27 March 2011

The Voice of the Poet?

Picture it: the old man sits in a tall wooden rocker in his second-floor bedroom, a wolf skin draped over the top of the chair behind his shoulders. He is a strikingly familiar figure, bragging once to friends that "no man has been photographed more than I have." Images of him have been widely distributed, appearing in newspapers and magazines and on products. His unruly white-shocked benevolent visage appears even on his own brand of cigars.

You and your partner hauled the heavy machine in from the wagon.  The old poet wants to do the recording upstairs in his own room, and you agreed to this.  A few more stairs made little difference after hauling the wood and metal contraption all the way across the state.

After clearing books and papers off the poet's work table and carefully setting the machine down on it, you help situate the rocking chair as close as possible to the machine.  You instruct the famous man to speak as loudly as he can into the large metal  horn.  His speaking voice has been somewhat frail, and you are concerned about the quality of the recording.  The boss is especially interested in getting a good, solid recording of this national luminary.

Instead of reading one of his more famous works, the poet has chosen a short, unknown poem from just a couple of years before.  The problem here was length.  This poem, he has assured you, is quite short and will do the job nicely.  Its subject and its title is, very simply, "America."

The famed poet is excited about this new venture.  His eyes glitter with alertness as he leans forward, his mouth poised perfectly inside the outer brass flange of the horn, as you instructed him.

You give the machine a good crank to start the process. Your assistant lifts the needle and places it near one end of the now-swiftly-spinning cylinder.  You cue the poet.

Clutching a paper in his good hand, he reads, in a marvelous, strong voice, his poem:


Centre of equal daughters, equal sons,
All, all alike endear'd, grown, ungrown, young or old,
Strong, ample, fair, enduring, capable, rich,
Perennial with the Earth, with Freedom, Law and Love."

Pulling the Plug.

That's how, on an otherwise ordinary 1890 day in Camden, New Jersey, the actual speaking voice of the Good Gray Poet, Walt Whitman, the most American of American poets, was preserved forever.   

Except that it never happened.

That's right. Kill the lights, strike the set, yank the fake beard and send the "old man" home! Hit rewind and erase the trumped-up scene from your mind, quick!  

Forget I ever said anything.

Edison's men never recorded Whitman.  The scene I described is just a product of my imagination.


The recording you listened to was a fraud.

(WAIT! Not so fast!!)

[Sound Effects:  Scream of a needle yanked across record grooves.  Tires screeching to a halt.]

At least, that is, according to some sources!  There are places on the internet where that piece of audio is referred to matter-of-factly as "the fake Whitman recording." National Public Radio carried a short piece called "Walt Whitman Hoax" five years after its much-heralded rediscovery in 1992.  Its major detractor concluded, "the supposed Walt Whitman recording is a fascinating fraud."  Reading my sketch of Thomas Edison's engineers capturing the voice of Walt Whitman, some out there would accuse me of perpetuating a hoax, of giving strength to a historical lie.


And yet the power of this 36-second recording persists.  Many people, including multitudes of Whitman scholars and a number of dedicated wax cylinder recording experts,  still believe it is Whitman's voice. Indeed, this conviction impresses many laymen hearing the recording for the first time. There is something about the speaker's delivery, its verve, its delight showing through the haze of scratches and rhythmic shoosh-shoosh of the wax cylinder that still, despite rancorous, shrill arguments to the contrary, seduces people.  

They want it to be the voice of Walt Whitman.

The recording keeps cropping up, rising to the top of our cultural stew. 

Researching his sprawling epic, Gangs of New York,  Martin Scorsese wanted to lend some period authenticity to his character's voices.  What kind of now-lost accent might such characters have spoken with?  According to Roger Ebert, Daniel Day-Lewis drew upon the unique voice in the Whitman recording for his  character, Bill the Butcher.

Does the recording sound familiar?  It was commercially (and, for some, controversially) enshrined in American culture and consciousness in the Fall of 2009 when Levi's used it as the soundtrack of a new commercial in their Go Forth! ad campaign.  You may have heard it as it played--unidentified--over a series of celebratory, freewheeling, mainly youthful, jean-clad,  and sometimes disturbing black-and-white scenes of mostly urban American life.

Walt & Me. 

I first heard the recording in 1996, when I purchased the collection In Their Own Voices:  A Century of Recorded Poetry. I remember being amazed and somewhat dubious that the 4-CD box set supposedly began with a recording of Walt Whitman, that very 19th century figure.  Could it be true?  Could the tail end of Whitman's life have intersected the invention of the phonograph?  I was thrilled listening to the scratchy, unclear recording, some few words in a bold, interesting voice coming through the ancient mechanical storm of noise.

When I bought the excellent book and CD set Poetry Speaks:  Hear Great Poets Read Their Work from Tennyson to Plath in 2001 (I am admittedly a nut for this stuff) I noticed that it also included the Whitman recording.  But, as if to best the incredible claim of Whitman on record, this collection demotes the recording of "America" to third, behind recordings of Alfred, Lord Tennyson and Robert Browning! In his introductory essay,  no less a poet than Galway Kinnell voices some doubt as to the recording's authenticity, but ends with the epiphanic image that, the moment he started listening to it out loud in a church, "several bats flew down out of the gloom of the cathedral's vaulted ceiling, and while the poem was being read they flitted about, seemingly both musing and attentive, and when the poem ended, they returned to their shadows."  While his brain voices several objections to the document's authenticity, he really wants to believe in it, and finally yields "to mammals with keener hearing than our own."

Five years later, I shelled out for another collection, Poetry on Record:  98 Poets Read Their Work.    Again, here are the voices of Tennyson, Browning, and Whitman.  But this time, editor Rebekah Presson Mosby, who also co-edited Poetry Speaks, confesses that "Of the three Edison recordings, only the one of the American giant of modern poetry, Walt Whitman, has had its authenticity questioned."

Questions of Epistemology and Slippery Fish.

So, what is the story of this mysterious, intriguing recording?  How far back can we trace it?  And why has it gotten such a bad rap in some circles. . .and is revered in others?

