I don't buy your claim that RWH is somehow more parsimonious than SH. it's easier to build a video game than it is a universe. With RWH you also need to explain both epistemology and ontology, whereas for SH, you just need epistemology and don't need to make ontological claims.
I'm not quite sure what you mean by epistemology here. I don't think a theory of the world has an epistemology built in. Rather I take it that epistemology is the method for choosing between theories of the world.
As for SH not needing to make ontological claims, I just don't think that's right. I mean, part of the theory will be specifying what exists (e.g. a brain, a vat, some scientists, etc.). But even if that's right, I suspect that ontological parsimony reduces to simplicity more broadly construed, and in terms of simplicity I hope I have shown that SH is worse off than RWH.
As a skeptic, you don't need to make claims on the external world, so ontology just isn't your problem. A skeptic just needs to explain the *possibilities* of perception and epistemology, not even commit to any particular ontology (whereas RWH needs to explain both ontology and epistemology).
So long as a skeptic can explain perception/epistemology in a way that's more parsimonious than RWH (video game universe is easier to make than an actual universe), then the skeptic's view would be the better explanation.
I'm not a skeptic, but I'll be writing my own rebuttal to it as well. I just don't take the same route you're going. Once you posit an external world outside of mind, then you're vulnerable to skepticism, at least to some (uncertain) extent.
Sure, the skeptic *in general* doesn't have to commit to some specific form of ontology, seeing as all they have to do is not have the RWH ontology. But still each particular skeptical scenario will need to commit to some ontology, be that a computer simulating the universe, an evil demon, or something else. The point is that for each of these, the explanation will be worse than RWH. This is because the simplest way of explaining our evidence is with the structure of RWH. But whereas RWH stops there, every single skeptical hypothesis will need to add some extra complications, in order to reach skepticism. Now it is true that RWH commits to this structure also has a certain sort of ontology (something like real existence, or whatever you want to call it), but every single SH will also have to commit to the real existence of something, it is simply that the real existence will be in the complicating extra element, which has been tacked on in order to make the hypothesis skeptical. So RWH beats out every single skeptical scenario in terms of complexity, it seems, and in terms of ontology it is on a par with each SH.
The brain in the vat argument doesn't *affirmatively* posit that we are actually brains in vats, only that it is *possible* that we are. It doesn't make an ontological commitment to brains in vats, only that such a commitment can't be ruled out. Big difference.
Once you raise the skeptical possibility, then RWH becomes just one of many possibilities, and it would require making more ontological commitments than other skeptical theories. So long as we can think of equally likely alternatives with less commitments, then those scenarios become more parsimonious
The argument, at least as usually formulated, raises the BIV hypothesis as an alternative hypothesis. The idea is then that you cannot rule out this hypothesis, and so don't have knowledge. While the argument itself doesn't make any ontological commitment, the BIV hypothesis does. And my point is then that BIV (and other SH's) all turn out to be much inferior hypotheses, and so we can rule them out, and know that RWH is true.
I don't grant that we can think of equally likely alternatives, with less commitments, which explain our evidence just as well (as I argue in the post). While the disjunction of all SH's doesn't make any particular ontological commitment, each individual SH does, and as I argue each individual SH is so inferior that the whole disjunction is much less likely than RWH (just like the whole disjunction of weird alternatives is ruled out in ordinary uses of abduction).
The only ontological commitment the skeptic has to make is the *possibility* of BIV, whereas RWH is committed to the *actuality* of the universe. RWH has to quite literally commit to everything in the universe. So SH has less commitments than RWH.
Still a good article, skepticism is just very slippery. But I will argue that once youre agnostic over the truth of the external world (and don’t rely on the external world for your philosophy) then skepticism is groundless.
My sense is that many people bring up SHs, not to convince you that they're more plausible than RWHs, but to have non-trivial amounts of plausibility and therefore ruin our ability to have certain knowledge of the external world. Best explanation arguments help us find the most plausible theories, but they don't reduce opposing theories' plausibility to zero.
