Hacker News Reader

Comment by axguscbklp | Hacker News Reader

Discussions of consciousness tend to frustrate me a bit. So many otherwise smart people, for some reason, simply do not understand the hard problem of consciousness. They'll say things like "well, consciousness is just a given pattern of physical stuff" but they do not realize what to me is an obvious objection - to say that consciousness is equivalent to a given pattern of physical stuff does not explain why that pattern of physical stuff is not simply present without any consciousness associated with it. I can imagine a being that is physically identical to a human in every way, but is not conscious - there is no reason to think that intelligent behavior requires consciousness. So what is the difference between such a being and a conscious human, if the two are physically identical?

View on HN

You may be able to imagine a body physically identical to yourself not having a consciousness, but you cannot deduce from that, that this is something that can actually happen in our Universe. It may just as well be true that any exact physical copy of a body containing a consciousness is also conscious. In fact, the latter seems like a simpler assumption than the former.

My own take is that we, as biased humans, put too much value on consciousness. I have not seen any evidence that consciousness is nothing more than a convenient way to organize a complex pattern of elementary particles for self preservation. In that sense the discussion of consciousness is an interesting exercise in analogies, and while it may very well put us on a track to better understand everything from quantum mechanics to the mind, it still, to me, has a very anthropic vibe to it.

>You may be able to imagine a body physically identical to yourself not having a consciousness, but you cannot deduce from that, that this is something that can actually happen in our Universe.

Agreed, but doesn't that just further throw us back to the mystery of consciousness as something that transcends the physical? Why would a given arrangement of physical stuff necessarily give rise to subjective experience?

>It may just as well be true that any exact physical copy of a body containing a consciousness is also conscious. In fact, the latter seems like a simpler assumption than the former.

Maybe, but that assumption brings us no closer to understanding consciousness, since consciousness is so radically qualitatively different from physical stuff.

What does "a convenient way to organize a complex pattern of elementary particles for self preservation" have to do with subjective experience, qualia, etc.?

> but doesn't that just further throw us back to the mystery of consciousness as something that transcends the physical?

No, I think it does the opposite - it dissolves the concept of consciousness. "Conscious" or "unconscious" is a label we put on complex physical objects to separate them into categories by their behavior, in the same way we use "hot" and "cold" to separate things by the degree of their apparent temperature. So an atom-perfect copy of your brain would be just as conscious as you, and a program running on a hardware, both of equivalent complexity to you, would be just as conscious as you.

No, we understand this objection, we just don't think it's valid.

Your thought experiment postulates a pattern of matter that is identical to conscious matter but lacking consciousness. But the fact that you can imagine this does not mean it possible or even meaningful. You give no proof that such a distinction can exist.

"Consciousness" is a label humans apply to a pattern of matter that behaves in certain ways, in the same way "liquid" is a label we apply to a pattern of matter that behaves in certain ways. It makes no more sense to imagine a being that displays conscious behaviour but lacks consciousness than it is to imagine matter that acts identically to liquid but is somehow not a liquid.

In your example, if there is no observable difference between the supposedly conscious and non-conscious beings, then there is, scientifically speaking, no difference at all. They are both conscious, in that they both exhibit the behaviours that we call "consciousness".

If you hypothesise that there is a difference, then the burden of proof is on you to demonstrate it, via measurement — again, your imagination does not count — at which point we can use scientific enquiry to find out exactly how this difference arises. However, since the difference is measurable, it is part of the pattern of matter, and so the outcome of this process will simply be a more refined definition of consciousness.

>I can imagine a being that is physically identical to a human in every way, but is not conscious - there is no reason to think that intelligent behavior requires consciousness.

I experience consciousness as having explanatory power for why I have conversations about the ineffability of consciousness. But if I can posit a being physically identical to myself but without consciousness, then, since the causal explanations for the duplicate's behavior are identical to my own, that means this being as well as myself have beliefs and say the things we do about consciousness, that are not caused by consciousness! My own beliefs about consciousness, supposedly incorrigible because they are directly caused or informed by consciousness, are necessarily not caused or informed by genuine consciousness, only some computational or informational imposter. But that means any reason I have to believe I am conscious, the zombie also has. It follows that either I am also a zombie or zombies are incoherent.

You’re aware, I assume, that you’re describing David Chalmers’ “philosophical zombie” thought experiment.

The “Hard Problem of Consciousness”— a term coined by Chalmers— is not universally agreed to actually be such a hard problem.

For anyone who wants to dig deeper into this philosophical discussion, Sean Carroll (who does not himself believe the problem to be so “hard”) has a very thorough and civil discussion with Chalmers about this on an episode of his fantastic podcast, Mindscape:

https://www.preposterousuniverse.com/podcast/2018/12/03/epis...

