Hacker News Reader

Comment by pontus | Hacker News Reader

Honestly, as weird as it might sound, to me it's even weirder to claim that consciousness somehow is just a manifestation of effect that we are already familiar with.

Saying things like "Consciousness is emergent" or "Consciousness is just a side effect of information processing" seems to miss the point. I'm definitely comfortable saying that consciousness is correlated with complexity and information processing, but claiming that it is fully explainable in some mechanistic way sounds suspect to me.

It would sort of be like saying that since large electric charges have only been observed in nuclei with proportionally large mass, then somehow charge is just a manifestation of mass. When we go further it turns out that charge is fundamentally distinct from mass, in fact so distinct that we have to add an entirely new attribute to fundamental particles. The same story goes with spin. Once you have spin and electric charge you get magnetic moments. Ultimately the reason why a magnet is magnetic is because all of these subatomic particles conspire in just the right way to have a macroscopic effect. Of course, not all materials have the property of being magnetic even though they all are made from protons and electrons, but some materials are.

To me, trying to say that consciousness is just something emergent is like saying that electric charge is emergent from mass. I would not be surprised if some type of proto consciousness would be needed in order to understand how macroscopic objects like human brains are self aware, have sensations, etc.

View on HN

> Saying things like "Consciousness is emergent" or "Consciousness is just a side effect of information processing" seems to miss the point.

It sounds suspect to me too, but the reason the idea is appealing is that it's worked fantastically for hundreds of similar phenomena. To vastly oversimplify thousands of years of philosophy, for most of human history the default position has been that every idea had to be reified into some metaphysical essence.

What is the essence that makes a dog a dog? That makes bread nourishing but rocks not? That makes rocks solid? That makes fire hot? That makes falling objects seek Earth? Again and again, we've found simple physical explanations for these puzzles, that once were regarded as just as inexplicable as consciousness. Indeed, consciousness is now the last essence that survives, and that's why some bet it'll go the same way as all the rest. "But this one is different", indeed, and every single time so far that argument has been wrong.

My background is in theoretical physics, so I absolutely subscribe to this perspective. In understanding my own consciousness, however, I find it a lot harder to see how such an explanation could work.

The only other problem that I feel may require a similarly "out-there" explanation would be the measurement problem in quantum mechanics. Why is it that a macroscopic object, made entirely of electrons, protons, etc. when interacting with an electron means that a measurement occurs, but somehow when a single electron interacts with that same electron we can just use the Schrödinger equation? The existence of some type of cutoff between macroscopic and microscopic seems to me to suffer from inevitable inconsistencies. More likely, I would think, is that something like the many worlds interpretation is correct.

Similarly, I wouldn't be surprised if something similar will happen when we understand consciousness more deeply. Simply saying that it's an emergent property is sort of cheating and discounting how different the sense of experience is from anything else that we have seen in nature.

We can imagine consciousness as a purely informational process, which can occur in any adequately complex substrate—real or virtual. In this regard (integrated information theory), the inherent “properties” of consciousness would be more readily described using graph theory or topology.

However, we can also view physics itself as something inherently informational, given that all the base particles in the standard model can ultimately be boiled down to scalar values. Would it be possible to say that those values are ultimately derived from the inherently mathematical properties which result in this reality “existing”? What does it mean for something to exist on the basis of its inherent mathematical identity?

If both consciousness and reality are purely mathematical phenomenon (maybe consciousness is more derived through the computation-generating properties of life), why would one such phenomenon be any less expected than the other? Isn’t it impossible for a non-mathematically-derived (informational) consciousness to exist at all in such a universe?

Last I heard, integrated information theory is _not_ neutral w.r.t. substrate, i.e. to compute the consciousness of a human being simulated on a computer chip, Tononi wants us to apply the IIT calculations to the computer chip.

Oh, certainly the processes occurring on the chip itself could bring about an entirely separate consciousness while at the same time simulating a human one on a different level.

> Why is it that a macroscopic object, made entirely of electrons, protons, etc. when interacting with an electron means that a measurement occurs, but somehow when a single electron interacts with that same electron we can just use the Schrödinger equation?

The way I understand it is that particles are fuzzy, but if they interact with sharper things they get less fuzzy themselves. Naturally large amounts of interacting particles (macroscopic objects) are way sharper than single isolated particles. So measurement is just theoretical concept that relates to system physically getting immensely less fuzzy to a point we can safely assume for the sake of easier calculations that it's not fuzzy at all. You can still describe such sharp system with Schrödinger equation, it still works even after measurement, but you don't do that since there are easier ways like Newtons mechanics. There's no artificial cutoff. There are just many orders of magnitude difference of how sharp things become through interaction.

You can see that experimental scientists managed to make pretty large systems to display quantum superposition if they are properly isolated from interaction with way larger systems that could knock them out of superposition.

As for consciousness my opinion is that it's just some specific algorithm that has some evolutionary utility for some animals. It's fairly good for modelling self, environment and the future. It's not in any way pinnacle of development. It's gradual like human has more consciousness than dog, and dog more than a hamster, but not to the point of absurd, that single electron or even single neuron has some amount of consciousness. Just the same as you shouldn't think that single electron or even neuron has some intrinsic ability to perform sorting or image recognition. You need hardware, and yes, more powerful hardware yields stronger results, but you still need to have at least rudimentary versions of correct algorithm running on that hardware before you can actually say it does the thing.

> So measurement is just theoretical concept that relates to system physically getting immensely less fuzzy to a point we can safely assume for the sake of easier calculations that it's not fuzzy at all.

It appears to be more complicated than that, see Wigner's friend[1] thought experiment and a recent experimental realization of it[2].

From the paper's abstract: "In a state-of-the-art 6-photon experiment, we realise this extended Wigner’s friend scenario, experimentally violating the associated Bell-type inequality by 5 standard deviations. If one holds fast to the assumptions of locality and free-choice, this result implies that quantum theory should be interpreted in an observer-dependent way."

[1]: https://en.wikipedia.org/wiki/Wigner%27s_friend

[2]: https://arxiv.org/abs/1902.05080

Wigner's friend experiment doesn't contradict my interpretation in any way. For me it's not the exchange of information that causes system to become more sharp. It's poking that system with immensely huge stick built out of billions and billions tightly and sharply interacting particles that you have to do to get that information.

So when Wigner's friend makes the measurement of quantum system through interacting with it he causes it to collapse to sharper form. Wigner unaware of the result may still think that there's still some fuzziness in the system that consists of his friend and the experiment he performed. But in fact all the fuzziness was gone when his friend performed the experiment. The fact that Wigner doesn't know that yet, changes nothing in reality. Sure, he can treat his friend and his experiment as if they were quantum system. But they are not. The math I think looks the same if you don't know things about sharp stuff or if you know all there is to know about fuzzy stuff.

The actual 6-photon experiment is a bit too technical for me but I don't supposed "Wigner's friend" in it is built of billions of particles interacting with each other. If the friend is something smaller, like just a few particles that perhaps loosely interact then it's perfectly fine to expect that after measurement of one fuzzy system by another there's plenty of fuzzines left for the macroscopic observer (Wigner) to see.

> The fact that Wigner doesn't know that yet, changes nothing in reality. > The actual 6-photon experiment is a bit too technical for me

There's a more approachable explanation, and lengthy discussion, over at Physics Forums: https://www.physicsforums.com/threads/a-realization-of-a-bas...

I'm no expert, but it seems to me the issue is of a different, more fundamental, nature.

> For me it's not the exchange of information that causes system to become more sharp. It's poking that system with immensely huge stick built out of billions and billions tightly and sharply interacting particles that you have to do to get that information.

That's exactly what I think about Schrödinger's Cat experiment. It's often implied that the hypothetical cat may be simultaneously both alive and dead, as if there are only two possible states for the cat.

There are approximately 7 x 10^27 atoms in the average human body. The figure for a cat wouldn't be much far from that I suppose. Taking that into account we can safely say that the quantum system (cat + radioactive source + poison) may actually simultaneously be in an abysmally high number of different entangled states, which may converge to, as you say, a "sharp" form.

Doesn't this experiment have a simple explanation that measuring something is essentially entangling yourself with it? So at the end of the thought experiment, both scientists and the test device are in a superposition together...

