The Symphonia of Hildegard of Bingen: Ambiguities in origin and intention
Catherine Jeflreys
The Symphonia harmoniae caelestium It appears that Hildegard began composing revelationurn [Symphony of the Harmony of Ce- music for liturgical use in the 1140s; two letters lestial Revelations] of German abbess and vision- written to Hildegard in 1148 indicate that her ary Hildegard of Bingen (1098-1179) occupies a compositional activities date fromas early as 1141 .3 unique position within the corpus of twelfth-cen- During the period in question, it was common in tury plainchant. This cycle of seventy-seven litur- monastic women's communitiesfor one woman to gical songs and a liturgical drama is one of the few assume responsibility for the musical activities of examples where two extant versions of the one the convent, and this often involved the composi- body of plainchant by a single identifiable com- tion of items for liturgical use. In smaller comrnu- poser are available to us from an age, the works of nities, such as that at Mount Saint Disibode, it was which are often characterised by the anonymity of not unusual for the abbess to assume that respon- their composers. The two Symphonia manuscripts ~ibility.~As such, Hildegard may have been in a have long been the subject of considerable debate position to compose items for liturgical use as as the Symphonia survives in two distinct versions, early as 1136. distinguishable in terms of the content of each The idea of collecting together composed items manuscript, that is the number and types of items into a song cycle was not new at the time; famous included in each version of the Symphonia; the examples include Notker of Saint Gall's Liber ordering of items in each manuscript; and the ymnorum from the ninth century and Abelard's intended purpose of each codex in which the Hymnarius Paraclitensis from the twelfth. Both Symphonia is preserved.' The issues concerning Notker of Saint Gall's cycle and the first two books the compilation and content of the two manu- of Abelard's Hymnarius are arranged according to scripts in question provide an insight into the the liturgical calendar; the third book of Abelard's ambiguities in origin and intention (whether the cycle resembles Hildegard's Symphonia as in both, Symphonia was intended as a collection of songs or items devoted to a particular subject are grouped a Gesamtwerk) of the musical output of one of t~gether.~Whether the concept of a song cycle history's most significant woman composers.2 implied performance as a cycle remains unclear. Some biographical details about the Abbess However, the concept of preserving liturgical items Hildegard are important, to establish the historical by providing a structure, that is, an ordering of locus of her Symphonia. Born in 1098 in liturgical items according to the liturgical calender Bermersheim (about 20 kilometers southeast of or by subject matter, is apparent. A distinction is Bingen, which is on the Rhine in the western part to be made here between a manuscript that has of Germany),Hildegard was placed into the care of been compiled as a miscellany, with items merely Jutta of Spanheim who occupied a small cell grouped together according to subject matter, and attached to the Benedictine monastery of Mount a manuscript in which the structuring of each Saint Disibode, near Bingen. Upon Jutta's death in subject group is based on aparticulardesign which 1136, Hildegard became abbess to the small com- displays a structural unity consistent with the no- munity of nuns at the monastery. The visionary tion of a Gesamtwerk. encounters, first experienced in Hildegard's child- The earliest reference to Hildegard's Symphonia hood, intensified with age, and after Jutta's death song cycle appears in the 1150s. In the preface to the matter was submitted to prelates who authen- her Liber vitae meritorum (1 158) Hildegard refers ticated Hildegard's claims to divine perspicacity. for the first time to the Symphonia as one of several This prompted a steady flow of written works items worked on during the 1150~.~As the texts which date from 1141 to her death on 17 Septem- to fourteen songs which appear in both versions of ber 1179. Part of her visionary experience resulted the Symphonia are found in the thirteenth vision of in one of the largest bodies of plainchant ascribed Book I11 of Hildegard's first written work, the to a single composer. Scivias (1 15I), it is generally assumed that, firstly, this vision formed the basis of the Symphonia song scripts, Dendermonde (Belgium), Saint-Pieters & cycle and, secondly, the Symphonia was com- Paulusabdij MS. Cod. 9 (hereafter D), dates from pleted by 1158. However, there is evidence of around 1175, some four years before Hildegard's Hildegard's compositional activities both before death in 1179. Two letters written to Hildegard in 115 1 and after 1158. For example, the thirteen around 1175 confm