Hacker News Reader

Comment by thriftwy | Hacker News Reader

Then what is consciousness?

People keep telling me that consciousness is just some illusion which happens inside brain neural network.

However, in case of mathematical universe hypothesis, nothing really happens. If you have a mathematical universe with seed 42, which you spend a lot of effort actually simulating, it is not in any way different from a mathematical universe with seed 43, which you don't. Neural networks have exactly the same zero amount of consciousness in both cases.

Why do we feel that we exist? Who does the feeling part?

View on HN

> People keep telling me that consciousness is just some illusion

In my view, consciousness is the only thing that can not be an illusion. If it's an illusion, who is being fooled? Consciousness isn't the illusion, it's the audience.

> If you have a mathematical universe with seed 42, which you spend a lot of effort actually simulating, it is not in any way different from a mathematical universe with seed 43

Yes, this is the key point of the mathematical universe (for me). No simulation has to occur.

There is no way to tell, from within the mathematical universe, whether or not anyone is running the simulation. The simulation helps someone outside find out what happens in the universe, but the stuff that happens in the universe happens whether you simulate it or not.

All of the possible mathematical universes "exist" and are equally valid.

"Existence" and "the passage of time" are things that only make sense within the mathematical universe. There is nothing outside.

Why do I have this long term feeling that I exist in a specific mathematical universe and is absent from all others, or at least represent an unique path distinct from all other paths which I am not?

Who does the feeling part? Who is the audience? When you run simulation there's at least some audience, but in case of math universe there is only archeology, by definition. Mathematical object has no original, only replicas.

I think for humans our memory is why we only feel connected to a particular causal timeline of consciousness. For myself I am conscious of the present moment and of my memories and of my imagination.

For us to be conscious of our experiences in other universes or timelines we would need a way to know what those experiences are like. In our universe memory is a cheap way of recording what the causal past was like vs. needing to compute/simulate what happens in other timelines with enough accuracy to be as fully conscious of it as we are of our memories.

> In my view, consciousness is the only thing that can not be an illusion. If it's an illusion, who is being fooled? Consciousness isn't the illusion, it's the audience.

I don't why people keep being taken by this silly argument. You've merely assumed that "illusion" requires a subject. Here's a definition of illusion that doesn't require a subject: a perception that entails a false conclusion if taken at face value. There, now "consciousness is an illusion" is simply, "consciousness is a perception from which we falsely infer the existence of subjective qualities".

Subjective qualities like color and pain? What about dreams and imagination? Illusions are misleading conscious experiences, usually perceptual. That's why it doesn't make much sense to say consciousness is itself an illusion.

> Subjective qualities like color and pain?

Colour is perceived. The "what is it like" to see colour is the qualitative illusion. Ditto for pain.

> What about dreams and imagination?

What about them? Does Midjourney plausibly imagine the scenes it generates?

> Illusions are misleading conscious experiences, usually perceptual

Illusions are misleading perceptions. There's no reason to throw "conscious" in there at all, it just muddies an otherwise simple concept.

> Who does the feeling part?

An even more intriguing question is if there are meta-consciousnesses on humans.

Let's assume humans have consciousness, and we are all just a bunch of cells.

Does a cell feel conscious? Maybe, maybe not. A bacteria doing gradient descent perhaps doesn't make that many decisions. But perhaps this hunt has something similar to our hormones, and they can feel happiness.

An arm? A brain? The neurons of just an eye? They can't really affect their survivability on their own, so perhaps there was no evolutionary advantage in being conscious.

For something with arms, feet and brain and two eyes? They can compare the world now, to what used to be, and reason about what should be. This awareness of time and the space I effect, I'd argue is a huge part of consciousness.

But what about some group of people, or the species as a whole? I'm fairly sure my cells don't understand that together they are "conscious." The same way, how could I tell if some meta-organism that includes me is considering itself conscious? Surely, if I'm capable of consciousness, anything that includes me would inherit that property?

If consciousness is transitive, that would be a huge result. Then the "who" is no longer of relevance, because it's all relative: a single consciousness could be a meta-part of multiple other consciousnesses at the same time.

And then we go all the way up to galaxy clusters. Are they conscious? I'm part of one. The slowness of information travel alone would make it incomprehensible for me to even observe it. If they are conscious, I would die long before I could even communicate the question.

I think this time scale problem will limit our understanding of consciousness for a long time. Even just looking at groups of humans, how would we recognize if they feel themselves? We can study biological systems on our scale, but anything beyond that... Well, it just sucks. Because I'd really, really, love to know the answer.

If you were an alien and showed up on earth, then hurt a human, you'd likely think we work at least like a semi-collective consciousness. Much like a cell that is injured starts chemically communicating problems, we'd start communicating at a meta organism against the new threat.

