Two Joes and a Jill

On December 12, 1942, The New Yorker published a 7000-word profile, entitled “Professor Seagull,” by Joseph Mitchell.  The subject was Joe Gould, a 53-year-old Greenwich Village eccentric, who was said to be writing an “Oral History of Our Times,” consisting of a record of conversations he had overheard over the last decades and essays related to these conversations. It was, Gould claimed, several times the length of the Bible and, most likely, the longest book ever written.  However, having learned that the Metropolitan Museum had stored its most valuable holdings into a bombproof shelter for the duration of the war, Gould had placed his history in the stone cellar of a chicken and duck farm on Long Island, owned by a woman who was out of the state, making it unavailable for Mitchell to read.

“Joe Gould is a blithe… little man,” is how Mitchell introduced him.  He had been born outside of Boston in 1889 into an old Yankee family.  After graduating Harvard, he had spent three years roaming around Canada and another winter measuring the skulls of North Dakota Indians as part of a eugenics study.  In 1916 he had come to New York and settled into the bohemian life.  He was a newspaper reporter.  He reviewed books.  He contributed to the The Dial.  E.E. Cummings, Malcolm Cowley, and William Saroyan were among his friends and admirers.[1] He was toothless and wild-bearded.  He wore castoff clothes and slept on benches and in flophouses.  He supported himself by begging for food and money from friends and strangers. His personal habits might have been off-putting, but he was writing this mammoth – perhaps consequential – book.

Then, on September 15 and 26, 1964, The New Yorker published Mitchell’s two-part, 37,000-word “Joe Gould’s Secret.”[2]

“Joe Gould was an odd… little man,” is how Mitchell now introduced him. [“(O)dd,” you will note, not “blithe.”] He was “unemployable.” He was “dirty,” “nonsensical,” “sarcastic,” and “scurrilous.”  He stole.  He was “a bummy-looking little red-eyed wreck of a barfly.” His behavior was often unbearable. But Mitchell still felt affection for and protective of him and not displeased that “Seagull,” had done him a good turn.  It had authenticated Gould, making him more able to reap “donations” from those he hit upon.  It had even attracted an anonymous benefactor, who had provided funds for Gould’s food and lodging for five years, until she had unexplainedly cut him off.  In 1952, Gould had collapsed on the street and been hospitalized.  He remained under institutional care until his death in 1957 from arteriosclerosis and senility.

But “Secret”’s major revelation was Mitchell’s belief that “The Oral History” did not exist.  Mitchell explained that he had originally been attracted to Gould because he viewed him as representative of a type of New Yorker, of which Mitchell was himself one: “the solitary nocturnal wanderer.”  And when Gould’s continued evasions about displaying “The Oral History” led Mitchell to conclude that was because had no notebooks to display, he saw himself again.  For, Mitchell wrote, he had long planned to write a “Ulysses”-like novel about a young man wandering through New York City.  He would mentally write – and discard – entire chapters.  He would envision its green, gold-lettered binding.  It was utterly “real” to him, but not one word was on paper.

This had led Mitchell to a “respect” for Gould.  The world had enough books, Mitchell concluded, and instead of adding another unnecessary one, Gould had created a single magnificent character, “The Eccentric Author of a Great, Mysterious, Unpublished Book,” who was “more complicated… than most of the characters created by novelists and playwrights of his time.” And out of this respect, Mitchell had kept his discovery to himself until Gould’s passing.


There things stood until July 27, 2015, when The New Yorker published “Joe Gould’s Teeth,” a 15,000 word article by Jill Lepore, which jarringly, enthrallingly re-evaluated Gould, Mitchell, and their work.  (It also proved to be a distillation of a 40,000-word similarly-titled book, published this year by Knopf, which comes augmented by an additional 60 pages of  endnotes, assiduously extracted from archives, collections, interviews, and privately held papers.)[3]

Lepore’s Joe Gould “could hardly have been more different” than Mitchell’s.  She believed Gould autistic.  (Ill-equipped for college, he was expelled from Harvard his senior year and did not receive a degree for another five years.) She considered him “unhinged,” “tormented,” “vicious,” “floridly mad,” and “terribly, terribly ill.” He may have had a syphilis-related “psychosis and dementia.” He was no “kindly eccentric,” but a “psychopath.”[4] She revealed information Mitchell had failed to uncover or had known but withheld.  Gould had a history of – and had once been arrested for – groping women.  He had been in held in mental hospitals multiple times.  The institution in which he spent his last years was one and the treatment to which he’d been subjected had been brutal.  And throughout his life, Gould had been possessed by a virulent form of America’s racial “madness.”

