Is consciousness everywhere?

March 15, 2021 9:27 AM

"Accounting for the nature of consciousness appears elusive, with many claiming that it cannot be defined at all, yet defining it is actually straightforward. Here goes: Consciousness is experience." Christof Koch, at MIT Press, discusses what Integrated Information Theory (IIT) has to say about consciousness: "Some level of experience can be found in all organisms, it says, including perhaps in Paramecium and other single-cell life forms. Indeed, according to IIT, which aims to precisely define both the quality and the quantity of any one conscious experience, experience may not even be restricted to biological entities but might extend to non-evolved physical systems previously assumed to be mindless — a pleasing and parsimonious conclusion about the makeup of the universe."
posted by beagle (112 comments total)

Thanks - this is an interesting piece. But it sure seems like "abduce" is being used as a ten dollar word for "guess," and many of the guesses in this piece seem unfalsifiable. In several places he substitutes cognition for consciousness (which perhaps is the core of Integrated Information Theory?) but then at others times acknowledges that they're different.

It's not really the author's fault, as coming up with an explanation for consciousness -- much less a way to detect and measure it -- seems in a category of difficulty so high that I cannot even imagine what a compelling explanation would look like, much less what it might say.
posted by zittrain at 9:40 AM on March 15, 2021 [3 favorites]

ctrl-f "animism"... Hmm, nope.

It's fun to imagine that everything is conscious, but like zittrain says above- unfalsifiable ideas aren't really the purview of science, are they? Still, fun read!
posted by BuddhaInABucket at 10:24 AM on March 15, 2021

Koch is an amazing theoretical/computational neuroscientist, so if I trust someone to approach this in a rigorous way, it's probably him. My uninformed bias (uninformed inasmuch as it's based on reading lots of his papers and chapters for academic work, and, more informed, probably the bias of a lot of his colleagues/reviewers/etc) is that this is outside the scope of science qua science, but that's his whole point, he thinks that it doesn't necessarily have to be.

I think his thing is more panpsychism than animism but maybe that's a distinction without a difference.
posted by supercres at 10:33 AM on March 15, 2021 [1 favorite]

Science is in a dire predicament. The only really important and interesting frontier left for it is the human mind. Yet none of the usual tools of science will open its fundamental nature to investigation. Having freed itself from philosophy and religion, science will not avail itself of their insights. Unable to advance and unwilling to back up, science is left stuck in the mud, with the whole of mankind stuck with it.
posted by No Robots at 10:58 AM on March 15, 2021 [5 favorites]

This is my jam. I strongly believe conscious is everywhere (or nowhere). I am not fundamentally different from the rest of the universe.
posted by hypnogogue at 11:00 AM on March 15, 2021 [10 favorites]

Bumble bees can even learn to use a tool after watching other bees use them.

And honeybees and hummingbirds can recognize faces -- the first I learned just now from reading just now, the second from eye to eye contact and listening.

As for consciousness is experience, yeah, well, if you say so. But I will add this aside:

Collectively and individually, this place is so so way way much smarter than it was when I first joined. I am in awe every day I open a thread. ...The mind reels, the intellect stands abashed indeed.
posted by y2karl at 11:16 AM on March 15, 2021 [1 favorite]

Science is in a dire predicament. The only really important and interesting frontier left for it is the human mind.

Nonsense, and self-absorbed. What an incredibly incurious approach to life.
posted by biogeo at 11:21 AM on March 15, 2021 [35 favorites]

Metafilter: Nonsense, and self-absorbed.
posted by Jonathan Livengood at 11:32 AM on March 15, 2021 [20 favorites]

I'd note that while Koch is a hard proponent of IIT, the idea has had heavy criticism from many other, similarly-prominent neuroscientists and philosophers, and I'm not sure IIT proponents have adequately addressed those criticisms or even adequately engaged with them. Interested folks may want to take a look at the "Reception" section of the Wikipedia entry for IIT for both endorsements and criticisms of the idea.
posted by biogeo at 11:41 AM on March 15, 2021 [12 favorites]

Echoes of Conway & Kochen's take that elementary particles may have as much free will as human beings?
posted by weston at 11:44 AM on March 15, 2021 [1 favorite]

I'm really not up to speed with IIT but in general over the years I've grown more amenable to ideas that postulate some sort of "mindedness" as a causal process alongside or in-between or superposed on physical causality. It really seems as if it is "up to me" to get out of the chair or not, to go visit a friend or not, to write this comment or not. Even for such trifles of the mind there don't seem to be any useful explanations in terms of e.g. quantum states or billiard balls or fields. Yet the moment I get up, that sets a lot of billiard balls and fields spinning. But caused by what?

Whether this idea will ever get anywhere, I mean anywhere more practical analytically than the persistent sense of bafflement and awe that thousands of years of philosophy have given us, I don't know. But it seems right that we're willing to glance in that direction again.
posted by dmh at 12:09 PM on March 15, 2021

What an incredibly incurious approach to life.

I am sure it seems so when regarded with the usual tools of science. However, viewed with the insights of philosophy and religion, we see that it is only by understanding the human mind that we can come to understand the rest of nature. As Spinoza put it:

[M]an conceives a human character much more stable than his own, and sees that there is no reason why he should not himself acquire such a character. Thus he is led to seek for means which will bring him to this pitch of perfection, and calls everything which will serve as such means a true good. The chief good is that he should arrive, together with other individuals if possible, at the possession of the aforesaid character. What that character is we shall show in due time, namely, that it is the knowledge of the union existing between the mind and the whole of nature…. Thus it is apparent to everyone that I wish to direct all sciences to one end and aim, so that we may attain to the supreme human perfection which we have named; and, therefore, whatsoever in the sciences does not serve to promote our object will have to be rejected as useless.

Know thyself, and thou shalt know all the rest.
posted by No Robots at 12:10 PM on March 15, 2021 [2 favorites]

none of the usual tools of science will open its fundamental nature to investigation

I don't think it has a "fundamental nature".

There's no undiscovered magic essence that makes it go, just a huge collection of behaviours whose details are really really difficult to measure.

It's completely obvious to me that consciousness is an umbrella term referring to certain kinds of behaviour exhibited by matter that happens to be arranged in a configuration similar to my own. To make the word "consciousness" useful it is simply not necessary to posit that mountains and rivers and stars either do or don't possess it. As for detecting consciousness, I'm essentially with Turing on this: if I want to know whether or not a system is conscious my best plan would be to have a conversation with it and then exercise my best judgement, same as I do for any of you.

Oh, and while my dander is up: unfalsifiable propositions are word play, no more and no less. Which is fine, I guess, if that's your jam, but really not terribly important in the grander scheme of things. And I think it's hilarious just how many of the so-called Big Questions that have bogged down finer minds than mine for countless generations turn out, on close examination, to be based squarely on unfalsifiable assumptions and/or underspecified or inconsistent referents for many of the words involved.
posted by flabdablet at 12:34 PM on March 15, 2021 [18 favorites]

n.b. I am right now having a wordless conversation with Alfie, a small hairy conscious system who is doing his best to make me understand that he needs to be let outside for a wee.
posted by flabdablet at 12:39 PM on March 15, 2021 [4 favorites]

Say what you will about the tenets of eliminative materialism, but at least it's an ideology
posted by thelonius at 12:55 PM on March 15, 2021 [9 favorites]

There's no undiscovered magic essence that makes it go, just a huge collection of behaviours whose details are really really difficult to measure.

I wonder what you think about the idea that some things appear not just difficult to measure but fundamentally immeasurable? E.g. there seems to be no way to "measure" what it is that makes a photon "choose" to reflect off a piece of glass. That behavior appears to be fundamentally probabilistic. Which isn't quite "undiscovered magic essence", but not very satisfying either.
posted by dmh at 1:00 PM on March 15, 2021 [1 favorite]

Science is in a double bind: it cannot use its measurement tools to apprehend the nature of the mind, nor does it recognize the ability of the mind to take its own measure.
posted by No Robots at 1:05 PM on March 15, 2021 [3 favorites]

Well... less “Animism” than “Nondualism”.

Or, as has been said before (was it Watts?):

At all points in history that we have access to, the following positions have been espoused:
The polytheist, who asserts “There are many gods”.
The monotheist, who asserts “The is one god”.
The atheist, who asserts “There is no god.”
The mystic, who asserts “There is nothing but god”.

Nondualism is in the last category.
—
The agnostic, who asserts “I do not know if there is or is not god” is a rather recent arrival in history compared to the other 4.
posted by Pirate-Bartender-Zombie-Monkey at 1:10 PM on March 15, 2021 [1 favorite]

Also seems like a framing problem: consciousness is a verb, a process in time, so understanding it as a thing is only by analogy. As the saying goes, it’s really hard to grab flowing water, so defining it as experience rather than thing should help better understand it on its terms.



[on preview: technically, an agnostic asserts that this can’t be known, right? (a-gnosis, outside of knowing/knowledge) because our temporal, embodied minds have no meaningful way to directly understand non-temporal, non-corporeal existence. So if god exists, we can’t even really notice, let alone apprehend; the blind can’t see blue.]
posted by LooseFilter at 1:20 PM on March 15, 2021 [1 favorite]

For a comprehensive treatment of the non-dualist position, see Harry Waton's A true monistic philosophy, full text at HathiTrust.
posted by No Robots at 1:21 PM on March 15, 2021 [1 favorite]

The problem with saying that everything has experience is that you immediately have to answer the rather difficult question what is a thing?

