Back | Contents

Thinking In Sound

As soon as I began to work with the spoken words the script began to change. Radio producers quickly learn that words that look good on paper can fall flat when actually heard, that what reads well does not necessarily sound well. A script composed from transcriptions can, and usually does, undergo many changes before it works in sound. As I listened through and timed the cuts it also became clear that the material was so rich I was going to have a hard time cutting out important actualities. At three to six minutes per single-spaced page, our 28-page script would be well over two hours, a bad length for radio broadcast. The need to cut wonderful material to fit into standardized radio time slots is something that all documentarians producing for broadcast must do. But our primary audience was academic, not general, and our primary carrier a CD, not radio. While editing the material and laying tracks it quickly became apparent that contorting the essay into half-hour blocks of time would destroy its integrity. As I continued to block out segments, I abandoned any effort to break the essay into 28-minute segments. Soon, too, the essay began to take on a life of its own. Interview segments that read great on paper proved boring, redundant, or insignificant when heard. Words that on paper appeared insubstantial, when heard carried unexpected meaning, power, and resonance. Some segments grew in length, others contracted.

The act of authoring in sound is different than that of writing, for the building blocks are not individual words, but recorded sounds and voices. As I began to work through the script some of the movements almost authored themselves, but I was finding the beginning of the essay hard to follow. Even worse, it wasn't holding my interest. Creating the bridges that lead the listener from one segment and idea to the next was also proving difficult. On paper, authors have free reign to analyze, draw conclusions, explain transitions, and create topic sentences to set up what follows. But unable to use a narrator reading a script I was finding it very difficult to segue between important actualities. I resolved the problem through the use of sound effects—the bark of a dog, closing of a door—or through musical bridges. Conversational asides were also used to demarcate a change of topic. But the ordering and analysis necessary for listeners to understand the flow of ideas was often just not there. I have often found this to be a problem with many, narrator-less actuality-based long-form sound pieces. In writing the initial script, Alessandro and I had minimized our use of tape from the June 15, 1997 interview that I had conducted with him. Alessandro wanted to keep his presence to a minimum, in part because of his natural shyness and in part due to his desire to make sure that the essay focused on the Kentuckians' stories and not his own. But as I poured through the transcripts looking for tape to bridge actualities and construct a coherent narrative, I found myself adding in more and more segments from our June 15th interview. It was the words in this interview that finally gave the essay shape and coherence. And here the split focus of the piece finally took shape.

My objective, only partially recognized as I was producing the piece, was to intertwine two stories, simultaneously engaging and sustaining listeners' interest in the lives of Annie, Chester, Liddy, and the other Harlan residents, as well as their interest in the Italian scholar Alessandro Portelli and his attempts to make sense of what he had discovered and recorded. This split focus made the piece much more interesting and richer than either story told separately or alone. Alessandro is fond of talking about field work as an experiment in equality. (You can hear him elaborate on this in his essay and at the end of Chapter 3). I think that this aural essay has been able to actualize what that means, for in the juxtaposition of conversations about conversations, many people offer interpretations, and listeners get to listen in as meanings are constructed. If we have been successful, Alessandro's words as "author" and "scholar" are, indeed, no longer fixed and authoritative. Rather, he becomes a guide who shares his search for understanding.

In working on the piece, I also found myself adding in more and more music. Music was one of the things that had drawn Alessandro to Harlan. In authoring this essay we were privileged to be working with people who have an exceptional and uncommon ability to tell about their world and lives through music. The songs that Alessandro had recorded serve many purposes. They reinforce the stories he was told, deepening and clarifying lives and events. They provide moments of rest in which a listener can absorb and contemplate what has been said. They bridge actualities, introduce and foreshadow what follows, engage and sustain listener interest, and more.

