The Passing of the Circus Parade, by Len Spencer and Gilbert Girard!
Acoustic Recording and Phonographic Transcription
Download 256.87 Kb.
|
Acoustic Recording and Phonographic TranscriptionBefore we get underway, there are still a few practical points we will need to consider with regard to representing phonograms on the printed page for purposes of quotation and analysis. Until the 1920s, nearly all phonography was based on what is known as the acoustic recording process, as opposed to electric. In acoustic recording, the palpable mechanical force of the sound vibrations themselves drives the movement of the recording stylus. Thomas Edison once illustrated the mechanical force of sound waves by inviting a newspaper reporter to place his hand on the top of his silk hat and “feel the rumble and the roar of the mighty city.” The reporter obliged and acknowledged that “sure enough it thumped away as though a living heart were beating against it.”131 Edison even devised a “phonomotor” which harnessed the mechanical power of the human voice to turn a small wheel attached to a gimlet—during an exhibition of 1878, someone had asked him if he could “talk a hole through a board,” and the design for the phonomotor was his immediate answer to the challenge.132 Certain sounds did not record well under the acoustic recording process. This had an effect on musical sounds, as seen in the arguments of Katz and Lastra, but it also impacted the recording of speech sounds. In 1878, William Preece remarked of the phonograph: “The s for instance at the beginning and end of a word is almost entirely lost—is entirely lost, although it is heard slightly in the middle of a word. The d and the t are exactly the same; and the same in m and n, mane and name are not distinguishable.”133 Soon after, he noted that both the phonograph and telephone had difficulty transducing sibilants: “Thus, if through the telephone you ask a person to ‘waltz,’ it will come out ‘walk,’ and names like my own, with the sound of ‘s’ in it, would come out ‘Pree,’ not ‘Preece.’”134 In the late 1880s, Edison’s improved phonograph was supposed to have initially “refused to say ‘specia’—it dropped the ‘s’ and said ‘pecia.’”135 “Well, I have about solved the problem now,” Edison claimed in 1889, “and the sound of ‘s’ is inscribed with the other letters.”136 However, the results were still imperfect. The phonogenic performer Russell Hunting noted some remaining flaws in an interview of 1903: “Are there any words,” I asked, “which present greater difficulties than others?” “There are lots. But two only will serve to prove my point. Just notice how the ordinary man enunciates ‘truth’ and ‘teeth.’ You know what he says, because you are watching his face. But let him talk into the trumpet. And what do you hear when you switch on the reproducer? Truth becomes merely an ‘oo’ sound, teeth an ‘ee’ sound; it is impossible to recognize them as words.”137
One result is considerable uncertainty over the words heard today on early phonograms, which can make transcribing them more of an art than a science. Debates have arisen over the correct interpretation of particular examples of acoustically-recorded speech and singing. Rumors still abound that Geraldine Farrar sang not “si, per la vita” but “he’s had a highball” in her performance with Enrico Caruso of the love duet from Madame Butterfly.139 The opening words of an 1889 cylinder documenting a piano performance by Johannes Brahms have been the source of another controversy, at issue being whether the composer is identifying himself in the first person or being introduced by someone else—and in what language.140 The spoken segment towards the end of a 1904 take of Turkey in the Straw by Silas Leachman has phonographicists disagreeing as to whether he shouts “Mister Booth,” addressing Victor Talking Machine Company house pianist C. H. H. Booth, or “Mister Coon,” consistent with the blackface genre.141 These are merely some of the most conspicuous examples of a general problem. Robert Cogswell comments on the difficulties he experienced in transcribing early “blackface” dialogue phonograms: Once the researcher has obtained a copy of some rare and long sought-after recording, poor sound quality, due either to the inadequacies of earlier technology or to the ravages of time, may render portions of the recorded text indecipherable. Whereas the folklorist who transcribes and studies his own field recordings can rely on memory to reconstruct all aspects of a performance, the student of commercial records lacks any recall from the collecting situation.142
This uncertainty over wording is not unique to modern-day researchers working with badly worn materials, nor does it necessarily imply that our ability to analyze them is flawed. Rather, lexical indeterminacy has been a part of phonographic listening from its beginnings. One of the first pieces of fan mail ever received by a recording artist—William Hooley of the Haydn Quartet, in 1899—centered on the following request: In “The Chapel,” I fail to distinguish all the words and my object in writing you, partly, is to ask if you would be kind enough to send me the words as you sing them and the composer’s name. I have a song by the same name, but the words are entirely different, nor can I find among my friends, any who know the one you sing, or I would not trouble you.145
