Elizabeth Aubrey

title.none: Rosenberg, Switten, and Le Vot, eds., Songs of the Troubadours (Aubrey)

identifier.other: baj9928.9807.012 98.07.12

identifier.issn: 1096-746X

description.statementofresponsibility: Elizabeth Aubrey, University of Iowa,

publisher.none: .

date.issued: 1998

identifier.citation: Rosenberg, Samuel, Margaret Switten, and Gerard Le Vot, eds. Songs of the Troubadours and Trouveres: An Anthology of Poems and Melodies. New York: Garland Publications, Inc., 1997. Pp. xi, 378. $85.00. ISBN: ISBN 0-815-31341-1.

type.none: Review

relation.ispartof: The Medieval Review

The Medieval Review 98.07.12

Rosenberg, Samuel, Margaret Switten, and Gerard Le Vot, eds. Songs of the Troubadours and Trouveres: An Anthology of Poems and Melodies. New York: Garland Publications, Inc., 1997. Pp. xi, 378. $85.00. ISBN: ISBN 0-815-31341-1.

Reviewed by:

Elizabeth Aubrey
University of Iowa

This is the first published comprehensive anthology of texts, English translations, and music of Old Occitan and Old French secular song, and it surely will find a significant niche in the libraries of scholars of these repertoires. One of the strengths of this collection is its presentation of the music and the poetry as equal partners in the complex of song. One of its ironies is that the essentially separate efforts of the three editors result in a somewhat eclectic product.

The anthology includes 63 songs by 23 troubadours (5 of them women), 34 with melodies, 5 of which are recorded on the accompanying compact disc; 26 anonymous songs in Old French (including chansons de toile, ballettes, and sottes chansons), 16 with their melodies; 2 Arthurian lais with music; 3 motets with their music, 2 of which are on the CD; and 50 attributed trouvere songs by 27 authors, 45 with their melodies, of which 5 are recorded on the CD. Each group of songs in the anthology is preceded by commentary on the composer (for attributed works) and genre, followed by specific remarks on each song. The volume also contains 13 photographic reproductions from 9 manuscripts. Each of the three editors contributed an essay, and these are followed by a guide to pronunciation of Old Occitan and Old French; at the end are a selected bibliography and discography, a general index, and indices of first lines and of genres represented in the anthology.

The most striking feature of this volume is its diversity: of contents (Occitan and French, monophonic and polyphonic, courtly and popular, anonymous and attributed), but also of viewpoint and editorial method. Samuel N. Rosenberg's "Introduction" spotlights the poetry, providing an overview of the repertoires, their histories, the social contexts in which they were created, and a wide-ranging discussion of the common themes, genres, and poetic styles. He addresses the manuscript and the performing traditions and invokes Paul Zumthor's concept of mouvance to explain "the innumerable instances of multiple redactions of a given song" and the impossibility of certitude regarding "authorial intentions" (4).

Gerard Le Vot's essay focuses on the music, first exploring manifestations of mouvance in the melodies, including "transformation" and "transposition." He adopts Friedrich Gennrich's four melodic types or "processes" (litany, sequence, rondel, and hymn), as well as terminology from Dante's De vulgari eloquentia (oda continua, pedes, versus, cauda, frons) to categorize musical structure. He briefly presents some of the questions about performance practice, including use of instruments, application of musica ficta, and rhythm (modal, mensural, declamatory; he does not mention the isosyllabic theory), but he does not take a strong stand on any of the prevailing theories that address these problems. He concludes with a short excursus on the medieval voice, opining that it "was no doubt of a quality that our modern taste would consider poor and untutored, with heavy breathing, 'unmusical' pronunciations, unrefined timbers," but that "the medieval singer possessed the ability to give the lyric poem its true voice." (12)

