I like Mike. But if you're going to talk about what's conceivable, then his rules for personal identity fail for identities in conceivable universes.
Consider Riker from ST:TNG. A transporter accident (pre-series) creates two copies of him: one on the Potemkin and one on Nervala IV. These copies are both William T. Riker (or as much so as any transporter-copy is). The version on the Potemkin goes on to become the Riker we meet in season 1, while the other version doesn't appear until series 6.
The two copies of Riker are both continuations of the Riker who was sent down to Nervala IV. They both have the same personal identity as the pre-away-team Riker. Nevertheless they have distinct personal identities afterwards. We have three identifiers:
(R1) Riker before the mission to Nervala IV.
(R2p) Riker on the Potemkin post mission to Nervala IV.
(R2n) Riker on Nervala IV after the Potemkin left.
R2p has the same personal identity as R1; R2n has the same personal identity as R1; but R2p and R2n have distinct personal identities. Personal identity is (conceivably!) not transitive. (Also not 1-to-1, since R1 is identical to both R2p and R2n, who are two beings.)
Time travel into your own lifetime shows this as well -- the present-me and the future-me have the same personal identity but also different personal identities.
Also cloning, as typically presented in science fiction. (Tho' at least here there's usually an original and a copy, and the original is typically taken to embody the "real" identity.)
Of course there are no Star Trek transporters or time travel or SF-style cloning in our universe, so Huemer's rules are actually fine for any real-life theory of personal identity.
Then again, Freaky-Friday type scenarios also don't happen in our universe, so the Ronnie Coleman story is just that -- something we can conceive of, but not evidence of the substance-hood of consciousness in reality.
One more thing. It would be possible to avoid the conceivability arguments above by holding that personal identies are not extended in time. You wouldn't have crapped your pants 21 years ago because that was an entirely different being. BONUS: no need to apologize for THEIR act of blaming Ellis. But then you'd be a completely different being now than the one that wrote the article I'm responding to. No chance of any two beings having anything like a conversation. Huemer should have noted that any reasonable theory of personal identities allows them to be extended in time.
I think in the Ryker scenario, I would rather say that neither of them are Ryker than abandon transitivity. Although, I suppose that commits me to saying Ryker is annihilated on being beamed, but that seems plausible to me anyway (not read much in that space though). Still, I would need to have very strong seeming that a counter example worked before I abandon the transitivity rule (or any of them, for that matter)
As far as the Ronnie example goes, it's not supposed to be evidence of souls per se, but rather a counter example to the body theory (also, I was using to show how personal identity is a hard question to answer)
And yes, that sure would be a nice out on the pant crapping front, but I think there is some sense in which our identity persists over time. Sadly, those were MY pants I soiled (I was just a kid, I swear!). I think having an identity persist over time is another plausible requirement for a good theory of PI
I actually lean toward the "teleport is murder" view myself. Drove one guy crazy that I wouldn't let him kill me in my sleep and replace me with a perfect copy and give the copy $100.
I say "lean toward" because I don't think personal identity is as simple as it's made out to be. Huemer doesn't mention this in his rules, but identity functions are also supposed to be commutative -- if x is identical to y, then y is identical to x. But in the teleportation universe, I'd say that R2p and R2n are both personally identical to R1, but R1 is personally identical to neither R2. R1 died in the transporter, and two new beings were created with copies of R1's personal identity. To make a computing analogy, R1 and either R2 are identical **by value** but not **by reference**. From **anyone else's** point of view, R1 and R2p (R2n) are the same person, but R1's p.o,.v. ended when the transporter activated (that variable has been garbage-collected). The ST:TNG episode shows how people struggle to accept R2n as personally identical to R1 given that they've already accepted R1p as p.i. to R1. Troi in particular struggles with the fact that R2n is still in love with her, even tho' she and R2p have moved on.
But here's another situation to consider -- one that gave Huemer pause. Suppose that instead of cloning by building an entirely new body as a copy of the original, the original is split down the middle (left/right) and the two halves are regrown into identical copies of the original. (Suppose that the two halves are kept alive but sedated thru the process.) Would you say that the original p.i. was annihilated? Would you arbitrarily take one to be the continuation (old p.i.) and the other to be the copy (new p.i.)? Or would you then give up on transitivity?