Does half of my arm count? A page in a book? The composite of myself and Jupiter?
posted by nosewings at 1:27 PM on March 15, 2021 [5 favorites]

Beyond a certain point, I am comfortable with residing in magic.. ignorance.. whatever.. And I can think about and discuss the topic endlessly, I just don't think we can ever have the type of certainty some people seem to need on certain definitions and assertions.

The very act of scrutinizing and discussing a thing or phenomenon is to acknowledge your distance/removal from that thing. I'm not sure we always appreciate the degree to which our descriptions of things are distinct from those things, we seem to get these 'things' confused at times. I like that how we conceive of 'consciousness' is expanding, this makes sense to me.
posted by elkevelvet at 2:00 PM on March 15, 2021 [6 favorites]

Beyond a certain point, I am comfortable with residing in magic.. ignorance.. whatever

Why not try a little reasoning? Or has our science come to hold even reason in contempt?
posted by No Robots at 2:16 PM on March 15, 2021

How about I put it this way: the greatest intellect in history will eventually exhaust the ability to reason and collapse in wonder. I'm not sure this is a controversial point, perhaps you are much closer to being the greatest intellect in history?

How would you take my comment as an opportunity to suggest I "try a little reasoning"? I'm not sure the word contempt has any place in the conversation either. Forgive me if the comment seemed trivial or misguided to you, but you might also search your impulse to jump to conclusions?
posted by elkevelvet at 2:24 PM on March 15, 2021 [10 favorites]

The word "panpsychism" has an unfortunate New Age ring to it, plus it is difficult for us to imagine consciousness in stones and sidewalks, let alone trees.

But it does solve the "hard problem of consciousness" quite easily. A little too easily, one might think. Part of the problem is that we habitually define consciousness as akin to observing, thinking and feeling. If one were to think of consciousness differently, then the problem is solved. How is one to think of consciousness differently? The mystic and the philosopher and the scientist all have different ways of approaching this question.
posted by kozad at 2:45 PM on March 15, 2021 [1 favorite]

The problem I have with the idea that consciousness is everywhere is also a problem that I have with the idea that consciousness doesn't exist at all. The two theories give opposite answers, but they do so consistently. Given the question "Is X conscious?" one always says "yes" and the other always says "no". That "always" is a serious problem. It precludes any possibility of conveying any meaningful information about what consciousness actually is or is not.

More generally, I think any theory of consciousness that proposes that a human being who is awake and active is no more or less conscious than one who is anaesthetised, or even dead, is a theory that can be safely discarded. Don't be led astray by simplicity or elegance in a theory. "Simple, obvious, and wrong" is very much a thing.

IIT seems to be a sort of panpsychism-with-filters, based on how much information processing is going on. I'm not sure if it's really saying anything about actual consciousness, but at least it's trying.
posted by swr at 3:06 PM on March 15, 2021 [2 favorites]

Why not try a little reasoning? Or has our science come to hold even reason in contempt?

Hey maybe take a step back and stop thinking of literally everyone else in this conversation as deluded/bad-faith/ignorant? Evidence (i know, blech) suggests otherwise, and your comments seem, at best, tangential to TFA.
posted by supercres at 3:14 PM on March 15, 2021 [8 favorites]

As a newcomer to IIT , it surprised me that this article never explained what it is really. I guessed from context, sort of, but given the level of the first half it’s weird that he just hand waved IIT. Also count me as someone who’s never seen “abduce” in print. Dictionary says it’s just “deduce”.
posted by freecellwizard at 3:43 PM on March 15, 2021

I think "abductive inference" is what Sherlock Holmes typically did, perhaps misheard as "deductive" by poor old Watson. I've never seen the word used outside philosophy papers. Maybe in law-talking academia they also use it.
posted by thelonius at 3:54 PM on March 15, 2021 [1 favorite]

Okay, consciousness is experience. Then, what is experience, and, yeah, what is a thing? I’ll hang up and listen.
posted by Don.Kinsayder at 3:58 PM on March 15, 2021

Abductive reasoning. Though it's not clear to me if Koch is really using abduction when he says in this essay that he "abduces" things, or if he's, as zittrain suggests, just using it as a synonym for "guess".
posted by biogeo at 4:02 PM on March 15, 2021

Is there any actual evidence of consciousness? Any reason not to take it as, like the self, merely a convenient narrative construct for the kinds of stories we like to tell about ourselves?
posted by Not A Thing at 5:32 PM on March 15, 2021 [1 favorite]

One might say consciousnesses is the only thing we have evidence of, for a certain well chosen definition of the word "evidence".
posted by grog at 6:16 PM on March 15, 2021 [5 favorites]

Oh, and while my dander is up: unfalsifiable propositions are word play, no more and no less.

The crossword takes offence!

'I think "abductive inference" is what Sherlock Holmes typically did, perhaps misheard as "deductive" by poor old Watson.'

Well, in the kidnapping cases, anyway.

But it does solve the "hard problem of consciousness" quite easily. A little too easily, one might think.

Ah, but can one think one has solved the hard problem of consciousness if they've determined that consciousness does not exist? Or make judgements about how hard the problem was in the first place?

Here goes: Consciousness is experience.

My 12th level drow sorcerer just leveled up in a big way.

(ok, i'll stop. sorry.)
posted by kaibutsu at 6:18 PM on March 15, 2021 [1 favorite]

it is difficult for us to imagine consciousness in stones and sidewalks, let alone trees

Actually, I've be more in for the trees than the stones. There's been stuff reported pretty widely lately about how trees and mycelium sort of work together to form pretty vast interconnected swaths of forest which can send chemical signals to communicate status or threat to others, and also food can be transferred through this network from older trees to younger or sick trees who need additional help.

I'd say, based on this stuff I've heard about probably only in the past year or two, I'm very much for believing that trees, especially groups of them growing in the wild, have a consciousness about them.
posted by hippybear at 6:33 PM on March 15, 2021 [3 favorites]

Can anybody shed light on what the author has in mind by claiming that IIT, and particularly the whole thing about Wholes, is "testable"? How would one test for the thingishness of Wholes?

(I can imagine a test based on felicity or narrative coherence, but I don't think that's the language he's speaking)
posted by Not A Thing at 6:49 PM on March 15, 2021

The polytheist, who asserts “There are many gods”.
The monotheist, who asserts “The is one god”.
The atheist, who asserts “There is no god.”
The mystic, who asserts “There is nothing but god”.

The flabdablet, who asserts "Make it clear to me what it is that you think the word 'god' refers to and maybe we can talk".
posted by flabdablet at 7:01 PM on March 15, 2021 [7 favorites]

Article: Apart from the occasional solitary solipsist

Me: Is there any other kind?
posted by Sparx at 7:11 PM on March 15, 2021 [9 favorites]

Given the question "Is X conscious?" one always says "yes" and the other always says "no". That "always" is a serious problem.

It is a problem, but couldn't it be a problem with the question? If we ask instead "is it helpful to attribute consciousness to X?" then the answer becomes largely a matter of choosing operational definitions of "helpful" and "consciousness" that make sense for a particular purpose.

(It has seemed to me that definitions of consciousness tend to be chosen with an eye to what the definer wants to exclude. Eg those who want to insist that consciousness is exclusively human will choose something they think is exclusively human as a marker of consciousness, such as tool use or language use, moving the goalposts as needed when it turns out that nonhumans do these things too. In this regard the author's insistence that a computer cannot be a Whole is somewhat striking.)
posted by Not A Thing at 7:13 PM on March 15, 2021 [2 favorites]

you immediately have to answer the rather difficult question what is a thing?

Having given that particular question some considerable thought over many years, the best I can do there is that a thing is whatever you choose to treat as distinct from all that it is not.

Does half of my arm count?

Yes.

A page in a book?

Yes.

The composite of myself and Jupiter?

Yes.

Personally I also use internally a somewhat idiosyncratic family of words to label various kinds of thing. I like "object" for things whose constituents are relatively stable (e.g. chairs, billiard balls) and "process" for things whose constituents are more usefully thought of as being in some kind of flow or state of change (e.g. candle flames, music).

People are complex things with many constituent objects and processes.

It makes much more sense to me to think of my consciousness as a process rather than an object.
posted by flabdablet at 7:16 PM on March 15, 2021 [6 favorites]

Following up on my previous comment, it seems to me like the author's answer to "does it make sense to attribute consciousness to computers?" would be "no, because the whole point of consciousness is to have a club that computers can't be part of." But perhaps that's unfair.
posted by Not A Thing at 7:19 PM on March 15, 2021

does it make sense to attribute consciousness to computers?

Certainly not to any computer I've ever encountered.

I can see no in-principle reason why it ought to be considered impossible to implement a conscious process on engineered hardware, but to the best of my knowledge it hasn't yet been done and I suspect that the energy consumption constraints on ways it could be done will turn out to be far more important than most people seem to think.
posted by flabdablet at 7:25 PM on March 15, 2021

Why not try a little reasoning? Or has our science come to hold even reason in contempt?

