When information needs to be communicated, Edward Tufte demands both truth and beauty.
Photo: Mark Ostow
By Fran Smith
In 1613, Galileo published Istoria e dimostrazioni intorno alle macchie solari, his remarkable observations of the sun. On a fall day, 393 years later, Edward Tufte stands in front of a packed hotel ballroom, holding up a first edition of that book.
The room could be in New York, San Francisco, Cleveland or any of the dozens of other cities where Tufte, ’63, MS ’64, teaches his daylong course Presenting Data and Information. Today he’s at the New Haven Omni, just blocks from the Yale campus where he taught for 22 years. Nearly 400 people have come, at $360 a head (half-price for students, and a set of his books is included), to hear the man who has been called the Leonardo da Vinci of data, the Strunk and White of graphic design, the George Orwell of the digital age.
If the course title sounds like a snooze, you might be surprised that an academic who lectures about data could attain such godhead status. But in four books on design, from 1983’s The Visual Display of Quantitative Information to the recent Beautiful Evidence, Tufte has offered the country’s foremost critique of the way information is depicted in graphs, tables, illustrations and (lately and relentlessly) PowerPoint slides. Graphics, as Tufte makes clear, are not mere sideshows to spruce up text, entertain readers or keep art majors employed. Graphics shape, and too often distort, our understanding of everything.
Tufte has demonstrated how confusing medical charts can lead to mistakes in treatment and how corporate reports that highlight years of rising revenue without adjusting for inflation can mislead investors. He has shown how a lawyer used a simple spreadsheet to defend mobster John Gotti and how 19th-century physician John Snow used detailed maps of London to pinpoint the cause of a cholera outbreak. Tufte is credited with turning chart-making into a discipline with intellectual credibility and moral weight. His course attracts not only visual professionals but also scientists, engineers, journalists, doctors, attorneys and financial analysts—pretty much anyone who analyzes and presents data.
In his lectures and books, Tufte invokes a variety of thinkers who have been models of precision, withering analysis and clarity. But his hero, “the master,’’ is Galileo, the mathematician and astronomer who challenged fiercely held misconceptions about the world by the simple, unprecedented act of looking at the sky through a telescope and drawing what he saw.
Beautiful Evidence opens with the words of a Galileo friend and patron, who wrote that those drawings “delight by the wonder of the spectacle and the accuracy of expression.’’ Tufte returns to the images again and again: sunspots, Jupiter’s moons, meticulously annotated diagrams of planets and stars. He says Galileo’s first published observations of Saturn’s rings, with word-sized sketches inserted mid-sentence (see below), “may be the best piece of analytic design ever done.’’
A white-gloved young man carries the book up and down the long rows of conference tables so all can get a close-up glimpse at this historical treasure, where Galileo stated the heretical idea that the Earth moves. And sure enough, every visual attribute Tufte promotes is on those old, mesmerizing pages: the integration of drawings and words; the efficient, elegant design; the straightforward image, almost as elemental as a child’s, capturing the soul-stirring richness of the universe.
But what inspires Tufte is more than aesthetics. Galileo’s observations, recorded in nearly 12,000 pages, marked an intellectual revolution. No longer was knowledge the dictate of church authorities, kings or the acolytes of Aristotle. Theories could be tested—doctrine could be upended—by what the eye can see. As Tufte sees it, what makes evidence beautiful isn’t artistry. “It’s all about discovering and telling the truth,’’ he says.
After an encounter with Tufte’s ideas, people can never again look at a chart, a map, a scientific table or a PowerPoint presentation quite the same way. In his romps through statistics, art, history, science, policy and anything else that grabs his interest, Tufte tackles a fundamental problem: how to accurately render complex, interrelated information on a two-dimensional paper surface or computer screen—how to, as he puts it, “escape flatland.’’ Tufte explains how to do it well and demonstrates the many, many ways it’s done badly.
Bad graphics mangle the truth or lie outright, Tufte says, by a myriad of design flaws. Lousy graphics omit context, bury critical information, cherry-pick data to advance a cause and heap on “chartjunk’’—a Tufteism (and there are many) for the smiley faces, irrelevant numbers and other doodads that distract us from grasping evidence, thinking about it and drawing smart conclusions. This can have catastrophic consequences. Tufte asserts, for example, that poorly designed charts played a decisive role in both space shuttle disasters (see sidebar).
