Hacker News Reader

Comment by TrainedMonkey | Hacker News Reader

Every time you go to sleep your consciousness shuts down. What wakes up is something slightly different due to biological processes. You are not tired, your mood is different, part of cells in your brain got replaced, memories become more stable and maybe linked to other similar memories, etc... This is way more pronounced for people waking up from a coma. Philosophers thought of this for a while, see https://philosophy.stackexchange.com/questions/66018/the-bra...

So, each morning is a copy of you wakes up? Or do you draw a line in the sand and say that is different.

View on HN

> Every time you go to sleep your consciousness shuts down. So, each morning is a copy of you wakes up

I think it doesn't "shut down", maybe fades away to a different, low-power mode. When you go to sleep and then wake up, here is still a continuity, because of the same underlying structures and processes from which the consciousness emerges. So it is like a 4D thing.

That continuity never really breaks (until death? which is like singularity) and I think this is what makes you "you". You can't copy/teleport it (kinda by definition), but you can extend/transform it.

Perhaps a "Ship of Theseus"-style approach would work — gradually replacing neurons one by one, or brain sections piece by piece. Alternatively, the brain could be extended with peripheral "devices" connected to it with bandwidth comparable to the interconnections within the brain itself, up to the point until the biological part becomes irrelevant. This is similar to how a portion of the living brain can be removed while still preserving consciousness — as neurosurgeons often demonstrate.

> Perhaps a "Ship of Theseus"-style approach would work — gradually replacing neurons one by one, or brain sections piece by piece.

You just blew my mind (no pun intended).

In this thread, we're basically alluding to the Hard Problem of Consciousness[0]. Consciousness is hard to define, and even harder to prove, and so conversations that talk about it become difficult.

There's certainly a certain subset of the brain you can take out and still have no effect on consciousness. Assuming you could replace small parts at a time, what would happen? At what point does the "mind", "self", or "consciousness" get lost?

[0] https://en.wikipedia.org/wiki/Hard_problem_of_consciousness

> There's certainly a certain subset of the brain you can take out and still have no effect on consciousness

If we don't even know what consciousness is, how do you measure the effect on consciousness of removing arbitrary amounts of brain matter?

Doesn’t seem that hard to measure. It isn’t like the subject will become a mysterious P zombie, they will probably just become increasingly incoherent and then non verbal and then die.

Damage a brain enough and brain death occurs. By the same token, after a certain number of neurons are lost, there is no self. Whatever consciousness is, it can’t be sustained with a sufficiently compromised brain.

This is observable in late stage dementia patients, it gradually becomes difficult to think and in the final dire stages their sense of self degrades too.

> This is observable in late stage dementia patients, it gradually becomes difficult to think and in the final dire stages their sense of self degrades too.

At best, you can be an external observer. You can't truly understand what the person is experiencing.

A common thread I see among people with brain damage is that they don't recognize it. For example, with patients with hemispatial neglect [0], if you ask them to draw a clock, and then only draw the right-hand side of one with the numbers 12 through 6 and they don't draw 7-11, then try to ask them where the 9 is, they might insist it's there. If you ask them to point at the 9, they'll either point to the wrong number or not point at all. And if you tell them they're not pointing or are pointing to the wrong number, they'll insist they're right.

But even with all that knowledge, it's still hard to imagine what that person is actually experiencing.

[0] https://en.wikipedia.org/wiki/Hemispatial_neglect

That is the common way of current understanding. But there is problem.

So called 'terminal lucidity' exists sometimes.

One theory is that is enabled by the immune systems shutdown, so inflammation ceases to interfere for a short time.

The problem with that is it still fails to explain the few cases, where shortly post-mortem it is found that there is not much healthy/able to function somewhat normally brain left to speak of.

Rather 'spooky'.

The thing is, it's not about removing brain matter, it's about replacing brain matter (with a manufactured copy, for example), and applying the same process in a secondary location (physical or digital, I guess). Will 'you' at some point stop being 'you'? Will the secondary copy at some point become you?

> your consciousness shuts down.

Consciousness not a single isolated process and describing a lull in activity as "shutting down" is incorrect.

