campionhorner

Mosaic, Vol. 27, 1994

Negotiating traditions of English song:
performance, text, history.

by Bruce Horner

As a musical entity, song seems beyond the ken of most literary scholars. It appears evanescent, intangible by virtue of the differing complex conditions of its performance. Thus despite the fact that song represents perhaps the most pervasive genre of literature across both time and space, literary scholars hav tended either to ignore song altogether or to attend to song in ways that evade its existence in performance.

Both practices depend on locating literature within the boundaries of verbal texts and separate from the domain of performance. Yet if we define the "work" of literature as the interaction between the text and the conditions of its production and reception, then song, by virtue of its existence in performance, illustrates in exemplary fashion the dynamics of all literary work. To study th interplay between the song text (score and lyrics), performers, listeners and the particular physical, social and historical conditions of song performance can make manifest how meaning is produced through the interaction between any text, responses to it and the social and historical contexts of those responses it enables us to see how not just song but all artistic work resides in the domain of performance.

Such an approach involves the erasure of the distinction between the "cultural" and the "material" by showing how the cultural "works" through the material. This type of song criticism would approach songs not as reified texts but as sites of social practice, treating them as performative activities responding t and acting on the cultural process. In what follows, I want to illustrate how one might engage in such a "cultural materialist" criticism by examining two types of Renaissance English song--madrigals and lute songs. These songs are especially pertinent to the question of performance for two reasons. First, critical response to these songs reveals the ways in which performance of the songs is evaded by some literary and music scholars. Second, the songs themselves represent in their tendencies two historically conflicting approache to "performance." "To perform" has meant "to complete, finish, perfect (an action, [or]...work)," though this meaning is identified as now being "obsolete (OED). Today the term usually means: "To carry out...[or] execute (that which i commanded....)" especially in the sense of doing "what one has to do; to...do one's part" (OED). It is this latter meaning which we more commonly think of when describing the performance of art. The OED's definition of the act of artistic performance as "To do...or execute formally or solemnly (...a piece of music, play, etc.)" continues the sense of performance as a matter of doing "what is commanded."

The two senses of performance are distinguished particularly by the relationships each posits among those parties involved. If performance entails completing the incomplete, all participants in the production of the "work" wil share creative responsibility: in the case of songs, composers, lyricists, singers, instrumentalists--and even publishers, audience members and patrons. The now more common sense of performance, however, places most participants in role of submission to one in command: they must do what is "commanded," and do so "formally." Criticism adopting this sense of performance imposes an "Author" on the work and, as Roland Barthes observes, "To give a text an Author is to impose a limit on that text...to close the writing. Such a conception suits criticism very well, the latter then allotting itself the important task of discovering the Author (or its hypostases: society, history, psyche, liberty) beneath the work: when the Author has been found, the text is 'explained'--victory to the critic" (147). In the case of song criticism, those critics who do not simply dismiss song as too evanescent to merit attention hav evaded the location of song in the activity of performance because of the threa this sense poses to the Author (or its hypostases) by its emphasis on the participation of numerous parties--singers and instrumentalists, listeners, interpreters--in the production of song.

In what follows, my own aim is to call attention to the importance of approaching any songs in the first sense of performance: as texts whose "work"--even today--is completed in their performance. Of course, it is a widel accepted truism that all genres of music "live" only in performance. Every performance of every song, however, raises the question of whether the performance should constitute a completion of what is incomplete or whether performance should be regarded as a ritual re-enactment of work taken as alread completed. This same issue, furthermore, applies to criticism not only of song but of all art; for the sense of performance as an act of completion challenges the hierarchy between "authors" and "readers," "artists" and consumers that is posited by the view of performance as re-enactment, a hierarchy that restricts some of these parties from participating in the creation and re-creation of works of art.

Students of English literature are most likely to be familiar with late Renaissance English song because of the songwriters' ostensible participation i a "golden age" of English culture and because of the convenient collections, editions and bibliographies which scholars have compiled. Paradoxically, however, much of the scholarship which has given us access to Renaissance English songs has also placed those songs in traditions in which they cease to exist as songs--at least if we define song as a genre that exists in performance. Most commonly, literary scholarship has simply omitted the songs' musical settings from consideration. Scholars have mined the period's songbooks and manuscripts for their lyrics and published these as poems without musical setting or in their discussions have placed the songs in a tradition of poetic miscellany publication. The songbooks, or at least the lutesong books, it is argued, carried on the tradition of the 16th-century English poetry miscellanie in publishing anonymous poetry (Lyrics; Seventeenth-Century; Doughtie, Introduction 10-30; Pomeroy 105-06, 119). The songs, in this view, are really poems in musical disguise. R. W. Short's advice that "whoever aims at justly appreciating [Thomas Campion's] poetry had best forget his music" provides an extreme example of this perspective (1004).

Other scholars do acknowledge the musical settings of the songs, but mainly wit a view to placing such works in traditions of poetic and/or musical stylistic experiment. Louise Schleiner and Winifred Maynard, for example, have examined the effect of song forms on lyric forms in the poetry of 16th- and 17th-century England. In English Renaissance Song, Edward Doughtie charts the mutual influence of the period's music and poetry on each other. Elise Bickford Jorgen has traced the influence of Renaissance French and Italian theories and practices of song setting on the practices of late Renaissance English song composers, and Philip Brett has identified the esthetic of William Byrd's songs with a native English tradition of song ("Word-Setting" 50-54). Insofar as social relations inhere in poetic and musical styles, such scholarship can of course contribute to an understanding of the performative character of the songs. Yet to the extent that the tendency is to focus on the development of compositional techniques (in both music and poetry) in terms divorced from the social conditions of song performance, such scholarship runs the risk of "closing" the writing through its discovery of an Author beneath the song "work."

All these critics are necessarily engaged in what Raymond Williams calls the "radically selective" process of tradition-making. By selecting "certain meanings and practices...for emphasis" to the neglect or exclusion of others, "the tradition" is established "which is intended to connect with and ratify th present" (115-16). By placing the songs into traditions of poetic miscellany publication, or poetic experiments in "lyric," or experiments in styles of setting lyrics, scholars remove the songs from the activating process of performance, thus risking the closing of the songs into "silence" while grantin "victory to the critic": the songs become poems with accompanying music or are understood as acoustic artifacts to be appreciated for representing a particula style of setting lyrics. Such closure positions the contemporary critic at the top of a hierarchy of authority over the songs by ignoring the "basically collaborative" nature of most Renaissance literature that, according to Stephen Orgel, has been demonstrated by scientific bibliography of Renaissance texts ("What" 6). Though Orgel writes primarily of dramatic literature, the well-know Renaissance practices of composing verse to pre-existing music and setting pre-existing poems and song lyrics to new music further illustrate Orgel's argument.

