Normally what I write in this column may sound like advanced science, but it stays fairly close to the here and now. This time I want to wander further afield, to a piece of technology that could be a century or more away.

A few weeks ago I pointed out some of the failings of the rockets used to launch payloads to space. In fact, there is a much bigger problem with all types of rockets, no matter their size or the source of their thrust. Even a fully reusable rocket (which we don't yet have) is a terribly wasteful way to send things to space from the surface of the earth, or to bring them back here.

To see why, think of an old-fashioned clock. Not the kind with a wind-up spring, but the kind where once a week you raise a weight that sits at the end of a cord. As the weight slowly descends, it provides the energy needed to drive the clockwork mechanism. The work that you put in, raising the weight against gravity, is recovered and used during the next week to keep the clock going. The same principle could be used, though it is not, in a modern elevator. Raising you to the top floor of a building calls for a certain amount of work, most of which could be recovered by driving a generator as you return to the ground floor.

Why not employ the same idea in sending things to space, and bringing them back? Most of the work needed to lift a payload during ascent could be recaptured during descent.

The problem, of course, is that we don't have elevators running all the way up to space. But one day we could have, and one day we surely will.

The idea of a "space elevator" is about as simple as an ordinary elevator. Imagine a cable that is anchored at a point on the equator and extends vertically upward. Naturally, the weight of the cable tries to pull it down; but if the cable turns with the earth, there will be a centrifugal force acting on each part of the cable pushing it upward. The longer the cable, the bigger the outward force. Make the cable long enough and you eventually get to a point where the total outward force on the cable from centrifugal forces exactly balances the total downward force due to gravity. Such a cable will hang stationary, tethered at a point on the equator, turning with the earth, and pointing straight up to space. We can treat it just like the cable of an elevator: attach cars, counterweights, a power system, and some safety precautions, and ride up or down the whole length. If a rocket is like a ferryboat to space, then this is like a bridge.

This idea has been around since at least 1960 (it has been independently "discovered" at least three times since). So why hasn't it been tried?

For two reasons. First, we don't have strong enough materials to stand the strain. However, stronger materials are being made all the time; at some point this will not be a showstopper. The second reason, and maybe a better one, is the enormous length of cable needed to balance gravity with centrifugal forces. When you do the calculation, you find the space elevator has to extend close to a hundred thousand miles upward. You can do better than this by attaching big ballast weight at the upper end, but there is no way you can make a space elevator less than about fifty thousand miles high.

A bridge to space, so huge that its load-bearing cable could stretch twice around the whole earth. Impossible? Preposterous? Foolish to think about, even for a hundred or two hundred or a thousand years from now?

Mm. Perhaps so; but I wonder what Sir Francis Drake's sailors would have said when they sailed by San Francisco Bay, if you told them that in less than three and a half centuries there would be a bridge spanning the Golden Gate - and people would cross it every day to go to work.

"Borderlands of Science" is syndicated by: