5 ms·

There’s no reason to believe that humans living as “digital beings” will ever be possible. We don’t even know what consciousness is yet, never mind how to repli

There’s no reason to believe that humans living as “digital beings” will ever be possible. We don’t even know what consciousness is yet, never mind how to replicate it in a non-human substrate. Most living humans will still argue that consciousness roughly equates to a soul. Leave that nonsense aside, and you’re still left with the mind-body problem: are those two things the same or different? If mind and body are the same — that is, if the experience of being a conscious human is emergent and/or dependent upon a human body — digital uploads are just copies of ephemeral state. Imagine a hell full of James T Kirk replicas, each regretting the time he stepped into a transporter and died, allowing a copy of him to be made elsewhere. If mind and body are different, you’re picking up the Cartesian problem — how do they influence each other at all? What is the mind made of that it can simultaneously affect the body, yet somehow exist separately from it? Neal Stephenson published “Fall” last year which treats this as a purely technical problem. Once our scanning resolution is good enough, he supposes, we’ll be able to create perfect digital replicas of brains. That’s tripe, of course. Human consciousness is dependent on much more than our brains. In fact, we have special words for how consciousness presents when disconnected from bodily sensation, words like “sleep” and “coma.” Sensory deprivation chambers isolate us from some of the sensations outside our bodies, but leave all internal sensations intact. These external and internal sensations make us who we are. For instance, we cannot emulate consciousness without first emulating the billions of gut bacteria, across thousands of distinct species, that affect the physical rhythms of our bodies. Even quadriplegic’s moods and consciousnesses are influenced by their gut bacteria. How much of the universe do you need to simulate in order to accurately simulate one human being? All of it, it turns out. And even then, there’s no reason to believe that human will be or feel the same as the meat substrate from which it was copied. The whole affair is nonsense. edit: typo