> So, each morning is a copy of you wakes up?

Obviously not. Many of these processes also happen in reverse when you are awake. So when I go to sleep I'm tired and in a different mood. When did that "copy" get made exactly?

> Or do you draw a line in the sand and say that is different.

You can easily ask "how much of my conscious state is controlled by my body's chemical state?" If that's a non trivial amount then simply copying and hoisting my memories into a different body is clearly going to have a non trivial impact. To the extent that it's easy to say "that would not be /me/."

> Consciousness not a single isolated process and describing a lull in activity as "shutting down" is incorrect.

but there is indeed a discontinuity in your consciousness when you sleep. This discontinuity cannot be distinguished from the discontinuity arising from a copy of your brain on first "restart".

So how do you reconcile this discontinuity?

>but there is indeed a discontinuity in your consciousness when you sleep. This discontinuity cannot be distinguished from the discontinuity arising from a copy of your brain on first "restart".

That amounts to saying "every consciousness discontinuity is of the same level".

I.e. that waking from sleep has the same distance from "you" before sleeping, as would a replica of you created by scanning your brain and uploading it to it.

Let's put it this way: if we can co-exist, it's not me.

> but there is indeed a discontinuity in your consciousness when you sleep.

Yet you dream and can remember your dreams. Your dreams can be influenced by things occurring around you in your sleep. Some people do this on purpose and find that they are more reliably able to do it with practice. So how are you defining "discontinuity?" I don't see one at all.

> arising from a copy of your brain on first "restart".

You're rather certain of the semantics of something that does not exist and that you haven't described any sort of process for. What if you can initiate a copy while I'm awake? Am I required to be asleep or sedated into an artificial coma in order to make this copy? Why is that so? Are my short term memories excluded? Is my stomach equally full? Do I have the same wounds? Do I need to pee?

> So how do you reconcile this discontinuity?

How do you explain sleep walking? Or narcolepsy. In your model of discontinuities?

>Yet you dream and can remember your dreams. Your dreams can be influenced by things occurring around you in your sleep. Some people do this on purpose and find that they are more reliably able to do it with practice. So how are you defining "discontinuity?" I don't see one at all.

Dreams aren't continuous throughout sleep, so this seems pretty irrelevant to whether there's a discontinuity in the first place, you're just saying there may be multiple during a given night's sleep.

> Dreams aren't continuous throughout sleep

What do you base this assertion on?

> so this seems pretty irrelevant to whether there's a discontinuity in the first place

Have you proven there is one? Or are you just saying "sleep is really weird so that must be a 'conscious discontinuity.'" I'm pointing out the obvious flaws in this logic.

> you're just saying there may be multiple during a given night's sleep.

No, you're saying that, and I'm asking you to define it.

>What do you base this assertion on?

It's a pretty well understood part of sleep research. Sleep has different phases, only REM, the lightest part of sleep, is associated with dreaming. Also, it matches up with my experience of dreams and basically all the reports I've seen from others on dreams: there's not a continuous experience from falling asleep to dreams, there's being in the process of falling asleep, then there's being in a dream, and the experience is discontinuous, the transition between wakefullness and dreaming is not experienced, nor is the transition between different dreams.

(I've also personally experienced a very sharp discontinuity around sleep, once or twice, where I have literally 'blinked' while looking out my bedroom window and seen it go from dark to bright light, experiencing nothing of the hours in between. Though I suppose in many ways this is a much more continuous conscious experience than I normally have when sleeping, where waking up feels much more disjoint from falling asleep)

Dreaming absolutely does not only occur during REM.

https://pmc.ncbi.nlm.nih.gov/articles/PMC7377375/

https://www.jneurosci.org/content/38/43/9175

Ah, TIL, and fair point then, though I think there's still not much support for continuous consciousness throughout sleep.

I mean personally I see continuity as mostly being about the hardware with the process running on it only being secondary at best and disjointed, though not completely irrelevant.

So funnily, I'm "perpendicular" to the issue of if continuous consciousness is needed -- i think the only relevant question is following the original structure and processes every "tick" of the universe to see what happened.

