Josquin's Missa Pange lingua is composed on material derived from the melody to which, from the 13th century onwards, Thomas of Aquinas's adaptation of a hymn by Venantius Fortunatus may have been most frequently sung. If anything is fascinating in the composer's setting of the Ordinary text, it is certainly the way in which he has conceived from this material a melodic framework in which the declamation of the text remains crystal-clear. In those sections with much text, the opening of a phrase is sung to a minimum of notes, which strictly follows the declamation of the text. Through this approach, in imitation in all voices, these clear-cut melodies clearly affirm what has just been stated. In general, melodic continuation either follows the principle of one note to one syllable, or, within this stream, slightly emphasizes a particular word by a few extra notes, which may stress its particular meaning as well as the apparently French pronunciation of the text. Towards cadences between two or more voices in imitation, the leading voice may approach the close of its line with a short improvisation on a foregoing melodic element, or by a subtle embellishment that not infrequently functions as exclamation sign. In those Ordinary movements with little text, the structure of melodic phrases in general leaves no doubt where repetition of text has been intended, particularly in the long-winded duos in the Sanctus. In these sections it is fascinating to rediscover the composer's step-wise presentation of text.

With these basic elements, the composer weaves a web of astonishing refinement, in which every melodic and rhythmic element has its place. Within the setting's variety of combinations of these elements - by way of an extremely balanced counterpoint - not a single note merely functions as a filler. As an additional introduction to this style of setting, the editorial underlay of the Ordinary texts (which sometimes deviates from the setting's edition in the New Josquin Edition) may demonstrate the way in which the composer generated his inspiration.

In general the Phrygian mode of the setting does not allow strict melodic imitation at the fifth below the final. However, the introduction of a b-flat before the third note of the 'Pange lingua' motive at the start of the Credo (Bassus/Contra, mm. 3/18), and the incorporation of melodic lines that clearly demand for a b-flat fa super la, invite more careful exploration of the setting, in order to locate these isolated escapes from the Phrygian hemisphere. In this respect the added editorial ficta in the score, as related to this particular aspect, is primarily meant as a stimulant for personal investigation.

Transmission

The transmission of Josquin's Missa Pange lingua in BrusBR IV.922 is one of a total of 26 instances, 14 of which present the composition in a complete or nearly complete reading. Some of these readings were copied in manuscripts produced in Josquin's lifetime or shortly thereafter; their provenance as well as their readings of the Missa Pange lingua offer a unique insight into the way Josquin's setting may have been transmitted throughout Europe in the first half of the 16th century.

One of the earliest readings of the complete mass seems to be preserved in the Vatican choirbook MS Cappella Sistina 16, copied in Rome around 1515-1516 by Claudius Gellandi for use by/in the Cappella Sistina. An almost identical reading, copied most probably from this source between 1518-1521, is transmitted by the Roman choirbook MS Cappella Giulia XII.2, as well as in some incomplete sets of partbooks in the Vatican Library, MSS Palatini Latini 1980-1981 and 1982, which were copied in Rome before 1523. These partbooks were in the possession of cardinal Giulio de' Medici. This isolated Roman transmission of the mass suggests that only in the second decade of the 16th century did a copy of Josquin's mass become known in circles directly related to the Vatican.

In contrast to these readings, the Roman choirbook MS Santa Maria Maggiore 26, copied by several scribes between 1516 and 1520, furnishes a heavily-edited reading of the mass, in which under-third cadences are ommitted, ligatures resolved and a large number of rhythmic substitutions introduced. Apparently, in some musical circles in Rome, certain aspects of Josquin's setting, as well of its notation, already held the odium of being "old-fashioned."¹ Equally distant from Josquin's original intentions is the reading of the Mass in the mid-16th-century MS Milan, Bibl. Ambrosiana, Ms E. 46 Inf. Apart from several copying errors, its reading offers both resolved and newly added ligatures and quite a host of rhythmic substitutions, which for the most part have little impact on the placement of text.

