Your comment would make more sense to me if I removed the word “not” from the sentence you quote. (Also, if I don’t read past that sentence of someonewrongonthenet’s comment.)
That said, I agree completely that the kinds of vague identity concerns about cryonics that the quoted sentence with “not” removed would be raising would also arise, were one consistent, about routine continuation of existence over time.
That said, I agree completely that the kinds of vague identity concerns about cryonics that the quoted sentence with “not” removed would be raising would also arise, were one consistent, about routine continuation of existence over time.
There are things that when I go to bed to wake up eight hours later are very nearly preserved but if I woke up sixty years later wouldn’t be, e.g. other people’s memories of me (see I Am a Strange Loop) or the culture of the place where I live (see Good Bye, Lenin!).
(I’m not saying whether this is one of the main reasons why I’m not signed up for cryonics.)
Hrm.. ambiguous semantics. I took it to imply acceptance of the idea but not elevation of its importance, but I see how it could be interpreted differently. And yes, the rest of the post addresses something completely different. But if I can continue for a moment on the tangent, expanding my comment above (even if it doesn’t apply to the OP):
You actually continue functioning when you sleep, it’s just that you don’t remember details once you wake up. A more useful example for such discussion is general anesthesia, which shuts down the regions of the brain associated with consciousness. If personal identity is in fact derived from continuity of computation, then it is plausible that general anesthesia would result in a “different you” waking up after the operation. The application to cryonics depends greatly on the subtle distinction of whether vitrification (and more importantly, the recovery process) slows downs or stops computation. This has been a source of philosophical angst for me personally, but I’m still a cryonics member.
More troubling is the application to uploading. I haven’t done this yet, but I want my Alcor contract to explicitly forbid uploading as a restoration process, because I am unconvinced that a simulation of my destructively scanned frozen brain would really be a continuation of my personal identity. I was hoping that “Timeless Identity” would address this point, but sadly it punts the issue.
Well, if the idea is unimportant to the OP, presumably that also helps explain how they can sleep at night.
WRT the tangent… my own position wrt preservation of personal identity is that while it’s difficult to articulate precisely what it is that I want to preserve, and I’m not entirely certain there is anything cogent I want to preserve that is uniquely associated with me, I’m pretty sure that whatever does fall in that category has nothing to do with either continuity of computation or similarity of physical substrate. I’m about as sanguine about continuing my existence as a software upload as I am about continuing it as this biological system or as an entirely different biological system, as long as my subjective experience in each case is not traumatically different.
I wrote up about a page-long reply, then realized it probably deserves its own posting. I’ll see if I can get to that in the next day or so. There’s a wide spectrum of possible solutions to the personal identity problem, from physical continuity (falsified) to pattern continuity and causal continuity (described by Eliezer in the OP), to computational continuity (my own view, I think). It’s not a minor point though, whichever view turns out to be correct has immense ramifications for morality and timeless decision theory, among other things...
What relevance does personal identity have to TDT? TDT doesn’t depend on whether the other instances of TDT are in copies of you, or in other people who merely use the same decision theory as you.
Ok I will, but that part is easy enough to state here: I mean correct in the reductionist sense. The simplest explanation which resolves the original question and/or associated confusion, while adding to our predictive capacity and not introducing new confusion.
Mm. I’m not sure I understood that properly; let me echo my understanding of your view back to you and see if I got it.
Suppose I get in something that is billed as a transporter, but which does not preserve computational continuity. Suppose, for example, that it destructively scans my body, sends the information to the destination (a process which is not instantaneous, and during which no computation can take place), and reconstructs an identical body using that information out of local raw materials at my destination.
If it turns out that computational or physical continuity is the correct answer to what preserves personal identity, then I in fact never arrive at my destination, although the thing that gets constructed at the destination (falsely) believes that it’s me, knows what I know, etc. This is, as you say, an issue of great moral concern… I have been destroyed, this new person is unfairly given credit for my accomplishments and penalized for my errors, and in general we’ve just screwed up big time.
Conversely, if it turns out that pattern or causal continuity is the correct answer, then there’s no problem.
Therefore it’s important to discover which of those facts is true of the world.
Yes? This follows from your view? (If not, I apologize; I don’t mean to put up strawmen, I’m genuinely misunderstanding.)
If so, your view is also that if we want to know whether that’s the case or not, we should look for the simplest answer to the question “what does my personal identity comprise?” that does not introduce new confusion and which adds to our predictive capacity. (What is there to predict here?)
Yes?
EDIT: Ah, I just read this post where you say pretty much this. OK, cool; I understand your position.
Suppose I get in something that is billed as a transporter, but which does not preserve computational continuity. Suppose, for example, that it destructively scans my body, sends the information to the destination (a process which is not instantaneous, and during which no computation can take place), and reconstructs an identical body using that information out of local raw materials at my destination.
I don’t know what “computation” or “computational continuity” means if it’s considered to be separate from causal continuity, and I’m not sure other philosophers have any standard idea of this either. From the perspective of the Planck time, your brain is doing extremely slow ‘computations’ right now, it shall stand motionless a quintillion ticks and more before whatever arbitrary threshold you choose to call a neural firing. Or from a faster perspective, the 50 years of intervening time might as well be one clock tick. There can be no basic ontological distinction between fast and slow computation, and aside from that I have no idea what anyone in this thread could be talking about if it’s distinct from causal continuity.
(shrug) It’s Mark’s term and I’m usually willing to make good-faith efforts to use other people’s language when talking to them. And, yes, he seems to be drawing a distinction between computation that occurs with rapid enough updates that it seems continuous to a human observer and computation that doesn’t. I have no idea why he considers that distinction important to personal identity, though… as far as I can tell, the whole thing depends on the implicit idea of identity as some kind of ghost in the machine that dissipates into the ether if not actively preserved by a measurable state change every N microseconds. I haven’t confirmed that, though.
Hypothesis: consciousness is what a physical interaction feels like from the inside.
Importantly, it is a property of the interacting system, which can have various degrees of coherence—a different concept than quantum coherence, which I am still developing: something along the lines of negative-entropic complexity. There is therefore a deep correlation between negentropy and consciousness. Random thermodynamic motion in a gas is about as minimum-conscious as you can get (lots of random interactions, but all short lived and decoherent). A rock is slightly more conscious due to its crystalline structure, but probably leads a rather boring existence (by our standards, at least). And so on, all the way up to the very negentropic primate brain which experiences a high degree of coherent experience that we call “consciousness” or “self.”
I know this sounds like making thinking an ontologically basic concept. It’s rather the reverse—I am building the experience of thinking up from physical phenomenon: consciousness is the experience of organized physical interactions. But I’m not yet convinced of it either. If you throw out the concept of coherent interaction (what I have been calling computation continuity), then it does reduce to causal continuity. But causal continuity does have it’s problems which make me suspect it as not being the final, ultimate answer...
Hypothesis: consciousness is what a physical interaction feels like from the inside. ... consciousness is the experience of organized physical interactions.
How do you explain the existence of the phenomenon of “feeling like” and of “experience”?
I agree that the grandparent has circumvented addressing the crux of the matter, however I feel (heh) that the notion of “explain” often comes with unrealistic expectations. It bears remembering that we merely describe relationships as succinctly as possible, then that description is the “explanation”.
While we would e.g. expect/hope for there to be some non-contradictory set of descriptions applying to both gravity and quantum phenomena (for which we’d eat a large complexity penalty, since complex but accurate descriptions always beat out simple but inaccurate descriptions; Occam’s Razor applies only to choosing among fitting/not yet falsified descriptions), as soon as we’ve found some pinned-down description in some precise language, there’s no guarantee—or strictly speaking, need—of an even simpler explanation.
A world running according to currently en-vogue physics, plus a box which cannot be described as an extension of said physics, but only in some other way, could in fact be fully explained, with no further explanans for the explanandum.
It seems pretty straightforward to note that there’s no way to “derive” phenomena such as “feeling like” in the current physics framework, except of course to describe which states of matters/energy correspond to which qualia.
Such a description could be the explanation, with nothing further to be explained:
If it empirically turned out that a specific kind matter needs to be arranged in the specific pattern of a vertebrate brain to correlate to qualia, that would “explain” consciousness. If it turned out (as we all expect) that the pattern alone sufficies, then certain classes of instantiated algorithms (regardless of the hardware/wetware) would be conscious. Regardless, either description (if it turned out to be empirically sound) would be the explanation.
I also wonder, what could any answer within the current physics framework possibly look like, other than an asterisk behind the equations with the addendum of “values n1 … nk for parameters p1 … pk correlate with qualia x”?
How do you explain “feeling like” and “experience” in general? This is LW so I assume you have a reductionist background and would offer an explanation based on information patterns, neuron firings, hormone levels, etc. But ultimately all of that reduces down to a big collection of quarks, each taking part in mostly random interactions on the scale of femtoseconds. The apparent organization of the brain is in the map, not the territory. So if subjective experience reduces down to neurons, and neurons reduce down to molecules, and molecules reduce to quarks and leptons, where then does the consciousness reside? “Information patterns” alone is an inadequate answer—that’s at the level of the map, not the territory. Quarks and leptons combine into molecules, molecules into neural synapses, and the neurons connect into the 3lb information processing network that is my brain. Somewhere along the line, the subjective experience of “consciousness” arises. Where, exactly, would you propose that happens?
We know (from our own subjective experience) that something we call “consciousness” exists at the scale of the entire brain. If you assume that the workings of the brain is fully explained by its parts and their connections, and those parts explained by their sub-components and designs, etc. you eventually reach the ontologically basic level of quarks and leptons. Fundamentally the brain is nothing more than the interaction of a large number of quarks and leptons. So what is the precise interaction of fundamental particles is the basic unit of consciousness? What level of complexity is required before simply organic matter becomes a conscious mind?
It sounds ridiculous, but if you assume that quarks and leptons are “conscious,” or rather that consciousness is the interaction of these various ontologically primitive, fundamental particles, a remarkably consistent theory emerges: one which dissolves the mystery of subjective consciousness by explaining it as the mere aggregation of interdependent interactions. Besides being simple, this is also predictive: it allows us to assert for a given situation (e.g. a teleporter or halted simulation) whether loss of personal identity occurs, which has implications for morality of real situations encountered in the construction of an AI.
The apparent organization of the brain is in the map, not the territory.
What do you mean by this? Are fMRIs a big conspiracy?
Fundamentally the brain is nothing more than the interaction of a large number of quarks and leptons.
This description applies equally to all objects. When you describe the brain this way, you leave out all its interesting characteristics, everything that makes it different from other blobs of interacting quarks and leptons.
What I’m saying is that the high-level organization is not ontologically primitive. When we talk about organizational patterns of the brain, or the operation of neural synapses, we’re taking about very high level abstractions. Yes, they are useful abstractions primarily because they ignore unnecessary detail. But that detail is how they are actually implemented. The brain is soup of organic particles with very high rates of particle interaction due simply to thermodynamic noise. At the nanometer and femtosecond scale, there is very little signal to noise, however at the micrometer and millisecond scale general trends start to emerge, phenomenon which form the substrate of our computation. But these high level abstractions don’t actually exist—they are just average approximations over time of lower level, noisy interactions.
I assume you would agree that a normal adult brain in a human experiences a subjective feeling of consciousness that persists from moment-to-moment. I also think it’s a fair bet that you would not think that a single electron bouncing around in some part of a synaptic pathway or electronic transistor has anything resembling a conscious experience. But somehow, a big aggregation of these random motions does add up to you or me. So at what point in the formation of a human brain, or construction of an AI does it become conscious? At what point does it mere dead matter transform into sentience? Is this a hard cutoff? Is it gradual?
Speaking of gradations, certain animals can’t recognize themselves in a mirror. If you use self-awareness as a metric as was argued elsewhere, does that mean they’re not conscious? What about insects, which operate with a more distributed neural system. Dung beetles seem to accomplish most tasks by innate reflex response. Do they have at least a little, tiny subjective experience of consciousness? Or is their existence no more meaningful than that of a stapler?
Yes, this objection applies equally to all objects. That’s precisely my point. Brains are not made of any kind of “mind stuff”—that’s substance dualism which I reject. Furthermore, minds don’t have a subjective experience separate from what is physically explainable—that’s epiphenomenalism, similarly rejected. “Minds exist in information patterns” is a mysterious answer—information patterns are themselves merely evolving expressions in the configuration space of quarks & leptons. Any result of the information pattern must be explainable in terms of the interactions of its component parts, or else we are no longer talking about a reductionist universe. If I am coming at this with a particular bias, it is this: all aspects of mind including consciousness, subjective experience, qualia, or whatever you want to call it are fundamentally reducible to forces acting on elementary particles.
I see only two reductionist paths forward to take: (1) posit a new, fundamental law by which at some aggregate level of complexity or organization, a computational substrate becomes conscious. How & why is not explained, and as far as I can tell there is no experimental way to determine where this cutoff is. But assume it is there. Or, (2) accept that like everything else in the universe, consciousness reduces down to the properties of fundamental particles and their interactions (it is the interaction of particles). A quark and a lepton exchanging a photon is some minimal quantum Plank-level of conscious experience. Yes, that means that even a rock and a stapler experience some level of conscious experience—barely distinguishable from thermal noise, but nonzero—but the payoff is a more predictive reductionist model of the universe. In terms of biting bullets, I think accepting many-worlds took more gumption than this.
I also think it’s a fair bet that you would not think that a single electron bouncing around in some part of a synaptic pathway or electronic transistor has anything resembling a conscious experience. But somehow, a big aggregation of these random motions does add up to you or me. So at what point in the formation of a human brain, or construction of an AI does it become conscious? At what point does it mere dead matter transform into sentience? Is this a hard cutoff? Is it gradual?
This is a Wrong Question. Consciousness, whatever it is, is (P=.99) a result of a computation. My computer exhibits a microsoft word behavior, but if I zoom in to the electrons and transistors in the CPU, I see no such microsoft word nature. It is silly to zoom in to quarks and leptons looking for the true essence of microsoft word. This is the way computations work—a small piece of the computation simply does not display behavior that is like the entire computation. The CPU is not the computation. It is not the atoms of the brain that are conscious, it is the algorithm that they run, and the atoms are not the algorithm. Consciousness is produced by non-conscious things.
“Minds exist in information patterns” is a mysterious answer—information patterns are themselves merely evolving expressions in the configuration space of quarks & leptons. Any result of the information pattern must be explainable in terms of the interactions of its component parts, or else we are no longer talking about a reductionist universe. If I am coming at this with a particular bias, it is this: all aspects of mind including consciousness, subjective experience, qualia, or whatever you want to call it are fundamentally reducible to forces acting on elementary particles.
Minds exist in some algorithms (“information pattern” sounds too static for my taste). Your desire to reduce things to forces on elementary particles is misguided, I think, because you can do the same computation with many different substrates. The important thing, the thing we care about, is the computation, not the substrate. Sure, you can understand microsoft word at the level of quarks in a CPU executing assembly language, but it’s much more useful to understand it in terms of functions and algorithms.
You’ve completely missed / ignored my point, again. Microsoft Word can be functionally reduced to electrons in transistors. The brain can be functionally reduced to biochemistry. Unless you resort to some form of dualism, the mind (qualia) is also similarly reduced.
just as computation can be brought down to the atomic scale (or smaller, with quantum computing), so too can conscious experiences be constructed out of such computational events. Indeed they are one and the same thing, just viewed from different perspectives.
The brain can be functionally reduced to biochemistry. Unless you resort to some form of dualism, the mind (qualia) is also similarly reduced.
I thought dualism meant you thought that there was ontologically basic conciousness stuff separate from ordinary matter?
I think the mind should be reduced to algorithms, and biochemistry is an implementation detail. This may make me a dualist by your usage of the word.
I think that it’s equally silly to ask, “where is the microsoft-word-ness” about a subset of transistors in your CPU as it is to ask “where is the consciousness” about a subset of neurons in your brain. I see this as describing how non-ontologically-basic consciousness can be produced by non-conscious stuff.
You’ve completely missed / ignored my point, again.
Apologies; does the above address your point? If not I’m confused about your point.
I’m arguing that if you think the mind can be reduced to algorithms implemented on computational substrate, then it is a logical consequence from our understanding of the rules of physics and the nature of computation that what we call subjective experience must also scale down as you reduce a computational machine down to its parts. After all, the algorithms themselves too also reducible down to stepwise axiomatic logical operations, implemented as transistors or interpretable machine code.
The only way to preserve the common intuition that “it takes (simulation of) a brain or equivalent to produce a mind” is to posit some form of dualism. I don’t think it is silly to ask “where is the microsoft-word-ness” about a subset of a computer—you can for example point to the regions of memory and disk where the spellchecker is located, and say “this is the part that matches user input against tables of linguistic data,” just like we point to regions of the brain and say “this is your language processing centers.”
The experience of having a single, unified me directing my conscious experience is an illusion—it’s what the integration process feels like from the inside, but it does not correspond to reality (we have psychological data to back this up!). I am in fact a society of agents, each simpler but also relying on an entire bureaucracy of other agents in an enormous distributed structure. Eventually though, things reduce down to individual circuits, then ultimately to the level of individual cell receptors and chemical pathways. At no point along the way is there a clear division where it is obvious that conscious experience ends and what follows is merely mechanical, electrical, and chemical processes. In fact as I’ve tried to point out the divisions between higher level abstractions and their messy implementations is in the map, not the territory.
To assert that “this level of algorithmic complexity is a mind, and below that is mere machines” is a retreat to dualism, though you may not yet see it in that way. What you are asserting is that there is this ontologically basic mind-ness which spontaneously emerges when an algorithm has reached a certain level of complexity, but which is not the aggregation of smaller phenomenon.
I think we have really different models of how algorithms and their sub-components work.
it is a logical consequence from our understanding of the rules of physics and the nature of computation that what we call subjective experience must also scale down as you reduce a computational machine down to its parts.
Suppose I have a computation that produces the digits of pi. It has subroutines which multiply and add. Is it an accurate description of these subroutines that they have a scaled down property of computes-pi-ness? I think this is not a useful way to understand things. Subroutines do not have a scaled-down percentage of the properties of their containing algorithm, they do a discrete chunk of its work. It’s just madness to say that, e.g., your language processing center is 57% conscious.
