The Muse Learns to Tape

[This essay was published originally in Reimagining Textuality: Textual Studies in the Late Age of Print, edited by Elizabeth Loizeaux and Neil Fraistat (University of Wisconsin Press, 2002).  The University of Wisconsin Press holds the copyright on this piece, and it is posted here with the permission of the press.]


Even for country kids like us attending “Calistoga Joint Union High School” seemed a joke to savor—at least in 1966 with San Francisco in driving range.  And on weekends we’d pool our money and head to the ballrooms for the lights and bands, and maybe even dance a little—in spite of our pretty short hair and lack of costume.  We heard bluesmen like Jimmie Reed and John Lee Hooker, jazzmen like Yusef Lateef, and an odd array of the soon-to-be-famous and never-to-be-heard-of-again (Cream to Sons of Champlin by way of Aum and Sanpaku)—and yes the Grateful Dead and Jefferson Airplane, and when Jerry Garcia trucked on to that great Fillmore in the sky I found myself thinking that those evenings had something to do with the nature of textuality.

Mostly we heard the San Francisco bands perform before we heard their records.  We assumed—so did the musicians I think—that the performances were the “real” music (and one reason for Bill Graham’s success may have been that his “dance-concerts” felt like louder, more kinetic coffee house readings instead of the rock shows at the armories and gymnasiums where teens came to watch their treasured 45s acted out).  So when one of our bands signed that contract that meant they were stars, we were pleased—we thought of the musicians as friends we just hadn’t met.  But the recordings that followed struck us as lesser than the performances and performing.  If they were more disciplined and more realized than what we heard on the weekends, they were also safer and more abstract.  We were spared Country Joe McDonald (too ripped to tune his dangling guitar) weeping over the invasion of the DMZ as the Fish swam on waiting for their leader to compose himself enough to continue, but we missed that risk and glee of Marty Balin and Grace Slick spinning multiple choruses of “It’s No Secret,” the high notes bouncing off the Fillmore walls and ceiling but, at least it seemed so then, harmonizing perfectly.

A music that stressed communal creation and consumption, spontaneity and individualism, the rejection of social and musical conventions, might have a great deal of vitality, even a certain authenticity, but of course record companies had little choice other than to distill what we valued—the fluidity of performance—if they were to construct marketable products, and we were too naive (for all our left-leaning politics) to respond with Gramscian critiques of this commodification of “our” music.  We just assumed that music that was packaged to fulfill commercial conventions was somehow lesser and skipped the pleasures of The Beach Boys and Motown acts—the costumes qua uniforms, the choreography, that ability to replicate in performance every note so precisely placed on those wonderfully realized recordings.  And while our “taste” then (or lack of it) may say more about our assumptions about race, class, and region than I might like, our reactions also point to a dichotomy between music as performance and music as composition, between music as a process and music as sound constructed into a preservable, reproducible object.  And it is here that Garcia and “textuality” begin to bear on each other.

A few days after his death I saw a clip of a young Jerry describing the bluegrass festivals of his folkie days—family affairs of picking around the pickup tailgates and savoring the locals and a headlining star—the Stanleys, whomever.  These festivals-as-occasions-for-community seemed precursors to his own aesthetic, and he noted that the promoters used to put electrical outlets along the stage for fans who wanted to tape the doings (early 1960s tape recorders weren’t as portable as today’s gear).  He didn’t mention what Decca Records thought about fans bootlegging Bill Monroe, but he assumes it was square with Bill.  To Garcia anyway, the point was to perform for an audience, and he apparently thought the point for the audience was to participate in the music as it happened—each unique performance fully “real” only as it was unfolding.  The reel of tape a fan might carry away wasn’t really the music, just a representation of an instance of it.  The music was the performance: the being there, the matrix of performer, sound, circumstance, and audience.  And the tape a fan might make would increase the desire to attend another performance (that it might reduce the demand for a commercially produced recording of the music was to Garcia, it seems, largely beside the point).

For Garcia, then, there could be no adequate text of “Dark Star” or even “St. Stephen” if by “text” we mean a pure, final, fully realized, storable and reproducible representation of the “work.”  The music is either a possibility (the skills, motifs, and patterns of interaction a set of performers use to enact a “work” by a semi-improvisational blend of creation and recreation), an actual moment of performance, or a memory of an array of performances (some documented, some not), among which one might have favorites but where none is the definitive representation or text of the work.  It is hardly surprising, then, that the Dead seem initially to have been unsure how to represent their performing on their records and that the group’s early albums explore different strategies for managing—or resisting—the seeming fixity, the seeming definitiveness, the pressure to distill introduced by the decision to “make” a record as opposed to “playing” music. [1])

For the most part our debates about editorial theory derive from, and reflect, the typical occasion of our practice: works which now have their primary existence in scripted or printed language.  Textuality, as we have debated it, is intertwined with writing.  Even with such pieces as The Iliad and the Bible’s various “books,” we primarily work from (or toward) the moment they were given written shape.  For one thing, we have no access to their prior states.  For another, their oral “form” was likely a succession of forms, each retelling a semi-improvisational reinvention, only partly captured by writing’s fixity and linearity.  A Grimm fairy tale, read from the printed page at bedtime, is not the same thing as the tale reinvented from memory with particular variations in performance for specific children before a particular cottage hearth.[2]

Our emphasis on written materials points to an axiom I think most of us share: that a work must be physically embodied to be textual; it must have a preservable form apart from its creator or performer.  This storage of the work in a stable medium can take the form of chirographic script, print, or musical notation (representations which can be reproduced and transmitted), or the work can itself be a unique physical object: a carving, a painting, a building (these are also experienced long after their production).  But unless the creative process yields some such physically preservable and relatively stable form, the work ceases to be even as it is being performed, since the performance leaves no record.  Instead of a “text” or “work,” there is only the potential (in the performers) to generate more or less analogous instances of the particular tradition through new, unique performances from the terms of practice that happen to define the “tradition” at a particular moment.  And even this potential survives only while the tradition continues to renew itself through succeeding generations of practitioners and their performances for succeeding generations of audiences.