Reasoning is useful but there's no need to make a religion out of it.
posted by flabdablet at 7:28 PM on March 15, 2021 [2 favorites]

Is there any actual evidence of consciousness? Any reason not to take it as, like the self, merely a convenient narrative construct for the kinds of stories we like to tell about ourselves?

Without consciousness, wherefore the question, and who is the audience? Consciousness has causal effects that seem difficult to explain without it. Like Metafilter comments.
posted by dmh at 7:45 PM on March 15, 2021 [1 favorite]

IIT offers a different chain of reasoning. The theory precisely answers the question of who can have an experience: anything with a non-zero maximum of integrated information; anything that has intrinsic causal powers is considered a Whole. What this Whole feels, its experience, is given by its maximally irreducible cause-effect structure. How much it exists is given by its integrated information.

To my way of thinking, the appearance of a Word in a text is a strong indication that we're about to embark on exploration of somebody else's idiosyncratic preferences in ways to label things that they're currently taking a bit too seriously. I did a lot of that kind of thing myself on the descent into psychosis; very very little on the climb out of it.

In other words, the theory doesn’t stipulate that there is some magical threshold for experience to switch on. The degree of consciousness is instead measured with Φ, or phi. If phi is zero, then the system doesn’t exist for itself; anything with Φmax greater than zero exists for itself, has an inner view, and has some degree of irreducibility — the larger this number, the more conscious it is.

I am personally dubious about the usefulness of a conception of consciousness that would allow us to arrange putative examples of it along a one-dimensional scale. That way lies the same kind of doctrinaire refusal to acknowledge value that's accreted onto the notion of IQ.

I would rather modulate my interest in the consciousness of other systems in ways that better reflect the interest they appear to be taking in mine.
posted by flabdablet at 7:48 PM on March 15, 2021 [2 favorites]

Consciousness has causal effects that seem difficult to explain without it.

Causality is overrated.
posted by flabdablet at 7:50 PM on March 15, 2021

Science is in a double bind: it cannot use its measurement tools to apprehend the nature of the mind, nor does it recognize the ability of the mind to take its own measure.

I have met many scientists. I can't think of a single one who appears to live in any such bind. All have been curious, playful, generous, thoughtful and a pleasure to spend time with.
posted by flabdablet at 7:57 PM on March 15, 2021 [2 favorites]

science is great for understanding the known and perhaps knowable universe, which may in fact prove to be everything. Maybe. Because I do worry when science (some of those pursuing it anyway) simply assume that everything can and/or will in time be knowable. I mean, that's a hell of an assumption.
posted by philip-random at 8:16 PM on March 15, 2021 [1 favorite]

Without consciousness, wherefore the question, and who is the audience?

What? How does "consciousness" help me to make any sense of either the question or the audience?

Or do you mean that a story about questions and answers is one that requires implicitly imputing "consciousness" to the entities involved, under the particular conventions of the genre in this time and place, in the same way that some kinds of stories require a "protagonist" or a "romantic interest"? Or the way that many modern genres require describing motivations in terms of character development rather than random neuron firings or divine intervention? In which case, yes, that's what I was inexpertly driving at.

. . . wait is this one of those "checkmate atheists" meme things that the kids are doing
posted by Not A Thing at 8:35 PM on March 15, 2021

I do worry when science (some of those pursuing it anyway) simply assume that everything can and/or will in time be knowable.

I've never met, in person, an actual living breathing scientist who has claimed that it would be. Every scientist I know has been struck by the same phenomenon in scientific inquiry as I have: that finding solid, testable answers to any question just prompts ten more questions.

It seems to me that the universe consistently provides overwhelmingly more opportunities for ignorance than for knowledge. Which, given that it's so much bigger than our tiny brains, strikes me as exactly how things should be expected to work.

Over the decades I've spent working in IT there's been a positively Cambrian explosion in IT I don't know about and that's just one tiny sliver of human inquiry. Are there other fields I'm unaware of in which this is not true?
posted by flabdablet at 8:58 PM on March 15, 2021 [2 favorites]

Or do you mean that a story about questions and answers is one that requires implicitly imputing "consciousness" to the entities involved

The way I see it there's two parts to it I guess. Firstly I take it that for you to have the idea that consciousness is a story, and commenting to that effect, to me those seem like conscious actions. It seems difficult to be conscious of the idea that consciousness isn't real.

At least that's how I understand it when you say consciousness is a story. I assume you mean it's not "real" in some sense. Perhaps that's not what you meant. I'd certainly agree it's a nebulous concept. But at the same time I seem to experience it directly, a lot like I can feel the sun or smell the air, and honestly I have no idea exactly how I do those things either. Maybe the sun and the air are not "real", but it doesn't seem that way. That brings me to the second point. How to even distinguish between what's real and what's not, without being conscious of the distinction? Isn't some sense of awareness of reality exactly what it means to be conscious?

To me personally the idea that consciousness can be done away with is akin to saying that the planets and the stars are really dabs of paint on the inside of a giant dome. I think it solves the problem of space by eliminating space as such. But who knows. Perhaps it's the other way around, and what I take to be consciousness, is just dabs of paint.
posted by dmh at 9:56 PM on March 15, 2021 [1 favorite]

Isn't some sense of awareness of reality exactly what it means to be conscious?

Having personally experienced states of awareness that didn't involve being aware of anything, I'm inclined not to define the word that way.
posted by flabdablet at 10:23 PM on March 15, 2021 [1 favorite]

I find it fairly intuitively acceptable that consciousness at various levels of complexity would be everywhere, but only because, inexplicably, there do seem to be things that exist of various levels of complexity everywhere, and there doesn't seem to be any particularly good explanation for that, either. It seems like it would be much easier for there to be nothing.
posted by Jon Mitchell at 10:45 PM on March 15, 2021 [1 favorite]

Easier for what?
posted by flabdablet at 10:48 PM on March 15, 2021

All have been curious, playful, generous, thoughtful and a pleasure to spend time with.

I assure you, some of us are real assholes. But I'm glad that hasn't been your experience!
posted by biogeo at 11:25 PM on March 15, 2021 [6 favorites]

Can anybody shed light on what the author has in mind by claiming that IIT, and particularly the whole thing about Wholes, is "testable"? How would one test for the thingishness of Wholes?

I think the argument goes something like this: integrated information is a quantity which can, at least in principle, be measured for a system. A system is a "Whole" if its integrated information is "maximally irreducible," by which I think it's meant that either adding or removing anything to the system would reduce its integrated information. It's unclear to me how this idea is supposed to be testable. It's also unclear to me that one can demonstrate that integrated information is, in fact, maximally irreducible for a human as opposed to a society of humans. However, (the argument goes) we can at least test the postulate that consciousness is related to integrated information by measuring it in humans who we know to be conscious versus one we know to be unconscious (e.g., due to anesthesia). If we find that the unconscious state has reduced integrated information, we can "abduce" that the IIT postulate is correct. This experiment has been done and support for IIT has at least been claimed as a result. My understanding of this argument may not be quite right: I find the mathematical formalism of IIT to be unnecessarily opaque in ways that make it hard for me to understand what exactly the actual claims are, and the non-mathematical argument strikes me as too vague to reason about clearly. But this is my best understanding.

A criticism in response seems to be that in practice, integrated information cannot actually be measured for any sufficiently complicated system. The IIT proponents respond that it can be reasonably approximated, and the IIT critics reply that different methods of approximation produce wildly different results suggesting that these approximations are not actually very good. Furthermore, there's a suggestion that even minor changes to the structure of a brain would produce wildly different levels of integrated information, which seems like a problem given that we know the brain is fairly robust to most small perturbations. Both of these seem to cast doubt on the idea that IIT can formalize the notion of a "Whole" in a way that captures our intuitions about human consciousness.

This isn't directly related to the question of "Wholeness" but this also strikes me as interesting, from the Wikipedia article:

Theoretical computer scientist Scott Aaronson has criticized IIT by demonstrating through its own formulation that an inactive series of logic gates, arranged in the correct way, would not only be conscious but be “unboundedly more conscious than humans are.” Tononi himself agrees with the assessment and argues that according to IIT, an even simpler arrangement of inactive logic gates, if large enough, would also be conscious. However he further argues that this is a strength of IIT rather than a weakness.

This seems to stand in direct contradiction to Koch's claim that conventional digital computers cannot be Wholes.

Honestly I'm inclined to think Mike Graziano is right: IIT smells like pseudoscience to me. Christof Koch is a smart guy and has done a lot of good science, but unfortunately that's no guarantee that someone won't fall into pseudoscientific thinking: look at Linus Pauling's work on vitamin C megadosing. Maybe I'm wrong, and it would be great if I am: if we had a real theoretical and mathematical framework for describing the level of a person's consciousness based solely on physiological measurements, that would be hugely valuable not just from a theoretical perspective, but would have real practical applications in anesthesiology and the study of persistent vegetative states. But I'm afraid right now I just don't see it.
posted by biogeo at 12:17 AM on March 16, 2021 [6 favorites]

Here is a reasonable seeming IIT introduction from a SciAm blog; I think it's where the Wikipedia quote from Scott Aaronson is from.
posted by thelonius at 1:47 AM on March 16, 2021

IIT smells like pseudoscience to me

I'd go for "reeks of" rather than "smells like" but basically yeah.
posted by flabdablet at 2:37 AM on March 16, 2021

Here is a reasonable seeming IIT introduction from a SciAm blog; I think it's where the Wikipedia quote from Scott Aaronson is from.