This is one thing that's always given me the 'concern of the unknown' when it comes to AI. The potential meta consciousness abilities of AI could vastly exceeded human capabilities.

Good point, though the reaction doesn't necessarily mean consciousness is required. It could be akin to a simple spasm.

> meta consciousness abilities of AI

Yeah. The more we create order out of chaos, the more likely something incomprehensible emerges.

> People keep telling me that consciousness is just some illusion which happens inside brain neural network.

Ask those people what is experiencing the illusion, and whether we can program a computer to have conscious experience using the same illusion generation.

It's like answering the question of "what created the universe" by saying "a god did it." It just pushes the question back.

If something can experience an illusion - i.e. to use the philosophical language, "there is something it is like" to have that illusion - then it has consciousness, and the illusion has nothing to do with it.

At high doses of psilocybin you experience the loss of ego. It's a weird feeling, where "you" are not there anymore. There is no "I". So it's definitely possible to be conscious but feel like "you" or whatever the idea you have of a self just isn't there anymore.

Beyond the ego, what is this conscious stuff providing awareness qualia? It seems we must assert one of these options:

a) consciousness is an emergent system of matter (materialism).

b) matter contains elements of consciousness (panpsychism).

c) matter/energy is consciousness with varying levels of complexity (monism).

Your option (c) is idealistic monism specifically, since physical monism corresponds to materialism.

And then there's:

d) matter, energy, and consciousness all arise from or are different aspects of some single "substance" (neutral monism)

e) consciousness and matter are "radically different kinds of things" (dualism)

f) the question is unanswerable and/or incomprehensible to us (mysterianism)

Thank you for that catch, yes, idealistic monism. Oh, I should have added E for as I already mentioned monism. D & F are also interesting positions.

Neutral-monism sounds like an hegelian dialectic [the gap]:

Many neutral monists hold that the terms “physical” and “mental” apply only to groups of neutral entities, not individual ones, "The stuff of which the world of our experience is composed is, in my belief, neither mind nor matter, but something more primitive than either. Both mind and matter seem to be composite, and the stuff of which they are compounded lies in a sense between the two, in a sense above them both, like a common ancestor". (Russell 1921: 10–11)

Is c really a workable hypothesis? It doesn't seem it could account for most of what we observe, at least not if we don't start believing in new sun gods or something like that.

Ultimately, if the world has a constant shape that looks the same to most people, it resembles what we call matter more than what we call consciousness.

Who knows?

https://en.wikipedia.org/wiki/Hard_problem_of_consciousness provides some interesting reading though.

Maybe we have it backwards, and consciousness is the underlying firmament of reality and the material world emerges from that (if it even exists, of course).

> Maybe we have it backwards, and consciousness is the underlying firmament of reality and the material world emerges from that (if it even exists, of course).

This is unsupportable, even in principle, and reduces very quickly to solipsism. There's no way to account for an objective world with this (bad) philosophy.

Not if you can think of some predictions from that theory that would later be proven true.

> reduces very quickly to solipsism

Not necessarily. Solipsism generally is the idea that only one's own mind exists (or at least, it's the only mind whose existence one can supposedly be certain of.) But the gp description seems to encompass something like idealistic monism, where consciousness is fundamental, but it's not necessarily that of one person's mind. This is not equivalent to solipsism in general.

I wasn't even trying to account for an objective world so much as to consider one possible explanation for the subjective experience of being. Though it is, admittedly, a bit of a cop out ("consciousness exists because it is existence").

I don't think it reduces to solipsism, though. I don't think anything I said precludes the existence of multiple minds?

In the mathematical universe consciousness is equivalent to a proper class (the sets that feel like something from the inside) and the intersection of that class with the possible states of our physical universe yields the possible consciousnesses in our universe.

How are some sets conscious? Probably some encoded self-reference or other self-similarity. E.g. there are infinitely many integers that under Gödel numbering refer to logical statements about the same integer. If anything, that should feel like some sense of self with reference to the choice of numbering and logic. Human consciousness is possibly many layers of parallel self-references that yield additional and comparative qualia. There are almost certainly other mechanisms of achieving consciousness as well.

But who does that achieving of consciousness? A random seed written on a piece of paper?

The "who" is the set that is conscious. Just like a prime number is prime without having any external force make it so or requiring anything to embody the primeness. Primality is in the nature of the natural numbers. The claim is that consciousness is in the nature of sets. The latter does not yet have experimental evidence. It seems as reasonable to me as other possibilities like electrons are conscious or quantum fields are conscious mostly because both of those are well-described by mathematics.

Why do I feel like I am living in time if I am actually not in a simulation, not a Boltzmann brain, but just a random seed that is not written on a piece of paper anywhere? What is the meaning of all this, then?

You are reiterating cheezy stuff about "nature" and quantum fields, ignoring the random seed part.