Critics of the songs, however, have used evidence of the many versions of a song's lyrics and musical settings primarily to justify treating song lyrics an musical settings as separate but fixed elements rather than using such evidence to reject the privileging of any one version of the lyrics and/or setting as "authoritative." In adopting this position, scholars participate in what Janet Wolff, in a critique of non-sociological approaches to music, calls the "ideology of autonomous art" and in what Williams, in a discussion of the rise of "aesthetics," notes as its evasion of "all problems of the multiplicities of intention and performance" (150). Most obviously, literary scholars evade connections between the songs and the immediate social, political and material conditions of their production both as bibliographical entities--song manuscripts and songbooks--and as acoustic performances. Musicologists, while acknowledging the musical settings of the songs, risk what Richard Middleton, i a critique of traditional musicology, calls "notational centricity," that is, a focus on the score which, he argues, encourages "reification: the score comes t be seen as 'the music', or perhaps the music in an ideal form....downgrad[ing] the vagaries of performance...and the influence of performance context; practic is frozen into symbol" (105). Whether the songs are "read" as poems or as authorized scores, however, the effect is equally to transform them from "incomplete" to "complete" through deletion of performance practices and contexts. The specialized knowledge of traditional musicology is brought to bea on the songs to authorize, and so limit, them, rendering actual musical performances of the songs into ritualized representations rather than creations

It may be objected that the "historical performance" movement in musicology offers a salutary antidote to this tendency--indeed, that it treats songs explicitly as acoustic, i.e., necessarily performed, cultural, historical phenomena. The movement has yielded valuable findings about different performance practices (and the recovery of much "early" music). Unfortunately, much of its research has been directed to establishing "normative" or "authentic" renderings of that music in performance in ways that ignore the participation of the modern critic, "performer" and listener in any modern performance of past music. As Joseph Kerman complains, many of those in the historical performance movement (like conservative interpreters of the U.S. Constitution, we may note) have attempted to use research either into the composer's intentions, so far as they can be deduced, or the performance practices common at the time the piece was composed, primarily to codify and prescribe "interpretation"--i.e., the playing or performance of a musical work (Contemplating Music 191-93).

We can see this aim expressed in David Wulstan's announcement that he will use historical research to clarify the "points of interpretation which confront the performer [of Tudor vocal music] and his editor--if they are to achieve a faithful representation of the composer's intentions" (156-57). Edward Huws Jones expresses a similar aim when he claims that "the purpose of [his] study i to explore how English song was performed during the period 1610-70 in order better to perform the music today," for, he assures us, "the music of any perio can only carry its full meaning if it sounds as the composer originally heard it" (2). The problem here, as Peter Rabinowitz has remarked, is that modern listeners of early music "are always hearing it in terms of esthetic categories that are foreign to the world in which that object was originally created" (171 see also Leppard 74-75). Scholars like Wulstan and Jones, however, deny the presence of the "modern"; they attempt to remove the modern listener, the performer and the "modern" itself from performance, either acoustic or interpretive. The ideal performance in these arguments is regarded as a set of acoustic phenomena ruled over by the critic, with the performers serving as passive, uninvolved media for the production of those phenomena, their role reduced to doing "that which is commanded," and doing so, of course, "formally or solemnly," as a "ceremony, or rite."

What characterizes the critical approaches I have described is the distinction they maintain between "text" and "performance" in the "work of art." The "work of an art" is seen not as an activity but as an entity residing in a reconstructed (printed, edited) "text" which is granted a status superior to an but an ideal performance of it. Focusing on the text thus has the effect of reifying the working of a song into a "work" found either in idealized space--a reconstructed "text"--or in idealized time--a purely conceptual moment of performance. The "works" are thereby placed into traditions composed either of Great Artists and their Artifacts or Great Moments in Art. To counter such tendencies in literary studies, Stephen Greenblatt recommends that we "look les at the presumed center of the literary domain [the text] than at its borders, t try to track what can only be glimpsed, as it were, at the margins of the text" (4). Instead of a tradition of great artists producing great art works, Greenblatt asks that we construct a tradition of cultural negotiation among artists, art "works" and audiences. Similarly, musicologist Richard Taruskin demands that we reject the investment of final authority for musical performances in texts, and he calls for the recovery of a performance tradition conceived of as "cumulative, multiply authored, open, accommodating, above all messy, and therefore human" (318, 323).

Of course, in thus attempting to negotiate a different understanding of the wor of art, both Greenblatt and Taruskin, like all critics, are themselves necessarily engaged in what Williams calls "a major part of all contemporary cultural activity," namely "the struggle for and against selective traditions" (117). They select certain meanings and practices to the inevitable neglect or exclusion of others. Such struggle is possible because what other traditions exclude is recoverable through historical research (Williams 116). Paradoxically, such recovery, at least in the case of Renaissance culture, ofte involves a reading of texts but in ways which redefine what counts as central and marginal in those texts.

In an analysis of artistic conventions, Williams reminds us that "convention" carries with it the sense of "agreement," a coming together: a convention "is a established relationship, or ground of a relationship, through which a specific shared practice--the making of actual works--can be realized" (173). Convention of song scores and lyrics work as gestures toward particular relationships, the realization of which depends on whether performers and listeners agree to the meanings and the relationships between the songwriter(s), the musical performer and the listeners that the conventions attempt to establish. Through their material construction and artistic conventions, song texts set up specific relationships between composers, lyricists, performers (singers and instrumentalists), consumers (listeners and purchasers of music) and larger social formations. By attending to such marginal features of late Renaissance English songbooks as the varying conventions of format, notational practices, lyrics and word-setting techniques, we may recover traces of attempts to negotiate various kinds of social relationships, situations, occasions, means and meanings for the songs among songwriters, singers, publishers, purchasers and listeners. These traces suggest a range of attitudes toward the interests o the parties with whom lyricists, composers and publishers negotiated, from highly accommodating to prescriptive.

