The Design Observer Group


Posted 05.09.05


Michael Bierut

On (Design) Bullshit


In Concert of Wills, the fascinating 1997 documentary on the building of the Getty Center in Los Angeles, architect Richard Meier is beset on all sides by critics and carpers: homeowners who don't want the Center's white buildings ruining their views, museum administrators who worry that the severe stone benches will be uncomfortable, curators who want traditional molding on the gallery walls. The magisterial Meier takes them all in stride, until one moment that is the hold-your-breath climax of the film.

The client, against Meier's advice, has brought in artist Robert Irwin to create the Center's central garden. The filmmakers are there to record the unveiling of Irwin's proposal, and Meier's distaste is evident. The artist's bias for whimsical organic forms, his disregard for the architecture's rigorous orthonography, and perhaps even his Detroit Tigers baseball hat all rub Richard Meier the wrong way, and he and his team of architects begin a reasoned, strongly-felt critique of the proposed plan. Irwin, sensing (correctly, as it turns out) that he has the client in his pocket, listens patiently and then says, "You want my response?"

His response is the worst accusation you can lodge against a designer: "Bullshit."

This single word literally brings the film to a crashing halt: a very long fifteen seconds of dead silence follows, broken at last by an awkward offscreen suggestion that perhaps on this note the meeting should end, which it does.

What is the relationship of bullshit and design?

In asking this question, I am of course aware that bullshit has become a subject of legitimate inquiry these days with the popularity of Harry G. Frankfurt's slender volume, On Bullshit. Frankfurt, Professor of Philosophy Emeritus at Princeton, is careful to distinguish bullshit from lies, pointing out that bullshit is "not designed primarily to give its audience a false belief about whatever state of affairs may be the topic, but that its primary intention is rather to give its audience a false impression concerning what is going on in the mind of the speaker."

It follows that every design presentation is inevitably, at least in part, an exercise in bullshit. The design process always combines the pursuit of functional goals with countless intuitive, even irrational decisions. The functional requirements — the house needs a bathroom, the headlines have to be legible, the toothbrush has to fit in your mouth — are concrete and often measurable. The intuitive decisions, on the other hand, are more or less beyond honest explanation. These might be: I just like to set my headlines in Bodoni, or I just like to make my products blobby, or I just like to cover my buildings in gridded white porcelain panels. In discussing design work with their clients, designers are direct about the functional parts of their solutions and obfuscate like mad about the intuitive parts, having learned early on that telling the simple truth — "I don't know, I just like it that way" — simply won't do.

So into this vacuum rushes the bullshit: theories about the symbolic qualities of colors or typefaces; unprovable claims about the historical inevitability of certain shapes, fanciful forced marriages of arbitrary design elements to hard-headed business goals. As Frankfurt points out, it's beside the point whether bullshit is true or false: "It is impossible for someone to lie unless he thinks he knows the truth. Producing bullshit requires no such conviction." There must only be the desire to conceal one's private intentions in the service of a larger goal: getting your client it to do it the way you like it.

Early in my life as a designer, I acquired a reputation as a good bullshitter. I remember a group assignment in design school where the roles were divided up. The team leader suggested that one student make the models, another take the photographs, and, finally, "Michael here will handle the bullshitting." This meant that I would do talking at the final critique, which I did, and well. I think I mastered this facility early because I was always insecure about my intuitive skills, not to mention my then-questionable personal magnetism. Before I could commit to a design decision, I needed to have an intellectual rationale worked out in my mind. I discovered in short order that most clients seemed grateful for the rationale as well. It put aside arguments about taste; it helped them make the leap of faith that any design decision requires; it made the design understandable to wider audiences. If pressed, however, I'd still have to admit that even my most beautifully wrought, bulletproof rationales still fit Harry Frankfurt's definition of bullshit.

Calling bullshit on a designer, then, stings all the more because it contains an element of accuracy. In Concert of Wills, Richard Meier is shown privately seething after Robert Irwin drops the b-word. "For one person to say," he tells the camera, "I want my object, I want my piece, to be more important than the larger landscape of the city...that my individual artwork is the controlling determinant, makes me furious, just makes me angry beyond belief." Of course, that same accusation could be leveled against Meier himself, who out of necessity had been nothing if not single-minded and obstinate during the endless process of designing and building the Getty. The difference is that each of Meier's victories was hard-won, with endless acres of negotiating, reasoning, and you-know-what expended in the process of winning over the project's army of stakeholders. On the other hand, Robert Irwin, flaunting intuition and impulse as his first, last and only argument, required no compensating bullshit: he's the artist, and that's the way the artist likes it. Can you blame Meier for finding this maddening?

Every once in a while, however, there is satisfaction to be had when design bullshit attains the level of art. I remember working years ago with a challenging client who kept rejecting brochure designs for a Francophile real estate development because they "weren't French enough." I had no idea what French graphic design was supposed to look like but came up with an approach using Empire, a typeface designed by Milwaukee-born Morris Fuller Benton in 1937, and showed it to my boss, Massimo Vignelli. "That will work," he said, his eyes narrowing.

At the presentation, Massimo unveiled the new font choice with a flourish. "As you see," he said, "in this new design, we're using a typeface called Ahm-peere."

I was about to correct him when I realized he was using the French pronunciation of Empire.

The client bought it.