Do Actualities and Potentialities Actually Work?

So, in reading Feser I noted that obviously the Scholastics make a big deal out of actualities and potentialities.  The reason is that they use this to make causation work and escape the dual seeming absurdities of Parmenides who argued that there can be no such thing as change and Heraclitus who argued that everything was in constant change and there’s no permanence in the world at all.  The Scholastics adopt, with some minor changes, Aristotle’s solution to this that argues that objects have actualities and potentialities, and so what explains their permanence is that they are actualized into the object with the actualities it has, and what explains their ability to change and allows for causation is that something can actualize their potentialities and thus cause them to indeed change how they are actualized.  So this allows for causation and for us to be able to avoid the issues Parmenides and Heraclitus raised, with issues around how we could possibly avoid having to rely on something that doesn’t exist or doesn’t exist yet to explain how things can change.  Actualities and potentialities exist as part of the objects, and since those are the things that driving the change we have clearly existent things that we can use to trigger and explain all the changes we see in the world.

But I wonder if we need them, and also wonder if the concept has some issues.

One issue that came up in the original comment thread that sent me down this path is the fact that for every distinct interaction between two objects, the cause has to have an actuality to active the potentiality in the effect, and the effect has to have a potentiality that can be triggered by the cause in order for causation to happen.  I’m not going to go and find the comment, but essentially the comment from that person was that this would result in an infinite set of potentialities and/or actualities, and that seemed rather odd.  I noted that we can split some off, so it wouldn’t be actually infinite or rather problematically infinite, in that we’d be able to essentially give a list of them and be able to derive sets of what can and can’t happen.  It wouldn’t mean that it was open season and anything could happen.  Even if that list was technically infinite — but since we likely have a finite set of types of objects and these would work more by type than by specific object — it would be a distinct set that would identify only the things those specific objects can do.

But then I thought about an issue with this, which is that the object, then, has to contain potentialities and/or actualities that relate to objects or types of objects that it may never actually interact with, and it might even have to hold potentialities and/or actualities for types of objects that don’t even exist yet.  The reason is that if the two objects suddenly come into contact, it would defeat the entire purpose of the actuality/potentiality system that once a cause comes into contact with an object that it wants to make an effect out of that the potentiality is suddenly created and then activated by the cause.  We can escape some of this by clustering potentialities and noting that a variety of types of objects can activate the same potentialities, and even insisting that the potentialities describe what the object can do and so would never align with types of objects, but then again we’d still have to have in the object all potentialities that any type of object could do, again even if those objects didn’t exist yet.  As noted, if the objects don’t exist yet we couldn’t suddenly add those potentialities to the potential effects when they come into existence, but we would need to in order to make it work.  Again, we can escape much of this by arguing that the potentialities are effect-object-specific, but since some of them do seem to be reliant on the cause object’s properties that’s not a very clean solution.  Regardless, we need the cause-object and the effect-object to have as part of them an actual set of these things that they have to have over and above their actual properties (even if they are derived by them).

Which leads to the second issue I had:  which is primary in a cause and effect relationship:  the actualities in the cause-object or the potentialities in the effect-object?  Feser tends to give the cause-object precedence, but that’s because he — problematically in my opinion — tends to talk about causation as creating ex nihilo, and so it must be the case that the thing with precedence be the thing that actually exists.  But this sort of situation is extremely rare and arguably never actually happens in the world we have, as what we see is always something that already exists being transformed by its cause.  We may create a rubber ball that never existed before, but what we are actually doing is creating it out of some rubber, and so transforming the rubber into the ball.  And the same would apply to melting the ball.  So for the most part we pretty much always have an interaction between two objects that exist, and so this would raise the question of which object is the primary one in this situation.

It’s easy to say that it’s still the cause-object.  After all, the cause-object is the one that is actually doing the work, and the effect-object is the one being transformed.  So the cause-object is active, the effect-object is passive, and if the cause-object didn’t initiate the interaction nothing would happen, so the cause-object is primary.  Except under this model it doesn’t really matter what the cause-object does if the proper potentiality doesn’t exist, so it seems like all the cause-object is doing is essentially asking the effect-object to do something or change, and it couldn’t actually do that if the effect-object didn’t give permission (by analogy).  So then the effect-object is primary.

So we have arguments that argue for the primacy of each object, and it seems like we should be able to answer the question.  I believe most Scholastics would probably simply say that neither is primary or both of them are and so it doesn’t really matter, and I’m making an issue out of nothing.

So I think that in both cases the Scholastics aren’t going to worry too much about it, especially since they will argue that there really isn’t a better solution.  But I think I have one by focusing on the properties themselves and not inventing actualities and potentialities to make this work.

What I would agree with the Scholastics on is that everything that exists has a substance.  I would also agree that every existing object has properties, which are the aspects that make that thing what it is instead of something else.  But when we start talking about causation, what I want to talk about are interactions between the objects themselves and between the properties of each objects, not about set potentialities.  I even want to add some properties that are relational that can be formed when objects start to interact and trigger what happens when those objects relate to each other.  So causation, then, would happen when two objects relate to each other in a certain way, and how those relations are formed is based on their properties.

So let me look at one of Feser’s common examples, that of a brick hitting a glass window and breaking it.  Feser would need to describe it as the brick-object having an actuality to activate the glass-object’s potentiality to break when struck, which is awkward, to say the least.  By contrast, what I’d do is say that the brick is hard, glass is fragile, and when hard things hit fragile things the fragile things tend to break.  I can derive fragility and hardness from other properties and so can make those properties purely descriptive.  I can also apply the same properties to multiple objects and thus use the same rule to describe the same changes.  So I can use the same descriptive rule to talk about stones hitting windows and bricks hitting windows.  More importantly, the rules don’t exist in the objects at all, whereas Feser’s actualities and potentialities have to exist in the objects themselves.  And if the effect fails to happen I don’t have to go around and explain away why the potentiality wasn’t activated and go looking for an actuality or something that prevented it.  I can go look at the environment itself or in this case things like the angle it was struck and create more rules to cover those exceptions and situations (for example, using a general rule about force to cover it instead of the simplistic “hard/fragile” rule).  This aligns quite well with what science does, far better than what the Scholastics would suggest, and Feser is clear that he wants and needs to preserve science while arguing for his position.

So, at a very shallow level, it looks like my option allows for causation and even talks about things that exist — in this case, the objects as they are in their bare minimum of having properties that say what they are and what state they are in — without having to invent a new set of existent things.  We don’t even really need relational properties and they and laws could be entirely descriptive and so not really real.  Feser would likely reply that leaving things as properties runs into issues with how we assign properties to objects and so the need for a formal cause, but I’d challenge that as well since I attach properties directly to the substance of the object and so the properties that are part of that substance and the ones that follow from those properties are the ones that belong to that object.  So I don’t really need a formal cause either, or a form.  Feser obviously has objections to this idea, but I’ll get into them in a later post when I defend at least my notion of conceptualism from Feser’s attacks.

So, to sum up, as with final causes to me it doesn’t seem like we really need actualities and potentialities to make causation work, and the concept adds oddities and complexities.  So I am skeptical of the move given that it doesn’t seem to be necessary and isn’t a very clean way to solve the problems.  But, again, that’s a fairly shallow analysis.