Yes, I am aware that I am describing what is sometimes called a philosophical zombie.

The "hard problem" is certainly not universally agreed to be hard, but so far I have not encountered any rejections of it that struck me as valid in the least bit. I'll check out your link, though.

If you’re curious, here’s Carroll’s case against the zombie argument:

https://nautil.us/issue/53/monsters/zombies-must-be-dualists...

> I can imagine a being that is physically identical to a human in every way, but is not conscious

I can't, and I think that's the central point. This means you're claiming the big c is not a physics process. But I suspect that the next sentence is a more accurate representation of your argument:

> there is no reason to think that intelligent behavior requires consciousness

To which I would say, what if consciousness is just a side effect of data integration. What if it's just what if "feels like" to be a system that has memory, external and internal data integration? Nothing magical about that.

> So what is the difference between such a being and a conscious human, if the two are physically identical?

Again, you're tacitly assuming that consciousness is not physical, thus assuming your earlier conclusion.

Your objection is far from being obvious. One objection to your objection is that your thought experiment is incoherent in the first place. Consciousness does not have be a property of a being in a separate ontology than material reality. In other words, consciousness is not something associated with conscious physical systems in a different reality, rather it is a property of these physical systems because of their particular function. So, it doesn't make sense to talk about something that behaves exactly like a human, but is not conscious. Something that is physically-indistinguishable from a human will also be "consciousnessly"-indistinguishable from a human.

My favourite book by Greg Egan is "Permutation City." It is a kind of reductio ad absurdum argument for exactly this: if you can simulate brains then where is the consciousness in all of this? Egan's book takes this idea to it's logical extremes. Very cool and mind-blowing when I first read it two decades ago.

Agreed. And I think it's even harder than that. When it comes to studying or thinking about consciousness - there is no way tor me to prove that any other person in the room is conscious except for myself. Given this, how can we even begin to talk about "AI consciousness" or in this case electrons being conscious.

> So what is the difference between such a being and a conscious human, if the two are physically identical?

One can imagine any number of fantastical creatures. A being identical to a conscious being is also conscious, there is no problem there.

I'm saying: imagine a being that is identical to a conscious being in every single physical way, but only in physical ways. Why would such a being necessarily be conscious?

The physical is all there is.

Do you not have subjective experience? If you do, then how is it physical?

Nobody knows how. But the fact that there's nothing else but fully known and understood physical forces pushing around the particles that make up my brain is rock solid science.

You're claiming that there's nothing but physical forces at work because we've yet to observe anything else. axguscbklp is claiming that subjective experience is such an observation.

To find out which position is correct we must attempt to find a link between physical phenomena and subjective experience. If we are not able to find such a link then perhaps they are not actually connected after all.

You are assuming such a link exists and has just not yet been scientifically articulated. For what it's worth, from my own review of the literature, I tend to agree we have been making progress on this front and will likely find a link eventually.

Consciousness could simply be an emergent phenomenon. Mathematical theorems and proofs are not physical things, yet we can create physical systems that can encode and prove theorems. Maybe we are conscious in the same way that these physical systems prove things without being made of mathematics.

For all you know, everyone except you is a philosophical zombie. Consciousness is subjective and you can only make observations about your own experience. Consciousness is an unsolvable problem because of that.

Forgive me if im being dense, but if that human replica still responds to stimulus in some way, and has basic wants and needs, then surely it is still conscious, just lower down on our human scale?

You can build a computer that responds to stimulus - we do it all the time - but that doesn't mean that it necessarily is conscious (that is, has subjective experience). If I build a computer that responds to stimulus, that doesn't necessarily mean (to paraphrase Thomas Nagel) that there is something that it is like to be that computer.

Well, sure, there is something it is like to be that computer, but it is only accessible to the computer, which does not pay a lot of attention to what it's like to be itself generally. When you ask me what it's like I can try to imagine becoming the computer, but in the process I would cease to exist, so for me there is only the objective computer seen from the outside.

The only reason "what it's like" seems to make sense with other humans is because we evolved this nifty faculty of empathy. But if I examine more closely what it's like to be you, I also find that at the point I'm you, and thus have subjective access to your consciousness, I'm not me anymore.

So, it seems that when looking more closely exactly the same problem occurs when asking what it's like to be a computer, or a human, therefore the "what it's like" question is not a valid argument to ascribe consciousness to one but not the other.

It is possible the computer pays as much attention to its own subjective experience as you or I but we have not given it the tools necessary to express itself.

Reply on news.ycombinator.com