...except it's not that simple, and here is why I don't like those experiments (and why in reality, they're always realized using atoms, and not cats): there's no Wigner, or Schrödinger's cat, as a unit in physical sense. They're made of atoms. Atoms that interact and radiate information all the time. That box with a cat with it, the cat radiates information at the speed of light, which gets absorbed by the box and reradiated away. If the result of your thought experiment could be changed or confused by the following setup:

- a) put a broad-spectrum camera suite around the box prior to the experiment

- b) have it record data on the hard drive

- c) have the experimenter do their experiment

- d) have someone pull the data from the hard drive and read out the actual state of the experiment subject

... then it means your explanation for the thought experiment is wrong.

Yes it's extremely difficult not to leak (potential) information. This is called decoherence in these situations. It doesn't matter if you don't become conscious of the measurement result as long as some atoms in your body gets contaminated by the measurement they get pulled into the same branch (or whatever concept you subscribe to, it all works the same).

People get hung up on this semantic concept of 'measurement' when that's not what it's about at all. It's not "if you measure a property of a particle then you change that property", it's "if you interact with a property of a particle then you change it, and that it's impossible to measure without interacting".

The electron doesn't care whether you're conscious and certainly not whether you're writing anything down. Its momentum or whatever is still going to change regardless.

Indeed. Measurement is a physical process, not metaphysical one. To measure something, you necessarily need to interact with it (and at this point it's worth remembering that photons bouncing off an object and hitting your retinas absolutely count as interaction, so you can't e.g. put a ruler next to the dog and read out the results, and say you weren't interacting with it, not in the sense physicists use this word).

How does that explain the quantum eraser experiment?

a photon that has been "marked" and then "unmarked" will interfere with itself and produce the fringes characteristic of Young's experiment.

https://en.m.wikipedia.org/wiki/Quantum_eraser_experiment

Did you mean to reply https://news.ycombinator.com/item?id=23221433? I don't see how QEE is incompatible with the notion that measuring something necessitates interacting with it. Photons are usually measured by absorbing them and turning into electricity.

Absolutely! I wish this were more commonly understood. And, if you take that train of though one step further, it implies that quantum mechanics exhibits non-local causality.

The electrons in ambient air are interacting but not measuring.

Anybody have a really bad negative feeling toward the deconstruction of science ? There was an obvious degree of existential improvement for a while, but then dissecting everything at this points .. kinda hurts my soul. I somehow wants to keep some magic unexplained phenomena to stare at without any theory, just feeling.

I recommend reading Joy in the Merely Real by Eliezer Yudkowsky (and its follow-ups). It might help you with that. https://www.lesswrong.com/s/6BFkmEgre7uwhDxDR/p/x4dG4GhpZH2h...

Alternatively, there's the obligatory xkcd (https://xkcd.com/877/), but I feel it kind of misses the point a little.

> Anybody have a really bad negative feeling toward the deconstruction of science ?

C.S. Lewis for one, this is from "The Abolition of Man":

> But you cannot go on 'explaining away' for ever: you will find that you have explained explanation itself away. You cannot go on 'seeing through' things for ever. The whole point of seeing through something is to see something through it. It is good that the window should be transparent, because the street or garden beyond it is opaque. How if you saw through the garden too? It is no use trying to 'see through' first principles. If you see through everything, then everything is transparent. But a wholly transparent world is an invisible world. To 'see through' all things is the same as not to see.

Well every layer man made transparent changed his life, and it became a chase, technologically driven mainly, to do so ad infinitum. Even though we're seeing resistance now because ignored side effects are biting us, I don't know if societies won't keep digging beyond QM so man can have room temperature super conductors and backyard fusion to power neverending fountain of youths. Until their sanity disappears.

You might like a site I made about this (or you might not...): http://www.lifeismiraculous.org/short.

>Saying things like "Consciousness is emergent" or "Consciousness is just a side effect of information processing" seems to miss the point.

On the contrary, I feel like those who think we need some fundamental consciousness property is missing the point. We expect to find some objective third-person property that is the substrate of consciousness because a phenomenon that seems so unlike every other property should have a fundamental basis. But the mistake is expecting to analyze consciousness in the same way we can analyze other objective properties of physics.

The only things we know to be conscious are complex macro-scale objects with highly complex and rare internal organizations. In fact, the only things we truly know to be conscious are ourselves. It seems to be fundamentally subjective; it is not something to be directly witnessed as a third-person observer. So our investigation should start there. What we want is a theory for deriving organizational subjectivity from a non-subjective substrate. What might this look like? A system with subjectivity will need to recognize the external from the internal. It needs an egocentric mode of representation with the ability to represent itself and its dispositions and intentional stances, as well as states to represent the external world. My intuition tells me such a system has a non-zero "inner life", i.e. there is something it is like to be it. But I see no reason to think "fundamental" subjectivity, whatever that is, could do any explanatory work here. The causal and representative power is in the organization.

It’s even worse: How can you be sure that you can separate the internal from the external? Both seem to have to be present in order to make sense of anything.

So, in a sense, everything is connected to everything else? What do you call that, hmm? ;)

This is an existential and observational query, not about industrial optimizations (ie. what works). What works is killing us!

Indeed. I think IIT is a good theory... of something, just not exactly consciousness. Maybe a precondition to consciousness or something like that. But the thing that we think of as our consciousness is to me best explained by the "global workspace" theory, which says consciousness is the process of the various specialized parts of the mind, which are constantly working separately and in parallel, communicating their state to each other. It's like a boardroom for the society of mind, where at any point one subsystem has the podium (although there is lots of chatter and crosstalk as well). For most of us a part of the language subsystem (Gazzaniga's "interpreter") is also giving a running commentary (the internal monolog) of the information it's receiving from the other parts (with a lot of its own interpretation thrown in)... but this is not an essential feature of consciousness! We have a tendency to identify our consciousness with this commentary, but that is obviously incorrect. I think that the communication in this global workspace occurs in its own "language", a language internal to organic brains, capable of abstracting and reducing to its barest essence information from any of its components.

This view of consciousness is phenomenologically best aligned with most of the (admittedly limited) objective information we have about human conscious experience, and is consistent with the experience of various altered states of consciousness such as meditation or the use of entheogenic substances. It explains how consciousness is only a small part of what happens in our mind and why the nature of the subconscious (which is most of what actually happens in the brain) seems to be so hard to nail down. It also means that any being with a "mind" that has numerous independent and parallel processes that need to be coordinated has some measure of conscious experience, even invertebrates, and probably even living things whose information processing uses an entirely different infrastructure, such as plants. However, I can't see any way that this definition of consciousness could apply to an electron.

Edit: I think that the global workspace theory of consciousness can probably be mathematically described by IIT, but not just any integration of information results in something that deserves to be called consciouness. The information that's being integrated should be combination of perceptions (feedback from the environment) with some kind of memories of previous states, resulting in new memories and predictions, and the integration should happen through independent pre-processing of this information by relatively independent subsystems. This is still general enough to apply to nearly everything living, but I think it puts conscious experience at a higher level than merely integrating information.

Yeah, I definitely like IIT and think its on to something important. But it doesn't strike me as a sufficient condition for consciousness. I have a lot of sympathy for GWT. One of its theoretical virtues is that it coheres with theoretical properties of consciousness with independent justification like integrated information, recurrence, self-modelling, etc. But it still lacks any direct theory of phenomenology, i.e. qualia. Although I can see why scientists would avoid attempting such arguments if at all possible. This would be a good place for philosophers to bridge the gap, but I guess it is easier to make a career out of promoting panpsychism these days than to come up with something insightful to say about mechanistic consciousness.

But to move the discussion forward, I think one obvious property of qualia is that it is representational. That is to say, it is structurally related to the thing being indicated such that it can inform about the thing. For example, the red quale tells you something about red substances in the context of the space of possible colors, the external world full of beneficial and harmful substances, and the bearer of the quale with drives, dispositions, preferred states, etc. This complex millieu of properties, states, dispositions, etc, all serve to inform the properties of a quale. Its representational power is one that gives the bearer certain competencies in the actual world, e.g. pain gives one the competency to avoid damaging states. But this representational power must be intrinsic to the structure that constitutes a quale. If this were not the case, then its power to confer competency would be contextual. Pain would only confer competency in the right environment (like a reflex that has meaning only in the right environment, e.g. the grasping reflex of an infant). But this isn't the case with qualia; the experience of pain is intrinsically representative and provides its bearer with competence universally. The same can be said for emotions and our senses. This suggests to me that some kind of recurrent structure is a necessary condition for a quale: to simultaneously be the producer and the consumer of a representative state, and consume in such a way that necessarily confers competent behavior. But this discussion sounds like a different level of description of coordination between different subsystems. Information from different subsystems bear on this central coordinator, and this information confers competent behavior on downstream subsystems, i.e. contextually relevant causal powers. I see the beginnings of the details required for mechanistic qualia in theories like GWT and others based on principled analysis of brain networks.

