The Two-lever Argument Against Free Will


There are many definitions of free will, so it’s best if we start with one before proceeding:

As you may have heard from my Absolute vs. Practical Free Will segment of this series, I describe this as the most important type of free will because it’s the only kind that can amply power Moral Responsibility.

Basically, if one could not have willfully chosen to do otherwise for any given previous decision then that decision cannot be said to be free. And because we live in a deterministic universe, we can not have chosen otherwise.

Quantum randomness leading to an alternative outcome does not count because humans don’t willfully influence quantum randomness.

The Two-lever Argument posits that there exist only two levers for humans gaining the ability to do otherwise during a decision, and that one must be able to control at least one of these in order to have true choice:

  1. The previous state of the universeHow the universe was configured at the moment prior to you making a decision.

  2. The laws that govern the universeThe physical rules that will determine how the universe transitions from one state to another, namely from the previous-state to the next-state.

If you do not have some measure of control over at least one of these two variables, you simply cannot control any future state of the universe. And if you are unable to control any future state of the universe, then—regardless of how it may feel to us—we are incapable of making a true, free decision. Instead, events are moving through you, and you are being given the perception that you made a choice.

Neither quantum randomness nor consciousness provide an escape from this. Randomness simply removes predictability from the universe; it does not provide humans any additional control of outcomes. Similarly, consciousness—since it does not offer the ability to control the previous state of the universe or its laws—offers no escape either, despite strong instinctual feelings to the contrary.

Here’s the argument in deductive form:

  1. Future states of our universe are defined by a combination of two things: 1) the previous state of the universe, and 2) the laws of physics that govern how one state moves to the next.

  2. Humans have not always existed and are a product of said universe.

  3. For any given instant in time, humans lack the ability to control the previous state of the universe.

  4. Humans lack the ability to control the laws of physics.

  5. Therefore, humans are unable to control any future state of the universe.

  6. And because humans cannot control any future state of the universe, humans cannot have free will.

Again, there are two levers for control: the universe’s state before you make a decision, and the laws that govern transitions to the next state. Demonstrate that you can pull one or the other and you have demonstrated free will.

Failing to do so keeps you within the natural outcome cycle of pre-existing state + physical law (including randomness) resulting in outcomes. This means that while humans are part of this mechanism, we are unable to control it, and thus we do not, and can not, have the type of free will that most people think we have or that is required for moral responsibility.


  1. Here’s why this matters.

  2. Here’s an illustration of the two lever argument in code: 25 Lines of Ruby That Illustrate Free Will is Impossible.

  3. Here is my explanation for why we evolved to have the sensation of free will.

  4. My two-lever argument is similar to Galen Strawson’s “Basic Argument”, but I believe his “because of what we are” needs to be unpacked more before it can be convincing to most.

  5. There’s actually a third lever: The supernatural. E.g. “God gave us free will”. The reason it wasn’t included should be obvious.

  6. I am not arguing for nihilism or anything like it. Quite the opposite in fact. See my other writing on the topic.

  7. States of the universe in this context mean the current, precise configuration of the molecules in your body and its surroundings at the point of decision, including location, speed, spin, etc.

  8. In item #5 of the deductive argument, I have added “willfully” to the statement because there seems to be some confusion about humans being able to “affect” things. It should be clear that rocks affect how rivers flow, but that’s not the standard we’re looking for here.

Related posts: