Tufte's books are spry and smart, due largely to the extraordinary visual examples he uses throughout -- everything from a reproduction of two pages from Galileo's notebooks to an annotated medical-costs list to a stamp from the San Francisco Match Co. This appeal is borne out by the books' huge sales (over half a million copies total, at $40 to $48 apiece) and by Tufte's financial success (by 1998, the year after his third book, Visual Explanations: Images and Quantities, Evidence and Narrative, came out, he'd grossed about $25 million, according to the New York Times). His first title has just been released in a second edition and has been translated into seven languages. All three have won numerous awards and been reviewed, mostly positively, in an absurd range of places, from I.D. to PC Magazine to the Journal of the American Medical Association.
Supplementing the books -- and a major source of income and publicity -- are Tufte's one-day courses, which he began giving in 1993. At first they concentrated on a few cities, but since he stopped teaching statistics and design at Yale two years ago after being named professor emeritus, he's taken the tour on the road, hitting almost a dozen towns around the country in 30 sessions a year. Each class costs $320 for about 5.5 hours of real instruction, plus copies of the books and a poster. He gave one such class in San Francisco last June, which I attended. (He'll come again on Dec. 3 and 4 and swing through Palo Alto on Dec. 5 and 6.)
Not to put too fine a point on it, but the course is a waste of money. Sure, it's entertaining, and Tufte graciously autographs each book, as long as you approach his table with your copy opened to the title page, as instructed in the handout. The class offers little beyond what you could read on your own -- and even then, Tufte says, "We cover about 15 percent of what's in the books. ... [After all,] we have just a day." The most serious problem with the session, however, is that it violates several of Tufte's own principles, as outlined in his books.
Edward R. Tufte has a lovely smile, producing dimples and crinkly eyes. He flashes it often during his June course at the Westin St. Francis, which is filled to capacity with almost 350 people. A smartass sign on his autograph table explains that he won't write personal inscriptions, such as, "I'll never forget our times together. E.T." Tufte tells a riveting story as he paces between two enormous screens, holding up as props original editions of books by Galileo and Copernicus that assistants wearing white gloves then walk around the room. Sometimes he speaks of himself in the third person or talks about his "author's vanity." The mild megalomania is no act: Tufte wrote, edited, designed, printed, marketed, warehoused, and distributed his books himself. Each took seven years to create, and each purports to cover a different aspect of information design, though there are many overlaps (including illustrations that appear in more than one volume).
As Tufte writes in Visual Display, "Every bit of ink on a graphic requires a reason. And nearly always that reason should be that the ink presents new information." But his courses do not add anything fresh to what can be gleaned from his books. Every image he projects in his presentation can be found on their pages -- appearing clearer on paper, as he points out in Envisioning Information, than they could on screen. (Tufte says that the latest classes include some information from an upcoming, unpublished book.) His anecdotes feel canned. His statements are pithy ("Good design is clear thinking made visible, and bad design is stupidity made visible"), but lifted straight from his publications. Eugene Kim, a 1997 attendee and admitted fan, describes the show on a Web site this way: "In some ways, Tufte used sleight of hand to make the course seem more substantial than it really was. ... Most of his talk and examples were derived from his three books." Still, Kim goes on, he's so "dynamic" that it was worthwhile.
Tufte's dynamism isn't in question. As he says, "I believe this stuff; I care about it; I'm enthusiastic. I don't regard that as a detriment." But as he points out in Visual Display, "[d]ata graphics should draw the viewer's attention to the sense and substance of the data, not to something else." These courses draw attendees away from the message and toward the messenger. In my experience, Tufte spends a lot of time tooting his own horn. He performs. He acts as a shill to sell stuff (you can buy his books and poster, along with special graph paper, art prints, and even a Tufte sculpture for $200,000, from his Web site, www.edwardtufte.com). Tufte explains, "It's certainly better than being a bad teacher, isn't it?" Of course this is true, but that's not the point.
What transpires after a day listening to Tufte is what he calls "non-information overload." One comes away with a vision of the man, and the rest is background noise. Grace Aquino, an editor at PC World, an admirer of Tufte since she took his class a few years ago, and a friend of mine, describes it this way: "The lasting impression was that he was a great speaker." Though she doesn't recall the details of what Tufte said, "He sticks with me." Tufte insists that he gets "tons of evidence" that people are using his information, including samples of revised work. But it's more likely that they are using his books, rather than his lectures, for guidance. What if people don't retain what they learn in class? "Well, isn't that too bad," Tufte says mournfully.
If the lesson is overwhelming, he says, "It's my fault, not theirs ... I don't worry about overload; I want to make it as thorough as possible." But in trying to make the course "thorough," Tufte spreads his information thin, making it so general that it can't be useful to the needs of his diverse audience. A handout seems to bear this out: It gives recommendations for chapters and pages to read "according to your interests." To me, the class was little more than an amusing day away from work. For $320 (minus the $133 cost of the books), one could see 23 movies and be just as entertained.
In Envisioning Information, Tufte gives grief to some engineers who put a note at the bottom of a group presentation stating that it "cannot be considered complete without the oral discussion." Tufte grumbles, "Such defensive formalisms ... suggest a corporate distrust both of the chartmaker and of any viewers of the chart." But Tufte's courses suggest he distrusts his own readers.
Like an actor who really wants to direct, Tufte is a scientist who yearns to be an artist. His next book, which on his current schedule should come out in 2004, will be Beautiful Evidence: Seeing and Showing. All of his books are beautiful, but they are clearly the work of a scientist -- albeit a witty, well-read, abundantly curious scientist. His shows, on the other hand, are the work of a ham.