> What we want is a theory for deriving organizational subjectivity from a non subjective substrate. What might this look like?

I think the Reinforcement Learning framework is useful here. An agent exists inside an environment, it has some rewards, it is capable of sensing and acting, and its goal is to maximise its rewards over time. This implies learning, exploration, planning and the ability to model itself and the world. Of course it is possible for many agents to share the same environment and have different bodies, rewards and needs.

I think sensing, learning and evolution are good filters for judging if something could be conscious.

this totally elides over the problem of consciousness, of qualia, which is the essential question

I'm personally a fan of attention schema theory. It's based on the idea that consciousness emerges from an advancing capability to create models of both the external and internal world. For me, it scratches a couple itches. First, it's not an all-or-nothing evolutionary gamble, but develops in steps and each step has advantages. Second, it seems pretty well rooted in neuroscience and psychology, rather than relying on handwaves like "all matter has consciousness".

Here's a couple articles from Michael Graziano, the founder (discoverer? namer?):

https://www.nytimes.com/2014/10/12/opinion/sunday/are-we-rea...

https://www.theatlantic.com/science/archive/2016/06/how-cons...

We all know what a forest is. But there is no "particle of forestness".

Most of the words we use don't need laws of physics and particles exclusive for them. What makes you think consciousness is more like electromagnetism than it is like a forest?

In my opinion consciousness is just an abstraction, like a football match or a forest. We have a word for these common configurations of matter, but it doesn't mean that they are fundamentally, qualitatively different from the rest of the universe.

Show me an elementary particle of forest then I can admit that consciousness needs one too :)

Consciousness is not just an abstraction, requires self replication/evolution and ability to create models of self and environment - both being extraordinary feats.

A perceptive self replicator existing inside a multi-agent environment that affords both cooperation and competition, and who's existence depends on its ability to adapt and act well - this configuration is what creates consciousness in my opinion.

There is no particle or fundamental quality of consciousness, but it is what self replication under limited resources leads to.

Consciousness seems like it should be special, but that’s not evidence that it is. Consider, after a lifetime of subjective experiences from exercise etc, a cardiac surgeon can still know far more about your heart than you do. Nobody is arguing some mysterious force is pumping blood through their bodies, but the feeling of blood pumping through your veins somehow feels primal not simple plumbing.

You can intellectually extend that to everyone else about our bodies, so why not consciousness?

Understanding how a brain works may only require a mechanistic model, but consciousness is different.

Consider the following thought experiment: suppose that we had a complete understanding of how neurons work (as well as the brain more broadly). I think this is totally reasonable. Then suppose that we created artificial neurons that acted identically to biological neurons. If we replaced the neurons of a brain with these manufactured neurons, I would imagine that the resulting brain would still be conscious. Now imagine taking some of these neurons and removing their interior structure instead routing the inputs / outputs to a machine somewhere else where the correct computation takes place in order to determine the output from the neuron. From the rest of the brain's perspective nothing has changed so presumably it's still conscious.

Now take this to the extreme: replace all of these neurons with empty shells and route the computation to a billion people in the world to perform on pen and paper. Is it still conscious? If so, what if you removed the brain altogether and just performed the computations directly on paper, is the paper conscious?

The paper would definiy still be intelligent, it would possibly come up with interesting inventions, claim that it was sad or happy etc, but would it actually be sad?

Sure, why not? That's literally what the materialist definition of consciousness means: it's a process that can be carried out on any appropriately-configured physical substrate; ours just happens to be neurons.

You're describing the systems reply to the Chinese Room Argument[1] and I agree with you. To go further, what else could possibly generate consciousness? A soul? Some phlogiston that breathes fire into beings? I legitimately don't understand what the alternative is to physicalism without positing some completely unsubstantiated Other Thing that also gets us no closer to understanding consciousness.

[1] https://en.wikipedia.org/wiki/Chinese_room#Systems_and_virtu...

I agree that other mediums (like artificial neurons) would likely still be conscious, but an abstract calculation on a bunch of pieces of papers, possibly spread over years and across multiple people performing the calculations? That seems odd to me. Definitely still "intelligent", but should I feel bad about ripping up those pieces of paper? Should I feel like I'm killing someone by erasing calculations off of the paper or introducing a mathematical mistake somewhere in the middle?

> should I feel bad about ripping up those pieces of paper?

Do you feel bad every time you have a beer and potentially kill off a few brain cells?

What does morality have to do with the nature of what is? How you feel about snuffing out some form of consciousness is irrelevant to whether it exists or not.

The idea that scribbles on a paper would be conscious is really hard to believe. It also sounds like it's own form of dualism that's functional instead of substance-based. I wouldn't consider it a good physical theory of consciousness.

Well, to a panpsychist, there's no reason why "conciousness" would need to stop at a human scale. The universe as a whole might be concious, and part of that concious experience might be isomorphic to the subjective experience of the brain that's being replicated on those pieces of paper. Though I can see how people might object that this is merely conflating "conciousness" with "computation", that's really inherent in what the thought experiment is doing.

The original claim was lots-and-lots-of-people calculating on many pieces of paper. But it's just an example of this endless stream of sort of ridiculous bait-and-switch arguments. First someone, says "so it's mechanical, so a big, fast, accurate speak-and-spell could simulate it" and you say "well, OK, it would have to be huge, incredibly fast and accurate and so-forth" and then the person "well, I doubt consciousness is just a speak-and-spell.

If the claim is that certain functions are what "generates" consciousness, then it's fair to ask whether any sort of functional arrangement will do, including writing stuff out on paper. I don't think it matters whether a billion Chinese are busy writing out 1s and 0s, or a computer is moving electrons around. None of that is conscious in my view, or at least I see no reason it would be.

I would sort of agree. Just avoiding first positing "well, even something silly would do" and then saying "well, that proves it's impossible 'cause it's silly". Then the argument is clearly unfair.

What makes electrochemical and chemical reactions in a brain so special that you see a reason for them to create consciousness? Or is it a lack of complete understanding of those processes that makes you think so?

Your billion-person computer certainly would not appear to be conscious in real time - maybe that's why you just can't get your mind around the picture.

Can you imagine a billion people, none of whom who have never even heard of chess, writing out 1s and 0s, and thereby beating a grand master at the game? (in an appropriately slowed-down game, of course.)

Yes.

There’s a short story from 1981 by Vernor Vinge titled “True Names”. Recommended reading.

>The idea that scribbles on a paper would be conscious is really hard to believe.

But it's not that the scribbles on paper are conscious; scribbles on paper have no causal powers for example. The scribbles on paper are the volatile storage of this causal chain: the causal cascade flowing through the actions of the person reading the symbols, manipulating them, then writing out new symbols. The bait-and-switch is that when you adjust the thought experiment, the focus shifts from the unified casual process performing certain computations to the inert substrates of the paper (or in the case of the Chinese room, the man performing the computations). But no matter the medium of computation, the causal chains are still instantiated and so there's no reason to think this process is not conscious.

Fair enough, but what makes the instantiation of causal cascade conscious? I see no reason to suppose it would feel pain or see color. What makes it so?

You're looking for a hard cut off where none exists. Most likely everything is conscious on many different levels and scopes of capability.

So the case which makes people really uncomfortable - are cows conscious (because we eat them) has the very unsatisfying answer of "yes, not in the same way we are most likely, but in some way".

Why wouldn't it? Supposing that it doesn't presupposes that somewhere along the gradual chain of replacing a human brain with this paper processing system those properties are lost either gradually or abruptly.

Assuming it's the functions the brain performs that are conscious. But it's part of the hard problem. Why and how is anything conscious in the physical world? What does that mean for other physical arrangements? How would we know?

> Assuming it's the functions the brain performs that are conscious.

As far as I know, if you turn off someone's brain functions you also turn off their consciousness. Moreover, consciousness can be manipulated by manipulating the brain. It stands to reason that consciousness is a subset of brain function (and body, sure). But I concede that this is not a rigorous proof of anything. Perhaps the consciousness treats the brain like a comfortable chair, and when it is destroyed gets huffy and leaves to go somewhere else. Seems unlikely though.