2. Say Ryker is still annihilated, and the other two aren't him. This feels wrong though, because if we only grew one side, then that would still be him - so that side can't lose their identity just from us growing the other side
I would be more prepared to reject transitivity than to reject that identity is extrinsic. However, couldn't we still say that the one that is still Ryker is just whichever one has his soul still? Isn't that the whole appeal of the soul view? That it helps us navigate these tricky situations without breaking very plausible rules?
Now, you might ask why it is that one side gets the soul and the other doesn't - and the answer would be I have no idea. But that's a separate question. Still, I'm definitely closer to rejecting transitivity now than when I wrote the article, seems vulnerable!
I suppose I could ask you which one got the soul, but I think I'd rather ask whether you think the soul is indivisible. We can certainly conceive of a soul being broken into pieces (see Voldemort). We can also conceive of one person's mind being in multiple people at the same time (see Lucrezia in Girl Genius or the Master in that Dr. Who story line where he was PM). Is the soul neither indivisible nor duplicatable?
At this point I think I'd like to mention the importance of the phrase "I don't care." None of these things we're discussing has ever happened, and I don't think any of them will ever happen. In such cases I try to cultivate the attitude of not caring what other people's theories say about those things. (Still, it's kind of fun to discuss those things, so thatnks for obliging!)
Yeah, I'm operating under the belief that a soul is indivisible - but if it isn't then it fails as a solution to personal identity. Just kicks the van one step down the road
Happy to oblige haha! Don't think I've ever chatted philosophy with such a well of relevant pop culture references lol
> But, if she knew all the physical facts prior, and has just learned something new, that must mean the conscious experience of seeing red isn’t physical.
I feel like this example exhibits a something that gives me a lot of discomfort in making hypothetical arguments.
In this example, Mary knows *literally everything* about seeing the color red. Every interaction of neurons, every wavelength interaction, every interaction between particles in the air and the retina and the deep brain structures that give rise to the sensation of “color” specifically in the brain.
I could go on and on and on. Really! Knowing this much information is *incomprehensible* to the ordinary human mind. So this example posits an intelligence that is superhuman. Very very superhuman.
I would say that a more fair hypothetical is this: Let’s say we have a redless Demiurge. It created all of the material universe from its primal desire and fine tuned all laws of nature as it pleased. Except it made sure that its own consciousness could never perceive the color red. After this creature creates the universe and observes it in its entirety for 13.8 billion years, it decides to give itself the power to see red.
Has it learned anything new? I think that the intuition of “yes it has” is *far* less strong.
I don't really see an issue with Mary's room. Even though learning all the physical facts is hard to comprehend, it just seems impossible that she can ever know what it's like to see red from inside the black room. Obviously, we don't know all the physical facts, but we know a lot of them, and they seem totally unrelated to our understanding what it's like to see colour.
Still, even if we use your example, I still have the same intuition. Definitely seems like the God learns something new. Doesn't really change it for me
I don't find the arguments for immateriality convincing. Your first argument seems to be that it's a category error to call consciousness physical, but why? Nothing seems ill-formed to me about this statement. I don't see why the property of being physical is automatically precluded from applying to consciousness or why consciousness is automatically precluded from being applied to physical things. Your reasoning is that you have an intuition that consciousness isn't physical, but I think this intuition can be debunked: It results from a difference in how you experience consciousness and other physical things, rather than a difference in their actual substance. Specifically, you observe other physical things indirectly, from the outside. But you observe your consciousness from the inside - you directly feel what it's like to be you. When you observe physical things, you don't feel what it's like to be them because you're not them. But that doesn't mean there is no what-it's-likeness there.
But what about observing your own brain as a physical object? If my brain *is* me, then shouldn't this argument not apply to it, since I do observe it in the first-person way? Well, no. When you talk about your brain, you are talking about it as viewed from the outside. You are imagining it the way it would look to a neurosurgeon if they cut open your skull and looked inside, or the way it looks to an MRI scan. And when imagining it this way, you don't feel the what-it's-likeness that is only felt from actually being the brain. There can be a difference between brain qua externally observable physical object and brain qua mind without abandoning the fact that the mind is (a process running on) the brain.
