Illusionism and functionalism

Mind and AI 6 Minutes

In the last thread, someone asked what exactly is it about consciousness that illusionists say is illusory?

One quick answer is that for illusionists, the properties people see in experience that incline us to think that consciousness is a metaphysically hard problem, are what’s illusory. In weak illusionism, the properties aren’t what they seem. In the strong version, which is usually what “illusionism” refers to, they don’t exist at all. But what exactly are these properties?

I’m a functionalist, someone who sees conscious experiences, and mental states overall, as more about what they do, the causal roles they play, than about any particular substance or constitution. It’s a view that I think provides a necessary explanatory layer between the mental and the physical, and so sees no barrier in principle to a full understanding of the relationship between them.

The usual argument against functionalism is that it doesn’t seem to account for qualia, the properties of phenomenal consciousness, the “what it’s like” nature of subjective experience, such as the redness of a red apple or the painfulness of toothache. Most functionalists, if they use the term, argue that qualia can be described functionally, such as pain being an automatic evaluation of a problem with a part of the body.

However philosophers have a number of thought experiments which claim to show that qualia and physics, including functionality, can be separated. This is where the illusionists come in. They argue qualia don’t exist, that the illusion is our impression that they do.

But that raises the question. What exactly are qualia? I gave the standard definition above, but it seems inadequate to settle this debate. The SEP article on qualia discusses four different versions, the simplest of which might be compatible with functionalism, but others that aren’t.

Daniel Dennett, in his 1988 Quining Qualia paper, a famous attack on the concept of qualia, provides the illusionist understanding, by noting four attributes commonly assigned to them. Summed up in the qualia Wikipedia article, they are:

  1. ineffable – they cannot be communicated, or apprehended by any means other than direct experience.
  2. intrinsic – they are non-relational properties, which do not change depending on the experience’s relation to other things.
  3. private – all interpersonal comparisons of qualia are systematically impossible.
  4. directly or immediately apprehensible by consciousness – to experience a quale is to know one experiences a quale, and to know all there is to know about that quale.

Dennett’s description of qualia is often decried as a strawman, something he constructs to easily knock down. However, we only have to look at the most popular qualia thought experiments to see these attributes confirmed. For example, consider Frank Jackson’s knowledge argument as described through the Mary’s Room thought experiment.

Mary is a brilliant scientist who is, for whatever reason, forced to investigate the world from a black-and-white room via a black-and-white television monitor. She specializes in the neurophysiology of vision and acquires all the physical information there is to obtain about what goes on when we see ripe tomatoes or the sky and use terms like “red”, “blue”, and so on. She discovers, for example, just which wavelength combinations from the sky stimulate the retina, and exactly how this produces via the central nervous system the contraction of the vocal cords and expulsion of air from the lungs that results in the uttering of the sentence “The sky is blue.” What happens when Mary is released from her black-and-white room or is given a color television monitor? Does she learn anything new or not? Jackson claims that she does.

Within the assumptions of this scenario, why is Mary unable to acquire the information she can only learn by having the experience? She supposedly can’t read descriptions of it, because no one can provide that, matching Dennett’s ineffable attribute. She can’t conduct experiments to detect it, because it’s scientifically inaccessible, meeting the private attribute. And Jackson argues that qualia are epiphenomenal, which seems to meet Dennett’s intrinsic attribute.

The same attributes are implied with the inverted spectrum concept, the idea that there’s no way to know if my experience of red looks like yours of green and vice versa. The fact that we seem unable to describe our experiences of color to each other, that they’re ineffable, private, and intrinsic, is what gives this scenario life. Likewise, the absent qualia / zombie argument, the idea of a being physically or behaviorally equivalent to a conscious one, but not itself conscious, only works if there’s no way to observe or deduce whether qualia are present.

And yet, in all these scenarios, the subject themselves still has first person access to these phenomenal properties. For that to be possible, for that access not to be prevented by the other attributes, it has to be special in some way, according to David Chalmers, in some non-causal manner, which gives us Dennett’s directly apprehensible property.

And what is it that makes Chalmers’ hard problem of consciousness metaphysically hard, if not these attributes? Remove them, and the thought experiments and metaphysical mysteries seem to disappear.

So the advocates of these thought experiments and illusionists seem to agree on what qualia are. They just disagree on whether they’re real. Functionalists and other physicalists, if they use terms like “qualia” or “phenomenal properties,” are referring to a concept with less theoretical commitments. Do Dennett’s attributes show up in the more reserved versions? As Dennett himself covers in the last section of his Quining Qualia paper, it becomes a matter of “in principle” vs “in practice”.

For ineffability, no one thinks describing experiences like the redness of red in a functional manner is obvious or easy, although it can be done to at least some extent, starting with the distinctiveness and high saliency of redness. For many experiences, the phrase, “a picture is worth a thousand words,” comes to mind. It might involve so much effort that it’s ineffable in practice, even if not in principle.

And there are limitations on our access to how these experiences are constructed. They are cognitively impenetrable, for the simple reason that it was never adaptive for our ancestors to be able to access the early processing, such as all the underlying associations and affects red stimuli trigger, but which figure in what red experience feels like. Which makes a full description of the experience impossible with only introspective information.

Mental content, until recently, was private due to technological limitations and our lack of knowledge about the brain. It still effectively is in virtually all cases. But with the progress in brain scanning technologies, we’re seeing the first cracks in this attribute. We still have a long way to go, but even though it’s early days, the idea that mental content is in a separate realm and utterly inaccessible seems less defensible with each passing year.

Without absolute ineffability or privacy, it’s not necessary to bring in direct apprehension. Which isn’t to say that we don’t have privileged internal access in practice, but it’s similar to the type of access the processors in the device you’re using right now have to read and write memory that aren’t easily observable from the outside.

And then there’s intrinsicality. Achieving functional descriptions of conscious experience typically requires looking at the upstream causes and downstream effects of what we think of as the experience. Intrinsicality assumes that there’s still something in between, something that remains with intrinsic properties, something distinct from the causal chain, somewhere where the prior causes culminate in the presentation, and from which the downstream effects flow, with some aspects still conceivably epiphenomenal. The functional shift here is to regard the experience as the whole causal chain, a more plausible stance in a massively parallel system with no central control point.

Clarifying these attributes as difficulties in practice, rather than absolute limitations in principle, both explains our impressions of them, and transforms conscious experience from an intractable metaphysical problem to a series of scientific ones.

This is one of the reasons I used to resist the illusionist label, and still prefer the functionalist one. The difference doesn’t seem that vast (a point David Lewis made in 1995), and mostly seems to amount to a lack of nuance in our initial understanding, rather than some deep unavoidable species-wide misperception.

And yet for a significant portion of the population, the strong intuition is that a functional description, while explaining behavior, still leaves out something important for experience. And here we run into an intuition clash. For someone convinced that an ineffable metaphysically private aspect remains, it doesn’t seem like something science can demonstrate is or isn’t there. It becomes an extra assumption some people hold and others don’t.

Which seems to leave us in the strange place where the two views are empirically identical, and the debate a purely philosophical one.

Unless of course I’m missing something. What do you think? Are functionalists overreaching for a non-gap explanation? Are there fact-of-the-matter differences between illusionism and functionalism I’m overlooking? And are there ways to demonstrate the reality or non-reality of ineffable private qualities?

Featured image credit

Published

130 thoughts on “Illusionism and functionalism”

  1. Traction might be gained with this problem by getting clear about what we mean by this single self that considers itself to be having an experience. What part of our distributed body and mind has both a singular nature and an interface to something it can consider to be its changing experience?

    Liked by 1 person

    Reply

    1. From what I’ve read, the single self is itself a model used by a system that is massively parallel, with no one region always in the loop. Instead there is a succession of circuit coalitions which come into being, accomplish their task (or at least try to), then fade away as the new coalition develops. Many of these coalitions include the circuits that model the self, providing an impression there’s always that indivisible self as part of the process.

      Like

      Reply

      1. Makes sense, and would imply that simulation might be a more accurate word than illusion. What we decide to do is driven by the prediction of the consequences for our simulated singular self…and then the simulation is updated to learn from real outcomes. This simulated self is therefore both illusory, and as real a reflection of how well we (this collection of cooperating cells) are doing as we have access to.

        Like

        Reply

      2. Have you read Max Bennett’s “A Brief History of Intelligence”? Among other things, it provides an anatomical model of how and where these coalitions develop, and explains how and why the model of self happens at the higher (highest?) level of the hierarchy. [See my op reply for an anatomical explanation of this hierarchy.]

        *

        Liked by 1 person

        Reply

        1. I have his book but haven’t read much of it yet. My early impression was that he wasn’t very careful with the facts. But a number of neuroscientists have recommend his book, so I do plan to swing back to it.