Interesting, thanks. The author John Horgan agrees with Searle's objection against IIT, that it's circular to invoke notions of information to explain consciousness:

The concept of information makes no sense in the absence of something to be informed—that is, a conscious observer capable of choice, or free will (sorry, I can't help it, free will is an obsession). If all the humans in the world vanished tomorrow, all the information would vanish, too.

For all the reasonable skepticism of IIT (which I share), this argument I can't follow. If all the humans in the world vanished tomorrow, there would still remain all kinds of things that seem to require the notion of information to explain. There would still be DNA to produce squirrels, and those squirrels would still remember where they buried their nuts, and there would still be a notion of a "nut" as something that provides the necessary nutrition for the production of more squirrels. All of these things seem to entail or require some notion of information, but not necessarily (human-like) consciousness.
posted by dmh at 7:59 AM on March 16, 2021 [3 favorites]

I'd say, based on this stuff I've heard about probably only in the past year or two, I'm very much for believing that trees, especially groups of them growing in the wild, have a consciousness about them.

So I have a dumb peach tree that yet again froze it's blooms and this year leaves off in a normal freeze. I'm waiting to see if it is dead, or capable of learning to set leaves and fruit a bit later like all other trees.
posted by The_Vegetables at 8:12 AM on March 16, 2021 [1 favorite]

this argument I can't follow

That's just Searle being Searle. He's got a referent for "consciousness" that he's sticking to come hell or high water and a reflexive tendency to bat away anything that might actually require him to challenge it.

It's not as well-defined as he clearly thinks it is; like the old story about porn being something that I know when I see it, consciousness for Searle is something he knows when he be it - but over the decades he's padded and protected what is really no more than an opinion inside such a lovingly crocheted web of circular argument that he now seems to take it as objectively true.

It's pleasingly hilarious to see him dinging somebody else's argument for circularity, though.
posted by flabdablet at 9:17 AM on March 16, 2021 [2 favorites]

Searle being Searle

I do sort of admire him for naming the philosophical position that he wants to refute, and that runs from at least Kant all the way into the 20th century, "The Bad Argument", in his recent book. (iirc, it is that minds have a sort of direct access to perceptions that they don't have to the objects that the perceptions refer to or are caused by, but Googling "Searle bad argument" mostly gave results from people attacking the Chinese Room, last time I tried it).
posted by thelonius at 9:37 AM on March 16, 2021 [1 favorite]

If all the humans in the world vanished tomorrow, there would still remain all kinds of things that seem to require the notion of information to explain

The central point is that if all the humans in the world vanished tomorrow, there would be nobody left to do any such requiring, and nobody capable of constructing a word like "information" and associating an appropriately abstract referent with it either.

I agree with Searle to the extent that if there's nobody around who is doing any thinking then there's nobody to make the distinctions that give things their thingness. Existence of an opinion-haver is a necessary precondition for the act of opining that "there would still remain all kinds of things". There cannot be anything at all to say about reality without meaning-makers to say it.

Where I depart from Searle is that I see no reason why other living things or indeed engineered structures bearing very little physical resemblance to human beings could never in principle behave in a way that they would themselves describe as conscious, that description being at least as convincing as anything I could come up with on my own behalf.

I'm conscious because I'm a particular kind of complicated, not because I'm infused with some kind of consciousness phlogiston.
posted by flabdablet at 9:40 AM on March 16, 2021 [2 favorites]

The argument that information requires a conscious observer seems to be missing the rather key point that the "information" in question is Shannon information, or at least a mathematical relative of it. I think there's a reasonable argument to be made about whether Shannon information is really information in the same sense as our everyday usage of the word, but it's a well-defined physical property that's entirely independent of conscious observers, or indeed of any particular function. This is a surprisingly fundamental point to miss in raising the objection that an information-based explanation of consciousness is necessarily circular. I think one could certainly object that IIT trades in a certain rhetorical advantage from the fact that when many people hear "information" they automatically think teleology and consciousness, but the term "information" as actually used in the formal construction of such a theory doesn't actually assume those things.
posted by biogeo at 9:53 AM on March 16, 2021 [4 favorites]

it's a well-defined physical property that's entirely independent of conscious observers

The point is that total removal of any entity that both makes and cares about definitions makes all definitions moot, regardless of how useful they might be to entities that do care about definitions.

The way I see it, all physical properties are attributes of the models that conscious entities like ourselves construct on an ongoing basis as part of the process of being conscious, and the degree to which they're well defined depends entirely on the rigour with which those models are maintained.

In other words, every physical property is in and of itself an abstraction of some part or aspect of reality. As such, properties belong to the map, not the territory. So I disagree that any well-defined physical property is in fact entirely independent of conscious observers for the simple reason that it's conscious observers who make and maintain and in fact contain the definitions.

The fact that we can use our understanding of well-defined properties such as Shannon information to make accurate predictions about the behaviour of parts of reality that we have no way to apprehend directly doesn't give the properties themselves any kind of existence independent of us; it just means our models work. The point is that they work for us.

Without model makers there is nothing at all to say about the reality that the models would otherwise be of.
posted by flabdablet at 10:18 AM on March 16, 2021

Eh, it's a bit of a dodge... does the tree in the forest exist without observers? Your answer says no, tree and forest are abstract concepts and without any observer to impose those constructs on the environment they don't exist. Whereas I would say that there's some underlying confluence of physical phenomena that we're assigning the tree+forest labels to, which certainly do exist independent of observation. (indeed, observer-free phenomena predate and gave rise to humanity in the first place.)

Shannon entropy+information are the same thing: We've got an abstract wrapper/model for a collection of observable phenomena. That doesn't mean that the things being observed don't exist independently of the model. As a more concrete example, we can measure the brightness of a pulsar once every second, assign the bit 'bright' or 'dim', and then measure the information content of the stream. The numerical measurement and methodology are absolutely brought by us, but they're saying something about the pulsar. Both the pulsar and the signal that we're sampling exist independently of any observer. Measuring the information content is similar to calling the tree 'green.'
posted by kaibutsu at 1:00 PM on March 16, 2021 [3 favorites]

fladblet, I think more than anything I'm enjoying your commentary on Searle. The Construction of Social Reality and The Rediscovery of the Mind were two primary texts in an undergraduate Philosophy of Mind course I took, close to the date they were published. I think the prof was smitten at the time, and the enthusiasm for Searle was fairly baked into the class.
posted by elkevelvet at 1:39 PM on March 16, 2021

flabdablet, I appreciate the argument and I think it's very well-stated. But I have to agree with kaibutsu. You could also argue that conservation of energy and momentum are just models, but regardless of whether or not there are people making those models, the universe behaves in a way consistent with them. This feels a bit like the use/mention confusion with statements like "love is just a word". Sure, information theory is just a model, but there is a reality described by that model, and it has certain behaviors and those behaviors have interrelationships. Information theory models a reality in which the quantity it describes is not dependent upon function or use by a conscious observer. That's why it's not circular to try to use information theory to describe consciousness. I don't think IIT actually succeeds at this, but that doesn't mean some other information theoretic approach couldn't.
posted by biogeo at 4:05 PM on March 16, 2021

The fact that we can use our understanding of well-defined properties such as Shannon information to make accurate predictions about the behaviour of parts of reality that we have no way to apprehend directly doesn't give the properties themselves any kind of existence independent of us; it just means our models work. The point is that they work for us.

I feel that kind of borrows the mundane everyday-ness of the word "work" to gloss the problem. Is there any notion of "to work" that doesn't in the limit resolve to "what we know to be true or real, to the extent we can know it"?

I also think that it's not so easy to distinguish between the map and the territory, because we ourselves exist as part of the reality that's being apprehended. So it's not so clear to me how saying something works "for us" if very different from saying it works "in reality". While we can harbor doubts about the existence of entities postulated by our understanding of physical properties, we can't very well doubt the existence of understanding itself (which is what I take to be Descartes cogito). Our understanding of reality has to exist within and as part of that reality, and it seems difficult to square that with the notion of reality as fundamentally/intrinsically unintelligible/unknowable.
posted by dmh at 4:10 PM on March 16, 2021 [1 favorite]

Part of the problem is that we habitually define consciousness as akin to observing, thinking and feeling. If one were to think of consciousness differently, then the problem is solved. How is one to think of consciousness differently? The mystic and the philosopher and the scientist all have different ways of approaching this question.

Yes... so much of the discourse about consciousness (including in this thread) conflates it with other forms of mental activity, which makes it very difficult to talk about, since people are each operating with different, often unclear, definitions.

Consciousness is best understood as something distinct from perception, thinking, and feeling -- something that exists beneath or behind all of those things, as a kind of mental foundation. It's inherently subjective -- it is the irreducible essence of subjectivity.

And so the materialist tools of science can't get a real grasp on it. Devoted adherents of scientism tend to either say that consciousness must be emergent from physical processes in a way which we'll soon be able to unravel and document, any day now, for sure, just you wait... or, even weirder, they deny that it exists at all.

Both answers are so manifestly inadequate. But that's what happens when people acquire a set of mental tools that are useful for certain tasks (the tools of science) and then believe those tools are the only tools they need to understand anything and everything.

