Rubik's Earth

The current boundaries of tectonic plates are completely unsatisfactory. On the one hand, the plates will insist on drifting willy-nilly. On the other hand, because they are irregular in shape, any movement of one relative to the other leads to lots of grinding, volcaning, subducting and general orogeny. The last thing you want in your neighbourhood is a bloody great mountain range popping up.

So. There are two things we need to do, probably simultaneously. First, we need to take a sharp implement and slice the crust up into neat segments. If you've ever seen the spherical versions of a Rubik's cube, that's the effect we're after - nice, regular tectonic plates.

Of course, things will still be messed up because the old plate boundaries would still exist, running through and across our newly-cut plates. This is where the second step comes in. We need to glue together the edges of the old, messy plate boundaries. The ways to do this are too obvious and simple to need describing here - suffice to say that a system of transverse alps would be an elementary solution.

In a short space of time, the old plate boundaries - stitched firmly together - would weld themselves shut permanently. The convective currents in the Earth's mantle will still be trying to shuffle the crust around, but this movement can now be accommodated without fuss or overt geology, since the new plate arrangement will allow smooth sliding.

Obviously, it would be silly to stop here. Convective flow (as of, say, a mantle) is quite susceptible to disruption, and hence to control. It wouldn't take much to harness the convective forces of the mantle, and therefore steer the newly- geometrized plates around.

Controlled Convective Tectonic Navigation would be a slow process. However, if you do the calculations it makes a lot of sense. Airplanes carry minute loads very quickly, and are economical. Trains carry larger loads fairly quickly, and are economical. Supertankers carry very large loads slowly, and are economical. A system for moving entire continents around could afford to be very slow indeed.

Traffic control for CCTN would need to be well thought out. For example, if you want to move northeast Africa (which will occupy a substantial part of one of the new plates) to the opposite side of the equator, you'd have to go through a fairly complex series of moves (northern hemisphere left quarter turn; eastern hemisphere half-turn anticlockwise; both polar segments half-turn east; northern hemisphere right quarter turn- the Greenwich Meridian would, of course, be fixed by convention). It's even more complicated if you want to avoid disturbing the other tectonic segments. However, the great thing about CCTN is that you need to think carefully, not fast.

Mathematically inclined geologists might even compete to see, for example, who could rearrange the continents into a reconstruction of Pangea in the fewest moves; or who could get Norway inverted without disturbing the other land masses.