The Rise of Polyphony : Magister Leoninus and Magister Perotinus

Cover of St. Martial manuscript.

I think we tend to take polyphony for granted.

We inherited a form of music which has been with us for centuries and, in fact, we have even seen major-minor harmonic polyphony peak in the 18th century and then gradually give way and break down completely.  However, what we take for granted took centuries to develop: from improvised melismatic elaborations of a plainchant over a drone tone to the first attempts to organize this free melodic movement with notated rhythm and early ideas of organizing the vertical sounds into progressive movement toward a cadence.

I will go back to the beginning, back before major-minor, before even the idea existed of chords and appreciating music vertically.  Our understanding of these early periods is somewhat hazy and unclear since the documentary evidence is scant, but it is sufficient for a firm grasp of how polyphony began and proceeded, ultimately, to produce a master composer like J.S. Bach whose manipulation of polyphony was arguably the greatest the world has seen.

“Resonemus Hoc Natali (Anon. St. Martial)”, The Age of Cathedrals, Theatre of Voices, Paul Hillier.

Beginning as an elaboration of plainchant, polyphony served almost exclusively as a means of increasing the splendor and solemnity of church services from the ninth century until well into the thirteenth.  Significantly, polyphony originated and was first cultivated extensively in the same centuries as in some of the same places that produced the large repertories of tropes and sequences.[1]

Early polyphony was produced or made, not “composed” in the present-day sense of composition: it was in the first instance sounded, not notated.  Polyphony arose and continued as a performance practice, a way of elaborating a known monophonic melody with a second line that was produced according to accepted conventions.[2]

However, it would be absurd to claim that polyphony was unknown prior to the ninth century.  We only have to imagine that as soon as man began to sing, soon more than one person was singing together.  If these two happened to be a man and a boy then the most natural occurrence would be for them to sing in octaves.  As man became more and more aware of intervals beyond the unison it does not take much imagination to see how improvised polyphony could have developed long before the ninth century.

Since plainchant was in use in the church for several centuries it was only a matter of time before singers might add a line above or below the main singer’s chant melody.  However, this kind of singing was done prior to the advent of written notation; we would have to wait until the ninth century before the first appearance of polyphony appeared in written sources.

“In Hoc Anni Circulo” (Anon. St. Martial), The Age of Cathedrals, Theatre of Voices, Paul Hillier.


We must depend on theoretical treatises, therefore, for knowledge of the earliest stages of its development.  Two nearly contemporary authors, Regino of Prüm (d.  915) and Hucbald (c. 840-930), wrote treatises with the same title, De harmonia institutione (Concerning Harmonic Instruction).  Both introduce the term “organum”, and both attempt to define consonance and dissonance.  More detailed descriptions with musical examples are given in two of the most important treatises of the period, the Musica enchiriadis (Music Manual) and the Schola enchiriadis (Commentary on the Manual).  While the exact dates of composition are in dispute for these two treatises, there can be little doubt that they both describe organum referred to in the authentic Hucbald treatise as it was practiced in the latter half of the ninth century.[3]

The earliest examples of organum are called parallel or strict because the basic principle is the duplication of an existing plainchant in parallel motion at the interval of an octave, a fifth, or a fourth.  The vox principalis (principle voice) is the main voice singing the plainchant and the additional voice is called the vox organalis (organal voice).  This kind of singing has been called simple organum.  As it developed, the further doubling of these voices at the octave produced composite organum.

Parallel organum, when done strictly, led to dissonant intervals which were unacceptable, e.g.  augmented fourths.  The solutions found to avoid dissonant intervals included the organal voice holding one note while the chant voice moves (drones), or the organal voice moving in contrary motion relative to the chant.  Contrary motion first began to be used as the preferred method of approaching a cadence, but using “drone tones” was something new.  These singing methods produced intervals such as seconds, thirds and even sevenths.  The style was eventually called free organum and could truly be said to approach what we would recognize as polyphony.

