Past Views of the Phonographic “Art”

One more obstacle impeding listening-based research into early phonographic practice has been a belief among influential critics that all the techniques that now make phonography a distinctive “art” worthy of study and appreciation did not come into existence until decades after the period I will be discussing here. In order for us to understand the origins of this belief and the perspective it reflects, it will be informative for us to compare and contrast the analytical treatment of phonography with the treatment of the motion picture. Whereas phonography has been widely equated with recorded music, classic film theory vehemently denied the parallel assertion that cinema is merely filmed theater. As Victor F. Perkins puts it:

The danger was that if the movie were not shown to be an extension of visual art, it would be seen as a corruption of drama. It would be exposed as ‘canned theatre’, drama without the power of speech and thus deprived of its most powerful resource…. ‘Theatrical’ became, and has remained, the most contemptuous adjective in the theorist’s vocabulary, being used to indicate that the filming has added nothing to the recorded event…. [The theorists’] model, fine art, imposed the view that the real scene or human figure had no relevance; what mattered was the way it was rendered in paint and marble, or on film. The resulting dislocations can be seen in the theorists’ inability to find the recorded action a place in the critical scheme or to allow it any artistic status.⁷⁰
The derogatory epithet “canned theatre” recalls “canned music,” a phrase John Philip Sousa popularized a century ago for lambasting the recording industry as a force corrupting American musical culture.⁷¹ In the case of cinema, visual art existed as an alternative frame of reference, a prestigious mode of expression with which film theorists could associate what filmmakers did, as opposed to what their machinery—or the subjects it recorded—did. Classic film theory accordingly foregrounded those areas in which cinema was furthest removed from sheer mechanical reproduction of a “theatrical” subject: camerawork and/or montage, depending on the theorist. In phonography it was music itself, a kind of recorded action, that seemed to offer the greatest prestige to the medium, notwithstanding the resistance of critics like Sousa. This led to marketing strategies during the early twentieth century in which “fidelity” with respect to highbrow music was advanced as the principal criterion of phonographic value—a campaign that was highly successful not just in expanding the market, but in shaping subsequent discourse about the medium as well.⁷² Because early phonography was less “faithful” to its subjects than later phonography, its products are less “valuable” when measured against this yardstick; and, again, if phonograms simply “reproduce” music, there is little to be said about how they represent music.

An alternative trajectory in phonographic theory has involved identifying and emphasizing phonographic equivalents to the cinematic practices of camerawork and montage. This approach was pioneered by theorists of film sound, whose goal has been to explore the artful use of recorded sounds—dialogue, music, sound effects—as a component of cinema. At first, film theory had treated soundtracks as a matter of transparent reproduction, not representation: sounds did not “represent” their originals, but were effectively identical with them. However, Alan Williams challenged this view in an article of 1980, suggesting that recorded sounds ought to be approached in the same way as recorded images, as representations rather than reproductions. Much as a flat image differs from its three-dimensional subject, the argument goes, so phonography records a three-dimensional complex of vibrations from a single vantage point. In Williams’ view, recorded sounds are distinguished from their originals by their subjective two-dimensionality, such that phonography “implies by definition a reading, a deciphering, an attending to a sonic event.”⁷³ Williams accordingly identifies the placement and strategic use of microphones as a signifying practice on a par with creative camerawork. Once recorded, he adds, phonograms are also subject to the same kinds of postproduction editing as images: “all manipulations possible in image recording have analogs in sound. There are sound edits, for example, as well as dissolves, super-impositions, and so on.”⁷⁴ He concludes that “if we ask whether messages of some sort are…transmitted from a source to a receiver via sound recording without directly depending on those languages simply relayed by the apparatus (spoken language, music), the answer must be yes.”⁷⁵ On the whole, Williams makes a compelling case for treating phonography as a system with its own communicative conventions and resources over and above its recorded subject matter. However, it is important to note that in his scheme the originals of recorded sounds are regarded as untransformed reality. It is only through what happens to them through the process of recording and afterwards that phonography adds its special layer of meaning.

