Past Items From the Home Page

Mid 2013

Now Available!
Cal Stewart: The Indestructible Uncle Josh (Archeophone 5009)

From A pioneer star of the talking machine, Cal Stewart emerged from humble origins in vaudeville to become one of the best known humorists in early 20th-century America. In his famous role as the "rube" Uncle Josh Weathersby, he entertained millions of listeners with tales of his antics both in New York City and at home in Punkin Center. This CD provides a snapshot of Stewart's repertoire as it stood at the height of his career, featuring all 25 of his 2-minute cylinders for the Indestructible company made in 1908 and 1909, along with a choice sampling of his work on U-S Everlasting cylinders made between 1910 and 1912. The production team of Patrick Feaster, David Giovannoni, Meagan Hennessey, and Richard Martin that brought you Actionable Offenses and Debate '08 has done another masterful job with The Indestructible Uncle Josh. [read more / order]



Early 2013

Latest: Slate ran a piece called "Listening to Records That No Longer Exist" about my playback last year of a paper print of a gramophone disc recorded in 1889 and published in 1890. A blog post I'd written on the same subject also got some new attention on Reddit, briefly reaching the number six position on their front page. Recent broadcast coverage of archeophonic projects includes a German-language report on Schweizer Radio und Fernsehen and Deutschlandradio Kultur, as well as an interview I did with Patt Morrison for Southern California Public Radio. I was also the "Indiana University sound historian" mentioned on NBC Nightly News for my part in discovering a recording of the voice of Alexander Graham Bell, although credit for the playback of the recording—which is the really newsworthy part of the story—belongs to Carlene Stephens of NMAH, Carl Haber and Earl Cornell of Lawrence Berkeley National Laboratory, and Peter Alyea of the Library of Congress. The same development has been covered in the New York Times and in various media abroad; here's Chinese.

Late 2012

The World's Oldest "Record"?

The print shown to the right attracted some media attention recently when I suggested that it's likely to be the oldest "record" in the world available for listening today. Let me clarify—I don’t mean it’s the world’s oldest sound recording; Scott's phonautograms, for example, are much older.  But nowadays when people use the word "record" colloquially to refer to sound media, they typically mean the specific format that includes LPs, 45s, and 78s—that is, the kinds of grooved disc you’d play on a "record player." Technically, these "records" are based not on the phonograph Thomas Edison unveiled in 1877, but on the gramophone invented by Emile Berliner in 1887. And this recitation by Berliner of Schiller's "Der Handschuh" appears to be the oldest gramophone recording currently available for listening—the earliest audible progenitor of all the world’s vintage vinyl.
[read more]


Early 2012

photo: Rich Strauss, Smithsonian

Voices from the early 1880s
at the National Museum of American History

Six early experimental sound recordings made between 1881 and 1885 by the Volta Laboratory Association and preserved at the Smithsonian Institution's National Museum of American History are now available for listening, thanks to a partnership between that museum, the Lawrence Berkeley National Laboratory, and the Library of Congress. You can read their official news release, a Smithsonian fact sheet, and an LBL technical report, as well as an Associated Press story about their accomplishments in sound recovery.

Meanwhile, I've been carrying out a complementary research project on the same material, drawing on it to help expand our understanding of the early history of recorded sound in America. In December 2010, I began studying the larger NMAH collection from which the six examples were taken, a project I continued from October through December 2011 through the support of a Lemelson Center Fellowship. With the help of curator Carlene Stephens, I examined each of the Volta recordings as well as all the other early experimental sound recordings at NMAH. I also scoured the written notes of the Volta Laboratory Associates at NMAH and the Library of Congress—not just the Home Notes of Charles Sumner Tainter, which have often been cited by historians of sound recording, but also the notes of his colleagues Alexander Graham Bell and Chichester Bell, which have generally been neglected. Combining these written and artifactual sources has given me a more comprehensive picture of the Volta Laboratory's activities than I believe has been available before.