They represent responses to particular orientations of Renaissance performers and listeners to song texts and scores. In performing songs, Renaissance performers evidently felt free to use whatever instruments and voices they pleased or happened to have available; they improvised descants, added melodic embellishments and even sang songs to meaningless syllables, ignoring the printed lyrics (see Duckles; Ferand; Horsley; Morley 206; Greer, "Manuscript Additions" 524). These habits reveal the appropriative attitude of performers toward the "text" of the songs, seeing it as so much malleable raw material for them to manipulate as they saw fit. As Thomas Campion complains, "Yet doe wee daily observe, that when any shall sing a Treble to an Instrument, the standers by will be offring at an inward part out of their owne nature; and, true or false, out it must, though to the perverting of the whole harmonic" ("To the Reader," Two Bookes of Ayres [c. 1613-14]; rpt. Works 55).

By way of accommodating these kinds of interests and attitudes (and so, presumably, boosting sales), the songbooks frequently present their contents in ways which allow for or encourage multiple performance possibilities. Such accommodation is particularly noteworthy in light of the control that some composers were able to exercise over the publication of their songs. For example, Brett finds Byrd to have exercised close supervision of the printing o his vocal compositions ("Text" 107-08). Yet Byrd published his songs, originall composed for solo voice and viol consort accompaniment, with words to all the parts so that they could be performed as part (rather than solo) songs. About a third of the so-called lutesong books provide, in table-book layout, part-song versions of songs most of which were originally conceived as solo songs. Table-book layout, as the term implies, makes it possible for performers seated around a table to read their parts from the open book, each part being printed at a different angle (see Greer, "Part-Songs," and, for examples, Dowland's Ayres). Those books published in sets as "part books," each containing the part for one of the voices, regularly describe their contents as "apt for viols and voices," implying the possibility of employing not just a mixture of the two in performance but of performing the songs a capella or with instruments alone (Westrup 25-27; see also Edwards 119-23). The composer Tobias Hume merely took to an extreme a common claim to the versatility of songs' performance potential when he boasted that his own "Poeticall Musicke" (1607) is "so contriued, that it may be plaied 8. seuerall waies upon sundry Instruments with ranch [sic] facilitie" (Lyrics 281).

Such accommodations show that song composers conceived of their song texts as far more "fluid," to borrow a term from Orgel, than modern critics are used to conceiving them. Orgel, noting differences between a Shakespearean play script, acting text, performances and printed versions, argues that "the realization of a Shakespeare text [in performance]...historically speaking, involves a considerable departure from the text....The text is the basis of the performance, but the performance is an independent entity" ("Authentic" 7). Similarly, in his study of late Renaissance vocal improvisation, Ernest Ferand finds that in music of the Renaissance and Baroque, "the 'plan'...i.e., the written composition, very often does not present much more than a general outline, the details of which...were left to the imagination of the individual performer or group of performers" (131). The common practice of composing new lyrics for pre-existing music and new settings for pre-existing lyrics--themselves representing new realizations of prior songs--illustrates th "fluid" conception of song texts obtaining during the period. Such a conception is further demonstrated by what Ian Spink has observed as the existence of widely varying versions of the same songs in manuscript and printed sources of the period, some variants produced under the composer's supervision (119-20). The fluidity of the term "song" is confirmed by what Warwick Edwards observes a the Renaissance use of the word to denote not just compositions for vocal performance, with and without verbal texts, but music "lessons" as well as poem with no musical setting (117).

The claims and accommodations found in the published songbooks generally suggest, in themselves, a relationship between composer and performer in which responsibility for such crucial elements of the song's content as instrumentation, lyrics and the number of voices is left to the discretion of the performers. A relationship in which creative responsibility is shared by songwriters and performers is also encouraged by those songs published specifically as part songs in part books, for such books make creation of the song as a whole dependent upon the participation of the performers. In the absence of any full-score version, each song is, as it were, dispersed into three or more parts, each printed in a separate book. This dispersal makes participation of a group necessary for the construction of the songs, and means in effect that the song is as much a product of the singers as it is of the composer or poet.

In this way, the part-book songs embody in their form the relations of a group rather than relations between individuals. Byrd's gentle reminder to performers in the preface to his 1611 book that "the well expressing of [songs]...is the life of our labours" (English Madrigal School 16: vii) suggests that part-song performers were expected through performance to complete the incomplete, and that "ownership" of the songs devolved as much on performers as on the songwriter. A number of the dedications of the part books illustrate this sharing of song authorship among patron-performers. George Kirbye reminds his patron that the songs belong to the patron by virtue of their having been "bredd," as he puts it, in the patron's home in performances by members of the patron's household (The First Set of English Madrigalls [1597]; English Madrigalists 24: xv). Henry Lichfield similarly reminds his patron that the contents of his First Set of Madrigals, 1613, are the patron's, being originall "presented by the Instruments and voyces of [the patron's] owne familie" (English Madrigalists 17: xi).

Further, insofar as the dispersal of a song into parts in part books makes a gathering of people necessary for the performance of the song, each song can be said to occasion such gatherings. Thus the song becomes a means to per-forming, in the sense of furnishing (par-fournir), and so comes to mean participatory recreation. Both Morley in the opening to his Introduction and Nicholas Yonge i the preface to his Musica Transalpina (1588) describe such gatherings. These descriptions suggest that such gatherings served especially as occasions for th display not only of the guests' musical ability as singers but also of their musical knowledge and the wealth and leisure necessary to acquire such knowledg and ability. Yonge describes the gatherings as including "Gentlemen and Merchants of good accompt," some if not all of whom spoke Italian as well as English and all of whom were interested in acquiring the latest music books "ou of Italy and other places." At an imaginary banquet that Morley describes in th opening to his Introduction, all but one of the guests can not only sing their parts but also dispute learnedly on music (9). (The one guest's ignorance of music leads the others to question his upbringing, shaming him into seeking instruction in music.) In participating in such singing and disputations, the guests not only display their musicality but also provide evidence that they have the leisure and wealth sufficient to have been trained to sight sing and t understand music theory. Equally, in the case of Yonge's gentlemen and merchants, they evidence that they had the means to purchase the latest songbooks which, as David Price notes, were out of the reach of the majority (184).

