Recording Enthusiasts Crack the Phonautograph Code

Nearly 50 years ago, phonograph historians Oliver Read and Walter Welch published a now-classic work, From Tin Foil to Stereo. In its treatment of the phonautograph, that pre-Edison scientific device for recording (but not reproducing) sounds, the authors claimed to have heard rumor of a phonautograph recording of Abraham Lincoln's voice. To this day, no one knows if an recording of Lincoln exists, or ever existed, but other recordings made on the Phonautograph have survived. Thomas Edison himself made a few, and these are preserved in the National Park Service's archives. Numerous short recordings survive, in facsimile form, in late 19th century textbooks on sound and hearing. The fact is, even well into the era of the phonograph, there were many users of the phonautograph. The appeal lay in the fact that a recorded sound was immediately visible to the naked eye, making it an ideal way to demonstrate the nature of sound and sound waves. But recordings were usually made in some fragile medium, such as a thin coating of soot on the surface of a glass cylinder or plate. The recordings were rarely kept, and when they were, they were often too fragile to survive for long.

In late 2007, audio enthusiast and well-known collector David Giovannoni teamed up with research David Haber at UC Berkeley and Archeophone Records to try to locate and play back some of these early phonautograph recordings. They were not the first to try. Using analog techniques, engineers had attempted to reproduce phonautograph recordings before, but found that the wave-forms captured by the device were of such poor quality, and the recordings so short, that there was simply not much information there to be reproduced. By searching internationally for surviving recordings, Giovannoni was able to find several made by phonautograph inventor Leon Scott in an archive in France. Scans of the recordings were sent to Carl Haber, a research at the Lawrence Berkeley National Laboratory in California, who has used advanced digital imaging to reproduce phonograph cylinders. Using similar techniques, Haber was able to recover sound -- low quality sound to be sure -- from the recordings. While two of the phonautograms from 1857 and 1858 did not have intelligible sounds on them, one 1860 recording held a recognizable human voice. These recordings are apparently the oldest extant sound recordings known.

MORE SOURCES:
New York Times Article
More on the Phonautograph