Is this all about Bayesian reasoning. That's tge chapter of Why I always get stuck on. I don't think it makes sense applying it to the whole universe because (wavey hsnds) the context is undetermined (or something).
I do think the Multiverse Objection is a good objection to the Fine-Tuning Argument (FTA), but mostly because it seems like certain theories in physics (Many Worlds Interpretation of Quantum Mechanics, Inflationary Theory, Big Crunch theories, String Theory, etc.) naturally posit a multiverse.
If the above weren’t true and the only reason we posited a multiverse was to explain fine-tuning, then I do think it would be a much weaker objection (and therefore, even if it’s not decisive, the FTA would be evidence for God*).
That said, since the multiverse is still only a possibility, I think the FTA is a good argument against the rationality of atheism even if it doesn’t prove theism (i.e., it’s like an argument for agnosticism).
* Technically it’s really just an argument for design. With simulation theories, it’s always possible we live in a simulation rather than a universe designed by an omnipotent God! I’m not sure how plausible the simulation theory is, but it’s at least something to consider.
In my Why? book I argue that if we're to go for eternal inflation but avoid this dodge reasoning, we need to postulate a multiverse in which all the universes have the same physics.
By replying here, I can address the main participants in this debate in one post!
For what it is worth, I agree with Philip insofar as I feel that the apparent fine-tuning of the universe is beside the point when the question is whether any multiverse hypothesis is plausible, and so it does not shift my credence on that matter one iota.
Having said that, I think there is a different argument that Philip should take seriously: the one from conceivability - multiverses are conceivable, and therefore metaphysically possible.
This is, of course, a riff on the zombie argument. As Philip endorses that argument, despite not thinking zombies are real, I feel it would be inconsistent for him to reject the corresponding argument for multiverses.
This would not, of course, rule out cosmic purpose, but, as he says above, "... it’s clear that the probability of cosmic purpose is significantly less if the multiverse is in the picture than it is if cosmic purpose is the only explanation of fine-tuning."
For what it's worth, I disagree with your agreement with Philip insofar as I feel that the apparent fine-tuning of the univere is relevant.
Philip does not mean to say that the multiverse is not conceivable (edit: or metaphysically possible). He thinks it is conceivable (edit: and presumably, at least conceivably, metaphysically possible), and has a non-negligible credence in it. That's not the point at stake.
What he means by "multiverse is in the picture" is not that the mutliverse is a viable option, but that we're considering what would follow if it were true. If it were true, then there would be much less reason to believe in cosmic purpose.
I don't doubt that what you say about Philip's position is correct, but I do not feel it goes so far as to address the issue of consistency over what conclusions can be drawn from metaphysical possibility. I see I need to be clearer in that respect.
In the case of zombies, he has often used them (for example, in his debates with Sean Carroll) to argue that the metaphysical possibility of zombies (as established by the conceivability argument) is as problematic for physicalism as their actual existence would be. Given that position, it would seem (at least to me) to be inconsistent to then avoid concluding that the metaphysical possibility of other universes (as established either by the equivalent conceivability argument or any other way) is as problematic for cosmic purpose as their actual existence would be.
The metaphysical possibility of zombies is supposedly a problem for physicalism because it demonstrates that, even conceptually, consciousness and physical behaviour are distinct, so the first cannot be reduced to the former even in principle.
I don't see a parallel argument that the metaphysical possibility of a multiverse is problematic for cosmic purpose.
It is not entirely clear to me whether you are saying you don't see a conceivability argument for the multiverse hypothesis, or you do, but one that challenges the hypothesis of cosmic purpose, so I will address both issues.
We can start with the bare bones of the zombie argument, beginning with a couple of definitions: (philosophical- or p-) zombies are physically indistinguishable duplicates of living humans, differing only in having no conscious experiences, and supervenience physicalism is summarized (perhaps loosely, but quite commonly, even in the philosophical literature) by the phrase "no mental difference without physical difference."
Clearly (I think!) the possibility of zombies would directly contradict supervenience physicalism, but are zombies possible? Some philosophers are convinced that mere conceivability entails possibility [1], and zombies certainly seem conceivable, at least as the word is used in everyday language (if they were not, would the above definition be comprehensible?)
After these preliminaries, the conceivability argument can be stated as two-propositions-to-a-conclusion: conceivability entails possibility, zombies are conceivable, therefore zombies are possible and supervenience physicalism is false.
