Can Almost See the Lights of Home ~
On the Making of an
Charles Hardy III
Part ethnography, part oral history, part radio documentary, "I Can Almost See the Lights of Home" is a hybrid work, an "essay-in-sound" designed to be heard, not read. This is a long work, so we have divided it into sections. The accompanying table of contents lists the order and time of the different chapters and movements. A transcription of the essay, complete with song lyrics and the names of speakers, is included for reference. This work, then, was an experimentan attempt to think in sound and to raise some questions about scholarship in the digital age. It is an experiment made possible by new digital technologies that are making it easier and more affordable to author in electronic media.
What follows here is a history of the creation of "I Can Almost See the Lights of Home," so that "browsers" for lack of a better word, can better understand our intentions, objectives, and the process of authorship: how two oral historians attempted to think "outside the box." Elsewhere in the Journal you will find another essay, written by Alessandro Portelli, providing background on this project and on the history of his own work in sound. The essay is best listened to with eyes closed and without distractions. To understand the potential of sound media for scholarship one must really listen. The essays and transcription should be read at a different time.
New Wine and Old Bottles
Media convergence being wrought by the ongoing digital revolution is breaking down and reshaping in significant ways the rules and conventions that structured discourse in print, radio, film, and other previously separate media. Sound, moving image, and mixed media now provide scholars opportunities to find creative solutions for some of the conceptual and stylistic problems they confront in their work, some of which are a product of written-word communication.
Scholars, journalists, novelists, poets, essayists, composers, and other authors who would work in the digital domain confront the age-old challenge of fitting content to container. There seems to be a pattern in which new media are initially used as containers for texts developed in other media, a phase in which users place old wine in new bottles. Authors have solved this misfit between content and container by creating new grammars, syntax, and formulas. We have seen this pattern in the use of writing as a carrier of the spoken, rhymed communication of oral cultures before the development of prose; early motion pictures as a carrier of stage performances before movie markers invented the art of cinematography; the phonograph as a carrier of compositions written for the sheet-music industry before the art of phonography made the recording rather than the printed notes or live performance the normative musical event; radio as a carrier of vaudeville and records; television as a container for old movies, radio soaps, and action adventure serials; the Internet as a carrier of images and text created for other media.
Prophets of the digital revolution tell us that the World Wide Web (WWW) will create a celestial jukebox of sight and sound that mixes and jumbles wines from previously discrete information carriers; a converging gumbo of written words, still and moving images, sounds, and manipulations; a unified docuverse that reintegrates our senses in a unified three-dimensional sensorium. Citing the utopian hopes that the pioneers of the phonograph, motion pictures, radio, and television placed in these earlier electrical media, skeptics rightly warn us to beware the claims of media hucksters intent on a sale and starry-eyed idealists still holding onto hopes that the Second Coming will arrive in the next new information carrier. So we make no great boasts or promises for our own foray into digital sound. We do, however, ask you to listen to our essay with ears open and eyes closed.
Having given workshops on sound documentary production for many years now, I also feel compelled to add one final caveat. It is not our intention to impose any new "rules." We offer no new Hammurobic Code or Chicago Manual of Style to follow. The wonderful thing about the digital domain, for the moment at least, is the opportunities it offers to break free from the "artificial" rules and restraints currently imposed by the publishers of books and scholarly journals, radio and television broadcasters, and other media gatekeepers. So on with the story.
Alessandro and I approached this project with different but overlapping agendas. A shared objective was to find out what one can do in sound when freed from the restrictions of written-word communication and radio broadcast. A perhaps more presumptuous objective was to demonstrate to oral historians and other scholars how sound can be a wonderful medium in which to author "serious" works of scholarship. We shared the belief that certain things are better communicated in sound than in writing and that the ear has an important role to play in our attempts to make sense of the world in which we live.
We came to this project from different nations, disciplines, and backgrounds. But we shared a love of voices, music, audio tape, and sound media. Alessandro was first drawn to the United States by the music of Woody Guthrie and other American folk musicians. During his first oral history project in Italy he produced an LP on Italian songs of resistance against Fascism (see Portelli's essay). I come from a family of musicians and spent much of the 1980s producing spoken-word documentaries for public radio. We both ended up teaching at the Columbia University Oral History Research Office's (OHRO) annual Summer Institute. Since 1992 director Ron Grele and associate director Mary Marshall Clark have provided at the Summer Institute a forum for participants from around the world to explore the evolving worlds of oral history scholarship.