This axiom points to textuality’s power and importance: unless practices and performances can be given a physically preservable form and thus be brought within what might be termed the textual horizon, they fade from the cultural record.  And much of the vernacular art of the past has, quite clearly, been irretrievably lost because it was never given a textual form.  However, as the example of the Grimm tale suggests, giving non-textual materials textual form alters them in significant ways.  At least with writing (historically the most powerful and pervasive textual medium), using it to preserve oral materials transforms the materials even as they are preserved.  Textualizing extends the circulation of oral traditions and practices; it prolongs their cultural life; but it also converts them from a set of elements, procedures, and occasions for performing a “work” (or rather enacting an occasion) into a single, fixed example of what those elements, procedures, and occasions might have produced.  Valuable, even necessary, this textual recording and recasting replaces what was a cultural process or praxis with a cultural “artifact.”[3]

In general we recognize that the conditions of textuality have changed across time.  The textual modes of Homer’s Iliad, of sixteenth century manuscript culture, and of our own world of print are not the same, and (reasonably enough) we approach the editing of ancient epics, poems circulated in manuscript, and late 19th century novels differently.  We recognize, that is, that textuality is historically conditioned and that its terms and processes are intertwined with both social and technological factors.  But at times we seem to forget (or at least take for granted) that textuality is “historical” in another sense as well.  Most of the works we study and their textual representations were produced within and for textualized traditions or textualized so long ago that it is difficult to imagine them as anything other than textual.  Moreover we are deeply enmeshed in the literacy that makes our textual practice possible.  Both training and habit conspire to obscure not only how fundamentally the textual dynamic changes what it preserves but also to obscure that textuality is itself an historical development—that there was a time (pre or post lapsarian) prior to textuality.

To textual scholars, this pre-textual condition is seemingly beside the point, but its existence (even if now past) poses a question: to what extent is textuality’s transformational impact inherent in the nature of textuality itself and to what extent does it change with (and reflect) particular textual media?  Until recently we might not have asked this question—at least in this form.  But two factors bring it into view and offer ways of exploring it.

The first (implicit in much of what I’ve said) is the exploration—initiated by the Homeric scholar Millman Parry and since extended by Eric Havelock, Walter Ong, and others—of orality and literacy as counterpointed cultural modes.  This research indicates that oral and literate cultures not only evolve different procedures for inventing and conserving verbal works but that “orality” and “literacy” encourage different modes of logic and support differing styles of cognition.  The oral performer (whether of something as grand as epic or humble as African American folk blues) relates to his materials and shapes his “work” in a fundamentally different manner than the literate creator (whether John Milton or Barry Manilow).

The second is the development of new modes of storage—film, audio tape, video tape, and (more recently) various digitized media.  Like written language, these can be used to store, multiply, and circulate cultural products.  They are, that is, “textual” (the boom, first, in VCR movie rentals and, now, DVD rentals and downloads shows that they are used as such).  These media, though, more nearly capture the actual texture and process of performance than writing; they offer a degree of transparency, a neutrality, that writing cannot manage.  Recording a work in these new media, that is, requires less abstraction than writing does or musical notation.  Reading the score of an opera is not the same as listening to a recording of it is not the same as watching a film of it (and none of these is quite a front row seat at the Met).  Or perhaps more to the point: a videotape of a Homeric performance would bring us significantly closer to the dynamic of oral creation and the reality of that practice than the recovery of even the earliest manuscript of The Iliad.

The greater transparency of these new textual modes highlights writing’s transformational power; it also moves us closer to being able to give non-textual works (folk and vernacular performances of various sorts) textual form while yet preserving something of their original, pre-textual dynamic.  If so, the impact (if any) of these new textual modes on the performances and performance traditions they preserve should help clarify the extent to which textuality itself (not just written textuality) is transformational—and perhaps something of the way it is transformational as well.  The nature and process of these newer textual modes might even have some relevance for the ongoing debate over the meaning of “work” and “text” and their relationship to each other, since these are terms that we have tended to address within the historical and technological context of writing without really doing much to determine the general validity of this specific and contingent textual mode.  At least within the textuality of recording (whether audio or video), the categories of “work,” “text,” and “performance” appear to be not simply interdependent but also relative; they seem to take their meaning from the specific structures and dynamics within which pieces are produced and consumed; they seem, that is, elements that both enact and derive from what might be termed a “textual rhetoric” (of which our models for written productions are specific instances).

2:19 BLUES:
Recorded music offers numerous illustrations of the contrast between music as a textual practice (where we assume either that a particular notated score or a particular construction of sound preserved on tape or vinyl is the musical work) and music as a tradition of performance.  And it may be that Lennon and McCartney, Dylan, Garcia, Jagger and Richards, and some others owe their relative creative longevity, something of their staying power, and much of their ability to catalyze other performers to their having early on sensed this dialectic (each probably in different terms) and chosen to exploit it.  Their work/working can be “read,” that is, partly as an ongoing exploration of the collision between the pre-textual and the textual, between the flux of the oral and the stability of the constructed object, between the dynamics of folk practice and those of pop art.  But rich as these examples are, they reflect what happens when performers whose heritage is primarily textual encounter practices that are essentially not textual and then try to accommodate to these new (actually old) possibilities—the first generation of British rockers collecting the 1950s blues 78s of Muddy Waters, Howlin Wolf, Slim Harpo, and others; Dylan, Garcia, and others cutting their eye teeth and molars on the compilations of 1920s 78s of rural white and black southern performers (from Dock Boggs and Uncle Dave Macon to Mississippi John Hurt and Blind Lemon Jefferson).[4] For clarifying the dynamic of textuality per se, the more significant question is what happens when performers whose practice is primarily oral have their music “textualized” by recording without themselves necessarily having had to master a textual technology—what happens, that is, when textuality, it might be said, overtakes a performance tradition.  And for this, the early history of jazz offers a particularly suggestive set of examples.

Histories of jazz typically note that it is a hybrid music that amalgamates African American and Western European materials and performance practices, and they often rehearse the probable terms of exchange in the century’s first decades between those African American musicians in New Orleans who were musically literate (i.e. could read music, had some knowledge of harmony, and had some training in how to play their instruments) and those who were musically illiterate (i.e. could not read music, whose instrumental technique was developed by trial and error, and whose approach to performing derived from the orality of African American folk musical practices).  The politics and sociologies of this interaction are rich and complex, and have been debated at length.  The point I would like to develop here is that this was a collision not only between the aesthetics of two different African American communities (one more middle class, one lower on the socio-economic ladder), between two different musical traditions, but also between an art that was fully textual (composed, preserved, and transmitted in musical script) and one that was non-textual (oral and performative).  It is also important that initially, at least, the non-textual dynamic seems to have been the force driving this transformation, and since this re-oralizing of the compositional elements occurred prior to this music and performance tradition being recorded (and performances were neither fully nor easily notatable in conventional script), the amalgam that resulted had to depend—initially—for its preservation and transmission on the resources that typify oral folk practice—memory, apprenticeship, and semi-improvisational recreation.