Many of the songs in the part books repay the expense of wealth and leisure by presenting opportunities for the testing and display of different kinds of musical ability and for appreciation of tle songs' counterpoint. Differences between the songs can be understood partly in terms of the kinds of musical abilities they test and reward. Madrigals offer music connoisseurs occasions fo the production, evaluation and appreciation of specific techniques of text setting. A short lyric is set with frequent repetition and overlapping of verba and melodic phrases, the melodic lines often highlighting the meanings of particular words--such as rising in pitch to the singing of the word "rising," pausing after the singing of "breath." Ballets are distinguished by their stron dance rhythms and rapid "fa la" sections that challenge the vocal agility of th singers. In his Introduction, Morley describes the madrigal as being, "next unt the Motet, the most artificial and, to men of understanding, most delightful" (294) and in contrast to ballets (e.g., Morley's own "Now Is the Month of Maying") which are among the "lightest" of vocal compositions (295).

In all cases, because no full score was provided and bar lines were not yet in common use, singers would face the challenge of both synchronizing their performance of the various parts and judging the accuracy of their own sight-singing, the composer's skill, and the printer's often questionable fidelity to that skill. Thomas Weelkes's madrigal "O Care" (and "Hence Care"), by setting ballet lyrics, complete with characteristic "fa la's," in madrigalesque fashion, simultaneously defines the kind of artificiality and understanding to which the madrigal appeals and distinguishes that appeal from that of ballets. Like the lyrics of many ballets, those in "O Care" call for music to dispel care and include the nonsense "fa la" syllables:

O CARE, thou wilt despatch me, If music do not match thee. Fa la. So deadly dost thou sting me, Mirth only help can bring me. Fa la.

Hence, Care, thou art too cruel, Come, music, sick man's jewel. Fa la. His force had well nigh slain me, But thou must now sustain me. Fa la.

(Madrigals, 1600; English Madrigal Verse 291; full score English Madrigal Schoo 11: 19-29)

These lyrics differ from ballet lyrics only in the avoidance of the first perso plural pronoun: ballets typically consist of group calls for communal participation in the pleasures of singing and dancing (e.g., "Sing we and chant it," "Say dainty dames, shall we go play," "Sing we at pleasure," "We shepherds sing, we pipe, we play" [English Madrigal Verse 148, 286, 287, 288]). Yet what chiefly makes Weelkes's "O Care "madrigalesque" rather than ballet-like--and therefore "artificial" and, presumably, "delightful" to "men of understanding"--is, most obviously, its treatment of the poetic text. It breaks the lyric into two separate madrigals, each setting one stanza; it breaks the lines into fragmented phrases and sections, sometimes repeated, which distort both the syntax and some of the formal pattern of the lyrics; and it gives a predominantly polyphonic--that is, multiple-voiced and imitative--rather than homophonic treatment of phrases.

This "madrigalesque" manner of treating poetic texts has been attacked by critics of madrigals generally as a distortion of the lyrics rather than being seen as evidence of artfulness (Kermode 267). At the same time, critics have also condemned the kinds of poetic text that madrigals typically set to music a unworthy of serious treatment. John Stevens claims that "to study, or sing, [Elizabethan madrigals] is to realize afresh how little the Elizabethan compose required of his text" (24), and that "in performance [the settings] made a complete jumble of the poem both to the listener and to the singer" (34). Kerma complains that madrigals contain little "to detain the connoisseur of Elizabethan poetry," the lyric being, as he puts it, "never without its elaborate conceits and rarely without the most exaggerated expression of lovers heartaches." Yet he also complains that madrigals, while "professing to follow text, actually smothered it with sophisticated musical devices" (Elizabethan 26 11; see also Maynard 42ff.). The madrigal is thus condemned both for the poor poetic quality of its lyrics and for its distortion of those same lyrics.

In defense of madrigal lyrics and their settings, Doughtie rightly notes that the lyrics served primarily to name "a mood or action or emotion that the composer could exploit," the madrigal's appeal being what he calls "mainly musical...composed for performers rather than for audiences" (Introduction 2). The jumble which Weelkes's setting makes of the poetic text for non-performing listeners and score readers in fact enhances singers' apprehension of the text through its repetitions. Performers, unable to review the lyrics as they sing, almost require such repetition. Doughtie's claim that madrigals' appeal is mainly "musical," however, requires qualification; we need to specify the kind of musical appeal, performer and understanding which the madrigal addressed. Th madrigal's "distortions" focus attention on "local" features of both the lyrics and what Morley calls the "artificial" counterpoint used to set them. Such distortions would delight the "understanding" of those interested in musical techniques of text-setting per se, an understanding likely to be cultivated by music connoisseurs such as Yonge's merchants and gentlemen.

The very real sense in which madrigal performers participated with the composer and poet in the creation of the madrigals through performance explains why setting techniques--use of unprepared dissonances, suspensions, chromaticisms--should have held performers' interests. Such techniques were offered up to cognoscenti as objects for them to wonder at or admire in ways which would display the depth of their understanding of music. Given, again, th absence of any full score against which to check the accuracy of their performance, any irregularity in counterpoint would arouse the singers' curiosity if not skepticism concerning the composer's intentions and the printer's fidelity to those intentions. Any striking setting of affective words or phrases would doubtless draw comment from the appreciative cognoscenti. "O Care," for example, is remarkable for the use of a double chromatic progression (see Kerman, Elizabethan 213-14); for the affective renderings of the words "Care," "music," "deadly sting," and "cruel"; and for the contrast achieved by using chains of harmonic suspensions for setting lines one, three and five, and using parallel thirds, sixths and tenths for setting the "fa la's" and lines two, four, six and eight. The employment of a number of composers of so-called "serious" madrigals as musicians in households, which would allow them to offer "authorized" assistance in the performance and appreciation of such passages, made the inclusion of such passages all the more appropriate. As both Kerman an Helen Wilcox have stressed, the steady patronage that musical families gave som English madrigal composers not only provided them with the leisure to work but also encouraged experimentation with stylistic devices (Kerman, Elizabethan 243 Wilcox 61).

We should therefore resist the temptation to dismiss madrigals like those of Weelkes as insignificant because of their concentration on form alone, for that concentration itself is significant. It illustrates how one sort of art occasions what Williams describes as a specific set of intentions and responses identified as "aesthetic"; through concentration on form, an art work can effec an evasion of immediate connections or even "an evacuation of immediate situation" (156). More interestingly, Weelkes's madrigal, though it may be used to carry out such an evacuation or evasion, is also precisely and explicitly about such an "evacuation," as its lyrics make clear: the singers command Care to leave and summon music to dispel Care and to sustain them. By calling attention to the strategy, this madrigal undermines the very activity of evadin Care through music that it purports to make possible. It thus combines an appea to interest in techniques of word-setting with an invitation to reflect on the evasion which taking such an interest attempts. What might be imagined as the specific "aesthetic situation" which this madrigal occasions is thus both invoked and itself dispelled.