I made a particular effort towards the end of my residence to contextualize and interpret the six specific recordings that had just been played back. You can read some of my conclusions at First Sounds. If you haven't yet heard the Volta recordings, I hope this will serve as an enjoyable introduction to them. And if you have, maybe what you read now will lead you to hear them in new ways. (It certainly had that effect on one blogger.)

[Older home page items]


More New Old Sounds

Work continues apace on the eduction (or automatic "playing") of primeval inscriptions of sound. Here for your consideration are just a few of my recent results.

Dasia notation

—First, here is the closest thing you're likely to hear to a thousand-year-old phonogram: four passages in Daseian notation, drawn from a tenth-century manuscript of the Enchiriadis treatises, which I've converted into sound just as though they were sound spectrograms (a technique I call paleospectrophony). The whole manuscript plays for about seven minutes at this speed. The Enchiriadis treatises, composed in the ninth century, feature the oldest notated polyphonic music in the Western tradition.

manometric flame recording

—Here is a manometric flame display of the spoken words "Doctor Koenig," as photographed on a strip of film and published in a Physical Review article of 1898 by Nichols and Merritt in homage to Rudolph Koenig, inventor of the flame manometer. If you're not familiar with Koenig's manometric flame apparatus, you can read an informative account of it on Wikipedia. To the best of my knowledge, however, this is the first time a recording made using this historically important instrument has ever been played back. These and other words are quite intelligible!

Paul Wendeler phonautogram

—And here is a phonautogram of the German word Karre ("cart" or "wheelbarrow") as recorded by Paul Wendeler—a student at the University of Kiel—about 1885 and first published in 1886. The instrument Wendeler used for his studies of consonant sounds was a modified phonautograph designed by his mentor Victor Hensen and most commonly referred to as the Sprachzeichner or "speech-depicter." The trace given here appears to be the oldest audibly identifiable recording of a word spoken in the German language.

(Stay tuned for information about a projected CD release that will contain these and many more pieces of elusive, paradigm-bending audio.)

June 2011 (left column)

Two Authentic Edison Playlists From 1907

The 27 July 1907 issue of Collier's featured an advertisement for the Edison Phonograph which included two recommended playlists ("sample programmes") calculated to show off the machine to best advantage. I've located online sound files for all twenty-four selections, all but one in mp3 format, so you can read the advertisement, see the illustration, and listen to both playlists in their entirety by following this link.

I've also updated my directories of online recordings by Arthur Collins, Byron G. Harlan, Collins and Harlan, Harlan and Stanley, Len Spencer, Spencer and Jones, and Edward M. Favor, to include material in the newly-launched National Jukebox, as well as the Belfer Cylinders Digital Connection from Syracuse University and the latest additions to the Cylinder Preservation and Digitization Project at UCSB.

You might also check out the Edison monthly release lists for May 1906, July 1906, and August 1906—none with a complete set of sound files yet, but it's a start.

Copyright Office Weighs Federal Copyright for Pre-1972 Sound Recordings

If you care about our shared cultural heritage of recorded sound (and the fact that you're here suggests that you do), you'll want to follow efforts currently underway to federalize copyright in pre-1972 sound recordings in the United States. One presumed result of federalization would be the establishment of an unambiguous public domain in the earliest recorded sounds, such as we enjoy for all other forms of creative work—novels, films, drawings, photographs, and so forth. See the Copyright Office's own website devoted to its study of the issue; the Historical Recording Coalition for Access and Preservation; a nice blog entry from the Future of Music Coalition; and the archived Twitter feed from the hearings of June 2-3 by K. Matthew Dames of Copycense.

Summer 2010 (left column)


"Schalldruck": Another Experimental Berliner Gramophone Recording From 1889

In April 2010, my optical film sound track method first retrieved sound from a spiral—a paper print of a Berliner gramophone disc recorded on December 14, 1889. I'm now pleased to offer a sound file created from an even older, longer, and more elaborate gramophone recording.