At the turn of the century when ensemble jazz first developed in New Orleans, its explicit musical elements were the melodies and voicings of brass band marches, ragtime, and blues as these were borrowed and synthesized by amateur and semi-professional musicians whose small ensembles provided functional music for neighborhood dances and picnics, funeral processions, and the like.  Some bands were as much social clubs as musical enterprises, and the musicians were essentially folk artists who evolved their practice through an improvisational give and take.  Initially, that is, jazz seems to have been a performance tradition, a practice, learned by imitation, in which no two performances would have been the same.  It was not a textual art, and the fact that the ragtime and march strands used in much early jazz had a written basis does not really contradict this, if we consider how these elements probably entered the vocabulary of these (for the most part) never recorded musicians: the way early jazz performers apparently recast these materials suggest they learned them aurally rather than from notated arrangements and that they circulated them aurally as well; these musically literate materials, that is, were functionally no different than the blues strands (oral material having no basis in written tradition) with which they mixed in practice.

Ironically, the musically literate jazz pianist Jelly Roll Morton (initially piano players were an all but separate tradition to the small brass and woodwind ensembles that played the parades) offers one of the clearer demonstrations of what might be termed the “oralizing” of written musical material.  In his 1938 Library of Congress interviews, he talks about both his development as a musician and the way jazz coalesced from, and out of, various performance traditions.[5] Although his tendency to mythologize his own centrality (at times it seems he’d have us think that his history and jazz history are one and the same) partly compromises his validity as a witness to early jazz, the musical demonstrations he offers with his tales still seem to convey something of the process by which he and others created the music.

Morton could read and write music, had been exposed to both opera and non-vocal classical music, and could handle the written popular piano pieces that were a mainstay in polite white and black, pre-Victrola middle class parlors.  He was also, though, a “professor” (i.e. a brothel piano player), a pimp, hustler, and gambler as occasion dictated.  Operatic airs were not the musical currency of this realm; Chopin’s harmonic intricacy was in less demand than the rhythmic drive of the rural piano stylings being developed in the barrelhouse joints by self-taught musicians who were adapting rural folk blues to the piano.  At the Library of Congress sessions Morton both recalled and demonstrated these two competing musical worlds—one written and polite; the other oral, improvisational.  And his key example was ragtime.

As elaborated by Scott Joplin and others, ragtime is a composed music with clearly defined sections, multiple melodic themes, and precise counterpoint.  Its chief innovation was using a mild syncopation (borrowed from the folk practice) as an element in written compositions that still foregrounded melody, harmony, and counterpoint.  Ragtime, that is, formalized (and popularized) an aspect of the folk practice, and the compositions—published as sheet music—could be found next to such pieces as “Nola” and “After the Ball Is Over” on the pianos in those polite parlors.  In the Library of Congress recordings, Morton plays Joplin’s “Maple Leaf Rag” twice: first (in a close approximation of the score) as he claims Joplin performed it, then as he used to play it in the brothels and bars.  Playing the piece á la Joplin, Morton articulates the melodic strands and counterpoint; he also carefully subordinates the syncopation to the melody and counterpoint.  In this approach the harmonic features are clearly ordinate, the rhythmic feature subordinate, and the piece seems primarily something to listen to and appreciate for its formal, yet sprightly elegance.  In offering the piece á la Morton, though, Morton abstracts the melodic themes to their basic intervals and mostly ignores the counterpoint.  He slights, that is, the elements that distinguish “Maple Leaf Rag” as composed music.  Yet he isn’t just reducing the composition.  His “Maple Leaf” has a rhythmic drive and variation that Joplin’s lacks, not simply because he emphasizes the syncopations but because he plays with them and because he uses his simplifications of the themes as occasions for both melodic and rhythmic improvisation; Morton replaces  the intricacies of Joplin’s harmonic counterpoint with an equally intricate “rhythmic counterpoint.”

If we equate sophistication with European art music and its foregrounding (at least until recently) of melody and harmony, ragtime as Joplin composed it and according to Morton performed it is a more sophisticated form than its jazzed interpretations (even when played by someone as sophisticated as Morton).  Yet it’s no surprise that the oral (i.e. unwritten) blues stylings continued to flourish in the bars and brothels and that composed ragtime (even though it borrowed from this oral tradition) had to be subsumed back into the dynamic of oral practice in order to have a role outside the parlors and concert halls.  There are several clear and important reasons for this beyond the technical limitations (for playing notated art music) of some of the musically illiterate bar and brothel pianists.  One is rhythmic.  In spite of its syncopations, composed ragtime is relatively static rhythmically, at least for the kinds of dancing in houses of ill repute.  But there is also a fundamental “rhetorical” issue:  With composed ragtime, the performer’s job is to render the text, and that text, that composition, that work, is the point.  For a whorehouse tickler the composition (when there is one) is a means to an end rather than an end in itself; his job is to use his musical skills to interact with his audience and support their interactions with each other (dancing and so on).  His musical materials and mode of performance have to support a certain improvisational freedom if he is to play to and for his patrons (even something as basic as duration is an issue here: when playing a text, the piece is over when it’s over, but in deploying materials improvisationally or semi-improvisationally, the piece can continue as long as there’s a need).

