Cataclysm begets change. It disrupts the inertia to which human beings and their institutions are prone. Oftentimes nothing less will do. As Niccolò Machiavelli pointed out centuries ago, individuals find it hard to change with the times and thus risk being left behind. Organizations are bodies made up of individuals allergic to change. Those who found and preside over institutions write their worldviews into standard routine. Institutional culture is intractable as a consequence. It takes a powerful jolt of some sort—defeat in the case of an armed service, or some other trauma of like magnitude—to set overdue reforms in motion.

Exhibit A: the loss of the nuclear-powered attack submarine USS Thresher in April 1963. The first of a new class, Thresher sank off the New England coast while conducting sea trials. The calamity came not long before the loss of USS Scorpion in 1968. Together the accidents shocked the silent service into remedial action. The reforms put in place were salubrious. Never again has the U.S. Navy submarine force lost another boat to unsafe matériel or practices.

The Thresher disaster supplies enduring lessons not just for undersea services but for naval forces in general. But first the facts of the case insofar as they can be known. Last week the U.S. Navy published the proceedings of the Court of Inquiry that convened at the submarine hub of New London, Connecticut immediately after the sinking. The report reviews the facts in language that’s chilling because it describes mass death in dispassionate and clinical terms. Early on the morning of April 10, it says, the boat “reported by underwater telephone that she was starting a deep dive.” According to observers aboard the submarine rescue ship Skylark, cruising in company with the sub, the deep dive went off without incident for just under ninety minutes. Then the leadership on board Thresher informed Skylark: “Experiencing minor difficulties. Have positive up angle. Am attempting to blow. Will keep you informed.”

Subs regain buoyancy by using compressed air to expel water from their ballast tanks. It appears Thresher was in distress and trying to surface by the time the initial message came in. Three minutes later, as the court put it, Skylark “heard a garbled transmission which was believed to contain the words “test depth,” suggesting the boat had descended to the maximum depth at which peacetime operating procedures permitted it to operate. (Submarine-force overseers allow a healthy safety margin between routine operating depths and a boat’s “crush depth,” where the water pressure is so intense that the hull can no longer withstand it. It collapses.) The boat kept plunging. Another garbled transmission came in four minutes later, “reported as containing the words ‘. . . nine hundred North.” A minute after that Skylark sonar operators heard “a high energy, low-frequency noise disturbance of the type which could have been made by an implosion.”

Concludes the court’s account, Thresher “was lost at sea with all on board at about [9:18 a.m.] on 10 April 1963, in the vicinity of Latitude 41-45 North, Longitude 65-00 West.” After taking testimony and reviewing the evidence the Court of Inquiry ruled that the loss of the boat stemmed “in all probability” from an “initial flooding casualty from an orifice between 2” and 5” in size in the engine room.” The flood apparently continued and was compounded by a loss of reactor power “due to an electrically-induced automatic shutdown.” The court further judged that operating procedures in place were inadequate “with respect to minimizing the effects of a flooding casualty and the loss of reactor power.” Lastly, it found the “main ballast tank blow system” deficient, meaning the equipment used to buoy the ship was “susceptible to freeze-up, with low capacity and low blow rate.”

In other words, a cascade of troubles beset the submarine. A flexible hose, a pipe, or, more likely, a “silver-brazed” joint connecting two sections of pipe probably gave way under pressure; saltwater may have splashed up into the switchboards, inducing a reactor shutdown; and the main-ballast-tank blow system iced up, keeping the crew from lightening the ship sufficiently to return to the ocean’s surface. Moreover, procedures for coping with such hazards were immature. Board members conjectured that three scenarios could explain how the series of events had unspooled but—given the paucity of evidence—concluded it was impossible to be certain which. The Court of Inquiry closed out its report by recommending that “early consideration be given to the establishment of an organization, similar to that employed in Naval Aviation, in the interest of safe submarine operating procedures.”

