Back at the beginning of the Cold War, with the threat of nuclear holocaust overhanging the nation, and the world, physicist Herman Kahn wrote a paper about "thinking the unthinkable."

It was about nuclear war, and it ultimately resulted in the theory of "Mutually Assured Destruction," with the appropriate acronym, MAD. No matter how horrible the consequences of some potential policy outcomes, they must be thought about, to minimize the probability of them occurring, or at least, of the damage resulting from their occurrence.

For the past month, NASA has been picking through the few recovered shattered shards of its first reusable spaceship, attempting to put together a tragic jigsaw puzzle, with misshapen pieces burned and warped, and most of them missing, in an attempt to try to find out definitively why it has to do so. But even before a conclusion has been reached, the agency has come under fire for alleged poor judgement during the flight, and for withholding information from the public in an attempt to protect senior management.

These accusations may turn out to be accurate. But, to me, they're premature. And, for now, they're misguided.

I say this not because I'm unable to believe NASA would do such a thing, or because I'm a reflexive defender of the space agency. Anyone who's been reading my columns in this space for the past year knows that, if anything, I'm the last person to defend NASA, an entity that I believe has delayed and held us back from our ultimate destiny in space.

In fact, I find it ironic and frustrating to be in the position of defending the Shuttle program, and the agency itself. I do so not because I am happy with it, or even because I would like to see it continue to exist in its current form.

My concern is that, in a witch hunt to find out "what did they know, and when did they know it," we'll lose sight of the real issues, the fundamental issues of space policy that led to this disaster. In an era in which many of our government institutions have shown themselves to have feet of clay, it's easy to point fingers at one that has lost two orbiters with 14 people, flown a supposedly orbiting vehicle into the Martian surface, and launched an ostensibly far-seeing telescope that turned out to need glasses, like some kind of multi-billion-dollar space geek. Moreover, it's one that doesn't seem to provide a lot of value for the money invested in it, at least as far as the ostensible purpose -- progress in getting humans into space.

And of course, it's always fun to play Monday-morning quarterback. When the apparent mode of destruction corresponds very closely with some engineers' pre-entry predictions, it's easy, and even gratifying, to cry "cover up"!

But in doing so, we fall into the trap of criticizing it for the wrong reasons, and once again being distracted from the real problem, which is that we don't, as a nation, really know what we're trying to accomplish in space. Until we resolve that issue, putting NASA into the stocks and throwing tomatoes at it is not only pointless, but counterproductive.

The people who work at NASA are good people, for the most part. But put yourself in their place for a moment.

Columbia is in orbit, for better or worse. There may have been damage to the tiles on the wing leading edge. The damage may or may not be critical, even catastrophic. If so, due to decisions made decades earlier, decisions made by people, most of whom are retired or dead, there's little that can be done about it. The thermal protection system has been accepted for years as a critical subsystem, just like a wing on an aircraft, and if it's lost, the vehicle is lost. The only solution to that problem is to prevent it from being lost.

The notion that it has been fatally compromised, with no realistic solution, may be discussed as a "what if" scenario in an email, but it's not something that can be easily contemplated as a reality.

To do so is to contemplate the loss of a quarter of the nation's fleet of orbiters, and to consign seven brave and exemplary men and women to an almost instantaneous incineration. To do so is to "think the unthinkable."

War gamers, military planners, RAND Corporation analysts, are used to doing so. NASA engineers are not.

Many will complain that this is a result of budget shortfall. To the degree that that's true, it has little to do with recent NASA budgets.

It's a result of a budget shortfall a quarter of a century ago, when we decided that we would have a single vehicle provide access for humans to orbit, when we decided that that vehicle must not only deliver humans to orbit, but sixty-thousand-pound payloads, and have a thousand miles of cross-range capability on landing, and act as a space station, rather than simply a delivery truck.

Most importantly, it's when we decided that we would only provide half of the money that engineers said would be required to develop such a vehicle, so compromises were made, that resulted in multi-segment solids that destroyed the Challenger in 1986. It also resulted in a fragile thermal protection system that looks increasingly like it doomed Columbia a little over a month ago, in a way that was beyond the means of frustrated engineers at NASA--and the United Space Alliance, and Boeing--to do anything about, to the point that it was painful to even think about.

Let me propose some (apparently) unthinkable thinking, that might get us out of the rut that we've been digging since the end of Apollo.

Let's think about multiple vehicle types for people to orbit, all American. Let's think about vehicles that can deliver anyone who wants to go, and has the money to do so, instead of government employees. Let's think about generating huge markets for space activities, instead of constraining ourselves to paltry notions of three, or six civil servants permanently in orbit at once. Let's think about building infrastructure, on the ground and in orbit, that will provide "tow-trucks," and hangars, and maintenance capability for vehicles in trouble, just as we do in every venue on earth.

Let's think the truly unthinkable--making space not a program, but just a place.

Rand Simberg is a recovering aerospace engineer and a consultant in space commercialization, space tourism and Internet security. He offers occasionally biting commentary about infinity and beyond at his Web log, Transterrestrial Musings.

Respond to the Writer