Recognition of this double movement may account for contradictions in the responses that individual critics have had to Weelkes's madrigal. Critics have attempted to explain the madrigal in terms of the authority of its author/composer Weelkes and/or in terms of de-socialized generic conventions, ignoring in either case the song's existence in performance. In The Elizabethan Madrigal, for example, Kerman vacillates between several contradictory extremes he first labels it "most astonishing....at the same time the greatest, the most extreme, and the least experimental of English madrigals" (218); he then goes o to dismiss it as a "sport...[un]typical...of the English madrigal...or even of Weelkes himself" (219); lastly he acknowledges it to be "the culmination of Weelkes' chromatic experiments" and "extremely affecting" (230). Weelkes's biographer David Brown gives it the oxymoronic label of being a "tragic ballet, claiming that "the fa-las do not contradict the couplets with unseemly gaiety, but instead intensify the sorrow by nostalgically reflecting the joy of which the lyric despairs" (99). These contradictory reactions implicitly register the peculiar esthetic situation madrigals can effect. The disparity of giving a madrigalian setting of ballet-like lyrics in "O Care" occasions self-reflectiveness by calling attention in the lyrics to the strategy of the music. It thus exemplifies explicitly both the kind of evacuation of situation which madrigals aspire to achieve and the dependence of such an achievement on social and material conditions: the immediate social gathering and the conditions making such a gathering and such a performance possible.

The publication of songs through part books accentuated a group relationship between composer and performers in which ownership of and creative responsibility for the songs was shared. Such "sharing" suggests the degree to which, for late Renaissance English songwriters and performers, song performanc was a matter of completing, finishing, perfecting, texts presumed inherently incomplete, unfinished, imperfect. Some songwriters, however, disputed this relationship, preferring that songs be performed as the songwriter commanded. Especially in the case of the lute songs and the lutesong books, and most strikingly in Campion's case, the sense of performance as completing the admittedly incomplete was contested. We can see this contestation even in the folio format in which lutesong books were published. Whereas part-book publication suggested a sharing of creative responsibility among composer and performers, the folio format established hierarchic relations between individuals. It is true that lutesong books commonly provided both solo version of the songs and part-song arrangements through table-book layout. Thus they di accommodate a variety of performance options and, for those choosing to perform the songs as part songs, they also occasioned a social gathering far more physically intimate than that occasioned by the performance of songs published in part-book form. The provision of the full score (in the solo version), however, and the presence of all the parts on one localized surface in the table-book layout, in effect gave authority to the text. By centering the focus of the performers on a single surface, though from different angles, it took away from performers any great responsibility for constructing the song and effectively raised the status of the published text itself (and, by implication the composer of the text) in relation to the performers. Performances could be "checked" easily against the authority of the full score. As a result, even performing these songs as part songs was transformed from an occasion in which participants created the song to one in which they attempted to adhere to the song, to be faithful to the score and the composer's intentions.

This imposition of the authority of the text coincides with an increased testiness in the prefaces of lutesong composers toward their art, and it coincides with composers' increasing identification of themselves with "their" compositions as "authors." In the preface to Hume's 1605 book (repeated in his 1607 book), he insists, "These are mine own Phansies expressed by my proper Genius" (Lyrics 197, 285). Alphonso Ferrabosco II boasts in the preface to his book of airs (1609), dedicated to Prince Henry, "J [sic] know [my airs] worthy of my Name: And, therein, J tooke paynes to make them worthy of Yours" (Lyrics 291). John Dowland and Campion, the two most prominent lutesong composers, provide the clearest examples of such testiness. In the preface to his first (1597) book of airs, Dowland goes to remarkable lengths to advertise what he describes as "the successe and estimation even among strangers I have found" (Lyrics 67-68). In the preface to his final (1612) book of airs, he again tells readers of his success abroad, complains of young upstarts and expresses concer for the status of the lute, his instrument (Lyrics 400-01). Further, he does so in spite of the fact that in 1612 he "still rode the tide of popularity" (Poulton 76). Similarly, Campion presents a virtual manifesto of the lute "air" in the preface to the 1601 book in which his first airs were published, and he presents the songs in the book (including Philip Rosseter's) with no provision for part-song performance, thus resisting the temptation to accommodate different performance options (Works 12). Though Campion does include part-song arrangements in his two next books of airs (c. 1613-14), he returns to his original position of providing no part-song arrangements in publishing his last two books of airs (c. 1617). Even when he does provide part-song arrangements, however, he seems less interested in accommodation than in control--pre-empting bystanders' attempts to improvise their own descants to the tunes by providing "authorized" parts for them to sing (see Works 55).

Lute songs typically provide a homophonic musical setting with a dominant melodic line giving a syllabic treatment of the text of the lyrics. Doughtie distinguishes the setting techniques of the lute airs from those of other song genres by suggesting that the lutesong lyrics were somehow more worthy. As he puts it, "If the words deserve to be heard, a composer is likely to be moved to set them to music which will allow them to be heard; conversely, if a composer prefers musical forms in which the words can be heard, he will probably select words he thinks worth hearing" (Introduction 3). Yet, if it is true that the lute airs are often designed to make their words "heard," it does not necessarily follow that the words set are somehow intrinsically more "worthy" than the words of other song genres. Doughtie's comment points to the isolation of the singer/performer from the listeners, those who simply hear the words rather than read them as they sing them. Campion himself points to this isolation of airs when he observes in his first preface that the air is "naked...without guide, or prop, or colour but his owne...easily censured of everie eare, and requires so much the more invention to make it please" (Works 15). The air, in Campion's description, is something "presented" to the censuring ear. The distinction between the performer and the listener established by the air makes its performance a presentation of the song--both the lyric text and its setting--and makes that performance into a formal rather than an informal one.

This very formalization of the performance occasion was intensified in some of the airs by the technical demands they made on performers. To the degree that they required virtuosity in playing or singing, their performance would require the presence of one or more professional musicians. The need to fix a time for the professionals to perform and the listeners to hear would in such cases itself formalize the occasions of the song performances into what we now commonly think of as performances: "presentations" rather than "playing," a carrying out of "that which is commanded." Madrigals and ballets could be performed at a dinner by the guests themselves or by family members for the enjoyment of the performers, as the prefaces of Morley, Yonge and others attest did indeed happen. In contrast, lute songs, though the performer might enjoy them, were also clearly directed at nonperforming, passive listeners.

