Experimental Eduction Projects:
making visual inscriptions of sound audible

Until the late nineteenth century, inscriptions of sound were typically earmarked for visual perception, not aural. But wouldn't it be exciting to hear them "played" right off the page anyway, much as we'd play an LP or an mp3? Since Fall 2008, I've been experimenting with ways of making various inscriptions "talk" or "sing" automatically—some made as recently as the 1940s, others dating back as far as the thirteenth century.

Paleospectrophony uses reverse Fourier analysis to play inscriptions that graph time against pitch just as though they were modern sound spectrograms, with no need for reperformance, transcription, or MIDI:

Balbastre's "Romance," as programmed by Engramelle (1778), regarded today as a groundbreaking effort to document a specific performance style in painstaking detail:  

Five "phonotactic" plates by Athanasius Kircher (1650), excerpted from his famous Musurgia Universalis:  

The optical film sound track method instead plays oscillographic inscriptions that graph time against amplitude, such as phonautograms. Here are two of the first inscriptions I tried in September 2008:

Eli W. Blake, Jr., "Ah, Ay, E, I, O, U. Brown University. How do you do? Brown University. How do you do?" (1878), the earliest identifiably recovered recordings of spoken English with a definite provenance:    [Read more]

A FRAGMENT of the actual tinfoil recording Edison used to demonstrate the phonograph to the editor of the Scientific American in December 1877, played at several different speeds with a gap inserted to reflect the missing parts:  

To experience the full range of these Experimental Eduction Projects—including a prototype Morse code message from the 1830s, medieval church music, and pitch contours of Shakespearean declamation dating back to the 1770s—start here.

The World's Oldest Alternate Take

At a meeting of the Indiana University Mediated Sound Group on September 30, 2009, a phonautogram of "Vole Petite Abeille" originally recorded by Édouard-Léon Scott de Martinville on September 15, 1860, was played in public for the first time. What makes this particularly noteworthy is that it's the second phonautogram of the piece we've heard.

Scott recorded this take of the song—actually the "Chanson de l'Abeille" from Victor Massé's comic opera La Reine Topaze—by attaching a stylus directly to an artificial chain of ossicles fixed to the tympanic membrane. The version we'd heard before was recorded a little differently: Scott had added an "amplifying lever" to the end of the chain of ossicles.

So now we can compare the recording without the amplifying lever:
...with the recording with the amplifying lever: .

What do we learn?

• The amplifying lever made a noticeable difference in terms of sound quality.
• The song itself is abridged and adapted almost identically in both cases—showing that a consistent fifteen-second version had been worked out for recording purposes.

Read more about the "Vole, Petite Abeille" phonautograms or check out the First Sounds initiative, without which none of this would have been possible.

A New Website Devoted to
"Recording Pioneers"

The recently launched www.recordingpioneers.com deals not with performers but with recordists, which is actually more consistent with early terminology (people like Billy Murray were generally known as "record makers" at the time; "recording" was what the machines did, not what the performers did).

If you want to see what's known about William Sinkler Darby, Leon Douglass, Cleveland Walcutt, or others of their ilk, this is the place to go—you'll find much original research, particularly of a genealogical nature.

Indiana University
Media Preservation Survey

Indiana University Bloomington holds more than 560,000 audio and video recordings and film reels, many of which are historically significant, all of which are actively deteriorating. And the window of time to save these materials is closing fast; most archivists agree that such audio and video materials could be lost forever in 20 years or less.

So begins a press release about the Media Preservation Survey for which I spent the 2008-9 academic year collecting data. The full report may be downloaded as a pdf file here, or you can watch a WTIU news broadcast on the survey's discovery of a cache of lacquer discs in the attic of Franklin Hall, including hundreds of episodes of the fabled "Indiana School of the Sky" series.

[Older home page items]