I like this passage from David Bentley Hart on the subject:

...[T]hese questions are not answered by trying to show how consciousness can be built up from the raw accumulation of the purely physical systems and subsystems and modular concrescences constituting conscious organisms. At some stage of organic complexity in that process -- amoeba, cephalopod, reptile, viviparous mammal, Australopithecus, what have you -- a qualitative abyss still must be bridged.

It might be tempting to imagine that we could imaginatively dissolve consciousness into ever smaller and more particular elements, until we reached the barest material substrate, and then conceptually reconstitute it again without the invocation of any immaterial aptitudes, just as we can make the image on that pointillist canvas I mentioned above dissolve before our eyes simply by drawing as near to it as possible and can then make it reappear simply by stepping sufficiently far back again. ... But, then again, there is something usefully recursive about this metaphor: Who does the standing back, after all? Where is this point of perspective that allows for the appearance of an ordered unity located? At what point does the chaos of sensory processes somehow acquire a singular point of view of itself? These are not facetious questions.

There is a troubling tendency among materialist philosophers of mind and cognitive scientists to indulge in analogies that, far from making consciousness more intelligible, are themselves intelligible only because they presume the operations of consciousness. It is not uncommon to find cameras or televisions mentioned as mechanical analogies of the mind's processes of representation; but, of course, a camera does not look at pictures and a television does not watch itself, and there is nothing even remotely representational in their functions apart from the intentions of a conscious mind, which is to be found not in those devices but in a person. ... All such analogies terminate precisely where they began: in the living mind, imperturbable in its incommunicable subjectivity and awareness, still the mysterious glass in which being shines forth as thought.
posted by Artifice_Eternity at 7:47 PM on March 16, 2021 [1 favorite]

I was a little flippant before, but I'm legit having a hard time distinguishing these two arguments:
A. It is absurd to doubt the existence of God, because God created everything including the doubter, and if God did not exist then the doubter would not exist either.
B. It is absurd to doubt the existence of consciousness, because all subjective experience including doubt rests on an "irreducible essence of subjectivity", and if that irreducible essence did not exist then doubt would not exist either.

I mean. If I don't believe in a deus otiosus up in the sky, why on earth would I believe in a homunculus otiosus inside my head?

The drive to believe in consciousness/Wholeness is understandable. People want something they can point to so they can claim membership in the No Nonhuman Animals Club, the No Plants Club, or more specifically to Koch, the No Computers Club. But Koch's article contains a number of flashing warning lights (just in case we needed them) about the ugly places that drive to exclude can take us. He cites facial recognition in bees as evidence of consciousness in insects. Does that mean that prosopagnostics like myself are not merely subhuman but sub-insect? I'm sure he would be suitably horrified at the implication, but the inference is straightforward enough. And closer to the heart of his argument, the implications of IIT for memory-related disabilities are alarming.

My own tentative operational definition of consciousness would be something like "having a relatable inner life". It is good to act as if we and other humans are conscious, because assuming that we and others can relate to each other's inner lives (even if we really can't) makes it easier to get by in the clamor and tumult of human society. And in a different way, it is also good to act as if all things are in some measure conscious, because only by attempting to relate to them do we have a fair chance of understanding how they do what they do. (The importance of desire and volition in premodern Western physics comes to mind here.) But this definition too can lead to dangerous places if we were to take it too seriously. After all, many neurodivergent people do not have inner lives that neurotypicals find relatable (and vice versa), see for example the widespread though absurd belief that autists lack a "theory of mind".

On reflection, I'm rather skeptical of any theory of consciousness (including my own, such as it is) that isn't firmly grounded in disability theory and anti-ableism. Are there any that are?
posted by Not A Thing at 9:10 PM on March 16, 2021 [2 favorites]

I'm sure he would be suitably horrified at the implication, but the inference is straightforward enough. And closer to the heart of his argument, the implications of IIT for memory-related disabilities are alarming.

I think the kinder reading is that consciousness tests are for 'sufficient but not necessary' conditions. In other words, you don't have to pass Test X as a human because we know* humans are conscious. But if we agree that Test X demonstrates some fundamental aspect of consciousness, and you can find a really smart [marmoset/neural network/piece of lint/david bowie album] that passes Test X, then we've expanded the size of the club. No one expects a blind person to pass the mirror test, but no one believes that says anything about consciousness.
posted by kaibutsu at 9:57 PM on March 16, 2021

Not A Thing:

It is absurd to doubt the existence of consciousness, because all subjective experience including doubt rests on an "irreducible essence of subjectivity", and if that irreducible essence did not exist then doubt would not exist either.

I mean. If I don't believe in a deus otiosus up in the sky, why on earth would I believe in a homunculus otiosus inside my head?

To me, the first sentence is self-evident. I gather the second sentence is meant as a critique of it... but I'm not quite sure I see how. You don't need to believe in a homunculus to contemplate the striking qualities of your own subjectivity.
posted by Artifice_Eternity at 10:19 PM on March 16, 2021

This is going to be a bit scattergun. Please don't assume that any of the positions I'm about to have a go at are ones that I think you personally hold or have argued for above. All I'm trying to do here is clarify my own position, not have a crack at yours, even when I'm quoting you as a starting point.

regardless of whether or not there are people making those models, the universe behaves in a way consistent with them.

I struggle to understand the "not" case here. If there were no people making models, and therefore no extant copies of them remaining, exactly what is it that you're claiming that the universe's behaviour would be consistent with?

I naturally have no difficulty in agreeing that it is reasonable to assert that the behaviour of even such parts of the universe as have no access to the models we've already made, including but not limited to those parts of it that existed before the models were made, those parts that have existed and do exist and will exist beyond our own Hubble radius, those parts that will continue to exist long after the last model maker has been strangled in the entrails of its last economic system, and such forests as contain unstable trees without conscious observers, would most likely be well described by the excellent models we do have already.

What I'm really getting at is that the reason our models work so well is that we have constructed them as efficient compressions of the things that we understand about the behaviour of the universe. Our models capture quite a lot of reality's behaviour very succinctly, but I am very very resistant to any notion that reality's behaviour is constrained by our models. To my way of thinking, the question of what it is that "breathes fire into the equations" is utterly wrong-headed. The equations exist in us and for us; the models themselves - as distinct from those aspects of existence referred to and related by the models - have no existence without us.

When people argue for example that it is remarkable that even a tiny alteration to the fine structure constant would preclude the existence of human beings altogether, and conclude that the fine structure constant must therefore have been fine-tuned by some kind of designer in order to secure our existence, I think they've got the wrong end of the stick. The fine structure constant is what it is because if it were different it would not fit. It would be part of a bad model.

When you say "regardless of whether or not there are people making those models", I read that as the opening of an argument about a hypothetical universe identical to this one except for the absence of conscious model makers.

The trouble with that is that the resulting universe is hypothetical. That is, it's a model. And this gives it a property that reality of which we form a part doesn't have: an ability to be examined, and reasoned about, from outside itself.

From the outside, any such model is naturally going to be full of the same kinds of things we identify inside our containing reality. But this collection of distinctions is entirely an artifact of the way the model has been constructed. It seems to me, therefore, that claims about the inevitable existence of things derived from such a model and then applied to the non-hypothetical universe that we actually occupy are on very shaky ground.

I can't see any reason to assume that the ways in which we sense and understand the behaviour of the universe are the only ones that could feasibly give rise to models with at least as much explanatory and predictive power as those we use. They're good models Brent.

More later. My back hurts now.
posted by flabdablet at 2:31 AM on March 17, 2021

On reflection, I'm rather skeptical of any theory of consciousness (including my own, such as it is) that isn't firmly grounded in disability theory and anti-ableism. Are there any that are?

I think any theory / praxis of justice has to acknowledge the "realness" of the subjective experience of harm, because we don't have a fully objective account of that experience. But even if we did have such an objective account, it's hard to see how it could, in general, provide a moral justification to refute or ignore the statement "I am suffering" (although I think it's possible to imagine such a justification in specific circumstances or for specific purposes).

I think Marx's theory of exploitation can be read as an attempt to secularize and objectivize the notion of harm as a rate of exploitation. I also think it fails in that respect, because for the theory to work, i.e. for us to be persuaded there is something unjust about the exploitation of labor as Marx theorizes, there has to be some moral subject towards which it is possible to commit injustice. It doesn't seem to make much sense, outside of ecological concerns, to say that there is anything unjust about exploiting sand, for example, or minerals, or an orchard. It's only by assigning some degree of consciousness / subjectivity to the object being exploited that we can begin to see how exploitation might cause harm, and conflict with the moral imperative to reduce suffering, or certainly at least not to inflict it for the sole purpose of personal gain.
posted by dmh at 2:36 AM on March 17, 2021

I've been following this thread with great interest and enjoyment. Thank you all. I'm a bit out of my depth and reluctant to make a comment that might drag the discussion down but I can't help myself. To borrow an analogy I read somewhere - We're in a cave, trying to make sense of the world by examining the shadows on the wall. I'm of the opinion that surmising that there might be things outside that are not currently casting shadows is not the same thing as believing in magic.
posted by night_train at 4:18 AM on March 17, 2021 [2 favorites]

Scattergun continues.