This madness had led Gould to North Dakota, where he measured, not only the skulls of Native Americans, but their “redness.” It had led him to be embraced by – and to embrace in return – the ardently anti-Semitic Ezra Pound.  (This embrace, which persisted after Pound’s arrest for treason, Lepore persuasively argues, accounts for Gould’s benefactor, whose identity she has also discovered, ending his support.)  And it led to Gould’s obsession with the Afro-American sculptor August Savage, whom Mitchell left unmentioned, but whom Lepore has positioned as a central figure in Gould’s life.[5]

But “Teeth”’s major revelation was Lepore’s belief that “The Oral History” existed.


Or maybe not.

The tease below Lepore’s by-line on her New Yorker article was, “Joseph Mitchell thought that Gould’s Oral History didn’t exist.  He was wrong.”  (Italics in original.)  Her book’s fly leaf asserted “The Oral History” “did in fact once exist.”  (Italics mine.) Lepore, however, never made that direct a statement.  But she posed the question, speaking of Gould, “Is a book a book if it has no readers?” Which seems to assume that someone has written it.  And she wrote, “(H)ardly anyone had read much of it… and no one had read all of it.” Which suggests there was an “all” to read.

So, I asked myself, who read or saw what when?

Gould told Mitchell he began writing his history in 1916 and had filled 267 composition books with 9,250,000 words (or 35,000 words per book).[6] Gould also told people he began in 1914, 1917 and 1918; and Lepore concluded he had actually started in 1912.  But since it is Mitchell’s account that’s in question, let’s take 1916 as the jumping off point.  It would mean that, by 1942 Gould had churned out 365,000 words (10.7 notebooks) annually.  Lepore has Gould writing through late 1947, which, at his prior pace, would have added another 50 for a total of 317 (11,000,000 words).

Mitchell, by my count, saw 12.  Seven were revised versions of a single topic, the death of Gould’s father.  One was about the death of Gould’s mother.  Three were revised versions of something about tomatoes.  One discussed being drunk.  Eliminating duplicates, that’s four books, or 1.3% of what Gould claimed to have written.  Aaron Siskind, a photographer and custodian of five of these books, told Mitchell he suspected Gould wrote about the same subjects repeatedly, tearing up what he’d written and beginning again.  Dwight MacDonald, the critic/essayist, believed Gould simply threw out what he’d written the previous year every January and began again, as in a ritualized cleansing in pursuit of a chance at life anew.

Lepore’s case for “The Oral History”’s existence is made through Gould’s friends, the novelist Millen Brand, and the poet Horace Gregory.  Millen wrote Mitchell, after the appearance of “Secret,” that he’d read about 35,000 words (or one notebook) of the history, (which he thought Gould had subsequently destroyed). Brand and Gould’s friendship had ended in 1934, so Brand was relying on a 30-year-old recollection when he wrote Mitchell and would have had no idea what part of the history existed when Gould and Mitchell met.  Assuming Brand’s memory was accurate and assuming he read what he read toward the end of the friendship, Gould would have been at work for 17-years and would have filled over 180 notebooks.  So credit Brand with having read less than 1% of what existed at the time and, having read only one notebook, would not know how much of the rest was duplication.

Gregory wrote, in The New Republic in 1931, that Gould’s oral history filled 50-to-100 notebooks.[7] Gregory added that the work was “in its eighth definitive version.”  (Why an “oral” history would be revised eight times is unclear.)  If Gould had been writing since 1916, he would have filled 160 notebooks by 1931, so either he was tearing them up and starting over, or he stopped tearing them up in 1931 and doubled his annual output to reach 267 by 1942, or Gould was fibbing when he met Mitchell, or Gregory was fibbing when he wrote what he did in 1931.  Furthermore, when Mitchell interviewed Gregory in 1942, Gregory said he had read “at least” 50 copy books,[8] which suggests he’d mis-estimated the low end of the total in 1931, or his memory was shaky when he spoke to Mitchell, or he was fibbing in 1931 or 1942 or both.

In assessing Gregory’s recollection, it is interesting to note Lepore’s suggestion that, by 1931, a number of Gould’s male literary friends had “contrived” to “protect” him from further institutionalization or arrest by fostering his “legend.” By promoting Gould’s genius, they hoped to help him sell his book.  The proceeds would get him off the street and, if he wouldn’t stop assaulting women, he “could more easily get away with it.”[9] Having Gould profiled in The New Yorker, she says, was part of this campaign.  The need to “protect” Gould was certainly no less by 1942 than it had been in 1931, so Gregory was likely to have been looking out for Gould when he spoke to Mitchell and inflating Gould’s productivity.