Many authors will not have such an extravagant wealth of music with which to work. Authors whose interviews do not create such wondrous music may question whether their own studies will work this well in sound. But the palate with which an author paints in sound contains many sound artifacts and documents that can be equally effective in bringing a study or story to life. An extremely spare use of sound can be as compelling as the most extravagantly lavish production. Rain, footsteps, dogs barking, machinery, breathing, and other sound effects all have their own musicality and can be wonderfully evocative. Jokes, rhymes, a cappella singing, conversational asides, interactions with others can similarly bring to life the world under scrutiny. Sound artifacts ranging from commercial musical recordings and newscast soundtracks to home tape recordings can be extraordinarily effective. For "I Can Almost See the Lights of Home" we chose not to use any pre-recorded sound documents, with one exception, which can be heard in Chapter 3.

Last summer, while back at Columbia, Alessandro provided me the dub of an interview he had completed a number of years before with former coal miner and political activist Mildred Shackelford. Alessandro had referred to this interview in our own June 15, 1997 interview, for Mildred was the first person to explain to him why people in Harlan County were so quick to accept and open up to him. His recall during our interview of two passages from his conversation with Mildred was so close to their original dialog that I placed her words directly on top of his—near the end of Chapter 3—to demonstrate the accuracy of his memory [28  56  ISDN]. This layering, impossible in print, enables listeners to simultaneously listen to the original interview and Alessandro's quite accurate recollection.

Authoring in Real Time

Communication on paper and in sound are also very different in how they move through time. When reading we naturally accelerate and slow our movement through the material in response to how tired or alert we are, in response to the complexity of the words or ideas, our level of interest, our objectives, and a host of other factors. Sound, however, unfolds in real time. Few of us have playback machines with controls that permit the listener to accelerate the pace—and change the pitch—of the sound programming.[6] Intelligibility, then, and the communication of information in a form that is also interesting enough to hold listener attention requires different skills of authorship. There is an art to authoring in sound, I would maintain, that requires the same sort of sensitivity and attention to the rhythm, cadence, and flow that is required of a musician or composer. To hold listener attention, authors working in sound media must pace the flow of information; at times, to pay as much attention to the space between the words as to the words themselves. An example: The pacing of a five-minute news piece will usually be much faster than that of a piece that runs a half-hour or hour. The pacing and rhythm of a sound documentary, the art of its composition, can make the difference between a piece that draws in and engages listeners and one that they quickly turn off.

The problem of pacing is especially critical at the beginning of a piece, for it is here that the listener's interest must be engaged and held. In the rough mix of "I Can Almost See the Lights of Home" it was the first three chapters and movements that gave me the most problems. In the original script Alessandro and I had agreed that near the beginning of the piece we needed to provide listeners some basic information about Harlan County and its residents. So in Chapter 1, at first titled "Kentucky Travelers," we chose cuts from interviews in which Gladys Hoskins described her impressions of Harlan County when she first arrived there with her family in the 1930s and three excerpts from Alessandro's interview with Bill Gent, who talked about the coal mining community into which he was born and raised, movement north to Detroit, and the insults that mountain people suffered in the North.

Concerned that listeners not stereotype the residents of Cranks Creek as ignorant and isolated hillbillies, we decided to begin the piece with two actualities in which Annie Napier and Gladys Hoskins acknowledge and then dispel the stereotype: Gladys telling of the naivete of a waitress out west who assumed that the Hoskins had never seen a television, and Annie expressing her pride in being a "hillbilly," and then telling of the many places to which she had traveled. Fearful that the opening segments not perpetuate stereotypes, participants of the 1997 Summer Institute concurred with this decision, and so this remained the opening segment through the rough mix. But the presence of these actualities at the beginning of the essay gave me unending problems. Try as I might to segue into the next segment, these actualities felt incongruous and out of place. They just did not sound right or make sense at the opening of the essay, for listeners had no context by which to make sense of the refutation of a misconception that had not yet been presented. After many experiments, I cut out all but one of the Gent actualities and moved Annie's and Gladys's hillbilly cuts to the end of the newly created Chapter 1, "My Home Up in the Hills," which now opened with a Bill Johnson song of the same title. Following the descriptions of life in Harlan, they now made sense and bridged well into the new First Movement, "Snakes." By trusting listeners to hear the humanity of the folks in Cranks Creek, we no longer began the essay with a segment that I suspected many listeners intuitively would have heard as preachy and out of place. (I also sent dubs of the essay down to Annie Napier. She and others who listened to the essay offered no criticisms. Her only suggestion was that I increase the volume of the music!)