> But it's part of the hard problem. Why and how is anything conscious in the physical world? What does that mean for other physical arrangements? How would we know?

Hard to state, perhaps. Hard to answer? I suppose we'll find out once stated whether the question is actually hard to answer.

Do you see a reason why a mechanical arrangement of atoms could feel pain or see color? Because they do.

Scribbles of paper serving as a part of a conscious system is indeed an implausible scenario. You'd need a huge number of pieces of paper, interacting in very interesting ways. It's an example specifically designed to be implausible, and so is a very poor guide for intuition.

> Do you see a reason why a mechanical arrangement of atoms could feel pain or see color? Because they do.

That's the hard problem.

There is indeed a lot about the workings of minds that neither I nor anyone else knows, but I do not mistake my lack of knowledge for strong evidence that certain things cannot be so. That would be an example of what Dennett calls the philosopher's syndrome: mistaking a failure of the imagination for an insight into necessity.

It would be like a pre-Archimedean philosopher asserting that iron boats are impossible, because how could they float?

Relevant comment that argues in this direction: https://news.ycombinator.com/reply?id=23221295

That was actually explored in https://xkcd.com/505/ but with rocks instead of paper. Also "Ship of Theseus".

My take on it, is that my consciousness is not the one I was born with, or the one I had a few years ago -- instead, it just thinks it is that same one, but the only link to previous versions is just memories.

It seems like the "thought experiment" has no palpable meaning. You are basically repeating the OP's argument "I don't think consciousness can arise mechanically" except more slowly. There's no more argument there. Well, except that an entire universe of paper and pencil calculations probably could not simulate a neuron in the age of the universe. But that's not even interesting.

People keep saying "consciousness" but it sounds like the topic is more specifically identity.

I have a brain, you have a brain, billions of others do, it seems like there is a reasonable amount of symmetry, but "I" look out of my eyes and not anyone else's. Replacing you with an AI is much more feasible from my perspective than replacing me with a program.

I think identity is definitely related to consciousness but you can have no sense of identity and still be conscious.

Yes! I completely subscribe on the idea that consciousness is an emergent property, but I can't begin to see how the asymmetry of me vs anyone else is reducible.

Your post sounds like it may be inspired by the work of Daniel Dennett and his riposte to the Searle Chinese room thought experiment.

It’s actually the China Brain scenario

Woah, that's awesome! I had no idea this was a thing :) Thanks!

Or go even further and say there is nothing special about the calculation being performed or series of actions. What is important is the mathematical structure. And the structure itself has no physical existence. Therefore, no one really exists.

>but would it actually be sad?

To the fullest extent that you are capable of being sad. There is no even remotely plausible alternative to this physicalist argument.

I would argue that you aren't nearly as conscious as you think you are. That's the conclusion I've come to after many hours and years on the meditation cushion cultivating awareness of my own cognition. Any thought, choice, or action that you make doesn't actually happen in your conscious brain. You just become aware of it after it's happened. That's all there is.

That doesn't solve anything though, there is still the "you" that becomes aware.

Not really. If you really start digging you realize that there isn't much "you" there left to be conscious. To a great extent I think consciousness is largely a bit of an illusion. Most of what people typically think of in terms of consciousness is really the contents of their attention. But you don't even truly control that attention. The decisions to switch from one unconscious information stream to another are themselves unconscious information streams. In essence, I'd argue that upon inspection, "attention" is the apparent last vestige where any consciousness may reside. But even deeper inspection reveals that it itself is quite an empty concept.

That’s the “Flashlight in the Dark” concept of consciousness as the focus of attention.

That's a good description of where I'm at, and I keep wondering if maybe the definition most people are using for "consciousness" just doesn't match mine. The longer I meditate, the less consciousness matters to me; it seems totally clear that I become aware of things after they've already been generated elsewhere in my brain, and that awareness is just a kind of self-reflection which maybe helps with higher-order planning or whatever. I'm still open to realizing I've missed the mark--especially because so many prominent meditators seem very focused on (or even enthralled by) mysteries of consciousness. Until then, I'm really struggling with why people would try to elevate this subjective experience to a fundamental aspect of physics.

> I keep wondering if maybe the definition most people are using for "consciousness" just doesn't match mine

I think this is true, when we talk about consciousness we talk about it as if consciousness has the ability to control the body and speak and that our thoughts are controlled by our consciousness. I think illusionist arguments are very strong here against that. We can change how someone thinks or acts based off of physical stuff (like lead poisoning someone) therefore those things must mostly be physical.

I have come to the conclusion that conciseness probably is only an experiential thing. It might not be able to control anything but there is something fundamental there experiencing something. We cant know that the world is real: we could a brain in a jar, or in the matrix, or on a massive DMT trip right now. But we can know "I think therefore I am", probably better written I experience therefore I am. It seems very strange to throw out the one thing we know to be true, that we are having a conscious experience, in favour of something we don't know to be true, that the world is real and causing that experience.

> I have come to the conclusion that conciseness probably is only an experiential thing. It might not be able to control anything [..]

Doesn't it cause us to at least have these kinds of conversations?

There is always still the _subjective experience_ of paying attention to whatever I'm paying attention to, of "being the one that sees my visual field", and so on.

There must be fundamental difference between the subjective experience I have of vision and that of a computer with a camera and processing software, I can't imagine that it has a similar experience.

How come there _is_ a subjective "me" that experiences things and can pay attention to them? Given that we are clearly bags full of extremely fancy chemical reactions.

I think there is a difference there, but it's largely because the computer with a camera is such a simple system at this point.

In contrast, you have many, many layers of very sophisticated and interconnected abstraction and reality modeling between that visual stream and other forms of processing. Typically, the higher the level of abstraction, the more "aware" of it you are as it gets filtered and dumped into your attention centers.

In short, even our most sophisticated state of the art "deep" learning algorithms are but puddles compared to the ocean of depth available in your brain. ...and almost none have any form of attentive aggregation and selection.

> There is always still the _subjective experience_ of paying attention to whatever I'm paying attention to, of "being the one that sees my visual field", and so on.

"Being me" is an experience or feeling. We have lots of other feelings both from within and outside our bodies. Could it be that consciousness is simply how it feels to have a focus of thoughts and attention in our brains?

Are you aware of the model of the mind system from The Mind Illuminated (a book that teaches meditation)?

First, I have to clarify concepts. There is a consciousness created by the mind. It's the place where sensory input is experienced. It's also the place where thoughts are experienced. It's basically the screen that allows the different parts of the mind system to communicate.

Then, there's the consciousness talked about in this article. It's a more primordial quality.

Now, that model of the mind system says that attention and awareness are the two modes of the mind. Consciousness as the screen of experience is created by the mind. Perception is created by the mind as well.

How could you possibly know that you "became aware" of anything without some information being sent back from that-thing-that-became-aware? You know it so well that you are even able to comment about it.

I think the fact that people ('s brains) are so sure that they are conscious, is pretty clear evidence that for some reason your brain is getting messages back from this consciousness thing. Either that, or brains are consistently are built so they can't shake the idea that they have an observer (see: how many comments are on every HN article about consciousness, including this one).

>How could you possibly know that you "became aware" of anything without some information being sent back from that-thing-that-became-aware?

I'm sorry, but I don't quite understand either your question or the following paragraph. However, I feel like it could be an interesting line of discussion if you can explain it to me better.

I'm a bit lost, it sounds like you're suggesting consciousness and your brain are too different things? Is that right?

Well the two things are the part where physics does stuff, then the other part is where you become aware of it (whatever that even is, but you experience it none-the-less. Let's call it the consciousness).

Random assumptions/observations that might be relevant:

- The universe seems to generally work under "simple" principles that can be explained with a few mathematical equations.

- The brain is made of the same stuff as everything else.

- You are only aware of a very very small amount of things that actually happen in your body: You are not aware of what happens in your visual cortex, just what comes out of it. You are not usually aware of the sound waves themselves, just the conceptual "sound". I would call all that stuff "processed data". Sound, images, touch, all that stuff, generally get blended together into one cohesive experience.

- All that looks like a flow chart when I think about it. The photons go into your eye, stimulating nerves, electrons flying everywhere. That cascade of nerve stimulations goes into the visual cortex where stuff is processed into actual stuff that you care about, but the data is still just essentially electrical signals at that point. But then suddenly, you are aware of that processed data. Sounds like it explicitly got that processed visual data and sent it out a transmitter, to be listened to by none other than my "self".