This paper print of a seven-inch disc is located in a "gramophone scrapbook" among the Emile Berliner papers at the Library of Congress and was brought to light by Stephan Puille in the June 2010 issue of The Sound Box. At 50 rpm, the total duration is just under two and a half minutes. Feel free to listen — — —or download the mp3 here. The date is announced in the recording itself—in German—as November 11, 1889, making this the oldest gramophone recording currently available to the public for listening, as well as the oldest known recording of singing and literary recitation in the German language. For an analysis of the recording and its historical context, together with a complete transcription and English translation, see the September 2010 issue of The Sound Box.

June 2010 (center column)


Techniques such as playing recordings backwards, superimposing them, excerpting and rearranging bits of them, altering playback speeds, and so forth were more widespread in early phonography than is generally thought. I'm just finishing up a new article that surveys this little-explored territory. In the meantime, here's a fun image from one of my sources—a Strand Magazine article of 1915—captioned "Playing a Gramophone Record Backwards":

Playing a Gramophone Record Backwards

Phonography in the News

April 11, 2010 (left column)

1889 Berliner

The Conquest of the Spiral

The optical film sound track method has now made a number of the world's earliest sound recordings "talk."

Until recently, it had only been applied to straight lines, such as traces on Scott phonautograms. But the technique has now been extended to spirals, greatly expanding its potential range of applications.

Read about the method and hear the results obtained from a paper print of a Berliner gramophone disc dated December 14, 1889, scanned for playback by Stephan Puille.

First Sounds "Facsimile Series" Launched

Just in time for the sesquicentennial of the April 9, 1860 recording of "Au Clair de la Lune," has unveiled a series of six new facsimiles of documents pertaining to the work of Édouard-Léon Scott de Martinville, inventor of the phonautograph.

David Giovannoni has not only produced improved facsimile editions of the Scott materials at INPI and the Académie des Sciences, but has also published some other recently discovered source materials for the first time, including the Scott dossier at the Société d'Encouragement pour l'Industrie Nationale (SEIN) and an important set of phonautograms found in the Regnault papers in the library of the Institut de France.

The new facsimiles are higher in resolution than those published earlier in the "First Sounds Working Papers" series, and they now also contain precise information about scale (mostly at 50% reduction) as an aid to researchers. Get them here.

April 11, 2010 (right column)


Joshing the Census Taker

On June 22, 1900, a census taker in Omaha, Nebraska, recorded a series of three names in a row: Vess L. Ossman, Daniel Quinn, Steven Porter. Four entries later came the name Cal Stewart. Those who know the history of the early American recording industry will immediately recognize these as the names of four of the most prominent phonograph performers of the era. Vess L. Ossman was a banjo virtuoso, Dan W. Quinn and Steve Porter were vocalists, and Cal Stewart was a monologist known for creating the character "Uncle Josh Weathersby."

But the information listed in the census doesn't match the biographical details of our phonographic luminaries. Vess L. Ossman is listed as a minister, Daniel Quinn as a coachman, Steven Porter as a blacksmith, and Cal Stewart as a clerk. The ages and places of birth aren't the ones we'd expect. And while an entry for the "real" Cal Stewart has yet to be found in the 1900 federal census, the other famous phonograph performers in the list turn up correctly elsewhere.

The overlap in names seems unlikely to be a mere coincidence. Probably a group of friends had agreed to play a trick on the census taker by assuming the names of talking machine "stars." Perhaps they'd have appreciated the following selections from the cylinder collection at UCSB (though none is quite as old as the incident in question):

Phono-Bretto Now available via Google Books is the Phono-Brettoa book containing the words to "about seven hundred" phonograms current in 1919. In some cases (especially monologs and descriptive sketches), the words were apparently transcribed by ear, offering insight into how listeners of the time understood what they heard.

October 2009 (right column)

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 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 [dead link April 2010] 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.

