Does the same apply to qualia? If so, if I touch a hot stove and it hurts so I pull it back, the Cliff notes of it hurting and thus pulling my hand back wouldn’t be needed to account for the pre-emergent (vastly more complex) account dealing with a trained neural network made of synapses, dendrites, axons, etc. Assuming you agree, what causal role does qualia play in our actions if it is just a Cliff notes version of what is actually happening at the neural net level?
That’s a lot to absorb, and I’m not sure I see the relevance. Perhaps you could take another shot at the question, “Can there ever be a truly basic description we can devise for anything?” I think you may be construing the concept of a “basic description” as more like a description of Kantian noumena, which wasn’t what I had in mind – not that basic! But I may not be understanding you.
1 Like
Taking from simulation theory, any description we devise is bound by what our local biological senses can pick up. You can create a basic relative description of the mechanisms for what our senses can pickup until you hit a limit. There are no known ways around this that are not being actively worked on at the time of this writing, which is us, trying to reinvent the wheel and hack our way out of what life is telling us represents our physical boundary. The brain in a vat thought experiment is no exception to this principle.
Does this help?
I don’t know what role qualia might play. I’d expect qualia to be multiple emergent levels up from the relatively ‘ground up’ discussion of intentionality which I have been having with Jay.
Supposing I do agree with all you say here, do you have an argument you want to make? As it is, it’s starting to seem as if you are conducting an interrogation, and it’s getting old.
Good, so we’re talking about “basic” within the limits of what we, as sensing creatures, can experience. That’s as basic as I’d intended to get; whether there’s a capitalized Basic Description of some X, beyond what we can know of it, is idle.
So, within our human world, do you think some items do have basic descriptions along the lines of the acoustical example I gave? The argument I was lining out in the OP depends on this being the case, in order to contrast it with what we find when we look at the “dual descriptions” of the mental and the physical, which I claim do not have such a basic description. What do you think?
We’re talking past each other. There’s no point. You’re not the original poster so I’m not sure why we’re even going over this.
The thought that occurs to me is that this is vulnerable to a more sophisticated form of Putnam’s ‘multiple realisability’ argument. Originally, this argument started with the observation that pain could be realised by totally different physical substrates in organisms of different kinds, yet still retain its identity as ‘pain’. So, the reasoning went, how could you establish the identity between the felt characteristics of pain — which is an unarguable fact of existence!—and the diverse physical interaction which give rise to the sensation.
I’ve long held that this basic idea can be extended. ‘My cup is yellow’ is a factual statement in English. But the same information can be represented in many languages, not to mention in binary code or some other symbolic form. It might also be written on paper, engraved in metal, or displayed on a computer screen. In all these cases, both the symbolic form and the physical medium are different - yet the information remains the same. This challenges what might be called ‘the equivalence thesis’ that is implied in the original statement.
When it comes to neuron-firing there’s another factor: the problem of localisation. We can identify some neural activity that correlates with particular perceptual or cognitive state, but that is not the same thing as identifying a specific, discrete neural pattern with the thought itself. Cognitive activity is distributed across networks, dynamically integrated, and capable of being realised through different configurations of the underlying neurophysiology. So even if we grant that “My cup is yellow” might be instantiated in neural activity, it doesn’t follow that the thought is identical with a specific pattern of neuron-firing. It might be instantiated in different neural patterns on every occasion that it is uttered.
And that seems to me to compound the multiple-realisability problem. “My cup is yellow” is information with semantic content. As noted, the same content can be expressed in English, French, binary code, Morse, writing, speech, or an inscription in metal. The physical realisation changes while the informational content remains invariant. Why should the brain be different in principle? Why should a particular pattern of neuronal activity be identical with the thought, rather than being one particular physical realisation of it?
And finally, we appeal to the rules of valid inference, and so on, in making all these claims and seeking these explanations. Even to understand if there is an equivalence between these rules and some putative ‘neural event’ requires constant appeal to those ‘rules of thought’. So how could we expect to find the basis of such judgements in the phenomena at issue in this debate?