          Like

          Reply

          1. [I failed to give the promised anatomical explanation in my other reply, so I’m putting it here]

            Given my thesis (consc. = pat. recog. + response), plus the concepts of (pattern) unitrackers and semantic pointers, I’ve been looking for the neural correlates of these functions in the anatomy of the human brain. Between Bennett’s book and a paper in which he theorizes the various functions of neurons in cortical columns (with Karl Friston as a reviewer: https://www.frontiersin.org/journals/neural-circuits/articles/10.3389/fncir.2020.00040/full) I think I have a good starting point to explaining human consciousness.

            My proposal (based on Bennett’s proposals in the paper) would be that a cortical minicolumn, constituted by an L4 pyramidal cell and its associated L2,3 cells, perform the functions of unitracker, tracking one pattern. The input from the senses comes via the core thalamus relay directly to the L4 and from the L4 to the L2,3’s. The L2,3’s interact with the other L2,3’s within the macrocolumn via coordination and competition and also w/ a particular set of L5 cells within the macrocolumn. This set of L5 cells constitutes a semantic pointer (as described by Chris Eliasmith), and as a unit represents the pattern of activation of unitrackers within the column. These L5 cells then activate a new (higher level) set of core relay cells in the thalamus, which core relay cells activate the L4 cells of a new, higher order, set of unitrackers in the cortex.

            In short, patterns in the thalamus (potentially starting with patterns of sensory data) are tracked by unitrackers (minicolumns) in the cortex, which then generate new patterns in the thalamus (via semantic pointers in the cortex) which are then tracked by new (higher order) unitrackers in the cortex. Bennett’s book essentially describes the evolution and “function” of these higher orders in the hierarchy.

            FWIW, most people considering consciousness are thinking about the highest order in this hierarchy, and refer to everything happening below/prior to the highest level as unconscious (or pre conscious or some such).

            *

            [whew]

            Liked by 1 person

          2. Thanks for the overview. I don’t have much to say on the details. I still suspect that the columns are more about ontogeny than anything else, more a consequence of evolution finding a way to generate a lot of substrate. That doesn’t mean functionality may not end up clustering in them, but I tend to doubt it will be as clean and consistent as you’re laying out here. And as I’ve noted before, I’ve read there are a large number of cross connections between cortical regions, as well as connections with other sub-cortical regions like the amygdala, hippocampus, basal ganglia, midbrain, etc, as there are with the thalamus. All of which makes me doubt the thalamus is the center of action in that way.

            But maybe I’m wrong. We’ll have to see.

            Like

  2. Fabulous piece. Well done.

    Re “What happens when Mary is released from her black-and-white room or is given a color television monitor? Does she learn anything new or not?”

    The key word here is learn, and of course, she learns something. Consider two friends, one has just come from a 31-Flavors type ice cream store and had tried some fabulous new flavor. She went on and on trying to describe it, so they went back to the store so her friend could actually taste it. When she got a taste, she now had sensory data to attach to her friend’s description.

    This is a common event because we do not have a way to share sensory information. We might be able to come up with one, maybe asking the friend to cobble the flavor together from memories of past tastings: “Imagine a really dark chocolate bit in your tongue, now add . . . ” But this is not the same were we able to share the sensations telepathically, I suspect.

    So, Mary would ‘learn’ something, if only through associations of her research numbers with her sensorium.

    And I have to ask, your illustration describes the philosophical question of do we see the same colors or are we taught that bananas are “yellow” and we just use that tag for what we sense? This question keeps coming up but it is frustrating because the question has been resolved. Because of brain scanners we know that when we see “banana yellow” we are seeing the same color and and are using the same visual cortex to process the signals. Yet the question keeps coming up quite frequently because people are not aware that it has been answered.

    Maybe it is just too good a question to gloss over and maybe this is why we should continue to teach philosophy to school kids.

    And I do not find consciousness to be mysterious. It seems a quite normal function of the brain as it copes with navigating its surrounds and surviving via evolutionary principles. I have a ladder of brain functions that lead quite normally to consciousness (or something like it) but, of course, the data are not yet available to fill in all of the blank spaces in such schemes.

    Keep up the good work!

    Liked by 2 people

    Reply

    1. Thanks Steve!

      I think the Mary’s Room story conflates two scenarios. One is the one you describe, which I fully agree with. It’s the one where someone has symbolic information (such as a language description) of the experience. They definitely learn something new when they have the actual experience. That’s how most people think of the story. Except that there’s nothing metaphysically significant in that version.

      For it to have the implications it’s supposed to, Mary’s knowledge has to be far more thorough. She has to have a sublimely complete understanding of the physics of the experience. That’s highly implausible, impossible in practice, but that implausibility cancels out the implausibility of her knowing so much that she learns nothing new with the experience.

      It’s worth noting that Jackson himself later changed his mind and decided it doesn’t have the implications for physicalism he thought it did. His new answer is that Mary gains a new ability, not new knowledge. But I think understanding what the premise requires makes even that answer unnecessary.

      On color perception, I hadn’t heard about those brain scans. Of course a non-physicalist would insist they’re only capturing the functional causes and effects of the experience, not the intrinsic aspects. But as a functionalist, I think those causes and effects are the experience, which makes the idea of inverted qualia a non-starter. It’s like saying my pain is your pleasure and vice versa. If I have all the reactions of it being pain, it what sense does it even make sense to discuss it as pleasure? It becomes a meaningless conjecture presupposing dualism.

      Agreed, I don’t think consciousness is mysterious in the way many take it to be. There remain interesting questions about how the mind works, but those are scientifically tractable. But as you note, some arguments never seem to go away.

      Like

      Reply

  3. @selfawarepatterns.com
    > What do you think?

    I think a philosophical-only understanding, while having a great historical body of work, misses what we now know about physics and neurons that makes up the conscious brain. Every brain has a unique arrangement of neuron connections based on experience and training which is what qualia physically _is_. You need a bridge of understanding between all we've documented of overall behavior from philosophy and how the physics works

    Liked by 2 people

    1. I agree with what you say here, for the functionality. The question is whether the functionality is a complete account. I think it is, but not everyone’s onboard. And it’s whether those indescribable undetectable acausal aspects exist that seems to still fall in philosophy. Myself, I think Occam’s razor culls it, but many disagree.

      Like

      Reply

      1. @selfawarepatterns.com
        > The question is whether the functionality is a complete account.

        That is fundamentally a religious question about the existence of an immortal soul and unanswerable by science or philosophy, you either believe or not but evidence and logic don't factor into it.

        The trick is figuring out when the question can be mathematically refactored as a religious one, and stop trying to apply logic and reason, it won't ever become consistent

        Like

        1. @selfawarepatterns.com the belief in religion is also a part of how the physics of biology works, superstition is baked into neurology

          I personally think that older animist tradition, belief in tree spirits and such, is actually close to the unknowable reality, because it's a fundamentally a recognition that living things have chaotic complex behavior, that can be described as a spirit. The difference between a live mouse and a dead one is physically very small, it's life

          Like

        2. I agree with the general sentiment. But not every non-physicalist is religious. Thomas Nagel, David Chalmers, and Bertrand Russell are (or were) atheists, but still think there’s a non-physical aspect to consciousness.

          Like

          Reply

          1. @selfawarepatterns.com IMO atheists _are_ religious, in that they have strong faith in the unknowable, but don't like how religion is organized and practiced, so they rebel against it in very specific ways, heavily influenced by the religion of their culture. Atheists care deeply about religion and a belief in non-physical consciousness is religious, not rational, even if they delude themselves about it.

            It's agnostics which actually don't care about religion.

            Liked by 1 person

    2. @selfawarepatterns.com you have the genes which describe the overall structure of the brain, how the cells are built, specific chemical recipes that affect overall function, and experience, input from all the connected senses and how those are learned to form connections and relationships with each other. You have a system primed for massive learning and feedback loops let loose

      Liked by 1 person

  4. Dennett’s qualia aren’t a straw man, but definitely not a Steel Man either. And there’s something wrong with your diagnosis of some of the classic philosophical scenarios. They aren’t all committed to Dennett’s four alleged pillars of qualia-talk. (The “philosophical zombie” thought experiments, however, are hopeless, and Dennett’s diagnosis of its proponents is probably correct.)

    First, let’s distinguish between concepts vs properties. A concept is an epistemic tool. A successful concept refers to a property (or process, or individual; but I take it we’re focusing on properties). A property is an aspect of the real world.

    And now we can see the problem with Dennett’s pillar number 1:

    ineffable – they cannot be communicated, or apprehended by any means other than direct experience.