You are your brain.

Your mind is a part of your brain; a part that combines a spatial-temporal sense with a future-oriented simulating function.

Your brain is modelable – not only is it "a structure" in an abstract sense, it is an ordered physical structure made of molecules, and, both in whole and in part, it is reducible to a mathematical structure of highly limited complexity. "Qualia" consist of your mind's representations of external things, immediately derived from sensory input, wedded to that sense of being in space and time and the associations raised by that sense.

As your brain exists within a logical system -- a structure with a mathematical model which includes an evolution function -- your brain is also a logical system which evolves through time.

As for "nothing actually happens" -- the block universe hypothesis, which is strongly implied by relativity, fully accounts for this.

As an aside: I think that Tegmark is basically correct, but his hypothesis is not well defined and could use further explanation.

It must feel really, really good when everything is so perfectly crystal clear and easily explainable; when there are only answers, and no questions. How it is possible, is a mystery in itself.

Unfortunately these questions may not be answerable. For example, let's imagine a universe where all these questions are answerable,how it started and why it works can be explained. Then that universe has a vacuum meta instability and collapses into a new form. In the previous universe it was completely predictable that it would happen, but the singularity effect of the state change left a low entropy universe. Information was lost, and new physics takes over.

An observer in the new universe would see a blank piece of paper and would have no idea why this paper existed and why it was blank. And could know how we got there.

This universe would be a very unsatisfactory place for the inquisitive to exist.

That was irony.

Inability to model a basic worm with 100 neurons and fully mapped out connections defeats this simplistic worldview.

I wouldn't go that far.

I don't know if we can model any biological structure with complete fidelity. There's a lot that modern science doesn't know about the cell. The Golgi apparatus, for instance, is still a mystery; there are models, but nobody really knows how it works, and all of the models are apparently incomplete. You'll note that the Golgi apparatus is one of the very largest cell structures.

Our cognitive processes rest upon an extremely complex biological substrate. That we're not able to simulate a nematode's brain, however, absolutely does not imply that it's not simulatable in principle.

...I'd say that it's made of fermions, which in a coordinate space approximate bits of information. Those fermions are mostly arranged into atoms, which are grouped together in molecules. As, contra Penrose, quantum effects don't seem important for cognition, if you have an absolutely complete molecular picture of the brain and an evolution function, you'll have a model of the brain.

You need more than "mapped out connections," but it's eminently possible in principle, which is to say that it's not infinitely complex or somehow ineffable. It's just decades beyond us, by my estimate.

Before we can rule out quantum effects, we need to understand that nematode at least. It's quite possible that the discrete computing model is completely unfit for simulating intelligence - a possibility that many scientists are afraid to consider.

As I say, we might not be able to "understand" the nematode's brain before we're able to understand its constituent parts. We don't really understand the neuron itself. Nor do we understand glial cells and their function.

...Nor, indeed, do we fully understand _any_ cell. Case in point: Both neurons and glial cells feature Golgi apparatuses, which remain mysterious...

As we neither understand nor can simulate a brain made up of neurons and glia, what makes you think that our lack of a nematode simulation is a big deal? Seems to me that it is rather to be expected. But when our biological models improve, simulating a nematode brain -- or _any_ brain -- will not be beyond us.

Not saying I agree with a mathematical universe hypothesis, however I have given it some thought. I'd argue that any 'operator' is consciousness in that setting.

There's some quote out there by some famous mathematician or hpyscisist along the lines of "it appears the operations themselves act as if they were conscious"

Even then, how we think of operations symbolically is just an emanation of a universal that we can't see in that sense; but in another sense, we are also emanations of that thing too in the sense that we (humans) are extensions of measurement and operations.

Another problem (which Tegmark writes about) is math will have to change. Some of the math we study will have to be part of illusion - there is no “one” math currently, and how could something exist in MUH which isn’t computable. His vision pairs down math, much to the ire of many mathematicians. But it is not unviable, we don’t need much of math for physics.

I think the view that mathematics and logic are static is incorrect. Computers are mathematical as well.

I think viewing mathematics as the study of structure, and structures can be transformed into other structures via a mathematical procedure we know of as computation resolves the difficulties.

You cannot transform a mathematical structure. The only thing you can do is find another structure with linked properties. But both structures are immutable.

It is not possible to transform a sphere so that it turns into a cube and no spheres can be reasoned about anymore.

> You cannot transform a mathematical structure. The only thing you can do is find another structure with linked properties. But both structures are immutable.

I think you're conflating the structures that describes mathematical structure with the structure itself. Like I said, mathematics is the study of structure, so the syntactic elements you're used to interacting with are themselves structures that describe other structures. These syntactic descriptions have isomorphisms to the underlying structures they're describing, but they are not the structures themselves (map is not the territory, etc.).