This formalization and the isolation of the song and particularly the performer that the lute song entails may help to account for some of the testiness of the lutesong book composers. Furthermore, such formalization and isolation distinguish the relations that these songs establish from those established by the part-songs as being those between individuals, often individuals of unequal status, and makes their function a "courtly" one: presenting one person's suit for favor from another, more powerful, person. Dowland recalls at length his pursuit of favor through his music at powerful courts in Europe. Ferrabosco presents his compositions as being ideally worthy of the Prince's name, to whom he dedicates them, as well as being worthy of his own. The "courtly" function o the lute song appears to operate in the composition of the lyrics as well. Though almost all the lyrics of the lutesong books remain anonymous, their very anonymity, as Peter Walls observes (239-40), as well as composers' remarks in some of the lutesong book prefaces, suggest that at least some of the lyrics ha courtly origins. Diana Poulton uses a contemporary account by Sir Henry Wotton to argue that Essex authored the lyrics of Dowland's song "Can She Excuse My Wrongs" as a way of airing his complaints to Queen Elizabeth through the medium of the court singer Robert Hales (226-29):

There was another time long after, when Sir Fulke Grevill (late Lord Brooke) a man in appearance intrinsecall with [Essex], or at the least admitted to his Melancholy houres, eyther belike espying some wearinesse in the Queene, or perhaps with little change of the word though more in the danger some wariness towards him, and working upon the present matter (as he was dexterous and close had almost super-induced into favour the Earle of Southampton; which yet being timely discovered, my Lord of Essex chose to evaporate his thoughts in a Sonnet (being his common way) to be sung before the Queene, (as it was) by one Hales, in whose voyce she took some pleasure.

(Reliquae Wottoniae, 2nd ed., 1654; qtd. in Poulton 227)

If Poulton is right in situating the origins of this song in a complaint Essex made to Elizabeth, one should also note that both the form and the formality of the song would help further Essex's suit. The formality of the occasion derives from the distinction between the song's singer--the professional musician Rober Hales--and the listener, Queen Elizabeth. Essex, as lyricist, would be concerne that the queen should be able to hear and easily comprehend his lyrics. It is i this sense, then, that the words were "worth" hearing in the way that Dowland's accompaniment makes possible. At the same time, however, the framing, or "evaporating," of Essex's thoughts in the form of a love sonnet and the presentation of that sonnet in the form of a song would effectively distance Essex from the lyrics. As Thomas Whythorne explains of songs he wrote for (and about?) his employer, "If she would take it to be written to herself, she might best do it....But and if she would not take it to herself or in good part...yet it is so made as neither she nor none other could make any great matter thereof (31-32). This distancing, he observes, is accentuated by the musical setting of the lyrics: "if [the song] were not to be well taken, yet inasmuch as it was sung, there could not so much hurt be found as had been in the case of my writing being delivered to her to read, for singing of such songs and ditties was a thing common in those days" (40).

The lyrics of "Can she excuse," like other of Essex's lyrics, are similarly "common" in their theme:

Can shee excuse my wrongs with vertues cloake: Shall I call her good when she proves unkind. Are those cleer fiers which vannish in to smoake: must I praise the leaves where no fruit I find.

No no where shadowes do for bodies stand, thou maist be abusde if thy sight be dime. Cold love is like to words written on sand, or to bubbles which on the water swim.

Wilt thou be thus abused still, seeing that she will right thee neuer if thou canst not ore come her will, thy loue will be thus fruitles euer.

Was I so base that I might not aspire Vnto those high ioyes which she houlds from me, As they are high so high is my desire, If she this deny what can granted be.

If she will yeeld to that which reason is, It is reasons will that loue should be just, Deare make me happie still by granting this, Or cut of delayes if that dye I must.

Better a thousand times to dye Then for to liue thus still tormented, Deare but remember it was I Who for thy sake did dye contented.

(Lyrics 72)

Note that these lyrics contain no explicit reference to either Essex or Elizabeth. The presentation of these words in a song performed by a musical servant of the royal household would have the effect of removing the lyrics yet further from immediate applications (versus, say, a hypothetical performance in which Essex might privately sing to the Queen). Yet, as Poulton observes, enoug parallels and allusions exist in the lyrics to have enabled the queen to get their message (228-29). The distancing achieved by the formalizing of messages of complaint through their presentation as songs would give the queen the optio of either admitting the parallels and the complaints or ignoring them while still "appreciating" the song. In this sense the lute songs operated in a particular sort of social "domain," an allowed space for the "airing" of an individual's complaints as well as for displays of his "courtliness."

This domain is distinguished first by the restriction of the complaints to thos of individuals--the speaker of "Can She Excuse" represents no one other than himself. It is further distinguished by the absence of contentment the lyrics express, their presentation as "complaints." Lute songs especially, and especially those by Dowland, are remarkable for their pleas for escape from care, as the first lines of a number of Dowland's lute songs illustrate: "Rest awhile, you cruel cares"; "Sleep, wayward thoughts"; "Go nightly cares"; "Come heavy sleep"; "Unquiet thoughts, your civil slaughter stint"; "Toss not my soul Love, twixt hope and fear." This is not to say, however, that all or even most of the lute songs, including the ones with melancholy-like titles, are dirges. Rather, though they express discontent, they often do so quite playfully, in a way which reveals the speaker's sprezzatura and so his right to have his complaints answered.

"Can She Excuse" is exemplary in this regard, being anything but melancholy or despairing either in its lyrics or its setting. As part of the rhetorical strategy of the song, the speaker assumes both the justness of his cause and th likelihood that the listener will agree with him and grant him his just deserts For most of the song, he "airs" his complaints not directly to the beloved but to a fictive third party, indicated by his use of the third person pronoun to refer to the beloved; but he then breaks through this fiction in the final six lines--when he addresses her directly as "dear." Such playfulness shows that th speaker believes he will not have to be too earnest in his argument and so displays a degree of sprezzatura which establishes the speaker's credentials as a noble and his right to the favor for which he is arguing. The setting accentuates the playfulness of this strategy in a number of ways: by the jauntiness of its galliard rhythm; by the melodic characterizations given the main speaker and the fictive third party, the former singing in wide leaps and the latter in a whispering monotone; and, finally, by the inclusion of the tune "The Woods so Wild" in the accompaniment to the final quatrain, perhaps an allusion to Essex's Wansted woods (full score English School 1.1-2:10-11).