I'm not myself a skeptic, though. It seems to me like the best way to tackle this problem is to argue against the notion that certain knowledge is only possible if all opposing theories have zero (or comically low) probability.
PS: The Descartes evil demon counterargument doesn't work. It could be that you do not need logical connectives to formulate the SH, and the demon is just tricking you into thinking that you do.
Yeah, I guess I was presupposing that knowledge doesn't require near-certainty, but only reasonable certainty, as I think the former is very implausible. If you want abduction to be able to provide knowledge in general, then you need to have a somewhat lax standard, and I think the argument against skepticism is on par in terms of strength with many other ordinary abductive inferences.
As for the evil demon, I don't think what you say works for the evil demon hypothesis I am actually considering. When I consider the hypothesis, I think that I am directly aware of the fact that I am thinking of it using logical connectives, in such a way that I couldn't be deceived about it. Now, it might be that there is some evil demon hypothesis *out there* which doesn't require logical connectives, but then that hypothesis is impossible for me to think about (or at least I haven't been able to think about it thus far).
> Yeah, I guess I was presupposing that knowledge doesn't require near-certainty, but only reasonable certainty, as I think the former is very implausible.
I think so too. Structurally, I think a lot of skeptics make the following argument.
Premise 1: We can't be near-certain that RWH is true. [SHs help to motivate this premise.]
Premise 2: If we can't be near-certain that RWH is true, we can't have knowledge of the external world.
Conclusion: We can't have knowledge of the external world.
I think Premise 1 might or might not be true - I lean towards it being false, but I think there are some good arguments in favor of it. But I think Premise 2 is much weaker. As you say, it seems pretty implausible. Of course, since the skeptic has to defend it, they've taken some pretty good shots at motivating Premise 2. For instance, when Descartes steelmans the skeptic position (I say steelman because he's not himself a skeptic), he goes on for a bit about how if we want to know anything it should be based on an undoubtable/invincible foundation, and implies that requires certainty.
I don't buy your claim that RWH is somehow more parsimonious than SH. it's easier to build a video game than it is a universe. With RWH you also need to explain both epistemology and ontology, whereas for SH, you just need epistemology and don't need to make ontological claims.
I'm not quite sure what you mean by epistemology here. I don't think a theory of the world has an epistemology built in. Rather I take it that epistemology is the method for choosing between theories of the world.
As for SH not needing to make ontological claims, I just don't think that's right. I mean, part of the theory will be specifying what exists (e.g. a brain, a vat, some scientists, etc.). But even if that's right, I suspect that ontological parsimony reduces to simplicity more broadly construed, and in terms of simplicity I hope I have shown that SH is worse off than RWH.
I’d also say that abduction *does* work necessarily. But whether your use of it justifies RWH depends on what data goes in
As a skeptic, you don't need to make claims on the external world, so ontology just isn't your problem. A skeptic just needs to explain the *possibilities* of perception and epistemology, not even commit to any particular ontology (whereas RWH needs to explain both ontology and epistemology).
So long as a skeptic can explain perception/epistemology in a way that's more parsimonious than RWH (video game universe is easier to make than an actual universe), then the skeptic's view would be the better explanation.
I'm not a skeptic, but I'll be writing my own rebuttal to it as well. I just don't take the same route you're going. Once you posit an external world outside of mind, then you're vulnerable to skepticism, at least to some (uncertain) extent.
Sure, the skeptic *in general* doesn't have to commit to some specific form of ontology, seeing as all they have to do is not have the RWH ontology. But still each particular skeptical scenario will need to commit to some ontology, be that a computer simulating the universe, an evil demon, or something else. The point is that for each of these, the explanation will be worse than RWH. This is because the simplest way of explaining our evidence is with the structure of RWH. But whereas RWH stops there, every single skeptical hypothesis will need to add some extra complications, in order to reach skepticism. Now it is true that RWH commits to this structure also has a certain sort of ontology (something like real existence, or whatever you want to call it), but every single SH will also have to commit to the real existence of something, it is simply that the real existence will be in the complicating extra element, which has been tacked on in order to make the hypothesis skeptical. So RWH beats out every single skeptical scenario in terms of complexity, it seems, and in terms of ontology it is on a par with each SH.