Mathematics is larger than this immutable descriptive layer IMO.

Edit: as a more concrete example, consider how we model mutable data structures and stateful session protocols in programming languages. One way is to project these into an immutable trace, as an immutable description of the object's state changes, but that trace is not the underlying structure. The underlying structure itself is still mathematical, but mutable, eg. cellular automata.

The idea here is that deterministic computations neither exist nor don't; regardless of whether they are being simulated. A Life pattern does not xare whether you simulate it or not. It will not become conscious in any way after your decision to simulate it.

> A Life pattern does not xare whether you simulate it or not. It will not become conscious in any way after your decision to simulate it.

Where's the proof of this? The game of Life is Turing complete. We have every indication that human beings can be modelled by Turing machines.

Turing machine is a purely mathematical concept, it does not need you to run it. All possible runs of Turing machines exist at the same time and are immutable.

What makes us special?

We're not special, nor do I agree with your Platonic view of mathematics.

Then you arrive at questions such as "Why do we exist?", even "Why anything exists at all, what was the need for all of this?"

> Why do we feel that we exist? Who does the feeling part?

There’s the non-helpful answer “I think, therefore I am.” I don’t think there’s any good answer to these questions. We can think about ourselves, we can think about existence, so we feel we exist? Let us imagine a person can not receive any external stimuli, can not send any instruction to control the movement of body, [e.g. a brain with blood circulation to keep it alive, but no input/output], in awakened state, now what should they feel? [Such a horrible state!] It seems they should feel their existence only on the basis of the functioning of their brain.

consciousness is just a real-time continuous representation of time/space, i.e. of the parts of the universe in front of us (and some of it already within us, i.e. memory and imagination).

No, it's not. I feel that I exist. I can represent time/space in a computer program. Nothing would feel that it exists there.

Overall, I feel I'm trying to push against another philosophical maxim that I know: "If somebody tells you that consciousness does not exist, it's all just measurement and neuron signals - believe them"

What is that maxim supposed to mean?

My first interpretation is that it's saying that people who say that might be philosophical zombies, lacking consciousness.

| I can represent time/space in a computer program. Nothing would feel that it exists there.

That's why I said real-time and continuous. All digital systems, at the lowest level of computations are discrete, not continuous (i.e. analog). And the changes happening in real-time, i.e. the transistors values changing in processing units are very far removed from a representation of reality. What I mean by that is that if you are watching a video of a desert on a computer screen and took a snapshot of the transistor values in a GPU/CPU happening at the same time, they is no good mapping from one to the other (Since that one screenshot may be computed by thousands of computation cycles in the GPU/CPU).

Memory is the current spacetime location's fuzzy view when looking towards the past. Imagination is the fuzzy view when looking forwards to the future or with the many-worlds view, imagination is the looking out(?) towards multiple outcomes.

Consciousness is the step towards the next point in the many world multiverse.

Without the ability to be blind to the future, life as we know it wouldn't exist. Nothing conscious would take risks and so complexity couldn't evolve.

If you could glance to the future and see all of the laid out spacetime, there wouldn't be any interesting exploration of it.

So we have to be blinkered just to exist with this level of insight into the complexities of the universe.

As someone else observed, many computer & robotic systems have some sort of real-time continuous representation of time/space. Do you think they're conscious?

By continuous I also meant analog. No digital system is continuous at the lowest level of computation.

You are a body. Consciousness is your UI to the world that you generate and you use. Don't over-complicate this shit.

That's one viewpoint, but it's not "over-complicated" to consider others.

You are a consciousness. Your body is your UI to the world.

That reduces to solipsism or idealism, which have been explored extensively. Defenses of them are definitely over-complicated, which is why they aren't considered serious positions in philosophy anymore. Your questions are orthogonal to my answers.

> they aren't considered serious positions in philosophy anymore.

Statements like this say more about your biases than about anything happening in philosophy. As recently as 2019, Chalmers published "Idealism and the Mind-Body Problem" (https://philpapers.org/archive/CHAIAT-11.pdf), a survey of some current positions in idealism.

Are computers conscious? If not, why not, and can they be programmed to be conscious?

> Don't over-complicate this shit.

One big issue that makes this difficult is that we have no understanding of how it's possible for physical matter to possess consciousness. Absolutely none, just lots of conjecture with no evidence or well-developed theory to support it.

If you have a non-complicated answer to all this, you will be world-famous for solving the hard problem of consciousness (https://consc.net/papers/facing.html).

I don't need to answer any of those questions to state what I did as facts. They aren't grounded in the same evidence.

Ok, but in that case what you wrote has no relevance to the questions asked in the comment you responded to: "What is consciousness? ... Why do we feel that we exist? Who does the feeling part?"

All you're talking about is its role, and making an unsupported assertion about it at that.

Reply on news.ycombinator.com