Witnesses aside, Lepore also argues for the existence of “The Oral History,” on a sort-of mini-infinite-number-of-monkeys-duplicating-Shakespeare basis.  Gould wrote, she says, “ceaselessly,” “endlessly.” He had, she diagnoses, “graphomania,” an obsessive compulsion to write.  So if he was compulsively writing, why not “The Oral History”?  (Lepore does not seem to be using the term in a strict “psychiatric context,” since, according to Wikipedia, graphomania   “results in writing rambling and confused statements, often degenerating into a meaningless succession of words or even nonsense…”)

Among the things Gould wrote were diaries.  Lepore located ten (800 pages), and, in them she found he often mentioned time spent writing.  “Wrote what?” she asks.  “Not the diary.  Something else.”  These diaries don’t prove the oral history didn’t exist, she says.  They “suggest exactly the opposite.” “(E)xactly” seems strong.  (Can a suggestion ever be “exact”?) If “The Oral History” existed, I wondered, why did no one see it? Why didn’t Gould show it to Mitchell?  Why didn’t he show it to the publishers who asked for it?  Why, when he applied for a Guggenheim and was asked to submit a portion, he submitted instead a nine-page synopsis, which even Lepore termed “gibberish.”[10] Why didn’t the owner of the duck and chicken farm have a name?  Why didn’t the farm have an address?  Why didn’t anyone find that? (Lepore suggests that the “farm” was, in fact, the Manhattan State Hospital for the Insane, on Long Island, where Gould was confined in 1929.)


Mitchell was a Southern gentleman writing in a decorous age.  He  offered readers a Joe Gould shaped by Mitchell’s humor and tolerance, unchecked by more than cursory validation.[11]  His Gould seemed witty and bright, if strange.  He could have put the touch on Nathan Detroit outside of Damon Runyon’s Mindy’s.  He might have sat down to dinner with the Sycamores of “You Can’t Take It With You.” Lepore, a contemporary academic, writes for an audience which accepts – indeed, expects – the acid-scarred, not the sugar-coated.  And Lepore’s attitudes and aims, her writing suggests, are different than Mitchell’s too.

Take Lepore’s treatment of Savage.  A woman of color, a talented artist, most of whose work was lost, destroyed, or forgotten, a documented victim of racial prejudice, and a target of white male aggression in the person of Gould, she must have exploded like a Roman candle in the dust of Gould’s archives for Lepore.

The daughter of former slaves, Savage had come to New York in 1921.  A bit of a fabulist herself, she claimed to be 19, not her actual 29.  She introduced her 14-year-old daughter as her younger sister. And she hid two marriages, one of which Lepore leaves undissolved when Savage married a third time.  In 1923, when Gould and Savage met, she was already a central figure in the Harlem Renaissance.  During the next several years, she became director of an Afro-American arts center and founded a progressive club devoted to issues of race and politics.

Gould proposed marriage to Savage.  Rebuffed, he claimed they were having an affair.  He claimed they were engaged. He besieged her with obscene, racially offensive phone calls and letters.  His conduct was so appalling and so frightening that in 1934 Brand, whom Gould had also turned on, told him to leave them both alone, or he would have Gould arrested.  Gould never contacted Brand again, but, Lepore writes, “for Savage, the end had not come.”  She “never… got free of Joe Gould’s hold.”  “He never left her alone.”

But how exactly? Three sentences after writing that second “never,” Lepore has Gould unsuccessfully trying to locate Savage.  It does not appear he succeeded. Lepore describes no further harassment of Savage by Gould that I could find.  And if Gould did harass her after 1934, Savage’s career did not suffer.  Though she was creating less art, through her teaching she became, “the most influential artist in Harlem.” In 1937, she received “her most important commission.” In 1938, she made the cover of “Life.”In 1939, she had her first one-woman show.

It is true that by early 1943, Savage had left Harlem for a small farm, without electricity or plumbing, in Saugerties, New York; and Lepore admits thinking that Gould had driven Savage to this wilderness. But, she learned, Savage had been an F.B.I. informant, causing at least one friend, the poet Gwendolyn Bennett, to lose a position she desired because of her political sympathies.[12] Savage told a minister in Saugerties that she had left the city because she had been threatened by Communists.  Maybe.  Or maybe it had become too uncomfortable for her  to have remained among those she had betrayed.  In any event, Gould was not the reason.