Williams focuses on the use of phonography in conjunction with film, but some critics whose interest lies in phonography on its own—“sightless” phonography, we might say—have pursued a similarly filmlike approach to their subject. This is how Evan Eisenberg introduces his chapter on phonography, which for him is “a new art, the art of recorded music”:⁷⁶

The word “record” is misleading. Only live recordings record an event; studio recordings, which are the great majority, record nothing. Pieced together from bits of actual events, they construct an ideal event. They are like the composite photograph of a minotaur. Yet Edison chose the word deliberately. He meant his invention to record grandparents’ voices, business transactions and, as a last resort, musical performances. The use we put it to now might strike him as fraudulent, like doctoring the records.⁷⁷

Eisenberg tacitly assumes that music is uniquely appropriate as subject matter for phonography, but another critic, Douglas Kahn, disagrees, proposing instead a future avant-garde “art phonography” based on overt mimesis and filmlike editing techniques. While he approaches phonography with very different aesthetic preconceptions than Eisenberg does, he too sees a “developed artistic practice of phonography” as necessarily predicated on splicing and montage. Like Eisenberg, he concludes that such a practice could not have existed until long after the period I will be covering here. Since early phonography did not involve splicing, from this point of view it was not an “art” in its own right, the argument goes; it was missing the one tool that makes creative phonography possible today. In his historical overview, Kahn claims that musical notions exercised a stranglehold on early phonography, barring a few “scattered novelties,” and he instead confines himself to considering why recognized avant-garde artists of the past did not pursue phonography, or why “there hasn’t been a phonographic art” yet. Elsewhere Kahn does acknowledge that phonography has been applied to subjects other than music, but he asserts that those uses have not lived up to the normative standard he is proposing:

The problem has been that phonography has not migrated over the expanse of sound, but has been limited to the reproduction of existing aural cultural forms—music, poetry and literature, theatre, reportage—when it could reproduce all these forms at once, inhabit their conventions and break them open to the general aural environment. In the audio-visual forms of film and video, sound recording has suffered from a subsumption under the visual and, within a hierarchy of sound itself, of a full range of sounds under speech.
In Kahn’s scheme, there seems to be no middle ground between the unimaginative recording of existing art forms and the programmatic shattering of generic boundaries. Kahn uses his bleak verdict on the phonographic past as a source of optimism for the future: “It signals an expanse of artistic possibility in a situation where other arts battle exhaustion.”⁸² His stance may provide inspiration for a new phonographic avant-garde more to his liking, but, like Eisenberg’s, it discourages one from expecting to find anything interesting in the actual practice of early phonography.

In terms of the “art” as Eisenberg and Kahn define it, they are largely correct about the pioneer recording era, though not entirely. Early phonography did involve some practices analogous to camerawork, among which the most important were the placement of recording horns, selection of diaphragms, and modifications to the acoustic environment of the recording studio. The importance of these practices, especially in helping to establish the professional legitimacy of the expert recordist, should not be underestimated. Rather than turning to cinema, which was not yet available as a point of reference, some early recordists drew legitimizing analogies between what they did and the art of photography:

Because many semiotic properties of secondary orality were traditionally unique to “live” enactments, they were long identified with them and invested with meaning accordingly. In particular, hearing a particular voice was once a sure sign that the person belonging to the voice was present and speaking; once phonography had appeared, such correspondences could no longer be assumed, and the resulting uncertainty, confusion, and paradoxes have been an enduring source of fascination. Some scholars have approached issues of this kind through Jacques Derrida’s critique of what he calls “phonocentrism,” the tendency of Western philosophy to operate in terms of a binary opposition between the speech of an immediately present speaker and the written text and to privilege the former over the latter. The phonograph challenges the speech/writing dichotomy by making the voice separable from presence and incorporating it into a new kind of independently circulable “text,” the phonogram.⁸⁹ A related concept in soundscape studies is schizophonia, which R. Murray Schafer defines as “the split between an original sound and its electroacoustical transmission or reproduction.” The term is modeled after schizophrenia; as Schafer explains, “I employ this ‘nervous’ word in order to dramatize the aberrational effect of this twentieth-century development.”⁹⁰ Unlike Williams, Schafer treats “reproduced” sounds not as two-dimensional representations, but as indistinguishable from their originals; this makes it all the more alarming that they “have been torn from their sockets and given an amplified and independent existence.”⁹¹ Whether the “split” of sounds and voices from their points of origin is regarded as a modern aberration (Schafer) or as further evidence that the differences between speech and writing have never been quite what they have been cracked up to be (Derrida), it has been a prominent part of the discourse surrounding phonography and so warrants consideration.