The idea for the project was born at the 1996 Summer Institute where Alessandro was discussing his ongoing work in Harlan County, Kentucky and I was teaching the basics of field recording equipment and recording techniques. For years I had been championing sound media as carriers in which oral historians should present their work. My premise was, and is, that oral historians should learn to communicate in the same medium in which they gather their interviews, and that in order to do so effectively they must learn how to think in sound. Knowing of Alessandro's love of the spoken word I asked if he would be interested in collaborating on a sound piece based upon interviews he had been conducting in Harlan County, Kentucky. Three months later, at the 1996 the Oral History Association annual meeting, I provided him with a portable digital audio tape (DAT) field recorder, one stereo microphone, an omnidirectional interview microphone, and a dozen blank DAT cassettes. Between October 14th and 16th Alessandro completed more than 14 hours of new oral history interviews and sound recordings in Harlan County.
In 1997, OHRO built its annual Summer Institute around the scripting of our sound piece. OHRO transcribed all of the interviews, which Alessandro and I reviewed in preparation for the two weeks we would have to work together on a script that June. Reading through the transcriptions and listening to the tapes I began to worry. Dates were few and far between, and I was disappointed at what I thought to be the paucity of the sorts of anecdotes and stories with which I was used to producing radio documentaries. Without a clearer story line scripting was going to be a real challenge. So one of the first problems we needed to solve was how to fill in the missing pieces and explain what was on the tapes.
For decades after the birth of the spoken-word sound documentary in the 1920s, radio producers used a narrator to tell a story that they then illustrated and brought to life through dramatic recreations and recorded "actualities" of real people and events. Primitive field recording equipment limited the use of actualities, which producers used in the same manner as historians and journalists used direct quotations: as evidence and illustration. This documentary formula soon became associated with American radio's most successful documentary series, Time and CBS's The March of Time. Characterized by its authoritative "Voice of Doom" narration, this documentary formula served radio well until the social and scholarly revolutions of the 1960s and 1970s made the traditional documentary formula sound old-fashioned and unnatural. Converting voices to uniform words on a page is a leveling and homogenizing process that can strip away difference. (Retaining some of those differences by the use of phonetic spellings and indications of dialect presents a whole other set of problems and challenges that most oral history programs prefer to avoid.) But a well-heeled member of the white media elite narrating the story of the Civil Rights, antiwar, student, or women's movements could sound so incongruous as to feel almost surreal. As the nation moved from consensus to pluralism the once-respected and "natural" voice of authority now had to share the microphone. And so female, minority, working-class, ethnic and other new voices of authority began to appear in broadcast journalism and documentaries. But the underlying and increasingly problematical question of who gets to interpret events still remained.
A second response to the authoritative narrator was for documentarians to eliminate the narrator altogether. When National Public Radio came on the air in 1970, the flagship program of this new, non-commercial radio system boasted a female host and championed the telling of stories through recorded sounds. At the heart of this new network was the radio "sound piece," in which people told their own stories. In this new formula the producer/author abandoned the microphone altogether, but continued to orchestrate from the studio. This was all made possible by new low-cost, high-fidelity, portable analog cassette recorders that enabled a producer/author to tell a story with voices, ambiances, and sound effects recorded in the field. These, then, were the two bottles into one of which most producers would pour Alessandro's interviews. Neither, however, seemed to fit our needs, either aesthetically or conceptually, for the field of oral history has itself undergone its own evolution, an evolution in many ways parallel to that of non-commercial radio.
In the 1950s and 1960s most oral historians used their tape recorders to fill in gaps in the historical record, gathering historical evidence that was no longer finding its way into the written record. As the field of oral history continued to expand and develop in the 1970s and 1980s, oral historians began to change the way they thought about the interview. The older scholarship based upon the 19th-century paradigm of scholarly objectivity worked on the premise that the historian was a neutral agent who collected evidence from the interviewee. This model, however, quickly broke down, as oral historians listened back to their own and other oral historians' interviews and heard how much their own world views, interests, experiences, moods, opinions, scholarly and personal agendas, biasestheir subjectivityshaped the content of the interview and the elicited "evidence."