There is a troubling tendency among materialist philosophers of mind and cognitive scientists to indulge in analogies that, far from making consciousness more intelligible, are themselves intelligible only because they presume the operations of consciousness. It is not uncommon to find cameras or televisions mentioned as mechanical analogies of the mind's processes of representation; but, of course, a camera does not look at pictures and a television does not watch itself, and there is nothing even remotely representational in their functions apart from the intentions of a conscious mind, which is to be found not in those devices but in a person. ... All such analogies terminate precisely where they began: in the living mind, imperturbable in its incommunicable subjectivity and awareness, still the mysterious glass in which being shines forth as thought.

There is a troubling tendency among mystics to indulge in analogies that, far from making consciousness more intelligible, serve to raise the inherent difficulty in understanding so complex a beast as a conscious being to the level of impossibility even in principle. It is not uncommon to find mirrors and mysteries artfully alluded to in vaguely poetic language so as to distract from the fact that attempting to understand consciousness as a complex emergent phenomenon is a prima facie reasonable endeavour, and this rhetorical manoeuvre is often coupled with criticism of analogies involving such things as cameras or televisions on the basis that these things do not operate in ways complex enough to yield explanatory insight. However, principled refusal to engage in a carefully introspective process of inquiry is not at all the same thing as a demonstration that no such process could even in principle eventually yield useful results.

A. It is absurd to doubt the existence of God, because God created everything including the doubter, and if God did not exist then the doubter would not exist either.

This argument is trivially sound as long as the referent of the word "God" is identical with that of the phrase "whatever it was that created everything including the doubter". But I have yet to see a longer piece featuring this, or similar arguments such as Anselm's ontological one, where that kind of identity is ever explicitly acknowledged. It's far more common to see those who say this kind of thing do so from a place of having a pre-conceived notion of what the word "God" refers to, then run a bait-and-switch to sub in their preferred referent at the end of the argument and act as if the argument itself actually supports their having done that.

In particular, a self-creating universe makes a completely suitable referent for "God" in both argument A and in Anselm's, neither of which has anything to say about whether the referent of "God" exhibits anything even vaguely resembling consciousness when considered as a whole.

B. It is absurd to doubt the existence of consciousness, because all subjective experience including doubt rests on an "irreducible essence of subjectivity", and if that irreducible essence did not exist then doubt would not exist either.

Again, this argument is completely reasonable as long as the referents of the word "consciousness" and the phrase "irreducible essence of subjectivity" are identical. But what it does not do is demonstrate that this is in fact the case.

"It is absurd for any conscious entity to doubt the existence of consciousness" is better, despite (or perhaps because of) the observable fact that conscious entities are well known to behave in completely absurd ways all the fucking time.

And again, I can see no reason at all to prefer a view of consciousness and/or subjectivity as necessarily requiring some kind of irreducible essence, to one that posits that consciousness and/or subjectivity might well be emergent behaviours of matter arranged in ways that we already know that matter can be arranged in that exhibit them, such as us, and quite probably in other ways as well.

Arguments for the irreducible essence that rest on difficulties in deciding exactly where consciousness might start along a scale of complexity from amoebae to humans are, to my way of thinking, nothing more than fancy-ass Sorites paradoxes and therefore not to be taken seriously.
posted by flabdablet at 4:28 AM on March 17, 2021

Kabbalistic panpsychism
posted by No Robots at 11:01 AM on March 17, 2021

I was looking for recent extensions or rebuttals in ITT, as a lot of the blog discussion linked here was 6-7 years ago, and found this article about efforts to empirically demonstrate IIT or Global Workspace Theory (GWT), a leading competitor. I'm not super aware of how technically good Quanta magazine is, but it seemed a fair overall gloss on the matter at hand, and some of the problems with this approach, which also took into account Artifice_Eternity's quoted points about making the distinction between the receptive machinery (eg, the human eye or a camera) and the handling of the experience itself.

GWT, to my mind initially, seems to handle things a lot better , at least in abstract - acknowledging the distinction between, say, an instinctual reaction to stimuli handled by the spinal column, and an act of creativity or emotional response. I think there are reductions to be made in terms of types of experience along those lines, and the level of reflection necessary for it, as they say, to be like something, that an owl would have and an electron or set of logic gates wouldn't. By isolating what constitutes this workspace as opposed to the machinery, at least for me, inutitively, it would be easier to come to a more specific understanding of what consciousness is.

That said, I wasn't previously aware of GWT and hope to explore it more fully. Thanks for everyone's considered posts above - much food for thought there.
posted by Sparx at 2:20 PM on March 17, 2021 [1 favorite]

Between Integrated Information Theory and Global Workspace Theory, I feel like there's a real need for theories of consciousness that don't sound like initiatives of a particularly dystopian HR department.
posted by Not A Thing at 3:07 PM on March 17, 2021 [10 favorites]

METAFILTER: theories of consciousness that don't sound like initiatives of a particularly dystopian HR department
posted by philip-random at 4:30 PM on March 17, 2021 [1 favorite]

So, if we're discounting human models ('tree', 'information'), why don't we consider the ego to be just another flawed human model? A helpful fiction for coordinating a vast array of interlocking physical phenomena?

It seems like if we don't trust humans to come up with good descriptive models of the world they inhabit, we should distrust the idea of consciousness at least as much as we distrust the concept of 'green', and maybe even as much as we distrust 'information theory.'

To finish reductio'ing the ad absurdum, If we don't allow any trust in any human models, we don't get to have a meaningful conversation. Any conclusion will itself be some linguistic model, and thus trash; we're left to go smoke some peyote on a ridgetop to further our understanding... But, hey, isn't that understanding just more trash?

When you say "regardless of whether or not there are people making those models", I read that as the opening of an argument about a hypothetical universe identical to this one except for the absence of conscious model makers. The trouble with that is that the resulting universe is hypothetical. That is, it's a model. And this gives it a property that reality of which we form a part doesn't have: an ability to be examined, and reasoned about, from outside itself.

But I argue that a world without observers isn't hypothetical at all, because it describes Earth before humanity, or before life, depending where you draw the line around 'observers.' We can see and interrogate evidence from that time long since passed to try to understand what it was like and what kind of muck we arose from. The proteins were folding long before there was anyone designing systems to predict how they would fold.

And why privilege shifting aeons? The unobserved universe suffuses us. I'm not currently doing anything in particular to observe The Ideological Construct I Call (TICIC) my left kidney, but it keeps on chugging along, regardless of the scope or quality of the model of it I possess. My model is very rough - it has something to do with being diabetic or not. Others have much more detailed models, and others still have even less detailed models than I have. But it keeps chugging along in the absence of direct observation. If we were to closely observe it, the models give a some ideas of what we would find. And if we never observe it, we know that it's capable of producing nasty surprises (stones! failures! other stuff beyond the limits of my personal model of the kidney!), providing evidence against a solipsistic universe.

When I own a company, I'm totally gonna name HR the 'department of kabbalistic panpsychism.'
posted by kaibutsu at 6:39 PM on March 17, 2021 [2 favorites]

So, if we're discounting human models ('tree', 'information'), why don't we consider the ego to be just another flawed human model? A helpful fiction for coordinating a vast array of interlocking physical phenomena?

I've been down this rabbit hole for bits of this morning, and this is very much a going assumption in some quarters. One metaphor GWT uses is the brain as a theatre, and consciousness a continuous spotlight that shines one on thing at a time, but the unconscious brain region activity and nervous system functions are the set dressing, the props, and the extras waiting in the wings - with neuronal activity between regions constantly juggling who's up next .

Consciousness is limited in its spotlight because it's useful to be - the brain doesn't know how its internal workings are behaving because a model of outside is far more useful to survival than one of inside since the days of single celled organisms, and focussing on one thing at a time also reaps practical rewards. But ultimately, what we experience consciously is just a cartoon or charicature of what we are fully sensorily experiencing, streamlined because its evolutionarily useful to be. The human consciousness distinction is thus the capacity, dexterity and integration of the various brain regions as they bring things into the spotlight.

I find myself quite liking GMT and its recent neuronal developments, as it demystifies and deprivileges consciousness yet maintains a model I can recognise in my own thought patterns, as opposed to IIT which just seems more nebulous, but I do have tendency to be impressed by the last thing I read, so take that with a grain of salt.
posted by Sparx at 7:21 PM on March 17, 2021 [2 favorites]

if we don't trust humans to come up with good descriptive models of the world they inhabit, we should distrust the idea of consciousness at least as much as we distrust the concept of 'green', and maybe even as much as we distrust 'information theory.'

I agree. The possibility that almost any idea could be incorrect is fundamental to my worldview. The only idea I'm actually sure is correct is that something's happening now, and the only thing I'm actually sure exists is the universe (aka everything; to some people aka God, though I personally don't favour that word because of its huge backwash of misleading connotations).

Every other idea, starting from the distinction between me and the rest of the universe and working on out, merits only degrees of confidence, not certainty.

Any conclusion will itself be some linguistic model, and thus trash

Now you're just being harsh! Many linguistic models are exceedingly useful. It just doesn't pay to accept them as certainties, is all, not even when they merit quite extreme degrees of confidence.