In my reading, Lepore is more tolerant of Savage’s deceptions than of Gould’s (and, as we shall see, Mitchell’s).  And Lepore seems less critical of Savage for the suffering she inflicted through her informing than she is of Gould for the suffering he inflicted on Savage (or, as we shall again see, Mitchell for suffering Lepore attributes to him.)  There is nothing as compelling as, say, a bloody axe in the hands of anyone involved to compel the tilt of  these scales.  The author’s thumb is decisive.


Then there are Gould’s final years.

Mitchell wrote that he did not probe the psychiatrist who told him about Gould’s last hospitalizations, as if to spare himself and his readers, as well as Gould’s memory, what he might discover. Lepore dove into them.  But having been denied access to Gould’s medical records, she is left to thought processes which range from reason to speculation. “Very likely…,” she says.  “I’m pretty sure…” “Chances are…” “I suspect…” “(I)t might possibly have…” “It’s not impossible that…” Her conclusions subject Gould to a torture chamber of personality destroying treatments, from electro-shock to neuro-surgery.

Lepore’s portrait of Gould is more complete, more complex, and, I’d venture to say, more accurate than Mitchell’s. But I am unconvinced that Gould’s degradation, no matter how probable, need to have been detailed.  The narrative does not compel it.  The likelihood of “The Oral History”’s existence clarifies not one whit. What readers gain in information about medical barbarism may not offset what they lose in spirit.  It seems important to Lepore that they be noted.  But why is less clear.


“What is biography?” Lepore asks at one point.  “A life in time,” she answers.

That is part of it. But biography is also the time and life of the biographer. Who someone is, when this someone is known only through the writing of another, depends, in part, on who that writer is, which depends, also in part, on that writer’s time.  Gould was one person when depicted by Mitchell and another when depicted by Lepore.  In 50 years, if depicted again, he is likely to be someone else.

By the time Lepore wrote about Joe Gould, it was well known that Mitchell made up scenes and quotes, that some of his characters were composites and others entirely fictional.  Lepore does not accuse Mitchell of directly falsifying anything he wrote in “Seagull.”[13] While she is critical of his omissions, which she believed inappropriately glorified Gould, she admires the piece as  “a defense of invention. Mitchell took something that wasn’t beautiful – the sorry fate of a broken man – and made it beautiful, a fable about art.” In “Secret,” the major fabrication Lepore finds is Mitchell’s claim that he had extensively researched “Seagull,” when he actually did this research after it had appeared.  But her overall assessment of Mitchell is damning.  “Passing off fiction as fact,” she writes, “is an act of deception.  And with deception, someone usually gets hurt.”

Who, I wondered. Not Gould.  His panhandling benefited. He won a benefactor.  Mitchell?  He became one of the most highly regarded magazine journalists of his era. True, he  produced no significant work after “Secret”; but this seems, according to most thinkers, in terms of causal connectivity, less a matter of bingo-bango punishment-for-his-sins than post hoc ergo propter hoc thinking.

Truth and fiction, in the writing of prose, lie along a continuum.  Some declarations are simple to place; others not so.  When a writer plants an assertion prefaced by “It is not impossible…,” what is that? When one writer omits a figure like Savage from a narrative while another writer gives her a featured role, is one doing anything different than the other?  And, hey, for that matter, how many contributors to oral histories, with their content compromised by the ravages of time and the compulsions of inner wishes would satisfy Diogenes?  As the novelist Wright Morris once remarked, “Anytime you are relying on human memory, you are writing fiction.”[14]

A few months before Lepore’s article appeared, Janet Malcolm, reviewing in NYRB a recent biography of Mitchell, quoted him in his own defense.  “My desire,” he said, “is to get the reader, well, first of all to read it…. Here and there, as I think a fiction writer would, I put things in that I know… that are going to keep the reader going.  (So) I can lure him or her into the story I want to tell.”

The story I want to tell motivates all writers.  As Lepore noted about “The Oral History,” for Mitchell “It made a better story in 1942 if it existed; it made a better story in 1964 if it didn’t.”  (My emphases.) So in 2016 it made a better story for Lepore if, once again, it did.  Otherwise, she had a reprehensible psychotic running a scam, not a sympathetic one.  So she doesn’t look too hard at the numbers her own investigation produced (or disown the contrary-to-the-evidence promotional statements of her publishers). So she elevated the role of one person Gould treated badly out of many because of the constituencies this victim, vice-presidentially delivered. So she traded Gould’s privacy for a politically appealing slam against mid-century psychiatric hospitals.