cyberpunk 6y ago
I’m not sure it’s so far fetched. Scanning sure, that’s a separate consciousness or a copy, but what if you replaced one neuron at a time with a digital/electrical component? Wouldn’t there be a point, around halfway, that you would be more robot than human? And so you can become totally digital if you accept that losing one neuron wouldn’t be enough to end your current stream of consciousness?
- staticman2 6y ago
  The electronic neuron would have to behave just like a legit neuron, and would have to be installable without damaging the legit neurons. Even if this is philosophically possible, that doesn't suggest it is inventible. How do you model and replace a neuron without tearing up and killing the host organism?
- sonofhans 6y ago
  Personal identity philosophy is full of examples like this: swapping brains, swapping half brains, swapping cells one at a time. It’s an interesting thought exercise. I don’t have a complete answer, but a few thoughts: First, there’s no reason to believe we’ll ever be able to replicate a neuron with non-meat technology. Nanomachines may be able to do this. Are they going to age and die as well, like real neurons? Second, replacing one neuron at a time looks hard as well. Best case is perhaps a self-replicating nanomachine that chugs through the brain, effectively eating neurons and building new ones. This process is going to change the brain, since it has to be carried out in constant time. How will it change the brain? Who knows? Third, it’s not just neurons. You’d need to replicate every cell in, and on, a human body. Including, again, the different species of fungus that live exclusively in our toenails. Fourth, how will this new 100%-technical human interact with its environment? Current thinking is that some auto-immune diseases are triggered by a insufficient challenge to young immune systems, effectively by an environment that’s too clean. Should we emulate this as well? So pretty quickly I think it’s turns out that we’re not emulating humans, we’re creating new things that are somewhat human and somewhat digital. How will this entity act over the long term? Ian McDonald wrote, “there are infinite ways of being human.” Perhaps we need to expand our definitions.
  - ben_w 6y ago
    > Third, it’s not just neurons. You’d need to replicate every cell in, and on, a human body. Including, again, the different species of fungus that live exclusively in our toenails. Do you? Don’t get me wrong, the chemical environment of the brain, and the wider body, can alter states of consciousness — I recall stories about people whose gut bacteria produces alcohol and permanently fail any blood alcohol tests as a consequence — but if you know what those chemicals are, do you really need to replicate the fungi making them?
lupire 6y ago
Consciousness isn't important for "digital" expansion. AI is. If people build self-sustaining/replicating robots and send them to stars, off they go.
arethuza 6y ago
Wasn't one of the points in Fall; or, Dodge in Hell that an isolated mind was pretty much impossible and that he had to build an environment at the same time that has mind was awakening?
- sonofhans 6y ago
  Yes, I actually love that awakening story. It’s been done in fiction before (e.g., “Cybernetic Samurai,” by Victor Milan) but Stephenson rendered it vividly. Even if we accept this on its face as possible, we still have no reason to believe that the digital Dodge will act the same way as the formerly-meat Dodge. Developmental trauma is a real thing. How did that experience of awakening change Dodge? If it changed him at all he’s no longer the same Dodge; if it didn’t change him, then he’s just an immutable copy. And then if we figure out some convoluted way to claim that Didi-Dodge will act/present the same as formerly-meat-Dodge, they’re not in any way the same person. Meat-Dodge is still dead, he just has a very lifelike digital replica.
dragonwriter 6y ago
> Neal Stephenson published “Fall” last year which treats this as a purely technical problem. Once our scanning resolution is good enough, he supposes, we’ll be able to create perfect digital replicas of brains That's...not at all accurate. > How much of the universe do you need to simulate in order to accurately simulate one human being? All of it, it turns out. And even then, there’s no reason to believe that human will be or feel the same as the meat substrate from which it was copied. That's actually much closer to the premise of Fall than your summary was.
- sonofhans 6y ago
  > That's...not at all accurate. Tell me how. My recollection is that Dodge was scanned from a frozen brain. Later in the book, folks were full-body-scanned. And conceding that we need to simulate a full digital universe in order to simulate one person gives away the argument entirely — there’s still no coherent way to say that it’s the same person, a digital afterlife of the same soul. Stephenson blithely writes as if this is true.
  - ben_w 6y ago
    > Stephenson blithely writes as if this is true. Forgive me, but this sounds like a criticism — is it? Because last I checked that novel was fiction, and that is what good authors do. It’s not like I’d criticise Diane Duane for writing Trek books as if Vulcans were real; if anything its kinda weird when shows like Trek break the fourth wall, which they avoided even in the two episodes which used the “the show is a fiction invented by one of the characters” trope.
  - dragonwriter 6y ago
    You claim: “Once our scanning resolution is good enough, he supposes, we’ll be able to create perfect digital replicas of brains” None of the scanned replicas in Fall is perfect. They are functional in that they have consciousness, and connected to the original by some fuzzy, vague, and very much incomplete transference of memory and personality, but that's notably imperfect even for the focal information where people most knowledgeable about the process are deliberately trying to carry key information and plans across the transition. That is implicitly (though it is hard to tell how much since neither is a viewpoint character, and the original was progressively becoming insane before the transition) somewhat less the case for Elmo Shepherd/El—the former of which spent enormous time and astronomical resources to achieve that—than anyone else, sure, but there is no indication of anything even vaguely approximating a “perfect digital replica”. > there’s still no coherent way to say that it’s the same person, a digital afterlife of the same soul. Stephenson blithely writes as if this is true. No, he writes as though the society of the not-yet-uploaded in the book comes to accept it as tru-ish and those of the imperfect-replicas-who-come-to-understand-their-nature view it as also (but, at least in my reading, somewhat less than the preceding) somewhat tru-ish. But even the characters in the book, on either side, don't seem to accept it as unqualifiedly true in general with the possible, again, exception of Elmo Shepherd/El. Who is a, or both of whom are (depending on your perspective on continuity of identity) unbalanced megalomaniac(s). (Even before things get dominated by the afterlife simulation, the book already has a strong focus on the socially-constructed nature of perceived identity and perceived nature of reality.)
Chris2048 6y ago
I think this argument is based on the false dichotomy of "The mind & body are either different or the same!". This denies things being similar, or multi-dimensional. Apply this same reasoning to hardware/software for example: "What is software code made of that it can simultaneously affect the physical hardware, yet somehow exist separately from it?" the mind, and software, are data, or information - and data can be replicated on multiple mediums. This data / medium distinction is no more or less perplexing. The "OK" hand-gesture can represent something in physical form, while the physical hand is distinct from the symbolic message it represents - if you want to get philosophical about symbology and representation I don't think there is a special case to be made for the mind. > Human consciousness is dependent on much more than our brains How can you support this statement when you first lead with "We don’t even know what consciousness is [yet]"? We have words for "god", "magic" and "spirits" too, and don't know what those are either; isn't it possible none of them exist? > we cannot emulate consciousness without first emulating the billions of gut bacteria Only is consciousness specifically depends on this aspect. It's like saying "The setting sun also affect our mood, so that must be emulated too!". That's not true though - It's be possible to emulate a human mind, without exactly reproducing the same personality and emotional profile of a modern human being. Simulating human environments and/or experience to feed a simulation data is a bigger project than replicating a mind. If you are simulating every detail of human existence the question is what you are trying to achieve; post-human AI being don't need to be anything like modern humans phycological; and I doubt the contributions to mood of gut bacteria are considered fundamental in that - and I've yet to hear a modern human consider the diet in questions of consciousness. For example, We live, and eat, very differently to our ancestors - Does that make us any less human, or conscious? I don't think so, and as such don't agree that: "All of it, it turns out".