Once I had made this adjustment, the opening fell together. Annie's and Gladys's hillbilly stories now formed a wonderful bridge into the First Movement, but the earlier actualities in which Gladys and Bill Gent described life in Harlan, vital as they were to the communication of place, were still too dry and unengaging. Even with Arthur Johnson's wonderful song as a segue and bed there was something missing. Pouring through the transcriptions I found a section in which Annie told Alessandro about old home remedies. One segment of this we had already used in Chapter 4, but there were still unused stories on laxatives and cures for the colic and asthma. On their own they were interesting but not compelling. Interspersed with Hoskin's and Gent's descriptions of Harlan, however, they worked wonderfully, adding comic relief, bridges between cuts, and a layering of voices that pulled the whole chapter together. The best way to assess the wisdom or folly of this decision is, of course, to hear both versions. To hear the first draft, you may access the rough mix in this separate sound file [28  56  ISDN].

Contrapuntal Radio and Multi-Channel Sound

One of the great challenges of authoring in sound is how to present essential, but time-consuming contextual information in real time without losing one's audience. The ear processes information differently than the eye. The initial tendency for scholars who begin to author in sound is to remain wedded to the linear aesthetic of the written word and so to author monaurally. (Today, most scholars and journalists using tape recorders continue to record their interviews and other sound documents in mono). Centuries ago, European composers and librettists discovered polyphonic writing, layering two or more voices in their musical compositions. Decades ago, the music industry learned to utilize the ability of stereo sound delivery systems to add depth and dimension to musical recordings—to make the music more lifelike and engaging. Scholars working in or with sound media will also need to learn how to think in sound and author in stereo if they are to use the media as more than a one-dimensional carrier for words read out loud, but in truth composed for the eye.[7]

Although they share many traits in common, communication on paper and in sound are also guided by different rules of construction; they operate under a different grammar and syntax. Authoring in sound one layers different elements on separate channels and uses a variety of cuts, fades, and pans to link actualities, continuity, beds, and sound effects into a seamless composition. One of the most important differences is between the linear mono-tony of written communication and the multi-vocality of sound. An essential part of the great pleasure we experience in music comes from the artful layering of sounds and voices: from overlapping rhythms and harmonies, from counterpoint and fugue.

In the early days of radio, documentary producers quickly learned how to take advantage of the multi-dimensionality of sound, using ambiances and music as "beds" that carry the listener to different times, places, and moods. Trained in written communication and needing to hold the attention of a radio audience, few, however, dared to push their use of synchronicity. It took a classically trained musician, Canadian pianist Glenn Gould, to finally rethink the spoken word documentary on a musical model.

In 1966 the Canadian Broadcasting Corporation commissioned Gould to create a spoken-word documentary for the Canadian Centennial. He used this opportunity to create an hour-long "oral tone poem" on the Canadian North. Coming from a musical, rather than a print background, Gould's search for auditory solutions to the presentation of information in spoken words led him to music rather than the printed word or motion pictures for models. All audio documentary producers face the challenge of how to condense essential information that is excessively time-consuming or dull into shorter, more engaging passages. Gould's solution was to layer and overlap voices in a fashion he called "contrapuntal radio." He became fascinated with the degree to which one could listen simultaneously to more than one conversation or vocal impression. Gould recognized that such multi-vocal montages would test listener patience and tolerance, but drawing on musical analogies he also knew that such concerns had done little to deter composers and librettists from utilizing vocal trios, quartets, or quintets in both operas and concert music. Confident in the formulas of western classical music, Gould built his documentary on a musical, symphonic model. By using his contrapuntal technique of multi-voice montages, Gould was able to convey a great deal of information in a short amount of time.[8]