    Some concepts might be ineffable in this sense while the properties they stand for are not. Thus, Jackson’s thought experiment might be illuminating something true and important, even if Chalmers and other property dualists are mistaken. Similarly, inverted qualia are conceivable (but not physically plausible) even if we rule out property dualism. (Digression: it has been pointed out that the subjective color spectrum is not symmetric to any rotations along the R-G-B circumference, so behavioral inverted color spectra are ruled out. But light and dark are symmetric around medium gray, so the battle can move on to that territory.)

    I would also like to renew my objection to the word “illusion” when referring to thoughts rather than perceptions. Do we perceive that subjective feels cannot be communicated? No, those who think so have thought their way into it. Of Dennett’s four pillars, only the first part of the last one is remotely plausible as a perception:

    directly or immediately apprehensible by consciousness – to experience a quale is to know one experiences a quale, and to know all there is to know about that quale.

    However, how would I know if a subjective experience flits by without making it into memory for long enough to get meta-cognized? The only thing we can say for certain is that sometimes we know some properties of experience.

    Liked by 1 person

    Reply

    1. Right, if Dennett has steelmanned qualia, he probably would have ended with his initial statement that they are the way things seem to us. Since he’s arguing that the seeming isn’t the actuality, that wouldn’t eliminate them.  Keith Frankish makes distinctions between “classic qualia” (the version Dennett’s attacks), “zero qualia” (basically a functional conception of qualia), and “diet qualia” (an in-between version that tries to do away with the problematic attributes while trying to preserve some mystery about qualia).  Overall he argues that diet qualia don’t work.  They either inflate to the classic version, or deflate to the functional one.

      But if we stick with Dennett’s initial identification, then we have qualia that exist subjectively, if not objectively.  This used to be my position, but to steelman Dennett and Frankish’s argument, it’s really just another way of talking about illusionism.  I have some sympathy with the objection to “illusion”, but the point remains that the seeming and the actuality aren’t the same, which is also another way of discussing your concept vs property distinction.

      My diagnosis of Mary’s Room was based on Jackson’s original takeaway, which was that it refuted physicalism.  I think that only works with a strong conception of qualia. If we take a weaker implication, similar to the one Papineau takes as a refutation of a priori physicalism, then we may be able to view in the way you describe.  But that’s not the author’s view, or the one I see from non-physicalist philosophers.  

      I’m with you on your comments about direct apprehension.  As described, it largely just begs the question in favor of dualism.  Any actual physical process would be fallible.  

      Like

      Reply

      1. Yeah, but Frankish is wrong: diet qualia do work. Papineau’s “phenomenal concept strategy” in particular, works just fine, walking the proper path between the Scylla of non-physicalism and the Charybdis of illusionism. His more recent, co-authored paper, concluding that the dispute is merely verbal, leaves me unimpressed – sometimes one way of talking is far more economical and/or explanatory than another.

        You could say that sometimes, the claim that a dispute is merely verbal is “merely verbal” (where the latter is a very expansive sense of “merely verbal”, in which we get to put awkward, complex ideas into people’s mouths).

        Like

        Reply

        1. I’m not sure diet qualia do work. When considering the concept we’re working with, we have to be sure we’re not using the one Frankish calls “zero qualia”. (Frankish could have less tendentiously called it “functional qualia”, but he wants to discourage use of the word “qualia”.). What attributes do the diet version retain that preserve a hard problem? How do they avoid implying the other attributes?

          The thing about illusionism is, it’s a negative proposition. Most illusionists are functionalists for their positive account of what is there, but not all. There are actually non-physicalist illusionists, like James Tartaglia. It wouldn’t surprise me if some illusionists aren’t a posteriori physicalists. So whether it really is a verbal dispute depends depends on the illusionist. But in most cases, I agree it’s more than just verbal.

          Liked by 1 person

          Reply

          1. Well, out of Dennett’s four pillars, the only one my understanding of Lite Qualia preserves is intrinsicality. The metaphor I keep returning to is internal combustion. A standard, gas-guzzling engine is not the only way to make a car go. Battery-electric, or fuel-cell-electric cars can perform all the functionality of a car just as well. The difference is purely internal, intrinsic to the engine. Other than price, availability of fuel, etc., few people would care about which kind of engine a car has. But Classic Car lovers would. I dare to suggest that most humans are Classic Psych lovers, when it comes to what kind of mental engine a person has. At least, I am.

            I think you’ve already spelled out how intrinsicality preserves the Hard Fact. (It’s a Fact that phenomenal concepts don’t have innate connections to our brain and behavioral concepts, but it’s not a Problem!) So I won’t bother repeating that.

            As to how to avoid the other Dennett pillars and similar baggage, it’s easy. Use the metaPhors(meta force), Luke. Engines aren’t private. You can open up the hood (though it might be considered rude, or a crime). They aren’t ineffable, although it can be really hard to convey to your mechanic how the feel of the engine has changed and made you suspect a problem. Etc.

            Liked by 1 person

          2. I do think whether the car is battery, gas, or hybrid makes a difference, in the sense of affecting the overall causal footprint of the car. You noted some, such as the availability of fuel, recharging time, range, etc, but also affects on the environment, noise, and other structural and causal relations, such as social reputation.

            I noted to Tina that I’m never sure what someone means by “intrinsic”. Often it means non-relational and/or epiphenomenal, but I take yours here to be irreducible. And certainly, experience is introspectively irreducible. We can’t examine the processing of neural circuits in our brain. As I noted in the post, it’s cognitively impenetrable. The question is whether we can objectively explain what we introspect. I think we can, but I’m in the a priori physicalist camp. I do grant it will likely never feel like we have to anyone convinced the functional account leaves out something.

            I guess the question is whether this really counts as diet qualia, which might come down to our definitions.

            Liked by 1 person

          3. Seems to me that “irreducible” is not much less subject to varied uses and meanings than “intrinsic”. Anyway, let’s say that a property, e.g. being water, is “strictly intrinsic” if we can carve out a region of spacetime and say that whether water is in the region depends only on what’s happening in that region. Let’s say a property is “for all practical purposes intrinsic” (FAPP intrinsic) if, even though there may be 1:1 necessary correlates outside the region, they are not the kind of things we naively think of as being relevant to the property in question. So for example, whether a person is in pain might be 1:1 correlated with certain images their brain makes on an fMRI machine, but that wouldn’t count against FAPP intrinsicality. The FAPP version is the one I’m going for.

            (I kinda suspect that no actual properties are strictly intrinsic, even chemical ones. Partly because metaphysical interdependencies can’t be divorced from causal interdependencies.)

            Liked by 1 person

          4. I think that’s the more common meaning of “intrinsic”, as something relative to how we’re interacting with something. So mass is often described as intrinsic to matter, and that’s true for a whole range of purposes. But ultimately mass involves the Higgs field and other interactions, which makes it relational. Which is to agree with your final point, there really is nothing intrinsic in the philosophical sense.

            Likewise, “epiphenomenal” often is used in an absolutist manner by philosophers that don’t reflect more common usage. Steam coming out of steam engine is epiphenomenal for engineering purposes, but not for philosophical ones.

            It seems like a lot of philosophical problems are avoided by not pushing common meanings too far.

            Liked by 1 person

  5. Great post, Mike. But I think Steve Ruis makes some good points. It seems, to me, that the “Mary’s Room” mind experiment, for example, reveals a deep commitment about a version of ontology that folks struggle mightily to hold on to. I’m tempted to regurgitate all my previous remarks to your previous essay—“Illusionism and Types of Physicalism.” I will just quote myself with a short distilled comment. “I fail to see the necessity of resolving our understanding of reality as having to be reduced to a foundational property. … Simply put, I have no problem with a greater evolving complexity producing a plurality of irreducible entities among the furniture in our universe—no problem with the complexity of a brain that causes a mind. Nor with a plurality of irreducible parts of reality that are understood only by explaining their complex systems as a whole.”

    I submit the argument on Mary’s Room (to me) appears to require minimal imaginative effort. In short, the description of an experience—no matter how thorough, is simply not the same as the experience itself. And could never be so—regardless of the amount of scientific knowledge about that experience. Otherwise, for example, so many young adolescent males would not so very ardently desire the experience of consummating an intimate relationship. They just don’t seem to be content, do they, with knowing all about it through books and films. Ok, so Jackson now calls that a new ability, not a new experience. I further submit that’s a distinction without a difference.

    Liked by 1 person

    Reply

    1. Thanks Matti!

      Yeah, we just disagree on what kind of explanation will satisfy us for understanding reality. To me, just saying “emergence” or “irreducible” and stopping amounts to giving up. I won’t be satisfied with that explanation. I may accept one like it provisionally, but I won’t see it as done until we can, at least in principle, relate the explanatory layers together. Which isn’t to say I expect us to use particle physics to predict the weather, just that we understand the relationship between them.