The conceivability argument for the multiverse hypothesis can be stated even more tersely, through a straightforward substitution: the multiverse is conceivable, therefore it is possible.
Moving on to whether Philip should be concerned by this argument, we can start by noting that he clearly has no problem with conceivability arguments, as he makes use of the zombie argument in debates. In addition, in the article we are discussing, he notes that the multiverse would present a significant challenge to cosmic purpose.
Are you thinking that nothing less than the non-modal, actual-world reality of the multiverse should give Philip concern? If so, then, on that basis, it would seem that he need not regard your arguments for the multiverse as a challenge to cosmic purpose, as they seek only to boost one's credence in it, not establish its reality.
Nevertheless, I feel there is a concern for Philip here. The hypothesis being debated here is that there is cosmic purpose, and, as Philip apparently acknowledges, the multiple universes hypothesis presents it with a significant challenge. As in all such cases, the burden rests on proponents of the original hypothesis to show that this challenge is unlikely to succeed, rather than on the challengers to show, beyond reasonable doubt, that it does.
In such a situation, it is not sufficient to dispute just one counter-argument; all reasonable ones should be addressed. Here, in addition to the argument for multiverses from fine-tuning, we have the argument from conceivability, and Philip’s response to the former does not appear to address the latter. The challenge for Philip here is to dispute the conceivability argument for multiple universes without using an argument that also applies, mutatis mutandis, to the zombie argument.
Nothing less than the non-modal, actual-world reality of the multiverse should give Philip concern.
But my arguments should be read as a challenge because they increase the credence that the multiverse is actual. Which then decreases the credence that there is cosmic purpose.
This sounds right to me, but something more narrow like the Fine-Tuning argument for Discoverability should still be fine and be reason to give an edge back to Cosmic purpose
I noticed in "Why?" you referenced John Leslie's fantastic "Universes". I was quite satisfied by his explanation for why the inverse gambler's fallacy doesn't refute the multiverse hypothesis.
As you know, the inverse gambler's fallacy goes like: a philosopher walks into a room in a casino and sees someone winning at roulette 20 rounds in a row. The philosopher says "ah, the casino must be busy tonight!" Supposedly, this is the same fallacy that people advocating for the multiverse make. But it isn't. Because you can ONLY be in the universe analogous to the gambler who wins 20 rounds, whereas the philosopher could have walked into any room containing any gambler.
Therefore what the multiverse argument is actually comparable to is having an alarm that rings any time a person in the whole casino wins 20 rounds of roulette. Obviously, if there are more people in the casino playing roulette, it's more likely that such an event will happen, and therefore it's more likely to ring. That's because the whole casino is relevant to the alarm, which means that the probability is going to be affected by the number of people in the casino, whereas for the philosopher stumbling into the room where the lucky gambler is, it's only the single lucky gambler that's relevant.
It's like the difference between the probability of something unlikely happening to someone among 8 billion people and the probability of something happening to just 1 person. The reason the reverse gambler fallacy isn't relevant to the multiverse argument is that if there were 8 billion universes and only 1 happened to be able to produce life, then you could ONLY be in the one that does. Therefore it's equivalent to having a buzzer that ONLY sounds when some specific unlikelihood happens to any person on earth. Obviously, the more people there are on earth, the more likely it is to buzz.
There is another way of countering the multiverse theory though, I think, which is to point out that it's kicking the can down the road. Just because you have an infinite number of universes, doesn't actually mean you avoid the necessity of cosmic purpose. Crucially, an infinite amount of something does not equal an infinite variety. If you have an infinite number of chocolate bars, one of them will not be a diamond. Similarly, even if the infinite monkey theorem were true, the monkeys would still never type out a living, breathing elephant, because it goes beyond the parameters of the infinity being conceptualised, which determines that there are just an infinite amount of different combinations of things that could be typed. So, even if there is a multiverse, it doesn't avoid cosmic purpose, because it would still be the kind of multiverse that has universes with conscious beings in them.
The Anthropic Principle is a fundamental part of the Multiverse Theory - without it, it's easy to disprove as you have often done.
White (2000, 268) dismisses the Anthropic Principle (without actually naming it): “it is not as though we were disembodied spirits, waiting for some big bang to produce a universe that could accommodate us.”
Paraphrasing White for our own universe: “it is not as though we were disembodied spirits, waiting for some planet that could accommodate us.” And yet we manage to inhabit one of the rare planets in our own universe that could support us, so White's statement is an unjustifiable assertion. If you allow it in our universe you can't really not allow it in the multiverse.