Unlike the filmlike approach to the phonographic “art” championed by Eisenberg and Kahn, the concept of secondary orality and the various arguments about “splits” between original sounds and “reproduced” ones seem potentially relevant to much of early phonography, and not just the phonography of later periods. At the same time, they must be handled with caution; we cannot simply apply them in cookie-cutter fashion to the history of the phonograph and expect to yield interesting or accurate results. Ruth Finnegan devotes much of her book on Literacy and Orality to rallying studies of particular cultural practices that contradict or complicate generalizations made by Ong and other proponents of grand theories. “These detailed studies remain essential,” she writes, “revealing as they do the specificities and contingencies, rather than the general laws, of human action.”⁹² However, she laments the lack of attention to what she regards as potentially illuminating transitions in the use of technologies of communication:

Amidst all the speculation about consequences of literacy, why are there so few studies which investigate the consequences of orality, or of the loss of literacy, or of a choice to use one rather than another? What about the changing balance and possible consequences when people rely less on written communication and more on oral forms? What happens when telephones are increasingly used for communication rather than letters….? If the implications of using different media of communication are indeed significant, then it is surely worth looking at the differing combinations and changes between them, rather than—as often in the implicit technological determinism model—just at apparently unidirectional and ‘natural’ progress based on ‘ascending’ technologies of communication.⁹³

I find that the language popularly used to talk about phonography, which has been built up over time through a gradual process of accretion, can become more of a hindrance than a help when we try to use it analytically. One of the greatest challenges I have encountered in trying to write about this subject has been that of developing an analytical vocabulary that is sufficiently flexible and neutral to accomplish what we will need it to accomplish.

The arguments I have surveyed so far reveal that there is some room for debate as to just what kind of relationship exists between recorded sounds, original sounds, and the sounds that come out of phonographs: for instance, does phonography “split” sounds from their sources, or does it “represent” them, or what? A working definition of phonography, and the vocabulary we use to talk about it, ought to be sensitive to uncertainties of this kind and allow various possibilities to be explored rather than imposing one or another a priori view. Here Jonathan Sterne offers a promising starting point. When trying to find a neutral way of defining “sound-reproduction technologies” in general—a term that, in my opinion, is itself already problematic—Sterne settles on the criterion of transduction: modern sound media such as the telephone, phonograph, and radio “turn sound into something else and that something else back into sound.” Furthermore, they transduce based on a principle Sterne refers to as tympanic, because it was initially modeled after the effect of sound waves on the tympanic membrane of the ear or “eardrum.”⁹⁴ Much as the eardrum passes the vibrations that converge on it along to the inner ear, the diaphragms of phonographs and telephones transduce them respectively into phonograms and transmissible signals. For this part of the process, I see no problem in employing the usual verbs record in reference to the phonograph and transmit in reference to the telephone, although the principle in general is sometimes known more abstractly as input transduction. This “drawing-in” aspect of transduction does not present any serious theoretical difficulties.