History is a slippery fish.  Trying to assign an accurate picture to past events is a protean undertaking.  Those who attempt it are wrestling with the shape-shifting god.  Great are the heroes who undertake the journey, and they are sure to meet with many wonders, terrifying monsters, and snake-oil purveyors.  Every simple fact or piece of knowledge we accept as "history" is soiled and wrinkled from being passed hand-to-hand down through the generations.   Finally, we choose to believe or not believe a picture of the past assembled from ragged, mistreated, very possibly suspect jigsaw puzzle pieces.

What follows is my examination of some of those puzzle pieces.  They are meager and contradictory. There is a spate of investigation, much supposition, and some voices raised in opposition.  There is no smoking gun.  Not even a badly-done pencil drawing of a smoking gun.

It comes down to a cassette tape re-issue of an older recording of an even older recording, plus a couple of letters from the time, plus what people have made of these shreds of evidence.

I'll line these objects and opinions up on the table in front of you.  Then I'll weigh their assertions, yea or nay.  Finally, I'll tell you what I think of the whole historical hash, what I think these puzzle pieces add up to.

I'll tell you right now, it is going to be a long slog. Depending on how much time you've got, you might want to skip all the evidence-weighing and jump to my conclusions, starting with the section titled "Enough!".

In Our Time.

The Walt Whitman recording floated to the surface of the illimitable ocean of history most recently in 1992.  In that year, long-held rumors of such a recording coalesced into the handy solidity of a cassette tape from Texas whose contents electrified Whitman scholars at the University of Iowa, then radiated far and wide to the general public.  The story was featured in the New York Times, on National Public radio and CBS "Sunday Morning."   An Associated Press story spread the tale across the country.

The story erupted in 1992, was bruited widely about, and then had the door slammed just as quickly on it.  There has been no significant writing on the subject of the Walt Whitman recording, and no new discoveries or revelations regarding it, in 18 years.

Two armed opposing camps, the True Believers, who have no doubt that the recording is the voice of the great poet and embrace it as an object deserving literary veneration, and the Naysayers or Debunkers, who just as surely know it is a base fraud, nail their certainty ultimately to the flagpole of  a mere three or four documents from that year, 1992.  Three written essays and one radio show.  And of these four documents, three present the positives and the negatives of the case and leave the matter open.  Only one of the four sources concludes unequivocally that the recording is, in the author's words, "a fascinating fraud."

Here follow those sources.  They tell in brief the tale of the controversial recording's recent history.  In each case I've given a synopsis of the article's significant points, and provided a link or instructions where to find it.

If you want to continue this journey with me, I recommend that you read at least one of these sources.  They narrate events that I only subsequently refer to.  (The first one I mention, the New York Times article, is easy to link to and contains most of the salient facts.)

After familiarizing you with these sources, I've drawn up a timeline of the events in the life of the Walt Whitman recording that partisans on both sides of the great Walt Whitman Divide can agree on.  (Except one, of course:  the hypothetical recording session itself!)  A listing of hard, known facts, it is called A Brief History of the Whitman Recording.


On March 16, 1992, the New York Times carries a story by William Grimes entitled "Poem is Whitman's.  Is the Voice?"  In concise, journalistic style, the article sketches the known history of the recording and speculates on its authenticity, with opinions by a number of audio and phonograph historians, specialists, and Whitman scholars.  The article finally leaves the door of possibility open, quoting a historian of the Edison era as saying, "It's a long shot, but more surprising things have happened."

Four days after the Times article, National Public Radio's "Morning Edition" runs a story about the Whitman audio document.  The story repeats the recording's brief recent history and features the voices of various players in the tale:  Larry Griffin, the Midland College, Texas professor who first turned up the recording on a cassette tape in the college library; Ed Folsom, Whitman scholar and editor of the Walt Whitman Quarterly Review at the University of Iowa, who heard about it from Griffin and boosted the recording's reputation;  Sam Brylawski, reference librarian for recorded sound at the Library of Congress, whom npr consulted concerning the cylinder's audio quality; and, of course, "Walt Whitman" himself.  It is on the "Morning Edition" story that most of the nation first gets the chance to hear the Walt Whitman recording. (Transcript available from National Public Radio, or contact this author.)

Some of the excitement sparked at the University of Iowa by the recording's rediscovery still reverberates in professor Ed Folsom's article "The Whitman Recording" in his Walt Whitman Quarterly Review (Spring 1992).  Unique among these documents, Folsom's article contains a passionate defense of the cylinder's audio quality from Dave Beauvais, in a voice authoritatively knowledgeable about early audio recordings, to counter Sam Brylawski's skeptical opinion of it. Beauvais also enthusiastically analyzes the dialect of the voice on the cylinder, concluding that "the notion that someone might have set out to imitate such a subtle and nuanced archaic inflection strains credibility just a bit."  Like the other articles cited, Folsom's concludes that more evidence is needed, but that "new and illuminating discoveries are bound to emerge, whether or not they lead to confirmation or denial of the authenticity of the recording."

Folsom's words eerily appear to be predicting or paving the way for the next article, the one that seems to have changed the game and in some minds sealed the fate of the alleged Whitman cylinder recording.  In a 1992 issue of his Antique Phonograph Monthly, Allen Koenigsberg, an authority on early recording devices and the history of sound reproduction, published his article "Walt Whitman (1819-1892) Speaks?" (The article is easily findable as a PDF file via Google.)

Koenigsberg's Article.

This has become, for many, the definitive article on the authenticity of the Whitman cylinder. People who matter-of-factly discredit the recording as a fake, a phony, and a fraud are consciously or unknowingly accepting Koenigsberg's conclusions.  

That seems to me like a lot of historical weight to place on the shoulders of one article. 

There is no doubt that Koenigsberg did a lot of research, ferreting out choice tidbits of information and tackling angles not explored by the other investigative sources.  

He spends a lot of time and ink digging into the life and legacy of New York City's mysterious collector, Roscoe Haley, from whose "remarkable" collection the wax cylinder was said to have come. (On the original 1951 radio show which first aired the recording, host Leon Pearson remarked that the  battered cylinder from which NBC sound engineers struggled to extract a clear recording came from "the remarkable Roscoe Haley Collection in New York."  No further information is given about such a collection.)  Haley, he discovers, was a retired elevator operator in New York whose apartment, upon his death, was "filled to overflowing with papers, books, and records."

