I just finished a marvelous book with the dry-as-dry-toast name The Visual Display of Quantitative Information, by a Yale statistician and sculptor named Edward R. Tufte. I’ve come to learn that the book is considered the Elements of Style of charts – a true classic.
This may be apocryphal, but he is known for the great quote — even if he didn’t actually say it — “every time someone uses Powerpoint, a kitten dies.”
One thing he attacks is “statistical lies”: for instance, a chart showing that government spending seems to be skyrocketing because the numbers are increasing radically – but failing to mention a little thing called inflation. Nearly all newspapers, magazines, and even academic journals are guilty of this, he writes. The Economist earns his praise for being one of the better magazines out there in this regard.
“Chartjunk,” as he calls it, is another disgrace to the graphics profession – badly designed and/or unnecessary stuff that clutters up all too many charts.
His key point is that over the years design and words have been separated, and not to either field’s advantage. He notes that in Leonardo da Vinci’s work, for instance, his writings and his art are intricately meshed, but that in the ensuing years the two fields have diverged. Since that time, graphic designers have little training in statistics and economics, and economics people have little training in graphics, and we poor English majors don’t know anything about anything.
I was thinking back at our recently completed work (hooray, hooray) on our Encyclopedia of Sustainability. One of the cardinal rules in our authors’ guidelines is to make sure to “not rely on tables and figures when text would convey the idea more clearly, and vice versa” – which I think Tufte would approve of.
What I don’t think Tufte would approve of is my previous attitude toward charts and tables. I consider myself a word person, and throughout work on the Encyclopedia of Sustainability I stuck to the credo, “if I can’t understand it easily, I don’t want it in the book.” This is all fine and well, and was well-intentioned, but the point of The Visual Display of Quantitative Information is that if graphics are confusing, they’re not done well, for a variety of reasons (which he goes into in some detail). As publishers, we owe it to our readership to assume that they will be intelligent enough to digest the data if they’re presented in a well-designed manner.
After reading this book I’ll never look at charts and tables in newspapers in quite the same way again. I also want to check out the author’s sculptures sometime: I understand that he had a show at a modern art museum nearby a couple of years ago. I just hope there were no kitten sculptures.