      Sounds like you read my response to Steve, so I won’t repeat it. I’ll just note that my discussion in the post was centered around Jackson and other non-physicalist’s takeaways, which I think requires the stronger premise I described in that reply. I understand many people take weaker implications from it. Some of those I’m fine with, like the specific one Steve described. I’m less onboard with David Papineau’s takeaway that it leaves physicalism intact but refutes a priori physicalism, basically the type of explanation I want and currently see no compelling reason for pessimism on achieving it. Maybe I’m just being naive.

      Liked by 1 person

      Reply

      1. Mike, I really fail to comprehend that saying “emergence” or “irreducible” amounts to giving up. Far from it. It accepts emergence, yes. It accepts the presence of irreducible properties at various levels of complexity. It accepts a basic physicalist notion that we live in but one reality. Yet it demands a more complex understanding of that reality, consistent, I believe, with a broader interpretation of evolution and modern science in general. The new science of “complexity” provides the template to that understanding. But I grant you it does give up something. It gives up an invented philosophical problem. It gives up an outmoded Cartesian background assumption that modern philosophy has struggled with since the dawning of the Enlightenment—the requirement that reality must be understood as consisting of (at most) two entities, the physical and the mental. And it gives up the various unsatisfactory “work-arounds” to deal with that assumption by way of idealism, dualism, or reductive materialism among others. I really do believe that we can, in time, relate or explain how the layers fit together so to speak from this slightly different and more promising starting assumption, an assumption that reality is a plurality of entities. But, then I’m not the sharpest ontologist in the box so to speak.

        Liked by 2 people

        Reply

        1. Matti,

          I have no issue with emergence in general, in the sense that it’s often productive to use different models at different scales. But I still want to understand how those models relate to each other. In my mind, accepting emergence without that understanding is effectively living with a different epistemic kind of dualism. We tell ourselves it’s all part of the same reality, but we end up having to take it on faith. I may have to do that at times, but I won’t consider it satisfactory.

          In IT, we have a phrase, “the definition of done”. How do we know when we’ve succeeded. (Usually asked in project management.) I think the issue here is we have different definitions of done. That’s ok, at least in my mind. But mine is what it is.

          Liked by 1 person

          Reply

          1. Mike, I get your concern. I certainly don’t think emergence leads to any form of dualism. Instead it leads to a messy pluralism—which I find fits our understanding of reality much better. Rather than taking something on faith, think in terms of questioning our starting background assumptions—something the philosopher Hilary Putnam talks about. We have to start somewhere. We can choose to start from that old Cartesian mind/matter conundrum or somewhere else. I think the Cartesian starting question is the wrong one—and it’s spent out. Changing background assumptions is difficult. And there’s no way to begin that is guaranteed. Beginning assumptions in philosophy are corrigible just as they are in every science. Moreover a new starting point, like emergence for example, means looking at the evidence differently or looking for different evidence. And, as we know, an evidence-frame tends to bias our thinking—the Cartesian formulation certainly has. But that’s the nature of any inquiry. Rather than seeing it as a change of faith, I prefer to think of it in terms of that great metaphor by the philosopher of science, Otto Neurath; “We are like sailors who must rebuild their ship on the open sea.” I think the ship is in need of more repair.

            Liked by 1 person

          2. Matti, I don’t see my view as Cartesian, which I agree is deeply problematic. I’m a mechanist all the way down. Quantum mechanics did use to shake my confidence in that view, before I discovered that wave mechanics fit our observations. Which isn’t to say that the mechanisms can’t be complex enough that we don’t need different strategies at different scales.

            I do think the study of the mind is polluted with remnant Cartesian intuitions. But it seems like both functionalism and illusionism resist that pollution.

            Liked by 1 person

          3. Mike, as you say, “I don’t see my view as Cartesian.” I would agree for the most part. I was awkwardly trying to make a slightly different point. When I refer to a Cartesian background assumption—and I should probably more accurately refer to it as the Cartesian-Lockean assumption—I am trying to refer to the starting premise or presumption that is assumed a good philosopher must contend with. In other words our recent intellectual history (since Descartes and Locke) presents us with a specific ontological option. In short, starting with every Phil 101 class we are taught that both Descartes and Locke adopt a dualistic substance metaphysics—that everything is a body or a mind—and that we must grapple with that option. Emergence rejects that ontological option. Emergent theories reject that ontology must be reduced to a foundational property and certainly not a duality. Instead, as I’ve said, emergent theories posit that a greater evolving complexity, which we are coming to better understand, produces a plurality of irreducible entities. And there are strong arguments that such an approach to ontology is a more satisfactory fit with modern science.

            Liked by 1 person

        2. Matti, I like your “messy pluralism” that you mention in a later reply. To put it in my own words, let’s do away with the ontological tower and pyramids that philosophers have loved to build. Just lay it all out flat and let the interdependencies flow in many directions at once.

          Liked by 1 person

          Reply

          1. Thanks Paul. It is a bit of an uphill battle, however, to try to shift a long-established interpretative framework as you can see.

            Like

  6. So here’s my (Psychule theory) understanding of the situation. Consciousness requires 1. a pattern recognition, and 2. a response to that pattern recognition.
    Functionalism seems to focus on the response without reference to the pattern recognition other than it being “the input”, whereas Illusionism is focusing on the pattern recognition itself. “Qualia” are simply the first person perspective on the pattern recognitions. A system with a set of recognizable patterns can access those patterns via (semantic) pointers, and can only differentiate patterns via those pointers, so “that one, not this one” or the like. So illusionism says (I think) there are no intrinsic properties to the “red” pattern, but we assign a (illusory) “redness” property when we refer to “that one” (the red one).

    Is this clear as mud? Otherwise problematic?

    *

    Liked by 1 person

    Reply

    1. I wonder what I’m saying about functionalism to give you these impressions (assuming you’re getting it from me). But as long as the pattern recognition has a causal role, functionalists will be interested in it as far more than “the input”. It sounds like a fairly substantial amount of work is happening in this recognition.

      Although I may not be entirely catching what you mean with “pattern recognition”. You may be working at too low a level for it to meaningfully map to illusionism. Or we could say that illusionism involves misinterpreting patterns for certain purposes, or maybe mis-recognition of patterns.

      Sorry, I’m probably showing how little I’m understanding?

      Like

      Reply

      1. I’m sure I’m strawmanning functionalism mostly to try to make various points. Psychule theory is, of course, functionalist. Although I guess I could point out a single “recognition” can lead to multiple different responses, such as a memory of the recognition as well as a movement response (flee!). And to be honest, I don’t think I have a grasp on what the pertinent Illusion is, although it’s (almost) definitely not a misinterpretation of patterns. It’s more like an assignment of a user interface icon, but I can’t really say how that would work, so I think I’ll shut up about it now.

        *

        Liked by 1 person

        Reply

  7. Of course it won’t be as clean and consistent as laid out here. Most complex object in the universe, and all that. But all those connections you mentioned I think will have supporting roles (predictive processing, memory, action planning, attention, etc.). My current questions include the promiscuity of thalamus to column connections and the reverse. Time will tell. I recommend you read the book to see if this makes more sense.

    *

    Liked by 1 person

    Reply

  8. I appreciate your earnest manner of laying out your views. I think we could get a lot further in these discussions with a bit more of this self-reflection and intellectual honesty. Unfortunately, those at the helm have academic careers and reputations on the line, which could be a real barrier to progress and mutual understanding.

    I’m beginning to think these thought experiments have outlived their usefulness. It seems to me the whole point is getting missed in this tedious struggle to tear apart one another’s thought experiments, and this is taking us further and further into verbal masturbation rather than real reflection. Can we not agree that I can’t literally and directly experience your pain? Nor you mine? My pain is not directly accessible—felt—to outside observers. That’s it! It doesn’t seem so hard!

    I do think Dennett is knocking down a strawman.

    ineffable – they cannot be communicated, or apprehended by any means other than direct experience.