Earth before humanity, or before life

is part of a universe of which we are also parts, and therefore doesn't count, to my way of thinking, as a universe empty of conscious modellers; again, to my way of thinking any half decent universe has to include its whole past and future relative to any given here-and-now of reference.

I think the only way to get a universe empty of conscious modellers is to construct one hypothetically and then refuse to look inside.

If we were to closely observe [a kidney], the models give a some ideas of what we would find.

Yes. They also suggest useful ways in which we might choose to observe it.

If models were useless we wouldn't bother making them. So you'll never find me using dismissive language like "trash" or "illusory" or "side effects" or "mere" to describe them (except for mere trash illusory side effect ones like those that working magicians exploit professionally).

But what our models are is efficient compressions of experience. I'm a physics descriptivist, not a math prescriptivist. Reality will do whatever the fuck it wants^*, occasionally surprising us in ways to which the healthiest reaction is to re-examine our models. And these compressions have to be both efficient and lossy, because reality is huge and complex and our brains can only cope with limited amounts of detail.

Hell, the best and most detailed models we have are explicitly predicated on the assumption that reality will do whatever the fuck it wants however the fuck it wants, the most accurate predictions we can make about it always remaining statistical.

^*poetic licence applied for. No actual endorsement of reality as a thing with goals or desires or any kind of plan implied or intended.
posted by flabdablet at 7:53 PM on March 17, 2021

why don't we consider the ego to be just another flawed human model? A helpful fiction for coordinating a vast array of interlocking physical phenomena?

This is an extremely good description of the way my own ego models itself, for what it's worth.
posted by flabdablet at 8:01 PM on March 17, 2021

part of a universe of which we are also parts, and therefore doesn't count, to my way of thinking

Fwiw, my favorite 'get high and look at your hands' idea is that the past doesn't physically exist. We only ever interact with ephemera of the past, like recorded bits in memory, or light in the sky through which we observe the events that already occurred billions of years ago. But that act of observations happens in the present, by interacting with little shock waves rippling it from those prior events. So I tend to think of the universe as an instantaneous membrane moving through the time dimension. Sorry, Marty, there is no future to go back to. It's just all the same pieces in a different configuration.

So looking back on the time before humans, I see a universe in a slightly different configuration, running the same operating system, and doing just fine without us. There /was/ an observer-free universe until we had the bad fortune of showing up.
posted by kaibutsu at 12:59 AM on March 18, 2021

I tend to think of the universe as an instantaneous membrane moving through the time dimension.

That's a pretty model, and I used to use it myself, but I can't find a way to reconcile it with the implications of Special Relativity, let alone General Relativity, so I abandoned it some while ago.
posted by flabdablet at 3:00 AM on March 18, 2021

The way I think of it now is that my present is the 3D slice of the universe that's normal to my worldline right now.

This is impossible for me to (aphantasic version of visualize), so I (aphantasic version of diagram) it with one of the spatial dimensions suppressed, making that 3D slice into a 2D slice. Since all of my interactions with other things come down to interactions between parts of fields where causality propagates at lightspeed, I can be causally acted upon only by things right on the boundary of my past light cone, and can act causally only upon things right on the boundary of my future light cone.

Causal interaction with stuff outside the light cones just can't happen. Causal interactions with stuff inside the light cones - that is, with stuff moving slower than the speed of light relative to me, which is most ordinary stuff - clearly does happen. I reconcile this with the idea of causality propagating at lightspeed within a field by (aphantasic version of picturing) my past and future light cones as not so much cones as insanely complex crumplings bounded by cones. That crumpling, it seems to me, is what matter is, and causality propagates along its thoroughly crumpled (3D equivalent of surface) at lightspeed.

I do not have anywhere near the mathematical chops that would be needed to formalize this model in any way, but I like it all the same; mainly because it doesn't require my past or my future to be your past or your future, while still allowing for the possibility that if you and I communicate sufficiently well, we'd both be able to identify parts of such regions of our crumpled light cones as overlap that we'd agree (are, were or will be) actually the same things.
posted by flabdablet at 9:13 AM on March 18, 2021

why don't we consider the ego to be just another flawed human model? A helpful fiction for coordinating a vast array of interlocking physical phenomena?

This is an extremely good description of the way my own ego models itself, for what it's worth.

I suppose I should counter this with the kind of unverifiable testimony that can only annoy someone who's trying to ... verify something.

That is, some decades ago I had a particularly intense and vivid and memorable psychedelic experience wherein you might say my psyche more or less divided. Though divided feels the wrong word. What happened was my so-called ego dis-entangled itself from my so-called id. That is, I witnessed and profoundly felt a very real separation of the two. You might say the id was an ocean and the ego a surfer.

- the id being all the stuff of memory, knowledge, information, all the complex files and files-within-files that contain the details of what I'd come to think of as my self, except for ...

- the ego, which I'd call the part of my self that is conscious of not being just a vastly complex accumulation of memory-knowledge-information-etc, but rather simply ... is and is conscious of it. It reconciles all of the id stuff, dances in point and counterpoint with it, rides it as a surfer rides a wave, sometimes even aims for a particular shore.

Except that particular night, overloaded with LSD, you might say the surfer had a wipeout, got swallowed and pounded and tossed around by a monster wave, almost drowned.

I realize this anecdote proves nothing in any kind of scientific sense, works better as poetry or whatever. Except it did happen. I did have the experience. And I've discussed it with more than one person who has reported similar. And I've read stuff that suggests the same. So yeah, I would posit that considering the ego (and thus the ego/id split) "... to be just another flawed human model? A helpful fiction for coordinating a vast array of interlocking physical phenomena?" -- well that just doesn't hold up to my particular data -- it's falsifiable.
posted by philip-random at 9:20 AM on March 18, 2021 [1 favorite]

What happened was my so-called ego dis-entangled itself from my so-called id. That is, I witnessed and profoundly felt a very real separation of the two.

Yeah, I had something very like that with shrooms and weed after a particularly intense bout of dancing. In my case I got a more than two-way disentanglement, experiencing my internal operations as something akin to a committee meeting with various subsystems taking on various responsibilities and vying for control of the physical body's agenda. Profoundly interesting state.

simply ... is and is conscious of it. It reconciles all of the id stuff, dances in point and counterpoint with it, rides it as a surfer rides a wave, sometimes even aims for a particular shore

Nicely put.

In my own case, I'm pretty convinced that Chairman Surfer is only actually in here during those moments in which he remembers to check whether he is; in that act of checking, he creates another flash instance of himself. The result is a thoroughly convincing though actually stroboscopic illusion of continuity.

The committee did get a bit rowdy during his somewhat extended departure.
posted by flabdablet at 9:31 AM on March 18, 2021

I'd say in everyday functioning, the surfer just keeps working the waves more or less unconsciously, more concerned with where they're going than how -- rather how the average physically healthy human walks to the store and back, or rides a bike, working complexities of proprioception that they couldn't begin to explain, and if they suddenly stopped and tried to, they'd fall down or off the bike, or in the surfer's case, wipeout.
posted by philip-random at 9:37 AM on March 18, 2021 [1 favorite]

I like to think of the waves as music, part of which the surfer is conducting as well as riding on.
posted by flabdablet at 9:41 AM on March 18, 2021 [1 favorite]

For what it's worth, I asked a colleague of mine who's both a philosopher and a neuroscientist if he thought there's anything more to IIT than it seems on the surface, and he just laughed.
posted by biogeo at 11:18 PM on March 18, 2021 [2 favorites]

(Small update on my end: The magic search term for 'there is only now' is 'presentism,' which leads directly to the special relativity argument against it, along with (imo) a bunch of metaphysical gobbledygook arguments against it as well. I'm about 90% on board with the special relativity argument. The remaining 10% consists of the feeling that what I'm /really/ arguing is that eternalism lacks evidence, and that there may still be some interesting other path.

Specifically, the special relativity argument shows that there's no simultaneity, so 'hard' presentism, in which there is a single simultaneous moment, is wrong. Two observers of two events can't say in what order they actually happened, but they /can/ say that they happened over there, in the past. Lacking a time machine, the past is the set of time points that I don't have causal access to, and both observers can agree that all of the observed events occurred in the past, regardless of disagreements about ordering. This eventually settles into a question of whether you can in special relativity create chains of events with reversed causation from the perspectives of different observers... And to this, SR says no, you can't. So, I'll define soft presentism as the non-existence of the past, defined as the set of space-time points in the past light-cone of all points in the universe. Or maybe in the past light-cone of /any/ point in the universe... still thinking it through.)
posted by kaibutsu at 12:36 AM on March 19, 2021

While you're thinking it through, I strongly encourage you to spend at least a little time on pondering whether what you're actually doing is (a) working out your most useful referent for "universe" or (b) working out your most useful referent for "existence".

Personally I find it quite useful to have a single label to refer to the totality of all that was, all that is and all that will be, and I'm comfortable using "universe" for that. But if you can find a more appropriate word I'm all ears.

Likewise, I find it useful to have a single label to refer to that property of a thing that makes it possible for you and I to agree that we're both able to identify it, each able to specify a set of spacetime coordinates that locate it with respect to ourselves, and mutually able to resolve each other's coordinates for it by taking into account our coordinates for each other. I'm comfortable using "existence" for that, but I can understand how others might prefer to reserve "existence" to label the intersection of this property with timewise separation from themselves by zero.
posted by flabdablet at 1:15 AM on March 19, 2021

I'll define soft presentism as the non-existence of the past, defined as the set of space-time points in the past light-cone of all points in the universe

Do those points have to exist, i.e. not be included in your own past light-cone, in order for the contents of their own past light cones to be removed from the set of things that also exist?