- I guess I would then relate my consciousness to a radio receiver that's waiting on the other end. And in some specific part of my brain is a transmitter.

- A radio station can't possibly know if anyone is listening to the signal they sent, unless the listener communicates back that they are listening. In the same way, it doesn't make much sense to talk about how aware you are if the "aware part" (the radio receiver) doesn't talk back.

- Random final observations if this model were true: To maintain conservation of energy, the receiver would likely have to act like a capacitor, storing energy that was used to send the message to it, and using that energy to send a message back. Either that or it communicates back by collapsing wavefunctions (okay now it's getting a little out of hand). Also, maybe using this model, you could say that your sense of the progression of time is proportional to the amount of information sent. Which is why time goes so quickly while you are asleep, because your brain stops sending stuff out the antenna (also getting out of hand).

Hopefully you find this idea at least a little interesting. I feel like there's a world of possibilities here that haven't properly been considered.

I would love to hear your take on it.

Ah yes ok I get it now.

There is feedback from the attention centers. In particular, when you pay attention to things, those connections get reinforced. So the real feedback occurs on a bit of a different timescale than the rest of your cognition, though there's probably some shorter term feedback too if I had to guess.

In a sense, I think there's a center of attention/awareness that consists of almost all of what we usually call consciousness largely because it's contents also get fed back in as an input. There's the self-awareness component.

>To the fullest extent that you are capable of being sad. There is no even remotely plausible alternative to this physicalist argument.

Actually that's just hand waving, and taking for granted what needs to be proved.

I feel like you could be equally dismissive to the statement "There is no remotely plausible alternative to a godless universe."

And yet, I think that statement stands pretty strong on the basis of what we do and what we can know.

It's premature to take for granted that a paper brain emulator can exist in principle.

Take something else less...emotionally laden. A black hole. Could we model a black hole on paper? If so, we ought to just be able to drop a computational particle in and see exactly what happens at the singularity. Except we can't -- the whole reason we call it a singularity is because we get divide by reality errors if we try to compute it.

But I would argue there's a different problem. A black hole is the most computationally efficient method of calculating itself. Any method of accurately emulating the black hole on the Planck level is going to take more energy (and mass) than the actual black hole. As a result, those pieces of paper would literally collapse into an actual black hole long before they could properly emulate the black hole they were modeling.

Which isn't to say that the brain is anything like that but the point is, it's not a foregone conclusion that we can construct a completely accurate emulator of a physical object.

Finally there's something of a category error in making the claim that pieces of paper can be sad. A paper emulation of a bar magnet would not itself be magnetic. In the world of the emulation it would produce an output that would model magnetism, yes, but that's the extent of it. Our paper brain emulator, if it were constructible, would produce output that would be a model of sadness in the emulation, but it would not actually be sad.

> As a result, those pieces of paper would literally collapse into an actual black hole long before they could properly emulate the black hole they were modeling.

That is wrong. Any method of emulating the black hole within a volume less-than or equal to the Schwarzchild radius of the black hole would collapse, but you could do the calculation over a larger volume and avoid that. The most massive stars are far more massive than the least massive black holes.

The smallest black hole known that I could find from a quick search is XTE J1650-500, at about 3.8 solar masses with a "15 mile"[1] diameter. There are a lot of stars more massive than that.[2]

[1] https://www.nasa.gov/centers/goddard/news/topstory/2008/smal...

[2] https://en.wikipedia.org/wiki/List_of_most_massive_stars

>Any method of emulating the black hole within a volume less-than or equal to the Schwarzchild radius of the black hole would collapse, but you could do the calculation over a larger volume and avoid that.

...This was my intuition as well, but I'm far from certain about it. It's not uncommon for the universe to find sneaky ways to prevent us from "cheating" so to speak. These often result from deep symmetries/conservation laws and fundamental limits on information.

I'm actually a big opponent of the simulation argument for that reason. I don't think you can accurately simulate the universe without having a universe to simulate it in. Otherwise, you could just 'turtles-all-the-way-down' the simulation. The information density would have to be unbounded, and this seems to be in disagreement with fundamental laws of the universe.

We don't have a theory of quantum gravity... it's definitely possible that to simulate the blackhole in a more spread out manner would be impossible. I could envision a fundamental tradeoff where to replicate the information exchanges between the microscopic constituents requires that either you satisfy the hoop conjecture, or you keep increasing the size of your model without bound, whereby for any finite size you still satisfy the hoop conjecture. (Particularly since the required mass/energy density goes down significantly as the radius increases.) I know that's wild speculation, but I feel like it's not quite crazy enough to take for granted.

I think I'm almost entirely with you except for these two statements:

>It's premature to take for granted that a paper brain emulator can exist in principle.

I think it is premature to say we could simulate all of quantum mechanics with sheets of paper, so on that technicality, I totally agree. However, I think it's quite unlikely to be the case that we couldn't in principle simulate the requisite components required for a faithful replication of consciousness. But you're correct that it's probably not a given.

>Finally there's something of a category error in making the claim that pieces of paper can be sad. A paper emulation of a bar magnet would not itself be magnetic. In the world of the emulation it would produce an output that would model magnetism, yes, but that's the extent of it. Our paper brain emulator, if it were constructible, would produce output that would be a model of sadness in the emulation, but it would not actually be sad.

I think I disagree with almost all of this. The bar magnet is a bit of a bait and switch because the system, the input, and the output weren't well defined. In consciousness, you need to define the system, system input and system output properly. If the system has persistence of thought/computation, and the inputs and outputs can be defined in a way identically to those of your own consciousness, then it would actually be sad, in every way that you are. In particular, if you replace bits of paper with "neurons", your argument should absolutely still hold. So the logical extension of this argument is that either YOU also aren't capable of being sad, or your consciousness does not originate in your brain. Both of those I think are pretty close to sufficiently absurd that we can accept them as givens.

Isn’t that just Searle’s Chinese Library?

Sadness along every other feeling is indeed a computation, its just that people don't like to face such bleak reality, same way we can easily understand why a machine can be turned off and never to be turned on again and yet most of us have issues understanding when the same thing happen to our brains, so we rather invent a lot of mythology around such event (death). Simplified example: If I perceive my father died (input) I get sad (output). Is a system of stimulus/rewards/pushinement that has worked pretty well for the evolution of our species, is extremely complex but that doesn't make it not one.

And we know that the animals behave "sad" when somebody close to them dies or just goes away.

Emotions aren't anything special to humans. Other behaviors too. The "being special" is what humans just want to believe. Most of those promoting "consciousness" as "something special, that can't be explained by physical interactions of electrons, atoms and molecules or above that by biological interactions of cells etc, are simply in some way still hanging on their religious beliefs and projecting them there. The rest just have a new book or whatever else to sell.

This argument begs the question and as such "Yes it would be sad" is a sufficient answer. Why not? We've failed to define "actually sad" in any interesting way.

Emotion is an especially odd angle as we can so easily manipulate it with chemicals. Why would that be relevant to consciousness?

>Now take this to the extreme: replace all of these neurons with empty shells and route the computation to a billion people in the world to perform on pen and paper. Is it still conscious? If so, what if you removed the brain altogether and just performed the computations directly on paper, is the paper conscious?

Yes, absolutely consciousness would exist in all of those scenarios — but you couldn’t claim that any one “thing” such as the paper itself, would “be” conscious. The notion of a separate self is still fundamentally an abstraction placed on top of billions of neurons interlinked with a musculoskeletal system, through which quadrillions of new atoms are constantly cycling through. On a long enough scale you only have the psychological “gestalt“ of a human.

Even the sensation of “self” is still ultimately just another internal cognitive model, because ultimately everything else we perceive is also just an internal model of reality.

In your example, that internal “self” model would continue to exist within the abstract computationally-mediated consciousness-generating process.

Each conscious process exists in its own fully separate ontological space—because it is inherently subjective.

Without wanting to cause too much controversy, this depends how broad your experience of consciousness is.

That "internal model of reality" is capable of some very interesting things if you're prepared to read past the first few chapter or two of the manual.

Looking around at other living creatures it looks as if consciousness is a gradient. Several factors could move you up and down the gradient (to simplify), and I'd imagine that the speed of data transmission would be one of those. So as far as the thought experiment goes, I'd imagine the pen and paper method of transmission would be slow enough to knock the "brain" off the consciousness scale completely.