E.g. to me if the structure is wholly or nearly-wholly preserved, "teleporting" by the mechanism of particles having an uncertain position is non destructive and retains continuity, stasis is the same, becoming part of a hivemind is the same -- though leaving one or being quickly ship-of-theseus'd would likely impair continuity, but less in a manner of "death" and merely making the question meaningless, because two new entities are born descended from the same source, akin to how who we were as children never dies, only undergoes a process of growth/change/refinement until it's something not only different but significantly more than the "original".

I think a very interesting line of thought is actually if severe brain damage essentially kills continuity / "the original", in context of how we can absolutely disable or impair parts of the brain with magnetism or drugs but then have the hardware return to original function along with apparent return of the original process. It actually feels much closer to the self-evidently wrong assertion that any mere identical copy would equally be the original, suggesting to me that damaging the original too far past a point may actively kill that thread of consciousness in a way I wouldn't consider to be the case from sleep / stasis / joining-but-not-leaving a hivemind / etc.

> This discontinuity cannot be distinguished

It can be, though. I don’t see how the discontinuity bit is the defining factor.

Regardless equating humans with digital computers and using analogies like that doesn’t make a lot sense to begin with.

I think the difference here is that we share our bodies with the copy and share our memories. The difference between us at night and us in the morning is small. A copy in the cloud doesn't pass memories on to us when it sleeps and it's experience is very different. It's digital and could technically have a billion instances with no experience sync between them. Which one is 'you' then?

The best way to go about this IMO, is nano bots first augmenting the brain, connecting to the cloud, then replacing the brain slowly and introducing syncing consciousness when online. Replace a little every night and then it would still be you IMO since you're replicating most closely what the brain already does.

There’s a difference when there could conceivably be two of you.

There will never be two of me in my bed when I wake up, so I’m the same.

If my mind could be copied and embedded in a robot or whatever, then there are two of me, each now diverging. That other one is not me, as far as I, the other, am concerned.

You'll need to be more clear in your last sentence.

If you are cloned while you are sleeping, and both the original and the copy are moved into different rooms before you wake up, both of you think themself is you, and both of you think the other is not you. So, who is correct?

You're assuming that there exists such a cloning process, and in doing so, silently ascribing properties to that process that completely change the outcome.

You're in effect saying "Assume there is a cloning process that is instant and creates a complete copy of you, both mentally and physically," and sure, in that theoretical world, you can talk about there being no way to know which was you, the clone or the original.

But IRL, we have no way of knowing what the properties of such a cloning process would be - maybe it's impossible to create a cloning process such that mentally, it is a complete copy. Maybe you can always tell whether you're the original or the clone. Similarly, maybe we simply cannot create a digital copy of a mind that cannot immediately tell it is the copy. We have literally no way of knowing.

So instead of living in the philosphical realm, we can have a lot more productive conversation in the practical one. Practically, it strongly appears to me that I have continuity every time I wake up, and practically, if you copied my mind and ran it on a computer, that mind would not be the "me" that appears to experience continuity. Practically, it would bring me no comfort to know that this other mind that used to be a copy of my own continues onwards while I die, no more so than it would bring me comfort to know a twin of mine continues onwards.

For me, the relationship is non-transitive.

If you tell me this will happen tonight, I would right now call both "me". Both of them after the event would agree that I am them. Both of them would agree they are different people to each other, even if they say so in perfect stereo unison.

Cue the "We are Bob" series, aka Bobiverse, by Dennis E. Taylor!

Was thinking more like John Crichton on Farscape, as Bob's copies diverge substantially and immediately, whereas John and his duplicate spend hours repeatedly drawing at rock paper scissors.

This entire thread just has me wanting to watch The Prestige one more time.

>If you are cloned while you are sleeping, and both the original and the copy are moved into different rooms before you wake up, both of you think themself is you, and both of you think the other is not you. So, who is correct?

Obviously the original one is.

Whether the clone also things that it's me doesn't make it any more so, than if a paranoid person thought it was Napoleon would make them Napoleon.

Yes, but we're suggesting a clean mental break in some senses from your cognitions POV. Both of you would be the real you, except to each other. The Hermetic tradition has some texts that delve into how God cannot be contained within the universe because he exists outside of it. In some senses, it would be downright insane to ponder the ramifications of this.