Alessandro Portelli has been a strong advocate and theoretician of the "dialogic" and subjective nature of the oral history interviewof the interview as a document that is co-created. So we faced a dilemma. Any outside narrative voice, be it his own or a ventriloquist reading his words, would reinforce the notion of the scholar as authority and of the interviewees as data banks rather than interpreters of history. To us, hiding authorship while maintaining it seemed dishonest and misleading; the act of selecting and assembling actualities is itself an act of analysis and interpretation. So the solution we came up with was to have me interview Alessandro, and to use excerpts from that interview to form the narrative thread that would hold the piece together. Our study, then, would be multi-layered: a conversation about conversations in which interpretation was tentative, evolving, subjective, and shared.
Alessandro arrived in New York from Rome on the evening of June 15, 1997. Late that night we completed a two-hour interview, and as we talked I, for the very first time, began to hear the essay take shape. While Alessandro spoke about the questions for which he had been seeking answers, about his relationship with the folks in Harlan, and about complex layers of connections that had previously been invisible for me, I was at last able to understand and make my own sense of the interviews recorded on the tapes. The interview also provided us both an opportunity to think out loud, and for me to ask him questions about things that I did not understand or about which I, as the first listener, wanted to know more.
Over the next two weeks we hammered out a script, cutting and pasting actualities late into the night for distribution to Columbia Oral History Summer Institute participants the following day. Each day we would talk through the previous day's installment with the Institute participants, who had also received all of the transcriptions the day they had arrived at Columbia. These discussions enabled us to talk the essay through its process of conception and to incorporate the Institute participants' comments and suggestions. And so the script took shape.
Form and Formula
The shape and structure of any work is determined by the carrier. Book reviews, journal articles, and monographs must all be configured to fit within predetermined word and page limits. Just as the time limits of sound recording media78 rpm phonographic discs, the 45 rpm vinyl "single" and long-playing record, 8-track and cassette tapes, the CDdetermined the length and shape of America popular musical composition, so the division of American radio into half-hour and hour blocks of time shaped the long-form sound documentary.
New digital media change the parameters. Our intention from the start was to create a hybrid, for our essay had to fit into two "containers." Our primary container was to be a CD that we wanted to be inserted inside the cover of a refereed, academic journal. Duplication costs limited us, we believed, to a single CD, which could hold 77 minutes of stereo sound. Publication for a non-broadcast, academic audience would permit us to focus on the story we wanted to tell without having to worry about creating a piece that would fit into the public radio container. But we also wanted to produce a work that could still be broadcast, and the best way to do so was to make it 56 minutes long or divide it into two 28 minute "programs." My assumption, too, was that the high cost of production time meant that we would only have enough money to produce a single composition, rather than tailoring different versions for CD and radio release.
The next conundrum we had to address was that of organization. One of audio tape's most serious limitations is the comparative difficulty of accessing material. A typed transcription collapses time into space and thereby permits a reader to move almost instantaneously back and forth through time, scanning, underlining and making notations on paper. Print is, in this sense, a highly interactive medium. Finding words on analog tape, however, requires winding through tape that is notated only by a distressingly inaccurate mechanical numerical counter and then listening to words in real time. So how, then, could we make sound more attractive to scholars by using the compact disc to increase interactivity?
The solution we came up with was, again, a hybrid. From the world of print we borrowed the convention of the "chapter," and from the world of music we took the convention of a "movement." Chapter headings would be assigned to sections that tended to be more linear and informational; movement titles to segments that borrowed more of their grammar from musical composition, that layered voices and sounds and communicated meaning more impressionistically or poetically.
Breaking the documentary into chapters and movements would also make it easier for listeners to cue CD players to specific sections and re-play particular segments. With the fast forward and high-speed rewind remote, a listener could quickly scan backwards and forwards, using the accompanying transcription to locate any desired passage. It became clear by this point, too, that we needed a new word to describe what we were creating. I began to call it an "essay-in-sound," for it is, in fact, an essay, not a documentary narrative, and I believe it is important to distinguish it from a "documentary," and the images that the word conjures up in most people's minds.
From Grapes to Wine: Editing and Composition
To write the script we had to sift through a tremendous number of words and songs: 700 pages of transcriptions drawn from close to 20 hours of tape. In addition, Alessandro had dozens of other interviews that he had recorded over the previous 15 years. I looked forward to adding additional strata of sound. Using segments from Alessandro's older recordings would enhance the essay's production valuesthe variety of sound elements that sustain listener interestand emphasize the sense of change over time. It would also permit listeners to hear the difference between an interview poorly recorded on analog cassette and one close-miked on DAT. The possibilities were very exciting, but also endless.