Furthermore, Koenigsberg investigates collectors of audio materials who might have had business with Haley, or might have purchased some of his "remarkable" collection after the eccentric New York pack rat's death.


At this point, however, there is a huge hole in Koenigsberg's investigation.  He tracks the oddball Roscoe Haley's estate and discovers two collectors who bought the bulk of the recordings from Haley's sister.  Koenigsberg diligently phones both collectors. One of these was Morton Savada, who, in Koenigsberg's words, bought "a cylinder phonograph and about 40 cylinders along with other personality recordings."  He then goes on to off-handedly state that "none of these could be located and he [Savada] remembered nothing special being set aside as a Whitman recording."

Despite the integrity of Koenigsberg's research, this seems like a large oversight to me.  The reclusive elevator operator packrat collector Haley is portrayed over and over again by Koenigsberg as being unwilling to show his cylinder collection to anyone, including potential buyers, such as Yale University.  Koenigsberg interprets this as suspicious behavior by a purveyor of fraud.  But couldn't it equally be the paranoid tendencies of a man overly protective of his "remarkable collection"? And if the packrat's collection was the confused mess it is portrayed as, would it not be easy to miss a single, maybe badly-labeled cylinder in the detritus?  Roscoe Haley, presumably, did not know he was going to die.  He most likely did not sort and label his collection and get it in order for future buyers.  Could not the cylinder in question, containing the conjectured Walt Whitman recording, have easily gone missing after collector Morton Savada's purchase of it?

Good Point.

Koenigsberg's other main point supporting his judgment of the recording as "a fascinating fraud" is a significant one:  no contemporary written accounts of the making of such a recording from Whitman's time have been discovered.  All of the 1992 sources explored or at least touched on this notable argument.

Whitman was by the time of the supposed recording (1890) a media celebrity.  His life and exploits were chronicled frequently in newspapers.  And no newspaper account of him being recorded on this relatively new invention has, as of yet, turned up.

Also, though the poet's daily doings were recorded by his "personal diarist," sometime nurse and assistant, Horace Traubel,  Traubel's  detailed 9-volume With Walt Whitman in Camden contains no references to a hypothetical Edison recording session either.  Nor do letters to or from the poet corroborate the story.

Not Entirely Silent.

The voices from the past are not entirely silent on the issue, however.   Though no news stories of the renowned poet submitting to the exciting new technology have appeared, two short letters clearly showing Edison's interest in pursuing such a project have.  These letters surfaced at the Edison National Historic Site as part of the flurry of interest surrounding the 1992 rediscovery of the recording.  All the sources mention these significant pieces of evidence, and Koenigsberg's article even includes facsimiles of the two letters, along with a concise description of the short correspondence.

On Valentine's Day, 1889, Edison's private secretary, A. O. Tate, replied to a request by one of Whitman's friends, Sylvester Baxter, to make a recording of the poet.  Tate wrote:  "Mr. Edison has received your letter of 8th instant, in regard to obtaining a phonographic record of the poet Whitman. He is very much obliged for your suggestion, and will endeavor to carry it out."

Tate then drafts a second letter to Edison's partner and president of the North American Phonograph Company, Jesse H. Lippincott, seeking approval for the project:  "In reference to the attached letter from Mr. Sylvester Baxter, Do you wish to act upon this gentleman's suggestion, and obtain a phonogram from the poet Whitman?"  Tate signs the letter for his boss, "T. A. Edison."

A Brief History of the Whitman Recording.

There turn out to be a number of pieces of the historical jigsaw puzzle that all four sources outlined above agree upon.  By placing them in front of us in chronological order, we can begin to glimpse a partial narrative of the central relic.  

Here, harvested from our four sources, are the main, agreed-upon events in the life of this mysterious audio artifact. [The single essential but still hypothetical event appears in brackets.]:

  • 1888:  Whitman writes short poem "America." 
  • February 11, 1888: "America" published in the New York Herald. Later that year also as part of an "annex" to an updated "impression" of Leaves of Grass.
  • February 14, 1889:  Edison's private secretary, A.O. Tate, drafts two letters discussing possible plans to record "the poet Whitman." (See facsimiles of these letters in .PDF of Koenigsberg article.)
  • September 3, 1889: Roscoe Haley, future eccentric audio collector, born in Tennessee.
  •  1890 [Hypothetical recording session.  Edison engineers capture voice of poet Walt Whitman.]
  • March 26, 1892:  Walt Whitman dies at his Mickle Street home in Camden, NJ.  He is 72.
  • August 5, 1951:  Radio show "Yesterday, Today and Tomorrow," narrated by Leon Pearson, features supposed recording of Whitman.  Pearson speaks of the difficulty encountered by NBC sound engineers in obtaining a clear recording off an old battered wax cylinder. Pearson says the cylinder came from "the remarkable Roscoe Haley Collection in New York."
  • 1974:  Center for Cassette Studies, CA, releases cassette version of the 1951 radio show.
  • January 15, 1982:  Roscoe Haley found dead in his NYC apartment, which was "jammed full of recordings, books, and papers."
  • circa 1982:  Audio items from the Roscoe Haley estate sold to 2 collectors, David Goldenburg and Morton Savada.
  • 1992: Article "Walt Whitman's Voice," by Professor Larry Don Griffin of Midland College, TX, appears in Walt Whitman Quarterly Review.  Offhand remark in the essay mentions the cassette featuring Whitman's voice, long available in the Midland College library under title "Voices of the Poets."
  • 1992:  Ed Folsom, Whitman scholar and editor of the Walt Whitman Quarterly Review becomes interested in the supposed Walt Whitman recording.
  • 1992:  Researcher at Edison National Historic Site finds two letters in the ENHS archives showing Edison's 1889 nascent plan to "obtain a phonogram from the poet Whitman." (See facsimiles of these letters in .PDF of Koenigsberg article.)
  • 1992: Investigative journalists and Whitman scholars consult a variety of audio experts and historians concerning the authenticity of the Whitman recording.  Sam Brylawski, reference librarian at the Recorded Sound Reference Center of the Library of Congress, appraises the recording's audio quality as "too good to be real."  Others, especially collectors of wax cylinder recordings, disagree.
  • March 16, 1992:  New York Times article "Poem is Whitman's.  Is the Voice?" by William Grimes.  Grimes concludes with the "long shot" possibility that the recording is real.
  • March 20, 1992:  National Public Radio's "Morning Editon" airs a story about the recording.  Features many of the important players in the controversy.  Listeners get to hear the recording itself. Concludes that more evidence is needed to rule on the recording's legitimacy, but finishes with an enthusiastic flourish from Whitman scholar Ed Folsom.
  • March 26-29, 1992:  Whitman Centennial Conference at The University of Iowa.  Whitman recording played as a featured event.
  • April 19, 1992:  CBS "Sunday Morning" features the recording.
  • Spring 1992:  "The Whitman Recording" by scholar Ed Folsom in his Walt Whitman Quarterly Review includes a solid defense of the artifact's audio quality and of the speaker's dialect, but concludes that more historical research is needed.
  • 1992:  Allen Koenigsberg publishes "Walt Whitman (1819-1892) Speaks?" in his Antique Phonograph Monthly.  Claims he "has been able to gather enough additional material to solve" the question of the recording's historicity.  Declares it, finally, "a fascinating fraud." (.PDF widely available online.)
  • 1996:  Recording appears, without apology, in cd collection "In Their Own Voices:  A Century of Recorded Poetry."
  • 2001:  "Poetry Speaks: Hear Great Poets Read Their Work From Tennyson to Plath" includes Whitman's "America."  In accompanying book, Galway Kinnell expresses qualms about the credibility of the recording, but finally decides to trust it.
  • 2006:  "Poetry on Record: 98 Poets Read Their Work" features the Whitman selection but notes that it "has had its authenticity questioned."
  • Fall 2009:  Whitman recording of "America" featured as soundtrack of Levi's "Go Forth" commercial.