      I would say some experiences can’t be defined without circularity. “Pain is a feeling of unpleasantness.” That’s pretty bad as far as definitions go. But this isn’t necessarily a failure of communication unless you assume communication must be the epistemic equivalent to being you or if you think it must not require any assumptions whatsoever. But then you’re taking communication into an unnatural place where I don’t think it can exist. Honestly, I think Chalmers is the one to blame for this sort of thinking. I have often found problems with his way of formulating scenarios. For instance, if I thought you experienced red where I experienced green, why wouldn’t I take that into account in my interactions with you? And if I didn’t know, who does? Where does this inverted color experience get brought into the light of day? (Chalmers does this kind of thing a lot. He starts by stipulating what is impossible to know but then includes that stipulation in his conclusion without taking perspective or limited epistemic reach into account. Another example is this bizarre idea that we can live in perfect virtual reality. But what does that even mean? In order for that to make sense I would have to be able to step outside the perfect virtual reality and see it as such—otherwise it’s just reality. But once I step outside the perfect virtual reality and see it as such, it’s no longer perfect because I’m aware of it. This is just nonsense to me. I can’t wrap my mind around this spurious situation. It asks me to imagine myself as having both a God-like perspective from outside the perfect virtual reality, but without feeling it’s any less perfect by virtue of being virtual, while at the same time imagining myself as a limited perspective within it who magically has no notion of itself possessing this God-like knowledge.)

      Anyway, back to the point. Apprehended is ambiguous; it can mean perceived, or it can mean understood. Perceived is closer to what is meant by qualia’s ineffability (I prefer experienced), which is far cry from communicated. If I say, “OW!” I’ve communicated my pain to you. If I’m rolling on the floor in agony, I’ve communicated my pain to you, and you have some idea that this pain is worse than the previous one. This isn’t to say that I’ve communicated everything there is to know about my qualitative experience, as I experience it. To do that I’d need to invent some sort of mindmelding technique that somehow circumvents the problem of the unity of consciousness, which I’m not sure is possible. But I communicated in a powerful way—if I assume you experience the world in roughly the same way I do.

      intrinsic – they are non-relational properties, which do not change depending on the experience’s relation to other things.

      I just think this is a totally inaccurate way of describing what is meant by intrinsic. Saying my experience is intrinsic—inherently and essentially mine, not yours—doesn’t mean my experience has no relation to my behavior or the world or others. I think Dennett is expanding a limited scientific view of causality to the sphere of human motivation and intention to make this point, a product of his scientism I suppose. I think if we’re being honest with ourselves about our beliefs, about what we are willing to live by, we’ll see why it’s misguided. I think science’s causally-closed deterministic universe may be a necessary assumption for scientific endeavors, but that assumption doesn’t work beyond the scientific realm. The results of a purely material explanation of desire make no sense and can only provide unsatisfying answers. If I want coffee and get up to make myself coffee, my desire is the appropriate explanation for why I got up to make coffee. You don’t need to have direct access to my lived experience of wanting coffee to understand why I got up to make coffee. But you do need to have desires yourself and a basic understanding of how motivation works, which is beyond what science’s current paradigm can explain (even if scientists implicitly rely on their own intuitions). To say my getting up to make coffee was the result of my neurons firing is an answer of some sort, but it’s only an answer to the wrong question.

      private – all interpersonal comparisons of qualia are systematically impossible.

      I would agree with this if it said ‘comparisons by way of direct experience’. But as it stands, it’s not quite right. Surely I can compare my experience to yours by talking with you. Of course I do have to assume your experiences as similar enough to mine and that you’re not lying to me, but so what? That’s not a huge ask. These philosophy of mind guys seem to enjoy setting up thought experiments that require a kind of god-like knowledge that’s absolute and certain. Of course my understanding of your experience can’t ever be certain, but if I want certainty, I won’t get very far.

      That said, I wouldn’t expect you to have a clear idea of what it’s like to have menstrual cramps because you’re a man, and I do expect women to have a better idea than you because I know they have likely experienced them too. Even then we women still have to discuss amongst ourselves what it’s like for each of us because there are variances amongst us. Not everyone gets pain so intense they throw up, etc. The reason we must discuss to understand one another is that we can’t directly experience each other’s pain.

      directly or immediately apprehensible by consciousness – to experience a quale is to know one experiences a quale, and to know all there is to know about that quale.

      “directly or immediately apprehensible by consciousness” —This is fine.

      “to experience a quale is to know one experiences a quale…” —This sounds right on the face of it and if it were stated by someone other than Dennett or if I wasn’t aware of what he intends to argue, I probably wouldn’t notice that it’s wrong. But it is. Surely I can experience a quale without thinking, “I know I am experiencing a quale”. Plus, to talk about a quale is far too atomistic for me. There’s something strangely pointillistic about qualia taken in isolation from all of experience that doesn’t jive with me. Maybe I’m just being pedantic though.

      “and to know all there is to know about that quale”—What? Do people say this? Surely we can add to our description of pain-as-experienced a physical explanation. Obviously! That’s not the problem, though. The problem is when certain scientifically-inclined philosophers want a physical explanation to take over, to dominate, to rule over everything there is to know. That’s an overreach on the part of certain people, but not on the part of science itself, which I consider to be neutral. Science generally operates by subtracting or eliminating the ‘subjective’ from the outset; a method that makes sense when it seeks to understand the world as it is in itself. But it seems crazy that we should expect to use the same method to grasp the very thing which has been eliminated from its purview.

      On the topic of that brain-scanning AI article, I wouldn’t say that AI has read phenomenal consciousness—or read anything at all. It first gets fed input that is then correlated to the brain as detected on MRI, then it spits out some version of whatever it was fed. It’s people who are doing the ‘reading’. On top of that, the technology seems to be capable only of reading an individual’s mind, and only after being fed input by that particular individual. It can’t take what it ‘learns’ from me and read your brain. So even as a mechanical feedback loop it doesn’t seem to work universally.

      I think if your purpose is to break the subjective privacy barrier and do some sort of mindmelding thing, doing it by giving computers input and correlating to brain scans is far less interesting to me than other methods that may become available in the future, such as hooking up human brains to one another, maybe through some sort of device. I thought I saw an article about research on that somewhere. Maybe you know what I’m talking about?

      Liked by 4 people

      Reply

      1. The idea of hooking up our brains to share experiences only goes so far. If I’m hooked up to someone so that when they chomp a mouthful of cilanto, I taste cilantro, am I having their experience, or my experience of their experience? What if they like the taste of cilantro and I don’t? Will I experience their enjoyment of the taste instead of my dislike?

        Like

        Reply

        1. Funny, I saw your comment in my email but I thought it was Mike talking.

          As to your point, totally. And I agree it can only go so far. I can’t even conceive of a total mindmeld…does someone actually die?

          Did you hear about these conjoined twins? I can’t seem to find the article, but apparently one of the twins likes ketchup and the other doesn’t, but the one who doesn’t like it has to taste it whenever the other eats it. I would hate that (not a fan of ketchup myself).

          https://scoop.upworthy.com/krista-and-tatiana-hogan-share-a-brain-can-see-though-others-eyes-craniopagus-twins-ex1

          Like

          Reply

          1. I wasn’t aware of the Hogan case. It’s potentially illuminating.

            I don’t mind the taste of ketchup, but the condiment is used to ruin a lot of good food. If people gob it onto flavourless frozen french fries to lend some interest, well, I get that.

            Liked by 1 person

            Reply

      2. Thanks.  Yes, I do think academic careers can weigh on the conversation.  When someone has become known for a particular view, it’s harder to update that view.  It does happen.  I mentioned Frank Jackson’s change of mind somewhere else in this thread. And Michael Tye went the other direction.  But they seem like exceptions.

        I think thought experiments are useful for flushing out our intuitions, but don’t really tell us anything more than that.  In that sense, calling them “experiments” is misleading.  Dennett calls them “intuition pumps”, but I think “intuition clarifiers” is probably more accurate.  Your reaction to each of these thought experiments basically just reveals your intuitions about the subject matter.

        On experiences, I’d say that if our definitions can only be circular, it reveals that we don’t really understand them yet, and we have to resist the temptation to give up too early.

        I do agree there’s a difference between having an experience and knowing about it.  We can never experience being a bat. But my laptop can also never be in the same information state as my phone, even if it can run a phone in an emulator.  However it can have information about a phone’s state, just as we can learn about the bat’s experience by studying it, an important distinction.  

        The problem with the inverted spectrum (which actually goes all the way back to John Locke) is that it isn’t supposed to make any difference in behavior.  I agree that’s implausible.  The purported conclusion seems built into the premise.  But it means thinking harder about what color is.  And in Locke’s defense, he lacked a lot of concepts we have today, like evolution.

        “Apprehended” is vague.  “Acquainted” is another word often used.  This is one of the things I’m often frustrated with in philosophy, loading common words with questionable assumptions, without clearly admitting what’s being assumed.  It wraps dubious propositions in reasonable phrases, that implies anyone who doubts them of doubting the more grounded version.  How unreasonable and radical!  “Experience” gets a lot of this treatment.

        I honestly am never sure what people mean by “intrinsic”.  It has a common meaning that is often tied to how we’re dealing with a particular concept.  But it also gets thrown around as though it’s a precise word.  Dennett actually admits in the QQ paper that he himself isn’t sure what it means, and notes a lot of philosophy having been spent attempting to clarify it.