By his count, 1.4 million copies of his books are in print, and 160,000 people have taken his one-day course. He is cited copiously in scholarly articles, design textbooks and general-interest books with Dilbert-worthy titles like Why Business People Speak Like Idiots. He has consulted with the U.S. Centers for Disease Control and Prevention, NASA, the Federal Energy Administration, IBM, Bose, Sun Microsystems, TV networks, leading newspapers, law firms, brokerage houses and hospitals. Devotees use his name, like Google’s, as a verb; to “Tufteize” your presentations is to scrap content-lite graphics, such as pie charts and decorative dingbats, and to create visuals that brim with data and state precisely what you know—nothing more and nothing less.
“We adore him,’’ says Nicolas Bissantz, managing director of Bissantz & Company, a software firm in Nuremberg, Germany. Bissantz stumbled onto Tufte’s books (in English—they’ve not been published in translation), and got so excited he developed software that makes Tufteizing a chart almost as easy as, well, creating a PowerPoint show. The software uses a Tufte idea for compressing huge amounts of data—say, the fluctuations of the exchange rate over several years—into a word-sized graphic called a sparkline.
Tufte posted sparklines on the Ask E.T. Forum of his website, free for the taking. He puts many ideas there and, although he prefers that people write open-source code, he doesn’t stop anyone from turning an idea into a commercial product. Tufte also posts draft chapters, photos of his art, graphics he loves or hates, and questions that intrigue him: Are bad PowerPoint displays the fault of presenters or the technology itself? What makes for a brilliant performer-audience relationship? Is concert music always too loud? What are the grand truths about human behavior?
Tufte calls the forum “open office hours’’ and the adherents who weigh in “Kindly Contributors.’’ Here he seeks comment on problems he’s working on. Last spring, after an airline security official asked him to analyze whether better airport runway maps would reduce runway incursions, Tufte posed the problem on his website; the discussion was still going in November.
Tufte dispenses advice and criticism unsparingly. “Sometimes the contributors are disappointed,’’ Bissantz says. “They’ll post an idea and he has three or four points of criticism, and they’re devastated. They want to defend their concept, but it’s ridiculous—he just demolishes the idea. It’s worthwhile to just sit back and say, ‘Thank you, Master.’ ’’
At 65, Tufte is not only a guru and a verb but also a cottage industry. He publishes, distributes and markets his books through his firm Graphics Press, run from a converted garage. (For more than 20 years, the press stubbornly had a single author: Edward Rolf Tufte. Last year, he added his mother, Virginia, publishing her book Artful Sentences: Syntax as Style.) Tufte also makes sculpture—huge abstract pieces, often made from steel, installed on 23 acres in Cheshire, Conn., where he lives, and a 122-acre spread, Hogpen Hill, that he recently bought in nearby Woodbury.
He was a professor of political science, statistics and computer science at Yale and senior critic in the School of Art when he retired in 1999, weary of what he calls the “bureaucratic bloat” of academia. He also stopped consulting, frustrated because managers forced to listen to his suggestions rarely followed them. (He once told an interviewer that products under development “are in one of two states—either too early to tell or too late to change.’’) He works on the occasional industry problem, like runway maps, pro bono. Running his own enterprise, Tufte says, allows him to work “elegantly, intensely, gracefully and incredibly efficiently.”
He was born in Kansas City in 1942, and graduated high school in Beverly Hills, Calif. “Until I got there I thought I was one of the smartest people around,’’ he recalls. “Then we moved and suddenly 20 people around me were smarter than I was.’’ His father was an engineer and public works director; his mother, a reporter, went to graduate school when Edward left for Stanford and later became an English professor at USC. “We had numbers and words in the same house,’’ Tufte says, as if that explains how he got to where he is.
Maybe it does. His career is built on the way he joined words and numbers under one intellectual roof, marrying quantitative reasoning with communication as nobody else. In an interview in the mid-’90s with the Computer Literacy Bookshops in San Jose, Tufte explained that to do statistical design, one has to be able to see and to count. He claimed he didn’t see as well as many graphic artists and didn’t count as well as the best statisticians, but he did the combo better than just about anyone. His great insight was to think about graphics not as art or statistical constructs but as stories. He challenged chart-makers to ask the question: what is the story we’re trying to tell?