Building the Historical Mosaic.

In my attempt to reconstruct and reanimate a historical picture from a scattering of factoids and opinions, I've begun by laying out this timeline on the flat plane of possibility before me.

Next, I'll take in hand one-by-one  the reasons and evidence arguing against or in favor of the historical authenticity of the supposed Edison cylinder recording of Walt Whitman.  These assertions appear in one, or more, or in all four of the sources I've been depending on.

These contentions are all based in facts, matters of historical record, and would therefore seem to appear as very solid pieces in our metaphorical puzzle.  But their hard edges blur and crumble when we realize that it is how each piece relates to the whole that is important.  A fact is a fact, but do these facts really support the larger picture, or contradict it, or are they powerless and irrelevant where the overall historical depiction is concerned?  Any of these pieces, in fact, might belong to an entirely different puzzle.

In any case, first I'll lay out the arguments against:

Reckoning: Fraud.

(From relatively minor points to more major ones):

CHOICE OF POEM:  Allen Koenigsberg wonders why the poem "America" presented on the cylinder is missing its final two lines as it appears in all printed versions.

LACK OF RUMORS:  Sam Brylawski of the Library of Congress (LOC) Recorded Sound Section remarked that there have never been rumors of a lost Whitman recording among historians of recorded sound.  Rumors of other lost celebrity voices are rife.

WHITMAN'S HEALTH:  Walt Whitman had suffered a series of strokes beginning in 1873 and was largely confined to a wheelchair at the time of the supposed recording (1890). Would he have been strong enough to produce the vigorous voice on the recording?

HISTORY OF FAKE RECORDINGS:  Such a phony recording would not be without precedent.  There is a long history of rumored recordings of famous people turning out to be hoaxes.  Famously, a supposed recording of President William McKinley's final speech before his assassination turned out to be a popular actor of the day.

PROVENANCE OF THE RECORDING:  Where did the hypothetical wax cylinder come from?  And mostly, where is it now?  What became of it after the 1951 radio show "Yesterday, Today, and Tomorrow," where it was supposedly played?  Examination of the actual cylinder would do a lot to settle the authenticity dispute.  Searches for it led to a series of dead ends with audio archives and collectors.  All searches led ultimately to. . .

ROSCOE HALEY & HIS "REMARKABLE COLLECTION":  Haley, a former elevator operator, is invariably described as "eccentric."  Upon his death, his apartment is found to be a chaotic clutter of old recordings, books, papers, and other debris.  Haley is notoriously tight-fisted about his recordings:  he won't let anyone see the cylinders, including prospective buyers, such as Yale University.  His negotiations with Yale broke down after he offered supposed recordings of Harriet Beecher Stowe, President Benjamin Harrison, and Whitman (reading a poem about Lincoln, a different work than that found on the cylinder in question).  Allen Koenigsberg says about the cylinder's authenticity:  "The provenance of the cylinder. . .is very much against it, associated as it is with the voice of Harriet Beecher Stowe, and other legendary, unauthenticated recordings" and "Roscoe Haley's unwillingness to show anyone the actual cylinders in his truly "remarkable" collection and his constant stonewalling on their whereabouts, certainly argues against their validity."  To further counter Roscoe Haley as a source of a genuine recording, Koenigsberg briefly sketches the source of "most of the legitimate historic cylinders from the early period of sound":  they were rediscovered in the basement of an Edison assistant and re-recorded between 1935-1937.  G. Robert Vincent, a former Edison employee heading this project, never made mention of a Whitman recording.

AUDIO QUALITY:  NPR's "Morning Edition" took the recording to Sam Brylawski, at the time Reference Librarian for Recorded Sound at the Library of Congress. His conclusion, summarized by reporter Katie Davis, is that the recording "sounds too good to be real."  Brylawski is often quoted as saying the recording "has either been exceptionally well-equalized or it's a fake."  Allen Koenigsberg devotes a short paragraph to the issue of audio quality as well, saying that he took the recording "to several sound experts, including the Rodgers and Hammerstein Archives, and they were unanimous that the recording contained too much bass response for a hundred-year-old recording, and the signal-to-noise ratio was much too high."