cyberpunk 6y ago

I’m not sure it’s so far fetched. Scanning sure, that’s a separate consciousness or a copy, but what if you replaced one neuron at a time with a digital/electrical component? Wouldn’t there be a point, around halfway, that you would be more robot than human? And so you can become totally digital if you accept that losing one neuron wouldn’t be enough to end your current stream of consciousness?

staticman2 6y ago
The electronic neuron would have to behave just like a legit neuron, and would have to be installable without damaging the legit neurons. Even if this is philosophically possible, that doesn't suggest it is inventible. How do you model and replace a neuron without tearing up and killing the host organism?
sonofhans 6y ago
Personal identity philosophy is full of examples like this: swapping brains, swapping half brains, swapping cells one at a time. It’s an interesting thought exercise. I don’t have a complete answer, but a few thoughts: First, there’s no reason to believe we’ll ever be able to replicate a neuron with non-meat technology. Nanomachines may be able to do this. Are they going to age and die as well, like real neurons? Second, replacing one neuron at a time looks hard as well. Best case is perhaps a self-replicating nanomachine that chugs through the brain, effectively eating neurons and building new ones. This process is going to change the brain, since it has to be carried out in constant time. How will it change the brain? Who knows? Third, it’s not just neurons. You’d need to replicate every cell in, and on, a human body. Including, again, the different species of fungus that live exclusively in our toenails. Fourth, how will this new 100%-technical human interact with its environment? Current thinking is that some auto-immune diseases are triggered by a insufficient challenge to young immune systems, effectively by an environment that’s too clean. Should we emulate this as well? So pretty quickly I think it’s turns out that we’re not emulating humans, we’re creating new things that are somewhat human and somewhat digital. How will this entity act over the long term? Ian McDonald wrote, “there are infinite ways of being human.” Perhaps we need to expand our definitions.
- ben_w 6y ago
  > Third, it’s not just neurons. You’d need to replicate every cell in, and on, a human body. Including, again, the different species of fungus that live exclusively in our toenails. Do you? Don’t get me wrong, the chemical environment of the brain, and the wider body, can alter states of consciousness — I recall stories about people whose gut bacteria produces alcohol and permanently fail any blood alcohol tests as a consequence — but if you know what those chemicals are, do you really need to replicate the fungi making them?

lupire 6y ago

Consciousness isn't important for "digital" expansion. AI is. If people build self-sustaining/replicating robots and send them to stars, off they go.

arethuza 6y ago

Wasn't one of the points in Fall; or, Dodge in Hell that an isolated mind was pretty much impossible and that he had to build an environment at the same time that has mind was awakening?

sonofhans 6y ago
Yes, I actually love that awakening story. It’s been done in fiction before (e.g., “Cybernetic Samurai,” by Victor Milan) but Stephenson rendered it vividly. Even if we accept this on its face as possible, we still have no reason to believe that the digital Dodge will act the same way as the formerly-meat Dodge. Developmental trauma is a real thing. How did that experience of awakening change Dodge? If it changed him at all he’s no longer the same Dodge; if it didn’t change him, then he’s just an immutable copy. And then if we figure out some convoluted way to claim that Didi-Dodge will act/present the same as formerly-meat-Dodge, they’re not in any way the same person. Meat-Dodge is still dead, he just has a very lifelike digital replica.