When in 1977 he produced The Quiet in the Land, his third and final piece in The Solitude Trilogy, the use of stereo enabled him to push his contrapuntal radio to new heights. Fascinated by what would become of a small group of Canadian Mennonites as they moved from the country to the city, Gould focused on their unique relationship to the world: that of being "in the world, but not of it." In an extended, ten-minute, multi-voice montage that begins about three-fifths of the way through the piece, Gould explored this dynamic tension by orchestrating a succession of overlapping voices. Soon it becomes impossible to follow any individual voice. Each new overlap causes one to lose track and wait for the appearance of the next new voice to again provide a single focus. Soon, however, one begins to follow the rhythm of spoken words and music of the voices, the ebb and flow of impressions, rather than the completed thoughts. A wonderful syncopation emerges.

Gould was quite consciously challenging the listener, and he sustained the contrapuntal montage long enough to enable the listener to move from frustration to meditation. After a while it becomes clear that the objective in listening is not to attach to any one voice as they flow in and out, but rather to be in this world but not of it! In effect, Gould created a sound piece that mirrored the subject of his meditation. To listen to this oral tone poem one must let go of individual voices, single ideas, and narrative strands of thought: all the talk about what it means to be a Mennonite. Surrendering to the contrapuntal presentation is like stepping through a door into a world in which time is suspended and one experiences how there is no single "Mennonite" belief, but many. [28  56  ISDN].

As mass consumers in a marketplace of competing ideas, we have all been socialized to read or listen for information that offers few challenges to our comprehension and that we can obtain from a single reading or listening, be it one line or one voice at a time. None of the programs in Gould's Solitude Trilogy are one-time listening experiences. These are not disposable information. Gould required that his audience listen in new ways, that they enter a new auditory space. (The experience is very much like trying to unfocus one's eyes to see the three-dimensional pictures buried inside two-dimensional images. Only after you are able to let go of the flat surface can you discover the pictures that have depth as well as outline.) Few have taken up Gould's challenge to cross over the perceptual threshold from monotony to polyphony. But his contrapuntal technique is a valuable tool for authoring in sound.

To take advantage of binaural hearing—the way we can hear and process different information simultaneously arriving in our two ears—I produced "I Can Almost See the Lights of Home" in stereo. I also made modest use of layered voices when I felt that this was the best way to present the material. One aspect of Harlan County culture that fascinated Alessandro was the paradoxical existence of both great violence and great care among the hill people of Eastern Kentucky. We devoted the Sixth Movement, "Violence and Care," to this theme. [28  56  ISDN]. We built it around two couples: mother and son, Omie and William Gent, and husband and wife Chester and Annie Napier. One of the major events in the Napiers' lives was a debilitating accident that almost cost Chester his life. Chester and Annie each told Alessandro their own versions of this accident and in their descriptions of what took place Alessandro had captured wonderfully compelling stories of caring. The question, then, was how to present them.

Linearly placing one after the other, as one might in print, simply did not work. The stories were so long that I found my own interest drifting—a bad sign—for if a segment does not hold the producer's interest what hope is there of it holding the attention of a listener? My next solution was to segment the two stories, moving back and forth from the one to the other. Again, too long, and now hard to follow. The third solution was play them both at the same time, Chester weighted to the right channel—coming in the right ear—and Annie to the left. Such multi-vocal passages can be challenging to the comprehend, so they cannot be sustained for long. But the overlapping worked. The interplay between their voices, the contrapuntal rhythms of their speech, the pauses in which one's attention shifts from one account to the next, made this segment come alive. I used this technique again in Chapter 8: "The Acts of Man Continued." [28  56  ISDN]. Here I juxtaposed Annie Napier and activist Joan Robinette speaking about their grass-roots organizing efforts against the mining companies. Here, too, I found the simultaneous voicing an effective way to present this material.

None of the contrapuntal passages in "I Can Almost See the Lights of Home" are nearly as challenging as those in Glenn Gould's Solitude Trilogy. One has only to briefly listen to his seven-minute montage in The Quiet in the Land to appreciate how sophisticated and complex contrapuntal passages can be. Gould's pieces are also extremely challenging to the untrained ear. Not wanting to lose listeners I tried to keep the montages restrained and inviting. I hope, however, that they do demonstrate how contrapuntal techniques can solve certain dilemmas of storytelling in sound media.