Only as it became obvious that the phonograph was a failure as a “talking” machine did a few adventurous (and probably desperate) investors begin to reconfigure it as a “singing” machine.146 Business dictation required clear, intelligible records that the early phonograph simply was not capable of producing, at least not without careful attention from the machine’s operator. Music, ironically, was in some ways well-suited to the phonograph’s limited sonic range and high levels of noise and distortion. Listeners often knew the words to songs already, or could recognize the melody of even a badly recorded song. Then, as now, it was not usually necessary for the recording of a song to be perfectly free of scratches, hissing, or distortion for it to be thoroughly enjoyable.147 When the goal of phonography was to produce written business correspondence, lexical indeterminacy was an obvious drawback. For instance, when George Gouraud, Edison’s agent in London, sent phonographic correspondence overseas to America during 1888 for purposes of experiment and publicity, he enclosed transcriptions prepared by his assistant H. de Coursey Hamilton as a backup, and the transcriptions contain a number of passages in which Hamilton had been unable to make out all the words and marked his guesses as uncertain, yielding relatively untidy documents: I sent you a cable immediately explaining to you that you are wrong in supposing that ----? deter ? you. You have last said ?, we should have the necessary means of making three complete machines.148 Each Phonogram will contain a message to you, informing you the route by which it has reached you. The bait one has left me, & consequently when you receive it, you will be able to judge, as to the effect, if any, of the various stages of private & conditions of handling, incident to the voyage which it has made.149 However, lexical indeterminacy did not necessarily make phonography inferior to writing as a medium of language in all cases, even if it made transcription difficult. Recorded utterances that cannot be understood or identified with certainty as specific words often still have other appreciable qualities, such as the ones Alan Durant identifies with secondary orality: paralinguistic and prosodic features, accent, intonation, vocal quality or “grain.” Indeed, much of what made the phonogram formally distinctive relative to the written or printed word was its embodiment of these qualities. Durant goes further, arguing that in commonly-encountered fast-tempo speech or melismatic singing under conditions of poor audio fidelity, a range of provisional senses are liable to intervene in hearing the spoken text, putting unusually to the test the ‘redundancy’, or surplus of identifying cues which exist in spoken communication. Effects of temporary polysemy of this kind amount to more than mere stylistic felicity, and can in fact create a potent—and often psychically invested—sub-script which it is left to following sounds to cancel out.150 The possibility of mishearing words through the phonograph was, minimally, tapped as a source of humor. In 1879, a tinfoil phonograph was reported as having garbled the phrase “In heaven y-clep’d Euphrosyne” into “In heaven she crept, and froze her knee.”151 In 1908, a customer entered an Edison dealer’s shop asking about a song called “Harry Warner” he had heard educed on some other company’s equipment. The song turned out to have been “Arrah Wanna,” and the customer obligingly concluded “that if he wanted good music and be able to get all the words right, he must have an Edison.”152 The relationship between the sounds a telephone mediated and what a speaker had originally uttered were similarly subject to playful reportage: What Beattie thought he said: “This is William J. Beattie. I want to get my bell ringers.” What the telephone made Beattie say: “Sizz Wibbum Jbeedy swantttergit—br—ingers whrrrrr.”153 Mr. Watson was asked to repeat some phrase loudly and slowly a number of times. The phrase was announced to be, “Do you understand what I say?” What came from the boxes was, “Oo, boo, boobooboo, boo, boo, boo.” Mr. Watson next tried to say “How do you do?” but only succeeded in transmitting “boo, boo—boo, boo.”154 Lexical indeterminacy may even shed some light on the uncertainty over the wording of the famous first spoken message transmitted by Alexander Graham Bell to Thomas Watson over the telephone on March 10, 1876. Bell wrote that he had shouted “Mr. Watson—come here—I want to see you.” Watson himself recorded the phrase as “Mr. Watson come here I want you,” with “come here” inserted with a carat.155 While this discrepancy might be attributed to faulty memory, it is also likely that what Bell said and what Watson heard were not identical on this occasion. If the telephone had failed to transmit some of the words intelligibly, then which part of this communication (if any) can be regarded as authoritative: the words Bell spoke, the imperfectly mediated sounds, or the message Watson understood? More practically: when transcribing a recorded phonogenic performance, enacted purely for recording, should the ideal be to transcribe what the person originally spoke into the phonograph horn, what part of it was recorded, or what we actually hear? For better or for worse, we are usually stuck with the last of the three options by default. The problem is that a listener’s ability to make out the words from a phonogram is often due just as much to cultural conditioning as it is to acoustics—for example, understanding a given set of sounds as “Hanna” or “Hannah,” as “Harry Warner” or “Arrah Wanna.” What was a hundred years ago