In a war zone, you have your vanguard. Then you have your tanks, your main body of troops, and your artillery. If all that fails, and you are being overrun, there is your rear guard. If they fail, and you cannot retreat, all is lost.

For the last 11 months, that has essentially been BP's explanation of what went wrong at the Macondo well in the Gulf of Mexico, where a mighty explosion killed 11 men and spilled five millions of barrels of oil into the water over a three-month period before the company managed to seal it in with concrete. But all along there has been the question -- what about that rear guard, in this case a much-trumpeted piece of technology known as the blowout preventer?

Whenever worriers have trotted out their concerns regarding offshore oil spills, the industry has in return trotted out trusty photographs of this gigantic, 50-foot-tall contraption that -- bolted to the top of the well, right at the sea bottom -- would save the day in any crisis. If there were a blowout, doubters would be assured, the preventer would act like gigantic shears, slicing through the well piping, and sealing in all the hydrocarbons below the surface.

In a new report prepared for the U.S. Interior Department, though, we get good news and bad news. The good news is that the blowout preventer does what is suggested in the photos and accompanying charts. The bad news, according to the 551-page report by Det Norske Veritas, a Norwegian risk management company, is that it only does so in photographs and charts, and not in real-life crises such as the Macondo blowout. (Here is volume 1 of the report. Here is volume 2).

In the case of Macondo, the intense pressure of the blowout bent the well pipe, and pushed it out of position. And when the mighty sheers were activated, they could not close entirely. It was through that "not entirely" -- a 1.4-inch diameter opening, according to the Wall Street Journal -- that much of the oil flowed. Read on for more details.

This modest report suggests that the industry consider the impact of a discombobulated pipe on "the ability of the blowout preventer components to complete their intended design or function." It proposes that "the findings of these studies should be considered and addressed in the design of future blowout preventers and the need for modifying current blowout preventers."(The blowout preventer image above is from the Christian Science Monitor)

Over at the Journal, Ben Casselman and Russell Gold get the message synthesized by Elmer Danenberger, who in 2009 retired as the Interior Department's overseer of U.S. offshore drilling rules: "They have to rethink the whole design."

In fact, this conclusion is not a surprise. As Casselman and Gold note, the oil industry has known for many years that the blowout preventers work in only a fraction of accidents, and that they have been

prone to failure, especially as drilling has moved into deeper water, requiring thicker, tougher pipe. In 2004, a study commissioned by federal regulators found that only three of 14 newly built rigs had blowout preventers that could squeeze off and cut the pipe at the water pressure likely to be experienced at the equipment's maximum water depth.

I contacted Cameron International, the maker of the blowout preventer on the Macondo well, which referred me to a statement it had issued. "The [blowout preventer] was designed and tested to industry standards and customer specifications. We continue to work with the industry to ensure safe operations," Cameron said. Here is congressional testimony in the same vein last May by Cameron CEO Jack Moore.

That's great to work to industry standards, but does the blowout preventer actually work in a crisis, I asked. "Cameron declines to comment further," a spokeswoman said.

At the Financial Times, Sheila McNulty says the industry will not welcome a re-examination of "these supposed machines of last resort." But, she goes on:

the bottom line is that if the blowout preventer can be compromised during an accident, then it cannot be guaranteed to serve its purpose of preventing a blowout of the kind that spilled oil into the Gulf of Mexico for three months last year. The Interior Department had been moving -- ever so slowly -- toward letting the industry back in the Gulf. It is likely that this latest report is going to slow -- if not reverse -- that effort.