A Response to Two Objections to the Pruss-Koons PSR
Joe Schmid of Majesty of Reason released a three-thousand subscriber special Q&A video on his channel a few months ago, and one of the questions he was asked revolved around a possible way that a naturalist could avoid skepticism while also not being committed to the Pruss-Koons PSR. For those who don’t know, the Pruss-Koons PSR states that all natural facts, where natural means bounded or limited, have explanations; Pruss and Koons have a phenomenal paper where they provide powerful reason for thinking that a denial of their PSR entails global skepticism (see here: https://link.springer.com/article/10.1007/s11098-020-01482-3). In the Q&A, Joe goes over two possible ways a naturalist could avoid this skepticism whilst remaining not committed to the Pruss-Koons PSR (see here: https://www.youtube.com/watch?v=p0QRtp07TbQ&t=9906s - the relevant section is at 2:43:35). In the paper, and in Koons’ discussion with Oppy, they argue that if one denies their PSR they have no reason to think that one’s phenomenal experiences have explanations, which is a defeater for all empirical knowledge; this skeptical scenario is elucidated very well by Koons in his discussion with Oppy on Joe’s channel (see here: https://www.youtube.com/watch?v=bHw61T-zaeQ&t=3089s). The two responses Joe touches on are: 1. an appeal to phenomenal conservatism and, relatedly, 2. the naturalist could say that their phenomenal experiences of the world are contingent and hold to a PSR where all contingent things have explanations and thus we can have explanations for these contingent phenomenal experiences while avoiding your PSR.
I’ll respond to each one in succession. In the video, Joe himself says that Koons would probably respond to the appeal to phenomenal conservatism by saying that in order for phenomenal conservative belief to be formed, it must be objectively probable that our phenomenal experiences have explanations, but a denial of the PSR entails that there is no objective probability regarding whether our phenomenal experiences have explanations or not, and this provides a defeater for all empirical knowledge. Joe says that he is agnostic on the effectiveness of Koons’ response, but his response seems right to me and Pruss agrees: “For us to have empirical knowledge, it has to be objectively probable that we are not being misled by our perceptions. If the PSR were false, it wouldn't be objectively probable, because probabilities could not be assigned to uncaused experiences.” Moreover, many proponents of phenomenal conservatism have added that on phenomenal conservatism one has an epistemic obligation to seek out possible defeaters for their knowledge. So the naturalist cannot simply claim "it seems x" and sit back while their metaphysical commitments pose possible challenges to their knowledge. This leads us into the second response that Joe offers.
The second response seems to misunderstand the point of the Pruss-Koons PSR: Pruss and Koons are not disagreeing with the naturalist that contingent things need explanations, but rather are pointing out that the language of contingency and necessity is too broad and does not tell us what it takes for something to be contingent or necessary, and the Pruss-Koons PSR does provide us with a metaphysical tool to determine which things are contingent and which things are necessary, namely, boundedness and unboundedness. The whole point of your PSR is that the language of contingent and necessary is not specific enough to adjudicate disputes between world-views for two world-views could hold that there are contingent and necessary things, while disagreeing about what makes things contingent or necessary. Therefore, we must appeal to more specific categories in order to determine which theory is ultimately correct, and the categories of natural and supernatural, where natural facts are bounded facts and supernatural facts are unbounded facts, are the categories we need in this case. Moreover, the skeptical challenge cannot be posed to someone who holds that all natural, bounded, facts have explanations because this category is fundamental; the fact that my phenomenal experience of the world is bounded is irreducible to simpler categories, whereas the fact that my solipsistic experiences are contingent seems to be grounded in more fundamental facts about the nature of my solipsistic experiences, namely, that my solipsistic experiences are bounded. Furthermore, this reasoning can be applied to Oppy's PSR where all non-initial things have explanations, for how are we to determine which things are initial or non-initial, without appealing to some more fundamental category. The theist can, and ought, to agree with Oppy that all non-initial things do not need explanations, but it still seems plausible to ask the question: why is something non-initial? If one responds by saying that something is non-initial just because it is, this seems to be intellectually unsatisfying. It seems very reasonable to suppose that non-initial things are non-initial because of something more fundamental. After all, it does not seem to be reasonable to suppose that just anything could be non-initial. This is not the same for the question: why is something bounded? A thing is bounded because it has bounds or limits. There is a primitiveness or fundamentality to boundedness which the notions of contingency and being non-initial lack, which provides reason to think that this possible response by the naturalist does not succeed.