Tufte majored in statistics at Stanford. He didn’t much like the “frat-boy rowdiness” of the campus. But three professors influenced him deeply—the late statistician Lincoln Moses and political scientists Richard Brody and Raymond Wolfinger, now at Berkeley—and he stayed in close touch with them.
Tufte “was always trying to devise user-friendly approaches to statistics that nonspecialists like me could use,’’ Brody recalls. Ray and Barbara Wolfinger remember that young Tufte was like a grad student in his seriousness, curiosity and enthusiasm for faculty dinners. “Most undergraduates had no interest in us at all,’’ Barbara Wolfinger says. “Most were only interested in mating. Ed was different. There are very few truly unusual people one meets in life. Ed was certainly one of them.’’
Tufte says his mentors showed him “how successful scholars lived,’’ and he loved it. “I wanted to get done with school as quickly as possible and become a professor. I realized the academic world is a more humane and ethical place with better values than most of the world. . . . It’s also much more tolerant of idiosyncrasy and independence. I’ve always been contemptuous of authority. There aren’t better places than the university to do that and get away with it.’’
He finished his bachelor’s and master’s degrees in four years and got his PhD in political science from Yale four years later. He started painting the day he completed his dissertation—balloons rising to the sky, “unintentional but heavy symbolism,’’ he says. He spent the next 10 years on the public affairs faculty at Princeton. He wrote two well-received books on democracy and political control and co-authored another, on politics, with his Yale adviser, political scientist Robert Dahl.
In the mid-1970s, Tufte was asked to teach a statistics course to visiting journalists. He found the literature in the field thin, “grimly devoted to explaining the use of the ruling pen,’’ with nothing to say about quantitative reasoning. He started doing research, and the noted Princeton statistician John Tukey suggested they give a series of joint seminars. Terrified about performing in front of Tukey, Tufte threw himself into preparing for each class. Soon, he began weaving his notes into a manuscript.
He finished the book in 1982, after moving to Yale. No publisher would print it to his exacting standards. Tufte wanted the book to exemplify the design principles he articulated. It had to have lavish, abundant, high-resolution images and footnotes alongside the text so a reader wouldn’t have to flip pages to find a reference. The book had to be printed on thick, creamy paper and sell for a reasonable price, about $30. “Publishers seemed appalled at the prospect that an author might govern design,’’ he later wrote. So he took out a second mortgage at nearly 18 percent interest and produced the book himself.
The Visual Display of Quantitative Information was instantly hailed as a classic. Orders poured in. He repaid the loan within six months. Envisioning Information followed in 1990 and Visual Explanations in 1997. Tufte likes to point out that Galileo, too, helped finance the publication of his own books.
I don’t want to sound too majestic, but my books are forever knowledge,’’ Tufte says. “People will be reading them a long time from now.”
On a late spring morning, Tufte sits on a couch in his Cheshire home, in a room cluttered with books, papers, magazines, art. One of his sculptures dangles from the ceiling, a large bird made from corrugated aluminum folded like origami. He wears shorts and a T-shirt, slumps into the cushions and flips through Beautiful Evidence, page by page. Nine years in the making, the book arrived from the bindery the day before.
Like the earlier books, Beautiful Evidence isn’t an instruction guide but a statement of Tufte’s design principles: Show comparisons. Show causality. Show data in their full complexity. Document and display your sources. Above all, respect the intelligence of your audience and tell the truth. “Serious presentations,’’ Tufte often says, “rise and fall on the quality, relevance and integrity of the content.’’
By “forever knowledge” Tufte means his principles “are indifferent’’ to culture, gender, nationality or history. They apply to a 6,000-year-old cave etching, to the latest web design, to every map, chart and graph in between. The images in Beautiful Evidence “come from 14 centuries, 15 countries . . . three planets and the innumerable stars’’ to underscore the point that his principles are universal, apparently as immutable as the laws of nature.