The problem is what to call the second process, the “drawing-out” (or output transduction) in which something other than sound is converted tympanically into sound. That process has usually been called reproduction: a telephone receiver or phonograph reproduces a person’s voice, and the technical name for the component that transforms a phonogram into sound is the reproducer. But Alan Williams and others who have followed his lead hold that phonography never “reproduces” originary sounds but instead subjectively represents them, an objection that makes a more neutral term desirable—even though these same critics do continue to use the word “reproduce,” apparently for want of a viable alternative. A more practical problem is the ambiguity inherent in statements such as “Mr. Edison is reproducing the phonogram,” which could mean either (1) that he is playing it, reproducing the sounds it embodies; (2) that he is duplicating it, making extra copies of it; or even (3) that he is producing a new phonogram from scratch in imitation of an earlier one. Finally, in situations where several phonograms have been edited together into an “ideal event,” it would be misleading to speak of the results being reproduced when they may never have been produced as such before, quite apart from the question of whether phonographs ever reproduce sound at all. The alternative term playback, or to play back, is open to the same criticism, insofar as its “back” implies an anterior “playing”; and output transduction has the practical disadvantage of requiring two words—a clunky way to express so fundamental a concept. Because of these objections, I prefer the verb educe, which has the more appropriate definition “to bring out, elicit, develop, from a condition of latent, rudimentary, or merely potential existence,” and the derivative words eduction and eductive.⁹⁵

Distinguishing eduction from duplication, rather than calling both processes “reproduction,” means reconsidering some popular generalizations that have been made about mechanical media in the past. In his influential essay on the work of art in the age of mechanical reproduction, Walter Benjamin treats the “reproduction” of static art objects and of live performances as equivalent acts. “Even the most perfect reproduction of a work of art,” he writes, “is lacking in one element: its presence in time and space, its unique existence at the place where it happens to be.”⁹⁶ Thus, a perfect reproduction of a famous painting is not that painting, and a perfect reproduction of a live speech is not that live speech. What is missing from the reproduced version, in each case, Benjamin calls the “aura” of the original, a term he associates with ritual value. Simultaneously, reproduction eliminates “distance”: people can experience these famous paintings and live speeches in their homes, where the originals either would not have been permitted to go or would not have taken place. However, there is a distinction to be made between the ability to “reproduce” a recorded enactment by educing a phonogram, and the ability to “reproduce” a phonogram by making duplicates of it. Multiple eductions of a unique phonogram might be compared to multiple viewings of a painting, rather than to copies of the painting. They might not, for that matter, but making the distinction at least allows us to articulate and explore different possibilities.

The word “educe” also helps reveal certain continuities that the word “reproduce” would tend to mask: for instance, a musical box might be said to “educe” the sounds encoded on its manually pinned barrel, but not to “reproduce” any originary enactment. Other examples of automatic “eduction” could include the projection of a motion picture and the running of a computer program—indeed, I am not quite sure what the boundaries of this concept might turn out to be. At the same time, we gain an opportunity to sharpen our understanding of the real difference between the eduction of a phonogram and these other categories of eduction. We can define tympanic eduction, as opposed to eduction in general, as artificially causing a surface to vibrate as though it were a point through which atmospheric sound waves were passing and thereby introducing a sound wave based on these vibrations into the surrounding atmosphere. In turn, I will define phonography as tympanic transduction between sounds and inscriptions and telephony as tympanic transduction between sounds and other signals transmissible across distances. Note that none of these definitions mentions “reproduction,” though none of them rules it out either.

Previous historians of the phonograph have overlooked Edison’s keyboard talking telegraph project, apparently failing to see in it any relevance to the subject at hand. Perhaps the idea behind it is simply too far removed from the way in which critics understand the phonograph today for the connection to be apparent. However, the evidence leaves little doubt that Edison first envisioned phonography as something useful not in terms of faithful “records” of past events that could be “reproduced,” but in terms of programs that could synthesize speech in the future at the push of a button—a handy new technique for generating complex sounds mechanically. Granted, the programs had to be tympanically recorded; there was, as yet, no other known means of inscribing the patterns of vibrations necessary to educe, say, the sound of the letter “a.” From the beginning, however, we find phonography linked to tasks that were not perceived primarily as “reproduction,” even if they happened of necessity to involve “reproduction.” Rather than thinking of all phonograms as “copies” of “originals,” it will often be more illuminating to understand them as programs waiting to be enacted, their social significance rooted in moments of eduction rather than in the moment of recording—which, in turn, has far-reaching implications for the nature of the originary events that still formed a necessary part of the process.