The experience of having a single, unified me directing my conscious experience is an illusion...
I agree with all this. Humans probably are not the minimal conscious system, and there are probably subsets of our component circuitry which maintain the property of conciousness. But yes, I maintain that eventually, you’ll get to an algorithm that is conscious while none of its subroutines are.
If this makes me a dualist then I’m a dualist, but that doesn’t feel right. I mean, the only way you can really explain a thing is to show how it arises from something that’s not like it in the first place, right?
I think we have different models of what consciousness is. In your pi example, the multiplier has multiply-ness, and the adder has add-ness properties, and when combined together in a certain way you get computes-pi-ness. Likewise our minds have many, many, many different components which—somehow, someway—each have a small experiential qualia which when you sum together yield the human condition.
Through brain damage studies, for example, we have descriptions of what it feels like to live without certain mental capabilities. I think you would agree with this, but for others reading take this thought experiment: imagine that I were to systematically shut down portions of your brain, or in simulation, delete regions of your memory space. For the purpose of the argument I do it slowly over time in relatively small amounts, and cleaning up dangling references so the whole system doesn’t shut down. Certainly as time goes by your mental functionality is reduced, and you stop being capable of having experiences you once took for granted. But at what point, precisely, do you stop experiencing at all qualia of any form? When you’re down to just a billion neurons? A million? A thousand? When you’re down to just one processing region? Is one tiny algorithm on a single circuit enough?
Humans probably are not the minimal conscious system, and there are probably subsets of our component circuitry which maintain the property of consciousness. But yes, I maintain that eventually, you’ll get to an algorithm that is conscious while none of its subroutines are.
What is the minimal conscious system? It’s easy and perhaps accurate to say “I don’t know.” After all, neither one of us know enough neural and cognitive science to make this call, I assume. But we should be able to answer this question: “if presented criteria for a minimally-conscious-system, what would convince me of its validity?”
If this makes me a dualist then I’m a dualist, but that doesn’t feel right. I mean, the only way you can really explain a thing is to show how it arises from something that’s not like it in the first place, right?
Eliezer’s post on reductionism is relevant here. In a reductionist universe, anything and everything is fully defined by its constituent elements—no more, no less. There’s a popular phrase that has no place is reductionist theories: “the whole is greater than the sum of its parts.” Typically what this actually means is that you failed to count the “parts” correctly: a part list should also include spatial configurations and initial conditions, which together imply the dynamic behaviors as well. For example, a pulley is more than a hunk of metal and some rope, but it is fully defined if you specify how the metal is shaped, how the rope is threaded through it and fixed to objects with knots, how the whole contraption is oriented with respect to gravity, and the procedure for applying rope-pulling-force. Combined with the fundamental laws of physics, this is a fully reductive explanation of a rope-pulley system which is the sum of its fully-defined parts.
And so it goes with consciousness. Unless we are comfortable with the mysterious answers provided by dualism—or empirical evidence like confirmation of psychic phenomenon compels us to go there—then we must demand that an explanation be provided that explains consciousness fully as the aggregation of smaller processes.
When I look a explanations of the workings of the brain, starting with the highest level psychological theories and neural structure, and working the way all the way down the abstraction hierarchy to individual neural synapses and biochemical pathways, nowhere along the way do I see an obvious place to stop and say “here is where consciousness begins!” Likewise, I can start from the level of mere atoms and work my way up to the full neural architecture, without finding any step that adds something which could be consciousness, but which isn’t fundamentally like the levels below it. But when you get to the highest level, you’ve described the full brain without finding consciousness anywhere along the way.
I can see how this leads otherwise intelligent philosophers like David Chalmers to epiphenomenalism. But I’m not going to go down that path, because the whole situation is the result of mental confusion.
The Standard Rationalist Answer is that mental processes are information patterns, nothing more, and tat consciousness is an illusion, end of story. But that still leaves me confused! It’s not like free will for example, where because of the mind projection fallacy I think I have free will due to how a deterministic decision theory algorithm feels from the inside. I get that. No, the answer of “that subjective experience of consciousness isn’t real, get over it” is unsatisfactory because if I don’t have conscious, how am I experiencing thinking in the first place? Cogito ergo sum.
However there is a way out. I went looking for a source of consciousness because I like nearly every other philosopher assumed that there was something special and unique which set brains aside as having minds which other more mundane objects—like rocks and staplers—do not possess. That’s so obviously true, but honestly I have no real justification for that belief. So let’s try negating it. What is possible if we don’t exclude mundane things from having minds too?
Well, what does it feel like to be a quark and a lepton exchanging a photon? I’m not really sure, but let’s call that approximately the minimum possible “experience”, and for the duration of the interaction continuous interaction over time, the two particles share a “mind”. Arrange a number of these objects together and you get an atom, which itself also has a shared/merged experience so long as the particles remain in bonded interaction. Arrange a lot of atoms together and you get a electrical transistor. Now we’re finally starting to get to a level where I have some idea of what the “shared experience of being a transistor” would be (rather boring, by my standards), and more importantly, it’s clear how that experience is aggregated together from its constituent parts. From here, computing theory takes over as more complex interdependent systems are constructed, each merging experiences together into a shared hive mind, until you reach the level of the human being or AI.
Are you at least following what I’m saying, even if you don’t agree?
That was a very long comment (thank you for your effort) and I don’t think I have the energy to exhaustively go through it.
I believe I follow what you’re saying. It doesn’t make much sense to me, so maybe that belief is false.
I think the fact that if you start with a brain, which is presumably conscious, and zoom in all the way looking for the conciousness boundary, and then start with a quark, which is presumably not conscious, and zoom all the way out to the entire brain, also without finding a consciousness barrier—I think this means that the best we can do at the moment is set upper and lower bounds.
A minimally conscious system—say, something that can convince me that it thinks it is conscious. “echo ‘I’m conscious!’” doesn’t quite cut it, things that recognize themselves in mirrors probably do, and I could go either way on the stuff in between.
I think your reductionism is a little misapplied. My pi-calculating program develops a new property of pi-computation when you put the adders and multipliers together right, but is completely described in terms of adders and multipliers. I expect consciousness to be exactly the same; it’ll be completely described in terms of qualia generating algorithms (or some such), which won’t themselves have the consciousness property.
This is hard to see because the algorithms are written in spaghetti code, in the wiring between neurons. In computer terms, we have access to the I/O system and all the gates in the CPU, but we don’t currently know how they’re connected. Looking at more or fewer of the gates doesn’t help, because the critical piece of information is how they’re connected and what algorithm they implement.
IMO, my guess (P=.65) is that qualia are going to turn out to be something like vectors in a feature space. Under this model, clearly systems incapable of representing such a vector can’t have any qualia at all. Rocks and single molecules, for example.
How do you explain “feeling like” and “experience” in general? This is LW so I assume you have a reductionist background and would offer an explanation based on information patterns, neuron firings, hormone levels, etc.
I indeed have a reductionist background, but I offer no explanation, because I have none. I do not even know what an explanation could possibly look like; but neither do I take that as proof that there cannot be one. The story you tell surrounds the central mystery with many physical details, but even in your own accont of it the mystery remains unresolved:
Somewhere along the line, the subjective experience of “consciousness” arises.
However much you assert that there must be an explanation, I see here no advance towards actually having one. What does it mean to attribute consciousness to subatomic particles and rocks? Does it predict anything, or does it only predict that we could make predictions about teleporters and simulations if we had a physical explanation of consciousness?
Hypothesis: consciousness is what a physical interaction feels like from the inside.
I would imagine that consciousness (in a sense of self-awareness) is the ability to introspect into your own algorithm. The more you understand what makes you tick, rather than mindlessly following the inexplicable urges and instincts, the more conscious you are.
Yes, that is not only 100% accurate, but describes where I’m headed.
I am looking for the simplest explanation of the subjective continuity of personal identity, which either answers or dissolves the question. Further, the explanation should either explain which teleportation scenario is correct (identity transfer, or murder+birth), or satisfactorily explain why it is a meaningless distinction.
What is there to predict here?
If I, the person standing in front of the transporter door, will experience walking on Mars, or oblivion.
Yes, it is perhaps likely that this will never be experimentally observable. That may even be a tautology since we are talking about subjective experience. But still, a reductionist theory of consciousness could provide a simple, easy to understand explanation for the origin of personal identity (e.g., what an computational machine feels like from the inside) and which predicts identity transfer or murder + birth. That would be enough for me, at least as long as there’s not competing equally simple theories.
What is there to predict here? If I, the person standing in front of the transporter door, will experience walking on Mars, or oblivion.
Well, you certainly won’t experience oblivion, more or less by definition. The question is whether you will experience walking on Mars or not.
But there is no distinct observation to be made in these two cases. That is, we agree that either way there will be an entity having all the observable attributes (both subjective and objective; this is not about experimental proof, it’s about the presence or absence of anything differentially observable by anyone) that Mark Friendebach has, walking on Mars.
So, let me rephrase the question: what observation is there to predict here?
So, let me rephrase the question: what observation is there to predict here?
That’s not the direction I was going with this. It isn’t about empirical observation, but rather aspects of morality which depend on subjective experience. The prediction is under what conditions subjective experience terminates. Even if not testable, that is still an important thing to find out, with moral implications.
Is it moral to use a teleporter? From what I can tell, that depends on whether the person’s subjective experience is terminated in the process. From the utility point of view the outcomes are very nearly the same—you’ve murdered one person, but given “birth” to an identical copy in the process. However if the original, now destroyed person didn’t want to die, or wouldn’t have wanted his clone to die, then it’s a net negative.
As I said elsewhere, the teleporter is the easiest way to think of this, but the result has many other implications from general anesthesia, to cryonics, to Pascal’s mugging and the basilisk.
I want my Alcor contract to explicitly forbid uploading as a restoration process, because I am unconvinced that a simulation of my destructively scanned frozen brain would really be a continuation of my personal identity.
Like TheOtherDave (I presume), I consider my identity to be adequately described by whatever Turing machine that can emulate my brain, or at least its prefrontal cortex + relevant memory storage. I suspect that a faithful simulation of just my Brodmann area 10 coupled with a large chunk of my memories would restore enough of my self-awareness to be considered “me”. This sim-me would probably lose most of my emotions without the rest of the brain, but it is still infinitely better than none.
You’ll need the rest of the brain because these other memories would be distributed throughout the rest of your cortex. The hippocampus only contains recent episodic memories.
If you lost your temporal lobe, for example, you’d lose all non-episodic knowledge concerning what the names of things are, how they are categorized, and what the relationships between them are.
That said, I’m not sure why I should care much about having my non-episodic knowledge replaced with an off-the-shelf encyclopedia module. I don’t identify with it much.
If you only kept the hippocampus, you’d lose your non-recent episodic memories too. But technical issues aside, let me defend the “encyclopedia”:
Episodic memory is basically a cassette reel of your life, along with a few personalized associations and maybe memories of thoughts and emotions. Everything that we associate with the word knowledge is non-episodic. It’s not just verbal labels—that was just a handy example that I happened to know the brain region for. I’d actually care about that stuff more about non-episodic memories than the episodic stuff.
Things like “what is your wife’s name and what does her face look like” are non-episodic memory. You don’t have to think back to a time when you specifically saw your wife to remember what her name and face is, and that you love her—that information is treated as a fact independent of any specific memory, indelibly etched into your model of the world. Cognitively speaking, “I love my wife stacy, she looks like this” is as much of a fact as “grass is a green plant” and they are both non-episodic memories. Your episodic memory reel wouldn’t even make sense without that sort of information. I’d still identify someone with memory loss, but retaining my non-episodic memory, as me. I’d identify someone with only my episodic memories as someone else, looking at a reel of memory that does not belong to them and means nothing to them.
(Trigger Warning: link contains writing in diary which is sad, horrifying, and nonfiction.): This is what complete episodic memory loss looks like. Patients like this can still remember the names of faces of people they love.
Ironically...the (area 10) might actually be replaceable. I’m not sure whether any personalized memories are kept there—I don’t know what that specific region does but it’s in an area that mostly deals with executive function—which is important for personality, but not necessarily individuality.
Ironically...the (area 10) might actually be replaceable. I’m not sure whether any personalized memories are kept there—I don’t know what that specific region does but it’s in an area that mostly deals with executive function—which is important for personality, but not necessarily individuality.
What’s the difference between personality and individuality?
Personality is a set of dichotomous variables plotted on a bell curve. “Einstein was extroverted, charismatic, nonconforming, and prone to absent-mindedness” describes his personality. We all have these traits in various amounts. You can some of these personality nobs really easily with drugs. I can’t specify Einstein out of every person in the world using only his personality traits—I can only specify individuals similar to him.
Individuality is stuff that’s specific to the person. “Einstein’s second marriage was to his cousin and he had at least 6 affairs. He admired Spinoza, and was a contemporary of Tagore. He was a socialist and cared about civil rights. He had always thought there was something wrong about refrigerators.” Not all of these are dichotomous variables—you either spoke to Tagore or you didn’t. And it makes no sense to put people on a “satisfaction with Refrigerators” spectrum, even though I suppose you could if you wanted to. And all this information together specifically points to Einstein, and no one else in the world. Everyone in the world a set of unique traits like fingerprints—and it doesn’t even make sense to ask what the “average” is, since most of the variables don’t exist on the same dimension.
And...well, when it comes to Area 10, just intuitively, do you really want to define yourself by a few variables that influence your executive function? Personally I define myself partially by my ideas, and partially by my values...and the former is definitely in the “individuality” territory.
OK, I understand what you mean by personality vs individuality. However, I doubt that the functionality of BA10 can be described “by a few variables that influence your executive function”. Then again, no one knows anything definite about it.
I take it you’re assuming that information about my husband, and about my relationship to my husband, isn’t in the encyclopedia module along with information about mice and omelettes and your relationship to your wife.
If that’s true, then sure, I’d prefer not to lose that information.
Well...yeah, I was. I thought the whole idea of having an encyclopedia was to eliminate redundancy through standardization of the parts of the brain that were not important for individuality?
If your husband and my husband, your omelette and my omelette, are all stored in the encyclopedia, it wouldn’t be a “off-the-shelf encyclopedia module” anymore. It would be an index containing individual people’s non-episodic knowledge. At that point, it’s just an index of partial uploads. We can’t standardize that encyclopedia to everyone: If the the thing that stores your omelette and your husband went around viewing my episodic reel and knowing all the personal stuff about my omelette and husband...that would be weird and the resulting being would be very confused (let alone if the entire human race was in there—I’m not sure how that would even work).
(Also, going back into the technical stuff, there may or may not be a solid dividing line between very old episodic memory and non-episodoc memory
Sure, if your omelette and my omelette are so distinct that there is no common data structure that can serve as a referent for both, and ditto for all the other people in the world, then the whole idea of an encyclopedia falls apart. But that doesn’t seem terribly likely to me.
Your concept of an omelette probably isn’t exactly isomorphic to mine, but there’s probably a parametrizable omelette data structure we can construct that, along with a handful of parameter settings for each individual, can capture everyone’s omelette. The parameter settings go in the representation of the individual; the omelette data structure goes in the encyclopedia.
And, in addition, there’s a bunch of individualizing episodic memory on top of that… memories of cooking particular omelettes, of learning to cook an omelette, of learning particular recipes, of that time what ought to have been an omelette turned into a black smear on the pan, etc. And each of those episodic memories refers to the shared omelette data structure, but is stored with and is unique to the uploaded agent. (Maybe. It may turn out that our individual episodic memories have a lot in common as well, such that we can store a standard lifetime’s memories in the shared encyclopedia and just store a few million bits of parameter settings in each individual profile. I suspect we overestimate how unique our personal narratives are, honestly.)
Similarly, it may be that our relationships with our husbands are so distinct that there is no common data structure that can serve as a referent for both. But that doesn’t seem terribly likely to me. Your relationship with your husband isn’t exactly isomorphic to mine, of course, but it can likely similarly be captured by a common parameterizable relationship-to-husband data structure.
As for the actual individual who happens to be my husband, well, the majority of the information about him is common to all kinds of relationships with any number of people. He is his father’s son and his stepmother’s stepson and my mom’s son-in-law and so on and so forth. And, sure, each of those people knows different things, but they know those things about the same person; there is a central core. That core goes in the encyclopedia, and pointers to what subset each person knows about him goes in their individual profiles (along with their personal experiences and whatever idiosyncratic beliefs they have about him).
So, yes, I would say that your husband and my husband and your omelette and my omelette are all stored in the encyclopedia. You can call that an index of partial uploads if you like, but it fails to incorporate whatever additional computations that create first-person experience. It’s just a passive data structure.
Incidentally and unrelatedly, I’m not nearly as committed as you sound to preserving our current ignorance of one another’s perspective in this new architecture.
I’m really skeptical that parametric functions which vary on dimensions concerning omelets (Egg species? Color? ingredients? How does this even work?) are a more efficient or more accurate way of preserving what our wetware encode when compared to simulating the neural networks devoted dealing with omelettes. I wouldn’t even know how to start working on the problem mapping a conceptual representation of an omelette into parametric functions (unless we’re just using the parametric functions to model the properties of individual neurons—that’s fine).
Can you give an example concerning what sort of dimension you would parametrize so I have a better idea of what you mean?
Incidentally and unrelatedly, I’m not nearly as committed as you sound to preserving our current ignorance of one another’s perspective in this new architecture.
I was more worried that it might break stuff (as in, resulting beings would need to be built quite differently in order to function) if one-another’s perspectives would overlap. Also, that brings us back to the original question I was raising about living forever—what exactly is it that we value and want to preserve?
Can you give an example concerning what sort of dimension you would parametrize so I have a better idea of what you mean?
Not really. If I were serious about implementing this, I would start collecting distinct instances of omelette-concepts and analyzing them for variation, but I’m not going to do that. My expectation is that if I did, the most useful dimensions of variability would not map to any attributes that we would ordinarily think of or have English words for.