        On the mind melding thing, Dennett actually discusses this somewhere.  He points out that if you hooked two brains together, you’d have to make calibration decisions about how idiosyncratic sensation patterns in one brain map to the other, many of which could be debatable, and ultimately might end up feeling arbitrary.  All of which is to say, it wouldn’t get around all these inverted qualia type conjectures, and you still couldn’t be the other person having their experience, just the joined minds sharing an experience.  Although I think calibrating them for their causal effects would at least put them in functional sync, which makes the whole inverted spectrum thing meaningless from a functional perspective.

        On a related note, there are conjoined twins which share some of their sensations.  I think I remember one of a girl tasting some of her twins food.  Of course, they developed together, so it’s not the same as attempting to join independent adult brains together.

        Like

        Reply

        1. I guess I do agree with Dennett about something! I prefer “intuition pump” as well. They do tend to have a curious circular nature that’s good for kicking thoughts into gear, but not so much for making an argument.

          “On experiences, I’d say that if our definitions can only be circular, it reveals that we don’t really understand them yet, and we have to resist the temptation to give up too early.”

          I don’t mean our definitions of experiences can only be circular, but certain perceptions—the smell of a rose, for example—might be impossible to define in the usual sense, though they can be described by alluding to other smells. But I do think we understand these experiences; rather, I don’t think they necessarily need to be understood, but are simply there. You might call these experiential facts in the sense that they’re just given to us. But to say phenomenal consciousness is fundamentally not the sort of thing that can be adequately captured by a scientific-mathematical reduction is not to say such a reduction is impossible, or that one should give up on the science of consciousness, only that we need to acknowledge that this method leaves something very important behind.

          I have to admit, I don’t get computationalism or the idea that information is somehow equivalent to mental state. I especially don’t get computationalism when mixed with physicalism. It seems to me information is just not physical, whether it can be measured or not.

          On the inverted spectrum, yeah, it’s not supposed to change behavior. Which I can wrap my mind around, though I think it’s best if you take it for what it is and don’t analyze it to death.

          “it wouldn’t get around all these inverted qualia type conjectures”—not sure I follow here.

          I did see the thing about the twins. Now I’m thinking about inverted qualia with conjoined twins. A new thought experiment. I mean, intuition pump.

          Like

          Reply

          1. John Keats famously chided Isaac Newton for destroying the poetry of the rainbow by explaining its spectral nature.  Keats was being light hearted, but my observation is that many aren’t when it comes to conscious experience.  There are a lot of people who feel like there’s something unseemly, maybe even immoral, with trying to unweave experience, like maybe something is being violated, similar to how people felt when medical doctors started dissecting human bodies in the 1500s, to learn how they worked..  

            I’m in the Newton camp.  I want to understand, with no bounds, or at least as far as can be understood in our time.  I personally don’t think such an understanding destroys the poetry, although it may alter it.  Which doesn’t incline me to accept that experiences can’t be defined, or that we’re over analyzing them.  Maybe I’m just being naive.

            On computation and information, consider the device you’re using right now.  Is it physical (or manifest physical in an idealist view)?  Does it hold information and do information processing?  If so, then in what sense would information not be physical?  In that case, it’s the dance of matter in motion that provides the device’s capabilities.  It seems like in principle human and animal behavior can be explained with a similar dance, albeit a more advanced one.  The question is whether experience is something wholly or partially outside of that explanation.

            On the inverted spectrum and brain joining, let’s suppose that your and my red and green are reversed, and our brains are joined.  That means when I receive your visual perception of a ripe strawberry, I’m seeing it as green.  Presumably, all my reactions to it would be the ones I have when I normally see green.  “Oh wait,” says the technician. “I need to calibrate the translation.”  So they make adjustments, and now when your view comes in, I see it as red, and react to it as such.  So which was the “right” translation, the one where I saw green, or where I saw red?  Is there a fact of the matter?  I think there only is to the extent we want the customary reactions to happen on each side.  If that’s the standard, there is no inverted spectrum.  If someone says it’s not the right standard, what argument can convince them otherwise?

            Liked by 1 person

            Reply

            1. I don’t see the scientific investigation of consciousness as such as immoral or destroying the poetry of experience, but rather that it has little to do with either, and that’s because it isn’t an investigation into those things. In other words, both morality and poetry are unaffected and remain the same as always. The problem I have is with the philosophical position that morality and poetry are really nothing and contribute nothing to knowledge in general simply because they cannot be taken into account by our current scientific methodology.

              That said, I realize it can’t always simply be a matter of engaging in science in one way as a particular activity off to the side and then divorcing that activity from one’s view on life in general. We once talked about how scientists tend to be scientific realists rather than instrumentalists. I can see why a realist would not be keen to take the view that science is a particular type of investigation, one that leaves room for other ways of knowing. Reality is what science says, nothing more, nothing less. I’m not sure how that outlook works within the psychology of such a person, but I suspect they have a way of compartmentalizing their own experiences of, say, being taken aback by a work of art or being morally offended by someone’s actions while at the same time thinking these experiences are not to be taken as reflecting any sort of reality, since according to such a person art can only have subjective meaning or value and moral judgments, though conceptually undermined by scientific determinism, should continue to be upheld for pragmatic reasons. Personally I’m not comfortable with that conceptual dichotomy. I can’t reconcile the powerful intuitions that I find within experience and which inform my everyday life to a view that those experiences aren’t real; I would rather take a step back to see the scientific investigation as one way to understand the world we live in, and a powerful one at that, but one that doesn’t have the final say.

              Another question could be, if we saw science in this way, how would this inform science?

              As for computation and physicalism. From an idealist perspective I’d say the device I’m using is indeed physical, but what counts as physical for the idealist will be available to experience. For the idealist, the physical is tangible, experienceable, and so that would include my laptop. The idealist sees the physical world the same way as the naive realist—as full of color, smells, and so on. In both, there is no assumption of dancing matter as the underlying reality. What differentiates the physical from the non-physical in an idealist view is the nature of the thing in question: the physical is visible, tangible, and the non-physical is not visible, not tangible, and both are real, just in different ways. This view overlaps with naive realism except for one big difference; the naive realist concludes that what is not tangible is in some sense not real, or at least not a part of nature, and for that reason may be a mere idea, something that exists only for human minds, etc. So the point is, the idealist sees both information and the laptop as real, but different and distinguishable. Information is not tangible, and so not physical, whereas the laptop is tangible and physical. Matter is intangible too, which means it’s not physical—but that makes no sense! For idealists, matter is a spurious concept because it is supposed to be the very rock bottom of physicality and yet is not itself physical. (But I get that the sophisticated physicalist will not insist that the physical must be tangible; quite the opposite, the physical turns out to be not at all tangible.)

              “Does it hold information and do information processing?”

              Here is where my trouble with computationalism begins. I don’t think my laptop literally holds or processes information. I see that as metaphorical language, but strictly speaking, we process information. Information exists in minds, or rather I would say it is meaning that gets passed between minds. My laptop is like paper with marks on a page; the meaning can only exist for those reading it. So that’s one area where I get confused about computationalism. It seems as though the metaphor is getting taken literally. But when you combine that with physicalism I get really confused. Is information itself an object for scientific investigation? Is it literally measurable? Or is information here getting mixed up with the equivalent to marks on a page? I mean, imagine if I thought I could communicate a sentence to you by measuring the letters and the spaces between each word and passing along my measurements to you. You wouldn’t know what the message said. The meaning of the message is not even in the letters or words. Does that make sense?

              On brain melding and inverted colors. “Presumably, all my reactions to it would be the ones I have when I normally see green.  “Oh wait,” says the technician. “I need to calibrate the translation.”

              Do you mean that you would see a green strawberry as unripe? Just as you see other green things that become red when they ripen? And therefore would be mistaken because the strawberry really is ripe, it just doesn’t look that way to you?

              Like

            2. I’m a poetry imbecile, so I probably shouldn’t say anything else about it.

              I’m not a moral realist, at least not in the sense of seeing morality as natural law. That doesn’t mean I take it not to exist at all. I see it more as a social tool, a way of living, which we all develop together. And while science can inform that tool, or learn about a culture’s particular consensus on it, it can’t dictate it. There will always be an element of deciding what we value, and discussing with others what they should value. (Or in the worst cases, trying to impose it, but that only ever seems to work short term.)

              But science to me overall is just the best ways we know of obtaining reliable knowledge. When people talk about other ways of knowing, I always wonder what they think science is, or what those other ways are. Science can’t tell us who to be friends with, or marry, or whether we’ll like a new dish, or many other things that are more about our own predilections than anything else. If for no other reason, the cost / benefit for the effort involved usually isn’t worth it. Put another way, sometimes the best way to discover something is just to try living it. I think most science oriented people understand that.