HISTORY IS SILENT: No contemporary sources from Whitman's time mention a recording session, even though the poet was by this time a national celebrity and, according to Koenigsberg, "Other incunabular recordings, like Barnum, Sullivan, Stanley, Nightingale, etc., were frequently commented upon in the press or in memoirs."  Scholars have yet to turn up a single private letter, memoir, or diary that documents such an event.  The lack of mention in Horace Traubel's multi-volume work, along with none in internal Edison or Lippincott records adds to the image of history as silent on the subject of a Walt Whitman Edison recording.

Reckoning: It's Real! It's Walt!

On the other side of our historical scale, let's now place the puzzle pieces in favor of the authenticity of the recording. These allegations each arose in one or more of the 1992 sources.  As we set them down, one at a time, the scales begin to slowly tip.  Where will they end up??

RUMORS:  Where Sam Brylawski of the Library of Congress said that there were no rumors among audio historians about a recording of Walt Whitman, the opposite was true among academics.  The National Public Radio story said that among Whitman scholars, "there had long been talk that a recording of the poet existed." 

LIKELIHOOD OF FAKE:  (New York Times article): "Most specialists in the history of the phonograph agree, however, that the possibility of outright fraud or a hoax is unlikely.  Audio experts who have heard the tape say they believe that it is a recording of a wax cylinder." 

NO CONTEMPORARY MENTIONS:  The Times article points out that at least some of our disappointment with the behavior of the past might just be a matter of perspective.  Specifically, it says that the fact that there is no further mention of recording Walt Whitman in the records at the Edison National Historic Site "is unremarkable, since neither Edison nor Lippincott kept logs or catalogues of cylinder recordings, which at the time were not generally made for commercial purposes, like records today, but as publicity vehicles for the company or keepsakes for those who recorded their voices."  Such recordings were one-of-a-kind artifacts in this era:  there was no process for duplicating or copying them. 

LEON PEARSON'S TESTIMONY:  Since the original cylinder itself is lost, we rely largely on the words of broadcaster Leon Pearson for information about it.  As the National Public Radio story said: "On the tape. . .the NBC anchor identified the voice as Walt Whitman's and explained that it had been taken off a wax cylinder that was originally recorded in 1890."  Although Allen Koenigsberg's article never overtly says it, his searches for the lost cylinder and his indictment of collector/pack rat Roscoe Haley dance on the brink of saying that the cylinder never existed at all.  Pearson clearly refutes that. He also fixes the date of the recording at 1890.  Clearly this does not authenticate the recording's age, but it is all we have to go on, and therefore functions as a working hypothesis.  As such, it meshes perfectly with the actual, real, authentic dates we do know about ("America" published 1888; Edison letters propose recording Whitman: 1889). 

AUDIO QUALITY:  Though Sam Brylawski of the Library of Congress branded the recording as sounding "too good to be real," the opinions of other experts in recordings of the era are sharply opposed.  First, and basically, the Times article said that "Audio experts who have heard the tape say they believe that it is a recording of a wax cylinder."  More than that, though, Ed Folsom says that "the consensus of those who have experience with wax cylinder recordings is that the recording is in fact an authentic 1890-era wax cylinder."  A longer quote in Professor Folsom's article is especially informed and enlightening. It is worth going back and reading the impassioned analysis of the recording's audio quality by Dave Beauvais.  Beauvais makes clear that the quality of wax cylinder recordings was ahead of its time, that, for particular technical reasons, they sounded much sharper and clearer than the dominant recording technology that followed, the standard shellac discs that most of us are familiar with. This opinion is echoed by other early recording experts, though not in any of the 1992 sources.

CHOICE OF POEM:  "America" is a relatively obscure Whitman poem, even among Whitman scholars.  If a fraudster were making a supposed Whitman recording, it is likely he would have chosen "O Captain! My Captain!" or "When Lilacs Last in the Dooryard Bloom'd" or some other iconic title with which the poet's name has been associated since his lifetime.  Despite his doubts about the recording's audio quality, Sam Brylawski of the Library of Congress calls this argument "compelling." Further, the poem is from 1888, with the recording supposedly from 1890.  This matches up well, especially considering the 2 authentic Edison letters discussing the possibility of making a recording, both dated Valentine's Day, 1889.
VOICE QUALITY:  The article that first sparked scholarly interest in the recording was "Walt Whitman's Voice" in Walt Whitman Quarterly Review.  Professor Larry Don Griffin of Midland College, TX cites dozens of historical references to the quality of Walt Whitman's voice from books, articles, letters, diaries, and so forth.  The sheer accumulation of these descriptions tolls audibly in the mind's ear:  one after another these witnesses from the past seem to be describing a voice very much like the one on the 36-second recording.  On National Public Radio, Professor Ed Folsom summed it up: "As I listen to the voice, one thing I'm struck by is how close. . .that voice matches the descriptions by Whitman's friends in--at exactly this time, 1889 or 90--friends like Horace Traubel, who described Whitman's voice as 'strong and resonant, full of music, a rich tenor, charms the ear and heart.'"

VOICE: DIALECT:  Cited in Walt Whitman Quarterly Review, audio expert Dave "Beauvais also offers a linguistic analysis of the voice on the recording:  'It exhibits a quaint and subtle regional inflection - a soft mix of Tidewater Atlantic and an Adirondack dilution of the contemporary New York accent - which has quite literally disappeared in our age.  No one speaks this way any more.  The notion that someone might have set out to imitate such a subtle and nuanced inflection strains credibility just a bit.'" 

EDISON LETTERS: The two internal letters in the Edison organization proposing a future recording of "the poet Whitman" are, in fact, the only truly documentary evidence in this entire historical squabble.  Other than the recording itself, they speak to us as the sole voices from history on this subject.  Once again, the date of these letters (February 14, 1889) is significant, dovetailing nicely with the known composition and publication date of the obscure Whitman poem "America" one year before,  and with the supposed date of the cylinder recording (one year later) as stated by Leon Pearson in its 1951 broadcast.


My friends, we have made it alive through this blistering, arid landscape of facts and sources.   Here we stand together on a plain of new possibility.

Our cyclopean scale has weighed all our carefully-circumscribed pieces of evidence. And still the answer is unclear.

