Last Updated Jan 30, 2011 1:20 PM EST
Technically, the cause of the shuttle's fatal explosion turned out to be a small rubber ring which cracked when subjected to exceptionally cold weather prior to launch and the extreme heat at launch. The fault was well understood -- so why had the shuttle been allowed to take off? Feynman described with some scorn the political pressure felt by engineers to deliver to then-President Reagan a story with feelgood PR:
"For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled," Feynman wrote. Today, most of us would be surprised to encounter anyone who felt that science regularly prevailed over PR. Feynman took it for granted that once everyone realized the trade-off between Nature and politics, they'd make the safer choice. Few would be so certain today.
But more specifically -- and, as usual, decades ahead of his time -- Feynman identified the central problem: When NASA took that fatal, PR-led decision, critical expertise was not at the table. Why not?
The people who really understood the 'O' rings weren't at the NASA meetings because they didn't work in NASA, only for NASA. The 'O' rings were outsourced to the manufacturer Morton Thiokol who in turn used plastic made by the Parker Seal Company. Moreover, the flight team was based not only at the Kennedy Space Center in Georgia, but also at the Marshall Space Flight Center in Huntsville, Alabama. The Morton Thiokol team was based in Brigham, Utah, while the Park Seal folks worked in Lexington, Kentucky. On no two conference calls - initiated to explore concerns - were the same people present. Information fell through gaps bigger and more important than the 'O' rings themselves.
But it wasn't only geography and rotating executives that posed the problem. Morton Thiokol engineers couldn't raise a red flag until they got support from their management. As mere subcontractors, they had no leverage; after all, NASA was their customer. So NASA was under political pressure to launch, Morton Thiokol was under commercial pressure -- and had no clout in the chain of command.
Even during the inquiries that followed the disaster, the problems of access and communication were obvious. A director from Morton Thiokol had to gatecrash one of Feynman's meetings to be able to share what he knew. Disaggregation of information was made worse by disequilibrium in power.
System failure -- and BP
And Feynman uncovered built-in blindness: the way relationships were structured made it impossible for the right people to use the right information effectively. This was not about bad, weak or stupid people: it was about power and how information flowed. When companies and projects are sufficiently so spread out and complex, you virtually guarantee that one hand cannot know what the other is doing - and may not even want to.
Feynman would have recognized the picture that is emerging from the explosion of BP's Deepwater Horizon rig. BP, of course, did not make the rig. It was built by Hyundai in Korea to a design by a Texas firm, R&B Falcon, bought by the Swiss operators, Transocean, who leased the rig to BP. Most of the workers killed were not BP employees but non-union subcontractors. Once you sub-contract at this level, you stand a high likelihood of being blind to how work gets done.
Not just shuttle missions
The most practical lesson we learned from the Challenger explosion was that outsourcing increases risks. Whatever we gain in expertise or save in time isn't worth what we jeopardize by distance and deference. And those dangers are bigger now than ever.
We outsource wars, through private security firms, and much of our policing: in the US and UK, the number of private guards is now more than twice the number of public police officers. Current debates over the outsourcing of clinical trials for new drugs to India, China and other cheap labor markets, focus primarily on whether ethnic difference will alter results. But those debates overlook the bigger issue: the imbalance of power between rich Western pharmaceutical companies and poorly paid, unprotected volunteers half a world away. Disregarding these inequalities of power is a form of blindness in itself.