Margaret Switten expresses her viewpoint succinctly at the outset of her essay, "Music and Words: Methodologies and Sample Analyses": "Song in the medium of the voice (and as received by the ear) does not separate music and words; it holds them together in a single artistic expression." (14) In an "attempt to seize the song in its very essence as a creation of the voice (14), she discusses two broad approaches to analysis, one emphasizing "structural" matters of poetic meter and melodic form, for which she also employs Dante's vocabulary (while pointing out some of its limitations), the other dealing with "rhetorical" elements of the text and the music which give a song its "meaning and expressiveness." (16)

She describes certain analytical tools, beginning with an economical and clear explanation of the principles of versification. Aspects of the music that can be analyzed include third chains and melodic contours, but especially "scales and modes," specifically the eight modes of liturgical plainchant, which Switten cursorily describes. She emphasizes the concepts of tonal hierarchy--final, subfinal, and reciting tone (tenor or, anachronistically, "dominant")--and range (authentic or plagal), but she barely mentions what to the ear is the most salient feature of a particular scale, the unique succession of whole tones and half tones that distinguishes it from another. Interval patterns, perhaps more than any other characteristic, underlie the very meaning of the term "mode," especially in a music whose orality she and Le Vot both emphasize. Other aspects of her treatment of mode are somewhat misleading, such as the rigid distinction she makes between plagal and authentic ranges (which are not so clear-cut in either medieval theory or practice), her mention of four modes in addition to the medieval eight (which did not receive theoretical validation until the sixteenth century), and the idea of modal transposition (another anachronistic concept, which Le Vot also invokes).

Although application of the modal system has a long tradition in the study of secular monophony, there are numerous instances where the system does not work at all, as Switten herself acknowledges (indeed, even medieval theorists often had trouble assigning liturgical antiphons to modal categories). I have pointed out elsewhere (The Music of the Troubadours [Bloomington: Indiana University Press, 1996], 32) that secular melodies developed within a different sociological environment from plainchant, for quite a different purpose, with an entirely different structural grammar, and were transmitted by a significantly different process over a much shorter period of time. Applying either the vocabulary or the underlying theories of the ecclesiastical modes to lyric song requires analytical gymnastics--not to mention the frequent irreconcilable exceptions--that leave one wondering whether any light has been shed at all. One example of ambiguity will suffice: A melody by Peire Vidal (112) with a range from low C to the D a ninth above, whose first four verses begin with three A's and whose next two verses begin with three high D's, whose phrases end on A, G, A, G, B, F, G, C, and F, with B- naturals throughout except in the last verse, defies modal categorization. According to the criteria that Switten establishes in her essay, the range and repeated D's suggest the D authentic mode, but in her commentary she asserts without explanation that "F tonality is suggested throughout," noting only the B-flat indication in the last line.

Switten's essay concludes with detailed analyses of two songs in the anthology, one by Marcabru and one by the Chastelain de Coucy. The anthology itself does not have space for such extensive examination of every song, but these are careful and thorough, and they can serve as guides for how to apply some of the methodologies that she outlines.

Even read attentively, the three opening essays do not fully prepare one for the diversity of approach that manifests itself in the following pages. Each repertoire, each composer or genre, each song, each text, each translation, each melody is presented as a more or less self-contained entity with its own critical orientation. Only gradually does one begin to realize what vast differences in editorial method, philosophy of translation, interpretation of musical rhythm, pitch, and style, and of the relationship between poem and melody are embodied in the presentation of the songs and in the commentaries on them. For a given song, the reader is offered no direction in applying, synthesizing, or assessing the relative merits of the different analytical possibilities.

The Introduction (5-6) provides a road map to the various roles played by the editors: Rosenberg edited and translated the French texts. Switten "assembled" the Occitan texts "from standard printed editions"and provided new translations. Le Vot transcribed the music from the manuscripts, "with final copy of the troubadour transcriptions prepared by Switten." Switten provided commentary on the Occitan texts and music "taking into account the musicological observations of Gerard Le Vot." Rosenberg provided commentary on the French texts and included "Le Vot's musicological remarks."