Perhaps what I have in mind can be said more clearly this way: there’s a certain amount of information that picks out the space of all human omelette-concepts from the space of all possible concepts… call that bitstring S1. There’s a certain amount of information that picks out the space of my omelette-concept from the space of all human omelette-concepts… call that bitstring S2.
S2 is much, much, shorter than S1.
It’s inefficient to have 7 billion human minds each of which is taking up valuable bits storing 7 billion copies of S1 along with their individual S2s. Why in the world would we do that, positing an architecture that didn’t physically require it? Run a bloody compression algorithm, store S1 somewhere, have each human mind refer to it.
I have no idea what S1 or S2 are.
And I don’t expect that they’re expressible in words, any more than I can express which pieces of a movie are stored as indexed substrings… it’s not like MPEG compression of a movie of an auto race creates an indexed “car” data structure with parameters representing color, make, model, etc. It just identifies repeated substrings and indexes them, and takes advantage of the fact that sequential frames share many substrings in common if properly parsed.
But I’m committed enough to a computational model of human concept storage that I believe they exist. (Of course, it’s possible that our concept-space of an omelette simply can’t be picked out by a bit-string, but I can’t see why I should take that possibility seriously.)
Oh, and agreed that we would change if we were capable of sharing one another’s perspectives. I’m not particularly interested in preserving my current cognitive isolation from other humans, though… I value it, but I value it less than I value the ability to easily share perspectives, and they seem to be opposed values.
My non-episodic memory contains the “facts” that Buffy the Vampire Slayer was one of the best television shows that was ever made, and the Pink Floyd aren’t an interesting band. My boyfriend’s non-episodic memory contains the facts that Buffy was boring, unoriginal, and repetetive (and that Pink Floyd’s music is trancendentally good).
Objectively, these are opinions, not facts. But we experience them as facts. If I want to preserve my sense of identity, then I would need to retain the facts that were in my non-episodic memory. More than that, I would also lose my sense of self if I gained contradictory memories. I would need to have my non-episodic memories and not have the facts from my boyfriend’s memory.
That’s the reason why “off the shelf” doesn’t sound suitable in this context.
So, on one level, my response to this is similar to the one I gave (a few years ago) [http://lesswrong.com/lw/qx/timeless_identity/9trc]… I agree that there’s a personal relationship with BtVS, just like there’s a personal relationship with my husband, that we’d want to preserve if we wanted to perfectly preserve me.
I was merely arguing that the bitlength of that personal information is much less than the actual information content of my brain, and there’s a great deal of compression leverage to be gained by taking the shared memories of BtVS out of both of your heads (and the other thousands of viewers) and replacing them with pointers to a common library representation of the show and then have your personal relationship refer to the common library representation rather than your private copy.
The personal relationship remains local and private, but it takes up way less space than your mind currently does.
That said… coming back to this conversation after three years, I’m finding I just care less and less about preserving whatever sense of self depends on these sorts of idiosyncratic judgments.
I mean, when you try to recall a BtVS episode, your memory is imperfect… if you watch it again, you’ll uncover all sorts of information you either forgot or remembered wrong. If I offered to give you perfect eideitic recall of BtVS—no distortion of your current facts about the goodness of it, except insofar as those facts turn out to be incompatible with an actual perception (e.g., you’d have changed your mind if you watched it again on TV, too) -- would you take it?
I would. I mean, ultimately, what does it matter if I replace my current vague memory of the soap opera Spike was obsessively watching with a more specific memory of its name and whatever else we learned about it? Yes, that vague memory is part of my unique identity, I guess, in that nobody else has quite exactly that vague memory… but so what? That’s not enough to make it worth preserving.
And for all I know, maybe you agree with me… maybe you don’t want to preserve your private “facts” about what kind of tie Giles was wearing when Angel tortured him, etc., but you draw the line at losing your private “facts” about how good the show was. Which is fine, you care about what you care about.
But if you told me right now that I’m actually an upload with reconstructed memories, and that there was a glitch such that my current “facts” about BTVS being a good show for its time is mis-reconstructed, and Dave before he died thought it was mediocre… well, so what?
I mean, before my stroke, I really disliked peppers. After my stroke, peppers tasted pretty good. This was startling, but it posed no sort of challenge to my sense of self.
Apparently (Me + likes peppers) ~= (Me + dislikes peppers) as far as I’m concerned.
I suspect there’s a million other things like that.
Like TheOtherDave (I presume), I consider my identity to be adequately described by whatever Turing machine that can emulate my brain, or at least its prefrontal cortex + relevant memory storage.
There’s a very wide range of possible minds I consider to preserve my identity; I’m not sure the majority of those emulate my prefrontal cortex significantly more closely than they emulate yours, and the majority of my memories are not shared by the majority of those minds.
Interesting. I wonder what you would consider a mind that preserves your identity. For example, I assume that the total of your posts online, plus whatever other information available without some hypothetical future brain scanner, all running as a process on some simulator, is probably not enough.
At one extreme, if I assume those posts are being used to create a me-simulation by me-simulation-creator that literally knows nothing else about humans, then I’m pretty confident that the result is nothing I would identify with. (I’m also pretty sure this scenario is internally inconsistent.)
At another extreme, if I assume the me-simulation-creator has access to a standard template for my general demographic and is just looking to customize that template sufficiently to pick out some subset of the volume of mindspace my sufficiently preserved identity defines… then maybe. I’d have to think a lot harder about what information is in my online posts and what information would plausibly be in such a template to even express a confidence interval about that.
That said, I’m certainly not comfortable treating the result of that process as preserving “me.”
Then again I’m also not comfortable treating the result of living a thousand years as preserving “me.”
Because the notion of “me” is not an ontologically basic category and the question of whether the “real me” wakes up is a question that aught to be un-asked.
I’m a bit confused at the question...you articulated my intent with that sentence perfectly in your other post.
Hrm.. ambiguous semantics. I took it to imply acceptance of the idea but not elevation of its importance, but I see how it could be interpreted differently.
and, as TheOtherDave said,
presumably that also helps explain how they can sleep at night.
EDIT: Nevermind, I now understand which part of my statement you misunderstood.
I’m not accepting-but-not-elevating the idea that the ’Real me” doesn’t wake up on the other side. Rather, I’m saying that the questions of personal identity over time do not make sense in the first place. It’s like asking “which color is the most moist”?
You actually continue functioning when you sleep, it’s just that you don’t remember details once you wake up. A more useful example for such discussion is general anesthesia, which shuts down the regions of the brain associated with consciousness. If personal identity is in fact derived from continuity of computation, then it is plausible that general anesthesia would result in a “different you” waking up after the operation. The application to cryonics depends greatly on the subtle distinction of whether vitrification (and more importantly, the recovery process) slows downs or stops computation. This has been a source of philosophical angst for me personally, but I’m still a cryonics member.
More troubling is the application to uploading. I haven’t done this yet, but I want my Alcor contract to explicitly forbid uploading as a restoration process, because I am unconvinced that a simulation of my destructively scanned frozen brain would really be a continuation of my personal identity. I was hoping that “Timeless Identity” would address this point, but sadly it punts the issue.
The root of your philosophical dilemma is that “personal identity” is a conceptual substitution for soul—a subjective thread that connects you over space and time.
No such thing exists. There is no specific location in your brain which is you. There is no specific time point which is you. Subjective experience exists only in the fleeting present. The only “thread’ connecting you to your past experiences is your current subjective experience of remembering them. That’s all.
I always wonder how I should treat my future self if I reject the continuity of self. Should I think of him like a son? A spouse? A stranger? Should I let him get fat? Not get him a degree? Invest in stock for him? Give him another child?
The root of your philosophical dilemma is that “personal identity” is a conceptual substitution for soul—a subjective thread that connects you over space and time.
No such thing exists. There is no specific location in your brain which is you. There is no specific time point which is you. Subjective experience exists only in the fleeting present. The only “thread’ connecting you to your past experiences is your current subjective experience of remembering them. That’s all.
I have a strong subjective experience of moment-to-moment continuity, even if only in the fleeting present. Simply saying “no such thing exists” doesn’t do anything to resolve the underlying confusion. If no such thing as personal identity exists, then why do I experience it? What is the underlying insight that eliminates the question?
This is not an abstract question either. It has huge implications for the construction of timeless decision theory and utilitarian metamorality.
“a strong subjective experience of moment-to-moment continuity” is an artifact of the algorithm your brain implements. It certainly exists in as much as the algorithm itself exists. So does your personal identity. If in the future it becomes possible to run the same algorithm on a different hardware, it will still produce this sense of personal identity and will feel like “you” from the inside.
Yes, I’m not questioning whether a future simulation / emulation of me would have an identical subjective experience. To reject that would be a retreat to epiphenomenalism.
Let me rephrase the question, so as to expose the problem: if I were to use advanced technology to have my brain scanned today, then got hit by a bus and cremated, and then 50 years from now that brain scan is used to emulate me, what would my subjective experience be today? Do I experience “HONK Screeeech, bam” then wake up in a computer, or is it “HONK Screeeech, bam” and oblivion?
Yes, I realize that in both cases result in a computer simulation of Mark in 2063 claiming to have just woken up in the brain scanner, with a subjective feeling of continuity. But is that belief true? In the two situations there’s a very different outcome for the Mark of 2013. If you can’t see that, then I think we are talking about different things, and maybe we should taboo the phrase “personal/subjective identity”.
if I were to use advanced technology to have my brain scanned today, then got hit by a bus and cremated, and then 50 years from now that brain scan is used to emulate me, what would my subjective experience be today? Do I experience “HONK Screeeech, bam” then wake up in a computer, or is it “HONK Screeeech, bam” and oblivion?
Ah, hopefully I’m slowly getting what you mean. So, there was the original you, Mark 2013, whose algorithm was terminated soon after it processed the inputs “HONK Screeeech, bam”, and the new you, Mark 2063, whose experience is “HONK Screeeech, bam” then “wake up in a computer”. You are concerned with… I’m having trouble articulated what exactly… something about the lack of experiences of Mark 2013? But, say, if Mark 2013 was restored to life in mostly the same physical body after a 50-year “oblivion”, you wouldn’t be?
Ah, hopefully I’m slowly getting what you mean. So, there was the original you, Mark 2013, whose algorithm was terminated soon after it processed the inputs “HONK Screeeech, bam”, and the new you, Mark 2063, whose experience is “HONK Screeeech, bam” then “wake up in a computer”.
Pretty much correct. To be specific, if computational continuity is what matters, then Mark!2063 has my memories, but was in fact “born” the moment the simulation started, 50 years in the future. That’s when his identity began, whereas mine ended when I died in 2013.
This seems a little more intuitive when you consider switching on 100 different emulations of me at the same time. Did I somehow split into 100 different persons? Or was there in fact 101 separate subjective identities, 1 of which terminated in 2013 and 100 new ones created for the simulations? The latter is a more straight forward explanation, IMHO.
You are concerned with… I’m having trouble articulated what exactly… something about the lack of experiences of Mark 2013? But, say, if Mark 2013 was restored to life in mostly the same physical body after a 50-year “oblivion”, you wouldn’t be?
No, that would make little difference as it’s pretty clear that physical continuity is an illusion. If pattern or causal continuity were correct, then it’d be fine, but both theories introduce other problems. If computational continuity is correct, then a reconstructed brain wouldn’t be me any more than a simulation would. However it’s possible that my cryogenically vitrified brain would preserve identity, if it were slowly brought back online without interruption.
I’d have to learn more about how general anesthesia works to decide if personal identity would be preserved across on the operating table (until then, it scares the crap out of me). Likewise, a AI or emulation running on a computer that is powered off and then later resumed would also break identity, but depending on the underlying nature of computation & subjective experience, task switching and online suspend/resume may or may not result in cycling identity.
I’ll stop there because I’m trying to formulate all these thoughts into a longer post, or maybe a sequence of posts.
It’s easier to explain in the case of multiple copies of yourself. Imagine the transporter were turned into a replicator—it gets stuck in a loop reconstructing the last thing that went through it, namely you. You step off and turn around to find another version of you just coming out. And then another, and another, etc. Each one of you shares the same memories, but from that moment on you have diverged. Each clone continues life with their own subjective experience until that experience is terminated by that clone’s death.
That sense of subjective experience separate from memories or shared history is what I have been calling “personal identity.” It is what gives me the belief, real or illusory, that I am the same person from moment to moment, day to day, and what separates me from my clones. You are welcome to suggest a better term.
The replicator / clone thought experiment shows that “subjective experience of identity” is something different from the information pattern that represents your mind. There is something, although at this moment that something is not well defined, which makes you the same “you” that will exist five minutes in the future, but which separates you from the “you”s that walked out of the replicator, or exist in simulation, for example.
The first step is recognizing this distinction. Then turn around and apply it to less fantastical situations. If the clone is “you” but not you (meaning no shared identity, and my apologies for the weak terminology), then what’s to say that a future simulation of “you” would also be you? What about cryonics, will your unfrozen brain still be you? That might depend on what they do to repair damage from vitrification. What about general anesthesia? Again, I need to learn more about how general anesthesia works, but if they shut down your processing centers and then restart you later, how is that different from the teleportation or simulation scenario? After all we’ve already established that whatever provides personal identity, it’s not physical continuity.
That sense of subjective experience separate from memories or shared history is what I have been calling “personal identity.” It is what gives me the belief, real or illusory, that I am the same person from moment to moment, day to day, and what separates me from my clones.
Well, OK. So suppose that, after I go through that transporter/replicator, you ask the entity that comes out whether it has the belief, real or illusory, that it is the same person in this moment that it was at the moment it walked into the machine, and it says “yes”.
If personal identity is what creates that belief, and that entity has that belief, it follows that that entity shares my personal identity… doesn’t it?
Well, OK. So suppose that, after I go through that transporter/replicator, you ask the entity that comes out whether it has the belief, real or illusory, that it is the same person in this moment that it was at the moment it walked into the machine, and it says “yes”.
If personal identity is what creates that belief, and that entity has that belief, it follows that that entity shares my personal identity… doesn’t it?
Not quite. If You!Mars gave it thought before answering, his thinking probably went like this: “I have memories of going into the transporter, just a moment ago. I have a continuous sequence of memories, from then until now. Nowhere in those memories does my sense of self change. Right now I am experiencing the same sense of self I always remember experiencing, and laying down new memories. Ergo, proof by backwards induction I am the same person that walked into the teleporter.” However for that—or any—line of meta reasoning to hold, (1) your memories need to accurately correspond with the true and full history of reality and (2) you need trust that what occurs in the present also occurred in the past. In other words, it’s kinda like saying “my memory wasn’t altered because I would have remembered that.” It’s not a circular argument per se, but it is a meta loop.
The map is not the territory. What happened to You!Earth’s subjective experience is an objective, if perhaps not empirically observable fact. You!Mars’ belief about what happened may or may not correspond with reality.
What if me!Mars, after giving it thought, shakes his head and says “no, that’s not right. I say I’m the same person because I still have a sense of subjective experience, which is separate from memories or shared history, which gives me the belief, real or illusory, that I am the same person from moment to moment, day to day, and which separates me from my clones”?
Do you take his word for it? Do you assume he’s mistaken? Do you assume he’s lying?
Assuming that he acknowledges that clones have a separate identity, or in other words he admits that there can be instances of himself that are not him, then by asserting the same identity as the person that walked into the teleporter, he is making an extrapolation into the past. He is expressing a belief that by whatever definition he is using the person walking into the teleporter meets a standard of meness that the clones do not. Unless the definition under consideration explicitly reference You!Mars’ mental state (e.g. “by definition” he has shared identity with people he remembers having shared identity with), then the validity of that belief is external: it is either true or false. The map is not the territory.
Under an assumption of pattern or causal continuity, for example, it would be explicitly true. For computational continuity it would be false.
If I understood you correctly, then on your account, his claim is simply false, but he isn’t necessarily lying.
Yes?
It seems to follow that he might actually have a sense of subjective experience, which is separate from memories or shared history, which gives him the belief, real or illusory (in this case illusory), that he is the same person from moment to moment, day to day, and the same person who walked into the teleporter, and which separates him from his clones.
If I understood you correctly, then on your account, his claim is simply false, but he isn’t necessarily lying.
Yes, in the sense that it is a belief about his own history which is either true or false like any historical fact. Whether it actually false depends on the nature of “personal identity”. If I understand the original post correctly, I think Eliezer would argue that his claim is true. I think Eliezer’s argument lacks sufficient justification, and there’s a good chance his claim is false.
It seems to follow that he might actually have a sense of subjective experience, which is separate from memories or shared history, which gives him the belief, real or illusory (in this case illusory), that he is the same person from moment to moment, day to day, and the same person who walked into the teleporter, and which separates him from his clones.
Yes. My question is: is that belief justified?
If your memory were altered such to make you think you won the lottery, that doesn’t make you any richer. Likewise You!Mars’ memory was constructed by the transporter machine in such a way, following the transmitted design as to make him remember stepping into the transporter on Earth as you did, and walking out of it on Mars in seamless continuity. But just because he doesn’t remember the deconstruction, information transmission, and reconstruction steps doesn’t mean they didn’t happen. Once he learns what actually happened during his transport, his decision about whether he remains the same person that entered the machine on Earth depends greatly on his model of consciousness and personal identity/continuity.
It seems to follow that he might actually have a sense of subjective experience, which is separate from memories or shared history, which gives him the belief, real or illusory (in this case illusory), that he is the same person from moment to moment, day to day, and the same person who walked into the teleporter, and which separates him from his clones. Yes. My question is: is that belief justified?
That sense of subjective experience separate from memories or shared history is what I have been calling “personal identity.” It is what gives me the belief, real or illusory, that I am the same person from moment to moment, day to day, and what separates me from my clones.
And yet, here’s Dave!Mars, who has a sense of subjective experience separate from memories or shared history which gives him the belief, real or illusory (in this case illusory), that he is the same person from moment to moment, day to day, and the same person who walked into the teleporter, and which separates him from his clones.
But on your account, he might not have Dave’s personal identity.
So, where is this sense of subjective experience coming from, on your account? Is it causally connected to personal identity, or not?