So in summary, attempting to ground or justify the factual content of such a proposition in ‘neural states’ will inevitably involve a kind of circular reasoning, as we are relying on the laws of rational inference and so forth, to even elaborate what is meant by a ‘neural explanation’ and to interpret the evidence for it.
Nevermind I mixed up you and wayferer. My apologies.
If an argument is intended to function as a viable general proof and stay internallly consistent, it cannot selectively discard mainstream methodological frameworks right out of the gate. This is my opinion.
When viewed strictly through the analytical lens you provided though, the musical analogy breaks down because demanding a basic physical description for a mental event assumes that all valid descriptive frameworks must ultimately converge on a single truth for monism to be valid. That being said, this brings me back to the thread that was created on how consciousness tunes different truths based on whatever based on you’re anologies.
From a bottom up perspective, yes items do have basic descriptions and overall I do believe this is true what you are pointing out. I do disagree however with your conclusions that the hard problem is opaque to us based on hereandabove.
Sure, getting from the “basic description” issue to a conclusion about the Hard Problem isn’t straightforward. I see them as facets of the same problem – the same ignorance, really – but I’ll be interested to see how others respond on this thread.
Very suggestive. Is a single occasion good enough to create token-identity, one of many possible multiple realizations? And is that, in turn, good enough to justify using language like “the thought is identical with a specific pattern”? I don’t have a strong argument here, but my inclination is to say 1) yes, a single occasion is good enough to establish token-identity (after all, that’s what a token is), but 2) using the "thought is identical to . . . " language isn’t justified by token-identity, no matter how that’s understood.
Another way to capture the shakiness of the “identical with” language.
We can’t. That is an extremely important issue – but not quite the one being raised here, I think. It may well be the case that a reduction of the rules of inference to a series of “neural events” would undermine the very justification for engaging in such a process. I think it would, and hence the dual functional descriptions of “reason” and “cause” are ineliminable. Davidson is right about this: The neural-event description is not more basic.
Another way to say it: "attempting to ground or justify the factual content of . . . a proposition in ‘neural states’ " will never work, precisely because neural states don’t “do” justification – it’s the whole Space of Reasons/category mistake issue raised earlier. At best, the neural events might provide a causal explanation, but we can’t accept such an explanation as final as regards human behavior without giving up the whole concept of “accepting an explanation” – which isn’t causal, but depends on reasons!
But we can grant all that and still ask, What makes the two descriptions “of the same thing”? If we reject a physicalist reduction, we haven’t solved the problem – we’ve only rejected one answer. Likewise, we can posit an invariant semantic-level description and (many) physical-level descriptions, but we still want to know what makes them descriptions of “the same thing.” We can’t even abandon language and point to that same thing, in this case! What would we point to?!
Nice job on the OP.
The music analogy is an interesting set up, and I think the analogy is actually better than you give it credit for. You said the analogy breaks because, whereas a chord has a more basic description in terms of Hz, thoughts do not have a comparably basic physical description.
But I don’t think that’s quite right. Hertz is ontologically/materially basic, but it’s not descriptively sufficient. You can’t derive the harmonic function of “dominant seventh” from the frequencies, because harmonic functions are functional/normative statuses within a musical practice.
So, I’d argue that music actually does have the Davidsonian structure. You have two levels of description – physical (Hz) and functional/normative (harmonic functions) – and the latter cannot be derived/deduced from the former.
Likewise with brain/thoughts you have two levels of description – physical (brain) and functional/normative (thoughts) – and the latter cannot be derived/deduced from the former.
Both harmonic functions and thoughts are defined holistically in terms of their role within a normative practice in a way that can’t be captured in terms of law-like relations with their physical substrates. And neither can be replaced by descriptions of their substrates and still function within their respective practices.
But I don’t think this is the hard problem. You’re identifying a gap between the normative and physical, but the hard problem concerns the gap between phenomenal and everything else. Chalmers would say that the former gap is one of the “easy” problems. Only the latter gap truly qualifies as “hard”.
Also, you asked how we could know that p and q describe the same thing without a third neutral description of that thing. I believe Davidson’s answer is that the identity is a conclusion rather than a starting point.