While such an assignment of tasks makes sensible use of each of the three scholars' expertise, the ensuing methodologies yield substantively different results. Switten's presentation of the Occitan texts, for instance, relies on a wide variety of previous editors, some better than others; she occasionally corrects or alters a prior edition, always justifying her reasons. Rosenberg's French text editions, on the other hand, are entirely his own, although most of them are not newly edited here but are taken from earlier publications, mainly Chanter m'estuet: Songs of the Trouveres, edited by Rosenberg and Hans Tischler (Bloomington: Indiana University Press, 1981) and Chansons des trouveres, edited by Rosenberg, Tischler, and Marie-Genevieve Grossel, Lettres gothiques (Paris: Librairie Generale Francaise, 1995).

Switten's elegant and readable translations are quite faithful to the Occitan texts, retaining verse and often word placement and even enjambment where it occurs; to maintain a sense of the original verse structure, she begins each verse of the translation with an initial capital. Her rendering of the first stanza of a song by Raimbaut d'Aurenga, for example, keeps enjambment and word order in place without making the English sound artificial (72): "Ar resplan la flors enversa/Pels trencans rancx e pels tertres,/Cals flors? Neus, gels e conglapis/Que cotz e destrenh e trenca,/Don vey morz quils, critz, brays, siscles/En fuelhs, en rams e en giscles;/Mas mi ten vert e jauzen Joys/Er quan vey secx los dolens croys" ("Now shines the flower reversed/Amidst the sharp cliffs and the hills,/What flower?/ Snow, ice and frost/Which burns and distresses and cuts,/So that I see dead chirpings, cries, calls, warblings/In the leaves, in the branches and in the twigs;/But Joy keeps me green and joyful/Now when I see dried up the sorrowing churl"). Rosenberg, on the other hand, allows himself a wider interpretive berth; his translations are less literal and are loosely laid out as verse without initial capitals so that they flow more like prose. His translation of the first stanza of a pastourelle by Thibaut de Champagne, for example, freely moves words about from verse to verse, and adds an emphasis to the last verse that is not necessarily implied in the French (314): "J'aloie l'autrier errant/Sanz compaignon/Seur mon palefroi pensant/A fere une chancon,/Quant j'oi, ne sai conment,/Lez un buisson/La voiz du plus bel enfant/C'onques veist nus hon;/Et n'estoit pas enfes, si/N'eust quinze anz et demi,/N'onques nule riens ne vi/De si gente facon" ("As I was out riding on my palfrey/all alone/the other day, thinking/of composing a song,/I heard near a hedge/(I can't say how)/the voice of the loveliest child/a man could ever have seen,/though she was not really a child/and must have been fifteen and a half;/never before had I seen a creature/of so pleasing a stamp").

Most of the musical transcriptions use the customary stemless note heads and, "for simplicity's sake" (7), treble clefs, which unfortunately places most of the melodies an octave higher than the manuscripts notate them; an "8" below the clefs would have indicated the correct pitch. The melodies are laid out verse by verse, sometimes with two verses side-by-side separated by a break in the staff, giving the appearance of two columns which nonetheless must be read from left to right, an unusual arrangement that verse numbers would have clarified.

The two editors diverge on the interpretation of certain important musical matters. To indicate plicas, for example, Switten gives the principal note followed by a smaller note head for the secondary note, interpreting the latter as a discrete pitch, whereas Le Vot signals only the existence of a plica by means of a comma-shaped tail on the principal note, "in order to underline [its] ornamental role" (8). The two editors suggest musica ficta differently as well, Switten with a sharp or flat above the system, Le Vot with a bracketed sharp or flat either on or above the system.

Switten's reworkings of Le Vot's transcriptions of the Old Occitan melodies are entirely non-mensural, in keeping with the non-rhythmicized neumes in the troubadour manuscripts. For the Old French songs, Le Vot occasionally offers mensural interpretations of the quasi-mensural notation that some manuscripts contain. Almost all of these measured transcriptions reproduce previous modal interpretations by Hans Tischler (in the Rosenberg/Tischler volume of 1981), which in their turn were based upon earlier editions (in particular those of Jean-Baptiste Beck in his edition of Paris fr. 846). This is unfortunate, since many of those interpretations make too free with the notated longs and breves in order to fit the melody into a rhythmic mode; Le Vot offers no explanation for such things as rendering a long as a breve or vice versa, or a breve-long neume as two semibreves.

