As a cub reporter in New York in the 1930s, Joseph Mitchell once listened in on the questioning of a prostitute who had allegedly allowed her body to serve as the altar for a Black Mass. Asked why she had become a prostitute, the woman replied, “I just wanted to be accommodating.” Mitchell adored this remark. “The best talk is artless,” he concluded. “Now and then … someone says something so unexpected it is magnificent.”

Mitchell made a legendary career out of listening to, recording, and shaping that artless talk. Of the first generation of New Yorker writers—A. J. Liebling, Janet Flanner, E. B. White—he remains the most influential. His immersive techniques and famously lucid style have given rise to countless imitators. In a foreword to a reissue of McSorley’s Wonderful Saloon, a collection that includes many of Mitchell’s famous profiles of figures on the margins of New York society, Calvin Trillin marveled that Mitchell “was able to get the marks of writing off his pieces,” making it seem as if the words had appeared magically on the page. His collected magazine work, published in the omnibus 1992 volume Up in the Old Hotel, remains in print and is required reading for any aspiring journalist.

Yet that same aspiring journalist, judging Mitchell by the stringent standards that now prevail—or are supposed to—in nonfiction magazine writing, might wonder just how much in his stories would survive a fact-checker’s scrutiny. How could the multi-paragraph quotes that characterize Mitchell’s work, recorded in shorthand notes on a single piece of copy paper that he folded to fit in his front pocket, possibly be accurate? What are we to make of his admitted use of a composite figure in at least one famous series of so-called profiles, about a man he called “Old Mr. Flood,” who turned out not to exist—at least not in the way Mitchell depicted him? Could there be a connection between Mitchell’s unconventional writing strategies and the long stretch of silence at the end of his rich career? Almost every day for 32 years—from 1964 to 1996, when he died—he put on his customary suit and hat, went to his office at The New Yorker, and sat down at his typewriter, but published not a single word.

In Man in Profile: Joseph Mitchell of The New Yorker, a thoughtful and sympathetic new biography, Thomas Kunkel—who delved into the magazine’s history for his earlier biography of its founding editor, Harold Ross—pores over Mitchell’s archive in an attempt to figure out how the writer created those memorable pieces, as well as what he might have been working on in those final decades. What Kunkel finds may distress Mitchell’s fans. Not only did Mitchell’s use of composite characters go beyond what he acknowledged during his lifetime, but some of the key scenes in his best-known profiles appear nowhere in his notes and may also have been invented. By current standards, which privilege fact-checkability over storytelling, and data over character, Mitchell—“a national treasure,” in Kunkel’s words—would likely be denounced as a fabricator. But if today’s literary journalism no longer has room for a category-defying writer like Mitchell, the problem isn’t him—it’s us.

Though mitchell once insisted that “most of the influences responsible for one’s cast of mind are too remote and mysterious to be known,” Kunkel convincingly traces certain character traits to Mitchell’s upbringing in rural North Carolina—the subject of a memoir he attempted but never finished. (It turns out, incidentally, that Mitchell’s problem was an inability to finish pieces he started.) Mitchell’s father, a prosperous farmer and merchant who dealt in cotton, tobacco, and other crops, hoped his eldest son would join the family business. But Kunkel portrays a curious boy who loved to spend hours rambling in the woods or soaking up the carnival atmosphere of the auction house and the traveling circus. Accompanying his father on a trip to New York City, the 10-year-old Mitchell supposedly gazed up at the skyscrapers and announced, “This is for me.” Later he would write of the Bowery as if it were a great river, noting the landmarks at its beginning and end.

Mitchell moved to New York in 1929, at the age of 21, and over the next decade worked for now-defunct newspapers like The New York World, the New York Herald Tribune, and the New York World-Telegram. He sometimes filed as many as three articles a day, writing of the city’s minor celebrities, mobsters, and laborers with sensitivity and nuance. At the same time, he tried to write fiction: he planned a novel that would take place on a single day in New York, inspired by Joyce’s Ulysses, and he managed to publish a few short stories. Aware that he had a rare talent on his staff, his editor at the Herald Tribune commented, “The elusive quality of the city is the despair alike of poets and short story writers and historians.” Mitchell’s early pieces revealed a writer who was already all of the above.