Codex Calixtinus: Benedicamus domino deo gracias

As scribes and composers attempted to capture polyphony on parchment the first problem was rhythm.  Improvising polyphony would probably often create dissonances and with more development towards composed polyphony composers and singers wished to avoid dissonance and have consonant intervals occur as much as possible.  As the desire to have certain notes converge it became obvious that some method of describing rhythm was needed.

Because there had been a long history of describing metrical feet for poetry, already in existence were a group of simple cells which described the rhythm of a metrical foot.  It was not a huge leap for medieval musicians and theorists to adapt these metrical feet to music.  They called them “rhythmic modes” mirroring the scalar modes.

In most sources there were six rhythmic modes, as first explained in the anonymous treatise of about 1260, De mensurabili musica (formerly attributed to Johannes de Garlandia, who is now believed merely to have edited it in the late 13th century for Jerome of Moravia, who incorporated it into his own compilation).[4] Each mode consisted of a short pattern of long and short note values (“longa” and “brevis”) corresponding to a metrical foot, as follows[5]:

  • Long-short (trochee)
  • Short-long (iamb)
  • Long-short-short (dactyl)
  • Short-short-long (anapaest)
  • Long-long (spondee)
  • Short-short-short (tribrach)

Although this system of six modes was recognized by medieval theorists, in practice only the first three and fifth patterns were commonly used, with the first mode being by far the most frequent.[6] The fourth mode is rarely encountered, an exception being the second clausula of Lux magna in MS Wolfenbüttel 677, fol.  44.[7] The fifth mode normally occurs in groups of three and is used only in the lowest voice (or tenor), whereas the sixth mode is most often found in an upper part.[8]

Theory is almost never pure description.  It is usually a representation not of the world the theorist sees but of a more orderly, more easily described world the theorist would like to see.  A persuasive theory, particularly one of suggestible human behavior or practice, can often to some extent reshape the world to conform, for better or worse, to the utopian image.  But an attractive theory uncritically accepted can also blind the believer to existing conditions, and lessen rather than enhance comprehension.  Uncritical acceptance of Garlandia’s six-mode scheme can obscure the actual history of musical practice at Notre Dame, and that is why it should be regarded as a secondary rather than a primary source of knowledge.[9]


st-martialTwo monastic houses are particularly important to our understanding of these new genres, the monastery of St Gall in the East Frankish kingdom, and the monastery of St Martial in the heart of the Aquitaine region of West Francia.  Their libraries preserve the largest collections of early tropes and sequences; their books are the foundation of our knowledge about this music before 1100.[10]

Tropes were musical methods of extending the duration of sections of the Ordinary of the mass.  Most often tropes were attached to the Introit, with the trope functioning somewhat as a commentary on what was happening during a specific section of the mass.  Tropes often would explain and even validate what was happening during the service.

Manuscripts that contained tropes were called “tropers” and the collections housed at Westminster, and monasteries at St. Martial and St. Gall were important depositories.

compostelaIf troping may be considered a performance of an existing work with accretions rather than the creation of a new work through borrowing, early polyphony was arguably also a manner of performance rather than a kind of borrowing.  Singing a chant with a drone or in strict parallel octaves, 5ths or 4ths does not result in a new piece but rather in a different way of presenting an existing one.  This is still true of the mixed parallel and oblique organum described in Musica enchiriadis, whose rules generate the added voice or vox organalis almost automatically below the chant melody or vox principalis.  In each of these styles, the polyphony enhances the presentation of the existing chant; the added voice lends greater resonance and thus weight and solemnity, and the use of a drone or oblique motion closing on a unison heightens the sense of melodic direction and cadential closure and thus clarifies the phrase structure.[11]

Before considering the Notre Dame school of polyphony there is one other monastery whose collection of manuscripts is also very important and unique: Santiago de Compestela in the northwestern Spanish province of Galicia.

codex_calixtinus-_inicio_del_prologoThis monastery was reputed to have housed the relics of James the Apostle and was the site of pilgrimage throughout the Middle Ages, rivaling Jerusalem and Rome.  Among the manuscripts which were preserved here, along with early polyphony, was a tourist’s map of the roads leading in and out and important sites along the way.   But what has been most important to music historians is Liber Sancti Jacobi which preserved entire services including their music as they were performed in the middle of the 12th century.