that D was originally in- items devoted to Saint Ursula were composed after tended as a gift of didactic texts and songs for a Hildegard's exposure to the cult of the Saint in community of Cistercian monks in Villiers, Bel- Cologne between 1158 and 1163.7 gium.10 D comprises four works: Hildegard's I As to the question of whether her compositions Liber vitae meritorum, the Liber viarum of existed prior to the song cycle, the genres repre- Hildegard's contemporary Elizabeth of Schoenau, 1 sented in the Symphonia suggest that the items the Symphonia and an anonymous dialogue. The included'in the cycle were composed specifically D codex predates the only other extant version of for liturgical use. As such, they could exist outside the Symphonia by over a decade. Included in the the song cycle. All but three of the items included Riesenkodex, Wiesbaden, Landesbibliothek Hs. 2 in the Symphonia (two Symphonies and an unclas- (hereafter R), the later Symphonia was compiled sified devotional song) can be integrated into the posthumously between 1180 and 1190.l prescribed services of the Mass and the Divine We know that the twoversions of the Symphonia Office. In addition to a Kyrie and an alleluia-verse, were compiled independently from one another as i Hildegardcomposed seven sequencesfor the Mass, D was in Villiers while R was being prepared in sung between the Alleluia and the Gospel. Compo- Rupertsberg. Furthermore, results of exarnina- sitions from the Symphonia associated with the tions of the textual and neumatic content of each i Divine Office include four hymns, eighteen re- manuscript suggest that the manuscripts were not sponsories and forty-three antiphons.* copied from the same source.l2 Even so, of Taking into consideration historical accounts Hildegard's seventy-seven extant liturgical songs, of Hildegard's compositional activities and the fifty-seven are includedin D and seventy-fivein R. genres included in the Symphonia, it is possible Fifty-five songs are common to both sources, that Hildegard's compositions from the 1140s (or although R includes the only extant version with perhaps earlier) and 1150s were collected into a neumes from the twelfth century of Hildegard's cycle once a number of items has been amassed. liturgical drama, the Ordo Virtutum.13 This seems likely as the two manuscripts in which The text of the Ordo Virtutum appears in em- the Symphonia survives date from between 1175 bryonic form in the thirteenth vision of Book 111 of and 1190. The order in which the items appear in the Scivias, alongside the fourteen song texts. the cycle, as Ruth Lightbourne suggests, could Hildegard subsequently expanded the text of the have been revealed to Hildegard through visionary Ordo Virtutum and gave it a musical setting for experience, as documented in the Scivi~s.~This performance by the nuns at Rupertsberg. The has important implications for the way we think of appearance of the liturgical drama in the Scivias the Symphonia, as it is commonly considered a seems to imply that the Ordo Virtutum formed a single work (as opposed to a compilation of items) part of the original Symphonia, as the ordering of which underwent a process of development, while items in the Symphonia appears to have been based the two extant versions of the Symphonia are seen upon the ordering of items in the Scivias. As such, to represent two stages of that developmental the Ordo Virtutum has been considered an integral process. However, if one takes a closer look at the part of the cycle. Its absence from D has prompted two manuscripts, problems associated with the the suggestion that it appeared before the surviving idea that the Symphonia underwent such a process Symphoniain that manuscript, on folios now lost.14 of development which was completed in the latter It is highly unlikely, however, that the Ordo manuscript come to light. Virtutum ever formed past of the Symphonia in D. The two manuscripts in which the Symphonia Firstly, the missing folios theory is based on the is preserved with musical notation were compiled presence of a psalm cadence in the top right hand at the Rupertsberg abbey (situated across the River comer of the opening folio of the D Symphonia. A Nahe from Bingen), Hildegard's official residence psalm cadence, or diferentia, is a melodic formula from around 1147. The older of the two manu- which provides a connection between the end of a