Those articles brought Mitchell to the attention of Ross, who hired him as a staff writer at The New Yorker in 1938. With more time to spend on each piece, Mitchell relaxed into the form that he pioneered: candid, sympathetic depictions of the barkeeps, fishmongers, and bohemians who haunted the city’s seedier neighborhoods. Mitchell’s interest in these characters wasn’t voyeuristic or exploitative; like a novelist, he portrayed his subjects in three dimensions. A profile of Mazie Gordon, the ticket-taker at the Bowery’s Venice Theatre, noted her unconventional looks—hair “the color of sulphur” and a “smudge of rouge the size of a silver dollar on each cheek”—as well as her kindness toward drunks and bums; her heart was legendary in the “all-night restaurants which specialize in pig snouts and cabbage at a dime a platter.” (Ever alert for artless talk, he quotes Mazie telling those who tried to take advantage of her generosity to “go steal a watch.”)

Not everyone appreciated his exploration of “lowlifes,” as some perceived his subjects. The Time-magazine reviewer of McSorley’s Wonderful Saloon sniffed that Mitchell studied “queer human weeds” with “the prying patience of a botanist”—an attitude Mitchell found “patronizing and repulsive.” He didn’t invade his subjects’ privacy; he seems to have genuinely befriended almost all of them. When he appears in his pieces, it’s not as a reporter but as an observer or an acquaintance—in Kunkel’s words, an “amiable travel companion.”

This quality of friendship distinguishes Mitchell’s greatest pieces, which combine a journalist’s real-time observations with a fiction writer’s creative empathy. He liked to begin with a declarative sentence as strong as a shot of whiskey. “Joe Gould is a blithe and emaciated little man who has been a notable in the cafeterias, diners, barrooms, and dumps of Greenwich Village for a quarter of a century” is the start of his famous profile of Gould, a bum who claimed to be working on a 9-million-word chronicle of overheard conversations called An Oral History of Our Time. Mitchell delighted in obscure facts: Who else would know that the “gelatinous air bladders” along a cod’s backbone are called sounds? In “Mr. Hunter’s Grave,” a portrait of an African American community on Staten Island, Mitchell listed the wildflowers and weeds he found in a graveyard: “catchfly, Jerusalem oak, bedstraw, goldenrod, cocklebur, butter-and-eggs, dandelion, bouncing Bet, mullein, partridge pea, beggar’s-lice, sandspur,” and many more. Some were unfamiliar to him, he took care to note, adding that he’d broken off samples to examine later under a magnifying glass.

It is this attention to detail, even fussiness, that is likely to trigger a sense of betrayal in readers dismayed by the liberties Mitchell took with facts. Why list all those flowers—and acknowledge the limitations of his expertise—and yet represent the conversation with Mr. Hunter as a single long encounter when it actually took place over many months? Why report that Mr. Flood eats the cod’s sounds for breakfast when Mr. Flood was actually a composite of various men Mitchell came to know at the Fulton Fish Market, one of his favorite hangouts? And how did Ross, the editor who essentially invented fact-checking and was notorious among writers for his fanatically literal-minded queries, let Mitchell get away with this?

Kunkel has a surprise: it was Ross who first suggested the composite approach and quietly urged Mitchell on. (In its early days, The New Yorker didn’t always distinguish between fiction and journalism for its readers.) Mitchell had been hanging out at the Fulton Fish Market for a decade, but his friends there didn’t want to be featured. Ross was ready to let this trusted star writer—whose rapport with sources was uncanny, and who had already based a much-praised profile of a “gypsy king” on a composite—focus his disparate material into a single figure. Ross even pressed for a fourth installment on Mr. Flood, but Mitchell was getting anxious: a publisher was clamoring for an autobiography from the “mayor” of the fish market, and fans wanted to visit with him. It’s telling that three years later, when the trio of pieces appeared as a book in 1948 and Mitchell revealed that Mr. Flood didn’t exist, no outcry ensued.

Mitchell’s readers and editor, and critics such as Malcolm Cowley, had come to expect his rare brand of literary journalism: he was portraying “some of the most memorable characters in the nonfiction canon,” as Kunkel puts it, revealing them in their distinctive habitats and full humanity. Mitchell liked to say that he aimed for the essence of his subjects, rather than for surface accuracy, and that he tried to avoid being “over-precise” or “over-documentary.” What this meant to him seems to have evolved over the course of his career. In Mitchell’s prolific early years at The New Yorker, Kunkel makes clear, his flexibility generally extended only to reshaping the contours of an encounter in the interests of narrative flow, or changing certain details about a subject who didn’t wish to be identified. Mitchell considered making up quotes to be out of bounds, but said he saw “nothing wrong in any way about” moving them around. Kunkel notes that reportorial transparency had yet to be enshrined as a standard, and that Ross was lenient about “sweetening” speeches.

