In its classic form, LFW posits the freedom to do otherwise. And this is the form of LFW which is deployed against Calvinism.
More recently, there are some libertarians who no longer regard this condition as a condition of LFW. If so, then that’s one less objection to Calvinism.
But, for now, I’ll address the classic formulation. Nowadays, the freedom to do otherwise is cashed out using the currency of possible worlds semantics. It ascribes to human creatures the godlike power to access and instantiate alternate possibilities.
So, the first question we need ask ourselves is what evidence is there that we enjoy this tremendous freedom? Indeed, what would event count as evidence for this sort of freedom?
The only direct evidence would be if I could travel back in time, reproduce the identical conditions under which I made a choice, and make a different choice.
This is not a straw man argument. This strictly follows from how LFW has been defined: “By ‘libertarian freedom’ is meant freedom such that the agent who makes a choice is really able, under exactly the same circumstances, to choose something different from the thing that is in fact chosen” (Hasker).
So the only way to put that proposition to the test would be to travel back in time. But in that event, there can never be any direct evidence for LFW. In that event, no one has ever had any experience of LFW. Indeed, no one could ever have any experience of LFW.
Why isn’t it possible to test this proposition? Because not all possibilities are compossible. You can’t do two different things in the same place at the same time. You can do two different things in the same place at different times, or do two different things at the same time in different places (by remote control).
One choice excludes another choice. By turning left, at that time and place, you didn’t turn right, at that time and place.
But if we have no direct evidence for LFW, do we have any indirect evidence for LFW? Well, if we had LFW, then there ought to be indirect evidence for LFW.
Although you can’t instantiate alternate possibilities simultaneously, you should be able to instantiate alternate possibilities successively—assuming that human beings enjoy LFW.
After all, alternate possibilities are just a special case of possibilities in general. They are differentiated by time. The freedom to do otherwise would be the freedom to do otherwise at the same time.
Even though this is unverifiable, it should also hold at a diachronic level as well as a synchronic level. For we’re talking about the general ability to realize abstract possibilities.
However, human experience doesn’t bear this out. If I had the ability to realize abstract possibilities, there are all sorts of things I should be able to do that I’m unable to do.
As a practical matter, if I want to do something, I have to use my hands and feet. Or I can use my voice to issue commands.
In other words, my choices are limited to what the physical world presents to me. And my field of action is limited to what the physical world presents to me.
In that event, the actual world delimits the range of the possible. I can choose from what is physically available.
But if I truly had the power to realize abstract possibilities, then the actual world shouldn’t pose a limit on my field of action. For, in that event, the actual world would be the sum total of what I and other free agents actualize.
The limiting factor on what is possible wouldn’t be the actual, but a competing possibility. If I want it to rain, and you want it to snow, at the same time and place, then those are incompossible possibilities.
But after we make allowance for all of the incompossible wishes of various agents, the world we live in still doesn’t look like the world I’d expect to find if we had the power to realize abstract possibilities. In that event, the real world should be a magical world. A world where every wish comes true—as long as your wish doesn’t conflict with my wish.
One reason I’m not a libertarian is that I don’t live in the sort of world predicted by libertarian freewill. And it’s easy to see why this couldn’t be true.
If LFW were true, you’d have all these counterfactual histories lined up in storage, waiting to be instantiated. A plurality of futures on tap. Take your pick! You don’t like this historical outcome? No problem! We’ll edit that out and splice in an alternate ending.
The problem with that notion is that it severs all the lines of causality, then picks up where it left off—as if the chain of cause-and-effect had never been disrupted.
Imagine if you tried that with a family tree. Selectively adding or deleting genealogical links between Abraham and David. The problem is that, if you make one of David’s forebears disappear, you make David disappear. You can’t very well swap out one historical segment, swap in another historical segment, then leave everything else intact. No, that dislocation will displace everything else down the line.
Imagine billions of competing agents with that power at their mental fingertips. Let’s destroy the world and recreate the world ten times a day.