Once he learns what actually happened during his transport, his decision about whether he remains the same person that entered the machine on Earth depends greatly on his model of consciousness and personal identity/continuity.
Yes, that’s certainly true. By the same token, if I convince you that I placed you in stasis last night for… um… long enough to disrupt your personal identity (a minute? an hour? a millisecond? a nanosecond? how long a period of “computational discontinuity” does it take for personal identity to evaporate on your account, anyway?), you would presumably conclude that you aren’t the same person who went to bed last night. OTOH, if I placed you in stasis last night and didn’t tell you, you’d conclude that you’re the same person, and live out the rest of your life none the wiser.
That experiment shows that “personal identity”, whatever that means, follows a time-tree, not a time-line. That conclusion also must hold if MWI is true.
So I get that there’s a tricky (?) labeling problem here, where it’s somewhat controversial which copy of you should be labeled as having your “personal identity”. The thing that isn’t clear to me is why the labeling problem is important. What observable feature of reality depends on the outcome of this labeling problem? We all agree on how those copies of you will act and what beliefs they’ll have. What else is there to know here?
Would you step through the transporter? If you answered no, would it be moral to force you through the transporter? What if I didn’t know your wishes, but had to extrapolate? Under what conditions would it be okay?
Also, take the more vile forms of Pascal’s mugging and acausal trades. If something threatens torture to a simulation of you, should you be concerned about actually experiencing the torture, thereby subverting your rationalist impulse to shut up and multiply utility?
Would you step through the transporter? If you answered no, would it be moral to force you through the transporter? What if I didn’t know your wishes, but had to extrapolate? Under what conditions would it be okay?
I don’t see how any of that depends on the question of which computations (copies of me) get labeled with “personal identity” and which don’t.
Also, take the more vile forms of Pascal’s mugging and acausal trades. If something threatens torture to a simulation of you, should you be concerned about actually experiencing the torture, thereby subverting your rationalist impulse to shut up and multiply utility?
Depending on specifics, yes. But I don’t see how this depends on the labeling question. This just boils down to “what do I expect to experience in the future?” which I don’t see as being related to “personal identity”.
This just boils down to “what do I expect to experience in the future?” which I don’t see as being related to “personal identity”.
Forget the phrase “personal identity”. If I am a powerful AI from the future and I come back to tell you that I will run a simulation of you so we can go bowling together, do you or do you not expect to experience bowling with me in the future, and why?
Suppose that my husband and I believe that while we’re sleeping, someone will paint a blue dot on either my forehead, or my husband’s, determined randomly. We expect to see a blue dot when we wake up… and we also expect not to see a blue dot when we wake up. This is a perfectly reasonable state for two people to be in, and not at all problematic.
Suppose I believe that while I’m sleeping, a powerful AI will duplicate me (if you like, in such a way that both duplicates experience computational continuity with the original) and paint a blue dot on one duplicate’s forehead. When I wake up, I expect to see a blue dot when I wake up… and I also expect not to see a blue dot when I wake up. This is a perfectly reasonable state for a duplicated person to be in, and not at all problematic.
Similarly, I both expect to experience bowling with you, and expect to not experience bowling with you (supposing that the original continues to operate while the simulation goes bowling).
The situation isn’t analogous, however. Let’s posit that you’re still alive when the simulation is ran. In fact, aside from technology there’s no reason to put it in the future or involve an AI. I’m a brain scanning researcher that shows up at your house tomorrow, with all the equipment to do a non-destructive mind upload and whole-brain simulation. I tell you that I am going to scan your brain, start the simulation, then don VR goggles and go virtual-bowling with “you”. Once the scanning is done you and your husband are free to go to the beach or whatever, while I go bowling with TheVirtualDave.
What probability would you put on you ending up bowling instead of at the beach?
Well, let’s call P1 my probability of actually going to the beach, even if you never show up. That is, (1-P1) is the probability that traffic keeps me from getting there, or my car breaks down, or whatever. And let’s call P2 my probability of your VR/simulation rig working. That is, (1-P2) is the probability that the scanner fails, etc. etc.
In your scenario, I put a P1 probability of ending up at the beach, and a P2 probability of ending up bowling. If both are high, then I’m confident that I will do both.
There is no “instead of”. Going to the beach does not prevent me from bowling. Going bowling does not prevent me from going to the beach. Someone will go to the beach, and someone will go bowling, and both of those someones will be me.
As I alluded to in another reply, assuming perfectly reliable scanning, and assuming that you hate losing in bowling to MarkAI, how do you decide whether to go practice bowling or to do something else you like more?
If it’s important to me not to lose in bowling, I practice bowling, since I expect to go bowling. (Assuming uninteresting scanning tech.) If it’s also important to me to show off my rocking abs at the beach, I do sit-ups, since I expect to go to the beach. If I don’t have the time to do both, I make a tradeoff, and I’m not sure exactly how I make that tradeoff, but it doesn’t include assuming that the going to the beach somehow happens more or happens less or anything like that than the going bowling.
Admittedly, this presumes that the bowling-me will go on to live a normal lifetime. If I know the simulation will be turned off right after the bowling match, I might not care so much about winning the bowling match. (Then again, I might care a lot more.) By the same token, if I know the original will be shot tomorrow morning I might not care so much abuot my abs. (Then again, I might care more. I’m really not confident about how the prospect of upcoming death affects my choices; still less how it does so when I expect to keep surviving as well.)
What is your probability that you will wake up tomorrow morning? What is your probability that you will wake up Friday morning? I expect to do both, so my probabilities of those two things add up to ~2.
In Mark’s scenario, I expect to go bowling and I expect to go to the beach. My probabilities of those two things similarly add up to ~2.
I think we have the same model of the situation, but I feel compelled to normalize my probability. A guess as to why:
I can rephrase Mark’s question as, “In 10 hours, will you remember having gone to the beach or having bowled?” (Assume the simulation will continue running!) There’ll be a you that went bowling and a you that went to the beach, but no single you that did both of those things. Your successive wakings example doesn’t have this property.
I suppose I answer 50% to indicate my uncertainty about which future self we’re talking about, since there are two possible referents. Maybe this is unhelpful.
That said, normalizing my probability as though there were only going to be one of me at the end of the process doesn’t seem at all compelling to me. I don’t have any uncertainty about which future self we’re talking about—we’re talking about both of them.
Suppose that you and your husband are planning to take the day off tomorrow, and he is planning to go bowling, and you are planning to go to the beach, and I ask the two of you “what’s y’all’s probability that one of y’all will go bowling, and what’s y’all’s probability that one of y’all will go to the beach?” It seems the correct answers to those questions will add up to more than 1, even though no one person will experience bowling AND going to the beach. In 10 hours, one of you will will remember having gone to the beach, and one will remember having bowled.
This is utterly unproblematic when we’re talking about two people.
In the duplication case, we’re still talking about two people, it’s just that right now they are both me, so I get to answer for both of them. So, in 10 hours, I (aka “one of me”) will remember having gone to the beach. I will also remember having bowled. I will not remember having gone to the beach and having bowled. And my probabilities add up to more than 1.
I recognize that it doesn’t seem that way to you, but it really does seem like the obvious way to think about it to me.
I was asking a disguised question. I really wanted to know: “which of the two future selfs do you identify with, and why?”
Oh, that’s easy. Both of them, equally. Assuming accurate enough simulations etc., of course.
ETA: Why? Well, they’ll both think that they’re me, and I can’t think of a way to disprove the claim of one without also disproving the claim of the other.
ETA: Why? Well, they’ll both think that they’re me, and I can’t think of a way to disprove the claim of one without also disproving the claim of the other.
Any of the models of consciousness-as-continuity would offer a definitive prediction.
Any of the models of consciousness-as-continuity would offer a definitive prediction.
IMO, there literally is no fact of the matter here, so I will bite the bullet and say that any model that supposes there is one is wrong. :) I’ll reconsider if you can point to an objective feature of reality that changes depending on the answer to this. (So-and-so will think it to be immoral doesn’t count!)
PS: The up/down karma vote isn’t a record of what you agree with, but whether a post has been reasonably argued.
It is neither of those things. This isn’t debate club. We don’t have to give people credit for finding the most clever arguments for a wrong position.
I make no comment about the subject of debate is in this context (I don’t know or care which party is saying crazy things about ‘conciousness’). I downvoted the parent specifically because it made a normative assertion about how people should use the karma mechanism which is neither something I support nor an accurate description of an accepted cultural norm. This is an example of voting being used legitimately in a way that is nothing to do with whether the post has been reasonably argued.
I did use the term “reasonably argued” but I didn’t mean clever. Maybe “rationally argued”? By my own algorithm a cleverly argued but clearly wrong argument would not garner an up vote.
I gave you an upvote for explaining your down vote.
I did use the term “reasonably argued” but I didn’t mean clever. Maybe “rationally argued”? By my own algorithm a cleverly argued but clearly wrong argument would not garner an up vote.
You are right, ‘clever’ contains connotations that you wouldn’t intend. I myself have used ‘clever’ as term of disdain and I don’t want to apply that to what you are talking about. Let’s say stick with either of the terms you used and agree that we are talking about arguments that are sound, cogent and reasonable rather than artful rhetoric that exploits known biases in human social behaviour to score persuasion points. I maintain that even then down-votes are sometimes appropriate. Allow me to illustrate.
There are two outwardly indistinguishable boxes with buttons that display heads or tails when pressed. You know that one of the boxes returns true 70% of the time, the other returns heads 40% of the time. A third party, Joe, has experimented with the first box three times and tells you that each time it returned true. This represents an argument that the first box is the “70%” box. Now, assume that I have observed the internals of the boxes and know that the first box is, in fact, the 40% box.
Whether I downvote Joe’s comment depends on many things. Obviously, tone matters a lot, as does my impression of whether Joe’s bias is based on dis-ingenuity or more innocent ignorance. But even in the case when Joe is arguing in good faith there are some cases where a policy attempting to improve the community will advocate downvoting the contribution. For example if there is a significant selection bias in what kind of evidence people like Joe have exposed themselves to then popular perception after such people share their opinions will tend to be even more biased than the individuals alone. In that case downvoting Joe’s comment improves the discussion. The ideal outcome would be for Joe to learn to stfu until he learns more.
More simply I observe that even the most ‘rational’ of arguments can be harmful if the selection process for the creation and repetition of those arguments is at all biased.
I won’t because that’s not what I’m arguing. My position is that subjective experience has moral consequences, and therefore matters.
OK, that’s fine, but I’m not convinced—I’m having trouble thinking of something that I consider to be a moral issue that doesn’t have a corresponding consequence in the territory.
PS: That downvote wasn’t me. I’m aware of how votes work around here. :)
Example: is it moral to power-cycle (hibernate, turn off, power on, restore) a computer running an self-aware AI? WIll future machine intelligences view any less-than-necessary AGI experiments I run the same way we do Josef Mengele’s work in Auschwitz? Is it a possible failure mode that an unfriendly/not-proovably-friendly AI that experiences routine power cycling might uncover this line of reasoning and decide it doesn’t want to “die” every night when the lights go off? What would it do then?
OK, in a hypothetical world where somehow pausing a conscious computation—maintaining all data such that it could be restarted losslessly—is murder, those are concerns. Agreed. I’m not arguing against that.
My position is that pausing a computation as above happens to not be murder/death, and that those who believe it is murder/death are mistaken. The example I’m looking for is something objective that would demonstrate this sort of pausing is murder/death. (In my view, the bad thing about death is its permanence, that’s most of why we care about murder and what makes it a moral issue.)
As Eliezer mentioned in his reply (in different words), if power cycling is death, what’s the shortest suspension time that isn’t? Currently most computers run synchronously off a common clock. The computation is completely suspended between clock cycles. Does this mean that an AI running on such a computer is murdered billions of times every second? If so, then morality leading to this absurd conclusion is not a useful one.
Edit: it’s actually worse than that: digital computation happens mostly within a short time of the clock level switch. The rest of the time between transitions is just to ensure that the electrical signals relax to within their tolerance levels. Which means that the AI in question is likely dead 90% of the time.
What Eliezer and you describe is more analogous to task switching on a timesharing system, and yes my understanding of computational continuity theory is that such a machine would not be sent to oblivion 120 times a second. No, such a computer would be strangely schizophrenic, but also completely self-consistent at any moment in time.
But computational continuity does have a different answer in the case of intermediate non-computational states. For example, saving the state of a whole brain emulation to magnetic disk, shutting off the machine, and restarting it sometime later. In the mean time, shutting off the machine resulted in decoupling/decoherence of state between the computational elements of the machine, and general reversion back to a state of thermal noise. This does equal death-of-identity, and is similar to the transporter thought experiment. The relevance may be more obvious when you think about taking the drive out and loading it in another machine, copying the contents of the disk, or running multiple simulations from a single checkpoint (none of these change the facts, however).
In the mean time, shutting off the machine resulted in decoupling/decoherence of state between the computational elements of the machine, and general reversion back to a state of thermal noise.
It is probably best for you to stay away from the physics/QM point of view on this, since you will lose: the states “between the computational elements”, whatever you may mean by that, decohere and relax to “thermal noise” much quicker than the time between clock transitions, so there no difference between a nanosecond an an hour.
Maybe what you mean is more logic-related? For example, when a self-aware algorithm (including a human) expects one second to pass and instead measures a full hour (because it was suspended), it interprets that discrepancy of inputs as death? If so, shouldn’t any unexpected discrepancy, like sleeping past your alarm clock, or day-dreaming in class, be treated the same way?
This does equal death-of-identity, and is similar to the transporter thought experiment.
I agree that forking a consciousness is not a morally trivial issue, but that’s different from temporary suspension and restarting, which happens all the time to people and machines. I don’t think that conflating the two is helpful.
It is probably best for you to stay away from the physics/QM point of view on this, since you will lose: the states “between the computational elements”, whatever you may mean by that, decohere and relax to “thermal noise” much quicker than the time between clock transitions, so there no difference between a nanosecond an an hour.
Maybe what you mean is more logic-related?...
No, I meant the physical explanation (I am a physicist, btw). It is possible for a system to exhibit features at certain frequencies, whilst only showing noise at others. Think standing waves, for example.
I agree that forking a consciousness is not a morally trivial issue, but that’s different from temporary suspension and restarting, which happens all the time to people and machines. I don’t think that conflating the two is helpful.
When does it ever happen to people? When does your brain, even just regions ever stop functioning, entirely? You do not remember deep sleep because you are not forming memories, not because your brain has stopped functioning. What else could you be talking about?
Hmm, I get a feeling that none of these are your true objections and that, for some reason, you want to equate suspension to death. I should have stayed disengaged from this conversation. I’ll try to do so now. Hope you get your doubts resolved to your satisfaction eventually.
I don’t want to, I just think that the alternatives lead to absurd outcomes that can’t possibly be correct (see my analysis of the teleporter scenario).
I really have a hard time imagining a universe where there exists a thing that is preserved when 10^-9 seconds pass between computational steps but not when 10^3 pass between steps (while I move the harddrive to another box).
Prediction: TheOtherDave will say 50%, Beach!Dave and Bowling!Dave would both consider both to be the “original”. Assuming sufficiently accurate scanning & simulating.
I’ll give a 50% chance that I’ll experience that. (One copy of me continues in the “real” world, another copy of me appears in a simulation and goes bowling.)
(If you ask this question as “the AI is going to run N copies of the bowling simulation”, then I’m not sure how to answer—I’m not sure how to weight N copies of the exact same experience. My intuition is that I should still give a 50% chance, unless the simulations are going to differ in some respect, then I’d give a N/(N+1) chance.)
I need to think about your answer, as right now it doesn’t make any sense to me. I suspect that whatever intuition underlies it is the source of our disagreement/confusion.
@linkhyrule5 had an answer better than the one I had in mind. The probability of us going bowling together is approximately equal to the probability that you are already in said simulation, if computational continuity is what matters.
If there were a 6th Day like service I could sign up for where if anything were to happen to me, a clone/simulation of with my memories would be created, I’d sign up for it in a heartbeat. Because if something were to happen to me I wouldn’t want to deprive my wife of her husband, or my daughters of their father. But that is purely altruistic: I would have P(~0) expectation that I would actually experience that resurrection. Rather, some doppelganger twin that in every outward way behaves like me will take up my life where I left off. And that’s fine, but let’s be clear about the difference.
If you are not the simulation the AI was referring to, then you and it will not go bowling together, period. Because when said bowling occurs, you’ll be dead. Or maybe you’ll be alive and well and off doing other things while the simulation is going on. But under no circumstances should you expect to wake up as the simulation, as we are assuming them to be causally separate.
At least from my way of thinking. I’m not sure I understand yet where you are coming from well enough to predict what you’d expect to experience.
@linkhyrule5 had an answer better than the one I had in mind. The probability of us going bowling together is approximately equal to the probability that you are already in said simulation, if computational continuity is what matters.
You could understand my 50% answer to be expressing my uncertainty as to whether I’m in the simulation or not. It’s the same thing.
I don’t understand what “computational continuity” means. Can you explain it using a program that computes the digits of pi as an example?
Rather, some doppelganger twin that in every outward way behaves like me will take up my life where I left off. And that’s fine, but let’s be clear about the difference.
I think you’re making a distinction that exists only in the map, not in the territory. Can you point to something in the territory that this matters for?
I come back to tell you that I will run a simulation of you so we can go bowling together
Presumably you create a sim-me which includes the experience of having this conversation with you (the AI).
do you or do you not expect to experience bowling with me in the future, and why?
Let me interpret the term “expect” concretely as “I better go practice bowling now, so that sim-me can do well against you later” (assuming I hate losing). If I don’t particularly enjoy bowling and rather do something else, how much effort is warranted vs doing something I like?
The answer is not unambiguous and depends on how much I (meat-me) care about future sim-me having fun and not embarrassing sim-self. If sim-me continues on after meat-me passes away, I care very much about sim-me’s well being. On the other hand, if the sim-me program is halted after the bowling game, then I (meat-me) don’t care much about that sim-loser. After all, meat-me (who will not go bowling) will continue to exist, at least for a while. You might feel differently about sim-you, of course. There is a whole range of possible scenarios here. Feel free to specify one in more detail.