The reasoning goes: since mental events can cause physical events, causality falls under strict laws and strict laws are physical, mental events must have a physical description. So we have one event with two descriptions, individuated by space and time (e.g. “thought by me at the specified place and time”). The individuation isn’t a third description, it’s what identifies and differentiates individual events.
That said, I think you have identified a genuine pressure point for Davidson. He gives two descriptions of one event and relates them through identity, supervenience, and anomalism, which opens the door to Kim’s causal-exclusion argument. Sure, Davidson can say that one event has both a physical and mental description, but if the causality is explained entirely by the physical descriptions then the mental starts to look epiphenomenal. It’s not clear to me that Davidson has an adequate answer, but perhaps someone more familiar with Davidson can pipe in.
Personally I prefer an Aristotelian-adjacent approach over Davidson’s. It’s not two descriptions of one event, but matter and form of one thing. The form is the intelligibile organization of the matter, and the matter at one level is already informed at the level below. “Basic” then means “no form without its matter”. But it doesn’t mean the form is derivable/deducible from its matter.
The soft spot in this approach is that form has to be causally efficacious in some sense. Higher level organization has to constrain what happens below rather than just redescribe it. That’s the emergence debate, and here I might look to authors like Juarrero, Deacon or Mossio/Moreno for support. I think a compelling case can be made, but I acknowledge that that debate is far from settled.
1 Like
That’s fine. I was curious how people would answer the questions, but I didn’t mean to sound like it was an interrogation or engaging in trolling.
I have not read the whole OP yet. I hope what I’m posting now applies. I wanted to get this idea down before I get distracted.
Obviously, this is coming from my beliefs about consciousness, which are not shared by anyone else that I know of. But here it is anyway.
We can describe vibrating piano strings in two ways. 1) The frequency at which each string vibrates; when each string is vibrating and when each is not; which strings are vibrating simultaneously; etc. 2) Chord progressions.
We can describe brain states/neuronal firings in two ways. 1) Ion gated channels transporting ions through the membranes of neurons, the build up of ions/electrical charge; molecules called neurotransmitters being released by one neuron and received by another; etc. 2) Ideas.
The second description of the vibrating strings has nothing to do with the first. They are not at all the same thing. Music is a different category of thing (I don’t know if that’s the right word) than changing frequencies of strings. Music would not exist if consciousness did not exist. Music only exists in consciousness. The vibrating strings can vibrate forever, in any order, combinations, or patterns, and there are no chord progressions, there is no music, if it is not assigned to them, if it is not created by, consciousness.
The second description of brain states/neuronal firings has nothing to do with the first. Again, we are talking about different categories of things. In Incomplete Nature, Terrence Deacon ponders “how a pattern of neural activity can be about something that it is not.” This is, of course, the Hard Problem of Consciousness. The vibrating strings are only about music because consciousness assigns that meaning to them. How is the meaning of brain states assigned to them? How is the very possibility of assigning meaning achieved?
I have used an analogy with vision when talking about consciousness. The consciousness of different things is not different. Not different kinds of consciousness, and not different degrees of consciousness. There’s no such thing as higher consciousness. The differences and details are in the nature of the things experiencing their own existence. Think of consciousness like vision. I can look at a blank sheet of paper. I can look at the Grand Canyon. I can look at my wife. I can look at a Monet painting. I can look at a bolt of lightning racing across the sky. I can look at a blade of grass. My vision does not change depending on what I’m looking at. The things being looked at are what’s different.
It’s the same situation with consciousness. When we compare two different things, it’s not the consciousness of the things that’s different; it’s the natures of the things that are conscious that are different. Take you and a dog. Your consciousness is not different than the dog’s consciousness. Rather, you experience your nature, and the dog experiences its nature. Of course the end results are different.
We do the best we can. Both of our analogies fail, because both hearing and vision only exist because of consciousness. Without consciousness, we only have chains and webs of physical events that begin with photons hitting retina and vibrations in the air hitting eardrum. There isn’t an analogy for consciousness.
Great response, thank you.