a witty reference to some commonly-known subject may today be unintelligible, not because we are hearing anything different acoustically, but because the intended word or name is not part of our vocabulary. “I’d like to see the Sandow that handles that pick,” says an Irish character in one version of Steamboat Leaving the Wharf at New Orleans.156 Anyone who did not already know that Eugen Sandow (1867-1925) was a famous body-builder would not only fail to understand the reference, but would probably be unable to give a confident phonemic rendering of the name—or the word, since it is not clear from context that it is a name. Except for the joke that follows—the Irishman is indicating the boat’s anchor and thinks it is a giant pick—the transcriber might also easily mistake the final word for “pig.” At the same time, the distinctive stage-Irish accent comes through loud and clear. One word that gave me pause for thought in my opening transcription of The Passing of a Circus Parade was pageant, which Len Spencer twice pronounces “PAY-geant”; I was unfamiliar with this mostly archaic pronunciation and was only confident about what I was hearing once I had found it listed as an alternative in the Oxford English Dictionary.157 Another word I thought might be either “silver wire” or “single wire,” though neither seemed quite right; only when I heard a different take of the selection issued on Columbia did the sounds finally resolve themselves into “slender wire.”158 As another example, Jonathan Sterne takes as one of his epigraphs the quotation “That shows that the phonograph can be…for a very long time,” identified as “THE VOICE OF JESSE WALTER FEWKES on a test cylinder, ca. 1890, as heard ca. 1980.” What amuses him is the ellipsis. “In a manner both geologic and poetic, some of Fewkes’s own ruminations on the preservative power of sound recording have eroded from the surface of his own recording,” he comments. “The transcribing engineer could no longer hear what he had to say.”159 Actually, this example is open to multiple interpretations. The phonogram has probably degraded over time, but perhaps Fewkes’ recorded words were no more intelligible when the cylinder was new, as Hamilton had found when trying to transcribe Gouraud’s dictations in 1888; the issue is not necessarily one of permanence. There is also the issue of subjective hearing: if Sterne had listened to the phonogram in question, would he have been able to make out the words? Would he or the unnamed transcribing engineer have understood “I’d like to see the Sandow that handles that pick” or “this PAY-geant of all its glorious magnificence”? The subjectivity of such transcriptions must always be borne in mind. In light of these considerations, I have felt it important to adopt some textual convention for indicating greater-than-usual uncertainty over wording in my transcriptions of early phonograms. Robert Cogswell explains his practice as follows: “The omission of words, phrases and sentences which are unintelligible because of poor sound quality is noted in brackets. In some of the dialogues I have deleted the words of song fragments which were especially difficult to understand.”160 Whenever I have been unable to make so much as a guess at the words being spoken at a particular point, I have noted it as [unintelligible]. But usually I can hazard a guess, and in those cases I have marked uncertain passages with dotted underlining. Of course, other words about which I felt confident when making my transcriptions could also be naïve mistakes on my part, in the sense of not being what the phonogenic speaker meant to utter. Some aspects of my transcriptions are almost always based on guesswork. Rarely is it possible to distinguish “said” from “says” on a worn phonogram other than from context, and often not even then. Unfamiliar proper nouns are especially difficult: in my opening transcription of The Passing of a Circus Parade, the name “Manzilvosky” is almost certainly “wrong,” but it is the closest I am likely to get. Another factor complicating phonographic transcription is the need to deal with the artful use of dialect. Cogswell writes of his transcripts of “blackface” dialogues: “The transcriptions use ‘eye dialect’ to express, as accurately as possible, the actual pronunciations of the comedians. The many inconsistencies reflect real variations in the Negro stage dialect; even the same comedian often altered his pronunciation from one line to the next.”161 However, “eye dialect” itself involves a set of conventions that overlap only in part with those of spoken dialect, and that have connotations of their own. I will discuss this more later on; for now, suffice it to say that this transcription practice is never completely neutral, but that leaving dialect out of transcriptions would be a distortion too. There is the further question of how best to format phonographic transcriptions on the page. In an article of 1994, David A. Banks proposes a method for transcribing early speech recordings by which he seeks to reflect the speaker’s “unique rhythmic speech pattern” by dividing the text “so that each line ends at a rhythmic beat in the speaker’s delivery.” His goal is to produce texts in which it is easy to read along while listening, and the example he gives is a transcript of an 1888 phonogram of the voice of Sir Arthur Sullivan, one representative segment of which runs as follows: DEAR MISTER EDISON, FOR MYSELF I CAN ONLY SAY THAT I AM ASTONISHED AND SOMEWHAT TERRIFIED AT