TL;DR: If the simulation will be the only copy of “me” in existence, I act as if I expect to experience bowling.
I’d have to learn more about how general anesthesia works to decide if personal identity would be preserved across on the operating table
Hmm, what about across dreamless sleep? Or fainting? Or falling and hitting your head and losing consciousness for an instant? Would these count as killing one person and creating another? And so be morally net-negative?
If computational continuity is what matters, then no. Just because you have no memory doesn’t mean you didn’t experience it. There is in fact a continuous experience throughout all of the examples you gave, just no new memories being formed. But from the last point you remember (going to sleep, fainting, hitting your head) to when you wake up, you did exist and were running a computational process. From our understanding of neurology you can be certain that there was no interruption of subjective experience of identity, even if you can’t remember what actually happened.
Whether this is also true of general anesthesia depends very much on the biochemistry going on. I admit ignorance here.
OK, I guess I should give up, too. I am utterly unable to relate to whatever it is you mean by “because you have no memory doesn’t mean you didn’t experience it” or “subjective experience of identity, even if you can’t remember what actually happened”.
Did I somehow split into 100 different persons? Or was there in fact 101 separate subjective identities, 1 of which terminated in 2013 and 100 new ones created for the simulations? The latter is a more straight forward explanation, IMHO.
I would say that yes, at T1 there’s one of me, and at T2 there’s 100 of me. I don’t see what makes “there’s 101 of me, one of which terminated at T1” more straightforward than that.
I don’t see what makes “there’s 101 of me, one of which terminated at T1” more straightforward than that.
It’s wrapped up in the question over what happened to that original copy that (maybe?) terminated at T1. Did that original version of you terminate completely and forever? Then I wouldn’t count it among the 100 copies that were created later.
Sure, obviously if it terminated then it isn’t around afterwards. Equally obviously, if it’s around afterwards, it didn’t terminate.
You said your metric for determining which description is accurate was (among other things) simplicity, and you claimed that the “101 − 1” answer is more straightforward (simpler?) than the “100″ answer. You can’t now turn around and say that the reason it’s simpler is because the “101-1” answer is accurate.
Either it’s accurate because it’s simpler, or it’s simpler because it’s accurate, but to assert both at once is illegitimate.
I’ll address this in my sequence, which hopefully I will have time to write. The short answer is that what matters isn’t which explanation of this situation is simpler, requires fewer words, a smaller number, or whatever. What matters is: which general rule is simpler?
Pattern or causal continuity leads to all sorts of weird edge cases, some of which I’ve tried to explain in my examples here, and in other cases fails (mysterious answer) to provide a definitive prediction of subjective experience. There may be other solutions, but computational continuity at the very least provides a simpler model, even if it results in the more “complex” 101-1 answer.
It’s sorta like wave collapse vs many-worlds. Wave collapse is simpler (single world), right? No. Many worlds is the simpler theory because it requires fewer rules, even though it results in a mind-bogglingly more complex and varied multiverse. In this case I think computational continuity in the way I formulated it reduces consciousness down to simple general explanation that dissolves the question with no residual problems.
Kinda like how freewill is what a decision algorithm feels like from the inside, consciousness / subjective experience is what any computational process feels like from the inside. And therefore, when the computational process terminates, so too does the subjective experience.
OK, cool, but now I’m confused. If we’re meaning the same thing, I don’t understand how it can be a question—“not running” isn’t a thing an algorithm can experience; it’s a logical impossibility.
Clearly, your subjective experience today is HONK-screech-bam-oblivion, since all the subjective experiences that come after that don’t happen today in this example… they happen 50 years later.
It is not in the least bit clear to me that this means those subjective experiences aren’t your subjective experiences. You aren’t some epiphenomenal entity that dissipates in the course of those 50 years and therefore isn’t around to experience those experiences when they happen… whatever is having those subjective experiences, whenever it is having them, that’s you.
maybe we should taboo the phrase “personal/subjective identity”.
Sounds like a fine plan, albeit a difficult one. Want to take a shot at it?
EDIT: Ah, you did so elsethread. Cool. Replied there.
How do you go to sleep at night, not knowing if it is the “real you” that wakes up on the other side of consciousness?
Your comment would make more sense to me if I removed the word “not” from the sentence you quote. (Also, if I don’t read past that sentence of someonewrongonthenet’s comment.)
That said, I agree completely that the kinds of vague identity concerns about cryonics that the quoted sentence with “not” removed would be raising would also arise, were one consistent, about routine continuation of existence over time.
There are things that when I go to bed to wake up eight hours later are very nearly preserved but if I woke up sixty years later wouldn’t be, e.g. other people’s memories of me (see I Am a Strange Loop) or the culture of the place where I live (see Good Bye, Lenin!).
(I’m not saying whether this is one of the main reasons why I’m not signed up for cryonics.)
Point.
Hrm.. ambiguous semantics. I took it to imply acceptance of the idea but not elevation of its importance, but I see how it could be interpreted differently. And yes, the rest of the post addresses something completely different. But if I can continue for a moment on the tangent, expanding my comment above (even if it doesn’t apply to the OP):
You actually continue functioning when you sleep, it’s just that you don’t remember details once you wake up. A more useful example for such discussion is general anesthesia, which shuts down the regions of the brain associated with consciousness. If personal identity is in fact derived from continuity of computation, then it is plausible that general anesthesia would result in a “different you” waking up after the operation. The application to cryonics depends greatly on the subtle distinction of whether vitrification (and more importantly, the recovery process) slows downs or stops computation. This has been a source of philosophical angst for me personally, but I’m still a cryonics member.
More troubling is the application to uploading. I haven’t done this yet, but I want my Alcor contract to explicitly forbid uploading as a restoration process, because I am unconvinced that a simulation of my destructively scanned frozen brain would really be a continuation of my personal identity. I was hoping that “Timeless Identity” would address this point, but sadly it punts the issue.
Well, if the idea is unimportant to the OP, presumably that also helps explain how they can sleep at night.
WRT the tangent… my own position wrt preservation of personal identity is that while it’s difficult to articulate precisely what it is that I want to preserve, and I’m not entirely certain there is anything cogent I want to preserve that is uniquely associated with me, I’m pretty sure that whatever does fall in that category has nothing to do with either continuity of computation or similarity of physical substrate. I’m about as sanguine about continuing my existence as a software upload as I am about continuing it as this biological system or as an entirely different biological system, as long as my subjective experience in each case is not traumatically different.
I wrote up about a page-long reply, then realized it probably deserves its own posting. I’ll see if I can get to that in the next day or so. There’s a wide spectrum of possible solutions to the personal identity problem, from physical continuity (falsified) to pattern continuity and causal continuity (described by Eliezer in the OP), to computational continuity (my own view, I think). It’s not a minor point though, whichever view turns out to be correct has immense ramifications for morality and timeless decision theory, among other things...
What relevance does personal identity have to TDT? TDT doesn’t depend on whether the other instances of TDT are in copies of you, or in other people who merely use the same decision theory as you.
It has relevance for the basilisk scenario, which I’m not sure I should say any more about.
When you write up the post, you might want to say a few words about what it means for one of these views to be “correct” or “incorrect.”
Ok I will, but that part is easy enough to state here: I mean correct in the reductionist sense. The simplest explanation which resolves the original question and/or associated confusion, while adding to our predictive capacity and not introducing new confusion.
Mm. I’m not sure I understood that properly; let me echo my understanding of your view back to you and see if I got it.
Suppose I get in something that is billed as a transporter, but which does not preserve computational continuity. Suppose, for example, that it destructively scans my body, sends the information to the destination (a process which is not instantaneous, and during which no computation can take place), and reconstructs an identical body using that information out of local raw materials at my destination.
If it turns out that computational or physical continuity is the correct answer to what preserves personal identity, then I in fact never arrive at my destination, although the thing that gets constructed at the destination (falsely) believes that it’s me, knows what I know, etc. This is, as you say, an issue of great moral concern… I have been destroyed, this new person is unfairly given credit for my accomplishments and penalized for my errors, and in general we’ve just screwed up big time.
Conversely, if it turns out that pattern or causal continuity is the correct answer, then there’s no problem.
Therefore it’s important to discover which of those facts is true of the world.
Yes? This follows from your view? (If not, I apologize; I don’t mean to put up strawmen, I’m genuinely misunderstanding.)
If so, your view is also that if we want to know whether that’s the case or not, we should look for the simplest answer to the question “what does my personal identity comprise?” that does not introduce new confusion and which adds to our predictive capacity. (What is there to predict here?)
Yes?
EDIT: Ah, I just read this post where you say pretty much this. OK, cool; I understand your position.
I don’t know what “computation” or “computational continuity” means if it’s considered to be separate from causal continuity, and I’m not sure other philosophers have any standard idea of this either. From the perspective of the Planck time, your brain is doing extremely slow ‘computations’ right now, it shall stand motionless a quintillion ticks and more before whatever arbitrary threshold you choose to call a neural firing. Or from a faster perspective, the 50 years of intervening time might as well be one clock tick. There can be no basic ontological distinction between fast and slow computation, and aside from that I have no idea what anyone in this thread could be talking about if it’s distinct from causal continuity.
(shrug) It’s Mark’s term and I’m usually willing to make good-faith efforts to use other people’s language when talking to them. And, yes, he seems to be drawing a distinction between computation that occurs with rapid enough updates that it seems continuous to a human observer and computation that doesn’t. I have no idea why he considers that distinction important to personal identity, though… as far as I can tell, the whole thing depends on the implicit idea of identity as some kind of ghost in the machine that dissipates into the ether if not actively preserved by a measurable state change every N microseconds. I haven’t confirmed that, though.
Hypothesis: consciousness is what a physical interaction feels like from the inside.
Importantly, it is a property of the interacting system, which can have various degrees of coherence—a different concept than quantum coherence, which I am still developing: something along the lines of negative-entropic complexity. There is therefore a deep correlation between negentropy and consciousness. Random thermodynamic motion in a gas is about as minimum-conscious as you can get (lots of random interactions, but all short lived and decoherent). A rock is slightly more conscious due to its crystalline structure, but probably leads a rather boring existence (by our standards, at least). And so on, all the way up to the very negentropic primate brain which experiences a high degree of coherent experience that we call “consciousness” or “self.”
I know this sounds like making thinking an ontologically basic concept. It’s rather the reverse—I am building the experience of thinking up from physical phenomenon: consciousness is the experience of organized physical interactions. But I’m not yet convinced of it either. If you throw out the concept of coherent interaction (what I have been calling computation continuity), then it does reduce to causal continuity. But causal continuity does have it’s problems which make me suspect it as not being the final, ultimate answer...
How do you explain the existence of the phenomenon of “feeling like” and of “experience”?
I agree that the grandparent has circumvented addressing the crux of the matter, however I feel (heh) that the notion of “explain” often comes with unrealistic expectations. It bears remembering that we merely describe relationships as succinctly as possible, then that description is the “explanation”.
While we would e.g. expect/hope for there to be some non-contradictory set of descriptions applying to both gravity and quantum phenomena (for which we’d eat a large complexity penalty, since complex but accurate descriptions always beat out simple but inaccurate descriptions; Occam’s Razor applies only to choosing among fitting/not yet falsified descriptions), as soon as we’ve found some pinned-down description in some precise language, there’s no guarantee—or strictly speaking, need—of an even simpler explanation.
A world running according to currently en-vogue physics, plus a box which cannot be described as an extension of said physics, but only in some other way, could in fact be fully explained, with no further explanans for the explanandum.
It seems pretty straightforward to note that there’s no way to “derive” phenomena such as “feeling like” in the current physics framework, except of course to describe which states of matters/energy correspond to which qualia.
Such a description could be the explanation, with nothing further to be explained:
If it empirically turned out that a specific kind matter needs to be arranged in the specific pattern of a vertebrate brain to correlate to qualia, that would “explain” consciousness. If it turned out (as we all expect) that the pattern alone sufficies, then certain classes of instantiated algorithms (regardless of the hardware/wetware) would be conscious. Regardless, either description (if it turned out to be empirically sound) would be the explanation.
I also wonder, what could any answer within the current physics framework possibly look like, other than an asterisk behind the equations with the addendum of “values n1 … nk for parameters p1 … pk correlate with qualia x”?
How do you explain “feeling like” and “experience” in general? This is LW so I assume you have a reductionist background and would offer an explanation based on information patterns, neuron firings, hormone levels, etc. But ultimately all of that reduces down to a big collection of quarks, each taking part in mostly random interactions on the scale of femtoseconds. The apparent organization of the brain is in the map, not the territory. So if subjective experience reduces down to neurons, and neurons reduce down to molecules, and molecules reduce to quarks and leptons, where then does the consciousness reside? “Information patterns” alone is an inadequate answer—that’s at the level of the map, not the territory. Quarks and leptons combine into molecules, molecules into neural synapses, and the neurons connect into the 3lb information processing network that is my brain. Somewhere along the line, the subjective experience of “consciousness” arises. Where, exactly, would you propose that happens?
We know (from our own subjective experience) that something we call “consciousness” exists at the scale of the entire brain. If you assume that the workings of the brain is fully explained by its parts and their connections, and those parts explained by their sub-components and designs, etc. you eventually reach the ontologically basic level of quarks and leptons. Fundamentally the brain is nothing more than the interaction of a large number of quarks and leptons. So what is the precise interaction of fundamental particles is the basic unit of consciousness? What level of complexity is required before simply organic matter becomes a conscious mind?
It sounds ridiculous, but if you assume that quarks and leptons are “conscious,” or rather that consciousness is the interaction of these various ontologically primitive, fundamental particles, a remarkably consistent theory emerges: one which dissolves the mystery of subjective consciousness by explaining it as the mere aggregation of interdependent interactions. Besides being simple, this is also predictive: it allows us to assert for a given situation (e.g. a teleporter or halted simulation) whether loss of personal identity occurs, which has implications for morality of real situations encountered in the construction of an AI.
What do you mean by this? Are fMRIs a big conspiracy?
This description applies equally to all objects. When you describe the brain this way, you leave out all its interesting characteristics, everything that makes it different from other blobs of interacting quarks and leptons.
What I’m saying is that the high-level organization is not ontologically primitive. When we talk about organizational patterns of the brain, or the operation of neural synapses, we’re taking about very high level abstractions. Yes, they are useful abstractions primarily because they ignore unnecessary detail. But that detail is how they are actually implemented. The brain is soup of organic particles with very high rates of particle interaction due simply to thermodynamic noise. At the nanometer and femtosecond scale, there is very little signal to noise, however at the micrometer and millisecond scale general trends start to emerge, phenomenon which form the substrate of our computation. But these high level abstractions don’t actually exist—they are just average approximations over time of lower level, noisy interactions.
I assume you would agree that a normal adult brain in a human experiences a subjective feeling of consciousness that persists from moment-to-moment. I also think it’s a fair bet that you would not think that a single electron bouncing around in some part of a synaptic pathway or electronic transistor has anything resembling a conscious experience. But somehow, a big aggregation of these random motions does add up to you or me. So at what point in the formation of a human brain, or construction of an AI does it become conscious? At what point does it mere dead matter transform into sentience? Is this a hard cutoff? Is it gradual?
Speaking of gradations, certain animals can’t recognize themselves in a mirror. If you use self-awareness as a metric as was argued elsewhere, does that mean they’re not conscious? What about insects, which operate with a more distributed neural system. Dung beetles seem to accomplish most tasks by innate reflex response. Do they have at least a little, tiny subjective experience of consciousness? Or is their existence no more meaningful than that of a stapler?
Yes, this objection applies equally to all objects. That’s precisely my point. Brains are not made of any kind of “mind stuff”—that’s substance dualism which I reject. Furthermore, minds don’t have a subjective experience separate from what is physically explainable—that’s epiphenomenalism, similarly rejected. “Minds exist in information patterns” is a mysterious answer—information patterns are themselves merely evolving expressions in the configuration space of quarks & leptons. Any result of the information pattern must be explainable in terms of the interactions of its component parts, or else we are no longer talking about a reductionist universe. If I am coming at this with a particular bias, it is this: all aspects of mind including consciousness, subjective experience, qualia, or whatever you want to call it are fundamentally reducible to forces acting on elementary particles.
I see only two reductionist paths forward to take: (1) posit a new, fundamental law by which at some aggregate level of complexity or organization, a computational substrate becomes conscious. How & why is not explained, and as far as I can tell there is no experimental way to determine where this cutoff is. But assume it is there. Or, (2) accept that like everything else in the universe, consciousness reduces down to the properties of fundamental particles and their interactions (it is the interaction of particles). A quark and a lepton exchanging a photon is some minimal quantum Plank-level of conscious experience. Yes, that means that even a rock and a stapler experience some level of conscious experience—barely distinguishable from thermal noise, but nonzero—but the payoff is a more predictive reductionist model of the universe. In terms of biting bullets, I think accepting many-worlds took more gumption than this.
This is a Wrong Question. Consciousness, whatever it is, is (P=.99) a result of a computation. My computer exhibits a microsoft word behavior, but if I zoom in to the electrons and transistors in the CPU, I see no such microsoft word nature. It is silly to zoom in to quarks and leptons looking for the true essence of microsoft word. This is the way computations work—a small piece of the computation simply does not display behavior that is like the entire computation. The CPU is not the computation. It is not the atoms of the brain that are conscious, it is the algorithm that they run, and the atoms are not the algorithm. Consciousness is produced by non-conscious things.
Minds exist in some algorithms (“information pattern” sounds too static for my taste). Your desire to reduce things to forces on elementary particles is misguided, I think, because you can do the same computation with many different substrates. The important thing, the thing we care about, is the computation, not the substrate. Sure, you can understand microsoft word at the level of quarks in a CPU executing assembly language, but it’s much more useful to understand it in terms of functions and algorithms.
You’ve completely missed / ignored my point, again. Microsoft Word can be functionally reduced to electrons in transistors. The brain can be functionally reduced to biochemistry. Unless you resort to some form of dualism, the mind (qualia) is also similarly reduced.
just as computation can be brought down to the atomic scale (or smaller, with quantum computing), so too can conscious experiences be constructed out of such computational events. Indeed they are one and the same thing, just viewed from different perspectives.
