Previous: losing trackNext: emotionContents

Society of Mind

15.11 the recursion principle

Let's consider one last time how a mind could juggle nonexistent furniture inside an imaginary room. To compare different arrangements, we must somehow maintain at least two different descriptions in the mind at once. Can we store them in different agencies, both active at the same time? That would mean splitting our space-arranging-agency into two different smaller portions, each working on one of those descriptions. On the surface, there's nothing clearly wrong with that. However, if those smaller agencies became involved with similar jobs, then they, in turn, would also have to split in two. And then we'd have to do each of those jobs with but one-quarter of a mind. If we had to divide agencies into smaller and smaller fragments, each job would end up with no mind at all!

At first this might seem to be an unusual situation. But it really is very common. The best way to solve a hard problem is to break it into several simpler ones, and break those into even simpler ones. Then we face the same issue of mental fragmentation. Happily, there is another way. We can work on the various parts of a problem in serial order, one after another, using the same agency over and over again. Of course, that takes more time. But it has one absolutely fundamental advantage: each agency can apply its full power to every subproblem!

The Recursion Principle: When a problem splits into smaller parts, then unless one can apply the mind's full power to each subjob, one's intellect will get dispersed and leave less cleverness for each new task.

Indeed, our minds don't usually shatter into helpless fragments when problems split into parts. We can imagine how to pack a jewelry box without forgetting where it will fit into a suitcase. This suggests that we can apply our full space-arranging resources to each problem in turn. But how, then, do we get back to the first problem, after we've thought about the other ones, without having to start all over again? To common sense the answer seems clear: we simply remember where we were. But this means that we must have some way to store,

and later re-create, the states of interrupted agencies. Behind the scenes, we need machinery to keep track of all the incomplete accomplishments, to remember what was learned along the way, to compare different results, and to measure progress in order to decide what to do next. All this must go on in accord with larger, sometimes changing plans.

The need to recall our recent states is why our short-term memories are short-term memories! In order to do their complex jobs so quickly and effectively, each micromemory-device must be a substantial system of machinery, with many intricate and specialized connections. If so, our brains cannot afford to make too many duplicate copies of that machinery, so we must reuse what we have for different jobs. Each time we reuse a micromemory-device, the information stored inside must be erased — or moved to another, less costly place. But that would also take some time and interrupt the flow of thought. Our short-term memories must work too fast to have any time for consciousness.