Some critics—and even some fans—don’t buy it. “Tufte, you’re a smart guy, with good points to make,’’ one blogger moans. “Why puff it up with this foolishness about universality?” Questions of timelessness aside, the power and the pleasure of his books lie in their unabashed reach. Each offers a stunning array of images, from the brilliant to the ludicrous to the heartbreaking. A favorite image—one now almost synonymous with Tufte’s name—is a once-obscure 1869 statistical map of Napoleon’s march on Russia, by the French engineer Charles Joseph Minard. Tracing the number of soldiers across the land, through months and dropping temperatures, the map shows the French army was devastated not by the enemy but, in retreat, by the cold. “This is War and Peace, as told by a visual Tolstoy,’’ Tufte says.
But it’s not a single image that makes Tufte’s work memorable; it’s the mix and multitude. “Escaping Flatland,’’ the opening chapter of Envisioning Information, has 40 graphics in almost as many pages—a fairly typical count. The images skip seamlessly from Galileo’s sunspots to a 1937 timetable for a Java railroad to air pollution charts to notations of dance movements. One spreadsheet shows the crimes committed by government informants who testified against accused Mafia boss John Gotti; the ledger of murder, drug sales and pistol-whipping helped persuade a jury to dismiss the testimony as that of sleazy stool pigeons and acquit Gotti. Is there any stronger proof that graphics isn’t just for designers? And that presenting data isn’t necessarily dull?
Tufte says he was drawn to the field precisely because it is so wide-ranging. Still, it’s hard to imagine another scholar approaching charts and illustrations with such mirth, passion and oddball appetites. Tufte once attended a Russian satellite auction and bought a visual diary made by cosmonauts during their three-month space voyage. When he worked on redesigning medical records, he donned a white coat and hung out in a hospital. In Visual Explanations, he co-authored a chapter on magic with a professional magician, Jamy Ian Swiss. It is vintage Tufte—an exuberant illustrated analysis of card tricks, vanishing coins, flying water glasses unmasked. “To create illusions is to engage in disinformation design, to corrupt optical information, to deceive the audience,’’ the authors write. “Thus the strategies of magic suggest what not to do if our goal is truth-telling rather than illusion-making.’’
Nine years ago, around the time he began work on Beautiful Evidence, Tufte set aside painting and started sculpting. Large abstract pieces now rise from the landscape in Cheshire and Hogpen Hill. Spring Arcs, a series of solid stainless steel arcs that seem to squash and stretch depending on your vantage point, span 12 by 67 feet. The Millstone pieces are curved rust-colored giants, 11,000 pounds of scrap from the nuclear power plant from which they take their name. Dear Leader, installed in the winter but deemed finished only in May when grass was planted around it, is two giant porcelain and steel cylindrical shapes, Tufte’s vision of missiles fired from North Korea that plonk down in a Connecticut field. Here and there graceful aluminum birds seem to lift to the sky, like the balloons he painted so many years ago.
Tufte has exhibited his sculpture at shows in Los Angeles and New York and discovered two things. “I learned that the art needs to be outdoors, and that I don’t like hearing what people say about it.’’ Still, the man who has made a career of visual display cannot resist displaying the visuals closest to his heart. Beautiful Evidence closes with the sculptor’s photo album: seven two-page spreads showing Tufte’s work, photographed in various seasons.
Some longtime Tufte fans have responded with impatience. “Beautiful—but not on topic without stretching the imagination,’’ Stephen Few, a consultant who specializes in data visualization, writes on the online Business Intelligence Network. Zach Gemignani, a founder of Juice Analytics, a data-consulting firm, says, “I wish that Tufte would focus more on the current state of information visualization in business today and encourage vendors to make better tools.’’
But making better tools has never been Tufte’s mission. His passion is fundamentals—the accuracy of expression and the wonder of the spectacle. And these sculptures, sitting on the grass or floating free in space, are wonderful spectacles. Changing with the shadows and the seasons, they grab a blade of grass, a buttercup, a mound of snow and reflect it back, transforming the familiar into an image to behold. Escapes from flatland, every one.
Read a May 2010 update on this story.
Read a November 2010 update on this story.
FRAN SMITH, a 1993 Knight fellow, is a writer and editor in New York.
The Effort Effect
What It Takes
What to do With Your VHS Tapes: Essential Answer
Bananas Are Berries?
The Menace Within
Data is from the past two weeks.