What is clear is that we cannot, from the evidence alone, prove the story, yea or nay.  That's why it remains an object of discussion! 
But we have heard enough!  We get the picture!  We are tired of other people's opinions and of the dusty relics of the past!  The time has come to topple the great Scale of Evidence and Opinion!  Topple it till the facts and theories come tumbling down, and then bulldoze them all under the forgiving earth!

(We will not forget them.  We have taken them into our bodies.  They are part of us now.)

We need new life in the argument!

It's time, finally, for me to tell you just what I think of all this!

What I think & Why.

I believe that the 36-second recording is truly the voice of Walt Whitman.  Here's why:

  • The timing of the thing.  The only pieces of solid, actually-confirmed historical evidence in this story are the two internal letters  from the Edison organization.  The letters clearly show that Edison was strongly considering recording Whitman.  Edison's secretary says his boss is "very much obliged for your suggestion, and will endeavor to  carry it out." The letters were dated Valentines Day, 1889, and the recording, according to broadcaster Leon Pearson, was done in 1890. The other key fact here is that the poem Whitman recorded, "America," was written and published in  1888, making it quite contemporary with the goings-on in the Edison organization. This confirmed chronology means that for a fraudster to have produced the recording, he would have had to know not only about the private Edison letters, but also have been aware of  the somewhat obscure poem "America," including its dates of composition and publication,  and would then have had to,  based on these, pass off his fraud as dating from 1890. A preposterously unlikely chain of hypothetical events!  So, if we are judging the case from hard historical evidence, we have to lean toward the recording's authenticity.
  • No credible theory about such a fraud has been presented by anyone.  In fact, there are no stories at all about who might have commited the hoax, nor when, nor why.  No one ever profited from the recording.
  • Koenigsberg's article doesn't prove anything.  I have to give kudos to Allen Koenigsberg for the sheer legwork and phone research he put into his take on the recording, but, finally, his research really doesn't add up to a solid conclusion.  The bulk of his article is a kind of guilt-by-association heaping of aspersions on the head of Roscoe Haley, the eccentric New York collector.  But as I pointed out, first, Koenigsberg's own words contain a huge hole into which the sought-for Edison cylinder could have dropped.  Secondly, his whole search for the cylinder is essentially pointless!  We have the voice of respected broadcaster Leon Pearson telling us that the cylinder existed and that NBC engineers had a hard time getting an airable recording off it.  Koenigsberg paints Roscoe Haley as an eccentric pack rat who tries to pass off fake recordings left and right, and makes the leap that therefore THIS recording is fake.  But the fact is that he doesn't even prove that Haley's recordings of others (Harriet Beecher Stowe, Benjamin Harrison) were fake, and so his guilt-by-association theory crumbles to pieces.  Finally, solid arguments (elsewhere in my jeremiad) refute Koenigsberg's other points that no contemporary documents mention the recording event (in fact the ONLY verified historical documents in the case, the Edison letters, included in facsimile in Koenigsberg's article, point clearly toward it) and that audio experts question its sonic quality (others stand firmly behind it.)  Mr. Koenigsberg's article, then, (the touchstone article that has discredited the Walt Whitman recording since 1992!), though well-researched and concisely-written, doesn't, finally, prove anything.
  • Whitman Scholars think it's real.  Professor Ed Folsom, of the Walt Whitman Quarterly Review at the University of Iowa, who has been the standard-bearer and champion of the Whitman recording since it surfaced there in 1992, says,  "In fact, virtually everyone I've talked to in the Whitman community (academic and more general) as well as in the larger American literature community considers the recording authentic." Though this, of course, does not prove anything, I'll put my trust behind these experts on Whitman's figurative literary "voice," just as Galway Kinnell put his trust in the superhuman hearing of the bats in the church.
  • Ed Folsom's perspective on the silence of history.  One of the strongest arguments against authenticity is the nearly-complete lack of documentary evidence from 1890.  Wouldn't Whitman have mentioned such a momentous event as his voice being preserved for posterity by a brand-new technological process?  Wouldn't some correspondent of his have reported it in a book or letter or diary?  But, Ed Folsom states, "The fact is, however, that by the 1880s, Whitman was something of a celebrity, and most days people would line up outside his little Camden home and wait to get an autograph or to simply talk with the infamously controversial poet.  There were disciples, gawkers, students, fans.  Whitman, depending on how he was feeling, would let them each come in for a few minutes, try to sell them a book that he'd sign on the spot, and move on to the next person.  I think that when Lippincott (or whoever showed up with the wax cylinder recorder) appeared, it would have struck Whitman as just another collector, someone who wanted an autograph of his voice.  There would have been no hint that he was actually recording his voice for posterity, since wax cylinders were not replicable, were one-of-a-kind items, and didn't seem particularly durable.  It's not something he would have dwelled on.  We look back on it now and think that surely Whitman would have made a big deal out of being recorded, but being "recorded" in 1888 was something so new and unknown that I'm guessing it was simply absorbed into the flow of Whitman's daily visitations from an endless stream of people who wanted a little piece of the famous poet before he died." 
No Smoking Gun.

I want to step back for a moment to repeat the bald truth of it:  we cannot, from the evidence at hand, prove unequivocally that the Whitman recording is either real or a "fascinating fraud."  Until a letter turns up with someone's memory of Whitman subjecting himself to some kind of new voice-recording machine. . .or else someone comes forward to tell how her grandfather, an actor,  bragged about how he put one over on decades of gullible rubes when he faked the recording, with specifics of how he did it. . .until some further scrap of evidence comes through, we are destined to speculate.
I've stated a few reasons why I think the artifact is a true recording of the Good Gray Poet.  Others will have other reasons for their opinions.

When we strip away everything--the "remarkable Roscoe Haley collection" and the opinions of audio experts and the rumors or lack of rumors and even the Edison letters--it all comes back, finally, to the recording itself. 

Listen to It!

Having come this far, Reader, listen to the recording again.  Go ahead, listen to it!

I just listened to it again, 2 more times.  I have listened to it hundreds of times in all.  It has, in a sense, become too familiar to me.  