Mary's Room is really just a result of human psychology - how our brain organizes and stores information - rather than metaphysics. We have multiple kinds of memory and as such multiple kinds of knowledge. Two of the types of memory are semantic memory and episodic memory. All of the facts Mary learns in the room would contribute to her semantic memory, but, having never experienced the color red for herself, they wouldn't give her an episodic memory of seeing the color. Once she does finally see red, she will gain an episodic memory of what the color looks like, and this allows her to simulate the color later on, to see it in her mind's eye. There's no reason that merely propositional knowledge about how the human visual system works would automatically allow her to do this, even if there are no further facts about redness.
As an analogy, imagine you created a program that ran an advanced AI (not necessarily conscious) and would access a certain folder on the computer running it if you gave it certain inputs. You could teach the AI what its exact source code was and show it how that code could be used to access the folder given some external inputs, but that still wouldn't actually lead the AI to access the folder. The same thing is true with Mary in her room: She is learning how her visual system works and how it experiences red given certain external stimuli, but that doesn't automatically cause her to experience red. She still has to actually get those external stimuli to do it.
For p-zombies, I will first note that you can't believe in them if you're an interactionist dualist. After all, under interactionist dualism, there would be a difference in our physical bodies if they weren't conscious - namely, none of the effects of the interaction with our conscious mind would occur. So accepting the possibility of a philosophical zombie requires epiphenomalism, which almost everyone agrees is extremely implausible - it would imply that you don't have free will, that your consciousness has no effect on your actions, so that it's a bizzare coincidence that you just happen to talk about consciousness anyway, etc. Surely, no matter how obvious it seems that p-zombies are metaphysically possible, it's even more obvious that epiphenominalism is false, so you will have to reject p-zombies.
But there is more we can say about why p-zombies are impossible. Although conceivability is usually a reliable guide to metaphysical possibility, there are some well-known cases where it fails. For instance, it was conceivable for people living before modern chemistry that water was not composed of tiny molecules containing two hydrogen atoms and one oxygen atom. They could imagine water being a continuous, infinitely divisible substance, and even we can, despite knowing this is false (this is how we usually model it in fluid dynamics). But it's not metaphysically possible for water not to be H2O, because that's just what water *is*. Any substance that was similar to water on the macroscopic scale but didn't have the chemical formula H2O wouldn't be water - it would be something else.
Another example: It was conceivable that the Morning Star and the Evening Star were two separate objects - after all, that's what everyone thought they were for most of history. But it turns out that both expressions "The Morning Star" and "The Evening Star" refer to the same object - Venus. Is it metaphysically possible that Venus not be Venus? Surely not. So, it can't be metaphysically possible that the Morning Star is the Evening Star, at least not if those names are taken to be rigid designators.
Both of these are cases where rigidly designating identity claims are conceivably false but in fact necessarily true. But this is exactly the kind of claim for which the philosophical zombie argument asks us to infer possibility from conceivability! We're asked to conceive of a being that is physically identical to us but not mentally identical, and conclude that such a being is possible. But both of these identities (physical identity and mental identity) must be based on rigid designators for the argument to work - otherwise, you could be imagining a being that isn't really physically identical to the way you are in the actual world (just as imagining a substance that seems watery on a macroscopic scale but isn't H2O isn't really identical to the way water is in the actual world) or one that is identical and does have a mind but you just don't realize it (like imagining that H2O isn't water).
Nice article. Interesting points about consciousness being immaterial and a substance. You might be interested in exploring the philosophy of Advaita Vedanta. It offers an answer to "What is the Self?" via inquiry based on experience and reason. Its conclusions might seem a little woo-woo at first to some, but it doesn't rely on wild metaphysical claims and (based on what I've learnt so far, which isn't a whole lot) seems philosophically sound. Swami Sarvapriyananda's has some great talks about it.
I like Mike. But if you're going to talk about what's conceivable, then his rules for personal identity fail for identities in conceivable universes.