Although including a number of editorial reworkings, the reading in Toledo, Biblioteca Capitular, MS B.16, copied 1542 in Toledo, may be a later descendent of the Roman tradition as transmitted in VatS16.

From about 1516 onwards the 'Alamire' scriptorium in Mechelen started the copying of several isolated fascicles, which around 1520 were assembled into a choirbook, today MS BrusBR IV.922. Apart from a few mistakes and some alternative cadential formulas, it delivers a reading of the Mass which, in principle, is quite in agreement with its reading in VatS16. An almost identical reading of the mass in the choirbook Jena, Thüringer Universitäts- und Landesbibliothek, MS 21, copied in the same scriptorium between 1521 and 1525, makes it highly probably that the readings in BrusBR IV.922 and JenaU 21 may have been copied from twins, originating from the same exemplar, but each with additional annotations which may have slightly differed from each other. Yet one of these twins must have included a rather peculiar anomaly: in BrusBR IV.922 Josquin's rather long and sophisticated two-voice settings of the 'Pleni sunt' and 'Benedictus' are replaced by shorter settings, taken over from the Missa Es hat ein sinn by Mathieu Gascongne. Moreover, the second Agnus dei, also for two voices, is not included in the manuscript. Therefore it looks as if the 'Alamire' scriptorium first obtained a copy of the mass in which the rather highly demanding settings for two soloists were replaced by other settings, and only somewhat later obtained a fair copy of Josquin's original version. That Alamire's copyists must have worked with several exemplars of the complete mass is verified by the redaction of a third reading of the mass, transmitted in the choirbook Vienna, Österreichische Nationalbibliothek, Handschriftensammlung, Ms 4809, copied between 1521 and 1525 in the same scriptorium. Its redaction makes clear that additional remarks and rewritings, as added by the copyists in their various exemplars, were not brought into line with each other. In VienNB 4809 most of the under-third cadences are suppressed, and several unpractical rhythmic substitutions as well as other unique readings have been introduced. Moreover, its reading of the mass includes far more copying errors than the other copies of the mass from the scriptorium.

This group of sources from the 'Alamire' scriptorium clearly demonstrates that Josquin himself was not directly involved in the dissemination of his mass by way of the 'Alamire' scriptorium. On the contrary, the various transmissions in these Alamire manuscripts suggest that, prior to their copying, performances of the setting outside the direct control of the composer had made clear that its various two-voice sections either were too demanding for the average singer, or that these sections did not suit his taste; hence the alternative sections in BrusBR IV.922.

Since the style of the composition points to a rather late work by Josquin, singing of the mass may have been restricted for a period to the church of Condé. Was it perhaps one of the singers in Condé who (secretly?) made copies of the setting, and distributed those copies among colleagues? In this respect the very unsatisfying underlay of text in the Sanctus and the Agnus dei as transmitted by the 'Alamire' manuscripts may point to some earlier copying in haste. For these sections, text may have been added later by copyists from the scriptorium. In any case, one of these early copies may have traveled to Rome, where it seems to have been treasured immediately. The setting's whereabouts in the Low Countries before Alamire first got his hands on one of its readings are still a mystery.

The reading of the mass transmitted in the choirbook Munich, Bayerische Staatsbibliothek, Musiksammlung, Musica MS 510, possibly copied in Augsburg or Munich around 1513-1519, seems to confirm this picture. Its reading is in remarkable agreement with that in VatS16, but shares variant readings with JenaU 21 as well. Although slightly edited, the setting's reading in the anonymous print of 1559, Missa super Pange Lingua, may also be a late descendent from an early copy. Together with MunBS 510 it shares the same variant readings with JenaU 21. Supposedly the East Slovenian partbooks from the third quarter of the 16th century, Budapest, MS Bártfa 8 (a-d), also descend from a comparable early copy.

In contrast to these sources, the edition in four partbooks by Grapheus of 1539, Missae tredecim quator vocum, offers several copying errors and a reading in which many under-third cadences and anticipations are ommitted and ligatures resolved. Moreover, a number of unique variant readings have been introduced. This edition must have functioned as model for the copying of the mass in the MSS Leipzig, Universitätsbibliothek, MSS Thomaskirche 49/50, Regensburg, Bischöfliche Zentralbibliothek, MS C100 and Rostock, Bibliothek der Wilhelm-Pieck-Universität, MS Saec. XVI-71/73. Their readings also include additional editorial reworkings.