After the first 'Clone Settlement' and the pains of splitting someone's belongings between the same person; it would have to be made illegal. No other way around it.

Which one is the original? Say we aren't making a computer copy but duplicate of your body. And we might have messed up the labels.

How do you tell which one is the original? Atoms don't have labels. But atom-level copy isn't possible. Say we made the scan at large enough scale that has fidelity but it isn't identical.

>Which one is the original? Say we aren't making a computer copy but duplicate of your body. And we might have messed up the labels.

Reality doesn't care about the labels. One of the two pre-existed, which is the original.

>How do you tell which one is the original? Atoms don't have labels.

That's a problem regarding our knowledge. There's still an original, regardless if we know which or not.

There's an original even regardless whether the original has a way to prove to themselves they are so (as opposed to simply believing they are, which the clone would believe too).

from my own perspective it would be impossible to tell given the parameters of this thought experiment, but any outside observer who witnessed the cloning process and the staging of the second bedroom would know.

Moreover, if randomly one of the two dies and the other survives then either the original or the clone experiences death. So from the point of view of the original before the experiment, he has a 50% chance of dying. If we change the experiment to say that he has a 100% chance of dying but the clone has a 100% chance of surviving, the original would not agree to the experiment unless he was suicidal; obviously he has no reason to care that a different person will live on assuming his identity after he dies.

It makes no difference. If it’s possible for the two versions to meet (they exist simultaneously) then what you’re describing is simple deceit.

A world where two of me exist is fundamentally and irreconcilably different than a world where I simply go to bed and wake up each day. It makes no difference if I personally am aware.

it's almost as if your thought patterns and language are a product of your experience of being the only copy of you...

if having copies of you walking around was a normal thing, our language and expectations would reflect that.

right now if another copy of you comes by and claims your wife and your house, you'd be miffed. but the whole concept of one of you owning the house is just an artefact of how individuality and possessions happen to be right now. it doesn't have to be forever. there was a time you could steal a picture and have the only copy - now you can save a copy on your phone and you don't deprive another of theirs. some day people and stuff could behave the same.

Hopefully IP and copyright laws will evolve in parallel to prevent or highly penalize stuff like that.

I imagine the poor will want their mind state covered by copyright/intellectual property laws, and the rich will ignore that if it can help them amass more wealth - just like LLM and diffusion model training was bootstrapped off the back of creatives without payment. i.e. the law doesn't apply if you're rich enough.

> just like LLM and diffusion model training was bootstrapped off the back of creatives without payment. i.e. the law doesn't apply if you're rich enough.

Or search engines.

I don't know how the legal cases will go, all I can say is that my suspicion is that, like search, such AI are simply too useful to be outlawed… but my armchair opinions are almost as far as possible from a guarantee, and even search had to change in response to copyright infringement laws.

>if having copies of you walking around was a normal thing, our language and expectations would reflect that.

Yes, and if the law of excluded middle didn't hold, then all kind of magical things would happen too.

But it does, and they don't.

None of that is related to my point.

A world where I get cloned and an unaware is distinctly divergent from a world where there is only one of me each time I wake up, regardless of how society handles it.

yes, a world where there are two of you is different from a world where there's one of you. I feel like this is something Aristotle must have covered, but it is important to point it out!

Based on this entire thread it did seem necessary to state.

Clearly, otherwise you would not have stated it.

> So, who is correct?

The one that wasn’t the clone.

Irrelevant, cloning doesn't and probably never will, happen.

I draw the line in the sand and say that is different because sleep has been working "forever", while the quote in the parent comment describes a process that, not only does not exist and therefore work now, if it ever gets to work, we won't know if it's the same as the daily process that happens when we sleep.

Otherwise I actually agree with your point ("every night another little death", as the Sonic Youth song says): I'm different every morning. Usually, the differences are subtle. Over the time span of a lifetime though, I've been through enough changes that I often say ghosts of former selves share this body, and what "I" am now will probably be another ghost by the time the biological substrate dies and takes whatever "I" will be then, along with all the ghosts, with it (in my case we'll go up in smoke since I'll be cremated).

