Reporter of Chaos and Information
A former Times science reporter who turned a meteorologist’s rounding error into a bestseller, co-founded an early New York ISP and fought for authors in the Google Books war.
Science Writer & Historian
An English-and-linguistics graduate with no laboratory training, James Gleick learned science as a New York Times reporter and then wrote a bestseller that opens on a meteorologist typing three decimal places where the computer had kept six. He has since run an internet company, helped negotiate the Authors Guild’s failed settlement with Google, led the Guild, and turned Claude Shannon’s mathematics into a history of everything that can be sent as a message. His method is reporting: find the person, the machine and the moment, and let the idea arrive through them. That method has a price, and his critics have named it.
James Gleick was born in New York City on 1 August 1954, the eldest of three children. His father, Donen Gleick, was a lawyer who dreamed of being a forest ranger; his mother, Beth, wrote children’s books and edited newsletters. His brother Peter (born 1956) became a water scientist and a MacArthur fellow, and his younger sister, Elizabeth, an editor. Everything here about the household comes from a 2009 profile of Peter, whose childhood it describes as spent in Manhattan; James himself has said little in print about growing up.
His mother’s best-known book, Time Is When (1960), was illustrated by Harvey Weiss, and The Marginalian describes it as her answer to the question of what time is, written when James was six. Readers can draw the line from there to Faster and Time Travel; that is an inference, not something Gleick claims. What he does say is that he was good at mathematics at a New York high school, and that at eleven, at summer camp, he met Dan Frank, the boy who would grow up to be his editor at Viking and then Pantheon.
At Harvard he majored in English and linguistics and, in his own words, “spent a lot of energy trying to wriggle out of the few science requirements they had.” What he gave his attention to was The Harvard Crimson: in 2002 he told Publishers Weekly, “I wasn’t thinking about science then, I just liked journalism.” His Crimson bylines run from 1974 to 1976. The one that points furthest ahead is a review of Leonard Bernstein’s Norton Lectures on music and Chomskyan linguistics (8 January 1975), in which the twenty-year-old linguistics student called the conductor “careless with words and with ideas”, mocked his “musicolinguistics” and decided that much of what he said about linguistics was “not so much wrong as it is just empty.” It is an early case of Gleick checking a famous man’s borrowed science against the field. Elsewhere he wrote a mock-epic on Bally’s Wizard pinball machine at the Underdog on Bow Street, which he compared to The Rite of Spring (30 September 1975), reported the opening of Harvard’s underground Pusey Library (February 1976), and on 7 April 1976 co-wrote with Thomas Blanton the Crimson’s report on the New York presidential primary.
After graduating in 1976 he moved to Minneapolis (one profile says after freelancing in Boston) and helped start Metropolis, a weekly that ran from 19 October 1976 to 6 September 1977, according to Minnesota Historical Society microfilm records quoted by the site Twin Cities Music Highlights. Almost everything known about the paper comes from one participant, Mark Simonson, its production manager from January 1977, who also names Gleick as managing editor and Scott Kaufer, later a television writer, as senior editor. In Simonson’s account the staff were mostly recent Crimson graduates modelling themselves on Boston’s Real Paper and Phoenix, and they picked Minneapolis because it seemed to have no alternative weekly. The Entertainer, an entertainment paper, did exist; threatened by the newcomer, it added news and became the Twin Cities Reader. “Competition for advertising between the two papers was a blood bath,” Simonson writes, and the Reader, “with all its homegrown business ties, completely shut out this outsider enterprise.” Metropolis’s business side was “much weaker than the editorial side”, it was mainly funded from its publisher Walter “Trip” Rothschild III’s trust fund, and by the fall of 1977 the money ran out and Rothschild shut it down. Gleick’s own one-line version is that he “helped found” it. This is a recollection posted on a fan-built history site; no contemporary reporting on the paper’s finances has turned up.
Gleick went back to New York. He was hired at the Times in 1979, and no available source says what he did between the paper’s closing and then. The order matters for what follows: Gleick came to physics and mathematics as a reader of language and a newspaperman who had already run an editorial desk and watched it fail, with no scientific credentials to lean on. Whatever authority his books would have had to come from reporting.
The Times took him on in 1979 as a copy editor. He stayed about ten years, first as an editor on the metropolitan desk and then as a science reporter. In 2002 he explained to Publishers Weekly how that led to science: “as a kid on the metro desk” he could not pitch the magazine a big piece on Middle Eastern politics, because “they had plenty of people who knew about that.” But he could say, “I want to write about this new thing called chaos theory,” and “they would say, ‘Okay.’”
He broke into reporting for the Sunday magazine with a profile of Douglas Hofstadter, author of Gödel, Escher, Bach, and went on to profile two mathematicians who were stretching their own disciplines, Mitchell Feigenbaum, who studied how simple systems can produce wildly unpredictable results, and Benoit Mandelbrot, the discoverer of fractals. The Times archive was out of reach, so these pieces are left undated. By his account Chaos “grew from two of those articles.” The reporting habits he credits to the writers he admired, J. Anthony Lukas, Tom Wolfe and Gay Talese, were “Telling the truth, going to the place and hanging out.”
Several publishers approached him; one was Dan Frank of Viking, whom he had known since summer camp at eleven. They talked over lunch and made a deal. Gleick took four months of leave from the Times to do the interviews and wrote the manuscript mornings, evenings and weekends, because, he said, “I had a finite leave of absence and a small advance.” Viking printed 20,000 copies, and by 1999 Publishers Weekly counted half a million sold in hardcover and paperback. The reporting also left him with a book he had not planned to read. Gleick has said he first heard of information theory while working on Chaos, bought Claude Shannon’s The Mathematical Theory of Communication, and has been thinking about it ever since; “I never heard anything about it when I was at Harvard,” he told the Crimson in 2011. The Information, twenty-four years later, grew from that purchase.
After Chaos he left the Times staff to report Genius, Publishers Weekly says, while continuing to write columns and articles. The same magazine’s 2002 profile counts only two years as the paper’s official science reporter, so the science beat was the shorter part of his years there.
Chaos: Making a New Science (Viking, 1987) opens at the Massachusetts Institute of Technology with Edward Lorenz and his Royal McBee, a loud vacuum-tube computer that broke down every week or so. Lorenz had written a toy weather of twelve equations; the machine printed a row of numbers for each simulated day. One day in the winter of 1961 he restarted a run midway, typing the starting values straight from the old printout, and went down the hall for a coffee. When he came back an hour later, the new weather had drifted so far from the old that within a few simulated months all resemblance had gone.
Gleick then gives the explanation. The computer’s memory held six decimal places, .506127; the printout, to save space, showed three, .506. Lorenz had assumed a difference of one part in a thousand could not matter. In his deterministic equations it did: the small error grew until it decided the whole outcome. Gleick reproduces Lorenz’s 1961 printouts as a figure and describes him laying one line of output on a transparency over the other to watch them separate hump by hump. That is sensitive dependence on initial conditions, the Butterfly Effect, and it was why Lorenz concluded on the first day that long-range weather forecasting was doomed.
The chapter does not stop at bad news. Lorenz stripped a convection model down to three nonlinear equations whose plotted path never settles and never repeats, yet never leaves its bounds: the double spiral later called the Lorenz attractor, published in 1963 as “Deterministic Nonperiodic Flow” on page 130 of volume 20 of the Journal of the Atmospheric Sciences, where, Gleick notes, no one thought to look for chaos. He quotes the MIT mathematician Willem Malkus dismissing it at the time (“Ed, we know—we know very well—that fluid convection doesn’t do that at all”) and then, a generation later, admitting “we completely missed the point.” The rest of the book works the same way: Mitchell Feigenbaum at Los Alamos finding that different functions all produced the same number on their way to disorder (about 4.669, now Feigenbaum’s constant), Benoit Mandelbrot’s fractals, experimenters in fluid labs. It was a finalist for both the Pulitzer Prize and the National Book Award, and became a national bestseller.
The scene-by-scene method is also the book’s limit. In Gleick’s opening chapter Lorenz is alone at the machine. Lorenz’s 1963 paper thanks “Miss Ellen Fetter for handling the many numerical computations”, and Joshua Sokol’s 2019 Quanta report shows that Margaret Hamilton ran Lorenz’s LGP-30 work from 1959 to 1961 before Fetter took over and helped plot the attractor; neither woman appears in Gleick’s telling. The historians David Aubin and Amy Dahan Dalmedico (Historia Mathematica, 2002) quote Gleick’s claim that twentieth-century science “will be remembered for just three things: relativity, quantum mechanics, and chaos” as an example of overblown assessments, while noting that he was faithfully reporting the scientists’ own enthusiasm. Mathematicians replied that Poincaré’s legacy had never been forgotten, and the book provoked a 1989 exchange in The Mathematical Intelligencer involving John Franks, Gleick, Morris Hirsch, Mandelbrot and Steven Krantz. The same historians call Feigenbaum the book’s “most conspicuous Romantic hero”; the physicist David Ruelle, a central figure in the field, credited part of chaos’s popular success to “Gleick’s bestseller.”
Gleick extended the book twice: Nature’s Chaos (1990) set his text beside the photographer Eliot Porter’s pictures of turbulence and pattern, and Chaos: The Software was made with developers at Autodesk.
On 17 February 1988 the Times ran Gleick’s obituary of Richard Feynman, who had died of abdominal cancer in Los Angeles two days earlier, calling him “arguably the most brilliant, iconoclastic and influential of the postwar generation of theoretical physicists.” After a year as McGraw Distinguished Lecturer at Princeton (1989–90), Gleick turned the obituary into a full life. Genius: The Life and Science of Richard Feynman (Pantheon, 1992) had to hold the atomic bomb, the Nobel-winning quantum electrodynamics and the Challenger inquiry together with the showman, the safecracker, the bongo player and the charmer, and to explain why colleagues thought of Feynman less as a genius than as a magician. Like Chaos, it was a finalist for both the Pulitzer Prize and the National Book Award.