Locating the performances of lute songs in the courtly domain offers a useful alternative perspective to the way many critics have interpreted expressions of "melancholy" in the lute songs. Poulton takes the "melancholy" of Dowland's songs to reveal Dowland's temperament (78); Lillian Ruff and D. Arnold Wilson interpret it as indicating his reaction, shared by other lutesong composers, to the downfall of Essex; Anthony Rooley explains it as evidencing Dowland's immersion in hermetic philosophy; and Robert Toft and Robin Wells argue that he exploits the Elizabethan cult of melancholy to experiment with new musico-rhetorical techniques in song composition (see Wells, "John Dowland" 523 and passim, a reply to Rooley; and Wells, "Ladder"). Though most of these explanations would seem to see the melancholy in terms of social contexts--Essex's career, contemporary cults and fads--they locate the songs in their texts--that is to say, in their scores and lyrics, which are then said in one way or another merely to reflect those contexts. The work accomplished through the performance of the songs is thus effectively evaded: attention is drawn not to the reworkings of the songs in every performance but to the presence of fads, personal temperament or stylistic traits in, or "beneath," static, and thus "silenced," texts.

If, however, we locate the songs in the activity of their performance, we can understand the "melancholy" as part of a negotiating strategy. In this light, the songs function not to express despair (whose language is silence) but discontent addressed to listeners with power to offer ease. Any allusions to Essex's downfall in the lyrics would function not as solitary outcries but protests aimed at redress. Exploitation of clever musical techniques for expressing melancholy would serve not as an end in itself but to display the abilities of the composer or performer to the listener. Daniel Javitch and Fran Whigham have shown how particular literary and social practices in the Elizabethan and Jacobean court enabled individuals to work upon and with existing social conventions in order to gain status and place. In song performance, I am arguing, lutesong conventions could serve to negotiate the individual ambitions and privileges of composers and poets by following a similar "courtly" esthetic of self-promotion.

To locate the "work of art" in performance redefines art as an activity and so as inevitably unstable, incomplete, historically specific, recurrent and yet, because always in process, never recurring. In the case of songs, to focus on performance is to see the songs themselves as sites of negotiation for listeners, performers and songwriters. Through specific conventions of bibliographic format, lyric content and musical setting style, madrigals and lute songs helped both to define and shape not only their genres but the social relations of those participating in their production. Yet those relations, it i worth emphasizing, are re-negotiated in each use of particular conventions and in every performance. Difficulties which critics have had with these songs result from the evasion of song performance, an evasion which blinds them to th strategic significance of particular conventions.

Of course, in adopting a cultural materialist approach, I too am participating in what Williams calls the "struggle for and against selective traditions" (116)--in this case, the struggle to negotiate certain kinds of contracts between critical theories, current dominant culture and the songs. The virtue o the contract I offer is its negotiability; it is admittedly an interpretive "performance" of the songs, a performance situated within a particular critical perspective in competition with a range of other late 20th-century Western perspectives. The critics I have faulted have themselves also been "performing" in the sense of completing the inevitably incomplete, but their strategy has been to deny the negotiability of their critical performances. Yet if we locate the work of songs, like any art work, not as something residing in their texts but as something negotiated and re-created in the play of each new performance, or "interpretation," we can never wholly be finished with that work. Instead, w continually work toward completion of the incomplete in accordance with the contemporary and in negotiation with the past. Tracing the interplay between th material production of a song and the social relationships embedded in its conventions not only illustrates how we might demystify our idealization of (an difficulties with) songs as reified art "works"; attention to that interplay, b foregrounding the role of the critic in producing the work song performs, also reveals the particular and strange historicity of the critic's response to the songs: the momentariness of that response as one situated performance. Attendin to that historicity should make us reluctant to accord final authority either t ourselves as listeners or to songs as texts, aware of the degree to which we rework both ourselves and the songs in every performance.*

* An earlier version of this essay was written for a National Endowment for the Humanities Seminar directed by Stephen Greenblatt. For their suggestions and encouragement, my thanks to Greenblatt, my fellow seminarians and to Donald Franklin, David Harris, Robert Hinman, Joseph Lenz, Bruce Martin, Marianne Novy Peter Rabinowitz, Philip Smith, Thom Swiss, and especially Barbara Hodgdon and Min-Zhan Lu.

WORKS CITED

Barthes, Roland. Image--Music--Text. Trans. Stephen Heath. New York: Noonday, 1977.

Brett, Philip. "Text, Context, and the Early Music Editor." Authenticity and Early Music: A Symposium. Ed. Nicholas Kenyon. London: Oxford UP, 1988. 83-114.

-----. "Word-Setting in the Songs of Byrd." Proceedings of the Royal Music Association 98 (1971-72): 47-64.

Brown, David. Thomas Weelkes: A Biographical and Critical Study. New York: Praeger, 1969.

Campion, Thomas. The Works of Thomas Campion: Complete Songs, Masques, and Treatises with a Selection of the Latin Verse. Ed. Walter R. Davis. Garden City NY: Doubleday, 1967.

Doughtie, Edward. English Renaissance Song. Boston: Twayne, 1986.

-----. Introduction. Lyrics from English Airs 1596-1622. Ed.Doughtie. Cambridge Harvard UP, 1970. 1-41.

Dowland, John. Ayres for Four Voices. Musica Britannica 6. Transcribed Edmund Fellowes. Ed. Thurston Dart and Nigel Fortune. London: Stainer, 1953.

Duckles, Vincent. "Florid Embellishment in English Song of the Late 16th and Early 17th Centuries." Annales Musicologiques 5 (1957): 329-45.

Edwards, Warwick A. "The Performance of Ensemble Music in Elizabethan England." Proceedings of the Royal Music Association 97 (1970-71): 113-23.

The English Madrigal School. 36 vols. Transcribed and ed. Edmund H. Fellowes. London: Stainer, 1913-24.

English Madrigal Verse 1588-1632. 3rd ed. Ed. Edmund H. Fellowes. Rev. ed. Davi Greer and Frederick W. Sternfeld. Oxford: Clarendon, 1967.