OK. The Bayes calc is reasonable, tho I have doubts that splitting weak and strong in the Multiverse case adds much since we already know the result is 1, hence the two terms have to be reciprocals which cancel each other out (as you say).
My objection is to your anthropic analogies, both of which assume our "spirit" comes from an ?infinite (or at least unemptiable) pool outside our universe (dualism, by def.) and that almost all fall on rocky ground, so my chance of existing is infinitesimal. This replaces the unlikelihood of fine-tuning by the unlikelihood of instantiation.
Any theory which postulates an infinite pool of potential spirits means your or my personal likelihood is zero. I think this completely rules out dualism, unless there's a targetting mechanism. (Similarly panpsychism has to explain how you or I beat the odds to be human rather than a sandgrain or electron.)
Now consider rainbows. How do rainbows get to Earth? Is it a Selection Effect or a Converse Selection Effect? Clearly either would be silly. They just emerge.
Maybe human awareness is just an emergent property. Maybe we invoke Occam and say all spirits are identical so there is only one. Maybe we go for Bundle Theory. Question for another day.
Firstly we already have the quantum multiverse, (whether it's literally real, or just a mathematical construct) so "multiverse" is not a novel concept. Ditto the ensemble.
Secondly, all the anthropic calculations about a multiverse containing multiple universes can equally be applied to our own universe of multiple worlds, except that here we already know there are multiple worlds. What's the chance that this world is life-supporting? Awarenesses get put in minds without incurring excessive improbability in this universe, so clearly the same mechanism could apply in the multiverse.
This Mark guy is a huge troublemaker.
Is this all about Bayesian reasoning. That's tge chapter of Why I always get stuck on. I don't think it makes sense applying it to the whole universe because (wavey hsnds) the context is undetermined (or something).
I do think the Multiverse Objection is a good objection to the Fine-Tuning Argument (FTA), but mostly because it seems like certain theories in physics (Many Worlds Interpretation of Quantum Mechanics, Inflationary Theory, Big Crunch theories, String Theory, etc.) naturally posit a multiverse.
If the above weren’t true and the only reason we posited a multiverse was to explain fine-tuning, then I do think it would be a much weaker objection (and therefore, even if it’s not decisive, the FTA would be evidence for God*).
That said, since the multiverse is still only a possibility, I think the FTA is a good argument against the rationality of atheism even if it doesn’t prove theism (i.e., it’s like an argument for agnosticism).
* Technically it’s really just an argument for design. With simulation theories, it’s always possible we live in a simulation rather than a universe designed by an omnipotent God! I’m not sure how plausible the simulation theory is, but it’s at least something to consider.
I think the inference to multiverse involves dodge reasoning, as I argue in my comments in response to Lewis and Barnes at this conference: https://sites.google.com/view/panpsychismandpanentheism/project-events/durham-workshop?authuser=0
In my Why? book I argue that if we're to go for eternal inflation but avoid this dodge reasoning, we need to postulate a multiverse in which all the universes have the same physics.
Thanks Philip,
I've replied here:
https://open.substack.com/pub/disagreeableme/p/trying-to-make-philip-goff-miserable
By replying here, I can address the main participants in this debate in one post!
For what it is worth, I agree with Philip insofar as I feel that the apparent fine-tuning of the universe is beside the point when the question is whether any multiverse hypothesis is plausible, and so it does not shift my credence on that matter one iota.
Having said that, I think there is a different argument that Philip should take seriously: the one from conceivability - multiverses are conceivable, and therefore metaphysically possible.
This is, of course, a riff on the zombie argument. As Philip endorses that argument, despite not thinking zombies are real, I feel it would be inconsistent for him to reject the corresponding argument for multiverses.
This would not, of course, rule out cosmic purpose, but, as he says above, "... it’s clear that the probability of cosmic purpose is significantly less if the multiverse is in the picture than it is if cosmic purpose is the only explanation of fine-tuning."
For what it's worth, I disagree with your agreement with Philip insofar as I feel that the apparent fine-tuning of the univere is relevant.
Philip does not mean to say that the multiverse is not conceivable (edit: or metaphysically possible). He thinks it is conceivable (edit: and presumably, at least conceivably, metaphysically possible), and has a non-negligible credence in it. That's not the point at stake.
