Loss of USS Scorpion

The loss of the U.S. submarine Scorpion in May 1968 became a striking demonstration of collective judgment under extreme uncertainty. Naval officer John Craven did not ask a handful of top specialists to agree on a single answer. He gathered independent estimates from people with different kinds of knowledge, combined them mathematically, and produced a location no individual had chosen—one that proved remarkably close to the submarine’s actual resting place.

Disappearance and the Search Problem

In May 1968, the Scorpion disappeared while returning to Newport News after a tour of duty in the North Atlantic. The navy knew the submarine’s last reported position, but not what had happened to it or how far it might have traveled after its final radio contact. The initial search area was a circle twenty miles wide and thousands of feet deep, an expanse that made the task seem nearly hopeless.

The obvious alternative was to locate three or four leading experts on submarines and ocean currents, ask where they believed the Scorpion lay, and search that spot. Craven, a naval officer, chose a different method. The central difficulty was not simply a lack of expertise: no one had enough information to reconstruct the submarine’s fate alone.

Craven’s Distributed Investigation

Craven devised a set of scenarios offering alternative explanations for what might have happened to the Scorpion. He then assembled men with varied expertise, including mathematicians, submarine specialists, and salvage men. Rather than having them deliberate together toward a consensus, he asked each participant to estimate how likely each scenario was.

The estimates took the form of wagers, with bottles of Chivas Regal as prizes. Participants bet on separate elements of the submarine’s fate: why it had run into trouble, how fast it was moving as it descended, and how steeply it fell toward the ocean floor. Each estimate addressed only part of the problem; none could locate the submarine by itself. The independence of the judgments allowed the group’s separate knowledge to be combined without requiring any member to possess a complete account.

Combining the Judgments

Craven used the wagers to assemble a composite picture of the Scorpion’s final moments and applied Bayes’s theorem to estimate its final location. The theorem provides a way to calculate how new information changes prior expectations about the likelihood of an event. In this case, the separate estimates were brought together into a collective estimate of where the submarine had come to rest.

The resulting location was not a point any member of the group had independently selected. No participant had the same picture of the Scorpion’s fate that emerged from the combined information. The answer was therefore not merely the judgment of the group’s most capable individual, but a genuinely collective one.

Discovery and Significance

Five months after the disappearance, a navy ship found the Scorpion 220 yards from the location Craven’s group had predicted. The result was striking precisely because the underlying evidence consisted of small scraps. No one knew why the submarine had sunk, how fast it had traveled, or how steeply it had descended to the ocean floor. Yet, as the account emphasizes, the group’s combined judgment supplied an answer no individual member had possessed.

The episode contrasts the apparent necessity of a lone expert with a method that makes useful knowledge from partial, distributed judgments. Its force lies not in any one wager, but in the distance between what the individuals knew and what their combined estimates could establish: the collective answer was both new and accurate enough to guide the search to within 220 yards.