This is true, but what you can derive is the “building-block description” I used to introduce the chord: three specified intervals built one on top of the other. This is not a functional description, but nor is it acoustical. It is pasting a different label on the relation “261.63 Hz - 329.63 Hz” (to pick the 1st interval of a chord based on C) – we now agree to call this relation a “major 3rd”. There isn’t a rule that can tell us whether this is a “derivation” or not, but I think it is, and I agree with you that the further step, into functional harmony, is not.
One more thought about why I would call the relabeling a derivation from the acoustic description: We understand it; we have no difficulty seeing why one would “translate” into the other, or how such translations are effected by a simpler labeling system. None of this could be said of the next move into functional harmony – and of course it also can’t be said of the “translation” (?) from physical to mental.
So with the addition of the “building-block” description above, we have three levels: physical (Hz); a simplified relabeling of the physical (intervallic relations); and the functional (harmonic theory). I’m suggesting that the first two aren’t actually two different levels: Both are physical descriptions, in the sense that if someone asked, “What is a major 3rd built on middle C?” a complete answer would be given by showing the Hz numbers.
So this is kind of housekeeping, because after tidying up we’re still left with the situation you describe: two levels of description that are anomalous in the Davidsonian sense. I agree. But when I said that the analogy breaks down, I didn’t mean it breaks down here.
My beef is about the term “basic description.” I’m agreeing with Davidson that there is no basic description for the mental/physical nexus; neither way of talking about it is more basic than the other. And I’m saying that this is unlike music, where the acoustic/physical description is the one we must have in order to make sense of the functional description. Granted, we can’t derive functional harmony from acoustical description; but we know exactly why functional harmony requires acoustical description, and why the reverse is not true. This, I’m claiming, is the big difference when we turn to physical and mental descriptions. We don’t know – and many/most of us wouldn’t agree – that we require a neuron-firing description in order to get to a “thought-level” description.
Yes, this is what follows.
This is really interesting. Yes, the hard problem is “bigger” than the normative/physical gap. But what I’m suggesting is that the reason we have difficulty understanding the normative/physical gap is that we don’t have an answer to the hard problem. In both cases, we’re stymied by the same conceptual set-up, which we perceive to be inadequate but do not know how to expand or replace. This is absolutely speculative, but I would bet that if/when we understand the relation of what we now call “the physical” and “the mental”, we’ll find that we also understand how meaning/intentionality/normativity should be conceived. I don’t know what Chalmers might say about this, but I don’t think the latter is a separate “easy” problem
Yes, that’s how I read him too. But regardless of whether “same-thingness” is a posit or a conclusion, my problem remains. “One event” or “same-thingness” is empty, without some description. It’s all very well to say that what we describe in physical language, and then in mental language, must be “the same thing” or “a single event.” But does this help us understand it? Not at all, I would say. Describing a chord in terms of what I’m claiming is its basic description, acoustics, does help us understand why harmonic functions describe the same events. Nothing like that happens here.
Yes. This is an important, if separate, reason to push back on anomalous monism. Or maybe not so separate, since once again, if we knew what to say about the role that the mental plays . . . but we don’t.
Ahah. Yes. We might say: higher-level organization can’t be epiphenomenal. Since you’re the second person to mention Deacon on this thread, I’d better read him.
1 Like
If the central subject is dual descriptions, then it boils down to whether there is a perfect language. A language which cannot be reduced to something base or closer to base, and can perfectly describe any phenomena. Is there a (possible) language that can describe perfectly concisely and perfectly informatively? Can it describe phenomena on all levels of abstraction?
Thanks for these good ideas!
I wonder if this “creation by consciousness” applies to anything with a pattern. I think that’s what you mean? – that what we call “music” is perceived patterns, and that it takes a perceiver to bring these patterns into being. This immediately reminds us of the controversy about math and whether it is discovered or created. In one sense, the patterns of music and math are already there, waiting to be identified, discovered. But some, such as yourself, prefer to think of a pattern as something that must be created, brought into being.
I’ve never been sure how much difference it makes (is it just a question of how to use the word “pattern”?), but you’re certainly right to raise the issue here.
Yes. It’s not clear whether that’s even a coherent question. If we pair a brain state with a thought, have we then “assigned a meaning” to the brain state?