The river I stand in is not the river I step in. (1)

(1) https://brandondonnelly.com/this-river-i-step-in-is-not-the-...

I do wonder sometimes if each period of being awake is the entire subjective experience of a sequence of independent conscious entities. Like I woke up today with the memory of yesterday, but possibly that version of me that had the subjective experience of being was an ephemeral phenomenon that ceased to exist when I fell asleep, just as I will tonight.

I raised a similar objection to the GP comment many years ago, and someone made the same response to me. I wonder if I'm alone in that instead of making me feel more optimistic about mind transfer, it just made me slightly afraid to go to sleep.

Same actually, if I follow the logic the conclusion is pretty grim. I've come up with silver lining to embrace the change. All life must change to survive and thrive. To wake up the same every time, to never change no matter what, to never learn - that is true horror to me.

You're absolutely not. There's two big differences as I see it - firstly, it's not clearly impossible for the instance to survive sleep. The brain exists continuously throughout. There is constant electrical activity. We don't know for sure that our awareness is "transmitted through" to when we wake up, but we can imagine ways it's possible.

In contrast, of course, transporters and mind uploads obviously kill us.

Secondly, we can't do anything about sleep. No matter how worrying it is. There's just no choice other than to accept it.

The first argument is nicer than the second, but, frankly, if I could avoid sleep I would and I really hope I never need general anesthesia.

Each morning I would not be a copy.

This is provable - only one instance can wake up, whereas with the other method, it’s conceivable that two or more copies (instances?) could wake up.

Thanks for feeding my existential dread. For that matter, the you of this moment is constantly being annihilated and replaced by a slightly different clone, in an illusory life that resembles stop-frame animation. And don't ask me about the me from yesterday - the one who didn't bother to do any of the things I am stuck with doing today - or the me from tomorrow who will likely squander everything I've worked so hard for. Neither of them is to be trusted.

On the other hand, I do seem to have a vast store of dream memories; wnen I think about them it seems like living some kind of second life, usually while fast asleep. Sleeping and waking selves are in agreement that being asleep is generally more pleasant than being awake, though being awake seems to have better continuity.

I don't think it quite the arbitrary distinction you're making it out to be. If you were cloned every time you went to sleep, and those clones awoke with the same memories you had before going to bed. Would you consider all of those clones' consciousnesses to be part of your own? Surely not, because you're not experiencing them, you're only experiencing your own. This means that the "you", in this case, is the self reflective portion of the collection of interpretations of sensory stimulus and memories, whereas consciousness is that collection as a whole. That's why these clones can't possibly be "you", because if there's no sensory stimulus being experienced or interpreted by your brain, then there is no consciousness and there certainly is no "you". What this ends up meaning is that "you" are necessarily bound to the mechanical component that interprets stimuli. That component can be adjusted over time, but the process must be interpreted in such a way that it seems gradual enough that your brain can still recognize those intermediary interpretations as "you".

So, when you state that robo brain is still "you", I have to disagree. While robo-brain can still believe it's "you", it can't possibly be "you", as "you" could potentially still exist in the original biomechanical device, which isn't interpreting the new body's stimuli, and thus not building a sensory profile based on the stimulus. Even if there was a transfer that resulted in the original device's (i.e. your body) death, the new body still couldn't be considered "you" as there isn't anything physically different about the two scenarios for the body.

Of course, if you consider every clone to be an extension of your consciousness, then you could argue all this. But at that point I think we'd have vastly different interpretations on what consciousness and "you" are. You could also argue that there is something physically different about the two scenarios presented in the robo-brain example, but the only way I can imagine a sincere argument for that is if you told me it was due to the transference of some soul-like entity. But, of course, if you were to tell me that, I would simply disagree.

Reminds me of Filmcow: https://www.youtube.com/embed/ur9jLK_6EcY?start=70&end=100

> Every time you go to sleep your consciousness shuts down.