14 Context 7 (Winter 1994) psalm and its accompanying antiphon. It has been antiphon. This occurs especially on the initial believed that this cadence does not belong to the sounding of this note which is described in neumatic opening antiphon 0 magne pater but to another notation by a pressus (a ), a neume which gives now lost. This assumption is based on the conven- emphasis to the upper of its two notes (see Exam- tional placement of psalm cadences after the anti- ple 1). phons they accompany. The practice of psalmody involved the singing of an antiphon, followed by the chanting of a psalm on a variable pitch deter- mined by the mode of the antiphon; the psalm - 0 mag- nepa- te.r, cadence provided a tonal link from the psalm back to its antiphon. As psalm cadences precede the antiphons they accompany in performance, it is - in mag- na aecessi- tr- Ie su- mus not unreasonable to expect psalm cadences to appear in written form before the antiphons they Example 1: Opening of 0 magne pater accompany. Given that only two antiphons and two psalm cadences appear on the opening folio of As such, a mode two psalm cadence could not be the D Symphonia, the cadence in question could considered inappropriate for an antiphon that em- well belong to the opening antiphon 0 magne phasised the final of mode two in its opening, pater, particularly as psalm cadences appear be- despite beginning and ending on the note a. It can fore the antiphons they accompany, and after items not, therefore, be said that the psalm cadence in the other than antiphons, at five other points in the top right hand comer of the opening folio of the D manuscript.15 Symphonia does not belong to the antiphon 0 Ultimately, the choice of psalm cadence was magne pater, especially as this cadence is more determined by the mode of the accompanying appropriate to this antiphon than the cadence fol- antiphon. The compilers of D utilised four basic lowing, which is particular to antiphons in modes types of psalm cadence: a common cadence ac- three and four (the following item on the folio is the companies antiphons in modes three and four, antiphon 0 eterne deus, which is in mode three). while items in modes one, two and eight are given As such, the notion of an imagined antiphon which separate cadences. The psalm cadence accompa- proceeded the D Symphonia and was apparently nying 0 rnagnepater may well have been consid- accompanied by the Ordo Virtutum, or other uni- ered inappropriate as the psalm cadence in ques- dentified items, can not be merited. tion accompanies two other antiphons, both in Psalm cadences aside, the original purpose for mode two (the final of which is the note d).l6 0 the compilation of D may provide a reason for the magnepater, on the other hand, begins and ends on absence of the Ordo Virrutum from that manu- the note a. The appearance of melodies which script. The Cistercians, for whomD was intended, closed on notes that existed outside the eight observed a particularly strict version of the Ben- church modes (that is, ended on notes other than d, edictine rule and rejected the decorative elements e, f or g) occurs frequently throughout the period introduced into the liturgy during the eleventh and in question;17 two other antiphons in D also begin twelfth centuries, of which liturgical drama formed and end on the note a (Spiritus sanctus vivificans a part.lg It seems unlikely that Hildegard would on fol. 157r and Et ideo puelle iste on fol. 168r), have included a version of the Ordo Virtutum, an but both are accompanied by a cadence associated example of liturgical embellishment, in a manu- with antiphons in mode one. As such, the appro- script intended for Cistercian monks. priateness of the psalm cadence accompanying 0 As to the presence of the Ordo Virtutum in R, magne pater could be questioned. it is important to recognize that the R codex consti- However from the tenth century on, the con- tutes an anthology of Hildegard's visionary works nection between psalm and antiphon was made not which was assembled presumably in anticipation between two notes but between two groups of of her canonization, for the manuscript only in- notes, that is between the end of the psalm and the cludes divinely 'inspired' works and omits those opening phrase of its antiphon.18 In 0 magne which were not the result of visionary experience. pater, particular emphasis is placed on the note d, It is thought that R was compiled as apresentation the final of mode two, in the opening of the manuscript for submission to the prelates who
The Symphonia of Hildegard of Bingen would have overseen the canonisation process.20 Ne~man.~~Firstly, the placement of the Virgin These church officials, in the search for signs of Mary before the Holy Spiritfrrinitycontradicts the saintliness, would have placed an emphasis on the more conventional placement of the Virgin after textual content of the Symphonia. This may pro- the Trinity as she appears in R. The order in D is, vide a reason for the presence of the Ordo Virtuturn as Newman suggests, indicative of Hildegard's in R. Rather than forming an essential part of the theological view of the Virgin as, begetting of the Symphonia, the Ordo Virtuturn may have been son, an essential link between God and the Holy singled out as deserving of particular attention, Trinity. Secondly, Saint Disibod, the seventh- and hence inclusion in the manuscript, as the text century Irish missionary after