If you have a slow simulation process you can just run it in slowed-down simulation time.

But there might be a link between the time-density of the simulation and the emergence of consciousness.

Ah, the classic "argument from incredulity by implausible substrate".

Step 1: point out that any number of wacky scenarios can be isomorphic to a human brain - a person in a room transcribing symbols, an unlikely cloud of dust in space

Step 2: Note that the fantastically implausible scenario you've constructed is fantastically implausible. Redirect the implausibility into the thing you want to claim is impossible.

Step 3: Wave hands, make argument about consciousness/AI

In case of paper it's not implausible at all. It's just ignoring a necessary part of the system - a human that executes the process on paper.

I don't think it's any more implausible (that this is a conscious proccess) than saying both personalities of people with split personality are conscious.

This is basically a convoluted restatement of Searle's Chinese Room argument. To me yes, since I think conciseness is an emergent property of the matter in my brain, and the behavior of matter is in principle computable, I think this is the logical consequence.

I don't think that this is practically possible though. Even a single brain cell is fantastically complicated and attempts to simply model neuronal behaviour from gross chemical simulations have failed. We can't even reliable model the behaviour of a single neuron yet. That's biological systems for you they are insanely complex with mind bendingly subtle feedback loops and cross-talk effects between seemingly unrelated systems that produce the macroscopic behaviour. That doesn't change the basic principle though, yes I think we're physical systems.

I guess if you replace a neuron with pen&paper calculation you have to keep the timing right. As in: A song played on a piano is just the various strings moving back and forth. So you could replace the strings with people walking forward and backward. But for it to be a melody: 1. the people have to move really fast 2. all movements must be synchronized sufficently 3. there must exist a point in space time at which all those vibrations come together, ie. all that movement must add up to the final frequency of the song.

So if all people in the world try to create consciousness by calculating on paper their would have to carefully watch their timing in each step.

What do you get if a lot of people synchronize their actions? Like in synchronized swimming? dancing? a concert? maybe some primitive form of consciousness?

/edit: Try combine this synchronization with a feedback loop, ie. the swimmers or piano player not following previously determined steps of a sequence but determining the next step based on the input from other systems around them and their own output.

>If so, what if you removed the brain altogether and just performed the computations directly on paper, is the paper conscious? The paper would definiy still be intelligent, it would possibly come up with interesting inventions, claim that it was sad or happy etc, but would it actually be sad?

One might argue that it would be sad, the way a human is sad, but lack the ability to immediately experience it in real time, while we can.

Isn't "real time" a concept that is somewhat biased by the frame of reference, i.e. it would just have a different concept of "real time" and consider our "real time" to very quick?

I very much doubt that a paper simulation could keep up with real-world events, and would immediately fall irrecoverably behind them, but if that simulated mind existed in a corresponding simulated world, then that world's pace would define realtime as experienced by the simulated mind.

In that context it’s the rules that define how you update the paper that give the symbols on the paper meaning. And in that context sad is both the symbols and system that gives those symbols meaning.

It’s the same situation as people playing a game of poker or go fish. The cards in their hands might be identical, but the game is what gives those cards meaning.

You'd like "Permutation City" by Greg Egan. It's about exactly this question, and he goes with it even further :)

For one example he considers information processing to be consciousnes, then asks about random patterns in cosmic dust that by chance encode another steps in a conscious process every 3 million years. Is that a conscious thing? What if it is?

It goes very abstract very fast, and I disagree with some conclusions, but I still recommend it.

> is the paper conscious?

Not paper, but the person implemented in such a way - yes. You don't say "my proteins are conscious", you say "I am conscious". The substrate only needs to ensure the process works correctly and the inputs and outputs are matched.

In your example it's like virtualization in IT - you can run a game on the computer, or run a virtual machine on the computer and run the game on that virtual machine. The game is running either way, it even runs on these same transistors, but it's a different process.

Besides, I think consciousness is very handweavy, and if I had to bet I would bet: - consciousness isn't a thing, it's just a bad abstraction (70%) - consciousness is emergent in a fully physical universe (25%) - consciousness is fundamental, and part of the laws of universe (5%)

As I mentioned elsewhere, your last step would not actually work in real time, so maybe that part of its implausibility.

As in Searle's "Chinese Room" (and also, though to rather different effect, in Jackson's "Mary the Neuroscientist"), you have proposed a wildly infeasible thought experiment, and then decided dismiss your rational conclusions about the outcome because... because it does not accord with your intuitions about normal situations? because it is infeasible? This simply is not the way to get to the truth of any matter.

Because unlike everything else in the body, we don’t know what sort of material explanation would possibly work. Instead assertions are made that subjectivity is identical to certain kinds of brain activity. What is needed as an explanation for how brain activity is conscious. How does neural activity result in the feeling of pain or experience of seeing a color?

> How does neural activity result in the feeling of pain

We can design synthetic chemical painkillers that work, suggesting at for this we actually know what’s going on. You can actually look up how these neurotransmitters etc work.

If you somehow want to suggest something else is going on you really need to support that argument extensively.

> You can actually look up how these neurotransmitters etc work.

Which doesn't explain having an experience of pain, only that certain neurotransmitters are involved which can be suppressed.

Sure it does, the signal is the feeling. You are your body, so when your body sends a signal that is you feeling something.

But how come people have different pain thresholds then?

I did a bunch of research on pain some years ago, and in order to "standardize" it, one would apply a treatment when the person said pain was greater than 7 out of 10 on a scale.

If pain were measurable "objectively", this wouldn't be necessary.

People are not uniform templates using identical DNA and histories. Our bodies are different which means our subjective experiences are different. That’s both consistent and expected.

As to measurement, we can design and build modern CPU’s but can’t inspect them in full speed operation. Similarly, mechanics often need to take something apart to test it. Living organisms present a more complex challenge, but the similarities are obvious.

Yes, but we cannot measure pain "objectively" at all. The equivalent would be if we could build a CPU, but could only determine how fast it could process by asking it to rate its processing power on a scale of 0 to 10.

Nothing is stopping a more technologically capable civilization from directly measuring our pain. You’re describing a purely technical limitation.

We have plenty of ways of directly measuring people‘s brains and CPU’s energy use etc. The issue is speed and resolution which is hardly a philosophical limitation.

> But how come people have different pain thresholds then?

Different signal sources and also differently configured/trained systems receiving and interpreting those signals.

But practically, it makes more sense to use the instrument already evolved and trained by their own experience to measure their pain: the person themselves.

> Different signal sources and also differently configured/trained systems receiving and interpreting those signals.

I think that this is a little too general a statement, given that it could be used to describe almost any human variability.

> Because unlike everything else in the body, we don’t know what sort of material explanation would possibly work.

I think that the closest we have to an explanation for something capable of thinking are CPUs and we're just not sure what the right software is to create independent thought.

This might imply that traits like universal computation are necessary (but not sufficient) for thought.

Of course there is no evidence available to you that anything is special about my consciousness. And vice versa.

Are you a solipsist or not? If you don't believe in the consciousness of other beings, I can hardly argue with that on rational grounds, but you can't disbelieve in your own consciousness.

I think the point is not that you disbelieve in your own consciousness but rather that you misinterpret what it is. It must exist: cognito, ergo sum, but this does not imply that you understand its nature.

Exactly, it's a mystery.

It's also basis for humanism, since if nobody has consciousness, there's no reason to keep humans around.

So it is existential, untestable and unprovable.

"a cardiac surgeon can still know far more about your heart than you do"

You're conflating the knowledge surgeons have with the subjective experience of a pounding heart. Surgeons study the heart qua plumbing, not as object of awareness related to primal feelings. Noting that physical changes in the brain cause changes in consciousness does not explain subjective experience.

He seems to have caught the issue dead on, since you seem to be arguing you know more about pumping blood than the surgeons based on a feeling - exactly what the people arguing consciousness can't arise from biology are doing. They feel consciousness is some big important thing that can't be produced conventionally and then resort to the equivalent of religions to explain it - feelings arrived at with no scientific method that don't really hold any weight.

"you seem to be arguing you know more about pumping blood than the surgeons based on a feeling"

No, I did not argue that at all. I said that there are multiple types of knowledge regarding hearts. Surgeons know one type in depth. The subjective experience of a racing heart is different from physical knowledge about how the heart pumps, and even if you know in depth the chemical/physical reactions that lead to and accompany a pounding heart, you do not thereby know that subjective experience.

