A set of wax cylinders recently digitized by Indiana University's Media Digitization and Preservation Initiative may be the oldest true stereo recordings in the world, predating the oldest previously confirmed stereophonic recordings by over a quarter century.
Berthold Laufer, a German anthropologist and historical geographer, made history with the sound recordings he made in 1901, though he may have been unaware of it. He recorded what are believed to be the world's oldest stereo recordings while on an expedition in China, where he was sent to collect various materials for the American Museum of Natural History. The pairs of recordings feature vocalists on one cylinder and an orchestra on the other, performing Chinese folk and opera music in Shanghai.
"The idea behind making two recordings, one of the voices and one of the instruments, was that when he listened back to the recordings to transcribe them, it would be easier for him to hear what he needed to hear," IU media preservation specialist Patrick Feaster said. "He could play the voice cylinders while he was transcribing the voices, and then the instrumental recordings while he was transcribing the instruments."
Feaster works with the Media Digitization and Preservation Initiative at IU, which preserves historical media for research and education. The majority of Laufer's wax cylinders are held at IU in the Archives of Traditional Music, though they technically belong to the American Museum of Natural History.
"They've been kept at IU for many, many years because of the high profile of the archives, one of the most famous, largest collections of ethnographic sound recordings in the country or in the world," Feaster said.
The cylinders were digitized and reconstructed as part of the Media Digitization and Preservation Initiative, which was charged with digitally preserving all of IU's valuable audio, video and film holdings.
"Only a few scholars seem to have been aware that Laufer had used two phonographs at the same time, based on letters in which he describes what he was doing," Feaster said. "Alan Burdette, director of the Archives of Traditional Music, was one of them and brought it to my attention last fall. He already understood then that it might be possible in theory to reconstruct stereo from them."
The reconstruction took a great deal of research, experimentation and coding. The recordings showed that it was possible to listen to the past in stereo 25 years further back than previously thought.
"There wasn't any software available for this kind of restoration," Feaster said. "Nobody had ever needed it before, so it had to be custom-written. Figuring out which pairs of cylinders matched each other meant going back to Laufer's handwritten notes in English, German and Chinese."
Laufer's recordings have an eerie quality and fuzzy sound, but after understanding their significance to the history of stereo recordings, they take on a new life.
"They don't sound quite like recordings that were made to be listened to in stereo because they weren't," Feaster said. "But they do give a wonderful sense of being in the space where these sounds were being made. You can really, at least in some of the cases, get a sense for the instruments being arranged in a space around you, and that's just something that we've not had at all for anything close to this age."
The stereo recordings were initially intended for research, and it's difficult to say whether Laufer and other researchers at that time considered that audio recordings would be shared for entertainment purposes in the future. In 1901, there was no technology that allowed for mass distribution of sound recordings. People like Laufer used the phonograph to record musical performances to study them.
"Their assumption was that people would not learn about them from the recordings," Feaster said. "Or if they did, it would only be someone playing them at a conference, or something like that. Any findings, any information communicated to the public, would be in print publications, journal articles and so forth. And to get them into that form, you'd have to transcribe them."
Feaster finds it unlikely that there is an older existing stereo recording than Laufer's, though it's technically not impossible. Laufer's use of a new technology 119 years ago went just slightly above and beyond what others were doing at the time. This idea brings a new hope for future technological advances.
"Today we have lots of other new technologies for recording various things that couldn't be recorded before, such as smells," Feaster said. "So, I also wonder if we might be capturing some things today that might be equivalent to the stereo recordings Laufer made, making recordings that someone in 100 years might be able to use for some equally transformative purpose."
Understanding the complex history of sound recordings opens up new possibilities for future discoveries. It emphasizes the importance of the Media Digitization and Preservation Initiative's projects in bridging the gap between scholarship and technological feats.