To keep the project manageable, Alessandro came up the idea of building the essay around a single field trip. Again, the decision made sense on many levels; it would also keep down costs and was consistent with Alessandro's conception of scholarly interpretation as tentative and evolving rather than permanent and fixed. We did, however, make two exceptions to the new rule. An important tape that Alessandro had recorded of Becky Ruth Brae singing her own songs was blank. Some of these songs were vital to the essay. In addition, Alessandro had recorded very little of the ambiances and sound effects which authors of sound pieces need to establish time and place, to set mood, and to create transitions between voices and scenes. So we decided that I would accompany Alessandro back to Harlan to record additional music and ambiances when he returned to the United States in October.
By the time the Institute ended in July, we had a 28-page first draft of a script. In late September, Alessandro and I traveled to Harlan County where in two days we recorded a church service, family picnic, two neighborly visits, the missing Becky Ruth songs, barking dogs, rain, clucking chickens, a sawmill, and other sound effects and ambiances. We now had all the elements I needed to bring the story to life. With tapes in hand, it was now time to start dubbing tape and editing sound files.
From Tape to Type to Tape . . . or PC
Although some radio documentary producers will work directly from the tape, most write their scripts from transcriptions, for it is easier to sort through and organize a large volume of material when it is on paper. Before the actual editing began, however, I had to decide whether to edit on open reel tape or digital sound files. From the 1950s through the 1980s, sound editing was skilled labor performed on open-reel tape recorders with an editing block, razor blade, and adhesive splicing tape. Interviews and other sounds recorded on cassette had to be dubbed onto open reel tape to be edited. To keep a piece moving most producers would edit spoken words, cutting out false starts, repeated words, "umms," "ahs," "you knows," and other speech crutches and non-essential words and phrases. Rocking the analog tape reels back and forth, the tape editor listens for hard consonants and clear spaces to make a clean, unheard cut. Certain producers edit more than others. I have known producers to make 50 or more edits in a three-minute actuality to "tighten" a passage, making it shorter and more listenable. Sometimes phrases are moved; sentences and paragraphs reconstructed. Not every edit is good, so deleted tape often has to be reinserted and an edit redone. When words or sentences run into each other, making an inaudible edit can be a real challengeand sometimes impossible. Editors routinely cut and reinsert slices one-sixteenth of an inch in length. Sometimes breaths have to be inserted to make the passage sound natural. So editing booths are often cluttered with strips of tape hanging from desk tops, walls, and the edges of machines. Open-reel tape editing takes a tremendous amount of time and a good deal of experience before it is mastered.
In the days when radio documentaries were produced on open-reel tape, the next step was to assemble the edited elements in their order of play onto reels in preparation for actual "mastering" of the program. Radio documentary production in the age of analog tape was a skilled craft. It was comparatively expensive, and both labor and time intensive. Production could be done one of two ways. In the early days of open-reel tape technology, feeder reels were placed onto two or more playback machines and mixed down onto a recording deck, performing fades, adjusting volumes and pans on the fly. Any mistake would require that the tapes be rewound and the whole process started over again. Longer pieces had to be mastered in sections, the good takes spliced together and then remastered onto a reel without edits. In the 1970s those who could afford it, moved to multitrack decks that permitted a producer to dub all of the sound elements exactly in place onto four or eight separate "tracks." The producer then had to "mix" the piece, adjusting the volumes and changing the panningthe assignment of elements to left and right channelson the fly. Again, sophisticated mixes often had to be done in sections, the best "takes" edited together into the final program.
The arrival of affordable digital audio workstations (DAWs) in the 1990s revolutionized sound documentary and music production. It is now possible to produce sophisticated, high-fidelity, stereo programming on a personal computer with software that costs only a few hundred dollars and that can be mastered in just a few days. DAWs have revolutionized editing. Loading recordings onto a computer hard drive converts them into digital sound files upon which one can perform "nondestructive," or reversible, editing with the keyboard and mouse of one's desktop computer. Digital editing of sound files also permits edits of less than one-hundredth of a second. Using a DAW, all the sound elements can be laid directly in place on a visual screen, moved at will, and volume and panning adjustments can be performed with clicks of the mouse. DAWs are revolutionizing sound production, democratizing it by reducing the cost of equipment and skills required to produce sophisticated, multi-track sound pieces.
Having left public radio in the late 1980s I had never used a DAW before, but recognizing that it was time to enter the digital age, I purchased DECK 2, a low-end Macintosh-based DAW that cost less than $400. Within a few days I had figured out how to edit and lay tracks. I was off and running, and now had to meet certain problems head on.
Notes for Part 1: