In May 1996, ValuJet Flight 592 crashed into the Florida Everglades six minutes after takeoff from Miami, killing all 110 people on board. Investigators traced the fire to chemical oxygen generators loaded into the forward cargo hold without safety caps on their firing pins. What they could not trace was a single point of failure, because there was not one.

We work through the layered collapse William Langewiesche documented in his landmark 1998 Atlantic article: work orders written in language the mechanics could not parse, safety caps that did not exist anywhere in the shop, paperwork signed off on work that was never done, a shipping clerk who put quotation marks around the word "Empty" on the manifest, and a copilot who recognized what he was looking at and said nothing. Drawing on Charles Perrow's theory of the normal accident and Diane Vaughan's concept of the normalization of deviance, we examine how the same mechanism that produced Challenger produced this, and where the two failures diverge.

The deeper question the episode keeps returning to is one Langewiesche raises and does not fully resolve: if the failure emerged from the gaps between organizations rather than from within any one of them, and if adding more procedure to a system can increase the complexity that makes these accidents possible, what actually closes the gap between what the paperwork says and what happened on the floor?

Links from the discussion:

"The Lessons of ValuJet 592" by William Langewiesche, The Atlantic (1998): https://www.theatlantic.com/past/docs/issues/98mar/valujet1.htm

Normal Accidents: Living With High-Risk Technologies by Charles Perrow: https://press.princeton.edu/books/paperback/9780691004129/normal-accidents

The Limits of Safety: Organizations, Accidents, and Nuclear Weapons by Scott Sagan: https://press.princeton.edu/books/paperback/9780691021010/the-limits-of-safety

The Challenger Launch Decision by Diane Vaughan:https://press.uchicago.edu/ucp/books/book/chicago/C/bo22781921.html