By your definition the subjective experience says nothing about how a heart actually works.

Except that not quite true, the subjective experience tells you there are pulses which increase in frequency under stress etc. It’s just the surgeon understands what that means where subjective experience is less useful. Further, the subjective experience fails to separate the causes of a fast heart rate with the response of a fast heart rate.

This should suggest that the experience of consciousness is of minimal value when trying to understand it.

"By your definition the subjective experience says nothing about how a heart actually works."

No, that's a reductionist assumption. When I say "subjective experience" I am explicitly not indicating its quantifiable aspects. I am indicating what Nagel talks about in "What It Is Like to Be a Bat," the what-it-is-likeness, the phenomenological aspect. Not the increased pulse, not the hormones being released, not the neurons firing in the brain that correspond to a feeling of anger--the anger itself. An idealized observer enumerating all of anger's physical correlates does not give that observer the experience of that anger as it is felt by the person experiencing it. There is a difference between understanding something and living something.

Only responding to a tiny piece of your very good comment.

In the Maier/Rechtin book on system design, they spend a few pages talking about the word “Emergent”, and how “weak” it is. They says that we use the word to denote things that are almost certainly explainable, but for which we do not yet understand the mechanism of its behavior.

They use (iirc) the example of a black box system that produces a whistle noise at seemingly random moments, which observers interpret as “emergent” when reality reveals itself on closer inspection. Their point is essentially that the description “emergent” is a cop-out, and that so far nothing we’ve described as “emergent” has actually turned out to be inexplicable.

I often hear smart people saying things like “consciousness is just an emergent feature of our brains”, in a way that seems to imply “that’s that”, when in fact it’s not an explanation but rather an admission of ignorance. Admitting ignorance is fine, but ignorance is not an explanation.

I personally don’t think we’ll understand the mechanisms that underpin consciousness within our lifetimes, but I certainly don’t think consciousness is inexplicable.

"Emergence" can certainly be abused in this way, but it is nevertheless a real, useful and important concept.

Take Darwin's theory of evolution, for example. Evolution is an emergent theory in that a purely reductivist approach to biology will not find a particle or field of evolution. This does not mean either that biology is inexplicable by physics, or that evolution is pseudo-science.

In your example, I don't think that "Evolution" matches the popular definition of "emergent properties" (i.e. properties that lack concrete explanations).

Evolution is natural selection (a well-understood mechanism) applied over time. The divergence of species, families, and genuses "emerges" over time, but I think we're agreeing that this is a different use of the word.

To be clear, I'm not saying that words like "emergence" have no use, I'm just recommending that we be careful using them, lest we devolve to being satisfied by increasingly hand-wavy ideas.

In situations where we're aware of our own ignorance, we should own up to the ignorance directly, instead of describing the systems as having "emergent" (i.e. tautological) properties.

> In your example, I don't think that "Evolution" matches the popular definition of "emergent properties" (i.e. properties that lack concrete explanations).

I do not know whether that is the popular definition, but whether it is or not, it is simply wrong.

My usage, with respect to evolution, is the usage that counts in discussions of consciousness. It is the valid response to naive misconceptions about what materialism entails, such as in Searle's dismissal of the systems reply to his Chinese Room argument.

I think we've been talking past each other a bit. If I'm continuing down that path, I'm sorry...

My point in responding to your point on evolution, is that we concretely understand the mechanisms by which evolution emerges. If you cut out the natural selection, you have no evolution.

We can use the word "emergent" to convey that the web of actions that leads to evolution is highly complex. But, for evolution, one can still isolate those actions into relatively orthogonal, self-contained narratives (mate selection, disease, symbiotic relations), that come together with the effect of evolution.

In contrast, we do not concretely understand the mechanisms by which consciousness emerges. We do not have a good methodology to break down consciousness into its constituent parts, such that you can union them iteratively and end up with consciousness.

To be clear, I'm not a materialist. I don't think you need to be a materialist to think that the phrase "emergent behavior" (as commonly applied to consciousness) is a cop-out explanation, that we shouldn't be satisfied with.

I do think it's possible to construct those orthogonal narratives, and union them to get consciousness. Again, I don't think it'll be accomplished in our lifetimes. I don't think we have the language to describe it currently. But I do think it's possible, and in the meantime, it's misleading to use phrases like "emergent behavior" to smuggle out that possibility.

To be clear, I'm not taking issue with your use of the word emergent (which is more nuanced), but with the OP's. And it's not the OP's fault, because I've seen it used identically elsewhere.

You are right; I see what you mean, and I agree that just saying, as some people do, that consciousness is an emergent phenomenon, does not explain it, and does not avoid the fact that there is a big hole in our understanding here.

The answer, of course, is that there is no consciousness: if you try to precisely define what consciousness is, you’ll end up with increasingly absurd “you know when you see it” kinds of definitions. Objectively there is no any “special” consciousness, it’s just information processing systems, simple or complex.

>The answer, of course, is that there is no consciousness: if you try to precisely define what consciousness is, you’ll end up with increasingly absurd “you know when you see it” kinds of definition

There's nothing absurd about “you know when you see it”. Those are just things we don't have (or can't have, or don't yet have) good definitions for, but do exist.

"Family resemblance" [1] classes of things are often like that.

"What is a game", is a classical example. Any definition you can give can be violated, including "having a goal", or "not having real impact" etc. And yet we know a game when we see one. And there's lots of other such distinctions.

[1] https://en.wikipedia.org/wiki/Family_resemblance

While such definitions are good for casual conversation, they are not good for scientific discussion. Or you end up with nonsense like “my doorknob has consciousness“, well, lets define what consciousness is first.

With “consciousness” in particular having good definition became important since there are a lot of discussions about “is AI conscious?” which might in the future dictate policy decisions.

Let me try an easy definition: consciousness is that which makes you get out of bed in the morning and get something to eat. Life is a series of experiences and choices and something needs to make those choices in such a way as to continue life.

If it weren't for consciousness your body would not survive, or grow to adult size, or reproduce. Of course, 'get something to eat' sounds simpler than it is in reality - you would need to be able to see, walk, grasp, understand the environment and objects at your disposal, shop, cook, earn money, be a functioning member of society - the whole bag.

But all organisms “get out of bed and get something to eat”, yet we don’t pretend that bacteria have consciousness. If you go down this road you’ll have to decide if apes, dogs, crocodiles, sharks and ants have consciousness and make arbitrary cut-off somewhere along this path.

> yet we don’t pretend that bacteria have consciousness

In my view, life is a process that is governed on the grand scale by evolution (species adapting to the environment) and on individual scale by consciousness (individuals adaptating to environment).

Bacteria do both, like humans. Some AI agents also do both - evolution and learning, such as AlphaGo, just that its environment is a Go board.

Is consciousness is not emergent from something - not "fully explainable in some mechanistic way" - what type of explanation for it would satisfy you? If you're not willing to brook the idea of breaking it into smaller components, then you've abdicated understanding it all - it's just an ineffable feature of the universe we'll have to live with, some shadow of the spirit world that has zero effect on any physics calculation. That seems weird to me, and lazy to boot.

I don't know, that's the problem :)

Btw, I would imagine that our experience of consciousness is emergent from something else, but the question is what that something else is. Is it nothing more than mass, charge etc. or is it something else?

> Is it nothing more than mass, charge etc. or is it something else?

It's more like chairs, beds, phones, cars and people. Consciousness is emergent from the environment. We learn to interpret visual stimulus by seeing, we learn to interpret sound by hearing, we learn how to relate to the world by being in the world and having consequences to our actions. Our brains grow on the rich sensations they receive from the body, world and the other people. Looking for the mystery of consciousness inside the brain is a mistake, we should look at the whole game instead. The brain is just doing the learning and acting part, but the game is much more.

Big corporations have a mind, but they aren't conscious in the way we are conscious. Lot more minds exist in the world than consciousness.

this is actually one of the problems I have with the theory that everything is conscious. The electron, like the company, is actually an abstraction and part of a model. We don't even really know if an electron is 'a thing in itself'. Maybe everything is just strings or quarks, or some other smaller particle or just the manifestation of some field (sorry I'm not a physicist in case I'm butchering the analogy).

Even every individual human could be said to be some sort of zoo of 'conciousnesses'. Panspychism posits that consciousness is essentially smeered across the universe like a sort of paste, and if consciousness applies down the ladder I don't see why it doesn't apply upwards.