His transcriptions of the three motets give no indication of original note forms or ligatures. Here again his treatment of plicas is unorthodox. In mensural music, these customarily are transcribed as a large note followed by a separate note either with a slash through its stem or smaller in size than the principal note, but Le Vot represents the symbol as a single small note with a slash through the stem to indicate both the principal note and the plica; this not only makes the entire complex look like an ornament, but also, as in his monophonic transcriptions, dodges the sometimes difficult perplexity of assigning a discrete pitch to the latter. He presents one of the motets with the three voices arranged as they are in the manuscripts, as separate parts rather than in score, which is, he says, "an arrangement which has a certain relevance to the synchronization of the voices. In the 13th century, such synchronization was in the main accomplished aurally and gesturally (by tactus) rather than visually (by reading), which of course entailed a degree of rhythmic imprecision; this was naturally increased by the musicians' ignorance of what we today call measure." (234) This latter assertion is extreme, but one cannot argue with the likely profitable results o be realized by performers who follow his suggestion to "give up part of their reliance on their eyes to achieve synchronization and give more attention to their ears."

The commentaries on the texts do not constitute a critical apparatus (for which the reader is referred to other published editions). They list the manuscripts sources for each song, for which Switten includes folio numbers in the Occitan manuscripts; modern editions are cited, and Rosenberg also lists specific studies of the French songs. Switten's commentaries on the troubadour texts are in general more extensive and detailed than are Rosenberg's on the French texts, and they include explanations of emendations, discussions of versification (often including a graph of rhyme scheme and syllable count), summaries of a composer's poetic style, exploration of the meaning of the text, and notes on specific words, phrases, and references in the poem. Except for brief introductions to genres and composers, most of Rosenberg's commentaries are taken from his 1981 publication, including discussions of some versification features, notes to specific verses, and a listing of salient dialectal features. Hints of Rosenberg's view of the relationship between music and words occasionally appear, such as his comment that the music of a monophonic song "has the subordinate function of communicating a verbal text," while "motets are primarily musical compositions, in which words provide a means for the expression of melody" (234).

Switten's commentaries on the troubadour melodies are incorporated into her discussion of the poems. These analyses consist mainly of observations on musical structure, chains of thirds and fourths, and the "modal" characteristics that she finds in the music. Notwithstanding the reservations I expressed above about the value of using the church modes to explain aspects of this music, Switten's analyses sometimes do provide canny insights into how some of the melodies seem to work. Frequent subjective characterizations, though, leave the reader unenlightened about anything other the author's own response to the melodies: "No flights of melodic--or emotional- -imagination characterize the song" (60); "The R version...offers a melodic vivacity that G does not have" (118); "The learned sound-architecture" of a melody "has the characteristics of a dance tune and remains joyously focused on c" (154); "The tonality is somewhat uncertain and the ending all the more poignant" (170).

Most of Le Vot's observations on the French melodies consist of brief statements about structure, mode, and cadential goals. Like Switten, he engages in subjective utterances, but declarations such as the following are veritable head- scratchers: "The melody, like most melodies of Arthurian lais, has a four-line scheme of three phrases (AABC), derived from the liturgical hymn pattern" (230) [the patterns may be similar, but there is no reason to believe that one was derived from the other]; "This is a kind of transposition of pitches that should in fact be regarded as a 'translation' of notes (in the sense meant by some of the music theorists of the Middle Ages)" (242); "In the opening of lines 1 and 3, we find a transposition to b of a formula in e, a formula often used in pieces meant for dancing" (306); "In its cleanly sculpted lines, with contrasting syllabism, ornamentation, and agogic rises..., this melody recalls the music of the troubadours" (277) [one is baffled by how to identify the agogic in a melody without rhythmic indications, let alone the idea of an agogic "rise"]; "The focused on c, the tone of instrument-playing jongleurs" (324); and in an explanation of the editorial addition of musica ficta, "Often in French folkloric traditions, a single composition may be played partly in a major tone and then continued in a minor tone, which is a way of re-energizing the piece or giving special emphasis to a spirited text" (347). All of these statement beg for amplification.