Consider Riker from ST:TNG. A transporter accident (pre-series) creates two copies of him: one on the Potemkin and one on Nervala IV. These copies are both William T. Riker (or as much so as any transporter-copy is). The version on the Potemkin goes on to become the Riker we meet in season 1, while the other version doesn't appear until series 6.
The two copies of Riker are both continuations of the Riker who was sent down to Nervala IV. They both have the same personal identity as the pre-away-team Riker. Nevertheless they have distinct personal identities afterwards. We have three identifiers:
(R1) Riker before the mission to Nervala IV.
(R2p) Riker on the Potemkin post mission to Nervala IV.
(R2n) Riker on Nervala IV after the Potemkin left.
R2p has the same personal identity as R1; R2n has the same personal identity as R1; but R2p and R2n have distinct personal identities. Personal identity is (conceivably!) not transitive. (Also not 1-to-1, since R1 is identical to both R2p and R2n, who are two beings.)
Time travel into your own lifetime shows this as well -- the present-me and the future-me have the same personal identity but also different personal identities.
Also cloning, as typically presented in science fiction. (Tho' at least here there's usually an original and a copy, and the original is typically taken to embody the "real" identity.)
Of course there are no Star Trek transporters or time travel or SF-style cloning in our universe, so Huemer's rules are actually fine for any real-life theory of personal identity.
Then again, Freaky-Friday type scenarios also don't happen in our universe, so the Ronnie Coleman story is just that -- something we can conceive of, but not evidence of the substance-hood of consciousness in reality.
One more thing. It would be possible to avoid the conceivability arguments above by holding that personal identies are not extended in time. You wouldn't have crapped your pants 21 years ago because that was an entirely different being. BONUS: no need to apologize for THEIR act of blaming Ellis. But then you'd be a completely different being now than the one that wrote the article I'm responding to. No chance of any two beings having anything like a conversation. Huemer should have noted that any reasonable theory of personal identities allows them to be extended in time.
I think in the Ryker scenario, I would rather say that neither of them are Ryker than abandon transitivity. Although, I suppose that commits me to saying Ryker is annihilated on being beamed, but that seems plausible to me anyway (not read much in that space though). Still, I would need to have very strong seeming that a counter example worked before I abandon the transitivity rule (or any of them, for that matter)
As far as the Ronnie example goes, it's not supposed to be evidence of souls per se, but rather a counter example to the body theory (also, I was using to show how personal identity is a hard question to answer)
And yes, that sure would be a nice out on the pant crapping front, but I think there is some sense in which our identity persists over time. Sadly, those were MY pants I soiled (I was just a kid, I swear!). I think having an identity persist over time is another plausible requirement for a good theory of PI
I actually lean toward the "teleport is murder" view myself. Drove one guy crazy that I wouldn't let him kill me in my sleep and replace me with a perfect copy and give the copy $100.
I say "lean toward" because I don't think personal identity is as simple as it's made out to be. Huemer doesn't mention this in his rules, but identity functions are also supposed to be commutative -- if x is identical to y, then y is identical to x. But in the teleportation universe, I'd say that R2p and R2n are both personally identical to R1, but R1 is personally identical to neither R2. R1 died in the transporter, and two new beings were created with copies of R1's personal identity. To make a computing analogy, R1 and either R2 are identical **by value** but not **by reference**. From **anyone else's** point of view, R1 and R2p (R2n) are the same person, but R1's p.o,.v. ended when the transporter activated (that variable has been garbage-collected). The ST:TNG episode shows how people struggle to accept R2n as personally identical to R1 given that they've already accepted R1p as p.i. to R1. Troi in particular struggles with the fact that R2n is still in love with her, even tho' she and R2p have moved on.
But here's another situation to consider -- one that gave Huemer pause. Suppose that instead of cloning by building an entirely new body as a copy of the original, the original is split down the middle (left/right) and the two halves are regrown into identical copies of the original. (Suppose that the two halves are kept alive but sedated thru the process.) Would you say that the original p.i. was annihilated? Would you arbitrarily take one to be the continuation (old p.i.) and the other to be the copy (new p.i.)? Or would you then give up on transitivity?