What consciousness? Yes, I am aware that "sleep" means that a person becomes unconscious (trivially demonstrated), but that is not the same thing as "consciousness shutting down". Supposing there is such a thing as consciousness, it doesn't seem to be anything that is suspended by the sleep process. And that's also a big "if", because consciousness is just woo-woo nonsense talked about but never defined by people who haven't quite gotten over the idea that humans don't have some immaterial soul.

Consciousness has, I'm told, 40 identifiable meanings. Makes it hard to get to grips with it.

For example, I sometimes have direct awareness of the nature of my consciousness changing with tiredness. Sometimes that has been what I can only describe as if my homunculus has switched to autopilot and is now just watching a cinema of my senses.

But that's probably a terrible description if you've never experienced that yourself.

If I'm not conscious, it's rather difficult to explain what I'm experiencing right now.

You're not really "experiencing" anything in the sense you mean. For a few thousand years, your species has communicated to others of its kind with a peculiar evolved language that causes a sort of narrative confusion.

I believe, if we could be copied to the cloud, that “me” is much more than neurons, and affected and defined quite a lot by the particular physical body we’re housed in. You can say losing an arm doesn’t change who you are, but it definitely does at some levels. Changes in health affect personality, mood, outlook, hope, enjoyment, etc, etc - that would all change when copied.

A gap in awareness is definitely not the same as creating a new copy of your consciousness and annihilating the old one.

Although, I'll grant that the subjective experience may be fairly similar. If you were to theoretically be copied, your copy would probably feel quite continuous with the original.

>Every time you go to sleep your consciousness shuts down.

Sure, but I'm ok with sidestepping that totally pedantic concern and still considering me the same person, like I do every single day.

Mere point-in-time-copying to another substrate, though, not so much.

Every nanosecond the set of atoms constituting you changes, each of those are a different you.

There's a difference between "I have changed" and "I have been copied." This seems obvious enough to me that I don't question it, but is it not obvious to everyone?

Copies are impossible because every moment is change with bounded but non-zero uncertainty.

The fact that "you are different" does not mean a "different you" now exists, though. Change of this specific kind is already intrinsically integrated into your existence and identity.

Every moment of consciousness is different. It is a different you in that sense. What is the strand that keeps it the same ‘you’?

Physical continuity, in the sense that there is no break so significant that at any point one could say, "Before this, smeej0, and after this, smeej1." Unlike the ship of the famous thought experiment, there is and can be no reconstructing of the replaced parts such that smeej0 could ever be reconstituted, so the analogy breaks down in any attempt to apply it to humans.

Change is ssential to human nature. A human exists--and the same human exists--fron fertilization to death. Fertilization is a change before which we can say "not smeej" and after which, "smeej." And death is also a change before which we can say "smeej" and after which, "not smeej." There will be vastly, vastly more cells at death than there were at fertilization (assuming smeej made it out of the womb and to an observable state), and the original fertilized cell will no longer exist, but there is no other point along the existence of smeej in which a change happens of such significance as to alter whether or not the same thing is present.

Honestly I'd like to have my mind copied, but may only be 'simulated' once the current me is dead. From my perspective I'll be dead and there's no changing that, but from the clone's perspective, life just goes on.

true quantum scale copies may be impossible but that resolution may not be necessary to fully copy a person.

Not if “you” is created by emergence.

Because you are more than atoms. You are a pattern. A pattern plays out over time. Like music. Music has a physical component, of course, but music is enjoyable because it's a pattern over time.

Which is the "real" Moonlight Sonata: Beethoven's original sheet music, the first performance, the photons carrying some radio broadcast, some physical vinyl or CD master copy, a lossless encoded digital copy?

The thing about a copy, it reproduces that pattern. The question is, how good does a copy need to be to retain the important features? For now, I suspect that any attempt to actually scan a brain would be a cargo-cult copy of a mind… but I don't know that for sure, I'm just going with what I hear from the occasional neuroscientist who bothers to comment on misconceptions in this field.

switching a sesame seed on my burger bun is not the same thing as scanning it and uploading the file to the cloud.

No it doesn’t.

You’re not unconscious while you’re sleeping.

Consciousness is probably an illusion anyway.

An illusion to whom?

That language implies a self does not mean a self exists in reality.

Reply on news.ycombinator.com