I thought dualism meant you thought that there was ontologically basic conciousness stuff separate from ordinary matter?
I think the mind should be reduced to algorithms, and biochemistry is an implementation detail. This may make me a dualist by your usage of the word.
I think that it’s equally silly to ask, “where is the microsoft-word-ness” about a subset of transistors in your CPU as it is to ask “where is the consciousness” about a subset of neurons in your brain. I see this as describing how non-ontologically-basic consciousness can be produced by non-conscious stuff.
Apologies; does the above address your point? If not I’m confused about your point.
I’m arguing that if you think the mind can be reduced to algorithms implemented on computational substrate, then it is a logical consequence from our understanding of the rules of physics and the nature of computation that what we call subjective experience must also scale down as you reduce a computational machine down to its parts. After all, the algorithms themselves too also reducible down to stepwise axiomatic logical operations, implemented as transistors or interpretable machine code.
The only way to preserve the common intuition that “it takes (simulation of) a brain or equivalent to produce a mind” is to posit some form of dualism. I don’t think it is silly to ask “where is the microsoft-word-ness” about a subset of a computer—you can for example point to the regions of memory and disk where the spellchecker is located, and say “this is the part that matches user input against tables of linguistic data,” just like we point to regions of the brain and say “this is your language processing centers.”
The experience of having a single, unified me directing my conscious experience is an illusion—it’s what the integration process feels like from the inside, but it does not correspond to reality (we have psychological data to back this up!). I am in fact a society of agents, each simpler but also relying on an entire bureaucracy of other agents in an enormous distributed structure. Eventually though, things reduce down to individual circuits, then ultimately to the level of individual cell receptors and chemical pathways. At no point along the way is there a clear division where it is obvious that conscious experience ends and what follows is merely mechanical, electrical, and chemical processes. In fact as I’ve tried to point out the divisions between higher level abstractions and their messy implementations is in the map, not the territory.
To assert that “this level of algorithmic complexity is a mind, and below that is mere machines” is a retreat to dualism, though you may not yet see it in that way. What you are asserting is that there is this ontologically basic mind-ness which spontaneously emerges when an algorithm has reached a certain level of complexity, but which is not the aggregation of smaller phenomenon.
I think we have really different models of how algorithms and their sub-components work.
Suppose I have a computation that produces the digits of pi. It has subroutines which multiply and add. Is it an accurate description of these subroutines that they have a scaled down property of computes-pi-ness? I think this is not a useful way to understand things. Subroutines do not have a scaled-down percentage of the properties of their containing algorithm, they do a discrete chunk of its work. It’s just madness to say that, e.g., your language processing center is 57% conscious.
I agree with all this. Humans probably are not the minimal conscious system, and there are probably subsets of our component circuitry which maintain the property of conciousness. But yes, I maintain that eventually, you’ll get to an algorithm that is conscious while none of its subroutines are.
If this makes me a dualist then I’m a dualist, but that doesn’t feel right. I mean, the only way you can really explain a thing is to show how it arises from something that’s not like it in the first place, right?
I think we have different models of what consciousness is. In your pi example, the multiplier has multiply-ness, and the adder has add-ness properties, and when combined together in a certain way you get computes-pi-ness. Likewise our minds have many, many, many different components which—somehow, someway—each have a small experiential qualia which when you sum together yield the human condition.
Through brain damage studies, for example, we have descriptions of what it feels like to live without certain mental capabilities. I think you would agree with this, but for others reading take this thought experiment: imagine that I were to systematically shut down portions of your brain, or in simulation, delete regions of your memory space. For the purpose of the argument I do it slowly over time in relatively small amounts, and cleaning up dangling references so the whole system doesn’t shut down. Certainly as time goes by your mental functionality is reduced, and you stop being capable of having experiences you once took for granted. But at what point, precisely, do you stop experiencing at all qualia of any form? When you’re down to just a billion neurons? A million? A thousand? When you’re down to just one processing region? Is one tiny algorithm on a single circuit enough?
What is the minimal conscious system? It’s easy and perhaps accurate to say “I don’t know.” After all, neither one of us know enough neural and cognitive science to make this call, I assume. But we should be able to answer this question: “if presented criteria for a minimally-conscious-system, what would convince me of its validity?”
Eliezer’s post on reductionism is relevant here. In a reductionist universe, anything and everything is fully defined by its constituent elements—no more, no less. There’s a popular phrase that has no place is reductionist theories: “the whole is greater than the sum of its parts.” Typically what this actually means is that you failed to count the “parts” correctly: a part list should also include spatial configurations and initial conditions, which together imply the dynamic behaviors as well. For example, a pulley is more than a hunk of metal and some rope, but it is fully defined if you specify how the metal is shaped, how the rope is threaded through it and fixed to objects with knots, how the whole contraption is oriented with respect to gravity, and the procedure for applying rope-pulling-force. Combined with the fundamental laws of physics, this is a fully reductive explanation of a rope-pulley system which is the sum of its fully-defined parts.
And so it goes with consciousness. Unless we are comfortable with the mysterious answers provided by dualism—or empirical evidence like confirmation of psychic phenomenon compels us to go there—then we must demand that an explanation be provided that explains consciousness fully as the aggregation of smaller processes.
When I look a explanations of the workings of the brain, starting with the highest level psychological theories and neural structure, and working the way all the way down the abstraction hierarchy to individual neural synapses and biochemical pathways, nowhere along the way do I see an obvious place to stop and say “here is where consciousness begins!” Likewise, I can start from the level of mere atoms and work my way up to the full neural architecture, without finding any step that adds something which could be consciousness, but which isn’t fundamentally like the levels below it. But when you get to the highest level, you’ve described the full brain without finding consciousness anywhere along the way.
I can see how this leads otherwise intelligent philosophers like David Chalmers to epiphenomenalism. But I’m not going to go down that path, because the whole situation is the result of mental confusion.
The Standard Rationalist Answer is that mental processes are information patterns, nothing more, and tat consciousness is an illusion, end of story. But that still leaves me confused! It’s not like free will for example, where because of the mind projection fallacy I think I have free will due to how a deterministic decision theory algorithm feels from the inside. I get that. No, the answer of “that subjective experience of consciousness isn’t real, get over it” is unsatisfactory because if I don’t have conscious, how am I experiencing thinking in the first place? Cogito ergo sum.
However there is a way out. I went looking for a source of consciousness because I like nearly every other philosopher assumed that there was something special and unique which set brains aside as having minds which other more mundane objects—like rocks and staplers—do not possess. That’s so obviously true, but honestly I have no real justification for that belief. So let’s try negating it. What is possible if we don’t exclude mundane things from having minds too?
Well, what does it feel like to be a quark and a lepton exchanging a photon? I’m not really sure, but let’s call that approximately the minimum possible “experience”, and for the duration of the interaction continuous interaction over time, the two particles share a “mind”. Arrange a number of these objects together and you get an atom, which itself also has a shared/merged experience so long as the particles remain in bonded interaction. Arrange a lot of atoms together and you get a electrical transistor. Now we’re finally starting to get to a level where I have some idea of what the “shared experience of being a transistor” would be (rather boring, by my standards), and more importantly, it’s clear how that experience is aggregated together from its constituent parts. From here, computing theory takes over as more complex interdependent systems are constructed, each merging experiences together into a shared hive mind, until you reach the level of the human being or AI.
Are you at least following what I’m saying, even if you don’t agree?
That was a very long comment (thank you for your effort) and I don’t think I have the energy to exhaustively go through it.
I believe I follow what you’re saying. It doesn’t make much sense to me, so maybe that belief is false.
I think the fact that if you start with a brain, which is presumably conscious, and zoom in all the way looking for the conciousness boundary, and then start with a quark, which is presumably not conscious, and zoom all the way out to the entire brain, also without finding a consciousness barrier—I think this means that the best we can do at the moment is set upper and lower bounds.
A minimally conscious system—say, something that can convince me that it thinks it is conscious. “echo ‘I’m conscious!’” doesn’t quite cut it, things that recognize themselves in mirrors probably do, and I could go either way on the stuff in between.
I think your reductionism is a little misapplied. My pi-calculating program develops a new property of pi-computation when you put the adders and multipliers together right, but is completely described in terms of adders and multipliers. I expect consciousness to be exactly the same; it’ll be completely described in terms of qualia generating algorithms (or some such), which won’t themselves have the consciousness property.
This is hard to see because the algorithms are written in spaghetti code, in the wiring between neurons. In computer terms, we have access to the I/O system and all the gates in the CPU, but we don’t currently know how they’re connected. Looking at more or fewer of the gates doesn’t help, because the critical piece of information is how they’re connected and what algorithm they implement.
IMO, my guess (P=.65) is that qualia are going to turn out to be something like vectors in a feature space. Under this model, clearly systems incapable of representing such a vector can’t have any qualia at all. Rocks and single molecules, for example.
I indeed have a reductionist background, but I offer no explanation, because I have none. I do not even know what an explanation could possibly look like; but neither do I take that as proof that there cannot be one. The story you tell surrounds the central mystery with many physical details, but even in your own accont of it the mystery remains unresolved:
However much you assert that there must be an explanation, I see here no advance towards actually having one. What does it mean to attribute consciousness to subatomic particles and rocks? Does it predict anything, or does it only predict that we could make predictions about teleporters and simulations if we had a physical explanation of consciousness?
I would imagine that consciousness (in a sense of self-awareness) is the ability to introspect into your own algorithm. The more you understand what makes you tick, rather than mindlessly following the inexplicable urges and instincts, the more conscious you are.
Yes, that is not only 100% accurate, but describes where I’m headed.
I am looking for the simplest explanation of the subjective continuity of personal identity, which either answers or dissolves the question. Further, the explanation should either explain which teleportation scenario is correct (identity transfer, or murder+birth), or satisfactorily explain why it is a meaningless distinction.
If I, the person standing in front of the transporter door, will experience walking on Mars, or oblivion.
Yes, it is perhaps likely that this will never be experimentally observable. That may even be a tautology since we are talking about subjective experience. But still, a reductionist theory of consciousness could provide a simple, easy to understand explanation for the origin of personal identity (e.g., what an computational machine feels like from the inside) and which predicts identity transfer or murder + birth. That would be enough for me, at least as long as there’s not competing equally simple theories.
Well, you certainly won’t experience oblivion, more or less by definition. The question is whether you will experience walking on Mars or not.
But there is no distinct observation to be made in these two cases. That is, we agree that either way there will be an entity having all the observable attributes (both subjective and objective; this is not about experimental proof, it’s about the presence or absence of anything differentially observable by anyone) that Mark Friendebach has, walking on Mars.
So, let me rephrase the question: what observation is there to predict here?
That’s not the direction I was going with this. It isn’t about empirical observation, but rather aspects of morality which depend on subjective experience. The prediction is under what conditions subjective experience terminates. Even if not testable, that is still an important thing to find out, with moral implications.
Is it moral to use a teleporter? From what I can tell, that depends on whether the person’s subjective experience is terminated in the process. From the utility point of view the outcomes are very nearly the same—you’ve murdered one person, but given “birth” to an identical copy in the process. However if the original, now destroyed person didn’t want to die, or wouldn’t have wanted his clone to die, then it’s a net negative.
As I said elsewhere, the teleporter is the easiest way to think of this, but the result has many other implications from general anesthesia, to cryonics, to Pascal’s mugging and the basilisk.
OK. I’m tapping out here. Thanks for your time.
Like TheOtherDave (I presume), I consider my identity to be adequately described by whatever Turing machine that can emulate my brain, or at least its prefrontal cortex + relevant memory storage. I suspect that a faithful simulation of just my Brodmann area 10 coupled with a large chunk of my memories would restore enough of my self-awareness to be considered “me”. This sim-me would probably lose most of my emotions without the rest of the brain, but it is still infinitely better than none.
You’ll need the rest of the brain because these other memories would be distributed throughout the rest of your cortex. The hippocampus only contains recent episodic memories.
If you lost your temporal lobe, for example, you’d lose all non-episodic knowledge concerning what the names of things are, how they are categorized, and what the relationships between them are.
That said, I’m not sure why I should care much about having my non-episodic knowledge replaced with an off-the-shelf encyclopedia module. I don’t identify with it much.
If you only kept the hippocampus, you’d lose your non-recent episodic memories too. But technical issues aside, let me defend the “encyclopedia”:
Episodic memory is basically a cassette reel of your life, along with a few personalized associations and maybe memories of thoughts and emotions. Everything that we associate with the word knowledge is non-episodic. It’s not just verbal labels—that was just a handy example that I happened to know the brain region for. I’d actually care about that stuff more about non-episodic memories than the episodic stuff.
Things like “what is your wife’s name and what does her face look like” are non-episodic memory. You don’t have to think back to a time when you specifically saw your wife to remember what her name and face is, and that you love her—that information is treated as a fact independent of any specific memory, indelibly etched into your model of the world. Cognitively speaking, “I love my wife stacy, she looks like this” is as much of a fact as “grass is a green plant” and they are both non-episodic memories. Your episodic memory reel wouldn’t even make sense without that sort of information. I’d still identify someone with memory loss, but retaining my non-episodic memory, as me. I’d identify someone with only my episodic memories as someone else, looking at a reel of memory that does not belong to them and means nothing to them.
(Trigger Warning: link contains writing in diary which is sad, horrifying, and nonfiction.): This is what complete episodic memory loss looks like. Patients like this can still remember the names of faces of people they love.
Ironically...the (area 10) might actually be replaceable. I’m not sure whether any personalized memories are kept there—I don’t know what that specific region does but it’s in an area that mostly deals with executive function—which is important for personality, but not necessarily individuality.
What’s the difference between personality and individuality?
In my head:
Personality is a set of dichotomous variables plotted on a bell curve. “Einstein was extroverted, charismatic, nonconforming, and prone to absent-mindedness” describes his personality. We all have these traits in various amounts. You can some of these personality nobs really easily with drugs. I can’t specify Einstein out of every person in the world using only his personality traits—I can only specify individuals similar to him.
Individuality is stuff that’s specific to the person. “Einstein’s second marriage was to his cousin and he had at least 6 affairs. He admired Spinoza, and was a contemporary of Tagore. He was a socialist and cared about civil rights. He had always thought there was something wrong about refrigerators.” Not all of these are dichotomous variables—you either spoke to Tagore or you didn’t. And it makes no sense to put people on a “satisfaction with Refrigerators” spectrum, even though I suppose you could if you wanted to. And all this information together specifically points to Einstein, and no one else in the world. Everyone in the world a set of unique traits like fingerprints—and it doesn’t even make sense to ask what the “average” is, since most of the variables don’t exist on the same dimension.
And...well, when it comes to Area 10, just intuitively, do you really want to define yourself by a few variables that influence your executive function? Personally I define myself partially by my ideas, and partially by my values...and the former is definitely in the “individuality” territory.
OK, I understand what you mean by personality vs individuality. However, I doubt that the functionality of BA10 can be described “by a few variables that influence your executive function”. Then again, no one knows anything definite about it.
I take it you’re assuming that information about my husband, and about my relationship to my husband, isn’t in the encyclopedia module along with information about mice and omelettes and your relationship to your wife.
If that’s true, then sure, I’d prefer not to lose that information.
Well...yeah, I was. I thought the whole idea of having an encyclopedia was to eliminate redundancy through standardization of the parts of the brain that were not important for individuality?
If your husband and my husband, your omelette and my omelette, are all stored in the encyclopedia, it wouldn’t be a “off-the-shelf encyclopedia module” anymore. It would be an index containing individual people’s non-episodic knowledge. At that point, it’s just an index of partial uploads. We can’t standardize that encyclopedia to everyone: If the the thing that stores your omelette and your husband went around viewing my episodic reel and knowing all the personal stuff about my omelette and husband...that would be weird and the resulting being would be very confused (let alone if the entire human race was in there—I’m not sure how that would even work).
(Also, going back into the technical stuff, there may or may not be a solid dividing line between very old episodic memory and non-episodoc memory
Sure, if your omelette and my omelette are so distinct that there is no common data structure that can serve as a referent for both, and ditto for all the other people in the world, then the whole idea of an encyclopedia falls apart. But that doesn’t seem terribly likely to me.
Your concept of an omelette probably isn’t exactly isomorphic to mine, but there’s probably a parametrizable omelette data structure we can construct that, along with a handful of parameter settings for each individual, can capture everyone’s omelette. The parameter settings go in the representation of the individual; the omelette data structure goes in the encyclopedia.
And, in addition, there’s a bunch of individualizing episodic memory on top of that… memories of cooking particular omelettes, of learning to cook an omelette, of learning particular recipes, of that time what ought to have been an omelette turned into a black smear on the pan, etc. And each of those episodic memories refers to the shared omelette data structure, but is stored with and is unique to the uploaded agent. (Maybe. It may turn out that our individual episodic memories have a lot in common as well, such that we can store a standard lifetime’s memories in the shared encyclopedia and just store a few million bits of parameter settings in each individual profile. I suspect we overestimate how unique our personal narratives are, honestly.)
Similarly, it may be that our relationships with our husbands are so distinct that there is no common data structure that can serve as a referent for both. But that doesn’t seem terribly likely to me. Your relationship with your husband isn’t exactly isomorphic to mine, of course, but it can likely similarly be captured by a common parameterizable relationship-to-husband data structure.
As for the actual individual who happens to be my husband, well, the majority of the information about him is common to all kinds of relationships with any number of people. He is his father’s son and his stepmother’s stepson and my mom’s son-in-law and so on and so forth. And, sure, each of those people knows different things, but they know those things about the same person; there is a central core. That core goes in the encyclopedia, and pointers to what subset each person knows about him goes in their individual profiles (along with their personal experiences and whatever idiosyncratic beliefs they have about him).
So, yes, I would say that your husband and my husband and your omelette and my omelette are all stored in the encyclopedia. You can call that an index of partial uploads if you like, but it fails to incorporate whatever additional computations that create first-person experience. It’s just a passive data structure.
Incidentally and unrelatedly, I’m not nearly as committed as you sound to preserving our current ignorance of one another’s perspective in this new architecture.