Everything is conscious and nothing is conscious, are equally valid as hypothesis.

Who is it, that is trying to establish a definitive answer to something unanswerable?

I think what you're lacking, if I recall correctly about panpsychism, is that only self-organized systems are conscious - not everything.

A computer, birds nest, a gun, a space ship: such things aren't self organized, therefore not conscious.

A cluster of cells, a planet, a star system, a galaxy, they are self organized therefore, permeated by consciousness.

Worth considering, I believe, is that many people do not have an inner voice.

For those of us who have conversations with ourselves, who can explore our consciousness throug dialog, it is easy to assume that all humans can do this.

Many cannot.

Many have no internal voice.

I find many folks are asking "where does my internal voice come from," and yet many conscious individuals lack it. "I think, therefor I am" has a different meaning when that question cannot be asked through internal monologue.

For myself, the concept that consciousness is an emergent and mechanical outcome became easier to accept after considering consciousness without monologue.

I absolutely agree, to the point of considering almost any materialistic (as well as theological) debate about consciousnesses pointless.

Just for fun, however: what if consciousness actually is just a manifestation of mass? (I've in fact heard a hypotheses it's a manifestation of gravity) Just trying to imagine consciousness of a mountain exceeding that of my own by orders of magnitude is an interesting experience. I imply that a consciousnesses of a different order does not necessarily has a function of acting or communicating in a way we could consider conscious.

Saying consciousness is emergent is definitely weird, but the problem is, so are all the other possible explanations. If you're trying to avoid supernatural explanations then emergence is the best of a baffling bad bunch IMHO.

(Then I would back this up a bit with "once you start investigating it, a lot of aspects of consciousness are quite different to what people would intuitively expect" and that makes it all a bit more feasible. E.g. saccades and the fovea are fascinating and counterintuitive, what else are we getting wrong?)

Is it though? Until we have more information, all these explanations are equally unsubstantiated. If you're speculating without the data and I sight needed to actually understand it, then the proposal isn't any better because it feels less supernatural. The actual explanation will probably seem really weird and supernatural to us now since we dont have the data and experiments to understand it. Today, quantum tunneling is still weird, still hard to wrap your mind around, but it's a phenomenon that we understand well enough to work with. We try to avoid it when making a CPU, try to make it happen when we use a scanning tunneling microscope. But 250 years ago trying to explain how a scanning tunneling microscope that just appeared into the world would have seemed impossible without invoking some supernatural sounding explanation. We have consciousness, we dont know where it came from yet, maybe there's a consciousness field we'll find when we figure out how to unite gravity and quantum mechanics, maybe theres an emergent property we'll understand once we can map a brain's functions and do some data science on it, maybe there's some explanation that we cant even imagine yet, but until there's data none are less supernatural than the others

> If you're trying to avoid supernatural explanations then emergence is the best of a baffling bad bunch IMHO.

Emergence is a general word, too general. I think what we need here is just self-replication. Self replication leads to competition, which given enough time leads to the evolution of perception and ultimately reasoning.

That's traditionally called a soul.

I see absolutely no evidence or suggestion that consciousness is anything else other than the summation of changes that happened in our brain in combination with the input devices we naturally have. You have decades worth of changes that occurred to get you to the point now and it's very clear and easy to see how much _less_ conscious you were when you were younger.

I agree that consciousness is correlated with this, but having a lot of experiences over years does not explain to me why I feel things.

Similarly I can definitely see how there would be a mechanistic explanation for intelligence (basically just build a machine that acts intelligently, perhaps even claiming that it's conscious), but there's a distinction between appearing conscious and actually feeling something.

Good luck getting to the roots of all your feelings when they can be based on any number of inputs and changes over who knows how long of a time.

The subjective feeling is just a brain state. The ‘why’ is due to whatever circumstances resulted in that brain state.

But one can definitely ask how something seemingly as complex as a "brain state" can be directly experienced as phenomenology. How is it that we can so easily describe some things about our brain states, that we'd never realize by physically looking at what brains do and how they behave.

That's the "hard problem" of conciousness stated simply, and the answer to it would seemingly have to involve some quite direct access to some part of brain state physics - meaning that the physical properties involved are quite "basic" in some sense.

> How is it that we can so easily describe some things about our brain states, that we'd never realize by physically looking at what brains do and how they behave.

As a software developer, this should be simple. Programs have their own state that you can understand without knowing the exact formulations of electrons in the processor that ultimately make up that state.

X = 5 sure but good luck figuring that out looking at the motherboard of a running computer.

So why are some brain states conscious and not others?

There is no reason to expect that the self-aware conscious mind has full access to everything that is going on in the brain. In fact, from both an evolutionary and systems perspective, that would be extremely surprising.

It would be impossible for the organism to operate effectively if they were conscious of every single thing going on in their brain at all times. In fact many things you aren't normally conscious of, if you focus, you can become conscious of for the moment.

A sleepwalker is not conscious, yet you can converse with them, ask them questions, get answers, you can tell them to do things.

In every external way they appear conscious, yet they are not. They are not aware of their own existence.

So consciousness is clearly more complicated than just "summation of changes that happened in our brain", there is something extra there, that we don't understand.

Perhaps they're simply not forming memories of their own existence. Can you ask a particular sleepwalker their own name and get answer? That might imply awareness of their own existence.

It seems most people want to make consciousness out to be more complicated in some way.

Ever heard of amnesia? Works the same way.

> but claiming that it is fully explainable in some mechanistic way sounds suspect to me.

What's the alternative? If it's not "fully explainable in some mechanistic way" then you need a mystical component to the explanation?

That is itself suspect, exceedingly suspect, to many people.

The alternative is that current metaphors are wrong and new metaphors are needed.

Will those new metaphors be "some mechanistic way" or not?

Proposing new mechanisms is all fine and well, But mixing in mysticism, and you lose respect from all of science.

As much as I'd like to believe that consciousness is unique, I don't think we have to look that far to show otherwise. There's an abundance of mental disorders that affect consciousness. Take for example something as (relatively) mundane as ADHD. Anyone with it that has a reasonable amount of insight will tell you that they are almost entirely driven by attention, which is the crux of ADHD.

I'd argue that "consciousness" is the illusion that's created from attention, which you're ultimately a slave to. While I don't have empirical evidence, considering how difficult it would be, I believe that attention is the heart of what we consider consciousness. What your attention focuses on is built on nature and nurture. Your attention is grabbed by what you're "interested" in and what you deem is important. It guides your thoughts, your perception, your internal representations, your attitude, your decisions, etc.

The illusion is that you have control over your attention. Again, anyone with ADHD and insight will conclude that you don't. If your mind will is a computer, then attention is the cursor that guides interactions with it. However, we're not in "control" of it since basically what we know of as the self is the mouse cursor of the mind. I don't think we move the mouse; I think that's the illusion.

Me typing this is simply a product of my attention and it leading me to the conclusion that I should type this because this topic and the representations I've constructed internally are significant to me as a result of my education, life experiences, genetics, etc.

At least, that's my opinion. I definitely don't want to come off as an authority on consciousness. I don't think anyone is (yet). Despite my view, I do believe we can uncover the mystery of consciousness from mere perseverance, even if consciousness is emergent.

---

Someone has already proposed a thought experiment, but I propose an alternative. Imagine all humans (and living creatures in general) as cells and neurons of an even greater being. What if what we understand as the universe is actually just the composition of a larger mind? As cells, would we comprehend or know that? Or would we assume we're independent beings? The infrastructure we build, the roads connecting clumps of societies, the internet connecting all humans together, etc.

Who is to say that we're not simply building more complicated versions of what we would consider synapses between neurons, nervous systems, circulatory systems, etc? Perhaps the progression of humanity, and the universe at large, is actually just a small organ during the gestation or development phase of a much larger being and we are completely unaware?

One notion that comes to mind is fractals. We see it everywhere in nature. What if consciousness does emerge from complexity and what we consider reality is just a single resolution of a much more complex and larger fractal?

For anyone interested in this kind of ponderings I strongly recommend Consciousness Explained by Daniel Dennett.

It may be the best book I've ever read.

The difference is that charge, spin etc have their own effects/manifestations, whereas consciousness does not cause anything.

You're not even wrong, it's logically impossible to derive temperature from velocity either because temperature is defined as a fundamental attribute of matter. You can only quantitatively connect them, and such connection is seen as explanation in mechanistic way.

Reply on news.ycombinator.com