              On information and computationalism, I think a lot depends on what we mean by “information”. My take is it’s causal differentiation, often pithily expressed as “a difference that makes a difference”. That seems true for information in minds, who are causal agents in the world, and computers, which are also causal agents.

              But a lot of this may hinge on our differing metaphysical views. We’re both monists, but I’m a mechanist. I think everything is mechanics, including minds. (In my view, Descartes took a wrong turn when he exempted minds from the mechanical philosophy.) For a mechanist, the idea that the information in our minds is the same stuff as the information in a computer isn’t jarring. Information in minds has to be implemented in some manner. But for someone with a mind first ontology, I can see where there would be a difference. Computationalism just won’t make sense in that worldview.

              On brain melding and the strawberry, I was actually thinking in earlier terms, such as the strawberry not looking as striking. But deciding whether it’s ripe gets at the same idea. Maybe another example would be if you saw a completely red room, something most people (and primates) find unsettling. If I perceived it as green, maybe I would be soothed by it. At least until the recalibration.

              Of course, under functionalism, the experience is those reactions in the pre-introspective causal chain. Which means the initial connection would not have been bringing through the entire experience.

              Like

            3. I’m not sure if this box will link to your comment thread properly (I can’t seem to find your comment here, but I see it in my email). Anyway, fingers crossed.

              By other ways of knowing I just meant ways that don’t use the scientific method to obtain knowledge. Deciding that science is the only way of knowing is not itself science, but philosophy—which is not a ‘way’ or ‘method’, but a stepping back from all ways and methods so that they can be evaluated.

              “Science can’t tell us who to be friends with, or marry, or whether we’ll like a new dish, or many other things that are more about our own predilections than anything else. If for no other reason, the cost / benefit for the effort involved usually isn’t worth it.”

              If we left aside all practical considerations, do you think science is in principle capable of answering all questions?

              “For a mechanist, the idea that the information in our minds is the same stuff as the information in a computer isn’t jarring.”

              I get what you’re saying about the difference between our views, but I actually have no problem with information being real. Idealism would be quite accommodating about that. It’s the characterization of information that I’m confused about.

              The thing is, I find computationalism most confounding when I think about it from a materialist’s perspective. It seems to me even a materialist would say that information is not some material thing, but is mind-dependent. Science hasn’t discovered the causal effects of some free-floating material entity that we could justly call information itself. Information as such would, according to the strict materialist, seem no different from qualia or phenomenal experience. In other words, I would expect them to call information an illusory by-product of our brains, nothing physical—nothing that in itself has causal powers—and therefore nothing real.

              Now I just happened to have Chalmer’s book beside me. I found him a bit too breezy on this issue. He says:

              “Information is physical—or, at least, structural information can be physically embodied. The basic idea of structural information as strings of bits is an abstract mathematical idea, but strings of bits gain causal powers once they’re embodied in physical systems, such as punched cards and computers.”

              So what is information itself, then? It sounds to me like there are various things being described here and swapped with one another willy nilly:

              One thing is called “information” and it’s characterized as abstract and mathematical. But is information causally empowered on its own? No definite answer is provided here, though there does seem to be a hint that it’s not.

              The other thing is called a “bit” (or “strings of bits”) which the dictionary tells me is a unit of information expressed as 1s and zeros. Is a bit a material thing? Does bit refer to mind-independent physical stuff in the world, stuff with causal powers to affect other material things? Or do those 1s and 0s as such live in the realm of abstract mathematics? I would guess the latter, though again, I’m not 100% sure I know what’s being referred to when people talk about bits. This is really far afield for me since I’m not familiar enough with computers.

              And then there’s another thing, the embodiment of the above in “physical systems” such as “punched cards” and “computers”. That seems clear enough.

              But embodiment?

              Imagine if we crossed out the word “information” and replaced it with “consciousness” in the paragraph above:

              “Consciousness is physical—or, at least, consciousness can be physically embodied. The basic idea of consciousness as strings of bits is an abstract mathematical idea, but strings of bits gain causal powers once they’re embodied in physical systems, such as punched cards and computers.”

              If we take “strings of bits” to be something not physical, then you can swap “consciousness” for “strings of bits” fairly easily. The two are not essentially different in nature, since they’re both “abstract mathematical ideas”. Which would read: “…consciousness gains causal powers once it’s embodied in a physical system”.

              I realize Chalmers is a dualist, so I wonder how you would change this description to make sense according to a monistic physicalist-functionalist view?

              Like

            4. On the comment, looks like you got it. Yeah, the Reply To link in the email works, but I wish WordPress gave a more reassuring visual indication of it. But I’ve learned to just trust the link when replying to comments it doesn’t let us reply to otherwise.

              “If we left aside all practical considerations, do you think science is in principle capable of answering all questions?”

              No. One thing I forgot to note were logical truths, which usually aren’t included in what we think of as science. So a mathematical or logical proof doesn’t seem like science. But for things in the world, if we regard science as the most time tested ways of acquiring reliable knowledge, what other ways in principle could you see? Again, doesn’t mean I’m going to conduct a double blind survey to figure out which brand of coffee I like. It will involve me actually tasting those brands to see which provide the best experience. But my results would have no necessary relation to which brands you prefer. (Although a careful nationwide taste test might pertain to the probabilities of which ones we do like.)

              For information, I think the first thing to acknowledge is that the word can refer to a range of concepts. I usually make a distinction between physical information, of the type that physicists discuss, and which they expect to be conserved (similar to energy conservation), and semantic information, information that carries meaning for some agent (not necessarily a conscious one). I generally see semantic information as a subset of the physical variety. Crucially, all physical information is potential semantic information. It doesn’t transform when an agent utilizes it, it just becomes…utilized.

              Eric Schwitzgebel, in a conversation with Philosopher Eric (now Eric Borg) once made a comment that information is causation. I’ve often thought he was right about that for physical information, but it’s a bit hazy. What we can say is that information and causation are closely related.

              Of course, in a modern electronic computer, information is largely treated as something inert. There is the doer of the system (the CPU) and the memory, the storage. The division between action and data makes it easy for us to understand what’s happening. It seems to have grown out of the procedures human computers once followed when doing complex calculations. But it’s a distinction nature doesn’t seem to bother with. In a biological neural net, or any physical neural net for that matter, action and information are the same.

              But I’m not a platonist. Or more precisely, I see platonism as a matter of personal philosophy. I used to wonder whether it’s true or false, before concluding that there’s no real fact of the matter. Abstract objects don’t exist in the causal sense. But platonists are upfront that abstract objects are acausal with no physical extent. So to say they exist is to use a different meaning for the word “exist”.

              In that sense, you could say information exists platonically, which wouldn’t be physical. But the nominalist can proceed as though the platonic version isn’t there, except that in their language, they’ll find it hard to avoid discussing it as though it were. But for them to discuss it, there have to be patterns in their brain (at least) that encode the information, with no causal relation to the platonic version.

              I have no idea if this is helping or just swishing the mud around.

              Chalmers is a dualist, but he’s property dualist, so his dualism is pretty thin. I once wondered if we could regard phenomenal properties as platonic in nature, and so find a kind of reconciliation between reductive physicalism and property dualism. But Chalmers himself doesn’t seem inclined to go there.

              On how I would reword what he wrote: I think information is physical and causal, as is consciousness. If I recall, that passage was referring to a line of reasoning to avoid Termark’s mathematics is reality conclusion. His point was that you only get structure and relations in the abstract version, but causation in the physical one. What I think he overlooks is that cause is itself a type of relation. Not that I’m a Tegmarkian. I just think his reasoning there didn’t really work.

              Like

            5. I’m not sure I’m following, and now there’s this idea that causation and information are the same or at least closely related? That’s beyond baffling to me. I just…can’t compute. 😉

              “Abstract objects don’t exist in the causal sense. But platonists are upfront that abstract objects are acausal with no physical extent.”

              Actually, abstract objects give physical objects their being, so they are the cause. The Good (the form of all forms) is the ultimate cause. It might be better to switch to Aristotle lingo and think of the abstract as “unmoved movers”. Meaning they don’t have to move to cause action. They’re like magnets that pull things toward them.

              “I once wondered if we could regard phenomenal properties as platonic in nature, “

              What made you stop?

              For what it’s worth, I don’t think consciousness is acausal. I have no idea where that notion came from. Of course it’s causal! That’s free will! Of course my version of consciousness is different from yours.

