ‘extra-liturgical’ or even ‘half-liturgical’. There is no evidence that in the earlier time of Musica enchiriadis any other standards, liturgical or ideological, would have applied in polyphonic practice. The problem comes down in the end to the choice of examples in early polyphonic theory. And for this purely pragmatic reasons can be adduced.
In the first place, what the theorists needed above all to demonstrate was the construction of beginnings and ends to phrases. These were not placed arbitrarily, but coincided with the natural caesuras of the chant. Hence it was sensible to fall back on chants with relatively short phrases. Other rules for the vox organalis could all be clearly demonstrated with comparatively few notes of chant, as in ex.1: parallel movement, adherence to the lower pitch limit and transfer to a neighbouring tetrachord in the course of a phrase. The extended melismas that feature in the main species of chant would have offered no appreciable gain in information, and would have wasted costly parchment. Moreover, the means of writing down that was first used was hardly suited to recording melismatic chant. It involved the placing of syllables on a grid of horizontal lines (see ex.1). Each note therefore had to have the syllable to which it belonged separately written out. If melismatic music examples were to be cited they would not only take up a great deal of space but would also suffer a great reduction in the legibility of the text: the notation of a Martianus Capella text, for example, although only lightly melismatic, appears in notation as: ‘su-ub-i-i-re cel-sa po-os-cit a-as-tra iu-up-pi-ter’ (the Bamberg Dialogue, ed. Waeltner, p.46).
If the selection of examples for early polyphony can be explained on pragmatic grounds alone, then the possibility should not be ruled out that polyphonic treatment was at least permissible, even as early as the 9th century, for the whole range of liturgical chants. The Winchester Troper transmits polyphony for the whole range of chants, solo and choral, responsorial and antiphonal, ‘old’ and ‘newer’, ‘standard’ and ‘local’. If the Chartres repertory reveals a preference for responsorial chant settings (and even suggests that Chartres may have developed a cycle of polyphony for the liturgical year analogous to the later Magnus liber of Notre Dame), the liturgical inclusiveness evident in the Winchester polyphony continued to be cultivated elsewhere for centuries to come.
6. ‘Organum’ and ‘discant’: new terminology.
Two improvisatory styles of performance are separately discussed in a group of early 12th-century treatises known as Ad organum faciendum, marking changes from the styles of the repertories discussed above (§4). Until then they had been regarded as a single improvisatory form: Diaphonia vulgariter organum. These two styles grew further apart as they developed, and became independent musical forms. In theoretical writings it became necessary to distinguish them by name. Two anonymous treatises (ed. Schneider and ed. La Fage) used the term ‘organum’ for one and ‘discantus’ for the other; but they continued to use ‘organum’ as a generic term to cover the two together. The ambivalent nature of the term ‘organum’ was first taken account of by Johannes de Garlandia, who qualified them as specialiter dictum and generaliter dictum respectively.
Immediately striking is that, of the two general terms traditionally used to signify polyphony, one (‘organum’) was taken over directly for use in a more specialized sense, and the other (‘diaphonia’) was merely taken over by analogy or by straight translation into Latin as ‘discantus’. The term ‘diaphonia’ quickly disappeared as a result. Thus both new designations, organum and discantus, were really more specialized usages of terms already previously in existence; and their new meanings rested on convention. This was possible because the specific meaning of the word ‘organum’ was still hardly known or understood at this time.
The real difference between organum and discantus at this stage lies purely in the relative amount of movement between the given part and the matching upper voices (which goes against Riemann’s theory that it is to be found in contrary motion in discantus). In organum the upper voice (vox organalis) is a melisma over the sustained single notes of the vox principalis; in discant on the other hand it forms a more or less strict note-for-note (or melisma-against-melisma) counterpoint.
It was not, therefore, the newer of the two terms, ‘discantus’, which was applied as one would expect to the newer style. It was the older term, ‘organum’, which was used for the style furthest from tradition. The reason for this is by no means obvious. Eggebrecht (III(3)1970, p.27) pointed to ‘discantus’ as a translation of ‘diaphonia’. ‘On the one hand’, he argued, ‘it retains the implication of a note-against-note progression, and indicates that the vox organalis is still plainchant-like in character. At the same time, it is a scientific term which reflects the transparency and rational nature of a note-against-note texture’. By contrast, organum, ‘as a word, was much less restricted in meaning, and could thus be applied much more easily to something that theory could not cope with and yet was successfully established in performance and in practical teaching: namely, the practice of singing melismas against single notes’. Another factor may have played a part in this. The 12th century was an era that believed in progress. It may be that the term ‘organum’ was kept for the style which was then considered most up-to-date. This would have been florid sustained-note organum, offering completely new scope for development which was being fruitfully exploited.