Whether we prefer Joplin’s composition or Morton’s improvisational treatment of it (or neither or both) is a matter of taste, but it seems clear that Morton, while able to render Joplin’s text, was finally aiming at something else.  His focus was less the composition as fixed and final text than how its elements could be redeployed and elaborated in the give and take of specific performance situations and for specific occasions.  In his approach text and work were not one and the same; the written material was raw material from which to generate a series of performances (some more “raw,” others less so), each distinct and different because specific to occasion, yet each related to the other.  It might be said that for Morton each performance is a work, while none of the performances is the work.  Morton’s interest was the fluidity of performance, not the fixity of texts.  For him, Joplin’s ragtime was only useful if he could free it from the page and recast its themes and phrases (which he did).  For all that he was himself a “composer” of a string of important jazz pieces, “King Porter Stomp,” “The Pearls,” etc. (some printed before he began recording) and an “arranger” of his music when he assembled small ensembles to record it, the pieces he wrote and the materials he borrowed (whether learned by eye or ear) were to be elaborated in performances that weren’t strictly limited to what was notated or fully represented by what was notated.  They were occasions for improvisational expression and what was being expressed was not solely the text.

As it happens, Morton’s responsiveness to the “oral” and performative in spite of his command of the written and textual can be heard in his December 16, 1939 performance of “Mamie’s Blues” for General Records.  While he plays a simple blues figure, Morton comments,

This is the first blues I no doubt heard in my life.  Mamie Desdumes, this was her favorite blues.  She hardly could play anything else more, but she could really play this number.  Of course to get in on it, to try to learn it, I made myself the can rusher.[6]

A notated transcription of this piece would offer no more than a simple and basic figure coupled with a few relatively standard blues verses, but as Morton savors the figures—both the ones he plays and sings—he not only infuses the piece with his own presence but projects something of what must have compelled him originally—not Desdumes’ “composition” nor the text she probably could not have written, but her ability to use these few elements to perform her moods.  What Morton seems to have learned from Desdumes and other figures who were—at least musically—of the folk world, was their stance, a stance that was part of his ability to co-opt and transform the written texts of Joplin and others and part of his ability to make something as formally threadbare as “Mamie’s Blues” seem emotionally and musically resonant.

Morton’s ability to read and write music means he is a somewhat tainted example for these issues.  More compelling would be, say, a turn of the century recording of Mamie Desdumes herself playing her blues or trying out one of “Maple Leaf Rag’s” strands by memory and ear (if she could).  Unfortunately, we have little direct evidence for how the earliest jazz musicians approached their materials in performance.  Such legendary figures as trumpeter Buddy Bolden did not write the pieces (as Morton sometimes did) that he and his small band played at dances and probably couldn’t write music in any case.  Nor were these earliest performers recorded.  When Bolden’s group was the rage, cylinders and 78s were for more “polite” musics.  African American musicians playing crude dance music in New Orleans weren’t candidates to record, and we “know” their work mostly through the descriptions of those who heard it and the recordings of slightly later figures (King Oliver, Bunk Johnson) who were influenced by it.  (A couple of Bolden’s near contemporaries, Freddie Keppard for one, did record a few sides but were well past their prime when they did so, and by that time the newer styles of Armstrong’s and others had already somewhat altered the way jazz was played.)  Like Homer, Bolden is beyond the textual horizon—legendary, foundational, and only partly knowable.

What we do know, though, suggests that early jazz was a music where the performers drew on a stock of figures (for the most part learned by ear or so transformed when gleaned from the written text that they quickly became part of the memorized pool of materials with an existence both separate and different from their original textual form) and that the performers elaborated this stock of figures in a collective and semi-improvisational manner as they responded to each other and their audience (usually an audience of dancers).  It seems, that is, to have been a music where the process of the performance (and one’s participation in it whether as player, dancer, or listener) was primary; the songs and structures were scaffoldings for the real “work” which was performing and the participation it enabled.  Although this early jazz was first and foremost an ensemble music, the logic of it somewhat parallels the practice of pre-literate oral poets whose art also seems to have been their ability to select elements from a learned stock in the flow of performance and embellish them in response to a specific occasion and audience (jazz riffs as Homeric epithets?).  And the logic of this early jazz practice (even though it is inferred and necessarily generalized) is enough to allow us to begin to identify the impact of recording on jazz—the effect, that is, of bringing the oral practice of early jazz within the textual horizon.

When record companies first began recording jazz extensively around 1923, the musicians were, it seems, primarily attempting to play in the studio as they would have performed for an actual audience.  For one thing, a group like King Oliver’s Creole Jazz Band would have had no reason to think of the recording process as anything other than a way to represent what they played live, since that had been their only experience of playing to that point.  Yet no one has ever mistaken these early 78s for full and transparent representations of what Oliver’s group played at dances.  For one thing, before the introduction of the “electric” microphone, jazz drummers could not play as they normally would without popping the needle out of the wax platter that was the actual recording medium.  Also, the size of the platter limited performances to about four minutes, and the sound captured by the horn into which the musicians played was thin.  But even with these limitations and distortions, those first 78s of King Oliver’s music are an infinitely richer record than we have of Bolden’s.

Moreover recording introduced a fundamentally new way to textualize a musical performance.  Instead of having to represent the performing by abstracting it into some variety of code (either by describing it in written language or translating it into musical notation), a performer could inscribe the music directly on the wax (later the acetate, then the tape) as he played.  King Oliver did not have to translate his performances into musical notation to produce textualized instances of “Dipper Mouth Blues,” “Snake Rag,” and “Alligator Hop”; he wrote them with his trumpet, and this simultaneity of the performing and the inscribing minimizes the difference between performance and the representation of performance in a way that differs qualitatively from earlier textual modes.

This dynamic, this blurring between the act of performing and the act of encoding a representation of the performing, seems to suggest that jazz recordings would become an increasingly full and faithful representation of what musicians actually created in live performances as the fidelity of the recordings improved.  In one sense this is what happened.  By the late 1920s, 78s could store a fuller image of the actual sound of the instruments.  By the late 1930s amateur recordings were being made of groups playing for actual audiences (one a complete 1940 dance the Ellington Orchestra played in Fargo, North Dakota).  And today’s tape and CD world can leave us wondering whether it’s Ella or Memorex.  We have, it seems, reached the point where recording reproduces performance so fully that a music like jazz could remain fully a praxis, a process, yet attain (or be given) the permanence and reproducibility of a textual artifact.

