Casual Causality: Beyond Proximate Cause; or, The O-Ring Syndrome Revisited

An accident is an event that is fortuitous and unforeseen. That’s how life works. Yet we try to figure out what we did to deserve it and why we didn’t see it coming. But if we can figure it out, then it was not an accident. We might know how, but we can’t know why.

First comes love, then comes marriage, then comes mama with the baby carriage. Love is an accident. BP’s drills appear as effective in avoiding undesired consequences as the Catholic Church’s promulgating the rhythm method for birth control. Both allow drilling to continue as a risky business with high promise of failure. For the Church, the method aimed to keep pews full; for BP, the deep water drilling without adequate protection aimed to keep its stocks on the rise.

Ellen Goodman, writing in the Tri-City Herald back in 1983, discusses an ultimate proximate cause she labels “the O-Ring Syndrome,” the tendency to view catastrophic events as triggered by the failure of minute considerations. The term comes from the dramatic and awful disintegration of the 1986 space shuttle Challenger. The mission failure was attributed to an O-ring that malfunctioned, a tiny rubber seal that failed to perform its job and set in motion a chain of events that ultimately ended in damage and death. A correlative cause may have been cold weather. In any case, working backward from the first moment of irreparable or irreversible harm, one stops at the O-ring because it appears to be the end (and thus the beginning, or proximate cause) of the unbroken chain of events that lead to the accident.

The problem with Goodman’s O-Ring Syndrome is that as an explanation it creates an illusion of control. If the O-ring is properly identified as the cause, we can avoid future accidents by ensuring the O-ring does not fail in the same way again, because, applying the “but for” test, had it not been for the failure of the O-ring, the accident would not have occurred. But the O-ring Syndrome as an explanation for accidents limits the explanation to events of physical or bodily damage. But what if an O-ring fails but does not lead to damage? Can the definition of accident be broadened to include events that do not necessarily lead to physical or bodily damage? If so, in the case of BP and the current Gulf oil leak, multiple accidents have already occurred.

Finding the O-ring responsible for BP’s oil leak isn’t difficult, and correlations from ineffective regulatory agencies to manipulated workers to greedy BP executives have been suggested. If we look beyond the O-ring, prior to proximate cause, to explain the accident, we find BP operating like one of E. O. Wilson’s massive ant hills, a single organism that is emergent from smaller parts, not predictable from any one part. BP’s ant hill includes what we value, and what we value isn’t necessarily good for us. We value oil. The oil leak in the Gulf is an accident of value.

James Surowiecki, on the regulation crisis in the June 14 & 21 New Yorker, speaking of the recent rash of financial accidents, Bernie and the [other] Jets, (as well as the Massey mining disaster) correctly estimates that “these failures weren’t accidents.” But he’s still caught in the O-Ring Syndrome: “They were the all too predictable result of the deregulationary fervor that has gripped Washington in recent years, pushing the message that most regulation is unnecessary at best and downright harmful at worst.” If the events were predictable, they were not accidents. The proximate cause is value, what we want. Moreover, in many cases, the failures did not lead to physical damage or bodily harm; to the contrary, the bailout, like some benevolent insurance policy, made some people whole again and more. The mistake is to assume that adequate regulation would have prevented the failures. Adequate regulation may have minimized the frequency and severity of the events, but the only way to avoid accident is to avoid risk, because of a true accident, we can never know why.