294. Sabine Hossenfelder — Existential Physics (6)
2 (1h 1m 33s):
Yeah, again, that's a pretty big if, if you could upload yourself to a computer, but you can't sorry. So I I'm afraid that at this point I kind of lose interest, but I don't see anything that fundamentally speaks against it. There's this little issue that strictly speaking, you can't copy a quantum state without destroying the original, which would seem to say that if you copy yourself identically into whatever kind of computer, you have to destroy yourself first. So you can't really be in two places at the same time that may or may not be relevant to consciousness.
2 (1h 2m 18s):
It's kind of unclear how big of a role quantum effect quantum effects actually play in the brain. So I'm quite sympathetic to the idea that you can actually copy consciousness onto a computer though. This is actually not what, what this chapter is about. I think it's more about, are there copies of us in the many worlds interpretation of quantum mechanics? Yeah. To which again, the answer is to which the answer is, well, it, again, it's a scientific, you, you can believe it if you want to, but there's no evidence that speaks for or against it.
1 (1h 2m 55s):
Yeah. This would be the, in some other university, you and I are having this conversation, but something is just slightly different. You're blonde and I have more hair or I'm taller and, and, and, and you're, and you're a professional rock singer rather than a physicist or something like that. I saw some video where you were singing, something like that. You know, there's just an infinite number of versions of us.
2 (1h 3m 18s):
Yeah. So if you believe the multiverse is real, then you have to believe that this is possible.
1 (1h 3m 25s):
I just find that w what did you call it? The money Python test, you know, it's just, you just have you just laugh at it, like, okay, what do I do with that?
2 (1h 3m 36s):
But it wasn't a, my terminology. I think it was Seth Lloyd.
1 (1h 3m 40s):
Hmm. Right. But again, these thought experiments, this is always kind of curious to me how somebody smart can take an idea like this and run with it. You know, like the, are we living in a simulation argument? You know, so the Copernican principle, we're not special. We're probably not first, it'd be civilization out of us. How far ahead of us would they be? It's not going to be 10 years or 10,000 years can be like 10 million years or a billion years. And by then they would have created what we've created a form of computers. They will have created these giant virtual realities where they could, you know, create entire universes. It's more likely than not that we're one of those. And it's all of a sudden, you're like, wait, what, how many steps did you have to go through there if then if then if then, boom.
2 (1h 4m 27s):
Yeah. Well, so for me, the most important part of the argument is entirely left out, which is the question, can you actually simulate our reality? The way that we observe it on any kind of computer to which the answer is, we don't know. And I, I kind of feel like the people who lead this argument and insist that it's somehow scientific that cheating to people. Because if you, if you knew how to reproduce all the observations that we have made today on some kind of computer, that that'd be a theory of everything. And I would really like to see it. We, we don't know how to do it. Like we D we, we don't even know exactly how to simulate climate change precisely we can do it approximately, but how we do it precisely, you know, there's this whole issue with chaos and how you truncate the equations.
2 (1h 5m 22s):
And then the people who lead this argument, they're always like, oh, we don't have to worry about the computational capacity of those computers, because there are so many parts in the universe where we can just stop stimulating them because no one's looking there. So it doesn't really matter. And I'm like, okay, well then you have to tell me, how do you know when someone is going to, to look there and how do you fill in the blank so that there'll never be an inconsistency that shows up. And of course those people don't actually have an algorithm. So, I mean, don't get me wrong. I, I, there's a certain appeal to the idea as a kind of a thought argument. And I think there's a philosopher, David Charmaz, he's recently written a book about it.
2 (1h 6m 8s):
So from a philosophical perspective, this is all well and fine with me. You know, there's certain, I haven't really read the book, but I, I have no doubt that there are moral questions that you can discuss about, you know, if you were to simulate beings and some kind of computer, are you responsible for them? That kind of staff philosophers can certainly write books about it is all fine with me. But the problem starts when they argue that this actually pretty much follows from science, which is how a lot of them put it. And I, and I am afraid I have to disagree because we don't know how to do it.