I’m really skeptical that parametric functions which vary on dimensions concerning omelets (Egg species? Color? ingredients? How does this even work?) are a more efficient or more accurate way of preserving what our wetware encode when compared to simulating the neural networks devoted dealing with omelettes. I wouldn’t even know how to start working on the problem mapping a conceptual representation of an omelette into parametric functions (unless we’re just using the parametric functions to model the properties of individual neurons—that’s fine).
Can you give an example concerning what sort of dimension you would parametrize so I have a better idea of what you mean?
I was more worried that it might break stuff (as in, resulting beings would need to be built quite differently in order to function) if one-another’s perspectives would overlap. Also, that brings us back to the original question I was raising about living forever—what exactly is it that we value and want to preserve?
Not really. If I were serious about implementing this, I would start collecting distinct instances of omelette-concepts and analyzing them for variation, but I’m not going to do that. My expectation is that if I did, the most useful dimensions of variability would not map to any attributes that we would ordinarily think of or have English words for.
Perhaps what I have in mind can be said more clearly this way: there’s a certain amount of information that picks out the space of all human omelette-concepts from the space of all possible concepts… call that bitstring S1. There’s a certain amount of information that picks out the space of my omelette-concept from the space of all human omelette-concepts… call that bitstring S2.
S2 is much, much, shorter than S1.
It’s inefficient to have 7 billion human minds each of which is taking up valuable bits storing 7 billion copies of S1 along with their individual S2s. Why in the world would we do that, positing an architecture that didn’t physically require it? Run a bloody compression algorithm, store S1 somewhere, have each human mind refer to it.
I have no idea what S1 or S2 are.
And I don’t expect that they’re expressible in words, any more than I can express which pieces of a movie are stored as indexed substrings… it’s not like MPEG compression of a movie of an auto race creates an indexed “car” data structure with parameters representing color, make, model, etc. It just identifies repeated substrings and indexes them, and takes advantage of the fact that sequential frames share many substrings in common if properly parsed.
But I’m committed enough to a computational model of human concept storage that I believe they exist. (Of course, it’s possible that our concept-space of an omelette simply can’t be picked out by a bit-string, but I can’t see why I should take that possibility seriously.)
Oh, and agreed that we would change if we were capable of sharing one another’s perspectives.
I’m not particularly interested in preserving my current cognitive isolation from other humans, though… I value it, but I value it less than I value the ability to easily share perspectives, and they seem to be opposed values.
I think I’ve got a good response for this one.
My non-episodic memory contains the “facts” that Buffy the Vampire Slayer was one of the best television shows that was ever made, and the Pink Floyd aren’t an interesting band. My boyfriend’s non-episodic memory contains the facts that Buffy was boring, unoriginal, and repetetive (and that Pink Floyd’s music is trancendentally good).
Objectively, these are opinions, not facts. But we experience them as facts. If I want to preserve my sense of identity, then I would need to retain the facts that were in my non-episodic memory. More than that, I would also lose my sense of self if I gained contradictory memories. I would need to have my non-episodic memories and not have the facts from my boyfriend’s memory.
That’s the reason why “off the shelf” doesn’t sound suitable in this context.
So, on one level, my response to this is similar to the one I gave (a few years ago) [http://lesswrong.com/lw/qx/timeless_identity/9trc]… I agree that there’s a personal relationship with BtVS, just like there’s a personal relationship with my husband, that we’d want to preserve if we wanted to perfectly preserve me.
I was merely arguing that the bitlength of that personal information is much less than the actual information content of my brain, and there’s a great deal of compression leverage to be gained by taking the shared memories of BtVS out of both of your heads (and the other thousands of viewers) and replacing them with pointers to a common library representation of the show and then have your personal relationship refer to the common library representation rather than your private copy.
The personal relationship remains local and private, but it takes up way less space than your mind currently does.
That said… coming back to this conversation after three years, I’m finding I just care less and less about preserving whatever sense of self depends on these sorts of idiosyncratic judgments.
I mean, when you try to recall a BtVS episode, your memory is imperfect… if you watch it again, you’ll uncover all sorts of information you either forgot or remembered wrong. If I offered to give you perfect eideitic recall of BtVS—no distortion of your current facts about the goodness of it, except insofar as those facts turn out to be incompatible with an actual perception (e.g., you’d have changed your mind if you watched it again on TV, too) -- would you take it?
I would. I mean, ultimately, what does it matter if I replace my current vague memory of the soap opera Spike was obsessively watching with a more specific memory of its name and whatever else we learned about it? Yes, that vague memory is part of my unique identity, I guess, in that nobody else has quite exactly that vague memory… but so what? That’s not enough to make it worth preserving.
And for all I know, maybe you agree with me… maybe you don’t want to preserve your private “facts” about what kind of tie Giles was wearing when Angel tortured him, etc., but you draw the line at losing your private “facts” about how good the show was. Which is fine, you care about what you care about.
But if you told me right now that I’m actually an upload with reconstructed memories, and that there was a glitch such that my current “facts” about BTVS being a good show for its time is mis-reconstructed, and Dave before he died thought it was mediocre… well, so what?
I mean, before my stroke, I really disliked peppers. After my stroke, peppers tasted pretty good. This was startling, but it posed no sort of challenge to my sense of self.
Apparently (Me + likes peppers) ~= (Me + dislikes peppers) as far as I’m concerned.
I suspect there’s a million other things like that.
There’s a very wide range of possible minds I consider to preserve my identity; I’m not sure the majority of those emulate my prefrontal cortex significantly more closely than they emulate yours, and the majority of my memories are not shared by the majority of those minds.
Interesting. I wonder what you would consider a mind that preserves your identity. For example, I assume that the total of your posts online, plus whatever other information available without some hypothetical future brain scanner, all running as a process on some simulator, is probably not enough.
At one extreme, if I assume those posts are being used to create a me-simulation by me-simulation-creator that literally knows nothing else about humans, then I’m pretty confident that the result is nothing I would identify with. (I’m also pretty sure this scenario is internally inconsistent.)
At another extreme, if I assume the me-simulation-creator has access to a standard template for my general demographic and is just looking to customize that template sufficiently to pick out some subset of the volume of mindspace my sufficiently preserved identity defines… then maybe. I’d have to think a lot harder about what information is in my online posts and what information would plausibly be in such a template to even express a confidence interval about that.
That said, I’m certainly not comfortable treating the result of that process as preserving “me.”
Then again I’m also not comfortable treating the result of living a thousand years as preserving “me.”
Because the notion of “me” is not an ontologically basic category and the question of whether the “real me” wakes up is a question that aught to be un-asked.
I’m a bit confused at the question...you articulated my intent with that sentence perfectly in your other post.
and, as TheOtherDave said,
EDIT: Nevermind, I now understand which part of my statement you misunderstood.
I’m not accepting-but-not-elevating the idea that the ’Real me” doesn’t wake up on the other side. Rather, I’m saying that the questions of personal identity over time do not make sense in the first place. It’s like asking “which color is the most moist”?
The root of your philosophical dilemma is that “personal identity” is a conceptual substitution for soul—a subjective thread that connects you over space and time.
No such thing exists. There is no specific location in your brain which is you. There is no specific time point which is you. Subjective experience exists only in the fleeting present. The only “thread’ connecting you to your past experiences is your current subjective experience of remembering them. That’s all.
I always wonder how I should treat my future self if I reject the continuity of self. Should I think of him like a son? A spouse? A stranger? Should I let him get fat? Not get him a degree? Invest in stock for him? Give him another child?
I think it matters in so far as assisting your present trajectory. Otherwise it might as well be an unfeeling entity.
I have a strong subjective experience of moment-to-moment continuity, even if only in the fleeting present. Simply saying “no such thing exists” doesn’t do anything to resolve the underlying confusion. If no such thing as personal identity exists, then why do I experience it? What is the underlying insight that eliminates the question?
This is not an abstract question either. It has huge implications for the construction of timeless decision theory and utilitarian metamorality.
“a strong subjective experience of moment-to-moment continuity” is an artifact of the algorithm your brain implements. It certainly exists in as much as the algorithm itself exists. So does your personal identity. If in the future it becomes possible to run the same algorithm on a different hardware, it will still produce this sense of personal identity and will feel like “you” from the inside.
Yes, I’m not questioning whether a future simulation / emulation of me would have an identical subjective experience. To reject that would be a retreat to epiphenomenalism.
Let me rephrase the question, so as to expose the problem: if I were to use advanced technology to have my brain scanned today, then got hit by a bus and cremated, and then 50 years from now that brain scan is used to emulate me, what would my subjective experience be today? Do I experience “HONK Screeeech, bam” then wake up in a computer, or is it “HONK Screeeech, bam” and oblivion?
Yes, I realize that in both cases result in a computer simulation of Mark in 2063 claiming to have just woken up in the brain scanner, with a subjective feeling of continuity. But is that belief true? In the two situations there’s a very different outcome for the Mark of 2013. If you can’t see that, then I think we are talking about different things, and maybe we should taboo the phrase “personal/subjective identity”.
Ah, hopefully I’m slowly getting what you mean. So, there was the original you, Mark 2013, whose algorithm was terminated soon after it processed the inputs “HONK Screeeech, bam”, and the new you, Mark 2063, whose experience is “HONK Screeeech, bam” then “wake up in a computer”. You are concerned with… I’m having trouble articulated what exactly… something about the lack of experiences of Mark 2013? But, say, if Mark 2013 was restored to life in mostly the same physical body after a 50-year “oblivion”, you wouldn’t be?
Pretty much correct. To be specific, if computational continuity is what matters, then Mark!2063 has my memories, but was in fact “born” the moment the simulation started, 50 years in the future. That’s when his identity began, whereas mine ended when I died in 2013.
This seems a little more intuitive when you consider switching on 100 different emulations of me at the same time. Did I somehow split into 100 different persons? Or was there in fact 101 separate subjective identities, 1 of which terminated in 2013 and 100 new ones created for the simulations? The latter is a more straight forward explanation, IMHO.
No, that would make little difference as it’s pretty clear that physical continuity is an illusion. If pattern or causal continuity were correct, then it’d be fine, but both theories introduce other problems. If computational continuity is correct, then a reconstructed brain wouldn’t be me any more than a simulation would. However it’s possible that my cryogenically vitrified brain would preserve identity, if it were slowly brought back online without interruption.
I’d have to learn more about how general anesthesia works to decide if personal identity would be preserved across on the operating table (until then, it scares the crap out of me). Likewise, a AI or emulation running on a computer that is powered off and then later resumed would also break identity, but depending on the underlying nature of computation & subjective experience, task switching and online suspend/resume may or may not result in cycling identity.
I’ll stop there because I’m trying to formulate all these thoughts into a longer post, or maybe a sequence of posts.
Can you taboo “personal identity”? I don’t understand what important thing you could lose by going under general anesthesia.
It’s easier to explain in the case of multiple copies of yourself. Imagine the transporter were turned into a replicator—it gets stuck in a loop reconstructing the last thing that went through it, namely you. You step off and turn around to find another version of you just coming out. And then another, and another, etc. Each one of you shares the same memories, but from that moment on you have diverged. Each clone continues life with their own subjective experience until that experience is terminated by that clone’s death.
That sense of subjective experience separate from memories or shared history is what I have been calling “personal identity.” It is what gives me the belief, real or illusory, that I am the same person from moment to moment, day to day, and what separates me from my clones. You are welcome to suggest a better term.
The replicator / clone thought experiment shows that “subjective experience of identity” is something different from the information pattern that represents your mind. There is something, although at this moment that something is not well defined, which makes you the same “you” that will exist five minutes in the future, but which separates you from the “you”s that walked out of the replicator, or exist in simulation, for example.
The first step is recognizing this distinction. Then turn around and apply it to less fantastical situations. If the clone is “you” but not you (meaning no shared identity, and my apologies for the weak terminology), then what’s to say that a future simulation of “you” would also be you? What about cryonics, will your unfrozen brain still be you? That might depend on what they do to repair damage from vitrification. What about general anesthesia? Again, I need to learn more about how general anesthesia works, but if they shut down your processing centers and then restart you later, how is that different from the teleportation or simulation scenario? After all we’ve already established that whatever provides personal identity, it’s not physical continuity.
Well, OK. So suppose that, after I go through that transporter/replicator, you ask the entity that comes out whether it has the belief, real or illusory, that it is the same person in this moment that it was at the moment it walked into the machine, and it says “yes”.
If personal identity is what creates that belief, and that entity has that belief, it follows that that entity shares my personal identity… doesn’t it?
Not quite. If You!Mars gave it thought before answering, his thinking probably went like this: “I have memories of going into the transporter, just a moment ago. I have a continuous sequence of memories, from then until now. Nowhere in those memories does my sense of self change. Right now I am experiencing the same sense of self I always remember experiencing, and laying down new memories. Ergo, proof by backwards induction I am the same person that walked into the teleporter.” However for that—or any—line of meta reasoning to hold, (1) your memories need to accurately correspond with the true and full history of reality and (2) you need trust that what occurs in the present also occurred in the past. In other words, it’s kinda like saying “my memory wasn’t altered because I would have remembered that.” It’s not a circular argument per se, but it is a meta loop.
The map is not the territory. What happened to You!Earth’s subjective experience is an objective, if perhaps not empirically observable fact. You!Mars’ belief about what happened may or may not correspond with reality.
What if me!Mars, after giving it thought, shakes his head and says “no, that’s not right. I say I’m the same person because I still have a sense of subjective experience, which is separate from memories or shared history, which gives me the belief, real or illusory, that I am the same person from moment to moment, day to day, and which separates me from my clones”?
Do you take his word for it?
Do you assume he’s mistaken?
Do you assume he’s lying?
Assuming that he acknowledges that clones have a separate identity, or in other words he admits that there can be instances of himself that are not him, then by asserting the same identity as the person that walked into the teleporter, he is making an extrapolation into the past. He is expressing a belief that by whatever definition he is using the person walking into the teleporter meets a standard of meness that the clones do not. Unless the definition under consideration explicitly reference You!Mars’ mental state (e.g. “by definition” he has shared identity with people he remembers having shared identity with), then the validity of that belief is external: it is either true or false. The map is not the territory.
Under an assumption of pattern or causal continuity, for example, it would be explicitly true. For computational continuity it would be false.
If I understood you correctly, then on your account, his claim is simply false, but he isn’t necessarily lying.
Yes?
It seems to follow that he might actually have a sense of subjective experience, which is separate from memories or shared history, which gives him the belief, real or illusory (in this case illusory), that he is the same person from moment to moment, day to day, and the same person who walked into the teleporter, and which separates him from his clones.
Yes?
Yes, in the sense that it is a belief about his own history which is either true or false like any historical fact. Whether it actually false depends on the nature of “personal identity”. If I understand the original post correctly, I think Eliezer would argue that his claim is true. I think Eliezer’s argument lacks sufficient justification, and there’s a good chance his claim is false.
Yes. My question is: is that belief justified?
If your memory were altered such to make you think you won the lottery, that doesn’t make you any richer. Likewise You!Mars’ memory was constructed by the transporter machine in such a way, following the transmitted design as to make him remember stepping into the transporter on Earth as you did, and walking out of it on Mars in seamless continuity. But just because he doesn’t remember the deconstruction, information transmission, and reconstruction steps doesn’t mean they didn’t happen. Once he learns what actually happened during his transport, his decision about whether he remains the same person that entered the machine on Earth depends greatly on his model of consciousness and personal identity/continuity.
OK, understood.
Here’s my confusion: a while back, you said:
And yet, here’s Dave!Mars, who has a sense of subjective experience separate from memories or shared history which gives him the belief, real or illusory (in this case illusory), that he is the same person from moment to moment, day to day, and the same person who walked into the teleporter, and which separates him from his clones.
But on your account, he might not have Dave’s personal identity.
So, where is this sense of subjective experience coming from, on your account? Is it causally connected to personal identity, or not?
Yes, that’s certainly true. By the same token, if I convince you that I placed you in stasis last night for… um… long enough to disrupt your personal identity (a minute? an hour? a millisecond? a nanosecond? how long a period of “computational discontinuity” does it take for personal identity to evaporate on your account, anyway?), you would presumably conclude that you aren’t the same person who went to bed last night. OTOH, if I placed you in stasis last night and didn’t tell you, you’d conclude that you’re the same person, and live out the rest of your life none the wiser.
That experiment shows that “personal identity”, whatever that means, follows a time-tree, not a time-line. That conclusion also must hold if MWI is true.
So I get that there’s a tricky (?) labeling problem here, where it’s somewhat controversial which copy of you should be labeled as having your “personal identity”. The thing that isn’t clear to me is why the labeling problem is important. What observable feature of reality depends on the outcome of this labeling problem? We all agree on how those copies of you will act and what beliefs they’ll have. What else is there to know here?
Would you step through the transporter? If you answered no, would it be moral to force you through the transporter? What if I didn’t know your wishes, but had to extrapolate? Under what conditions would it be okay?
Also, take the more vile forms of Pascal’s mugging and acausal trades. If something threatens torture to a simulation of you, should you be concerned about actually experiencing the torture, thereby subverting your rationalist impulse to shut up and multiply utility?
I don’t see how any of that depends on the question of which computations (copies of me) get labeled with “personal identity” and which don’t.
Depending on specifics, yes. But I don’t see how this depends on the labeling question. This just boils down to “what do I expect to experience in the future?” which I don’t see as being related to “personal identity”.
Forget the phrase “personal identity”. If I am a powerful AI from the future and I come back to tell you that I will run a simulation of you so we can go bowling together, do you or do you not expect to experience bowling with me in the future, and why?
Yes, with probability P(simulation), or no, with probability P(not simulation), depending.
Suppose that my husband and I believe that while we’re sleeping, someone will paint a blue dot on either my forehead, or my husband’s, determined randomly. We expect to see a blue dot when we wake up… and we also expect not to see a blue dot when we wake up. This is a perfectly reasonable state for two people to be in, and not at all problematic.