The polyphonic repertory at Compostela consists of 21 pieces which appear in the Codex Calixtinus, and its appendix, called that because it was attributed to Pope Callixtus II, which is dated to 1140.  Most are monophonic chants to which a second voice has been added: a few conducti, tropes for the Kyrie, settings of the Benedicamus and other stophic poems that do not seem to have a liturgical function.

The conductus, the sequence and the discant were some of the forms in which plainchant was expanded into polyphony.  Often the lowest voice would have an elongated chant line; over this, a melismatic upper voice would be sung.  The evolution which saw the flowering of polyphony from monophony took place over a period of approximately 200 years, from the 9th through the 11th centuries.  But this early polyphony was written anonymously.  It was not until the 12th century that the first two names of composers writing early polyphony appear.


leoninLeoninus (fl Paris, 1150s–c1201) was a composer of polyphony, including organum and, probably, conductus.  He is credited by the theorist Anonymus IV with perhaps the single greatest achievement in the development of early polyphony, the creation of the so-called Magnus liber.  This collection is believed to constitute the matrix in which polyphony was transformed from performance practice into ‘composition’ in the modern sense of the word.  In its repertory appeared a system of consonance and dissonance that would dominate polyphonic composition for the next three centuries, and also a coherent rhythmic language that, for the first time in Western music, expressed itself in its notation.[12]

The composer Leoninus is a pivotal figure in the history of Western art music, yet to the moment almost nothing is known of his life.  Leoninus claims a special place in musical historiography on several counts.  For instance, prior to Leonin’s arrival on the intellectual scene of 12th century Paris, liturgical polyphony was characterized by an almost wholesale anonymity.  Aside from a handful of compositions in the Codex Calixtinus, few previous polyphonic works can be attributed to specific individuals.[13]

Leoninus, we are told, wrote a cycle of two-part settings of the most important chants in the liturgical year—Christmas, Easter, Assumption and other feasts; this cycle was called the Magnus liber organi (‘The great book of organum’).  Organa of the type that make up Leoninus’ Magnus liber organi are polyphonic settings of plainsong.  The original chants employ two musical styles: the solo sections are elaborately melismatic and contrast with the simpler, more syllabic, sections sung by the schola.  It is the melismatic solo sections of the chant that are set polyphonically.  The result is that a performance of organum involves polyphony and plainsong.[14]

Performing the duplum lines in organum per se is a skill that is difficult to regain in the 21st century.  The music notated in the original manuscripts gives a mixture of information: some idea of what the composer’s overall structure might have been, and an idea of at least one (and probably more than one) performer’s view of the music.  And it has to be remembered that a ‘performer’s view’ of this music would almost certainly have entailed changes to pitch and rhythm, and a sense of what a thirteenth-century editor would have done in trying to copy down and render consistent a wide range of material.

So these sections which, in their floridity, resemble late-eighteenth- and early-nineteenth-century coloratura vocal lines, differ in that they are not just blueprints to be ornamented; they are blueprints that have already been partially ornamented, and the singer of the duplum part treads a very careful path between the slavish duplication of a medieval performer’s view of the work and the complete recreation of Leoninus’ music.[15]

perotinFollowing Leoninus there was Perotinus (fl Paris, c1200).

Virtually everything known about Perotinus’s musical activity is extrapolated from a passage in the treatise of Anonymus IV.  Noting that Leoninus, the “optimus organista” (‘best man with organum’), ‘made’ the Magnus liber to embellish the liturgy, he remarks:

[This liber] was in use up to the time of the great Perotinus, who made a redaction of it [‘abbreviavit eundem’] and made many better clausulas, that is, puncta, he being the best discantor, and better [at discant] than Leoninus was.  … This Magister Perotinus made the best quadrupla, such as Viderunt and Sederunt, with an abundance of striking musical embellishments [colores armonicae artis]; likewise, the noblest tripla, such as Alleluia, Posui adiutorium and [Alleluia], Nativitas etc.  He also made three-voice conductus, such as Salvatoris hodie, and two-voice conductus, such as Dum sigillum summi Patris, and also, among many others, monophonic conductus, such as Beata viscera etc.  The book, that is, the books of Magister Perotinus, were in use in the choir of the Paris cathedral of the Blessed Virgin up to the time of Magister Robertus de Sabilone, and from his time up to the present day.”[16]