1 (1h 6m 46s):
Yeah. I had David on the show, I read the book, it's 500 pages of, of what a virtual world would be like, essentially. He said he tackles all the great questions in philosophy in a virtual world. Although even he admits right there in the book and on the show, none of this is testable. I have no idea if any of this is real and we never will know. It's like, okay, then what are we talking about? You know, sort of a Harry Potter world, you know, we're just making it up. Okay. It's fun. But it's not really science. I think you would say, so here, you know, it kind of gets to the problem of who you are and the replication issue. I dealt with this in my book, heavens on earth, like, you know, the copy of you, let's say instead of after you're dead, the copy is uploaded, but we have a sophisticated MRI machine and we slide you in it.
1 (1h 7m 37s):
And we make a copy of Sabina, Haas, and Felder, and then slide you out. And then, and then say, look, there you are, you're up in the cloud. And you'd say, no, I'm standing right here. This is me. Right? So this is the distinction between the memory self mem self, which is just the copy of everything you've ever thought and all your emotions, memories, and so on. And then the point of view, self, the POV self of you looking through your eyes, experiencing the world from one moment to the next. And when people think of mortality, they don't think of a copy of me going somewhere. They think I close my eyes for the last time and I opened them up and I'm in this new, it's still me looking out at the world.
1 (1h 8m 18s):
Right. And people seem confused about this. Like, well, the, the copy is going to be me. It's not going to be you. You will, you'll have no experience of the copy.
2 (1h 8m 28s):
Yeah. Well, depends on what you mean by you. Yeah. So I don't discuss this in the book because I don't really have anything to add. I made a video body accorded, the transmitter problem. So what, what happens to Kirk when he goes through the transmitter, you make a copy of her mother and you destroy the original. Did he die or did he not die? And I, I don't have a good answer to it. I mean, I think there are arguments for, for either side. And I don't know. I mean, maybe it's just, it just a question that doesn't make any sense to us because we have this insight perspective to consciousness, kind of like we have this insight perspective to the moment of now.
2 (1h 9m 13s):
So maybe that's where the problem lies, but, but really, I don't know.
1 (1h 9m 18s):
Yeah. Well, I don't think anybody knows again, it's just kind of one of those thought experiments that you run out and see how far you can go with it. To me, it feels like here I quote Woody Allen, you know, I don't want to live on through the, the memories of, of my countrymen and I want to live on in my apartment. Right. I, I want to just keep going, you know? So something like cryonics where you're frozen and then you you're woken back up and it's just like a long sleep. That would be a POV, self continuation, but uploading to the cloud, any kind of digital copy, unless you mean by you just, you know, there's multiple copies of you, but that can't be you in any sense that we mean it traditionally, because it would be like your twin, your twin, the moment the split happens, the copy is made that the copy starts to leave lead a slightly different life than yours.
1 (1h 10m 13s):
Right. And they have a different life path, different memories, different experiences. That's not you,
2 (1h 10m 18s):
But of course, this brings up the question, like if your path splits, which one do you take?
1 (1h 10m 25s):
Mm Hmm. Right, right. Yeah. So there you'd have to define you as more than just you, right. It'd be kind of a new definition of you. Yeah. All right. Let's hit a couple more of the big existential questions here. I had Tim Palmer on the podcast last week, although we will release this after this one, because his book comes out October 18th. So we'll release it on a pub date, but he argues for freewill or, or sort of a compatible ism through chaos theory, which I agree with. And you disagree with that. So can you give us your best argument for why freewill doesn't exist or even compatible ism and your determinants?
2 (1h 11m 5s):
Well, I just don't find free world useful. The issue with talking about free will is that no two people really the same thing with it, like we all have this idea of what we kind of vaguely mean by it, but then we don't really know how to formalize it. So the way that I'm, I'm trying to approach the problem is to say, let's first talk about what we know about what our brains are made off. And the laws that describe what happens to the brain and for a particle physicist, the brain is just a big collection of particles and we know how those particles interact and we can write it down. And when we know the laws and okay, so the details are kind of complicated, you know, all the Synopsys and neurons and so on, which I don't really know a lot about, but basically it's all just particle physics.
2 (1h 11m 57s):
And we know that those laws are partly deterministic. So what happens in the future is determined by some initial set in the past. If you want to be extreme, you can roll it back all the time to the big bang or whatever was briefly after the big bang. And then there might be some random element from quantum mechanics, depends on how seriously you take quantum mechanics, but whether or not you add quantum mechanics into the mix, you're stuck with some mixture of determinism and randomness. And then seeing chaos doesn't really change anything about it because chaos is still deterministic makes it difficult to predict certain situations, but theoretically, it's still determined by the past.