But this isn’t quite where we’ve ended up.  Ironically, as recording technology improved and made it more possible to capture and transmit jazz as a practice, composition also became a more central feature; the music itself gradually became more of an artifact (and not just with such self-conscious efforts as Paul Whiteman’s to legitimize jazz as a concert music).  First, the collective improvisation of Oliver’s generation gave way to the practice of the individual musicians soloing in a predetermined order over a more or less composed accompaniment largely determined before the performance.  Then as ensembles grew in size (and more jazz musicians could read), written arrangements became common.  By the late 1930s (only fifteen years after jazz recordings became common) jazz had developed from being essentially a folk practice (a process for generating performance) into a popular art (a commodity where each performance of a piece by a musician or group was expected to closely resemble a previous performance of the piece) and in some instances (Ellington, for example) even a composed art music.

The various socio-economic and biographical elements jazz histories cite to explain these developments were certainly factors.  But so was the dynamic of textuality that recording added to jazz, and it is all the more telling that this seems to have happened even though this textual technology enforced less abstraction than writing.  Recording did more, that is, than just make it possible to store actual jazz performances (albeit in imperfect form) and thus allow them to be reproduced and transmitted.  Recording—by making an artifact of performance out of actual performance—fundamentally shifted the performer’s relationship to his materials and audience and thus to his practice.  With the advent of recording, performance could be understood not only as a means and end in itself but also be a means to a specific end (the distillation of the recording) or it could take the form of a self-conscious recasting of a previously recorded performance that both performer and audience might already hold in mind as a model for what the piece was supposed to be.

Recording’s impact is apparent as early as Louis Armstrong.  There were few (any?) jazz musicians in the late 1920s who did not spin his Hot Fives and Hot Sevens over and over, and his bravura flights were carefully analyzed and imitated (they could be precisely because they were recorded), and this mode of learning was quite different from standing next to the bandstand trying to catch and remember the details of a live performance or later talking over the performance (even with the performers).  Textualizing the music, even with a medium as neutral and transparent as analogue recording, thus helped create, then enforce and accelerate, the move in jazz toward working from composed materials, toward more fully elaborated compositions, and toward the adoption of musically more and more complex and abstract materials.  And this shift is analogous to the shift that Ong and others (as noted above) contend occurred with the transition from orality to literacy, a transition that (like jazz’s transition to textuality) inclines us to become more committed to the distillation of “finished” aesthetic objects than the participation of praxis.

The analyses Ong and others offer of literacy help explain why this might be so.  The key element in the development of literacy (the intertwined technologies of writing and reading) is the capacity it provides to store verbal performance in a medium other than individual memory (and the complementary capacity to derive an image of the performance by processing the stored visual marks), and this capacity encourages a progressively greater awareness of our verbal performances as things in themselves rather than transitory (if powerful) gestures.  Verbal performances become potentially created objects that have a fixity beyond the dynamic of performance.  And like the shift to literacy, the shift to textuality encourages the sense that the records of our various performative acts can be scrutinized, revised, manipulated, perfected.  It provides a necessary basis, and in part the impetus, for a folk art to evolve from a process or practice into a self-conscious art tradition that preserves the aesthetic objects produced within it.

If this is so, it suggests that two factors drive textuality’s transformational capacity: one is the greater (as with writing) or lesser (as with tape recording) abstraction of the storage medium, which can gradually drop from view (both for the practitioner and consumer) as the practitioners learn to think primarily in terms of creating specifically for the medium or as the medium becomes increasingly able to capture actual performance without diminishing its texture or resorting to abstract encoding.  But there is also the way textuality itself (of whatever sort) changes the practitioner’s relationship to past gestures within the tradition by “storing” and foregrounding a particular array of works—by encouraging, that is, the view that the “art” is more a product than a process and that these products can be copied or be an occasion for further refinement or extrapolation.  The example of jazz suggests that this transformational dynamic is even more fundamental than the greater or lesser abstraction enforced by the particular medium of storage.  Textuality seems, that is, to encourage (and certainly enables) a self-conscious mixing of styles and gestures that have been learned through the study of more or less stable artifacts, and practitioners of textualized arts thereby stand in a different relationship to their materials, their audience, and their tradition than do practitioners of arts that are not textual.

The apparent impact of the capacity for textuality on the development of jazz suggests that textuality enforces a gradually greater hegemony of the compositional and that this progressively eclipses the performative.  If the focus here were literary we might be inclined to consider such pieces as The Waste Land and The Cantos as projects where (triumphantly to some) a textually supported aesthetic of composition and construction nearly effaces the personality of performance.[7] But just as the aggressively performative “Howl” demonstrated that the high modernist aesthetic was not the end of the story, part of the significance of jazz (for this study anyway) is the way its performative origins (its orality) exist in a still unresolved dialectic with the compositional capacity that the textuality of recording encourages but does not enforce entirely, and this dialectic between the impulse to value improvisation and the personal expression of performance yet move toward the distillation of recorded composition is potentially not only a way to clarify the evolution of jazz as a music but may even be a key element of what jazz expresses, since the jazz creator often seems to be negotiating a liminal space between the performative and the textual and seeking out strategies that will either express both or (in some way and if only temporarily) resolve their potential dissonance—a space evoked by the old jazz joke about the musician who interviews for a job with a big band and is asked can he dot (i.e. read the dots of a musical score) and answers “Yes, but not enough to hurt.”  And this same bifurcation between the oral and compositional can be found in the practice of the quite real and modern John Coltrane who—when handed a transcription of his famous solo on “Giant Steps” and asked to play it—replied that he couldn’t play what was notated because it was too hard (and this in spite of his ability to sight read and play intricate and demanding written exercises on his tenor).[8]

Jazz, that is, in both its history and current array of practices seems not only to help clarify the nature of textuality (in the sense of encoded storage and reproducibility) and to argue that textuality’s impact is fundamental and unavoidable once introduced, but jazz also suggests that the tendency toward construction and composition that textuality encourages does not (even over time) eliminate the emphasis on performance on which oral practices necessarily build but instead leads to a situation where the performative and compositional interact in varying ratios in the work of different individuals and traditions.  And these ratios, along with the interacting dynamics of different modalities of storage, transmission, and consumption might be said to define specific, distinctive rhetorics of textuality.


