I gripe a lot about the sorry state of American defense procurement, and sometimes about the even sorrier state of Western European defense procurement. But there have been successes. In thinking about a few of the recent ones, namely the Virginia-class nuclear-powered attack submarine, the M1 Abrams tank, and the F/A-18E/F Super Hornet, I noticed something: all had followed a gross failure.

First let’s define terms. The obvious: failure. A program is a failure when it is cancelled while the need remains. If an army decided it needed no more tanks and cancelled its latest tank design program, that would make sense. At least from a logical standpoint. It follows. If you don’t need a thing, you shouldn’t be buying a thing. But sometimes a program is such a massive overbudget clusterfuck of mismanagement, it gets cancelled even though the service still needs it. And that usually forces some ranking officers to be “forced into retirement” and a lot of soul searching. We’ll see that this is important later.

Next, let’s talk success. I’m going to be mean and set the bar high. A successful program delivers a quality product at a reasonable price. On time. On budget. But it must also be a product that compares well to its peers, regardless of how much gold-plated nonsense is baked into them. Each of our aforementioned projects fulfills both criteria. They are seen as good by both the bean-counters and the warriors. All designs are compromises, and these appear to have made the right ones. Why?

Almost certainly, because the need was great, and the service in question had already tried an “everything and the kitchen sink” gold plated design that failed. Each predecessor was super expensive. Two of the three were cancelled outright. The third just barely made it out of the gate before being terminated unceremoniously. Let’s look at these failed programs.

The Abrams was preceded by the MBT-70, a case study in multinational mismanagement. It was a joint German-American tank project, but the Germans and Americans couldn’t agree on anything. Rather than actually make hard decisions, the project team let each country do its own thing. Since work was duplicated and the project had to work with both, costs skyrocketed. The Germans wanted a 120mm smoothbore gun. The Americans wanted a 152mm short-barrel gun/missile system. So they compromised. Both were developed and integrated. The Americans wanted a gas turbine. The Germans wanted a diesel. So they compromised. American versions had a gas turbine; German versions had a diesel. The design teams couldn’t even agree on whether to use metric or SAE measurements on bolts and nuts. You guessed it, both were used. Plus, they wanted to integrate an autoloader, which had never been done in the West. They also wanted an active hydropneumatic suspension that could “lean” and “kneel”, another novelty. Costs spiraled out of control, and eventually, Congress and the Bundestag agreed on something: the MBT-70 had to go.

The Super Hornet had an ill-fated predecessor in the A-12 Avenger II. The Navy wanted to replace the A-6 Intruder and A-7 Corsair II attack aircraft with a cool new stealth attack aircraft. Stealth was cool. The USAF had the awesome F-117A and B-2A. Stealth meant you could go anywhere, and the pesky Soviet air defense systems could do nothing to stop you. But stealth was expensive. Very expensive. And the A-12 program was probably the worst-managed aircraft program in history. Composites were new, and screwups led to the plane coming in overweight, and the weight growth never stopped. The multifunction radar had development problems as well and started to rapidly consume the navy’s budget. Delays in the prototype design pushed back early flights, and added to the cost. Then Secretary of Defense Dick Cheney cancelled the program in 1990.

Finally we come to the Seawolf class, the predecessor of the Virginias. At first, you might object. Seawolfs were commissioned! And yes, they were. But only three of them ever put to sea. THREE. They were supposed to replace the Los Angeles class attack submarines. But how can they do this when there are more than twenty times as many of the Los Angeles class boats? Yes, it failed. Get over it. Loaded with everything from a fancy new sonar with battle management system and newer hull construction techniques, and even new steels, it came in overbudget and at the wrong time. Even though they’re really great boats, there’s only so much you can do with three hulls instead of sixty two.

After failure, each service went back to the drawing board. They thought long and hard about compromising to get the price down. What did they really need now, what could they add later, and what could they do without. They relearned that the perfect is the enemy of the good, and a piece of hardware that you have is infinitely better than a cancelled project. It’s a pity this lesson is so very hard to remember. The results are excellent vehicles that have received a large number of upgrades.