And indeed the silent service instituted such a program, titled SUBSAFE, shortly after the loss of Thresher.

And the lessons for today? First of all, getting the hardware dimension right on the first ship in any class—subsurface or surface—is hard. Blueprints for a ship amount to a theory about how machinery will perform under real-world circumstances. They have to be reduced to engineering, built, and subjected to the test of reality. Problems arise when evaluating even the finest design. Only after they have been fixed (or sometimes circumvented) is it possible to say field trials have ratified designers’ vision. Vetting a submarine class was doubly troublesome in 1963, when not just the Thresher class but nuclear-powered submarines, full stop, remained a new and somewhat mysterious quantity. After all, only eight years had elapsed since USS Nautilus, the world’s first nuclear-powered sub, sent the famous radio signal “underway on nuclear power.” The U.S. Navy was still learning about its new creations.

Or as the Thresher Court of Inquiry noted, “the complexity of modern submarines has increased at a rapid rate” with the “advent of nuclear propulsion, ballistic missiles, and greatly increased speeds and operating depths.” Nuclear submarines “make many more excursions to test depth” than their diesel-electric forerunners had in the past. The unknowns hidden within this “increased number of cycles” between the surface and the intense water pressure found in the depths demand “periodic surveillance of submarine hulls.” And of their internals.

Second, basic workmanship matters. The court paid intimate attention to the number and soundness of silver-brazed joints on board Thresher, viewing them as a likely source of seawater flooding and thus the reactor shutdown. A boat thus equipped had approximately three thousand brazes in hazardous systems; some 13.8 percent of brazed joints originally installed in Thresher—some 400 junctures—had been rejected during non-destructive testing. For the court the rejection rate constituted “a clear indicator that additional action was required.” Board members took shipyard managers to task for declining to strip away “lagging”—heavy padding that encases pipes or machinery to inhibit unwanted transfer of heat—to test brazed joints underneath for flaws. Defective joints may have gone unnoticed.

The court urged shipyards to replace brazed joints in hazardous systems with welds, a stronger method for joining piping systems, when brazes came up for replacement. Members also admonished yard managers to redouble their commitment to quality assurance, embracing ultrasonic testing in particular. By the time of Thresher’s post-shakedown yard period, they declared, “there had been a sufficient number of serious failures of [silver]-braze piping joints in submarines to require thorough investigation by all responsible” for the boat’s safety. Exacting standards are a must.

Third, battle testing amplifies the likelihood of an engineering failure. USS Thresher underwent “shock trials” in the vicinity of Key West during the summer of 1962. Shock testing exposes a ship from a new class to nearby underwater explosions to determine how the hull and innards will stand up under combat conditions. It vindicates a design while identifying improvements that need to be incorporated into subsequent copies of the ship. While the Court of Inquiry observed that “there was no loss of main power” during Thresher’s trials “and no hull rupture was suffered,” yard workers and crewmembers kept unearthing more damage throughout the post-shakedown yard period—even in its “late stages” just before the boat went down.

Nor was finding additional damage some bizarre departure from past experience. In fact, the board members noted that “this pattern of continuing discovery of shock damage during post shakedown [yard maintenance] parallels that found” in previous submarine classes after shock testing. All the more reason for stringent standards when repairing subs exposed to detonations during simulated or actual battle. They warrant special vigilance.

Fourth, damage control at sea is tough on any oceangoing vessel. It’s tougher on board shallow-diving diesel submarines, and tougher still on deep-diving nuclear-powered subs. The deeper a sub ventures, the fiercer the water pressure when flooding occurs, the faster the floodwaters rush in, and the harder it is to stanch the flow and patch a rupture. The nature of the operating environment makes prompt action imperative to a degree unknown in surface vessels, which generally float except after extreme damage. Time is a critical factor. As the Court of Inquiry maintained, “the increasing operating depths of submarines” have “compressed the time available in which to take effective damage control action with respect to flooding. The shortness of time available to control flooding is not well recognized.”