Contrapuntal radio montages also raise interesting questions about how we listen differently than we read. An unspoken rule of most journalistic and much historical writing is that the reader should be able to comprehend all of the words and extract the full meaning in a single pass. This is all well and good for the transmission of useful information, but it has its downside, for the same text read a second time may often offer little that is new. Poetry, music, and richer written texts improve with repeat exposures. Every time we listen to a good piece of music we get more from it, and our enjoyment grows as we learn the rhythms, cadences, harmonies, and words. Moby Dick read at age 45 or 60 is a profoundly different text than the same words read at the age of 18. And so it should be with good aural history. The multi-vocality of contrapuntal audio creates an aesthetic richness akin to music that clearly separates sound from written communication, and that offers scholars an opportunity to present sonic texts that are as rich and textured as those on paper.

By presenting discrete information to the left and right ear, stereo permits the layering of voices and other sound elements in a manner that is both aesthetically engaging, and that changes the ways one hears and comprehends. New interactive, multi-channel technologies and the development of three-dimensional sound systems promise to offer scholars authoring in sound an even more broader and deeper canvas upon which to work. Studies authored with sound elements moving in a three-dimensional field can rivet people's attention.[9]

Multiple Media Publication

Alessandro and I conceptualized this essay as a study for publication in a print journal, (a CD with accompanying articles). Our idea was to use texts in print and sound in a complementary fashion, playing to the strengths of each. On-line publication has expanded our ability to work affordably in multiple media. Had we thought about on-line publication at the beginning of this project we could have easily and affordably integrated videotape and more still images as well. As it is, the photos remain add-ons, unintegrated into the central body of the work. Hypertext and multimedia present scholars the ability to author in numerous media combinations. To do so effectively, we will all need to develop new forms that integrate the different media and play to their separate strengths. I hope that "I Can Almost See the Lights of Home" demonstrates how, within these new media combinations, sonic texts can be an important component.

"I Can Almost See the Lights of Home" was an experiment: our attempt to explore what it means to think in sound. The decisions that we made were not the only, or necessarily the best solutions to the problems that we were attempting to solve. The essay includes a series of probes, designed to explore how authoring in sound may help unlock some doors—some very old and some perhaps new—to the wonder of the human condition.

Back | Contents

6. In the early 1980s I bought a Soundpacer, a tape playback machine that permitted tape playback at anywhere from half to twice as fast as normal speed and adjustment of the pitch so that the voices, music and other sound elements retain the same pitch! Speed and pitch adjustment are even easier to do in digital media and make it quite easy for listeners to scan through sound files and programming. On-line sound delivery systems also permit users to stop and start programming and to adjust the speed and volume of sound programming. [Return]

7. A fascinating introduction to systems of acoustic communication and both natural and man-made soundscapes can be found in Barry Truax, Acoustic Communications (Norwood, NJ, 1985). [Return]

8. Glenn Gould's Sound Trilogy: Three Sound Documentaries (CBC Records, PSCD 2003-3. Canadian Broadcasting Corporation, 1992). [Return]

9. Multichannel sound has enormous potential. A number of museums, zoos, and other institutions are already using computer-based multi-channel sound systems—some using eight or more separate sound sources—to create "sound and light" shows and jungle soundscapes. In 1988, I produced This Car to the Ballpark, an 18-minute quadraphonic audio arcade assembled from oral histories, archival recordings, and sound manipulations. It was exciting to hear how sound originating from four separate speakers could move through a large space and create an extraordinarily rich and complex listening experience. The age of stereo sound may soon give way to four- and five-channel sound delivery systems. Digital video disc (DVD) technology is unleashing a multi-channel revolution in sound. DVD-Audio players will soon be on the market and record companies are already preparing to release multi-channel music. On the history and future of multi-channel sound see Dan Sweeney, "Multi-Channel Music: Silence at the Top," The Absolute Sound (Issue 16, February 1999): 41-51. [Return]