Terrorists, Microchips, and other Half-Baked Frankfurters
Frankfurt cases are classic thought experiments used to explore the necessary conditions of a free choice. Allegedly, they demonstrate that the principle of alternative possibility (PAP) is not a necessary condition. In other words, even if something is all you can do, you can still do it freely. Here are a few common examples to get the gist.
- Scenario A: Suppose you are running out of a burning building and you reach the ground floor. Ten feet to your left is a door labeled Exit A and ten feet to your right is a door labeled Exit B. You choose Exit A and escape to safety. Unbeknownst to you, Exit B was actually blocked on the outside by a giant beam and could not have been opened.
- Scenario B: Suppose there is an evil scientist who has been hired by the Donald Trump campaign in the battleground state of Colorado. This evil scientist is systematically installing microchips into people’s brains under the guise that they are Pokémon Go updates. In reality, they are chips to ensure that the individual votes for Donald Trump. If the person attempts to vote for Gary Johnson, the chip activates and changes their mind to Trump. Come Election Day, some voter, Ash, has a chip in his brain. He’s in the booth and decides that he wants to Make America Great Again™ and freely votes Trump. However, he could not have voted for anyone else but Trump.
- Scenario C: Every evening from 6pm – 7pm, I make a cuppa joe and read from a book from my personal library. Suppose at 6:15pm, a terrorist runs into my house, points a gun at my head and says “Read that book for the next 45mins or I shoot!” I look at him and tell him it won’t be a problem because that what I wanted to do anyways. I freely read my book even though I can’t do otherwise. (This example is adapted from the great Protestant pope Ronald Nash).
I have two problems with the Frankfurt cases. First, I don’t find them convincing and second, they don’t bear the argumentative weight that some people think.
As stated, the purpose of these thought experiments is to show that the PAP is not necessary for a choice to be free. However, the strength of the argument is completely dependent on how broadly or finely one defines a possibility. Here, I appeal to Sarte’s notion of “radical freedom” in that every choice is a free choice and every situation is a freedom permitting circumstance. The standard example is of a group of hikers who going up a mountain encounter a giant boulder blocking their path. “We have no choice but to turn back”, says the leader of the group. “False,” retorts Sarte “for you have the choice to jump off of the mountain to your near-certain death”. Sarte is correct in that there are alternative possibilities in this case and in all of the above Frankfurt examples. You can choose to die in the burning building, get shot by the terrorist, or kick over the voting machine and urinate on it. There is nothing about the thought experiments that precludes any of these behaviors, thus, the hypothetical scenarios don’t truly demonstrate a lack of alternative possibilities.
The above retort is only true in the finely grained sense of “alternative possibilities”. On a more broad understanding, this isn’t the case. (By “broad”, I’m referring to the way we typically use “I had no choice” e.g. “I had no choice but to swerve off the road – I was going to run over a child”. Sure, you could run over the kid, but c’mon). In the broad sense, these scenarios might eliminate alternatives because (i) given that I want to escape the burning building, I must choose Exit A, (ii) given that Ash is going to vote, he must choose Trump and (iii) given that I don’t want to die, I must stay in my chair. Even still, I don’t think that Scenarios A and B truly capture a lack of alternative possibilities in the broad sense. Surely there is a difference between picking Exit A first and choosing Exit A after learning Exit B is locked. Equally, there is a difference between voting for Trump because the chip detected a Johnson vote and voting for Trump first. The terrorist example comes closest to truly creating a no-alternatives scenario, but, also shows the limitation of Frankfurt-style arguments.
It creates a problem for divine determinists (e.g. Calvinists) because it is by mere happenstance that the choice and the external limits coincide. My decision-making process is entirely independent of the terrorist’s, yet, they both line up by luck. The divine determinist wants to say that God sovereignly determined all events, not that His decree happens to line up with what His creatures have already decided. Notice how the terrorist has no causal relevance in the scenario. If he were or were not there, I would still read my book. To say that this scenario is some kind of parallel to the divine decree is to make the decree causally irrelevant, which is the exact opposite of what the divine determinist is wanting.
To use another illustration, suppose there is a grid of numbered squares in a large field. Each odd-numbered square has a land mine underneath but each even-numbered square is clean. Suppose Jane, who knows nothing of the land mines, is walking through the field and on a whim decides to cross the field by walking only on even numbers. The person who placed the mines didn’t determine Jane’s path – he only determined that < if Jane steps on an odd numbered square, she will die >. The fact that Jane made it to the other side seems to be by luck.
What happens if a creature tries to go against the divine decree? Well, there are two options. Either (a) she can but won’t because God’s decree is compatible with her choices by luck or (b) she can’t because the divine decree didn’t just remove alternative possibilities but actually determined the specific choices she is going to make. The Frankfurt cases do not independently support such a strong position as (b). At most, this argument only shows that the PAP is not a necessary condition for free will which is not the same as demonstrating that a free decision is compatible with causal determinism. Don’t get me wrong – should Frankfurt-style arguments go through, the conclusion is non-trivial; however, the conclusion is often overstated. Compatibilists will need additional argumentation for their position.