Until fairly recently the textual rhetoric of most jazz recordings has actually been relatively simple.  A Coltrane album is supposed to allow us to feel that we are participating in the spontaneity of an actual performance to an actual audience (minus the cigarette smoke, clinking glasses, and the boor at the next table who thinks he’s suave but is only loud and pretentious).  At times this requires at least some distillation (studio performances tend to be shorter than those for a physically present audience), and at times what one hears on record has actually been honed in the studio across various takes.  But even so, much of the recorded jazz legacy has involved musicians recording performances and then choosing which ones to release in a process that echoes King Oliver “waxing” his notes.  The reason for this is partly economic.  Jazz records sell modestly compared to pop music; studio time costs money, and jazz records have typically had to be produced cheaply to be commercially viable (the studio constructions of Miles Davis and Teo Macero after Davis began experimenting with fusion and became something of a pop star stand as the exception that proves the rule).  That both jazz musicians and patrons have wanted to continue to believe that the music’s significance is that it is at root performative (expressive, improvisational) is also a factor; no matter how many times we might play Armstrong’s “West End Blues” or Coleman’s “Lonely Woman,” we still want to believe the lines are being invented right then for us as we listen.  These factors may explain why textuality’s impact on jazz has tended to be more on how the music is performed and what is written for performance (with recording as transparent as the technology permits) than on experimenting with the ways recording can itself be used to construct compositions.  (Even when the technology of tape recording matured to the point that it allowed for multi-tracking and building a “performance” from pieces spliced from different takes and partial takes, the jazz recordings actually released continued to be typically integral performances with here and there a chorus or two snipped out for reasons of length or structure or, when units from several different takes were spliced together, were carefully managed to convey the illusion of an integral performance.)[9]

With pop music, though, the sense that recording was actually a way to construct a product rather than to preserve and circulate a performance more broadly took hold more quickly and completely—at least with the producers and executives pitching 45s to the teen market and looking for a hit.  These producers were not simply constructing pieces of music to sell but constructing the performers into marketable commodities as well.  As Bill Parsons (aka Bobby Bare) has it in “All American Boy” (a ditty cut right after Elvis got drafted): “Come here cat, I’m gonna make you a star,” and our tendency to see Phil Spector as the star of the 45s he produced (not the ostensible performers) is not only because he had the ego to claim he was a musical genius.  The textual rhetoric of the pop 45, then, is about as straightforward as the one governing most mainstream jazz recordings—but happens to be its opposite.

Yet as the example of The Grateful Dead (to round at last toward the beginning) suggests, not all textual rhetorics are as direct as those of mainstream jazz and 1950s rock and roll.  Of the 1960s San Francisco bands, The Dead proved not only the longest-lived but also the most committed to the improvisation and experiment of live performance, and their allegiance to praxis, their decision to emphasize the participatory give and take of the moment, meant that each dance concert could (and did) involve substantially different renderings of the musical skeletons on which they wove their performances.  Those of us who trekked to the ballrooms for the group’s amplified amalgam of jug band, blues, and bluegrass came to accept that the price of what seemed at times stunning and revelatory flights were other times when the music was (to be kind) amorphous and meandering (the band that lives by the risk of a collective semi-free-for-all dies by the risk of a collective semi-free-for-all, amen, so be it).  At issue here, though, is not so much what The Grateful Dead was like in its early years for the Fillmorites and Avaloners but rather what happened when the band came to record—when its emphasis on praxis collided with Warner Brothers’ goal of recasting that praxis as the product of salable records.

When Warner Brothers signed The Grateful Dead (in part because the media had proclaimed that a “San Francisco Sound” might be the next market trend) AM radio was the main way record companies marketed their product (FM “underground” radio wasn’t yet a factor much beyond San Francisco).  Radio listeners bought the 45s that caught their ear (and sometimes albums that included these songs), and radio programmers expected songs to be no more than about three minutes (longer songs left too little time for commercials).  It’s thus not a surprise that The Grateful Dead (1967),[10] recorded in three days, has six songs less than three minutes long, and that only the ninth and final track, the ten-minute “Viola Lee Blues,” is long enough for the improvisational elaboration that typified the group’s performances (another track, “Good Morning, Little School Girl,” fades precisely just as the band begins to modulate from the basic song into improvisational territory).  Producer Dave Hassinger’s goal seems to have been (not surprisingly) to pull some brisk pieces from the band’s repertoire and distill them into something that would have some chance of radio play—and thereby create a market for the album.  But though brief and upbeat, the track released from the album as a single for radio play, “The Golden Road (to Unlimited Devotion),” was too strange, ragged, and swirlingly kinetic to pass as a pop object.  It wasn’t “Good Vibrations” or “Mickey’s Monkey” (each in their different way perfectly constructed AM pop); it wasn’t even “White Rabbit” or “Somebody to Love” (the two instances where a 1960s bay area band successfully constructed something for the pop marketplace).  The only region where either “The Golden Road” or The Grateful Dead sold even modestly when first released was, not surprisingly, San Francisco.

In the months after this first encounter with the recording industry, the group (if anything) became even less of a marketable rock and roll band.  Expanding the group to include a second drummer added complex polyrhythms to their performances and rendered the songs themselves even more secondary, and the commitment to improvisation increased with pieces stretching out until a single improvisation at times might fill an entire several hour set.  The Dead had, that is, become even less recordable—at least if by “recordable” we mean constructing recorded product for either the market of AM radio or the market of record stores, and at one point the band declared that its first album would be its only one and that it would henceforth focus exclusively on performing.  Yet in 1968 the band (and Warner Brothers) released Anthem of the Sun.[11]

On the first album, the group’s improvisational performing had been secondary (if that) to the company’s agenda of recasting the band’s “songs” as AM product.  Anthem of the Sun, however, was constructed as if the AM market didn’t exist (or had been dismissed as simply irrelevant).  Instead of short pieces, it presents what are essentially two compositions, each taking a whole side of the album (one a medley of three “songs,” the other of two) and fashioned by splicing and layering units recorded in concert (the cover lists the fourteen performances as sources) with units generated in the studio (including pieces of “Electronic Tape” influenced by the experiments in electronic music and electronically manipulated sound then current in classical music).  The way these discrete units have been amalgamated resembles the way a film maker might use film stock generated at different points under different circumstances to fashion the illusion of a continuous action from the various bits and thus construct a scene or even an entire movie.  Ironically, the result of all this assemblage is an album that more fully evokes the band’s improvisational performing practice in this period than the first album and actually projects something of its sound at full rumble—even as one registers the electronic effects and the splices where pieces from different taped performances butt against each other (in fact Anthem of the Sun’s manipulated material probably sounds more like the band in concert than the unmanipulated concert recordings of the band’s fourth album, Live Dead).