Intrinsic*
Hmmm, I'm not sure. We'd have a few options
1. Reject transitivity, like you say.
2. Say Ryker is still annihilated, and the other two aren't him. This feels wrong though, because if we only grew one side, then that would still be him - so that side can't lose their identity just from us growing the other side
I would be more prepared to reject transitivity than to reject that identity is extrinsic. However, couldn't we still say that the one that is still Ryker is just whichever one has his soul still? Isn't that the whole appeal of the soul view? That it helps us navigate these tricky situations without breaking very plausible rules?
Now, you might ask why it is that one side gets the soul and the other doesn't - and the answer would be I have no idea. But that's a separate question. Still, I'm definitely closer to rejecting transitivity now than when I wrote the article, seems vulnerable!
I suppose I could ask you which one got the soul, but I think I'd rather ask whether you think the soul is indivisible. We can certainly conceive of a soul being broken into pieces (see Voldemort). We can also conceive of one person's mind being in multiple people at the same time (see Lucrezia in Girl Genius or the Master in that Dr. Who story line where he was PM). Is the soul neither indivisible nor duplicatable?
At this point I think I'd like to mention the importance of the phrase "I don't care." None of these things we're discussing has ever happened, and I don't think any of them will ever happen. In such cases I try to cultivate the attitude of not caring what other people's theories say about those things. (Still, it's kind of fun to discuss those things, so thatnks for obliging!)
Yeah, I'm operating under the belief that a soul is indivisible - but if it isn't then it fails as a solution to personal identity. Just kicks the van one step down the road
Happy to oblige haha! Don't think I've ever chatted philosophy with such a well of relevant pop culture references lol
Better call soul
Literally was my back up subtitle lol
> But, if she knew all the physical facts prior, and has just learned something new, that must mean the conscious experience of seeing red isn’t physical.
I feel like this example exhibits a something that gives me a lot of discomfort in making hypothetical arguments.
In this example, Mary knows *literally everything* about seeing the color red. Every interaction of neurons, every wavelength interaction, every interaction between particles in the air and the retina and the deep brain structures that give rise to the sensation of “color” specifically in the brain.
I could go on and on and on. Really! Knowing this much information is *incomprehensible* to the ordinary human mind. So this example posits an intelligence that is superhuman. Very very superhuman.
I would say that a more fair hypothetical is this: Let’s say we have a redless Demiurge. It created all of the material universe from its primal desire and fine tuned all laws of nature as it pleased. Except it made sure that its own consciousness could never perceive the color red. After this creature creates the universe and observes it in its entirety for 13.8 billion years, it decides to give itself the power to see red.
Has it learned anything new? I think that the intuition of “yes it has” is *far* less strong.
I don't really see an issue with Mary's room. Even though learning all the physical facts is hard to comprehend, it just seems impossible that she can ever know what it's like to see red from inside the black room. Obviously, we don't know all the physical facts, but we know a lot of them, and they seem totally unrelated to our understanding what it's like to see colour.
Still, even if we use your example, I still have the same intuition. Definitely seems like the God learns something new. Doesn't really change it for me
I don't find the arguments for immateriality convincing. Your first argument seems to be that it's a category error to call consciousness physical, but why? Nothing seems ill-formed to me about this statement. I don't see why the property of being physical is automatically precluded from applying to consciousness or why consciousness is automatically precluded from being applied to physical things. Your reasoning is that you have an intuition that consciousness isn't physical, but I think this intuition can be debunked: It results from a difference in how you experience consciousness and other physical things, rather than a difference in their actual substance. Specifically, you observe other physical things indirectly, from the outside. But you observe your consciousness from the inside - you directly feel what it's like to be you. When you observe physical things, you don't feel what it's like to be them because you're not them. But that doesn't mean there is no what-it's-likeness there.
But what about observing your own brain as a physical object? If my brain *is* me, then shouldn't this argument not apply to it, since I do observe it in the first-person way? Well, no. When you talk about your brain, you are talking about it as viewed from the outside. You are imagining it the way it would look to a neurosurgeon if they cut open your skull and looked inside, or the way it looks to an MRI scan. And when imagining it this way, you don't feel the what-it's-likeness that is only felt from actually being the brain. There can be a difference between brain qua externally observable physical object and brain qua mind without abandoning the fact that the mind is (a process running on) the brain.