1) I fail to see how this response is any different to the one Oppy gave in the debate. The same applies here as it did back then, it's nothing but footstomping. As your friend aptly put it, the mere insistence on the reliability on one's faculties does nothing to actually show it. In the worst case it's just shutting out the external world. While I agree with the point about probabilities, I want to suggest that phenomenal conservatism itself requires something like the PSR in order for the position to be reasonable, after all an observation of the external world could be brute without any of us noticing it. It's not like a brute fact always need to present something extraordinary, but it can be something believable as well, a slight deviation from actual reality. How could we deny that? While Koons and Pruss in their argument only mention brute sensory experiences of a particular individual, the threat of global skepticism isn't evaded by merely claiming that the own sensory experience seems reliable. While it's true that our sensory faculties are reasonably conceived as reliable and their unreliability is an assumption that needs arguments, even conceding that won't help here. The first event in time, to follow an example given by Koons, can have the exact properties the event currently, while writing these words, can have, there's no metaphysical impossibility here. The issue isn't the reliability of our faculties, but the (missing) causal link in brute occurences. And these are external to us. The threat of global skepticism can be formulated in such a way that it doesn't depend of skepticism about our own experiences.
ReplyDeleteI honestly have a hard time understanding what's there to be agnostic about here. Especially if we keep in mind the point by your friend, that no metaphysical underpinning of phenomenal conservatism has been taken place. The argument by K&P has been met with question begging.
2) You have made great points here. I want to add though that there's a hidden premise in the reference to a PSR limited to all contingent facts, namely the impossibility of explaining necessity. This is quite significant. First of all, the explanatory principle is weaker than that of K&P. Unless we're presented with defeating evidence for the stronger version, it's not rational to just affirm the weaker one.
ReplyDeleteFurthermore, let me supplement your argument here by giving an example. In a possible world in which are only penguins and the first moment in time is the first penguin coming into existence. Following the reasoning Oppy applies and to which Joe often refers, given an Aristotelian account of modality, the first moment of the first penguin is necessary. What this is supposed to show is that "being the first event" is insufficient to ascribe necessity. Putting all the reasons about the nature of necessity aside, the first penguin isn't sufficiently different to all following events to coherently affirm its necessity. Not only do we do we have abundant evidence in subsequent penguins that they are contingent beings who come to be and die, we also observe how they are "produced" so to speak, letting us believe that, at least in theory, this would have been possible for the first penguin as well. This leads us to conclude that the first event wasn't necessary, but contingent and brute.
We can broaden this reasoning. In our universe the first event has particular qualities, like size, density, weight or constitution, which only change in degree and not in kind in following events. It would require convincing arguments as to how we could ever ascribe ontological priority to a particular quantity over another.
This can be further broadened to account for eternal, supposedly necessary, fundamental physicals. The PSR as Joe wants to propose as a way out, needs to overcome the argument from arbitrary limits, even though the entailment wasn't immediately obvious.
The arguments you give are absolutely correct, what could be a non-arbitrary requirement for necessity? "Unboundedness" certainly is one of them, since the entity in question doesn't share any of the arbitrarily limited properties we demand explanation for. Honestly, I can't even conceive of another one.
I want to conclude with an observation. None of the options given actually attack the argument by K&P, but only offer possible ontologies the naturalist might accept instead of the one they offer. However no argument has been given against the ontology which actually provides 1) a metaphysical underpinning for rational believability which isn't solely based on circular reasoning and 2) which provides more explanatory power.
Sorry for getting back to you so late. I like all of the things that you offered and I appreciate your compliments :)
DeleteIf we can know apriori that "All-natural facts are contingent" then I dont see why we can't know the principle that all natural mental facts are contingent as well? There's no threat of skepticism here right?
ReplyDeleteA fact is just a state of affairs, so I'm not sure I understand the distinction you're trying to propose. The threat of skepticism is only evaded if we can rule out that such contingent events could be uncaused
Delete