Late June 2009 (right column)

"The Monster by the Boys"

Ladies and Gentlemen, we are standing at the site of an enormous crisis.

Here in Catalina, Arizona, we have discovered a new kind of insect. It is enormous—almost unbelievable in size and strength. It can tear, rip, or mutilate any tank or instrument of war that this earth has ever been able to construct.

So begins a recent addition to the Feaster Collection—an amateur disc recording that deserves to become a cult classic. If Ed Wood of Plan 9 From Outer Space had tried his hand at radio drama, it might have come out something like this. Experience the three-part tour de force here, if you dare.

More on the Phonautograph

My First Sounds colleagues and I recently unveiled a batch of new findings that fundamentally reshape how we think about the world's oldest recorded voices. Here are some of my own ruminations (and the sounds, too, of course). The "script" and audio of my presentation on New Directions in Phonautographic History at the 2009 ARSC Conference are also now available online.


"PsychoPhone" Episode
of History Detectives airs

Their question: "Did Thomas Edison invent a machine to listen to the secrets of the dead?" Well, no, he didn't, but it gave the TV watching public a fine opportunity to experience phonographic luminary Tim Fabrizio talking about the PsychoPhone. (The best line, however, is Jerry Fabris's remark, "It would have to have been a very loud ghost.") Watch online here.

Taylor Made Recordings

The Center for Southern Folklore in Memphis, Tennessee, has recently launched a website featuring an interesting set of amateur original and off-the-radio sound recordings made in the Memphis Delta Region by Rev. L. O. Taylor. Listen to them here.

The Lost Photographs of Thomas Edison

Lewis Lueder was Edison's official photographer from 1913 through the 1920s. Robin and Joan Rolfs have made Lueder's photograph collection the subject of a new CD-ROM, and the Hearthstone Historic House Museum in Appleton, Wisconsin is home to a related exhibition through November 9, 2009.

Mid June 2009 (right column)

2008 National Recording Registry Announced

I'm happy to see they kick the list off with Nat Wills's inimitable "No News, or What Killed the Dog." See the whole list here.

A Phonographic Time Capsule

Some intriguing things are scheduled for the August 1, 2009 Phonovention in Auburn, Indiana—and for the year 2109 as well!

Peter Dilg will be recording the world premiere performance of a brass quintet ragtime number on wax cylinder which will go into a time capsule along with a CD documenting the same performance. It's expected that the cylinder will be playable in 2109, but that the CD won't.

See details here. And in the meantime, have you heard the cylinders extracted back in 2001 from a time capsule sealed in Colorado Springs in 1901?

"For the Record" documentary available on DVD

Leah Biel's much-anticipated documentary profiling record collectors was premiered to acclaim at the recent ARSC conference in Washington DC. Now you can own a copy! Check it out here.

More on the Phonautograph

My First Sounds colleagues and I recently unveiled a batch of new findings, fundamentally reshaping how we think about the world's oldest recorded voices. Here are my own ruminations (and the sounds, too, of course).

June 2009-April 2010 (left column)

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.

Early June 2009 (right column)

More on the Phonautograph

In case you haven't yet heard, the version of the "Au Clair de la Lune" phonautogram my First Sounds colleagues and I released to the world in March 2008 turns out to have been played back at twice the speed at which it was originally recorded. What we thought was the voice of a young girl was really a "chipmunk effect"—played here after two other examples at the same speed for comparison:   . Here it is at what we now believe to be the correct speed:   . When I imitated the new version during a trip to Paris in April, the response I got was: "Ah! That's how we sing 'Au Clair de la Lune' as a lullaby!" So we may have to give up our romantic notion of Scott recording the voice of his young daughter, but in return we may have a record of the way he sang his children to sleep.