Analyses of the variant readings of the mass in BrusBR IV.922 against JenaU 21, VatS16 and MunBS 510 (see Table 1) seem to underline that transmission's isolated position. Apart from the absence of the original 'Pleni sunt celi,' the 'Benedictus' and the second Agnus dei sections, BrusBR IV.922 has no particular errors, but includes 8 unique readings: three melodic variants, one rhythmic substitution, two unique ligatures and two variant cadence formulas. Given the quality of these unique readings, most of them may be the result of some performing practice outside the composer's control.

Table 1: Variant readings in JenaU 21, VatS16 and MunBS 510 against BrusBR IV.922

Kyrie
Measure/Notes	Voice	Variant	JenaU 21	VatS 16	MunBS 510
95-7	S	mi sm sm	X	X	X
156	C	S L			X
345-351	T	no lig.			X
48	B	mi mi c d S A	X	X	X
512	T	no round b	X
631	T	f [error]			X
Gloria
Measure/Notes	Voice	Variant	JenaU 21	VatS 16	MunBS 510
72	T	mi d' dot sm c'	X	X	X
443	T	b		X
664	T	sm sm b a		X	X
713	T	S S		X
822-5	T	S e' mi d'	X
86	T	mi mi		X	X
944	S	sm sm g' f'		X	X
Credo
Measure/Notes	Voice	Variant	JenaU 21	VatS 16	MunBS 510
354-362	S	S b' mi a' dot sm g'	X
46	S	lig.		X	X
564	C	sm sm c' b		X	X
582-3	T	mi g dot sm sm sm f f e	X	X	X
593	S	mi d'' dot sm b'	X	X	X
73	C	S S	X
981-2	C	Br		X	X
1305	S	round b			X
1383	C	S e' mi-rest		X	X
1641-2	C	mi mi		X	X
194	T B	no mensuration sign	X	X	X
2004	C	sm sm c' b		X	X
201	T B	no 3	X	X	X
Sanctus
Measure/Notes	Voice	Variant	JenaU 21	VatS 16	MunBS 510
52	S	mi d'' dot sm c''		X	X
174-6	S	mi dot fu fu		X	X
21-91	Missing in BrusBR IV.922
1041-2	B	no lig.		X	X
1201-2	S	lig.	X	X	X
1261-2	S	no lig.		X	X
1301-2	B	lig.	X	X	X
138	B	E + e			X
139-86	Missing in BrusBR IV.922
Agnus dei
Measure/Notes	Voice	Variant	JenaU 21	VatS 16	MunBS 510
51-2	C	Br	X	X	X
212	S	no round b		X	X
26-81	Missing in BrusBR IV.922
982	B	mi <line break> sm		X
1163	C	mi mi		X
1261	B	mi b [error]	X
1364-1371	T	mi mi d' d'	X
1401-1411	B	S-rest mi dot sm mi mi	X
1411	B	mi mi		X	X
148	T	no lig.		X	X
157	C	e only		X

Pre-existent material: Plainchant hymn Pange lingua gloriosi, treated as basis of melodic lines and points of imitation

Cambrai, Bibliothèque municipale, Impr. XVI C 4, f. 73v (Cambrai antiphoner, c. 1508-1518)

Editorially supplied material: Plainchant intonations of Gloria and Credo supplied from Paris, Bibliothèque nationale, MS lat. 14452, f. 243r-v (Gradual of St. Victor of Paris, 13th century, with additions dated 1567)

Notes
[1] As a specific example of early 16th-century editing, a complete list of the manuscripts' contrasting readings has been included in the New Josquin Edition Critical Commentary: Masses based on Gregorian chants 2, ed. Willem Elders, NJE 4 (Utrecht: Koninklijke Vereniging voor Nederlandse Muziekgeschiedenis, 2000).