I remember that when I first heard it, I could not be sure of the words.  I had to look them up in the accompanying book to confirm what I was hearing.  But now the words, and every subtle nuance of them, are so familiar to me that in a sense I cannot hear the recording as I did on a first listen.

How is it for you, Reader?  Assuming you had not heard this recording before you started reading this blog, could you, at first, understand all the words?

Here, once again, is the full version of Whitman's 1888 poem, "America":

Centre of equal daughters, equal sons,
All, all alike endear'd, grown, ungrown, young or old,
Strong, ample, fair, enduring, capable, rich,
Perennial with the Earth, with Freedom, Law and Love,
[A grand, sane, towering, seated Mother,
Chair'd in the adamant of Time.]

Mystery of the Missing Lines.

You'll notice that the speaker on the recording leaves off the final 2 lines of the poem.  And that is an important point.  The speaker comes to a  natural ending cadence, a sort of hoorah of finality, on the words, ". . .with Freedom, Law and Love."  The last two lines are not chopped off the recording, they are unspoken.

So, whoever did this recording decided that the poem ended nicely with those words, "Freedom, Law and Love."

Once again, there is no proof here.  But does it seem likely that  a hoaxer would make such a decision, to omit the last two lines of a poem that he would have had to unearth from where it reposed in its obscurity in a particular edition of Leaves of Grass?

And doesn't that seem like something the poet himself would feel perfectly justified in doing?

A Better Job 

What else do you hear, Reader?

I am a big fan of the sound of poets reading their own works.  But I have to admit that they are often not the polished "best" interpreters of their words.  They are usually not trained in elocution or dramatic interpretation, so they do not follow the rules.  And yet, in another sense, they are the best readers of their works, or at least the most interesting, simply because this is their poetic voice, this is the cadence of their breathing and their words.  They have a deep connection with the words that a professional interpreter might struggle for years to understand.

So that's another thing I have to ask about this recording:  wouldn't a trained actor or someone dedicated to producing a convincing hoax have done a better job?

What do we hear here?  Delivered in an interesting dialect that sounds like a mixture of New York urban with a rural softening at the edges, we hear a heartfelt pronouncement of the poem, with nearly oozing loving care on the typically Whitmanian list of words, especially, 

"Strong, ample, fair, enduring, capable, rich,"

each of which he honors with a power and a place of its own; he slams each word powerfully down in front of us, a chunk to consider by itself.   And he feels the individual power of these carved auditory hills:  "Strong, ample, fair. . ."


Listen to that one very broad, very Whitmanesque word, expressing so much of his embrace of plurality, of his desire to encompass everything American:  "ample"!  He fairly barks it out here, placing that intial 'a' in his throat and rocketing the word forth!  "Ample"!!

I am convinced by the sound.  By the powerful delivery.  The speaker loves these words and knows them intimately.  

An actor or a trained speaker would have done a more practiced, perhaps more varied, more trained job.  It would have been at once more dramatic, but not as not as heartfelt, not as loving, not as embracing, not as married to each word.  What we hear here is natural and affectionate, not studied and dramatic.  The speaker knows these words so deeply, knows the complex relationships and the nuances behind them, and loves them so clearly, that he does not have to dramatize them with vocal interpretation techniques.


As I said, I am a sucker for the spoken word, and especially for archival recordings of poets reading their own works.  So for me, sound trumps all. The sound of the voice on the recording blows all logical considerations out of my head like a purging wind. There are good, very good reasons to believe that this recording is a fraud.  But in my childish way I won't hear of it.  After all, I have heard that voice!

And yet. . .

It is the measure of the power of this archival recording that it continues to cause controversy.  The True Believers are still lined up against the Debunkers or Naysayers.  All stick firmly to their guns.  

I have been working on this article for a long time, Reader.  So I have been living with the Whitman Recording and all its history for a while.  

And so it is with a kind of amazement and certainly amusement that I take a look at my own feelings and thoughts about the artifact.  It is powerful indeed.

It even has power to shake my faith.

Creeping Doubt. 

Before I wrote the section "Listen to It," there was a pause of some weeks, Reader. Because I, myself had gone back and listened to it. 

And doubt began to creep into my mind.

Now, don't get me wrong:  I firmly stand behind all I've written above about the natural quality of the voice on the recording.  It is not that.

It is the audio quality.

I started comparing this recording with other nearly-contemporary (1890) recordings I could find on the internet.  And it really does sound different, in the long run.

There is much more of a bassy quality to the voice.  That's the main thing.

There are, as I have said, disagreements about this.  The strongest authority in defense of the recording's audio quality came from a very long and detailed, very authoritative letter from a man named Dave Beauvais, which he sent to most of the major players in this epic drama back in 1992.  Ed Folsom quoted Beauvais in his article, and the quote is reflective of the rest of his long letter.

I know of at least one other wax cylinder collector who agrees with the assessment.

But listening to the evidence myself, doubt began a-creeping.

Put side-by-side with other recordings of the era, this one sounds far more present, with a bass quality to the voice missing in the others.

I emailed a few wax cylinder collectors.  They added to my doubt by declaring the recording a phony!

Audio Quality Questions.

But I look at at this way.  As long as there is doubt, then the other considerations I've cited win the day as far as I am concerned.

Mainly, I think of all this recording has been through, some of which we do not really know about for certain.

What I have linked for you to listen to is not an untouched original version of the recording.  There is, in fact, no such thing.  We do not have it available.

This is part of what launched Allen Koenigsberg's search for the lost cylinder.  From the actual original cylinder, the question of authenticity would have been relatively easy to settle.  

But what we have to listen to is such a latter-day artifact, an audio palimpsest, compromised by a history of copying and re-copying and audio processing.  People all along the way have sought to make the recording more listenable.  

Most of these efforts since the 1992 surfacing of the recording can be traced.  The one that cannot dates back to 1951.

Historical Hiccup.

We are told, over and over, that NBC engineers had a difficult time lifting a hearable voice off of a badly-damaged wax cylinder in order to air it on the "Yesterday, Today & Tomorrow" program.

This is the crux of the matter, as far as I am concerned.