What he means by "multiverse is in the picture" is not that the mutliverse is a viable option, but that we're considering what would follow if it were true. If it were true, then there would be much less reason to believe in cosmic purpose.
I don't doubt that what you say about Philip's position is correct, but I do not feel it goes so far as to address the issue of consistency over what conclusions can be drawn from metaphysical possibility. I see I need to be clearer in that respect.
In the case of zombies, he has often used them (for example, in his debates with Sean Carroll) to argue that the metaphysical possibility of zombies (as established by the conceivability argument) is as problematic for physicalism as their actual existence would be. Given that position, it would seem (at least to me) to be inconsistent to then avoid concluding that the metaphysical possibility of other universes (as established either by the equivalent conceivability argument or any other way) is as problematic for cosmic purpose as their actual existence would be.
The metaphysical possibility of zombies is supposedly a problem for physicalism because it demonstrates that, even conceptually, consciousness and physical behaviour are distinct, so the first cannot be reduced to the former even in principle.
I don't see a parallel argument that the metaphysical possibility of a multiverse is problematic for cosmic purpose.
Let's see if I can get it right this time!
It is not entirely clear to me whether you are saying you don't see a conceivability argument for the multiverse hypothesis, or you do, but one that challenges the hypothesis of cosmic purpose, so I will address both issues.
We can start with the bare bones of the zombie argument, beginning with a couple of definitions: (philosophical- or p-) zombies are physically indistinguishable duplicates of living humans, differing only in having no conscious experiences, and supervenience physicalism is summarized (perhaps loosely, but quite commonly, even in the philosophical literature) by the phrase "no mental difference without physical difference."
Clearly (I think!) the possibility of zombies would directly contradict supervenience physicalism, but are zombies possible? Some philosophers are convinced that mere conceivability entails possibility [1], and zombies certainly seem conceivable, at least as the word is used in everyday language (if they were not, would the above definition be comprehensible?)
After these preliminaries, the conceivability argument can be stated as two-propositions-to-a-conclusion: conceivability entails possibility, zombies are conceivable, therefore zombies are possible and supervenience physicalism is false.
The conceivability argument for the multiverse hypothesis can be stated even more tersely, through a straightforward substitution: the multiverse is conceivable, therefore it is possible.
Moving on to whether Philip should be concerned by this argument, we can start by noting that he clearly has no problem with conceivability arguments, as he makes use of the zombie argument in debates. In addition, in the article we are discussing, he notes that the multiverse would present a significant challenge to cosmic purpose.
Are you thinking that nothing less than the non-modal, actual-world reality of the multiverse should give Philip concern? If so, then, on that basis, it would seem that he need not regard your arguments for the multiverse as a challenge to cosmic purpose, as they seek only to boost one's credence in it, not establish its reality.
Nevertheless, I feel there is a concern for Philip here. The hypothesis being debated here is that there is cosmic purpose, and, as Philip apparently acknowledges, the multiple universes hypothesis presents it with a significant challenge. As in all such cases, the burden rests on proponents of the original hypothesis to show that this challenge is unlikely to succeed, rather than on the challengers to show, beyond reasonable doubt, that it does.
In such a situation, it is not sufficient to dispute just one counter-argument; all reasonable ones should be addressed. Here, in addition to the argument for multiverses from fine-tuning, we have the argument from conceivability, and Philip’s response to the former does not appear to address the latter. The challenge for Philip here is to dispute the conceivability argument for multiple universes without using an argument that also applies, mutatis mutandis, to the zombie argument.
[1] Chalmers, "Does Conceivability Entail Possibility?" ( https://consc.net/papers/conceivability.html )
Nothing less than the non-modal, actual-world reality of the multiverse should give Philip concern.
But my arguments should be read as a challenge because they increase the credence that the multiverse is actual. Which then decreases the credence that there is cosmic purpose.
Oops! I posted a reply to a different question here by mistake - I will make a relevant reply soon.
This sounds right to me, but something more narrow like the Fine-Tuning argument for Discoverability should still be fine and be reason to give an edge back to Cosmic purpose
I noticed in "Why?" you referenced John Leslie's fantastic "Universes". I was quite satisfied by his explanation for why the inverse gambler's fallacy doesn't refute the multiverse hypothesis.
As you know, the inverse gambler's fallacy goes like: a philosopher walks into a room in a casino and sees someone winning at roulette 20 rounds in a row. The philosopher says "ah, the casino must be busy tonight!" Supposedly, this is the same fallacy that people advocating for the multiverse make. But it isn't. Because you can ONLY be in the universe analogous to the gambler who wins 20 rounds, whereas the philosopher could have walked into any room containing any gambler.
