DanielCC wrote:

Miguel wrote:

Let's argue. How could someone deny PSR or PC and still preserve:

1) The order in the world;

2) The legitimacy and justification of our explanatory practices, IBE etc;

3) The reliability of sensory experience and reason.

They would say that the last two are best accounted for by there being an explanation for most contingent states of affairs. To avoid ad hocness it would be better if the principle covered all contingent states of affairs, but, such a person would say, we have strong to reason to hold that it cannot due to the problems with the PSR i.e. modal collapse. From a philosophers belief standpoint more light needs to be shone on the problems such as that.

 
Then we can quite simply use a PC instead, seeking causes or explanations for the existence of contingent things. It wouldn't exactly be ad hoc, since one may hold there can be significant differences between facts about actions, operations, relations etc and facts about existence. And then we can explain the order in the world, and also justify our explanatory practices at least to existence (which covers a lot of ground), maybe even avoid more troubling skeptical scenarios.

But even with full blown PSR I just don't see why people think the issues are on par. There are proposed solutions to the modal collapse problem (explanans need not entail explanandum; or some contingent actions can be self-explanatory) while the "most contingent state of affairs" principle is completely ad hoc, at least if used to avoid explanations of existence and so on. It is ad hoc, has less explanatory power, and is just plain weird. Why would "most" contingent states of affairs have an explanation if not all of them do? What could account for the difference between a contingent state that does have an explanation and one that doesn't? It gets even worse if we are discussing the existence of contingent things.

Edit: also, http://alexanderpruss.com/papers/PSR-prob.pdf

It seems localized principles would give us the same ontological commitments and face the same problems of modal collapse, and are not sufficient to avoid the issues of explaining order and justifying our explanatory practices. So what would even be the point.

Last edited by Miguel (6/20/2018 12:55 pm)

Miguel wrote:

Let's argue. How could someone deny PSR or PC and still preserve:

1) The order in the world;

2) The legitimacy and justification of our explanatory practices, IBE etc;

3) The reliability of sensory experience and reason.

As far as I can tell, someone committed to phenomenal conservatism could reject PSR without (2) or (3) becoming problematic; but phenomenal conservatism is sort of uniquely situated to resist skeptical threat arguments, so perhaps that's not surprising.

But surely they'd just appeal to PC regarding the fact that it seems brute facts aren't prevalent, even if possible; so you'd have to apply the defeater to that. I imagine you'd have to show ~PSR entails brute facts are likely, but I'm aware of no such argument, so I'm not convinced.

Sorry. I thought by PC you meant principle of causality, not phenomenal conservatism.

Correct, phenomenal conservatism could lead us to believe brute facts are not prevalent. But the problem is that if ~PSR, then either brute facts are very likely, or no meanigful probabilities attach to them. What this means is that ~PSR leads us to believe brute facts are not unlikely, and that is a defeater against our epistemic seeming that brute facts are not prevalent. Phenomenal conservatism is quite welcome here; we can just say that if ~PSR however then it would be a miracle if reality were as orderly as it is, and so in order to maintain our epistemic seeming we should accept PSR.

One can argue that if things can come into existence with no explanation whatsoever, then we should expect to observe af least some things coming into existence in the next second, from the sheer amount of simples that could come into being uncaused. See for example Pruss and Rasmussen's argument from chaos in the third chapter of their "Necessary Existence" book (the chapter is available in Google Books).

Another way to argue would be from the fact that for every possible world with a regular universe like ours, there are infinitely many possible worlds in which some thing or other comes in or out of existence with no explanation whatsoever. So the observed order would be extremely unlikely. (For more on that, check Pruss's book on PSR, specifically his section on the everyday applicability of psr)

Or it could be that there are no meaningful probabilities at play here. See, for instance: http://alexanderpruss.com/papers/PSR-prob.pdf

But if we cannot say anything objective about the probability of brute facts, surely we also have a defeater for the "seeming" that brute facts aren't prevalent. It would be like saying that there are no probabilities for whether a bowling ball will come into being with no explanation whatsoever in front of you now, but nevertheless it seems to you it "won't happen". That's absurd.

Phenomenal conservatism works because when we use it against something like traditional skeptical scenarios, we are not simultaneously saying there may be no explanation whatsoever to justify our belief; the belief just seems true, even if we can't directly argue for it (doesn't mean there is no ulterior explanation here beyond our ken). More importantly, it works because we are not simultaneously holding the possibility of a thesis that basically implies there are no probabilities whatsoever that the skeptical scenario is false; or that the skeptical scenario should be probably true despite our seemings.

But if the arguments are right, if PSR were false, then plausibly we shouldn't be observing such an orderly world. Or, alternatively, no probabilities whatsoever would attach to it. But that would be a very big problem for our seeming that the world won't be chaotic in the next seconds. Best explanation would be that PSR is true, which confirms our phenomenal seeming of consistent order. In face of this, the burden of proof is on the denier: after all, if brute facts are a real possibility and things can come into existence with no explanation whatsoever, then how on earth could this occurrence be improbable?

Last edited by Miguel (6/21/2018 6:30 pm)

UGADawg wrote:

What's the best developed version of the argument that admitting the possibility of brute facts entails one cannot attach 'objective' probabilities to the likelihood of their occurrence? I know Feser and Pruss have both advanced that claim, but it mostly seemed motivated by intuition. Admittedly it is plausible, but ideally I'd prefer something a bit stronger than prima facie plausiblity.

 
Check Pruss's book on PSR. In his chapters about the "everyday applicability of PSR" and "inference to the best explanation" he discusses these issues at length.

See also the article I linked: http://alexanderpruss.com/papers/PSR-prob.pdf

These slides can also be useful: http://alexanderpruss.com/psr-acpa-12.pdf

His article on Leibnizian cosmological arguments also briefly mentions problems with assuming brute facts are improbable.

The issue seems very plausible to me, though. It seems that those who deny PSR really have a big problem when it comes to justifying our explanatory practices, and explaining the apparent order in the universe. It doesn't seem like we can attach any probabilities to such events the way we do to other things; we cannot ground it in any laws, something coming into existence with no explanation isn't an event governed by laws. We can try non-nomic probabilities with a principle of indifference, but then there's no way you can make brute contingency unlikely - in fact, it would seem overwhelmingly more likely for all sorts of random things to come into being uncaused in the next minute, since for every possible world in which things are seemingly "orderly", there are infinitely many more worlds in which this order is violated by the brute existence of different contingent things.