Listening to the “Dog Society Song” or Trying to Decipher an Audible Trace of the Past
“Item number 14. Cylinder number 0039. Strips a and b,” an authoritative but smooth male announcer recites the library catalogue number of the sound recording I am about to hear. The historical recording itself then starts with a series of crackles, and there seems to be some semblance of a rhythm to that crackling noise. After several seconds, a male voice and drum gradually emerge. The drum beats out a steady and repetitious series of “duh-duns” like a heartbeat. The voice starts out on a high note and gradually meanders down and lingers, sometimes disappearing in the lower registers, and his melodic descent is sung with syllables that repeat with successive phrases. This occurs in variation three times within the one-minute sound recording that is labelled as the “Atsina Dog Society Song.”
The “Dog Society” song, one of the thousands of wax cylinder recordings made by Edward Curtis for The North American Indian project, offers the modern listener a tantalizing auditory trace of the Gros Ventre (or A’ani) culture. On the recording, I am barely able to make out the distant voice of an anonymous man, singing to the accompaniment of a drum and perhaps rattles. Listening to this historical recording is a frustrating endeavor as I must strain to hear the melody and make out the syllables. Indeed, my initial hearing of the “Atsina Dog Society Song” leaves me with many questions and very little information. Are the crackles at the beginning a rattle? Or is it just static from an analog recording made from old, and perhaps deteriorating wax cylinder? Is the warble in the voice due to the quality of the recording or present in the original performance? What is happening when the voice almost disappears from the recording? What words are being sung and what do they mean? Is the entire song on the one-minute recording or is it just a short snippet of a longer song? What kind of drum is being used? The enigma of the recording forces me to search for more information.
Edward Curtis’s monumental series The North American Indian was part of a nation-wide effort to collect and enshrine the cultures of American Indians that were deemed to be “dying out.” Although this “dying out” was a direct result of the conquest of the American Indian populations by the United States army and the diseases brought by the settlers, the myth of the “vanishing race” portrayed the disappearance of Native Americans as an evolutionary inevitability and hence spurred on efforts to record the cultures before they disappeared. Edward Curtis stated this view eloquently in his introduction to the North American Indian:
The passing of every old man or woman means the passing of some tradition, some knowledge of sacred rites possessed by no other; consequently the information that is to be gathered, for the benefit of future generations, respecting the mode of life of one of the great races of mankind, must be collected at once or the opportunity will be lost for all time.
In addition to creating the photographic images for which he is renowned, Curtis recorded approximately ten thousand wax cylinder recordings and produced twenty volumes that combined the encyclopedic information with ethnographic description and artistic representations in order to achieve such a comprehensive representation. The “Dog Society” song from the Atsina Dog Society ceremony as it was practiced or at least remembered in 1907, represents but a tiny piece of the auditory museum that Curtis collected “for the benefit of future generations.” Unlike the images that were reproduced for subscribers and subsequent publications, the wax cylinder recordings generally were not reproduced for commercial sale. If a song was reproduced at all, it was most likely in the form of a musical transcription of the wax cylinder recording by composer Henry F. B. Gilbert and included in the volumes that accompanied the images. Thus, unlike the recordings of American Indians made by ethnologists like Frances Densmore, which were archived by the the Smithsonian and reproduced commercially in the 1960s, these recordings languished and deteriorated in relative obscurity.
As described in the accompanying volume about the Plains Indians (Vol. 5), Atsina culture had several ceremonial societies, and each society, including the Dog Society, consisted of male members of a certain age. The volume even includes a brief ethnographic explanation of the Dog society dance:
The dance of the Dog society resembled in its principal features the ceremonies of the other societies. There were two leaders, who wore shirts, the one yellow and the other red, trimmed with crow-feathers. Each of the other members wore, over one shoulder and under the other, a baldric of red cloth, whose end, trailing the ground, was, in conflict, transfixed with a staff thrust into the ground. Thus the warrior was not permitted to retreat until some one not a member of the organization drove him away, speaking as if to a dog. All Dog Dancers carried eagle-bone whistles and small wands from which hung deer-hoof rattles, and all wore eagle-feather head-dresses. This dance lasted four days and took place in the usual enlarged lodge.
From this description, one could presume that the crackles heard at the beginning of the recording are indeed rattles, but the faint rhythmic pattern noticed at the beginning fade into static as the voice and drum enter, and the presence of rattles is not certain. However, it becomes clear that this one-minute recording cannot be “typical” in any way since it represents only one minute out of 5760 (4 days x 24 hours/day x 60 minutes/hour)!
The volume on the Plains Indians also includes a transcription of a “typical Dog Society song” (see transcription). This musical transcription, which features passages that seem to correspond to the sounds of the recording, helps me as a musicologist to pick out particular melodic contours and to visualize the relationship between the voice and the drum in the recording—in other words, to get a handle on the faded sounds. The transcription, however, also leads me to match the sounds with particular notes even though the pitches I hear seem to fall between notes on a Western scale. The transcription also assigns a particular meter and definite note values to the sounds even though the performance features more complex and nuanced rhythms. Curiously enough, the transcription seems incomplete: The recording features three distinct parts whereas the notated music only shows two. And finally, there are no words or syllables indicated in the score.
Thus more questions arise. Is the transcription offered in the volume even based on this recording or based on some other recording of a Dog Society song? How much has the recording deteriorated such that I can no longer hear recorded sounds that the transcriber working over a hundred years ago was able to hear and notate? If indeed the transcription is based on the recording, why did the transcriber take so much liberty with the transcription, especially considering that the aim of the project was to record these traditions for posterity? Why are no words (transliterations and/or translations) offered in the transcription? And how did this song fit into the larger Dog Society ceremony and what kind of dance did it accompany?
I offer some ready answers to some of my own questions: Yes, the transcriber probably was able to hear sounds no longer available to me. Wax cylinders deteriorate significantly with repeated playings over time. And historically, many musicologists did take liberties with the transcription of non-Western melodies, often forcing the music to fit Western scales and rhythmic structure. No words are included in the transcription because the composer doing the transcription probably did not understand the Algonquian language and possibly did not deem the words valuable. Indeed my limited experience with Plains Indian music leads me to guess that the singer in the recording could have been singing vocables, which are important parts of the song performance but without discrete semantic value.
In the end, I am left wondering about how the musical transcription represents or mis-represents the undoubtedly complex and lengthy ceremony that was the “Atsina Dog Society Dance.” On the surface, one could blame the limitations of a frail and imperfect recording for the poor representation. But even if the recording had been of high quality and the transcription more precise, it still would have been an extremely brief excerpt accompanied by numerous omissions of information. Among other questions, I would still be left asking about the individual whose voice I hear on the recording, about the circumstances of the recording session, and about the significance of the song and the entire ceremony in Atsina society of 1907. Indeed, the audible trace stubbornly remains merely a trace—a historical artifact that cannot answer questions that did not plague Curtis or many of his contemporaries, but which figure prominently in our own historical moment.