Suppose I believe that while I’m sleeping, a powerful AI will duplicate me (if you like, in such a way that both duplicates experience computational continuity with the original) and paint a blue dot on one duplicate’s forehead. When I wake up, I expect to see a blue dot when I wake up… and I also expect not to see a blue dot when I wake up. This is a perfectly reasonable state for a duplicated person to be in, and not at all problematic.
Similarly, I both expect to experience bowling with you, and expect to not experience bowling with you (supposing that the original continues to operate while the simulation goes bowling).
The situation isn’t analogous, however. Let’s posit that you’re still alive when the simulation is ran. In fact, aside from technology there’s no reason to put it in the future or involve an AI. I’m a brain scanning researcher that shows up at your house tomorrow, with all the equipment to do a non-destructive mind upload and whole-brain simulation. I tell you that I am going to scan your brain, start the simulation, then don VR goggles and go virtual-bowling with “you”. Once the scanning is done you and your husband are free to go to the beach or whatever, while I go bowling with TheVirtualDave.
What probability would you put on you ending up bowling instead of at the beach?
Well, let’s call P1 my probability of actually going to the beach, even if you never show up. That is, (1-P1) is the probability that traffic keeps me from getting there, or my car breaks down, or whatever. And let’s call P2 my probability of your VR/simulation rig working. That is, (1-P2) is the probability that the scanner fails, etc. etc.
In your scenario, I put a P1 probability of ending up at the beach, and a P2 probability of ending up bowling. If both are high, then I’m confident that I will do both.
There is no “instead of”. Going to the beach does not prevent me from bowling. Going bowling does not prevent me from going to the beach. Someone will go to the beach, and someone will go bowling, and both of those someones will be me.
As I alluded to in another reply, assuming perfectly reliable scanning, and assuming that you hate losing in bowling to MarkAI, how do you decide whether to go practice bowling or to do something else you like more?
If it’s important to me not to lose in bowling, I practice bowling, since I expect to go bowling. (Assuming uninteresting scanning tech.)
If it’s also important to me to show off my rocking abs at the beach, I do sit-ups, since I expect to go to the beach.
If I don’t have the time to do both, I make a tradeoff, and I’m not sure exactly how I make that tradeoff, but it doesn’t include assuming that the going to the beach somehow happens more or happens less or anything like that than the going bowling.
Admittedly, this presumes that the bowling-me will go on to live a normal lifetime. If I know the simulation will be turned off right after the bowling match, I might not care so much about winning the bowling match. (Then again, I might care a lot more.) By the same token, if I know the original will be shot tomorrow morning I might not care so much abuot my abs. (Then again, I might care more. I’m really not confident about how the prospect of upcoming death affects my choices; still less how it does so when I expect to keep surviving as well.)
Your probabilities add up to more than 1...
Of course they do. Why shouldn’t they?
What is your probability that you will wake up tomorrow morning?
What is your probability that you will wake up Friday morning?
I expect to do both, so my probabilities of those two things add up to ~2.
In Mark’s scenario, I expect to go bowling and I expect to go to the beach.
My probabilities of those two things similarly add up to ~2.
I think we have the same model of the situation, but I feel compelled to normalize my probability. A guess as to why:
I can rephrase Mark’s question as, “In 10 hours, will you remember having gone to the beach or having bowled?” (Assume the simulation will continue running!) There’ll be a you that went bowling and a you that went to the beach, but no single you that did both of those things. Your successive wakings example doesn’t have this property.
I suppose I answer 50% to indicate my uncertainty about which future self we’re talking about, since there are two possible referents. Maybe this is unhelpful.
Yes, that seems to be what’s going on.
That said, normalizing my probability as though there were only going to be one of me at the end of the process doesn’t seem at all compelling to me. I don’t have any uncertainty about which future self we’re talking about—we’re talking about both of them.
Suppose that you and your husband are planning to take the day off tomorrow, and he is planning to go bowling, and you are planning to go to the beach, and I ask the two of you “what’s y’all’s probability that one of y’all will go bowling, and what’s y’all’s probability that one of y’all will go to the beach?” It seems the correct answers to those questions will add up to more than 1, even though no one person will experience bowling AND going to the beach. In 10 hours, one of you will will remember having gone to the beach, and one will remember having bowled.
This is utterly unproblematic when we’re talking about two people.
In the duplication case, we’re still talking about two people, it’s just that right now they are both me, so I get to answer for both of them. So, in 10 hours, I (aka “one of me”) will remember having gone to the beach. I will also remember having bowled. I will not remember having gone to the beach and having bowled. And my probabilities add up to more than 1.
I recognize that it doesn’t seem that way to you, but it really does seem like the obvious way to think about it to me.
I think your description is coherent and describes the same model of reality I have. :)
Yes. Probabilities aside, this is what I was asking.
I was asking a disguised question. I really wanted to know: “which of the two future selfs do you identify with, and why?”
Oh, that’s easy. Both of them, equally. Assuming accurate enough simulations etc., of course.
ETA: Why? Well, they’ll both think that they’re me, and I can’t think of a way to disprove the claim of one without also disproving the claim of the other.
Any of the models of consciousness-as-continuity would offer a definitive prediction.
IMO, there literally is no fact of the matter here, so I will bite the bullet and say that any model that supposes there is one is wrong. :) I’ll reconsider if you can point to an objective feature of reality that changes depending on the answer to this. (So-and-so will think it to be immoral doesn’t count!)
I won’t because that’s not what I’m arguing. My position is that subjective experience has moral consequences, and therefore matters.
PS: The up/down karma vote isn’t a record of what you agree with, but whether a post has been reasonably argued.
For many people, the up/down karma vote is a record of what we want more/less of.
It is neither of those things. This isn’t debate club. We don’t have to give people credit for finding the most clever arguments for a wrong position.
I make no comment about the subject of debate is in this context (I don’t know or care which party is saying crazy things about ‘conciousness’). I downvoted the parent specifically because it made a normative assertion about how people should use the karma mechanism which is neither something I support nor an accurate description of an accepted cultural norm. This is an example of voting being used legitimately in a way that is nothing to do with whether the post has been reasonably argued.
I did use the term “reasonably argued” but I didn’t mean clever. Maybe “rationally argued”? By my own algorithm a cleverly argued but clearly wrong argument would not garner an up vote.
I gave you an upvote for explaining your down vote.
You are right, ‘clever’ contains connotations that you wouldn’t intend. I myself have used ‘clever’ as term of disdain and I don’t want to apply that to what you are talking about. Let’s say stick with either of the terms you used and agree that we are talking about arguments that are sound, cogent and reasonable rather than artful rhetoric that exploits known biases in human social behaviour to score persuasion points. I maintain that even then down-votes are sometimes appropriate. Allow me to illustrate.
There are two outwardly indistinguishable boxes with buttons that display heads or tails when pressed. You know that one of the boxes returns true 70% of the time, the other returns heads 40% of the time. A third party, Joe, has experimented with the first box three times and tells you that each time it returned true. This represents an argument that the first box is the “70%” box. Now, assume that I have observed the internals of the boxes and know that the first box is, in fact, the 40% box.
Whether I downvote Joe’s comment depends on many things. Obviously, tone matters a lot, as does my impression of whether Joe’s bias is based on dis-ingenuity or more innocent ignorance. But even in the case when Joe is arguing in good faith there are some cases where a policy attempting to improve the community will advocate downvoting the contribution. For example if there is a significant selection bias in what kind of evidence people like Joe have exposed themselves to then popular perception after such people share their opinions will tend to be even more biased than the individuals alone. In that case downvoting Joe’s comment improves the discussion. The ideal outcome would be for Joe to learn to stfu until he learns more.
More simply I observe that even the most ‘rational’ of arguments can be harmful if the selection process for the creation and repetition of those arguments is at all biased.
OK, that’s fine, but I’m not convinced—I’m having trouble thinking of something that I consider to be a moral issue that doesn’t have a corresponding consequence in the territory.
PS: That downvote wasn’t me. I’m aware of how votes work around here. :)
Example: is it moral to power-cycle (hibernate, turn off, power on, restore) a computer running an self-aware AI? WIll future machine intelligences view any less-than-necessary AGI experiments I run the same way we do Josef Mengele’s work in Auschwitz? Is it a possible failure mode that an unfriendly/not-proovably-friendly AI that experiences routine power cycling might uncover this line of reasoning and decide it doesn’t want to “die” every night when the lights go off? What would it do then?
OK, in a hypothetical world where somehow pausing a conscious computation—maintaining all data such that it could be restarted losslessly—is murder, those are concerns. Agreed. I’m not arguing against that.
My position is that pausing a computation as above happens to not be murder/death, and that those who believe it is murder/death are mistaken. The example I’m looking for is something objective that would demonstrate this sort of pausing is murder/death. (In my view, the bad thing about death is its permanence, that’s most of why we care about murder and what makes it a moral issue.)
As Eliezer mentioned in his reply (in different words), if power cycling is death, what’s the shortest suspension time that isn’t? Currently most computers run synchronously off a common clock. The computation is completely suspended between clock cycles. Does this mean that an AI running on such a computer is murdered billions of times every second? If so, then morality leading to this absurd conclusion is not a useful one.
Edit: it’s actually worse than that: digital computation happens mostly within a short time of the clock level switch. The rest of the time between transitions is just to ensure that the electrical signals relax to within their tolerance levels. Which means that the AI in question is likely dead 90% of the time.
What Eliezer and you describe is more analogous to task switching on a timesharing system, and yes my understanding of computational continuity theory is that such a machine would not be sent to oblivion 120 times a second. No, such a computer would be strangely schizophrenic, but also completely self-consistent at any moment in time.
But computational continuity does have a different answer in the case of intermediate non-computational states. For example, saving the state of a whole brain emulation to magnetic disk, shutting off the machine, and restarting it sometime later. In the mean time, shutting off the machine resulted in decoupling/decoherence of state between the computational elements of the machine, and general reversion back to a state of thermal noise. This does equal death-of-identity, and is similar to the transporter thought experiment. The relevance may be more obvious when you think about taking the drive out and loading it in another machine, copying the contents of the disk, or running multiple simulations from a single checkpoint (none of these change the facts, however).
It is probably best for you to stay away from the physics/QM point of view on this, since you will lose: the states “between the computational elements”, whatever you may mean by that, decohere and relax to “thermal noise” much quicker than the time between clock transitions, so there no difference between a nanosecond an an hour.
Maybe what you mean is more logic-related? For example, when a self-aware algorithm (including a human) expects one second to pass and instead measures a full hour (because it was suspended), it interprets that discrepancy of inputs as death? If so, shouldn’t any unexpected discrepancy, like sleeping past your alarm clock, or day-dreaming in class, be treated the same way?
I agree that forking a consciousness is not a morally trivial issue, but that’s different from temporary suspension and restarting, which happens all the time to people and machines. I don’t think that conflating the two is helpful.
No, I meant the physical explanation (I am a physicist, btw). It is possible for a system to exhibit features at certain frequencies, whilst only showing noise at others. Think standing waves, for example.
When does it ever happen to people? When does your brain, even just regions ever stop functioning, entirely? You do not remember deep sleep because you are not forming memories, not because your brain has stopped functioning. What else could you be talking about?
Hmm, I get a feeling that none of these are your true objections and that, for some reason, you want to equate suspension to death. I should have stayed disengaged from this conversation. I’ll try to do so now. Hope you get your doubts resolved to your satisfaction eventually.
I don’t want to, I just think that the alternatives lead to absurd outcomes that can’t possibly be correct (see my analysis of the teleporter scenario).
I really have a hard time imagining a universe where there exists a thing that is preserved when 10^-9 seconds pass between computational steps but not when 10^3 pass between steps (while I move the harddrive to another box).
Prediction: TheOtherDave will say 50%, Beach!Dave and Bowling!Dave would both consider both to be the “original”. Assuming sufficiently accurate scanning & simulating.
Here’s what TheOtherDave actually said.
Yes, looks like that prediction is falsified. At least the first sentence. :)
I’ll give a 50% chance that I’ll experience that. (One copy of me continues in the “real” world, another copy of me appears in a simulation and goes bowling.)
(If you ask this question as “the AI is going to run N copies of the bowling simulation”, then I’m not sure how to answer—I’m not sure how to weight N copies of the exact same experience. My intuition is that I should still give a 50% chance, unless the simulations are going to differ in some respect, then I’d give a N/(N+1) chance.)
I need to think about your answer, as right now it doesn’t make any sense to me. I suspect that whatever intuition underlies it is the source of our disagreement/confusion.
@linkhyrule5 had an answer better than the one I had in mind. The probability of us going bowling together is approximately equal to the probability that you are already in said simulation, if computational continuity is what matters.
If there were a 6th Day like service I could sign up for where if anything were to happen to me, a clone/simulation of with my memories would be created, I’d sign up for it in a heartbeat. Because if something were to happen to me I wouldn’t want to deprive my wife of her husband, or my daughters of their father. But that is purely altruistic: I would have P(~0) expectation that I would actually experience that resurrection. Rather, some doppelganger twin that in every outward way behaves like me will take up my life where I left off. And that’s fine, but let’s be clear about the difference.
If you are not the simulation the AI was referring to, then you and it will not go bowling together, period. Because when said bowling occurs, you’ll be dead. Or maybe you’ll be alive and well and off doing other things while the simulation is going on. But under no circumstances should you expect to wake up as the simulation, as we are assuming them to be causally separate.
At least from my way of thinking. I’m not sure I understand yet where you are coming from well enough to predict what you’d expect to experience.
You could understand my 50% answer to be expressing my uncertainty as to whether I’m in the simulation or not. It’s the same thing.
I don’t understand what “computational continuity” means. Can you explain it using a program that computes the digits of pi as an example?
I think you’re making a distinction that exists only in the map, not in the territory. Can you point to something in the territory that this matters for?
Presumably you create a sim-me which includes the experience of having this conversation with you (the AI).
Let me interpret the term “expect” concretely as “I better go practice bowling now, so that sim-me can do well against you later” (assuming I hate losing). If I don’t particularly enjoy bowling and rather do something else, how much effort is warranted vs doing something I like?
The answer is not unambiguous and depends on how much I (meat-me) care about future sim-me having fun and not embarrassing sim-self. If sim-me continues on after meat-me passes away, I care very much about sim-me’s well being. On the other hand, if the sim-me program is halted after the bowling game, then I (meat-me) don’t care much about that sim-loser. After all, meat-me (who will not go bowling) will continue to exist, at least for a while. You might feel differently about sim-you, of course. There is a whole range of possible scenarios here. Feel free to specify one in more detail.
TL;DR: If the simulation will be the only copy of “me” in existence, I act as if I expect to experience bowling.
Hmm, what about across dreamless sleep? Or fainting? Or falling and hitting your head and losing consciousness for an instant? Would these count as killing one person and creating another? And so be morally net-negative?
If computational continuity is what matters, then no. Just because you have no memory doesn’t mean you didn’t experience it. There is in fact a continuous experience throughout all of the examples you gave, just no new memories being formed. But from the last point you remember (going to sleep, fainting, hitting your head) to when you wake up, you did exist and were running a computational process. From our understanding of neurology you can be certain that there was no interruption of subjective experience of identity, even if you can’t remember what actually happened.
Whether this is also true of general anesthesia depends very much on the biochemistry going on. I admit ignorance here.
OK, I guess I should give up, too. I am utterly unable to relate to whatever it is you mean by “because you have no memory doesn’t mean you didn’t experience it” or “subjective experience of identity, even if you can’t remember what actually happened”.
I would say that yes, at T1 there’s one of me, and at T2 there’s 100 of me.
I don’t see what makes “there’s 101 of me, one of which terminated at T1” more straightforward than that.
It’s wrapped up in the question over what happened to that original copy that (maybe?) terminated at T1. Did that original version of you terminate completely and forever? Then I wouldn’t count it among the 100 copies that were created later.
Sure, obviously if it terminated then it isn’t around afterwards.
Equally obviously, if it’s around afterwards, it didn’t terminate.
You said your metric for determining which description is accurate was (among other things) simplicity, and you claimed that the “101 − 1” answer is more straightforward (simpler?) than the “100″ answer.
You can’t now turn around and say that the reason it’s simpler is because the “101-1” answer is accurate.
Either it’s accurate because it’s simpler, or it’s simpler because it’s accurate, but to assert both at once is illegitimate.
I’ll address this in my sequence, which hopefully I will have time to write. The short answer is that what matters isn’t which explanation of this situation is simpler, requires fewer words, a smaller number, or whatever. What matters is: which general rule is simpler?
Pattern or causal continuity leads to all sorts of weird edge cases, some of which I’ve tried to explain in my examples here, and in other cases fails (mysterious answer) to provide a definitive prediction of subjective experience. There may be other solutions, but computational continuity at the very least provides a simpler model, even if it results in the more “complex” 101-1 answer.
It’s sorta like wave collapse vs many-worlds. Wave collapse is simpler (single world), right? No. Many worlds is the simpler theory because it requires fewer rules, even though it results in a mind-bogglingly more complex and varied multiverse. In this case I think computational continuity in the way I formulated it reduces consciousness down to simple general explanation that dissolves the question with no residual problems.
Kinda like how freewill is what a decision algorithm feels like from the inside, consciousness / subjective experience is what any computational process feels like from the inside. And therefore, when the computational process terminates, so too does the subjective experience.
Non-running algorithms have no experiences, so the latter is not a possible outcome. I think this is perhaps an unspoken axiom here.
No disagreement here—that’s what I meant by oblivion.
OK, cool, but now I’m confused. If we’re meaning the same thing, I don’t understand how it can be a question—“not running” isn’t a thing an algorithm can experience; it’s a logical impossibility.
Clearly, your subjective experience today is HONK-screech-bam-oblivion, since all the subjective experiences that come after that don’t happen today in this example… they happen 50 years later.
It is not in the least bit clear to me that this means those subjective experiences aren’t your subjective experiences. You aren’t some epiphenomenal entity that dissipates in the course of those 50 years and therefore isn’t around to experience those experiences when they happen… whatever is having those subjective experiences, whenever it is having them, that’s you.
Sounds like a fine plan, albeit a difficult one. Want to take a shot at it?
EDIT: Ah, you did so elsethread. Cool. Replied there.