Aha. So his arguments use deduction, but his premises are the result of induction right? Like deductive arguments inferring from inductive conclusions?
Well, I would go one step further and say that even his premises are not the result of induction. Take for instance a summary of Aquinas’ 1st way:
Argument from change
- Some things change.
- Everything that changes is made to change by something else.
- The chain of causes can’t go back to infinity.
- Therefore, there must be a cause of change that does not itself change.
Which of these 1-3 would be the inductive premise(s)? Well, let’s first be clear on the definition of induction. Taken from
Philosophy Dictionary, one of the definitions reads ‘A method of reasoning by which one infers a generalization from a series of instances.’ But more poking around on the net led me
here and
here. Which leads me to believe that there are a number of different kinds of inductive reasoning, generalization only being one kind. And it also leads me to believe that “infers” from the Philosophy Dictionary definition must introduce the notion of probability–and as taken from your post discussing your changes.
I really think the notion of probability is key, while inferring from one group to another group (of however many or few entities one wants).
So, is #1 taken from induction? Well, at the very least #1 is taken from experience. But there doesn’t seem to be any inductive inference; it seems to be an immediate kind of thing. Do I infer from the group of one thing changing to the conclusion that some things change? Well no, not if you take “infer” as indicating some level of probability. That some things change is just a logical deductive corollary of my experience of one thing changing. So, #1 is not reached by induction. I daresay someone could still try to argue this by questioning how we could know that any object of experience exists in the first place and that it might just be an illusion: for now I will not travel down that road unless someone wants to.
I really think #2 is the best candidate for a premise arrived at by induction. But does one infer from one instance of change (something being changed by another) that everything that changes is being changed by another? Well, first off, this conclusion is not what #2 states. #2 states that ‘Everything that changes is made to change by something else’, which means it’s not talking about the actual state of affairs directly but any
possible state of affairs. So is Aquinas’ guilty of such a horendously weak induction, where he goes from what he sees here and now to not only what he doesn’t see, but that which will ever be?
I don’t think so. In metaphysics, if you follow an aristotelian line of reasoning about being from our everyday experience (which includes change) you come to two principles that must hold for any possible change to occur. #2 is just the result of abstraction and understanding the nature of change, at least enough to declare what is necessary for change. So instead of Aquinas seemingly making an induction from one instance of change to every possible instance of change; rather, Aquinas has understood the nature of change enough to say that if you have an instance of change, these principles must be in effect.
Now the further question remains, is induction used in an aristotelian line of reasoning about being? As far as I can tell, no. Inductive reasoning seems to remain probable because the attribute in question is not a necessary feature of the entity or at least not known to be a necessary feature. With change we can find necessary features. Therefore, induction does not apply.
Is #3 arrived at by induction? I don’t think so. It seems #3 follows upon the nature of change, at least regarding the two fundamental principles of change. So for the same reasons as with #2 I would have to say no, unless I am failing to see something.
So it would seem that Aquinas’ 1st way (and I would argue all of the other ways) does not involve induction. If one takes induction as merely getting a general principle from a particular thing and excludes the notion of probability… then maybe, just maybe one could claim induction in Aquinas’ arguments. But I think that’s a stretch.
Any probability in Aquinas’ 1st way is due to any probability that one might not actually experience change, but might be tricked by an evil demon, ect. And there doesn’t seem to be any reason why we shouldn’t trust our immediate experience, in general.
These kind of mistakes, of supposing abstraction and apprehension of necessary features of a nature are actually discovered by induction and therefore not necessary, can be traced to nominalism and a denial of essences in modern philosophy. One way modern philosophers fell into this trap was an erroneous conception of experience: many times they thought that we only have access to “representations” of objects rather than the objects themselves (come to think of it, that might be due to a rejection of final causality, leading credence to a recent book I read). Cue much of the havoc and confusion in philosophy.
Questions?