What does this version of soft presentism have to say about the contents of your future light-cone?

Does it admit of anything like degrees of existence, depending how far the candidate things are from the (3D equivalent of a) surface midway between your past and future light-cones?
posted by flabdablet at 2:15 AM on March 19, 2021

I'm admittedly a bit preoccupied with the basic physical questions here. One of the things that people don't like about the many-worlds hypothesis is that it requires the basically-infinite duplication of all the energy/matter/other conserved quantities in the universe on a continual basis. The eternalist position has a similar problem, on a smaller scale: all the conserved quantities need to be duplicated endlessly into the future, which strikes me as a bit too magical...

I feel like a good final answer will be along the lines of 'the universe is a bunch of causally-interlinked stuff with a conserved mass as time progresses.' So for the existence question, we have two separate questions of whether Marie Curie can be referred to (linguitic-philosophical: sure, we do it all the time), and the physical question of where's the mass? (for which the answer is, the mass is scattered around at this point, and prooooobably not what we would refer to as Marie Curie anymore, so 'no.')
posted by kaibutsu at 2:33 AM on March 19, 2021

all the conserved quantities need to be duplicated endlessly into the future, which strikes me as a bit too magical

If I have an apple in my hand, and its mass is 0.1kg, and its mass ten seconds later is still 0.1kg, then its mass is conserved regardless of any opinion I might hold about its past or future existence before, during or after this experiment.

If I were to consider an eternalist model for the apple I could perhaps integrate its mass along its worldline to get a number that might be useful for comparing it with other things. Perhaps I could call this a measure of "substance" until somebody else came along to tell me I was using that word all wrong. But whatever I called it, that measurement wouldn't be a mass; it would be in units of kilogram seconds, not kilograms.

Mass remains a measurement that applies to any given object at a particular point in time, regardless of anybody's opinion about that object's state of existence at other points in time. And the same applies to any of the other physical properties we might wish to assume are conserved.

In fact the concept of conservation depends on time in order to be a coherent idea in the first place. It's simply not an idea that's applicable without having made a local split of spacetime into space and time for perceptual convenience reasons. Which is, of course, exactly why General Relativity (let alone many-worlds) can do such violence to assorted conservation laws (including conservation of energy) without apparent objection from respectable physicists.
posted by flabdablet at 5:00 AM on March 19, 2021

a single label to refer to the totality of all that was, all that is and all that will be, and I'm comfortable using "universe" for that. But if you can find a more appropriate word I'm all ears.

I find I just go with "everything", which goes back to another acid trip. Two friends arguing about a definition of reality, getting really quite heated about it. At some point, I concluded, "Well obviously it's everything. Everything that's ever happened, and is currently happening, and everything that everyone has ever thought about it, felt about it, done about it, is thinking about it, is feeling about it, is doing about it. There really is a whole lotta everything." Or something like that.

Notice, the future doesn't figure in this everything (except insofar as it can be imagined, used as a concept). I guess that puts me in the "future doesn't exist" crowd, because well, this puts it rather well:

One of the things that people don't like about the many-worlds hypothesis is that it requires the basically-infinite duplication of all the energy/matter/other conserved quantities in the universe on a continual basis. The eternalist position has a similar problem, on a smaller scale: all the conserved quantities need to be duplicated endlessly into the future, which strikes me as a bit too magical...

Because for the future to be as real as the now (and the previous), the many-worlds hypothesis would have to be so. Wouldn't it? Because if the future is, in effect, already set, then say goodbye to free will. Which I realize is definitely a thing these days. I just don't happen to buy it.
posted by philip-random at 8:04 AM on March 19, 2021

I find I just go with "everything"

Main reason I rejected "everything" is the same reason I rejected "God". Didn't like the connotation baggage.

To me, "everything" connotes some kind of collection: everything is all the things. And to me, that makes "everything" seem like the things are primary and the "everything" collection refers to some kind of construct that needs to be built up from those, like "the composite of myself and Jupiter" only worse.

The idea I want to label is the idea of a thing to which all other things, myself included, stand in a part-to-whole relationship. This thing has the distinct advantage that defining it requires attaching only a single attribute to it - that of existence - which can be done by direct inspection of experience. It's not necessary to specify or understand or comprehend what kind of experience one is having in order to do this.

My preference here is motivated partly by my having personally experienced states of awareness where the prospect of distinguishing one thing from another is past ludicrous. It's nice to have a metaphysics with a foundation that still works in a state like that. So I'd rather use a word that connotes the thing I label "universe" as primary, and its parts derived by the cognitive act of making distinctions, rather than the other way around.

This is. Sometimes, that's all a body needs to know.

for the future to be as real as the now (and the previous), the many-worlds hypothesis would have to be so. Wouldn't it? Because if the future is, in effect, already set, then say goodbye to free will.

I don't think of it that way. I can't see how an event that I can choose to make happen will fail to happen solely because I chose it. I can have all the free will I like, but that doesn't change the fact that actually exercising that will is always going to make something happen.

It seems to me that things that will be have far more justification for being labelled as existent than those that won't be, regardless of whether or not my free will is a factor in why they will be or won't be and regardless of the extent to which they're predictable or identifiable beforehand. Facts first, causal analysis second is my metaphysical motto.

I also note that the word "free" is doing possibly even more heavy lifting in the phrase "free will" than it does in "free speech". In my experience, absolute freedom is not a thing. I can't choose to be a hundred kilometres west of here a millisecond from now, regardless of how desirable that outcome might seem.
posted by flabdablet at 9:25 AM on March 19, 2021

Well, another free-will friendly alternative to MWI is to accept quantum uncertainty and allow that you don't know what's up with any cat that you've intentionally causally isolated yourself from... This doesn't seem at all unreasonable from where I sit.

> kilogram-seconds

Agreed; integrating mass over time is different from measuring instantaneous mass.
But that doesn't fix the problem: my straw-man eternalist is Robert Zemeckis, who says the past is a physical place you can visit if you have a cool enough car or sufficiently complicated system of wormholes. That means that the total (instantaneous) universal mass is duplicated constantly - all of the mass that existed at Nov 12, 1955 at 6:38am is still there.

a) This seems equally miraculous to the idea that the universe exists at all, along with being suspiciously unobservable.

b) Creating a 'new' observation of this 'old' matter already has some weird implications: I need to send a photon back to 1955 and bounce it off of something. This breaks causal sequencing, since the photon I've sent has a physical interaction with the past and could cause a butterfly to flap its wings or something. It seems that we need a Zemeckis time machine in order to prove physical persistence in the first place.

c) Physical persistence of the past creates another way in which time is very different from spatial dimensions: the thrown ball doesn't leave a worm of copies behind it as it moves through any other dimension. (though this is a weaker objection, since time is already weirdly different from the other dimensions.)
posted by kaibutsu at 10:23 AM on March 19, 2021

all of the mass that existed at Nov 12, 1955 at 6:38am is still there

Sure. Still there, in Nov 12, 1955 at 6:38am. It's not here, getting drunk in dive bars and throwing its weight around. I have no problem with it as long as it stays where it is. I've never heard of an existing (as opposed to fictional) phenomenon whose explanation would require that some instantaneous mass or other is failing to do that.

Creating a 'new' observation of this 'old' matter already has some weird implications

Just as well, then, that that's something we only know how to do for objects with far larger spacelike separation from us.

the thrown ball doesn't leave a worm of copies behind it as it moves through any other dimension

Quite reasonable to think of it as a a single object with extension along a worldline as well as length, width and height, though, as opposed to a collection of 3D movie-frame analogs. There are no "copies", there is only the 4D ball. Difficulties in visualizing this can generally be dealt with by temporarily deleting length, width or height but there's no particular need to do that if you're conceptually comfortable with it.
posted by flabdablet at 10:42 AM on March 19, 2021

Hulk head hurt now.
posted by y2karl at 7:13 PM on March 21, 2021

Is consciousness everywhere?

No.

Consciousness is here and now.
posted by flabdablet at 8:55 PM on March 21, 2021 [1 favorite]

...and then.
posted by y2karl at 1:07 AM on March 22, 2021

Now and then.
posted by flabdablet at 3:13 AM on March 22, 2021

I thought therefore I am?
posted by fullerine at 4:16 AM on March 22, 2021 [1 favorite]

You are, but what was I?
posted by flabdablet at 5:50 AM on March 22, 2021

Schroedinger's cat box.
posted by y2karl at 2:44 PM on March 22, 2021 [1 favorite]

OP just stopping back here to say, thanks to everyone for a great week-long discussion, one that will be well worth revisiting from time to time.
posted by beagle at 5:53 PM on March 22, 2021

Visualizing Time Dilation
posted by flabdablet at 1:54 AM on March 28, 2021

Many people confuse consciousness with ego, particularly their own exaggerated ego. From this perspective, it would seem obvious that consciousness is unique to mankind, and perhaps even then only to a select few.
posted by No Robots at 9:58 AM on March 29, 2021 [1 favorite]