dragonwriter 6y ago

> Neal Stephenson published “Fall” last year which treats this as a purely technical problem. Once our scanning resolution is good enough, he supposes, we’ll be able to create perfect digital replicas of brains That's...not at all accurate. > How much of the universe do you need to simulate in order to accurately simulate one human being? All of it, it turns out. And even then, there’s no reason to believe that human will be or feel the same as the meat substrate from which it was copied. That's actually much closer to the premise of Fall than your summary was.

sonofhans 6y ago
> That's...not at all accurate. Tell me how. My recollection is that Dodge was scanned from a frozen brain. Later in the book, folks were full-body-scanned. And conceding that we need to simulate a full digital universe in order to simulate one person gives away the argument entirely — there’s still no coherent way to say that it’s the same person, a digital afterlife of the same soul. Stephenson blithely writes as if this is true.
- ben_w 6y ago
  > Stephenson blithely writes as if this is true. Forgive me, but this sounds like a criticism — is it? Because last I checked that novel was fiction, and that is what good authors do. It’s not like I’d criticise Diane Duane for writing Trek books as if Vulcans were real; if anything its kinda weird when shows like Trek break the fourth wall, which they avoided even in the two episodes which used the “the show is a fiction invented by one of the characters” trope.
- dragonwriter 6y ago
  You claim: “Once our scanning resolution is good enough, he supposes, we’ll be able to create perfect digital replicas of brains” None of the scanned replicas in Fall is perfect. They are functional in that they have consciousness, and connected to the original by some fuzzy, vague, and very much incomplete transference of memory and personality, but that's notably imperfect even for the focal information where people most knowledgeable about the process are deliberately trying to carry key information and plans across the transition. That is implicitly (though it is hard to tell how much since neither is a viewpoint character, and the original was progressively becoming insane before the transition) somewhat less the case for Elmo Shepherd/El—the former of which spent enormous time and astronomical resources to achieve that—than anyone else, sure, but there is no indication of anything even vaguely approximating a “perfect digital replica”. > there’s still no coherent way to say that it’s the same person, a digital afterlife of the same soul. Stephenson blithely writes as if this is true. No, he writes as though the society of the not-yet-uploaded in the book comes to accept it as tru-ish and those of the imperfect-replicas-who-come-to-understand-their-nature view it as also (but, at least in my reading, somewhat less than the preceding) somewhat tru-ish. But even the characters in the book, on either side, don't seem to accept it as unqualifiedly true in general with the possible, again, exception of Elmo Shepherd/El. Who is a, or both of whom are (depending on your perspective on continuity of identity) unbalanced megalomaniac(s). (Even before things get dominated by the afterlife simulation, the book already has a strong focus on the socially-constructed nature of perceived identity and perceived nature of reality.)

Chris2048 6y ago

I think this argument is based on the false dichotomy of "The mind & body are either different or the same!". This denies things being similar, or multi-dimensional. Apply this same reasoning to hardware/software for example: "What is software code made of that it can simultaneously affect the physical hardware, yet somehow exist separately from it?" the mind, and software, are data, or information - and data can be replicated on multiple mediums. This data / medium distinction is no more or less perplexing. The "OK" hand-gesture can represent something in physical form, while the physical hand is distinct from the symbolic message it represents - if you want to get philosophical about symbology and representation I don't think there is a special case to be made for the mind. > Human consciousness is dependent on much more than our brains How can you support this statement when you first lead with "We don’t even know what consciousness is [yet]"? We have words for "god", "magic" and "spirits" too, and don't know what those are either; isn't it possible none of them exist? > we cannot emulate consciousness without first emulating the billions of gut bacteria Only is consciousness specifically depends on this aspect. It's like saying "The setting sun also affect our mood, so that must be emulated too!". That's not true though - It's be possible to emulate a human mind, without exactly reproducing the same personality and emotional profile of a modern human being. Simulating human environments and/or experience to feed a simulation data is a bigger project than replicating a mind. If you are simulating every detail of human existence the question is what you are trying to achieve; post-human AI being don't need to be anything like modern humans phycological; and I doubt the contributions to mood of gut bacteria are considered fundamental in that - and I've yet to hear a modern human consider the diet in questions of consciousness. For example, We live, and eat, very differently to our ancestors - Does that make us any less human, or conscious? I don't think so, and as such don't agree that: "All of it, it turns out".