The brain in the vat argument doesn't *affirmatively* posit that we are actually brains in vats, only that it is *possible* that we are. It doesn't make an ontological commitment to brains in vats, only that such a commitment can't be ruled out. Big difference.
Once you raise the skeptical possibility, then RWH becomes just one of many possibilities, and it would require making more ontological commitments than other skeptical theories. So long as we can think of equally likely alternatives with less commitments, then those scenarios become more parsimonious
The argument, at least as usually formulated, raises the BIV hypothesis as an alternative hypothesis. The idea is then that you cannot rule out this hypothesis, and so don't have knowledge. While the argument itself doesn't make any ontological commitment, the BIV hypothesis does. And my point is then that BIV (and other SH's) all turn out to be much inferior hypotheses, and so we can rule them out, and know that RWH is true.
I don't grant that we can think of equally likely alternatives, with less commitments, which explain our evidence just as well (as I argue in the post). While the disjunction of all SH's doesn't make any particular ontological commitment, each individual SH does, and as I argue each individual SH is so inferior that the whole disjunction is much less likely than RWH (just like the whole disjunction of weird alternatives is ruled out in ordinary uses of abduction).
The only ontological commitment the skeptic has to make is the *possibility* of BIV, whereas RWH is committed to the *actuality* of the universe. RWH has to quite literally commit to everything in the universe. So SH has less commitments than RWH.
Still a good article, skepticism is just very slippery. But I will argue that once youre agnostic over the truth of the external world (and don’t rely on the external world for your philosophy) then skepticism is groundless.
My sense is that many people bring up SHs, not to convince you that they're more plausible than RWHs, but to have non-trivial amounts of plausibility and therefore ruin our ability to have certain knowledge of the external world. Best explanation arguments help us find the most plausible theories, but they don't reduce opposing theories' plausibility to zero.
I'm not myself a skeptic, though. It seems to me like the best way to tackle this problem is to argue against the notion that certain knowledge is only possible if all opposing theories have zero (or comically low) probability.
PS: The Descartes evil demon counterargument doesn't work. It could be that you do not need logical connectives to formulate the SH, and the demon is just tricking you into thinking that you do.
Yeah, I guess I was presupposing that knowledge doesn't require near-certainty, but only reasonable certainty, as I think the former is very implausible. If you want abduction to be able to provide knowledge in general, then you need to have a somewhat lax standard, and I think the argument against skepticism is on par in terms of strength with many other ordinary abductive inferences.
As for the evil demon, I don't think what you say works for the evil demon hypothesis I am actually considering. When I consider the hypothesis, I think that I am directly aware of the fact that I am thinking of it using logical connectives, in such a way that I couldn't be deceived about it. Now, it might be that there is some evil demon hypothesis *out there* which doesn't require logical connectives, but then that hypothesis is impossible for me to think about (or at least I haven't been able to think about it thus far).
> Yeah, I guess I was presupposing that knowledge doesn't require near-certainty, but only reasonable certainty, as I think the former is very implausible.
I think so too. Structurally, I think a lot of skeptics make the following argument.
Premise 1: We can't be near-certain that RWH is true. [SHs help to motivate this premise.]
Premise 2: If we can't be near-certain that RWH is true, we can't have knowledge of the external world.
Conclusion: We can't have knowledge of the external world.
I think Premise 1 might or might not be true - I lean towards it being false, but I think there are some good arguments in favor of it. But I think Premise 2 is much weaker. As you say, it seems pretty implausible. Of course, since the skeptic has to defend it, they've taken some pretty good shots at motivating Premise 2. For instance, when Descartes steelmans the skeptic position (I say steelman because he's not himself a skeptic), he goes on for a bit about how if we want to know anything it should be based on an undoubtable/invincible foundation, and implies that requires certainty.
Great post, and also yes that's how I thought video games worked as a kid too lmao
Good to hear I'm not alone!