Anthem of the Sun’s two sides are, that is, meticulously constructed yet seem to unfold as performances; they are—paradoxically—compositional representations of the group’s improvisational aesthetic, and as such they reflect an intriguing attempt to use the long playing album to mediate the compositional and performative and to fuse them.  Anthem of the Sun derives from concert material, evokes a concert, but is not a concert.  It mimes the band’s performance practice in the way it incrementally modulates and elaborates the rudimentary materials of the “songs” but has a compositional density, a kind of architecture, and a clearer sense of form than an actual performance.  In performance, the music is invented in real time as played and is meant to be heard/experienced once; on the album the music has been chosen, shaped, revised, and reconsidered and is meant to be listened to over and over.  For all that the album conveys the spontaneity of improvisation, it neither rambles nor detours.  The album might be termed a “composed performance” (as opposed to a performed composition), and it is this textual hybridity that is finally significant (and no, this is not been an attempt to canonize Anthem of the Sun as a crucial post World War II musical episode).

Without textuality, we have only performance.  When textual media entail considerable abstraction (as do both the alphabetic representation of spoken language in writing and the notation of music as dots and lines on a page when they recast transitory sound as more permanent visual codes), textuality seems to encourage an emphasis on construction and composition.  More transparent textual technologies (such as tape recording), however, tend to promote a blurring of these two capacities by promoting a compositional consciousness in oral performers and a fascination with the oral mode among at least some whose primary reference has been the compositional realm that abstract notation supports and helps enforce (it would perhaps be worthwhile in this context to consider as well the quite different experimental agendae of such composers as Partch, Cage, Feldman, Harrison, Crumb, and more recently the minimalists, perhaps especially Terry Riley).

In spite of some fine Garcia solos and stirring kazoo cadenzas, The Grateful Dead’s attempt to be both live and “memorex” on Anthem of the Sun was a commercial flop, which may be one reason why the group never tried this hybrid mode again.  Its later albums either use the studio to render songs—Aoxomoxoa, Working Man’s Dead, etc.—or are compilations of recordings of live performances— Live Dead, Europe ’72, etc.  For The Dead, that is, the solution to the polarity of pop artifact and oral performance was, finally, to treat the polarity as a given and explore its two possibilities separately.  Its communal aesthetic, a sort of tie-dyed update on those bluegrass festivals Garcia fondly remembered, may have encouraged it to see recording and the marketing of recordings as a complement to performing.  One made music because one enjoyed the process of making music and enjoyed drawing in those who came to listen—to participate.  Recordings were another outlet for the music, but they weren’t the music.  Recordings could be treated as secondary.

In part, The Dead’s resolution to the polarity depended on the willingness of those in the group not to be pop stars (that Garcia had become a kind of folk icon even before his death complicates but doesn’t contradict this) and to work out their own (eventually highly successful) sub-economy on the margins of the pop music industry.  For a figure like Jimi Hendrix whose ambition was to be a pop star, the challenge, though, was to confront the polarity between pop artifact and oral performance, do so within the context of commercial music, and do so in such a way that one somehow altered or mastered it if one was to be fully an “artist” (as opposed to being one’s self a pop product).  And for all his experience as a blues, rhythm and blues, and rock performer; his flamboyance as a performer; and his improvisational energy, Hendrix seems to have sensed that this synthesis had to be through, by, and of the recording studio.

A radio documentary on Hendrix (later marketed as a set of CDs) offers a gloss on this that can stand as a sort of coda to this blurring of composition and performance and of work and text.  At the end of the second disc Eddie Kramer, a recording engineer who worked with Hendrix on Electric Ladyland, describes the mixing of “1983…(A Merman I Should Turn to Be)”:

It [Electric Ladyland] was a double album and Jimi had much more time to develop his ideas.  One of the songs that sticks out in my mind would be “1984 a Mermaid I Should Be” [sic] which I worked on with him for about 18 hours straight.  And we mixed the entire thing—the entire side of the record—in one go with no interruptions, so it was a complete piece.  It was like a performance, and Jimi and I mixed it together, where he would grab his vocals and some of his guitar effects and I would do the drums and his other guitar effects and generally hold on to the whole thing so it didn’t fall apart.  And we’d be flying around the board like lost flies.  It was wonderful.  It was the creation of a piece of music in addition to what had already been recorded.[12]

The “board” is the mixing board through which Hendrix and Kramer could simultaneously play the different taped pieces of vocals, rhythm tracks, guitar solos, and sound effects that had previously been recorded for the piece.

Unlike Anthem of the Sun, none of the taped material for “1983” derives from an actual performance for an audience (though the strands have been created by the musicians), and this suggests that “1983” is primarily and inherently compositional: Hendrix started with an idea of what he wanted to construct; he and his colleagues generated “tracks” of tape toward this end; and he and his engineer then selected the tracks to mix into the final recording and (as the final compositional step) mixed them.  But in describing this final step, Kramer stresses that they mixed the piece “with no interruptions” in what seemed “like a performance.”  To Kramer, it seemed that he and Hendrix were actually performing a piece of music using the mixing board and taped tracks as their shared instrument.  Their decisions (as each chose which tracks to bring in and the level and tonal qualities of these tracks) were responses to the unfolding piece in response to each other and what they were discovering as they went; these were real time reactions made without stopping to deliberate and plan.  The notes to the CD derived from the documentary identify the version of “1983” that follows Kramer’s comment as “an alternate mix of a song that appeared on Electric Ladyland” and characterize it as “one of a series of experimental mixes—each totally unique…”