In her book review, Malcolm wrote few “writers of nonfiction… have gone as far as Mitchell in bending actuality to our artistic will.  This is not because we are more virtuous than Mitchell.  It is because we are less gifted…” In the epilogue to “Teeth,” Lepore describes an imaginary room where her study of Gould and Mitchell has led her.  Allen Ginsberg lurks beside the door.  Ezra Pound shrieks hatred of Jews from the radio. A nineteenth century phonograph utters a seagull’s screech. Handwriting covers the floor and walls. The detritus includes a tattered raincoat, an empty liquor bottle, Joe Mitchell’s typewriter, stacks of magazines, and one notebook – only one.  There is not a single “fact” in these imagined four pages, but they amount to a sizable one.

These pages are stunning.  They are elevating.  They sweep you through the imps and goblins to which investigative creativity exposes one.  When you write a biography, you choose your subject because, out of all those available, this one calls to you the most seductively.  It whispers of thrills and dangerous pleasures, not always beforehand clearly seen or sensed.  “Joe Gould lived a life of dreams and brilliance and psychoses,” the (semi-fictional) psychoanalytic social critic Ruth Delhi, PhD., said to me.  “He made people want to write about him, but when they did, he took them places they did not want to go.  ‘What am I doing with this guy,’ he left them asking.  There but for the grace of God…, they may have thought.”

Some writers of biography, like Mitchell in “Secret,” investigate how their subject ensnared them. More, like Lepore, resist this unveiling.  But either way, the writer tries to extricate him/herself from their subject’s grip through exercising a creative vision which reshapes him, whether through a pat on the back or by his neural pathways.  This reshaping restores the writer’s dominance and is offered to readers for a validating response.  Both bind the bricks of fact with a mortar whose fictive proportion is not always easily measured.  But always the aim is that the story be told and the writer freed.



1 A more complete list would have included John Dos Passos, Marianne Moore, Ezra Pound, and William Carlos Williams.

2 The three articles were published, “reworked,” in book form as “Joe Gould’s Secret,” by Viking in 1965. I have relied upon that volume in this writing. (They also appear in “Up in the Old Hotel,” a collection of Mitchell’s work, published by Vintage in 1993, three years before his death.)

3 These endnotes are not as funny as Will Cuppy’s or as mind-expansive as David Foster Wallace’s.  Most are of no use to anyone not intending to follow Lepore into library stacks and check up on her.  But an occasional nugget makes them worth mining.

4 Lepore notes she “tried to avoid diagnosing Gould” because it was “impossible” “on evidentiary grounds” and “unsupportable” as a “historical method.” Still, she made it clear she would not be supporting him for a tenured position.

5 Savage appears on over 50 pages in the book.  Aside from Gould and Mitchell, the next most frequently mentioned person, E.E. Cummings, appears on 37.

6 Mitchell was willing to credit Gould with, maybe, “a couple dozen.”

7 A few months later, a Dallas newspaper wrote that it filled 500 copy books. Let’s assume this was a typographical error and disregard it.  But in 1934, Gould told the Herald-Tribune his history had 7,300,000 words, which, taking the high end of Gregory’s estimate of 100 notebooks in 1931, meant Gould had produced as many words in the following three years as he had in the previous 15.  Then, in 1937, Gould told that same newspaper his history held 8,800,000 words, meaning he’d added another 1,500,000 in three years (500,000/year), but that in the five before he met Mitchell he’d only managed 400,000 (80,000/year).

8 This total is what Lepore says she found in Mitchell’s notes of his interview of Gregory.  In “Seagull,” though, Mitchell quotes Gregory as saying he’d only read “twenty-odd” books.

9 This theory, particularly the “get-away-with-it” part, is not endnoted and seems Lepore’s creation.

10 Hmmm.  Graphomania: A meaningless succession of words… even nonsense…

11 From quotes encountered in “Seagull”’s text, he seems to have spoken to three others about Gould.

12 The information about Bennett is in an endnote.  It is not in Lepore’s New Yorker article.

13 I would note that Mitchell’s description of the “Oral History” as “a great hodgepodge and kitchen midden of hearsay, a repository of jabber, an omnium-gatherum of bushwa, gab, palaver, hogwash, flapdoodle, and malarkey,” seems, if wonderful, an over-reach, since he hadn’t seen a word of it.

14 At least I think he said that.  I am relying on my own memory here.