THE RESULTS OF THIS EVENING’S EXPERIMENT.162 As a means of entextualizing one aspect of performance style, Banks’ practice is similar in its conventions and rationale to Dennis Tedlock’s use of line breaks to represent pauses in the delivery of Zuni oral narrative. Tedlock asserts that “prose has no real existence outside the written page,” and that it is “unfit for representing spoken narrative” because “it rolls on for whole paragraphs at a time without taking a breath: there is no silence in it.”163 Phonograms of speech, like Sir Arthur Sullivan’s, likewise incorporate pauses. While there is no reason to regard all spoken-word phonograms as “poetry,” I will nevertheless format all transcribed phonograms in this way, first as a means of drawing attention to their aurality, and second because there is no compelling reason to format them in any other way. In general, I have divided transcriptions into lines based on longer pauses which may or may not correspond to taking breaths. Occasionally, when a shorter pause conspicuously affects the rhythm of a particular line, I have indicated this by an m-dash (—), and I have shown “latching,” or the juxtaposition of speech by two speakers without pause, with an equal sign (=). I have also indicated conspicuously emphasized words or syllables through the use of italics. These efforts do not, of course, represent the full load of prosodic and paralinguistic features found in the phonograms themselves, but I regard them as a minimal effort to entextualize those details for which conventions are easiest to establish. Apart from matters of transcription, another textual convention I have adopted is the use of small capitals to indicate titles of individual phonograms, e.g. Uncle Josh on a Street Car, and italicized small capitals to indicate larger units such as albums, e.g. Emile Berliner’s Gramophone: The Earliest Discs, 1888-1901. I have several motives for adopting this practice, apart from the precedent set by some publications on early cinema.164 First, I want to distinguish clearly between compositions and their recorded-sound manifestations. “The Preacher and the Bear” refers to a musical composition, whereas The Preacher and the Bear refers to a phonogram centered upon it, including any announcements, spoken interludes, and sound effects such as—in this case—the growling of a bear. In any case, I will be referring to specific phonograms very frequently, and multiple quotation marks tend to produce cluttered-looking sentences. There is also the troublesome issue of discrepancies between the way a title appears written on a label, printed in a catalogue or discography, and in a spoken announcement; in my citations, I have used the version of the title appearing earliest in this list. Finally, whenever I have drawn my conclusions from listening to a phonogram, I have indicated this in the endnotes with the mark §. When this mark is absent, it means I am basing my conclusions on the visual inspection or mere existence of a phonogram. * * * * * It remains for us to preview the contents that will make up the body of this thesis. Chapter one covers the first public demonstrations and exhibitions of Thomas Edison’s phonograph in 1877-78. During this time, phonograms were typically recorded and educed on the spot, and whatever the audience heard back from a phonograph it had generally just heard “live.” However, these events were far from simple as sounds were educed at different speeds or backwards, layered one over the other into elaborate montages, and otherwise manipulated to create novel aural and linguistic effects. As we will see, the practice of phonography did not start out as mere “reproduction” and become complex only later; rather, it proves to have been extraordinarily complex from its very beginnings. Chapter two takes a fairly traditional approach to the history of the early commercial recording industry in the United States, although much of the specific information presented is new. It begins by surveying the popular speculation of the 1870s into how the future industry might unfold and then describes the key technological and business developments through which it first became a reality. It also treats in some detail the practices on which the industry was founded: the imperfect methods of duplication that restricted the number of copies any single master phonogram could generate and the new arts of sound recording, phonogenic performance, and phonographic eduction. In chapter three, I expand upon the linguistic side of phonogenic adaptation. To introduce the idea that distinctive speech conventions have evolved over time to fit the special constraints of sound media, I trace the origins and early development of two relatively familiar examples: the use of “hello” in telephony and the wording of outgoing telephone answering machine messages. The second half of the chapter deals with a less well known phenomenon, the formulaic spoken announcements with which commercial phonograms typically opened until being phased out between 1903 and 1908. I find that medium-specific speech conventions such as these initially helped users to orient themselves to unfamiliar new media which might otherwise not have “worked” properly. However, they invited critical reevaluation once these same media had been successfully integrated into social life, pitting ingrained but now supposedly unsophisticated habits against a new ideal of transparency and immediacy. Download 256.87 Kb. Share with your friends:
|