Might this apply to brain states and thoughts? I can become conscious of a particular neuron-firing pattern (at least theoretically), and also become conscious of the thought that accompanies it. Are you suggesting it might be the case that these “things being looked at” are indeed different, and so my unchanging human consciousness would see them differently? It’s a way out of the “same thing” box, that’s for sure.
I don’t think you quite understand me. When we compose music, yes, it generally has patterns. Bceause that’s how we compose music according to the tonality of the common practice period. But I am more talking about the beginnings of music. How many millennia ago did anyone first think of the noises as music? Probably in different parts of the world independently. Someone hunting with a bow and arrow noticed the difference between his bow string’s pitch and that of someone else in his group. Maybe in another part of the world someone heard a rock fall onto a hollow log, and the sound struck him differently than any other sound had ever struck anyone before. Listening to birds could certainly apply. I’ve heard it said that the first person to hear any of these things as music should be considered a natural musical talent greater than Mozart. I can’t argue with that.
And we can consider John Cage’s 4’33".
None of this is music without a conscious listener. No piece by Bach, Louis Armstrong, or The Beatles is music if there is not a conscious listener. Not just because they were composed to be music, but because the vibrations in the air must be interpreted as, given the meaning of, music every single time, or they are just vibrations in the air.
Right. Seems like the bootstraps issue to me.
I don’t see how, even theoretically. Is anyone ever aware of their neurons firing? I recently posted about how I think consciousness is causal. I mentioned that, using the example of writing with a pencil, we never think about moving particular nerves, or muscles, much less creating action potential. I think we are only aware of the thoughts and emotions that accompany the neuron-firing patterns.
Yes. But it’s difficult to make the comparison I want, because, as I said, there’s really no good analogy for consciousness. My vision is one conscious entity experiencing the different groups of photons hitting the retina.
To compare consciousness with vision, maybe atman is the way to go. Atman is Brahman. The atman known as Patterner is in no way different from the atman known as Jay. Your consciousness is not different from my consciousness, because we are both the same Brahman. Rather, Brahman experiences your physical makeup, events, memories, sensory input, etc., and it experiences my physical makeup, events, memories, sensory input, etc. What is experiencing is not different. What is being experienced is what is different.
I can’t argue against the Brahman/atman idea. I just don’t happen to believe it. I believe consciousness is a fundamental property, or capacity, of reality. I guess it plays out the same while we’re alive.
Ah, I see. Yes I agree, but with the same qualification that you already mentioned. Interval size is a relabeling of a frequency ratio, but interval spelling isn’t. For instance, minor 3rd versus augmented 2nd is the same ratio, and choosing between them is already taking into account their harmonic function. But I’m being nit-picky.
I just want to make sure I’m following, because “must have” can be read in two ways:
The first reading is that there can be no harmonic function without a sounding. That’s clearly true. But the parallel claim, that there can be no thought without neural activity, is one most people would accept, Davidson included. That’s what his token identity is for.
The second reading is that we need the acoustic description to make sense of the harmonic one. But it seems like your own OP tells against that in the case of music. I’m not totally up-to-speed on the history of music theory, but my guess is that people understood what a dominant seventh was long before anyone new about the physics of sound. What they needed was the “sounding”, not the physical/acoustic vocabulary. In that respect music and mind end up looking alot alike: we’re constantly making sense of thoughts without any reference to neural descriptions.
So I wonder whether the difference you’re pointing to is actually a third thing. In music, once we do have the acoustic description, we can see why the physics is suited to the functional distinctions that musicians make, even though they don’t give us the norms of tonal harmony.
We don’t yet have anything comparable to that in the case of neurons and thought: there’s no account of why this kind of activity is suited to carry that kind of content. If that’s your point, I agree, and it’s a real disanalogy. But it’s a difference in how well we understand the fit between matter and function, not a difference in which description is basic.
Is that close to what you were getting at?
I am aware of a few thinkers who would agree with you about the close connection between phenomenal consciousness and intentionality (Horgan and Tienson, Loar, Kriegel). If they’re right, then it would make Chalmers’s split between “easy” and “hard” a bit misleading. To be fair, Chalmers has himself expressed some sympathy toward a close link between experience and representation, though he never ties them together via argument as far as I know.