The English Madrigalists. 36 vols. Transcribed and ed. Edmund H. Fellowes. Rev. ed. Thurston Dart. London: Stainer, 1956-68.

The English School of Lutenist Song Writers. 16 vols. Transcribed and ed. Edmun H. Fellowes. London: Stainer, 1920-32.

Ferand, Ernest T. "Improvised Vocal Counterpoint in the Late Renaissance and Early Baroque." Annales Musicologiques 4 (1956): 129-74.

Greenblatt, Stephen. Shakespearean Negotiations: The Circulation of Social Energy in Renaissance England. Berkeley: U of California P, 1988.

Greer, David. "Manuscript Additions in Early Printed Music." Music & Letters 72 (1991): 523-35.

-----. "The Part-Songs of the English Lutenists." Proceedings of the Royal Musi Association 94 (1967-68): 97-110.

Horsley, Imogene. "Improvised Embellishment in the Performance of Renaissance Polyphonic Music." Journal of the American Musicological Society 4 (1951): 3-19 Javitch, Daniel. Poetry and Courtliness in Renaissance England. Princeton: Princeton UP, 1978.

Jones, Edward Huws. The Performance of English Song 1610-1670. 1978. New York: Garland, 1989.

Jorgens, Elise Bickford. The Well-Tun'd Word: Musical Interpretations of Englis Poetry 1597-1651. Minneapolis: U of Minnesota P, 1982.

Kerman, Joseph. Contemplating Music: Challenges to Musicology. Cambridge: Harvard UP, 1985.

-----. The Elizabethan Madrigal: A Comparative Study. New York: American Musicological Society, 1962.

Kermode, Frank. Rev. of Music and Poetry of the English Renaissance, by Bruce Pattison. Review of English Studies 25 (1949): 265-69.

Leppard, Raymond. Authenticity in Music. London: Faber, 1988.

Lyrics from English Airs 1596-1622. Ed. Edward Doughtie. Cambridge: Harvard UP, 1970.

Maynard, Winifred. Elizabethan Lyric Poetry and Its Music. Oxford: Clarendon, 1986.

Middleton, Richard. Studying Popular Music. Philadelphia: Open UP, 1990.

Morley, Thomas. A Plain and Easy Introduction to Practical Music. 1597. Ed. R. Alec Harman. New York: Norton, n.d.

Orgel, Stephen. "The Authentic Shakespeare." Representations 21 (1988): 1-25.

-----. "What Is a Text?" Research Opportunities in Renaissance Drama 24 (1981): 3-6.

Pomeroy, Elizabeth W. The Elizabethan Miscellanies: Their Development and Conventions. Berkeley: U of California P, 1973.

Poulton, Diana. John Dowland. Rev. ed. Berkeley: U of California P, 1982.

Price, David C. Patrons and Musicians of the English Renaissance. Cambridge UP, 1981.

Rabinowitz, Peter J. "Circumstantial Evidence: Musical Analysis and Theories of Reading." Music and Literature. Ed. W. John Rempel and Ursula M. Rempel. Winnipeg: Mosaic, 1985. 159-73.

Rooley, Anthony. "New Light on John Dowland's Songs of Darkness." Early Music 1 (1983): 6-21.

Ruff, Lillian M., and D. Arnold Wilson. "Allusions to the Essex Downfall in Lut Song Lyrics." Lute Society Journal 12 (1970): 31-36.

-----. "The Madrigal, the Lute Song and Elizabethan Politics." Past & Present 4 (1969): 3-51.

Schleiner, Louise. The Living Lyre in English Verse from Elizabeth through the Restoration. Columbia: U of Missouri P, 1984.

Seventeenth-Century Songs and Lyrics Collected and Edited from the Original Music Manuscripts. 1959. Ed. John P. Cutts. Freeport, NY: Books for Libraries, 1969.

Short, R. W. "The Metrical Theory and Practice of Thomas Campion." PMLA 59 (1944): 1003-18.

Spink, Ian. "Sources of English Song, 1620-1660: A Survey." Miscellanea Musicologica: Adelaide Studies in Musicology 1 (1966): 117-36.

Stevens, John. "The Elizabethan Madrigal: 'Perfect Marriage' or 'Uneasy Flirtation'?" Essays and Studies 11 (1958): 17-37.

Taruskin, Richard. "Tradition and Authority." Early Music 20 (1992): 311-25.

Toft, Robert. "Musicke a Sister to Poetrie: Rhetorical Artifice in the Passionate Airs of John Dowland." Early Music 12 (1984): 190-99.

Walls, Peter. "'Music and Sweet Poetry'? Verse for English Lute Song and Continuo Song." Music & Letters 65 (1984): 237-54.

Wells, Robin Headlam. "John Dowland and Elizabethan Melancholy." Early Music 13 (1985): 514-28.

-----. "The Ladder of Love: Verbal and Musical Rhetoric in the Elizabethan Lute-song." Early Music 12 (1984): 173-89.

Westrup, J. A. "Domestic Music under the Stuarts." Proceedings of the Royal Music Association 68 (1942): 19-52.

Whigham, Frank. Ambition and Privilege: The Social Tropes of Elizabethan Courtesy Theory. Berkeley: U of California P, 1984.

Whythorne, Thomas. The Autobiography of Thomas Whythorne. c. 1576. Modern spelling edition. Ed. James M. Osborn. London: Oxford UP, 1962.

Wilcox, Helen. "'My Mournful Style': Poetry and Music in the Madrigals of John Ward." Music & Letters 61 (1980): 60-70.

Williams, Raymond. Marxism and Literature. New York: Oxford UP, 1977.

Wolff, Janet. "The Ideology of Autonomous Art." Music and Society: The Politics of Composition, Performance and Reception. Ed. Richard Leppert and Susan McClary. Cambridge: Cambridge UP, 1987. 1-12.

Wulstan, David. Tudor Music. Iowa City: U of Iowa P, 1986.

Yonge, Nicholas, ed. Musica Transalpina, London, 1588.

ABOUT THE AUTHOR

BRUCE HORNER is Associate Professor of English at Drake University, where he teaches writing, song criticism and literacy studies. He is a specialist in 17th-century English songs and has also written on the teaching of writing and relations between English studies and music. His publications include essays appearing in Rhetoric Review, Journal of Advanced Composition and English Education. Currently he is completing a co-authored book on "Representing the 'Other.'"