While Le Vot restricts most of his measured transcriptions to melodies that are given in a manuscript with mensural notation, he makes numerous allusions to what he perceives as a "dancing" quality in a melody, including many for which he does not offer the mensural transcription that would be necessary for such a performance; he sometimes makes other explicit performance suggestions as well, for none of which he provides evidential support: A melody "readily lends itself to interpretation in the ternary meter...that some modern musicologists have tended to ascribe to it. No doubt intended for dancing, this son d'amors may well be accompanied in performance by such instruments as the fiddle, flageolet, or tambourine" (195); "These ouverts and clos, like the syllabic character of the music, make it clear that the song is meant for dancing" (272); "The melody, in g with a major third in the first six lines, calls for a dance rhythm." (347)

The compact disc, which features baritone Peter Becker and fiddler Robert Eisenstein, is an estimable bonus to the anthology. The selection, newly recorded from the texts and melodies printed in the anthology, represents several generations and genres, and varying theories of musical interpretation. Included are performances of five troubadour songs, two motets, five trouvere songs, an instrumental version of an anonymous Occitan dance song ("A l'entrada del tens clar"), and two thirteenth-century instrumental estampies. The musicianship is impeccable: Becker's singing is pleasing and expressive, and Eisenstein's handling of the fiddle (and lute on one song) tasteful. Becker sings four of the monophonic songs (which have no indication of the use of instruments in the manuscripts) without accompaniment, a fashion that is enjoying increasing popularity among singers of these repertoires, and one that effectively brings out the rhetoric of the text. On the rest of the monophonic songs Eisenstein improvises a gentle accompaniment, mostly consisting of two sustained tones that occasionally move to different pitches following the line of the melody. I have argued against the use of drones to accompany medieval monophony (in my 1996 book, cited above), and I did not find myself persuaded by these performances to change my view. There is no compelling medieval evidence in favor of the practice, and the constant presence of non-melodic tones gives every pitch of the melody a color--major, minor, consonant, dissonant--and thus establishes a hierarchy of tones (which would be completely different if the drone were on a different pitch), and a musical shape that ignores other structural constituents of the melody (such as, for instance, motivic development). Even more than a single drone, two drones create frequent major or minor triads that sound less like composed polyphony of the period than like modern folk songs--appealing, perhaps, but not medieval.

Becker sings a few of the songs with a measured (but flexible) rhythm, but for most of them he uses the "declamatory" style proposed over twenty-five years ago by Hendrik van der Werf, without measurable rhythms assigned to individual notes. This allows for a very expressive rendering of the poem. Now that it has become de rigueur for singers to perform most or all of the stanzas of a lyric song, one can begin to see how a melody acquires a shape of its own as it is reiterated over and over to changing words. In every piece sung with "declamatory" rhythm here, certain pitches in the melody are always emphasized, others always treated as quick ornaments--creating a kind of macro-rhythm that is consistent in all stanzas.

An anthology like this is generally intended for classroom use, presumably undergraduate. The presentation of the materials lends itself well to such use, with full texts and parallel translations, music laid out with the first stanzas of the poems, and little clutter of detailed critical apparatus. Undergraduates will need supplementary explanations of many terms and concepts sprinkled throughout that are insufficiently explained or contextualized, as well as guidance to thread their way through the varying editorial and analytical procedures. On the whole this is a book that students and scholars of medieval secular song will want to own, particularly because it collects in one place such a broad and diverse representation of the various repertoires, and because its contents are accurate and accessible. The sum of its rich array of divergent approaches is a broader and more nuanced understanding of how scholars might study these songs, although not necessarily a deeper understanding of the songs themselves. It nonetheless promises to fill an important pedagogical function by introducing to a wider audience the full spectrum of the songs of the troubadours and the trouveres.