So I agree with you about the shared set-up. Both gaps really do look unbridgeable when “the physical” is defined as whatever is left over once purposes, norms, and qualities have been removed from nature. So if that’s the starting point, then I would agree that neither meaning nor experience can be recovered from it.
However, I still think they pull apart. We can already see, at least in outline, how physical organization yields things being “for” a system. A self-maintaining organism has conditions that it can succeed or fail at meeting. That’s basically a standard by which things can either go well or badly for it, which is a foothold for normativity.
Building up from that to full rational normativity is a long road, but seems at least plausible to me. The further question of what organization is sufficient for phenomenal presence-to-a-subject is something that I don’t think anyone has a convincing account of yet. So while I’d agree the two problems have a common source, I don’t think solving the first automatically solves the second. Your bet goes the other way, which is fair, because I don’t think it can be settled right now.
That’s fair. Davidson’s identity is a thin metaphysical claim rather than a full explanation. And if your complaint is that it leaves the relation between mind and physics unintelligible, then I think that’s the right complaint, and it actually connects to the musical point above. In the case of music, we can understand the relation between the sounding and harmonic function. And that understanding actually enriches our understanding of harmonic function itself. What’s missing for mind is that kind of understanding. As I mentioned before, the hylomorphic approach is an attempt to supply that understanding, though whether it succeeds is obviously still very much an open question.
If you’re looking to read Deacon, Incomplete Nature (2011) is the one I’d recommend (although Symbolic Species is excellent too). But be prepared: it’s dense and meandering. If you’re looking for a lighter treatment, Juarrero’s Dynamics in Action (1999) is shorter and more direct. Moreno and Mossio’s Biological Autonomy (2015) is probably the most technically demanding of the three.
I certainly didn’t, sorry. Thanks for clarifyihg.
So naturally I have to ask: What does it mean to you to “think of (or hear) the noises as music”? I don’t have an answer at the ready – I’ve often puzzled over this and am not sure.
No, I agree that’s unlikely. What I had in mind by “become conscious of a particular neuron-firing pattern” was learning a description of such a pattern in the same way we might learn that the Hz numbers were a description of a chord. Neither the pattern nor the Hz numbers can be distinguished consciously under that description, but they represent possible objects of cognition. I wondered whether you might consider the neural pattern and the thought as “different things”, therefore, in the same way that the lightning and the blade of grass are different things.
Yes, that’s one of my takeaways from all these many discussions.
I’m kind of in the same boat. It may turn out to be good way to conceptualize consciousness, but of course we don’t know yet . . .
That’s… leaving out the theory that goes into creating types. So, yes - the htz-description would be a description of the token, but it would not include the information that makes it a token.
In olden times, sheet music often precedes the performence, but in recorded music sheet music is optional: people often just play chords and sing, so to speak. That is, the experts can then discuss what types these tokens are tokens of.
So if you have the hertz-description of the chord, the chord could be either a token of a dominant chord of an augmented 6th chord as you. These are competing analysis experts actually fight over sometimes.
So by saying the thought is token identical to the brain patterns, we’re not saying much, if you don’t have anything to talk about in what both things are supposed to be a token of.
There is then also the problem that your thought is only accessible to you, while the brainpattern (if recorded) is accessible to both of us. If we could somehow record experiences, and replay them in other brains we’ve created a type of the token: we assume that what I am experiencing is your experience, but by making that assumption we’ve now created a type “your experience” and two tokens (your original and the replayed recording in my brain). That makes the token-status interesting. Both the sounds, the transcription, and the acoustic analysis are available to all interested parties. Your thought (as the actually in situ token) is only accessible to you. Every other representation is a different token, and by being a token of the “same thing” we are already making a theoretical assumption. The token-status of your thought, then, is in itself theoretically interesting in that it cannot be shared as a token, only as a type via a range of tokens assumed to be under the type.
And that assumption is already a move in theory game to begin with. Calling your thought a token feels like a rhetorical move, but it might not be one. It’s interesting.