With a piece like “1983…(A Merman I Should Turn to Be)” the studio becomes both an instrument for performing (albeit to an absent audience) and a tool for “composing” a record, and the process of producing this piece involves, finally, a different logic than the process for Anthem of the Sun.  With “1983” Hendrix and Kramer—having previously created the material to be used—”performed” a series of mixes, of compositions, then chose one for the album.  The sides of Anthem, though, seem each to have been composed and constructed as a single mix; the mix was, that is, deliberated step by step.  The difference is like the difference between performing a sketch and constructing a collage.  The mixes of “1983” are a series of closely related sketches; each side of Anthem is a collage.  In The Dead’s experiment, acts of performance become the themes and elements of the composition, and the results might be termed a composed improvisation (the raw tracks—both those derived from concert performances and those constructed or played in the studio—have been used to represent a performance that never occurred and would never occur in quite this fashion).  However, with “1983” and Hendrix the distinction between performance and composition blurs even more and is even more problematic.  The mixes of “1983” are both our only record (in the sense of textual representation and textual object) of the “composition” (a “composition” that exists as a series of realized, stored, reproducible, and transmittable “texts”); and they are performances of it (one of which Hendrix and Kramer have privileged over the others by choosing it for the album).  The track on the album represents neither a possible performance of “1983” nor a composition called “1983” which others might perform; it is simply “1983” (in much the way a painting is simply itself, though in the case of “1983” there is no possible distinction between the “original” and “copies” as there would be with a painting and reproductions of it)—Or as Paul Simon might have sung it (but didn’t): “one man’s composition is another man’s performance” (and vice versa), and both and neither may be texts.

While neither Garcia nor Hendrix (or Buddy Bolden, for that matter) saw themselves as textual theorists, the encounters with the dynamic of analogue recording that their examples help highlight suggest that recording (especially audio tape and its attendant technologies) has a capacity to fuse performance, composition, and text into a single phenomenon in a way that the code of written language does not.  If so, the models of textual rhetoric we will need to guide the textual analysis and editing of recorded material will be different than the ones we have developed from and in response to written and printed material.  And if this is the case, it at least raises the possibility that “text,” “work,” and “performance” are more fluid phenomena than we may have realized and that these terms are less names for stable categories and qualities than markers whose meaning is historically contingent (and shaped as well by the differing dynamics of the various textual media).  And while this isn’t the place or time to launch a deconstructive reading of the famous “Is It Live or Is It Memorex” Ella Fitzgerald commercial, that could be a worthwhile effort to which the answer might also be “both” and “neither”.

[1] See The Grateful Dead, Warner Brothers (WS 1689), n.d.; Anthem of the Sun, Warner Brothers (WS 1749), n.d.; Aoxomoxoa, Warner Brothers (WS 1790), n.d.; and Live/Dead, Warner Brothers (2WS 1830), n.d.  Also, see below for a discussion of these issues in the first two of these albums.

[2] For a systematic survey of these issues and summaries of relevant research, see Walter J. Ong’s Orality and Literacy: The Technologizing of the Word (New York: Methuen, 1982).

[3] Eric Havelock’s recasting and extension of the work of Millman Parry and Albert B. Lord provides a sustained examination of this process and its cultural and social impact in the context of Homeric and Classical Greece.  His The Muse Learns to Write (New Haven: Yale University Press, 1986) both summarizes and extends the analyses he developed in such studies as Preface to Plato and The Greek Concept of Justice: From Its Shadow in Homer to Its Substance in Plato.

[4] The British rockers frequently note their interest in the blues and rhythm and blues of the 1950s and early 1960s in various interviews and fanzine stories.  In the notes to the recent CD compilation, Dave Van Ronk, The Folkways Years: 1959-61 (Smithsonian/Folkways SF 40041: 1991) Van Ronk recalls the impact of Harry Smith’s Anthology of American Folk Music (three double record sets from the early 1950s that gathered some of the key blues and country 78 recordings of the 1920s and 1930s):

The scope of this collection [Smith’s anthology] was panoramic: Uncle Dave Macon, Mississippi John Hurt, Blind Willie Johnson, Buell Kazee, Lemon Jefferson, the list goes on and on.  The Anthology was our bible.  We [the performers of the urban folk revival] knew every word of every song on it, including the ones we hated.  They say that in the 19th century British Parliament, when a member would begin to quote a classical author in Latin the entire House would rise in a body and finish the quote along with him.  It was like that.

[5] Morton’s spring 1938 Library of Congress sessions have been released commercially several times and are currently available on CD.  His illustrations of “Maple Leaf Rag” are part of the selection “Discourse on Jazz,” Jelly Roll Morton: The Library of Congress Recordings, Volume Eight, Swaggie Records (S1318), n.d.

[6] Morton’s sides for General were later acquired by Commodore Records and have been reissued several times and are currently available on CD.  Jelly Roll Morton, New Orleans Memories Plus Two, Commodore Records (XFL 14942), 1979.

[7] It is certainly true that these poems can be read aloud (and both Eliot and Pound “performed” them for audiences), but Eliot’s insistence on the impersonality of art and his various strategies for suppressing narrative and deflecting any identification of the poem’s language with a stable “voice” (especially his own) suggests the extent to which his “performances” were distinctly secondary to the poem’s status as composed and fixed artifact.  Similarly, Pound’s role in helping Eliot distill “The Waste Land” and his decision to discard the so-called “ur-Cantos” (which feature the sort of narrative voice, a persona in which we are invited to take a dramatic interest) suggest his rejection of the oral and performative (in spite of his fascination with pre-technological culture).

[8] Somewhere in my years of reading album liner notes, articles in Down Beat and the like, and books on jazz I ran across this anecdote but do not at this point know its source.

[9] Some of the tracks on Charles Mingus’s Mingus Ah Um (Columbia CL 1370) have had, for instance, material deleted (a solo or two here, a bridge or chorus there), and it isn’t clear who decided to snip these pieces out or why (because of length? to improve structure? something else?).  What is clear is that the listener is expected to listen to this processed performance as if it were simply an unprocessed performance (unedited versions of these tunes were released after Mingus’s death in Nostalgia on Times Square, Columbia JG35717).  The pieces on Miles Davis’s Miles Ahead (Columbia CL 1041) are the result of even more radical manipulation, yet here too for the listener, these constructions (a complex pastiche of spliced takes and overdubs) play as if they are integral performances.  Phil Schaap’s painstaking analysis of the construction of these tracks is included in the booklet to the Mosaic Records LP version (MQ11-164) of the just released Sony/Columbia CD set, Miles Davis/Gil Evans: The Complete Columbia Studio Recordings.

[10] The Grateful Dead, Warner Brothers (WS 1689). n.d.

[11] Anthem of the Sun, Warner Brothers (WS 1749), n.d.

[12] Lifelines: The Jimi Hendrix Story, Reprise (9 26435-2), 1990.


Back to Critical Work