whom the monas- of the drama was deemed significant enough to be tery at Mount Saint Disibode was named, is placed extended and given a musical setting. among the Apostles. In R, Saint Disibod is placed In relation to the question of whether the more appropriately among the confessors, along- Symphonia was intended as a collection of items or side Saint Rupert. Thirdly, in D Saint Ursula and as a Gesarntwerk, the most telling evidence can be the Eleven Thousand Virgins are placed among the gained from an examination of the variant struc- Holy Innocents. This prompts an analogy between tures of the Symphonia in D and R. The order in the Innocents slaughtered at the hands of Herod which the songs appear in the two manuscripts is and the eleven thousand virgins slaughtered by governed by a similar hierarchical structure in thugs in C~logne.~~Newman also points out the which the individual items are grouped together significance of the placement of the sequence 0 according to their subject matter, beginning with Ecclesia near the items devoted to ecclesia: the God and descending through the celestial ranks to virgin martyr portrayed in the sequence 'becomes Ecclesia. However, the songs within each liturgi- a prototype of the virgin church, which occupies cal category more often contrast than coincide. the next and final place in the ~ycle'.2~In R, the The structure of R is complicated by the separation items devoted to Saint Ursula are placed among the of the cycle into two parts.21 The order of the items devoted to virgins. liturgical categories included in D and the first As D was more than likely compiled under cycle of R (which incorporate fifty-seven and Hildegard's supervision, it seems possible that the fifty-eight songs respectively) is outlined in Table structural 'inconsistencies' found in the D 1. Symphonia were in fact purposeful to the creation of a song cycle that could be considered more than an assemblage of works. As Barbara Newman D R points out, the D Symphonia, on account of its God God structural peculiarities, gives the impression that vngin Mary Holy Spirifliity Hildegard 'ultimately meant the Symphonia as a Holy smlinity Virgin Mary Gesamtwerk rather than a mis~ellany'.~~ bk htimhand Fmpkts Pauiiand Propkts In contrast to D, one could easily use the title Apostles Apostles 'Musical Output' in place of the title 'Symphonia' Saint John the Evangelist Saint John the Evangelist SaintDisibod to describe the musical content of R. Admittedly, Miutyres carem the more normative hierarchy of the R Symphonia carem -Saint Disibod is not enough to dismiss it as an assemblage of Saim Rupert Saint Rupert Virgins virgins liturgical items; rather it is the appearance of two Wiw (Widows reserved for 2nd cycle) cycles in R that suggests that this manuscript Holy tmmm Sainturmla preserves not a single work but Hildegard's entire Saintursuk Holy hawus Eceksii Eceksii musical corpus. Kyie The division of the Symphonia in R into two cycles appears simply to separate Hildegard's shorter liturgical items (her antiphons and respon- Table 1: Order of liturgical categories in D and the first cycle of R sories) from her longer items (sequences, hymns, symphonies and other songs), while the structures There are three main structural differences be- of both cycles in R are based on the one hierarchy. tween the two hierarchies as outlined by Barbara However there is a more subtle difference between
16 Context 7 (Winter 1994) the two cycles in R. The second cycle includes five stages of development of a completed song cycle, items dedicated to four saints of local significance it is perhaps preferable to regard the cycles in D to religious communities in Trier (situated about and R as two different orientations of the one idea, lOOkm east of Bingen), where Hildegard gave a that of a cycle of songs arranged according to sermon in 1160.26 The second cycle in R is subject matter. outlined in Table 2. NOTES Ihe Holy Spii sequence,hymn Ihe Viio Mzay alleluia-verse, sequence, song, hymn There are also textual and neumatic discrepancies Saint Manhi hymn Saint Bonifaci anriphon between the two versions of the Symphonia, which give rise to questions associated with how the Symphonia came to be Saint Diiibod seq- Saint Mu9 responsory, sequence recorded on parchment. The most complete study of the Saint Maximum sequence textual discrepancies was undertaken by Barbara Newman Saint Rupen sequence as part of her edition of the Symphonia texts. See Barbara Saint Ursula sequence, hymn Newman, ed., trans., Saint Hildegard ofBingen: Symphonia Virgins symphony (Ithaca and London: Cornell University Press. 