Home » philosophy » Free will revisited

Free will revisited

In a previous post I set out my position on the topic of free will. It’s a view called compatibilism, which is commonplace in philosophical circles and mostly unknown everywhere else. Compatibilism remains my position, but nowadays I find it more useful to come at the topic from a different direction. In this post I shall write less about the view I hold, and more about some mistakes of reasoning that can lead to other views.

These mistakes have a peculiar property: they are most effective at tripping up smart people, and all the more so when smart people put quality time and effort into falling for them. They do not, as a rule, confuse the person on the street. So, before I get us all confused, let me plant my feet firmly on the street.

Ordinary modern opinion

Most of us, or at least most of us in the modern world, have no trouble holding the following two ideas at once:

A. Common sense about decision-making: We accept that people make decisions, and in many cases a person’s decision to do X causes the person to do X. Moreover, these decisions can be influenced by the prospect of (and a history of) rewards and punishments and praise and blame, including moral praise and blame. Furthermore, it is often true that the person who did X had the power to do otherwise, which is to say: if she had decided to do Y instead of X, then she would have done Y instead of X.

B. Open-ended acceptance of scientific explanations: We trust the scientists to figure out how decision-making really works inside our brains, right down into the cells of our brains and the atoms of the cells, and so on, as far as the science can go. Maybe one day the white coats will find some level of description where everything obeys strictly predictable laws, like clockwork; or maybe they will reach consensus that there are fundamental uncertainties due to quantum mechanics. Whichever way it goes, we’ll take their word for it. Right now, we don’t have to commit to specifics.

A and B are the views of ordinary people, and I’m on side with both of them. But it’s sometimes said that you can’t have both at once. Why?

Mistaken objection #1: puppet strings and constraints

Some people look at the science and say “Aha! Here is where the causes of our behaviour lie: in the chemistry, or in the cell biology, or in quantum randomness, or what-have-you. These things cause our behaviour, therefore our decisions don’t. So our commonsense view of decision-making is all an illusion.”

I’m perfectly happy to accept the stated premise, that the causes of our behaviour are all physical, chemical, etc. But it’s a mistake to jump from there to “Our decisions don’t cause our behaviour”. It’s a mistake because it assumes that our decisions are one thing and our brain chemistry etc. is something else. That is a very old-fashioned view, and if it was spelled out as a premise then most of us would reject it. The trouble is that it wasn’t spelled out; it snuck its way into the argument under cover of darkness. We need to be on guard against it.

One sure way to tell that someone is making this mistake is that they use the analogy of puppets and strings. In this image, the strings (representing the physical goings-on in our brains) are one thing, and the puppet (representing ourselves) is something separate. If you take away the separation, as you should, then the analogy collapses.

Occasionally you will see analogies to handcuffs or other constraints. The idea is similar: our physical/chemical/neural life is a force that opposes our will, like handcuffs opposing muscles. Just recall that our will is physical/chemical/neural, and straightaway the analogy of constraint collapses.

Mistaken objection #2: “She couldn’t have done otherwise”

If we accept science in an open-ended way, then one of the things we must allow for is the universe turning out to be deterministic clockwork. Whether it’s at the level of brain cells, or much deeper down in some “hidden variable” version of quantum mechanics, it might turn out that the complete state of the world up to midnight last Tuesday “locked in” what happened at one minute past midnight, and what happened at two minutes past midnight, and so on for the rest of time.

Now, if that is so, where does it leave someone who decided, at a minute past midnight, to raid the refrigerator, and at two minutes past midnight raided it? According to this clockwork kind of science, the raid was locked in at midnight, so she couldn’t have done otherwise. Yet common sense says she could have done otherwise. And so, the argument goes, we can’t have the science and have the common sense. Something’s got to give.

I say that this argument is mistaken too, but the mistake is interestingly subtle. It is what logicians call a fallacy of equivocation. That means that the argument relies on an ambiguity. It uses an ambiguous expression in a way that makes the argument appear sound, but if you take care to avoid the ambiguity, then the appearance of soundness vanishes.

The ambiguous phrase here is “could have done otherwise”. When common sense says that the late-night fridge raider could have done otherwise, it actually means that she had the power to do otherwise, which in turn means:

If she had decided to do otherwise, then she would have done otherwise.

This type of if-then statement is what philosophers call a forward-looking counterfactual, and it can be true even if the laws of physics guarantee that, at midnight, the decision and the raid were both “locked in”.

Do you find this surprising? Then I’d ask you to consider another forward-looking counterfactual. This one is about the asteroid that killed the dinosaurs 66 million years ago:

If Jupiter had been in a different position, then the asteroid would have missed the Earth.

This can be true even if the laws of physics had both the position of Jupiter and the asteroid’s impact locked in from 67 million years ago. The forward-looking counterfactual does not claim that Jupiter could (in any sense of “could”) have been in a different position from its actual one. It only makes a claim about how things would have played out if it had been in a different position.

So, too, with “If she had decided to do otherwise, then she would have done otherwise”. It can be true even if the laws of physics had both the decision to raid and the raid itself locked in at midnight. It does not claim that she could (in any sense of “could”) have decided to do otherwise. It only makes a claim about how things would have played out if she had decided to do otherwise.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s