Mary's Room is really just a result of human psychology - how our brain organizes and stores information - rather than metaphysics. We have multiple kinds of memory and as such multiple kinds of knowledge. Two of the types of memory are semantic memory and episodic memory. All of the facts Mary learns in the room would contribute to her semantic memory, but, having never experienced the color red for herself, they wouldn't give her an episodic memory of seeing the color. Once she does finally see red, she will gain an episodic memory of what the color looks like, and this allows her to simulate the color later on, to see it in her mind's eye. There's no reason that merely propositional knowledge about how the human visual system works would automatically allow her to do this, even if there are no further facts about redness.
As an analogy, imagine you created a program that ran an advanced AI (not necessarily conscious) and would access a certain folder on the computer running it if you gave it certain inputs. You could teach the AI what its exact source code was and show it how that code could be used to access the folder given some external inputs, but that still wouldn't actually lead the AI to access the folder. The same thing is true with Mary in her room: She is learning how her visual system works and how it experiences red given certain external stimuli, but that doesn't automatically cause her to experience red. She still has to actually get those external stimuli to do it.
For p-zombies, I will first note that you can't believe in them if you're an interactionist dualist. After all, under interactionist dualism, there would be a difference in our physical bodies if they weren't conscious - namely, none of the effects of the interaction with our conscious mind would occur. So accepting the possibility of a philosophical zombie requires epiphenomalism, which almost everyone agrees is extremely implausible - it would imply that you don't have free will, that your consciousness has no effect on your actions, so that it's a bizzare coincidence that you just happen to talk about consciousness anyway, etc. Surely, no matter how obvious it seems that p-zombies are metaphysically possible, it's even more obvious that epiphenominalism is false, so you will have to reject p-zombies.
But there is more we can say about why p-zombies are impossible. Although conceivability is usually a reliable guide to metaphysical possibility, there are some well-known cases where it fails. For instance, it was conceivable for people living before modern chemistry that water was not composed of tiny molecules containing two hydrogen atoms and one oxygen atom. They could imagine water being a continuous, infinitely divisible substance, and even we can, despite knowing this is false (this is how we usually model it in fluid dynamics). But it's not metaphysically possible for water not to be H2O, because that's just what water *is*. Any substance that was similar to water on the macroscopic scale but didn't have the chemical formula H2O wouldn't be water - it would be something else.
Another example: It was conceivable that the Morning Star and the Evening Star were two separate objects - after all, that's what everyone thought they were for most of history. But it turns out that both expressions "The Morning Star" and "The Evening Star" refer to the same object - Venus. Is it metaphysically possible that Venus not be Venus? Surely not. So, it can't be metaphysically possible that the Morning Star is the Evening Star, at least not if those names are taken to be rigid designators.
Both of these are cases where rigidly designating identity claims are conceivably false but in fact necessarily true. But this is exactly the kind of claim for which the philosophical zombie argument asks us to infer possibility from conceivability! We're asked to conceive of a being that is physically identical to us but not mentally identical, and conclude that such a being is possible. But both of these identities (physical identity and mental identity) must be based on rigid designators for the argument to work - otherwise, you could be imagining a being that isn't really physically identical to the way you are in the actual world (just as imagining a substance that seems watery on a macroscopic scale but isn't H2O isn't really identical to the way water is in the actual world) or one that is identical and does have a mind but you just don't realize it (like imagining that H2O isn't water).
Nice article. Interesting points about consciousness being immaterial and a substance. You might be interested in exploring the philosophy of Advaita Vedanta. It offers an answer to "What is the Self?" via inquiry based on experience and reason. Its conclusions might seem a little woo-woo at first to some, but it doesn't rely on wild metaphysical claims and (based on what I've learnt so far, which isn't a whole lot) seems philosophically sound. Swami Sarvapriyananda's has some great talks about it.
Thanks for the recc! Yeah, I've always had a passing interest in Hinduism, so sounds up my alley