Therefore what the multiverse argument is actually comparable to is having an alarm that rings any time a person in the whole casino wins 20 rounds of roulette. Obviously, if there are more people in the casino playing roulette, it's more likely that such an event will happen, and therefore it's more likely to ring. That's because the whole casino is relevant to the alarm, which means that the probability is going to be affected by the number of people in the casino, whereas for the philosopher stumbling into the room where the lucky gambler is, it's only the single lucky gambler that's relevant.
It's like the difference between the probability of something unlikely happening to someone among 8 billion people and the probability of something happening to just 1 person. The reason the reverse gambler fallacy isn't relevant to the multiverse argument is that if there were 8 billion universes and only 1 happened to be able to produce life, then you could ONLY be in the one that does. Therefore it's equivalent to having a buzzer that ONLY sounds when some specific unlikelihood happens to any person on earth. Obviously, the more people there are on earth, the more likely it is to buzz.
There is another way of countering the multiverse theory though, I think, which is to point out that it's kicking the can down the road. Just because you have an infinite number of universes, doesn't actually mean you avoid the necessity of cosmic purpose. Crucially, an infinite amount of something does not equal an infinite variety. If you have an infinite number of chocolate bars, one of them will not be a diamond. Similarly, even if the infinite monkey theorem were true, the monkeys would still never type out a living, breathing elephant, because it goes beyond the parameters of the infinity being conceptualised, which determines that there are just an infinite amount of different combinations of things that could be typed. So, even if there is a multiverse, it doesn't avoid cosmic purpose, because it would still be the kind of multiverse that has universes with conscious beings in them.
Leslie was one of the first people to respond to this stuff, but the debate has moved on since then. Check out my response to Barnes and Lewis in this conference https://sites.google.com/view/panpsychismandpanentheism/project-events/durham-workshop?authuser=0
The Anthropic Principle is a fundamental part of the Multiverse Theory - without it, it's easy to disprove as you have often done.
White (2000, 268) dismisses the Anthropic Principle (without actually naming it): “it is not as though we were disembodied spirits, waiting for some big bang to produce a universe that could accommodate us.”
Paraphrasing White for our own universe: “it is not as though we were disembodied spirits, waiting for some planet that could accommodate us.” And yet we manage to inhabit one of the rare planets in our own universe that could support us, so White's statement is an unjustifiable assertion. If you allow it in our universe you can't really not allow it in the multiverse.
You need to engage with the Total Evidence Requirement stuff, which Draper et al 2007 builds on. I lay this out in terms of anthropic analogies and the maths in my responses to Lewis and Barnes in this conference: https://sites.google.com/view/panpsychismandpanentheism/project-events/durham-workshop?authuser=0
OK. The Bayes calc is reasonable, tho I have doubts that splitting weak and strong in the Multiverse case adds much since we already know the result is 1, hence the two terms have to be reciprocals which cancel each other out (as you say).
My objection is to your anthropic analogies, both of which assume our "spirit" comes from an ?infinite (or at least unemptiable) pool outside our universe (dualism, by def.) and that almost all fall on rocky ground, so my chance of existing is infinitesimal. This replaces the unlikelihood of fine-tuning by the unlikelihood of instantiation.
Any theory which postulates an infinite pool of potential spirits means your or my personal likelihood is zero. I think this completely rules out dualism, unless there's a targetting mechanism. (Similarly panpsychism has to explain how you or I beat the odds to be human rather than a sandgrain or electron.)
Now consider rainbows. How do rainbows get to Earth? Is it a Selection Effect or a Converse Selection Effect? Clearly either would be silly. They just emerge.
Maybe human awareness is just an emergent property. Maybe we invoke Occam and say all spirits are identical so there is only one. Maybe we go for Bundle Theory. Question for another day.
Two other points:
Firstly we already have the quantum multiverse, (whether it's literally real, or just a mathematical construct) so "multiverse" is not a novel concept. Ditto the ensemble.
Secondly, all the anthropic calculations about a multiverse containing multiple universes can equally be applied to our own universe of multiple worlds, except that here we already know there are multiple worlds. What's the chance that this world is life-supporting? Awarenesses get put in minds without incurring excessive improbability in this universe, so clearly the same mechanism could apply in the multiverse.