Let’s go back for a moment to the actual world, and leave aside possible worlds. When I say (1) below, it is false.
(1) I am president of the United States of America.
But when Barack Obama says it, right now, it’s true. In a short time, though, when he says (1), it won’t be true. It’ll be true when someone else says it. In order to evaluate the truth value of (1) we have to know two things. We have to know, first, who is saying (1); and we have to know when they’re saying it.
If I say it at any time, it isn’t true. If Obama says it right now, it is. If Reagan says it in 1983, it’s true.
We say that the word “I” is a deictic. It’s a word whose semantic value points (deixis in Greek means “pointing”) to something in the actual (or a possible) world. Without knowing what it points to, you can’t evaluate its meaning. Other deictics include “here,” “there,” “now,” “then,” “tomorrow,” “he, she, it”, and so on.
Let’s look at another statement:
(2) I have a new car.
(2) is true just in case the speaker, the person pointed to by “I”, has a new car. Let’s imagine that I is me, and I don’t have a new car. It’s false, then. But let’s recall configuration space from my last post. How far away are those points in the possible worlds in which I have a new car? Not very far away, really. All I have to do is move the indexical world, the “actual” world, to one of those worlds in which (2) is true. I’m not even changing the world: I’m just moving the index.
How will this look from the outside? Well, the index in configuration space doesn’t move in jumps and starts, so a new car won’t appear in my driveway. I’ll probably get a perfectly plausible windfall. I’ll find a good sale. I’ll get a high trade-in value. All quite possible, even coincidental-seeming, things, and I’ll end up in a possible world in which I have a car.
This is, of course, exactly how we know magic to work: by seeming coincidence. In fact, if we trace back the chain of cause and effect, we often find the causes preceding our act of doing magic, as if the universe is weaving together a consistent story. Some have tried to explain this weirdness of magic by calling it “retroactive enchantment.” But if we think of magic as just moving the index across the configuration space of possible worlds, it makes absolute sense that such a thing would be. We could not imagine magic working any other way than that, if that’s what it is.
We also know that magic in order to make me president is unlikely to work. Sometimes people evoke probability, but there’s a problem with that: in classical probability, statistical certainty is merely the measurement of our ignorance. To say there’s a 10% chance of something is only to say that in the past we have observed that one out of ten times have come to this outcome. It says nothing about the situation itself, and if we knew more about it, we could get a more accurate number. It’s a very deterministic view of the world. If instead of invoking probability, we think in terms of distance in configuration space, we can see that some “improbable” things are actually fairly close to us in configuration space. For example, there’s about a nearly zero chance that I’ll ever get married to a woman, but it’s only changing a few numbers describing my position in configuration space. I know that if I did magic to do that, I’d probably succeed (gods forbid).
We can also see how the focus of magic changes when we think of it this way. Winning the lottery is a magical goal that some people have. But if you think in terms of configuration space, winning the lottery isn’t really a number that defines an index. The amount of money you have is. So doing magic to increase the amount of money you have will move you to a closer point in configuration space than doing magic to win the lottery.
So that raises the question — how do we move our index? In other words, if magic is the selection of a possible world in configuration space, a world that already exists in the abstract sense, how do we get there?