As a presenter and analyst, Tufte ranks extremely high. He has a way of deflating the pompous and seeing through the clutter, then conveying effectively what he sees and what you should see. “Beautiful Evidence” picks up the thread of the three earlier books, and expands deeper into some of the concepts, and suggests some new ideas.
“Mapped Pictures,” the first chapter, is an attempt at applying formal map-style methods of labelling for the analysis of graphics. The point made is that “mappings help tell why the image matters.” Examples range from a Hockney painting to renaissance engravings. He leans heavily in favor of captioning and mapping directly on the picture, and not as a subsidiary table somewhere else on the page. In fact, a recurring theme in the book is to keep explanaory text, explanatory illustrations, and captions close together.
There’s an amusing debunking of one man’s method of formally analyzing sculptures based on key structural points which generate universal symmetrical geometries. The problem is, many possible mappings are available, applicable to many different pieces of art. Mappings must be “specific, coherent, credible and testable.” This is also applicable to certain types of research, covered in the chapter on “Corruption in Evidence Presentations.” The same cherry-picking process that led to the meaningless though pretty geometries can also lead to meaningless though pretty analytical conclusions. It gets serious when applied to, say, pharmaceutical research.
“Words, Numbers, Images” picks up the theme of proximity, with some fine examples of the artistic ways that text and illustration can be seamlessly combined. An especially amusing example is the “Government Sentencing Memorandum” issued in the “Duke” Cunningham case. Nested in with the dry legal text are full-color photos of the antique furniture, mansions, the yacht, as well as the notorious “menu” of bribery options. The chapter ends — I want to say climaxes — with Galileo’s beautiful diagram of the Pleiades as seen through his newly invented telescope. The chain of stars as if by magic breaks the grid of the book and streams across the center.
That’s the good news; the bad news is much of the rest of it feels like padding. “The Cognitive Style of Powerpoint” is a wonderful monograph; if you, like me, purchased it separately you’ll be dismayed that you also get it in this book in its entirely. I guess it’s akin to including the hit single in the album. Now if you haven’t read it by all means you should. Powerpoint’s insidious way of not representing your thoughts, but instead channeling and limiting them to fit its own particular toolset, is disturbing. And again, it can be a life-and-death matter; the slideshow used to decide what to do about the space shuttle Columbia, while it was still in orbit, was done in Powerpoint and abetted in totally disguising the most important facts.
“The Fundamental Principles of Analytical Design” picks up with the famous Minard drawing of Napoleon’s retreat from Russia. Now just in case you missed the three reproductions of it in “The Visual Display of Quantitative Information,” he reproduces it five more times in this volume, including an “actually very nice” foldout version. Obviously he’s demonstrating an earlier principle that the proximity principle applies even if it means reproducing an image or graphic more than once. Maybe so; but five times just makes it feel like, well, Phluff.
The last section is very nice but I have no idea why it’s here. Tufte is also a sculptor, working in large-scale outdoor metal works. His suggestion that pedestals take up their own space and so modify the experience of the sculpture is well taken. I was hoping for more of a helpful analogy to the “big all-purpose picture frames,” perhaps even an application of Edward Cone’s suggestion of the silence before and after a performance of music as a structure in its own right. The example of the enormous toppled Stalin, where the empty plinth took on a symbolic life of its own, could be lined up with Cage’s 4’33”, where the empty sound becomes the work, or Rauschenberg’s white paintings.
But none of this has anything to do with statistical analysis of graphic descriptions of information.
All that said, it’s worth a look at least. I absolutely endorse studying his work (my copies of his first three books are autographed!) This one is for a serious student of the craft who wants to explore further examples. But if you’re just beginning to explore Tufte’s work, I’d suggest you go first to “The Visual Display…”, which is dense with fine examples and counterexamples. It’s like comparing “Animal Crackers” to “A Night At The Opera”; the jokes are just as funny, but there’s an awful lot more of them in the former than in the latter.
Additionally for the graphic artist, “Envisioning Information” is terrific, with many absolutely elegant examples.
Visit http://www.edwardtufte.com for more about these books, along with some great unpublished material and discussions of these ideas.