Anticipating postmodernism, Mitchell believed that “all biographies and autobiographies are fiction,” because of the writer’s role in constructing the narrative. But for him, that authorial craftsmanship—which we now recognize as a part of any journalist’s work—wasn’t just a way of evoking scenes, controlling the pace, or creating suspense. “When I’m working on a story,” Mitchell once said, “before I’m through I have a huge collection of notes … And then I realize: all these are facts, and the truth lies within these collections of interviews.”

Any writer—even a Joe Gould—can write down facts and quotes. Mitchell chose subjects who were at once outliers and figures whose lives opened up ways to explore broad themes close to his own concerns: experiences of exile, of “otherness,” of nostalgia for disappearing worlds. He was searching for truth, unironically, in much the way a fiction writer does: by drawing on immediate experience to illuminate the human condition. Mitchell was faithful to the grist he gathered about actual individuals, whose perspectives he probed: his stories were, he said, “solidly based on facts.” At the same time, he considered his subjects to be characters whose presentation on the page reflected his own unique way of seeing the world.

How much to trust his sources, and his own intuitions, could be a hard call, as he discovered at what seems to have been a crucial turning point in his career. Mitchell was deeply unsettled when one of his characters got away from him, and his subsequent struggle to navigate the boundaries of fact and fiction is instructive. After his piece on Joe Gould was published, in December 1942, Mitchell learned that Gould was a fraud. There was no Oral History of Our Time; Gould’s hundreds of composition books were filled with the same few essays written over and over.

Despite his dismay, Mitchell kept Gould’s secret—out of respect, he said, though Kunkel observes that along with feeling “betrayed, embarrassed, used,” Mitchell was also more than “a little panicky” at the prospect of the magazine having to take it all back. It was around this time that Mitchell was working on the Mr. Flood series, and over the next several years he seems to have relied more extensively on fictional techniques than he had before. Perhaps not coincidentally, his pace began to slow. Between 1938 and 1942, he published close to 30 pieces in The New Yorker; from 1943 on, barely half that number appeared, while expectations of Mitchell’s one-of-a-kind creations rose. Depression also seems to have played a significant role during Mitchell’s later decades: at one point, Kunkel quotes him worrying about memory loss from “all the Seconal and Valium … I have been taking over the years.”

It’s too simple to conclude that Mitchell’s method of mixing fact and fiction—the two genres that he’d hoped to pursue, separately, from the start of his career—led him fatally astray. Yet while rightly resisting that verdict, Kunkel is also alert to a subtext of uneasiness in a writer whose pride in his work was perhaps haunted by a failure to be more forthright about Gould and about his own increasingly inventive approaches. Mitchell pulled back, Kunkel writes, after “bending the boundaries of nonfiction to the point where these particular stories were more accurately seen as fact-based fiction.” And in 1964, in what turned out to be his last appearance in The New Yorker, Mitchell finally revealed that Gould had duped him.

He didn’t apologize in “Joe Gould’s Secret.” In vintage Mitchell form, he instead identified with his betrayer. As he later said in an interview, he held Gould up as an emblem of “so many people who bit off more than they could chew”: in Gould’s unrealized opus, Mitchell saw a parallel to the big New York City novel that remained trapped in his own head. Yet thanks to the constraints of nonfiction, he got many parts of that book out in other ways during a time of flux in journalistic standards. In his depictions of the people and places of an already disappearing New York, Mitchell sought to elevate their daily routines into something universal. He saw the city itself, in Kunkel’s words, as a “big, messy, living novel.” David Remnick, The New Yorker’s current editor, has called Mitchell’s portraits “as pointed, various, and haunted as the stories in Dubliners.”

Naming a contemporary journalist who sounds much like Mitchell—or a publication that would grant such a person the time and space to roam so freely—is hard. His true heirs might be writers who forthrightly defy clear-cut labels, such as W. G. Sebald and Teju Cole. Both of them spin a profusion of facts into something that resembles both fiction and journalism, and whose narrators speak in a register similar to the tone of melancholic nostalgia Mitchell strikes in his later pieces (“When things get too much for me, I put a wild-flower book and a couple of sandwiches in my pockets and go down to the South Shore of Staten Island”). Given how little room today’s rapid-fire media offer for a style this ruminative, Mitchell’s legacy may be as important for an aspiring novelist as for an aspiring journalist. To say so is no insult to his memory: such a verdict, one suspects, would have made him proud.