Libertarian free will, Frankfurt cases, and the ability to do otherwise

Anybody who believes in free will thinks the will is free from something or free to do something. Libertarians thinks the will is free from absolutely all antecedent conditions, including one's own psychological states. That means if you have free will in the libertarian sense, then for any free act, you could have done otherwise even if everything in the universe prior to and up to the moment of choice had been exactly the same, and that includes all of your psychological states, including your beliefs, desires, preferences, biases, motives, etc.

Since the ability to do otherwise is so wrapped up in the notion of libertarian free will, many have taken to defining free will as the ability to do otherwise. A philosopher named Harry Frankfurt came up with some counter-examples to show that the ability to do otherwise is not necessary for libertarian free will. These are thought experiments designed to show that one can have libertarian freedom even if they lack the ability to do otherwise.

For example, imagine you're sitting at a table with a can of Coke on one side and a can of Dr. Pepper on the other, and you are given the option to drink one or the other. Imagine that unbeknowst to you there's a guy hiding behind the curtain watching you closely, and if he sees you reach for the Coke, he's going to jump out from behind the curtain and slap the Coke away, preventing you from drinking it. That never happens, though, because you choose to drink the Dr. Pepper instead of the Coke.

Thought experiments like this are meant to show that one can make a libertarian free choice without having the ability to have done otherwise. Even though you couldn't have chosen to drink the Coke, your choice to drink the Dr. Pepper was still a free choice.

One can nit pick about the particulars of the thought experiment (e.g. if you were free, then your choice wasn't just between Coke and Dr. Pepper, but between drinking Coke and not drinking Coke, etc.), but setting those quibbles aside, I think what Frankfurt thought experiments show is that the ability to do otherwise is not what is meant by libertarian free will. It shouldn't be part of the definition of libertarian free will.

However, Frankfurt cases almost never happen in real life. In real life, we make choices continuously every day without there being Frankfurt cases. In the absense of a Frankfurt case, if you have libertarian free will, then you do have the ability to do otherwise. The ability to do otherwise, then, is a consequence of libertarian free will in the real world. So it does make sense to talk about libertarian free will as entailing the ability to do otherwise, at least in the real world as opposed to imaginary scenarios.