Language shows us very clearly that memory is not an instrument for exploring the past, but its theatre. It is the medium of past experience, as the ground is the medium in which dead cities lie interred.[1]

I had set out to make a film based on artifacts. It would be based on three sound recordings made onto wax or celluloid plastic cylinders. It would be an archeology or maybe a disinterring. I wanted to chronicle the excavation of three stories recorded on cylinders dating from the late 19th century or early 20th century: a recording of an aboriginal language; a recording of early music; and a recording made for the purposes of either commercial or political propaganda. These three categories were my initial attempt to classify these recordings. I was interested in exploring the physical cylinders themselves, the technologies that actuated them and which sounds people were first interested in preserving. I hoped that this investigation might answer what it is about sound that made it so desirable to capture, represent and preserve.

Cylinders are the physical storage objects onto which sound was recorded, and also the objects from which sound was played back. Imagine a vinyl LP shaped like a hollow tube. Cylinders were placed onto the mandril or rotating drum of a phonograph (appendix C). As these rotate around the fixed axis of the spindle, sound vibrations would be carved onto the surface of the cylinder by a metal stylus. The cylinder is inscribed with a picture of sound. During playback, the process is reversed, the stylus reads back the carved image and projects sound from the phonograph horn. The oldest existing recording, Charles Lambert’s Experimental Talking Clock (1878,) was written onto a lead cylinder.

The first successful recording medium was tin foil, and the first patent for a viable sound recording machine, US Patent no. 200,521 Improvement in Phonograph or Speaking Machines is credited to Thomas Alva Edison and was submitted to the United States Patent Office in December 1877 (appendix C.) This recording was of Edison reciting Mary Had a Little Lamb [2]. Experiments investigating potential recording materials led first to various wax substances,  and then to celluloid plastic which was adopted by the Edison Manufacturing Company in 1912 and remained the standard into the 1930’s. This material was suitable for both cylinders and the pressed flat disk. By 1888, cylinder records were approximately two inches in diameter and four inches long and would yield a spoken recording of about four minutes or a musical performance of about two to two and a half minutes.[3] The very earliest sound reproductions include recordings of instrumental music, poetry, and prose. Ethnographers and linguists of the time recorded aboriginal languages for study. Presidential addresses and the campaign speeches of politicians were recorded for inspiration and posterity.

Professor Lisa Gitelman, an expert on early sound recording technology, was helpful in initiating my own research on this topic. In an on-camera interview granted for the project, she offered the following comments:

Early on, entertainment became the dominant function and recordings were musical or spoken word. We could think of them as representations of performances. It’s not that you heard a band concert on an early phonograph record, you heard a two-minute selection of a few band instruments that were standing in for a band concert. Similarly in spoken word, what you were hearing was a representation of what would have been a fifteen-minute vaudeville turn. Those are two dominant genres of the pre-recorded musical cylinders.

People did use them for other things, for stenography, and an outgrowth of that would have been ethnographic recordings of indigenous populations in the US and Canada…They are connected in kind to another genre of the pre-recorded cylinder. There’s really no good label for it, but it is called “ethnic music” or “old-time music” by the collectors. The old-time records or ethnic records are particularly hard to categorize. In the 1890’s, there was an unfortunate popularity of what was called the “coon” song in the United States based on blackface minstrel traditions. So you could have those records, but you could also have what was then called “negro poetry” or spirituals recorded with far more seriousness. There’s another section of the stenographic functions, ethnographic home recordings, and recordings of businessmen but few of those have survived.[4]

I began to build the first story around some Onondaga language recordings owned by the Royal Ontario Museum (ROM), which date from either 1902 or1903. I saw these as points of origin for the recording industry and useful for examining the ongoing results of colonialism and the mechanism of propaganda. The Ethnology Department at the ROM sent the CBC cassettes of their collection of Edison wax cylinders containing voices of Iroquois speakers telling stories and singing songs. The CBC at the time was broadcasting a radio series called Moments in History: Lost and Found Sound based on old recordings that people had sent in. The ROM recordings were over a hundred years old. The voices on the recordings had been translated at the Six Nations Reserve in a joint project with the Woodland Cultural Centre and the ROM

The cylinders were deposited at the ROM in 1949, along with related and explanatory documents from linguists and ethnologists. These recordings remained at the ROM until the mid 1990’s when funding permitted the inclusion of the artifacts in a new First Nations display. The cylinders were transferred to audiocassette at Syracuse University’s Belfer Audio Laboratory, a lab and archive that specializes in this kind of work. The CBC based a segment of the series on these wax cylinders, and called it Iroquois Voices. The show stated that the difficulties in translating the recordings were due to the scarcity of people who could speak the languages spoken in the recordings.[5] In fact this is not the case, there are people on the Six Nations Reserve who speak each of the six Iroquois languages. The translation/transcription problems were due to the quality of the recordings.[6] Since their involvement with this project, members of the Six Nations reserve have been able to reclaim a number of stories and oral histories previously lost, and have put these to use in their community. This difference in the description of the problem directed my interest towards the differences in archival practices between the ROM and the Woodland Cultural Centre and the CBC.

I visited the Six Nations Reserve and the Woodland Cultural Centre where I interviewed Amos Key, the Director of Language Programs. My visit to the Woodland Cultural Centre, and research related to the Onondaga cylinder recordings had a profound effect on the outcome of the thesis that eventually became Mr. Edison’s Ear.

I was particularly struck by a story that Amos Key recounted in order to demonstrate the differences of opinion within his community regarding preservation: There are antique wampum ceremonial belts displayed at the centre. A group of elders wanted to use the belts in a ceremony—as they had for generations—and liberated them. After the ceremony, the belts were simply returned to the exhibition case. This practical and spiritual use of the belts caused controversy within the community over their value as useful and historical display artifacts.

I also spoke (informally) with Trudy Nicks, a curator at the ROM, about the recordings. At this meeting she outlined the plans for exhibiting the recordings. I was shown the Edison phonograph that recorded the original Iroquois singers, the cylinders (and the suitcase that had stored them) some documents—all the items that became the physical elements of the display. As well, the intention was to have an audio component, so patrons would push a button and listen to the recorded voices. These would be accompanied by explanatory text and translation. In the end it seems the ROM was unable to secure rights to the recordings, because the display contains no audio component, only the above mentioned artifacts.

At around the same time I was looking into the Onondaga/ROM recordings, I began experiments with time-lapse animation. I began to download images of the new construction at the ROM from the ROM’s website. I thought of this process as a way of starting to do work on archival practices and on the archive. These images were being uploaded to the web by a stationary camera mounted opposite the construction. I collected the jpegs at regular intervals, first daily and later weekly, as construction slowed. On each occasion I downloaded thirty jpegs over a span of sixty minutes. Thirty corresponds to a video second, or 29.97 frames per second. I intended to reassemble the images into a time-lapse animation of the construction. I understood this as a way of documenting and interrogating the passage of time. Time was beginning to interest me as a related and important subject matter particularly because the perception of time passage was beginning to shift radically during the mid-19th Century. This phenomenon related in part to the invention of the railroad system and the establishment of the Greenwich time system, but also to emerging technologies that made possible the capture and then replay of a previously lived moment. This early work clarified issues around the ownership of artifacts, the uses of the archive, and the practice of archiving, and remained present for the entire thesis.

Unfortunately, it became overwhelmingly difficult to secure permission to use the Onondaga/ROM recordings and to obtain the required permission from the ROM and from the Woodland Cultural Centre to investigate the Onondaga cylinder recordings in a manner that would do justice to the complex social and political issues that were being excavated. To properly address the racial and moral issues of colonialism, would have required collaborative work, and neither the ROM nor the Woodland Cultural Centre were interested in pursuing the investigation much further. So the video took another path, one that veered toward a more direct inquiry into the nature and recording of sound.

The act of collecting, led me to consider ways that I could translate this activity formally and conceptually and reflect it on screen. I decided not to worry about acquiring visually perfect material, but rather to allow the archival medium (video, film, digital file) to be exposed and in so doing, develop an aesthetic rooted in the process of collection. I wanted the distressed qualities of the source material to be visible because each recording media disintegrates or decomposes in a way that is specific to its medium, and that evidence grounds the preserved material both as a physical object and within a historical context. This decision turned out in fact to be correct one for the project—correct because archival practices change as technologies evolve and become obsolete. This process of evolution is important to a component of the record.

1. Walter Benjamin, “A Berlin Chronicle,” Reflections, ed. Demetz Peter, Trans. Jephcott, Edmund. New York: Schocken Books, 1978, pp. 25–6
2. The recording of Mary had a Little Lamb heard in the video is not the original recording. Edison re-recorded a demonstration in 1878. The original 1877 recording does not exist except in documentation and Anecdote. (NPS)
3. www.tinfoil.com. Duration was dependent on recording and playback speeds.
4. Gitleman. Interview excerpt, July 16, 2005.
5. CBC Radio (English), “Moments in History: Lost and Found Sounds: Iroquois Voices,” This Morning, Broadcast date:  June 2, 2000.
6. Amos Key stated that the recordings were difficult to make out because of the noise on them and therefore were difficult to translate. He confirmed that there are speakers of these languages and there is an immersion school on reserve. Interview date: March 2, 2004.