When I was a young undergrad living in the student slums of Auburn, Alabama I had a neighbor named Pat Duke. Pat was short, stocky, full of energy, and irresistible to the ladies, but that’s not where I’m going with this. Pat drove a Honda CVCC that was ancient even in that long ago time. It ran like a top. There wasn’t a linkage, belt, hose, or bearing that Pat hadn’t personally removed, inspected, and reinstalled. Pat repaired electronic equipment to make pocket money. Once, when I took him an amplifier that was misbehaving he examined it for a few minutes and said: “Oh yeah, they always break like that.” What I’m trying to say is that Pat had a finely honed sense of all the modes of failure of the many imperfect things in this world. So does Henry Petroski, author of Success Through Failure.
As he examines the evolution of such things as slide presentations, band-aids, and automobiles as well as buildings and bridges, over and over again he posits the notion that failure or the observation of an imperfect solution is more instructive to the designer than the emulation of past success: “Our built environment is populated by structures that have survived proof tests of all kinds. We term such structures successful designs, and they are. However, successful designs do not necessarily tell us very much about how close to failure they are. Sometimes cracks develop, which signal problems, but they can be attributed to settlement and interpreted to be the cracks of age. Any failure, however, is incontrovertible evidence that weaknesses existed – in the design, the workmanship, the materials, the maintenance, or the defense against terrorists. Failure is the counterexample to the hypothesis of success. This again is the paradox of design: Things that succeed tell us little beyond the fact that they have been successful; things that fail provide incontrovertible evidence that the limits of design have been exceeded. Emulating success risks failure; studying failure increases our chances of success.”
Petroski borrows his definition of failure from the American Society of Civil Engineers, stated as: “an unacceptable difference between expected and observed performance.” The troubling part of avoiding failure is the expectations part. In my experience, most problems can be solved if they are correctly defined. With the tools and expertise to solve a wide array of problems, understanding the problem in many ways is the problem.
In the world of architecture every building is its own prototype. There are no test runs and no do-overs. In architecture “fail early and fail often” is not an option. And so, unfortunately, some small degree of failure in the sense of unmet expectations is the norm in most projects, whether it’s a consequence of design or workmanship or confusion about what was intended or wanted. Clients frequently don’t tell us everything, either because they think we should already know or because they’re just too busy or distracted to give their project the attention it deserves. Oftentimes vendors, eager to sell their products, aren’t forthcoming about the limitations. Workmen show up drunk or hung over and cover up faulty work because they’re tired or indifferent. Usually these problems get resolved as they arise and good designers have learned to avoid the big failures and adapt to the small ones.
How do you ask? Imagine disaster. Imagine tornados; imagine wet dogs and sick kids; imagine driving rain and burrowing insects. Imagine lover’s quarrels. If being a successful designer is avoiding failure then every successful design team needs a Jeremiah. That guy or girl who imagines Armageddon in every decision from the color selections to the bolt patterns. That’s me. Living every imaginable excruciating mistake in my head so that you don’t have to.