Then the reporter became, briefly, a businessman. In December 1993 Gleick and the programmer Uday Ivatury started the Pipeline, one of New York’s first internet service providers. At a time when getting online usually meant Unix command lines, the Pipeline gave subscribers point-and-click software for email, chat, Usenet, the web and FTP, and licensed it to other providers. Gleick told Publishers Weekly in 1999 that the itch was personal: the scientists he interviewed all had e-mail and he wanted it, but the Unix commands got in the way of his passion for contract bridge on the computer, so he asked a programmer he knew to build something friendlier. Gleick was chairman and chief executive until 1995. PSINet bought the company in February 1995 and took it national; in July 1996 PSINet sold its consumer business to MindSpring and the Pipeline name disappeared. Publishers Weekly reported the 1995 sale at about $10 million; Gleick would not give his share but admitted to “some millions”, which with his book income paid for a house above the Hudson.
In December 1997 Gleick, an experienced pilot, was flying his canard-winged Long EZ with his eight-year-old adopted son, Harry, when the plane crashed short of the runway while he was bringing it in to land. Harry was killed. Gleick was pulled from the wreckage in critical condition, lost one leg and nearly lost the other, and spent months in hospital and many more in rehabilitation. He declines to discuss it beyond saying that it “shattered my life”. At his editor’s urging he began writing again from his hospital bed, resuming the Fast Forward column on technology and society that he had been writing for the Times Magazine; the Publishers Weekly writer who visited him in 1999 called the work therapeutic.
He covered the network culture he had helped sell as a columnist: Fast Forward in The New York Times Magazine ran from 1995 to 1999 and fed Faster (1999), on acceleration and the shrinking of waiting time, and What Just Happened (2002), a chronicle of the years when computers and networks entered ordinary life. He was the first editor of The Best American Science Writing (2000). Isaac Newton (2003), a short life that keeps the physics, the alchemy and the theology in one man, was another Pulitzer finalist.
In September 2005 the Authors Guild sued Google over its scanning of millions of library books. Gleick helped negotiate the settlement announced in October 2008, and on PBS NewsHour in March 2010 defended it as putting “an end to the idea that you can just scan books freely without the permission of authors.” The authors’ side lost. Judge Denny Chin rejected the amended settlement on 22 March 2011, then ruled Google’s scanning fair use on 14 November 2013; the Second Circuit affirmed on 16 October 2015, and the Supreme Court declined to hear the case in April 2016. Gleick served as the Guild’s president from 2017 to 2019.
The Information: A History, a Theory, a Flood (Pantheon, 2011) begins with the talking drums of the Kele-speaking peoples of the Congo, as recorded by the English missionary John Carrington in The Talking Drums of Africa (1949). The drums reproduce only the tones of speech, so a single drummed word could mean many things; drummers resolved the ambiguity by stretching each word into a long, fixed phrase, about eight drummed words for every spoken one, as Freeman Dyson summarises it. Dyson’s review quotes the call Carrington’s wife drummed to bring him home for lunch: “White man spirit in forest come come to house of shingles high up above of white man spirit in forest. Woman with yam awaits. Come come.” That redundancy is Claude Shannon’s theory in miniature. Shannon’s 1948 mathematics measured information while setting meaning aside, and that separation let engineers build the networks that now flood us; Gleick’s epilogue, “The Return of Meaning”, asks what the bargain cost. The book won the Royal Society Winton Prize in November 2012.
Time Travel: A History (2016) argues that the idea is barely older than H. G. Wells’s The Time Machine (1895), and that physics and fiction have traded it back and forth ever since. Gleick states flatly that time travel is “not real. It’s an impossibility.” Reviewing it in Physics World, Brian Clegg objected that dismissing relativistic time dilation as not really time travel “smacks of ignorance”, that crediting Wells with a new mode of thought was excessive, and that the line “the wave function is timeless” sits badly with the time-dependent Schrödinger equation.
He now writes long essays for The New York Review of Books. “The Parrot in the Machine” (2025), a review of Emily Bender and Alex Hanna’s The AI Con and James Boyle’s The Line, opens with Shannon asking his wife Betty, a mathematician, to guess a detective-story passage letter by letter (she was right 69 percent of the time) and argues that chatbots extend the same method “with the advantage of prodigious computing power and terabytes of training material”, plagiarism being “one of their chief skills.” “How the Web Was Lost” followed in December 2025. The Telephone: A New History is due from Farrar, Straus and Giroux on 10 November 2026, with a British edition in January 2027; its jacket copy sets Gleick against Octavio Paz’s line that solitude is the profoundest fact of the human condition and gives his reply: “Connection is the profoundest fact of the human condition.” He lives in New York and London.
Claude ShannonThe mathematical theory at the centre of Gleick’s history of information.
H. G. WellsA literary precursor to the cultural history of time travel.
