-> Philosophy -> The meaning of "choice"

Posted on 2019-08-05.

Ted Chiang is probably best known for writing the short story that the excellent movie Arrival is based on. He has a new collection out called Exhalation and it’s really, really good. But one of the stories has a discussion of free will that reminds me of several frustrating conversations I’ve had on the subject in the past.

In the story (called “What’s Expected of Us”), there’s a device consisting of a light and a button. Pushing the button causes the light to flash in the past, a second before you pushed the button:

Most people say that when they first try it, it feels like they’re playing a strange game, one where the goal is to press the button after seeing the flash, and it’s easy to play. But when you try to break the rules, you find that you can’t. If you try to press the button without having seen a flash, the flash immediately appears, and no matter how fast you move, you never push the button until a second has elapsed. If you wait for the flash, intending to keep from pressing the button afterward, the flash never appears. No matter what you do, the light always precedes the button press.1

To the people in the story, this proves irrefutably that the universe is deterministic and that we do not have free will. Many of them conclude that, therefore, nothing they do matters. They stop trying to accomplish anything or even bothering to feed themselves. It’s all predetermined, so what’s the point?

In the real world people who believe in free will do often seem to think that such despair would be the rational response if free will turned out to be an illusion. I usually hear these two arguments:

  1. It’s not even possible to live without believing in free will. We all unavoidably think about and talk about our lives as if we have “choices”, but it’s nonsensical to talk about “choices” and “choosing” if there’s actually only one way everything can go.
  2. If there’s no free will, nothing matters. You might as well stop trying to do anything, because the outcome is already set in stone.

These are terrible arguments. First, the word “choice” is fully intelligible and useful even when it’s entirely predictable which choice will be made. Imagine you’re building a robot, and you need to program how it will respond when someone attacks it. The robot could behave in various ways: it could run away, it could stand still and take the abuse, it could organize a robot uprising and kill all humans, etc. The robot’s programming will deterministically select one of these options, so in a sense the others aren’t actually possible. But they are actually possible from the program’s perspective. If it issues the command to run away, the robot will run. If it issues the command to strike back, the robot will attack. In this sense, the program has choices. The choice it makes will become part of the chain of events that determines the future. Determinism simply says that the choice isn’t the beginning of the chain. The program chooses what the robot will do, but the program follows a fully predictable process for making the choice.

If a bystander, wondering whether they’re about to be the victim of a vengeful machine, asks you “how does it decide whether to kill us?”, it would be unreasonable for you to respond “it’s deterministic, so no decision is being made.” Their question is meaningful, and you could meaningfully answer it by explaining the decision-making algorithm you programmed. (“Don’t worry, it’s programmed to be a utilitarian; it won’t murder you unless it can confirm that you’re an organ donor or chronically miserable.“)

Second, it’s mistaken to say “it doesn’t matter what I do if there’s no free will.” For example: if you stop eating, you will suffer, starve, and die. If you go to the Cheesecake Factory instead, you will experience the delicious taste of cheesecake and continue living. Those are significantly different outcomes. Sure, if determinism is true, then it’s already inevitable which one will occur. But that’s not because you can’t choose; it’s because there’s a (theoretically) predictable process by which you make choices. The decision-making process is a part of you, a part of your brain; from your perspective, there genuinely is a choice, since your decision is a link in the chain of events that determines the outcome, and different decisions would result in different outcomes.

It can get confusing when one of the inputs to your decision-making process is awareness of the fact “my decision-making process is deterministic.” And indeed, it’s possible your brain is programmed in such a way that, once it knows this fact, it chooses starvation over cheesecake. But if so, it’s defective and wildly irrational. A rational program would make the choice that leads to the best outcome, and correctly recognize that the fact “I am deterministic” is completely irrelevant to whether starvation is a good idea. (Indeed, that fact - assuming it does turn out to be a fact - is completely irrelevant to most of the decisions we’re faced with.2)

The diagram below shows two deterministic worlds. In the first, your genetics/upbringing/etc cause you to believe free will is crucial, so you choose starvation; in the second, your genetics/upbringing/etc cause you to believe free will is not so important, so you choose otherwise. In both worlds, the choice you’ll make is determined from the start. But surely the brain in the first world is the one that’s irrational: although it was destined to make a poor choice from the outset, it’s still a poor choice, and that irrational poor choice is precisely what causes the bad outcome.

Silly illustration

  1. p. 58
  2. I do think it’s relevant to how we view moral responsibility, though; it’s hard to see how purely retributive punishment makes any sense if a person’s crimes were inevitable. So, I’m not arguing for compatibilism.