I'm doing a series at my own blog based on the content of an intro to philosophy course that I teach from time to time called Theories of Knowledge and Reality. This is designed to reflect that course, so it's aimed to be at a fairly introductory level, though later posts in the series reflect what was covered earlier (as should be the case in a course). I'm now up to the cosmological argument. I didn't cross-post them here due to their aim at being introductory level, and I didn't foresee them being of interest to most readers of this blog, which tends to operate at a higher level of detail. The comments have clearly into that level, however, so I thought it might be of interest to readers of this blog. The first post presents what I think is the strongest version of the cosmological argument, and the second responds to what I think are the two strongest objections.
Technorati Tags: blog, philosophy course, cosmological argument, cross post, level of detail, foresee, aim
Hi Jeremy-
One worry about PSR. It does seem reasonable to require that, for each event e or object o, there is some explanation for e's occurrence and o's existence. But it does not seem reasonable (not to me, anyway) to require that, for each e and o, there is some final cause (or point or purpose) for e and o.
As you know we can explain why something happened or exists by appeal to (among other causes) efficient causes and final causes. PSR I think rightly urges that there must be some efficient cause for event e (I'd say so, even under the assumption that nature is indeterministic). But PSR is not plausible in it's demand for a final cause for each event e. Lots of events and objects simply have no point.
But this is exactly what PSR demands when applied to the infinite series of efficient causes. When Leibniz argues that there must be some reason for series S rather than series S' he is demanding not the efficient cause of S (there is no efficent cause of S in addition to the infinite series of efficient causes), but the final cause of S. Leibniz's answer: it is the best series, and that gives us the point or purpose for God's choosing this series (i.e., this world) rather than another.
But I think we should balk at the idea that PSR demands a final cause for the infinite series or for any event or object.
While I share Leibniz's intuition that there is a final cause explanation for every event, I agree with you that that's not something we can assume from an intuitively plausible PSR that someone might use in an argument like this. But I don't think the argument as I've presented requires a final cause for the existence of the complete set of infinite causes in the past. I think it does require some sort of explanation besides the full set of efficient causes, but that need not be a final cause. (It especially need not be a final cause in the sense of an intelligent designer.)
As long as you don't assume that an efficient cause must be before its effect, then an efficient cause can explain an infinite series. Most theists do think of God as an efficient cause of all dependent things anyway, and most theists historically have considered God atemporal. Thus God on that view would be an efficient cause explaining the (assumed for the sake of argument) infinite series of efficient causes that each have a prior efficient cause.
In general, PSR as I've been thinking of it requires some sort of explanation. If something has no explanation, it violates PSR. Efficient causes can serve as an explanation, but some things can satisfy PSR by having a final cause explanation and no efficient cause explanation. Augustine, Aquinas, and Leibniz would insist that human free choices are like that. That doesn't amount to seeing PSR as requiring a final cause, however. It just requires some sort of explanation for any individual thing or any positive fact.
"As long as you don't assume that an efficient cause must be before its effect, then an efficient cause can explain an infinite series".
I can't follow that. I'm not assuming that a cause must precede its effect. Causes and effects can be simultaneous, of course. Take any event in the actual series e. Explain to me how it is not redundant to have e efficiently caused by the event e' and e also efficiently caused (now as a member of the efficiently caused series S) by, say, God. How is it that e is not overdetermined? And surely PSR does not require overdetermination.
I'm talking about an atemporal cause, actually.
It's not a redundant cause, because each is caused in a different sense. There's the immediate cause and the ultimate cause, where both are efficient causes. If A causes B, and B causes C, then A causes C. It doesn't causes C in the same sense that B does, but both are efficient causes. There is no ultimate cause if the infinite succession in the past is correct. But then there's no ultimate cause for anything, and that's a kind of efficient cause that Aquinas thinks there must be. This isn't a final cause. It's an efficient cause. It's not redundant because it's explaining the ultimate grounding of all things, which none of the intermediate causes do, and which the sum of the intermediate causes does not do.
Ok, I see. So let 0 be the ultimate cause in the sequence S, but not part of the sequence S. Two quick points. First, this does not entail that S is a finite series. S might well be infinite in the very same way that the real interval (0,1] is infinite. 0 is the greatest lower bound on the interval (not contained in the interval) and God is the initial atemporal cause (and, since atemporal, not contained in the sequence). Perfectly consistent with an infinite sequence of causes.
Second, if God is the ultimate cause of series S, we still do not have an explanation of S occurring rather than S', do we? I'm happy to to say that PSR leaves this a brute fact, and it does seem just a brute fact at that point.
I suppose everything I've said is consistent with its being like (0,1]. But that's not what I was thinking. I was thinking that it would go back infinitely in the past, with no lower bound. There is no initial cause. There's an atemporal cause.
What's your argument for your second point? If God explains why it's S rather than S', then it's not a brute fact. It's explained.
If there's an atemporal cause, then there is a lower bound on the set of causes. That's unavoidable.
On the second point, the fact that God initiated S rather than S' does not provide an explanation as to why he did so. "Why not S'?" is the question that goes unanswered. But as I said, this is no worry for me.
If the explanation requires is literally a sufficient explanation, as Leibniz wanted, then there must be a reason God would choose one or the other. PSR would then lead to more than simply a self-existent thing but must require something that can choose on the basis of reasons. I wasn't assuming that PSR required what we would call literally a sufficient reason, though. It might just be a necessary reason. B couldn't have happened without A, but that doesn't mean A required B. John Hawthorne and Andrew Cortens have a good paper on this in Faith and Philosophy called "The Principle of Necessary Reason". If I remember correctly, the main motivation is to avoid this exact problem (while also avoiding theological determinism, I suspect). As I present PSR to undergrads, I'm trying to leave it open whether it's Leibniz's PSR or the Hawthorne/Cortens version.