1988). For a Widows WP~O~Y study of the neumatic discrepancies see Catherine Jeffreys, The Symphonia of Hildegard of Bingen, diss. University of Melbourne, 1993. Table 2 Second cycle in R Newman, Saint Hildegard, p. 59. Peter van Poucke, Hildegard of Bingen: Symphonia The five items devoted to Saint Matthias, Saint harmoniaecaelestium revelationum (Peer: Alarniere, 1991). Bonifaci, Saint Eucharius and Saint Maximum are p.6. In separateletters to Hildegard both Odo of Soissonsand not included in the D Symphonia, possibly as these Volmar of Disibodenberg mention her musical activities. See J. Miscele Edwards, 'Women in Music toca. 1450', four saints were of little significance to the Women and Music, ed. Karin Pendle (Bloomington: Indiana Cistercians in Villiers, and, unlike Saint Disibod University Press, 1991). pp. 21-25. and Saint Rupert, of little local significance to the Newman, Saint Hildegard, p. 57. community at Rupertsberg. As this cycle includes van Poucke, Hildegard of Bingen, p. 6. an antiphon to Saint Bonifaci and a responsory to Ruth Lightbourne, 'The Ordo of Hildegard of Bingen', diss., University of Otago, 1990, p. 17. Saint Eucharis, the separation of the two cycles in For a disscussion of the items included in the medieval R can not be said to have eventuated from the Mass and Divine Office see John Harper, The Forms and perspective of genre alone. It seems likely that the Orders of WesternLiturgy (Oxford: Clarendon Press, 199 l), second cycle was created to separate the genres as pp. 67-125. well as accommodate the items devoted to these Lightbourne, 'Ordo', p. 18. lo Marianna Schrader and Adelgundis FiihrkUtter, Die four saints within the Symphonia. This may have Echtheit des Schriftums der Heiligen Hildegard von Bingen been done by compilers who wished to include all (Cologne and Granz: Boehlau-Verlag. 1956). p. 49. The of Hildegard's compositions for inspection by the monks in Villiers thanked Hildegard for the manuscript in a prelates overseeing her canonization, without dis- letter to her from the 1170s. A letter to Hildegard from her turbing the original subject matter of the Symphonia. secretary Guibenof Gembloux also refers to the manuscript. See van Poucke, Hildegard of Bingen, p. 6. The liberties taken with the form of the Symphonia The dates cited here are based on research by Schrader are what has given R appearance of an assemblage, and FiihrkUtter, Die Echtheit, pp. 154-170. that is a collection of songs ordered according to l2 See in particular Newman, Saint Hildegard, pp. 6-12. subject matter. l3 The Ordo Virtutum appears with neumes in a manu- As a consequence, it is possible to regard the script now housed in London (British Museum, Cod. Add.) which was copied from R in 1487. Symphonia as an anthology of Hildegard's musi- l4 van Poucke, Hildegard of Bingen, p. 10. The last three cal works; the older Symphonia survives as a fascicles of D, on which the Symphonia is preserved, are no Gesammerk, while the other version survives as lon er affixed to the rest of the manuscript. an assemblage, as the structure, while not random, 9 Other instances where a psalm cadence precedes an 1 suggests only a means of organisation and lacks antiphonand follows another typeof liturgical itemoccur on: fol. 154r (between repsonsory 0 clarissima muter and anti- the structural unity found in D. This allows for the phon 0 splendidissirnugemma), fol. 160v (between respon- varied manifestations of the Symphonia as well as sory 0 vos felices radices and antiphon 0 chohors milicie ~ the existence of compositions outside the song floris), fol. 161v (between responsory 0 lucidissima cycle. While many regard the Symphonia as two apostolorum and antiphon 0 speculum colwnbe), fol. 163r (between sequence 0 presul vere civitatis and antiphon 0
The Symphonia of Hildegard of Bingen undetermined. See P. Barth. M. I. Ritscher and J. Schmidt- 0 victoriosissimi), and fol. 161r (between responsory vos Gdrg. Hildegard von Bingen: Lieder (Salzburg: Otto Miiller itnitatores and antiphon 0 successores). Verlag. 1969). p. 3 18. l6 Both these antiphons, De patria eriam and Sed 22 Newman, Saint Hildegard, pp. 58-59. diabolis, appear on fol 168r. 23 Although no historical account identifying the slaugh- l7 WiUi Apel, Gregorian Chant (Bloomington: Indiana tered virgins exist, it is thought that Ursula and her compan- University Press, 1958). pp. 157-165. ions were British pilgrims who were slain by thugs in 451. l8 Apel, Gregorian Chant, p. 222. The number of virgins slain is said to actually have been l9 Lightbourne, 'Ordo', p. 98. eleven. The number eleven-thousand is purportedly the 20 Schrader and Fiihrk6tter. Die Echtheit, pp. 154-70. result of a printing error. 21 The structure of the D Symphonia is also complicated E. Day. 'Ursula, St', New Catholic Encyclopedia (New as it is missing three folios, probably due to the separation York: McGraw-Hill, 1967). p. 490. of the Symphonia from the rest of the manuscript. While the 24 Newman, Saint Hildegard, p. 58. majority of missing material can be accounted for, the 25 Newman, Saint Hildegard, p. 59. content of seventeen lines on one missing folio remains 26 Newman. Saint Hildegard, p. 58.
Context 7 (Winter 1994)