Because what we hear on the recording today is quite audible.  In fact, it is TOO audible to be an 1890 recording, according to some audio experts.  

Therefore, something happened between what NBC engineers could discern on the badly-damaged cylinder and what we hear today.   

That's all I'll say.

I do not have any knowledge about the state of the art of audio processing in 1951. As of yet, I don't know what was or was not possible then.

But the historical record hiccups:  NBC engineers could barely lift an audible voice from the badly-damaged cylinder. . .and we have a recording that is, according to some, "too good to be true."


So, though I firmly believe that the mysterious recording really is the voice of Walt Whitman, there is another character living in me who has some doubt, and some questions.

A famous quote of Whitman's finishes it off:

"Do I contradict myself?
Very well then I contradict myself,
(I am large, I contain multitudes.)"
       -Whitman, "Song of Myself"


  1. This is a wonderful analysis, Jim. I love how you began by placing us in the moment -- then said it never happened. You had me coming and going in both directions. I finally found myself with heels dug forward, reluctant to be pulled in by your final logic that this may be too good of a recording to be true. But ultimately--I want to believe it's Walt. Thank you for this masterpiece. Bravo!

  2. Wow - I thoroughly enjoyed reading this!

    I will admit that within seconds of listening to the Tennyson and Browning recordings that you posted in your comment on my blog, I literally felt my heart sink. Though I know very little about Edison's wax cylinder method, the stark disparity in the recordings caused a flood of doubt to well up inside.

    That said, in my "gut" I believe it's Whitman. (really logical and scientific of me, I know!) In regard to his choice of poem, as a man who was passionate about his country and it's people, I think "America" is precisely the kind of poem that Whitman would have chosen for a recording. Like you, I am also convinced by the speaker's delivery and the natural, unhesitating intimacy that he seems to feel with his words.

    The fact that the Edison letters exist without a doubt is also compelling to me. Whitman was fascinated by the act of preservation through photography and he was also passionate about oratory - those combined factors make me believe that if a recording session was in the works, he wouldn't have let the opportunity pass.

    I also think you make an excellent point about the absence of a fraud story. If someone were going to fake a recording, certainly there's something that they are trying to gain. Why would someone go to the trouble, if they weren't going to try to profit from it?

    Again, I very much enjoyed reading your analysis. Thank you for sharing.

  3. Really interesting post here, Jim. I didn't know there was so much controversy over the authenticity of the recording. My feeling? It's Whitman. And if not, his voice is alive in the words -- so vocally or not, he lives in the text.

  4. Jillian: I like what you said about Whitman's voice living in the words here, whether or not it is really the man himself. I thoroughly believe that, being a big memorizer of poems myself. To accurately commit a poem to memory you have to get inside the poet's breathing and cadences of language and subtle relationships of words. I often feel that the poet is speaking through me when I do that. It is not my voice, in a way.

    And yet, there is something even deeper going on here, I feel. Maybe just wishful thinking on my part.

    Nicki, you spoke of the "natural, unhesitating intimacy" of the speaker here. Very well put. Exactly what I feel listening to this. This is not a delivery or oration of this poem, it is just some guy speaking it. He loves the words, he caresses them and gives them weight. But his familiarity with them and all the subtlety of every relation is complete. I went over all that in my piece, of course.

    If a piece of hard evidence finally turned up showing that the recording was indeed not Whitman, I would be disappointed but in a sense doubly impressed. The speaker so lives inside the words here.

  5. Very stimulating. Hats off to you on a great article.

  6. Fascinating article. As to your "creeping doubt"....

    If Dave Beauvais claims Edison cylinder recordings of comparably high quality existed in 1890 - it seems to me his assertion is either true or false. Either Beauvais is right about that or he isn't - regardless of whether or not this particular recording is the authentic voice of Walt Whitman.

    I believe it IS Whitman. The Tennyson and Browning recordings are poor, but that doesn't prove all early recordings were likewise poor. Could the richness of the speaker's voice have something to do with the recording's quality?

    Listen to this Youtube upload, which is labelled as an opera coloratura performing for a cylinder recording of 1895. If this is authentic, I have no problem believing the Whitman one is too:


  7. Here's a youtube upload that purports to be an 1890 Edison cylinder recording of Tchaikovsky. I don't know if it's been authenticated, but if it's not a fake, the clarity is comparable to the clarity of the purported Whitman:


  8. Believe it or not, I came upon this website by searching Roscoe Haley...not Walt Whitman. Roscoe is my mother's uncle. I knew nothing of his life... Wow this is interesting stuff !

  9. Now THAT is fascinating, Anonymous!

    I would love to hear ANYTHING you can pass along about Haley. Have you found out anything else since reading my article?

    I think the man bears some investigating. A "character," certainly, but who isn't? Allen Koenigsberg definitely did a hatchet job on him. (Have you contacted HIM, by the way? Maybe he would share with you Haley's letters to Yale. . .He did write back to me.)

    If you have anything to say about Haley, write me at: jimhermanson@q.com


    1. I am always glad to correspond with those who think this is a 'Whitman' cylinder. The triumph of hope over experience, as they say.

      Allen Koenigsberg

  10. I came here today because of the Language Log link to your post. I have a purely subjective and anecdotal remark to make about your comment that "I do not have any knowledge about the state of the art of audio processing in 1951."

    Back in the 1980s, I was a jazz DJ at KZSU at Stanford. For the most part, I played things from about 1955 on, with very little stuff from earlier. At no point was this a conscious decision; in fact, it was only after a couple years that I noticed the pattern. When I thought about why, I decided it had to do with an unconscious recognition of mine that the sound quality of recordings increased considerably about 1955; if you know the history of jazz recording, then you would connect that with the studio owned by Rudy Van Gelder, which was the best studio from then through the 1960s. Earlier stuff just didn't sound as good!

    So my purely anecdotal and ungrounded observation is that sound quality improved in the mid-50s, so that whatever else is going on with this recording and its 1951 connection, it's just before a significant improvement in recording techniques and technology.

  11. He sounds like FDR, who Gore Vidal always said had a very common Hudson River Valley accent for the time. It's definitely a vintage dialect, and the personality sounds very much like many late 19th-century Easterners I have known.