The word “possibility” can be used in two different ways: ways that are quite often confused and conflated, leading to some huge errors in thought. This is even done by very intelligent people.
One way has to do with our uncertainty about the future. Due to our limited prediction capabilities, we often look at and call future events in which we think at the time “could happen” as a “possibility”. This type of possibility I’ll call “epistemic possibility” as “epistemic” assesses our “knowledge or lack of knowledge” over the possibility.
It’s important to note that “possibility” in this epistemic sense does not necessarily align with whether something was a real possibility. It’s more of a perspective over our uncertainty regarding what is and what is not a real possibility. For something to be a real possibility it needs to have an ability to be actualized. In other words, it must be able to happen. Keep in mind that this doesn’t necessarily imply that it will happen, only that it “can”. It’s a real possibility. This is what possibility actually means – that it may in fact happen. To distinguish this from epistemic possibility it can be qualified as well. We can call it an “ontic possibility” meaning the possibility actually “exists” in reality. Another way to qualify this is simply to say “real possibility” or say that it’s “really possible” rather than simply an assessment of uncertainty in our head (epistemic possibility).
The distinctions between these two ways to use the word “possibility” should not be conflated, because they are very different. When people do conflate them they make some huge mistakes. For example, they may believe that when thinking about options as “possibilities” at the time of epistemic uncertainty, that they in fact are “real” ontic possibilities. This isn’t, however, necessarily the case.
The way we’d determine if they were real possibilities is a different process than how we assess epistemic possibilities. Right now I can think “it could be possible that I never finish writing and never post this article” but at the same time, in reality, it could be impossible that I do not finish or post this article. I just don’t know that for certain, while writing it, due to my limitations over future knowledge. I can recognize the distinction between these two things even before I finish writing or posting this article.
For real possibilities, we need to assess the ways in which events can and cannot happen. Does something cause the event to happen? If not, what does it mean that an event happens without a cause. If all things have a cause, what would that mean for possibility? If some events can just pop into existence, what does that mean for possibility? A large portion of the book Breaking the Free Will Illusion for the Betterment of Humankind deals with causal and uncaused event “possibilities” and what they mean.
This article will not be going over those argument, but rest assured that causality is not compatible with multiple options that are all “real possibilities”. To understand this you can read here:
This article could also be (indirectly) of use even though it’s more technical: Ontic Probability Doesn’t Exist: Assessing “Probability” for the Free Will Debate
My book explains how if there are events that do not have a cause (what I call acausal events) then those events can push the future to a different “possibility”, but such a push can never be “up to a chooser”. In other words it allows for real, ontic possibilities that can never be willed (let alone freely willed).
Within the book, I purposely use a present tense definition of free will, in which I use the word “viable” to basically mean “real possibility” (rather than epistemic possibility):
“The ability to choose between more than one viable option or action, in which that choice was up to the chooser.”
The reason I moved my definition from the more common past tense version is so people do not confuse the fact that “it already happened” with the fact that even before it happened the alternative options were either (given determinism) “never viable/really possible” or (given indeterminism) “not up to the chooser“. So though we might deliberate on options and think they are all possible, it doesn’t mean they actually are (unless an acausal event can lead to such a possible future – which we’d have no say in).
The more common definition of free will is a past tense version that addresses the word “otherwise” and addresses if an otherwise could be “up to the chooser” or “of one’s own accord. My definition can be moved to this definition which I often use as well:
“The ability to have, of ones own accord, chosen otherwise than they did.”
The nice thing about this past tense definition is that epistemic possibility cannot be conflated with ontic possibility – because one already knows ahead of time what had happened – so at least for our assessments of deterministic scenarios there is no longer future uncertainty. If we were to postulate that everything that exists is entirely causal (a deterministic universe), then asking after someone has eaten a banana if the could have not eaten that banana?… to say “yes” in response (under the scenario that every event has a cause) is a mistake. It’s a logical error.
Epistemic possibilities don’t get revived for “could have done otherwise” assessments given an entirely causal universe. The information is already available that they were caused to eat the banana, and therefore they couldn’t have been caused to not eat the banana, as that would lead to causes (variables) that were inherently a self-contradiction. Yet people tend to revive such ideas even given an entirely deterministic scenario, which lends to the incoherent free will abilities many lay-people think they have:
This is problematic and leads to incorrect intuitions about free will abilities that people truly do not possess. Given determinism, real alternate possibilities are logically impossible.
My book makes the strong case that the only type of event that can lead to a real alternate possibility would be an acausal event that would come into existence if we were taken back to a point in time – in which case that acausal event can influence the causal chains of events leading to a different output. Of course this presupposes an indeterministic universe where such acausal events are allowed in – which we simply have no evidence of (I hold an agnostic position in regards to indeterminism even if I have leanings toward determinism). For more information on that you can read here:
And though an acausal event can grant real alternate possibilities, whether one of those possibilities comes about would be entirely due to such an acausal event, an event that we would have absolutely no say over if it happens or how it happens. In other words, such an event can never be a “willed” event, and if such an event had any say over our thoughts and actions it would potentially have a detrimental effect as all consistency of thought gets thrown out:
But let’s be perfectly clear that, if we want to be logically coherent, the only type of event that is logically compatible with real alternative possibilities are events that acausally pop into existence for no rhyme or reason at all. If we postulate a universe that is entirely causal (determinism), real alternative possibilities are….impossible. Only one of the options were really possible, even while we deliberate and have no clue which option is the real possibility. In other words, while we deliberate between epistemic possibilities, only one path translates to an ontic possibility. The others never translate over unless we postulate indeterminism in which our deliberation process has no say over.
In the philosophy of free will, the principle of alternate possibilities (PAP) states that if there are not real alternate possibilities (if only one option is ever a real possibility and all of the rest are merely epistemic but can never be actualized) – that has a say over whether we have “moral responsibility”. And if the only way an alternative real possibility can happen is for a random or acausal event (or multiple acausal events) to push to the alternative, our incapacity to have any say over such an event does not grant moral responsibility – due to the mechanism of obtaining that alternate possibility. Due to this, free will skeptics such as myself say that moral responsibility in any strong sense cannot exist.
Take a look at this infographic to understand the moral responsibility we don’t have:
Note: I’ll eventually address the failing of Frankfurt cases which try to reject PAP, but that will be saved for another post.
It’s important to also note that not having “moral responsibility” in the strong sense above does not imply a lack of morality or lack of a moral system. The book I’m currently in the process of writing is on morality without free will, and explains this important distinction.
For this post I just want to create a preventative block on certain free will advocates who tend to confuse the very important distinction between epistemic possibilities (due to uncertainty over the future) and ontic possibilities (possibilities that have the ability to be actualized) for their own contrivances. Not only are these different, when given a deterministic universe scenario and addressing the question about if someone “could have done otherwise” – there is no rational way to confuse the two given that there is no longer “epistemic uncertainty”. And of course, indeterministic events that would be out of our control that could have lead to an alternate possibility wouldn’t be “up to us” in any way, shape, or form.
Here are a couple more articles on or related to the “possibility” word:
- The Word “Possibility” in a Deterministic Universe?
- (Real) Possibilities and Our Decisions – INFOGRAPHIC
Keep in mind that understanding what is or is not a real alternate possibility, and what it implies to have a real alternate possibility are important to the free will topic.