Even so, the new version of "Au Clair" lacks the audible charm our initial playback had. Fortunately, another recently educed phonautogram makes up for it: Scott's last known phonautogram, an exuberant rendition of "Vole, Petite Abeille" ("Fly, Little Bee"):   . This is one of two 1860 phonautograms played back so far using my "optical film sound track" method. The other is a recitation in Italian of the opening lines of Torquato Tasso's pastoral drama Aminta . . : "Chi crederia che sotto forme umane e sotto queste pastorali spoglie fosse nascosto un Dio? Non mica un...." Scott writes at the bottom of this sheet: "I was wrong; it should be umane forme." By taking responsibility for the mistake, Scott indirectly identifies himself as the speaker here and, in all likelihood, in other phonautograms as well.

Find out more about these latest discoveries at

Phonograph Patents

Explore U.S. phonograph patents issued from 1913-1919, linked to images of the complete documents on the United States Patent and Trademark Office website: 1913 | 1914 | 1915 | 1916 | 1917 | 1918 | 1919. See also a list of all phonograph patents issued in the United States from 1878-1919 organized by current U. S. classification. [More details...]

Some new information on phonographic performer Clarice Vance

Despite what you may have read elsewhere, she wasn't born in 1871, and she wasn't born in Louisville, Kentucky either.  She was married at least three times, and her third husband died in 1928 under circumstances that suggested suicide.  Read more.


April 2008

phonautograph cartoon

"First Sounds" Makes Phonautograms Talk

In Fall 2007, David Giovannoni, Richard Martin, Meagan Hennessey, and I founded First Sounds, an informal collaboration of audio historians, sound archivists, scientists, and others who share our goal of making the world's earliest sound recordings available to all people for all time.  Working together with Carl Haber and Earl Cornell of Lawrence Berkeley National Laboratory, we recently succeeded in retrieving the oldest sound recordings audible today: phonautograms made between 1857 and 1878 and deposited at the French patent office (INPI) and Académie des Sciences, including a ten-second snippet of Au Clair de la Lune recorded on April 9, 1860 and premiered at the annual conference of the Association for Recorded Sound Collections at Stanford University in March 2008.  You can read more about our initiative and listen to all the recordings we've recovered to date at

Read and listen:
The First Prize-Winning Record
(sort of)
Climax 63 label

One of Charles Smith's tinfoil phonograph
Homemade Tinfoil Phonographs
Charles Smith turns a set of instructions from the Scientific American Supplement of July 20, 1878, into some working machines.

November 2007

An important addition to the Phonozoic Text Archive is a full English translation by George Brock-Nannestad of Leon Scott's French phonautograph patents of 1857 and 1859, made available here with the translator's kind permission.

Chronological List of
U. S. Phonograph Patents,

Patent texts are among the most significant official documents associated with the history of phonography, documenting a vast array of technical innovations—often ingenious, and occasionally whimsical or bizarre.  Here you will find links to all the American phonograph-related patents I have been able to identify for the period between January 1913 and December 1919.  Over 2,200 relevant patents were issued in the space of those seven years—almost as many as had been issued during the preceding thirty-four years.  While I hope the information will be useful to researchers in this form, it is also one step towards a broader goal of compiling a comprehensive directory of phonograph-related patents organized by type—stay tuned.

Show and Tell
 July 7, 2007
Lillian Halvosa Stewart
Lillian Evelyn Halvosa:
the first "Mrs. Cal Stewart."

I finally tracked down a nice copy of the sheet music to Roger Harding's "Pretty Kitty Clover," published in 1899 with a dedication to "Mrs. Cal Stewart," who is pictured on the front cover (above; click for larger version).  The famous phonographic storyteller Cal Stewart had married Lillian Halvosa in Westerly, Rhode Island, in 1898; however, Cal's second wife, Florence, is the "Mrs. Cal Stewart" heard on sound recordings made a few years later.  

Listen to an interview on Cal Stewart's life and work I did with Jerry Fabris on "Thomas Edison's Attic": Part One and Part Two



Original content copyright © 2007-2010, Patrick Feaster.