As interesting as Anonymous IV’s statements are they would be of little value to us unless confirmed by other sources.  Such sources do exist in several manuscripts that preserve the repertory of the School of Notre Dame.  There are three primary manuscripts that preserve the Notre Dame polyphony, two are referred to as W1 and W2 named for the northern German location where they were found, Wolfenbüttel.  One of these manuscripts made its way to France and the other ended up at St. Andrews, Scotland.

The names Leonin and Perotin do not appear in these manuscripts, nevertheless, the correspondence between their contents and the statements of Anonymous IV is extraordinary.  The core repertory consists of two-voice settings of responsorial chants for the great feast days of the Church year.  All manuscripts arrange these settings in two groups, each of which follows the order of the Church calendar beginning with Christmas Day.  These musical sources vindicate Anonymous IV as an historian and make it possible to reconstruct Leonin’s Great Book of Organum.  They also allow us to follow Perotin’s process of revision and improvement, and to study the first polyphonic pieces that can be attributed with certainty to individual composers.[17]

[1] Hoppin, Richard H. 1978. Medieval music. New York: W.W. Norton, p.187
[2] Everist, Mark. 2011. The Cambridge companion to medieval music. Cambridge: Cambridge University Press. eBook.
[3] Hoppin, Richard H. 1978. Medieval music. New York: W.W. Norton, pp. 188-189.
[4] Baltzer, Rebecca A. 2001. “Johannes de Garlandia [Johannes Gallicus]”. The New Grove Dictionary of Music and Musicians, second edition, edited by Stanley Sadie and John Tyrrell. London: Macmillan Publishers.
[5] Reece, Gustave. 1940. Music in the Middle Ages. New York: W. W. Norton & Company. Pp. 207-209.
[6] Apel, Willi. 1961. The Notation of Polyphonic Music, 900–1600, fifth edition, revised and with commentary. Publications of the Mediaeval Academy of America, no. 38. Cambridge, Mass.: Mediaeval Academy of America, p. 223.
[7] Hughes, Dom Anselm. 1954a. “Music in Fixed Rhythm”. In New Oxford History of Music, vol. 2: “Early Medieval Music up to 1300”, edited by Dom Anselm Hughes, 311–52. London, New York, & Toronto: Oxford University Press, p. 320.
[8] Ibid.
[9] Taruskin, Richard. 2010. Music from the earliest notations to the sixteenth century. Oxford: Oxford University Press. eBook.
[10] Everist, Mark. 2011. The Cambridge companion to medieval music. Cambridge: Cambridge University Press. eBook.
[11] J. Burkholder, Peter . “Borrowing.” Grove Music Online. Oxford Music Online. Oxford University Press, accessed February 9, 2017,
[12] Roesner , Edward H. “Leoninus.” Grove Music OnlineOxford Music Online. Oxford University Press, accessed January 27, 2017,
[13] Wright, Craig. “Leoninus, Poet and Musician.” Journal of the American Musicological Society, vol. 39, no. 1, 1986, pp. 1–35.
[14] Everist, Mark. “Magister Leoninus, Vol. 1 – Sacred Music from 12th-century Paris”, booklet with recording, Hyperion CDH55328, p. 3.
[15] Ibid.
[16] Edward H. Roesner. “Perotinus.” Grove Music Online. Oxford Music Online. Oxford University Press, accessed February 9, 2017,
[17] Hoppin, Richard H. 1978. Medieval music. New York: W.W. Norton, pp.217-219.

Recordings of early polyphony

The Age of Cathedrals, Theatre of Voices, Paul Hillier,  Harmonia Mundi

Magister Leoninus: Sacred Music from 12th Century Paris (2 vols.), Red Byrd, Hyperion 66944

Polyphonic Aquitaine of the 12th Century, Ensemble Organum / Marcel Pérès, Harmonia Mundi 1901134

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s