              Like

            6. On the causal part, consider if you wake up one morning and discover doggie poop on the living room floor. The poop itself is information, it tells you about an event during the night. It’s also causal, because it leads you to react in various ways (cleaning it up, interacting with Geordie in some manner, etc.) All patterns have prior causes, and they all have causal effects, although it may take other coincident causes to generate particular actions, such as someone opening a book and reading a passage for it to cause an effect in their mind.

              On platonism, we may be talking about different versions. My take is based on what I read in the SEP article on platonism. In particular, this passage for the acausal remark.

              Platonism is the view that there exist abstract (that is, non-spatial, non-temporal) objects (see the entry on abstract objects). Because abstract objects are wholly non-spatiotemporal, it follows that they are also entirely non-physical (they do not exist in the physical world and are not made of physical stuff) and non-mental (they are not minds or ideas in minds; they are not disembodied souls, or Gods, or anything else along these lines). In addition, they are unchanging and entirely causally inert — that is, they cannot be involved in cause-and-effect relationships with other objects.

              https://plato.stanford.edu/archives/spr2016/entries/platonism/#1

              I think the idea that consciousness is non-causal comes from intuitions that make inverted spectrum or absent qualia (zombies) seem plausible. They implicitly assume qualia make no difference in behavior. It’s the non-causal part that makes me lose interest. It seems vulnerable to being pruned by Occam.

              Like

            7. I forgot to say what I think information is. I take it in the broadest sense to mean knowledge or facts or beliefs that are shared. Information as such is not physical or mathematical—its meaning is almost the same as ideas, but with an added ‘shared’ emphasis. Ideas or information can be expressed mathematically. But when they are expressed mathematically, the math is either the message itself (if the idea being expressed is mathematical) or the language that must be translated into natural language which must be read by minds that share knowledge.

              Maybe that will help you see where I’m at. 🙂

              Like

            8. Thanks. The difference is I see knowledge, facts, and beliefs as all physical information. It’s just that the way our brains instantiate it, as neural firing patterns, is different from the symbols we’ve developed to record it outside of our brains. But for me, it’s all physical, just in different forms or causal differentiation. Put another way, I don’t have the assumption that the mental is non-physical. (I mean, I did when I was younger, just like everyone else, but moved past it a long time ago.)

              Hopefully that helps explain the different assumptions we’re working with?

              Like

            9. Just came across this video, which does a good job at explaining physical information. You don’t have to watch the full thing if you’re not interested in the black hole information paradox. The relevant parts are in the first half. (Warning: he does a lengthy sales pitch for his sponsor between the 4 and 6:20 minute marks.)

              Like

            10. Thanks! I’ll check it out. Sorry if I’ve been tedious. I just can’t help but insert my Platonic forms into the functionalist mix. That may be where my lack of comprehension is coming from.

              Like

        2. I take it back when I said Dennett was knocking down a strawman. I went back to Chalmers paper and he’s quite confusing, but you can see where Dennett gets these ideas that consciousness is acausal if you look very closely:

          “I suggest that a theory of consciousness should take experience as fundamental.”

          Ha. So would an idealist! Except he doesn’t mean what he’s saying here.

          “We know that a theory of consciousness requires the addition of something fundamental to our ontology, as everything in physical theory is compatible with the absence of consciousness.”

          Again, so would an idealist.

          “We might add some entirely new nonphysical feature, from which experience can be derived, but it is hard to see what such a feature would be like.”

          So here C says, basically, “I’m a physicalist…nothing nonphysical allowed” but he’s being very slippery about it.

          But then:

          “More likely, we will take experience itself as a fundamental feature of the world, alongside mass, charge, and space-time. If we take experience as fundamental, then we can go about the business of constructing a theory of experience. Where there is a fundamental property, there are fundamental laws. A nonreductive theory of experience will add new principles to the furniture of the basic laws of nature. These basic principles will ultimately carry the explanatory burden in a theory of consciousness.”

          Experience is not, in his scheme, fundamental. It’s epiphenomenalism, but with physicalism taken for granted:

          “In particular, a nonreductive theory of experience will specify basic principles telling us how experience depends on physical features of the world.”

          That sounds reductive to me! Specifically, causally reductive. But there’s no mention of causation until…

          “These psychophysical principles will not interfere with physical laws, as it seems that physical laws already form a closed system.”

          Here it’s implied. The physical laws form a closed system, ergo…no mental causation.

          “The new basic principles postulated by a nonreductive theory give us the extra ingredient that we need to build an explanatory bridge.”

          It would be an explanatory bridge, then, between an acausal conscious experience (baffling to me!) and the physical world.

          Here he’s speaking to physicalists (everyone pretty much):

          “This position qualifies as a variety of dualism, as it postulates basic properties over and above the properties invoked by physics. But it is an innocent version of dualism, entirely compatible with the scientific view of the world. Nothing in this approach contradicts anything in physical theory; we simply need to add further bridging principles to explain how experience arises from physical processes. There is nothing particularly spiritual or mystical about this theory—its overall shape is like that of a physical theory, with a few fundamental entities connected by fundamental laws. It expands the ontology slightly, to be sure, but Maxwell did the same thing. Indeed, the overall structure of this position is entirely naturalistic, allowing that ultimately the universe comes down to a network of basic entities obeying simple laws, and allowing that there may ultimately be a theory of consciousness cast in terms of such laws. If the position is to have a name, a good choice might be naturalistic dualism.”

          I don’t think he’s expanding ontology at all, but diminishing it from Cartesian dualism. The ontology is simply that everything is physical, but there are two ‘properties’ that arise from the physical: one physical, the other experiential. Is this what you got from reading him? I feel like when I read him the first time, I completely misunderstood what he was saying. This paper was the only writing of his that I thought I understood…his other stuff I found very obscure. It’s like he’s wavering between contradictory views and either can’t make up his mind or doesn’t want to be pinned down.

          Here’s the paper if anyone’s interested:

          https://consc.net/papers/facing.pdf

          Liked by 2 people

          Reply

          1. I guess it would be too cynical to say that conjuring up an air of mystery around the hard problem has kept Chalmers in leather jackets and keynotes at conferences in exotic locations for years!

            Liked by 1 person

            Reply

            1. It can be hard for someone who publicly espouses a position to change it, particularly if they become well known for it. Although even back in the initial paper, his position seemed pretty minimal and nuanced, so there’s a good chance it’s still his actual position. But his statements, as he’s gotten older, do seem more ecumenical.

              Like

          2. That is what I get from reading him. He’s a property dualist, which seems like a very thin minimalistic sort of dualism. As you note, it’s basically physicalism plus non-physical properties. To me, it’s like the deism of the philosophy of mind. It can’t be detected by science, but at the cost of making no difference in the world.

            Chalmers, like most people who hold this view, resists the epiphenomenalist conclusion. (Can’t remember if he addresses it in this particular paper. It might have been in the response paper (Moving Forward) to the commentaries on this Facing one. His defense it to invoke our old friend “things in themselves”. Physics only addresses what matter does, not what it is. Maybe what it is, is consciousness, and physics is consciousness expressing itself, or something along those lines. It’s an old line of argument going back to Bertrand Russell (at least), and I’m not really doing it justice.

            Interestingly, Chalmers’ Facing paper came out seven years after Dennett’s Quining Qualia one. Dennett actually gets the version he attacks from philosophers writing in the 1970s and 80s, although he alludes to much of it being repackaged from sense data theory going back to the early 1900s. But Thomas Nagel implies epiphenomenalism in his 1974 Bat paper, and Frank Jackson is pretty clear on it (the paper where he introduces Mary’s Room actually has the title “Epiphenomenal Qualia”).

            But as to Chalmers himself, as I noted in one of these recent posts, he seems to basically be a physicalist and functionalist, just one with an extra assumption of something extra coming along with the functionality It’s almost down to a verbal dispute between him and Dennett. Almost. And Chalmers’ fascination with the meta problem and illusionism in recent years had me wondering if maybe he’d turned but wasn’t ready to admit it yet.

            But it’s hard to say. He seems more interested in AI these days, at least based on his X posts.

            Like

            Reply

            1. Now I know why my husband ran screaming from analytic philosophy. I’m finally starting to realize I have been using the principle of generosity beyond its limit! I guess where I stand is really really really far from anything these guys are talking about, farther than I could comprehend.

              “It can’t be detected by science, but at the cost of making no difference in the world.”

              It’s so funny, for the longest time I had no idea what you were getting at. Now I do. If ‘dualism’ these days means mind is acausal, I have no sympathy with it. What in the world is the motivation for that stance other than to hold hands and play nice with physicalists? LAME.

              Chalmers can deny the epiphenomenalist position, but I imagine that will involve redefining words and boy am I getting tired of this verbal masturbation.

              Liked by 1 person

            2. There are still some interactionist dualists around, but it seems like a hard position to defend rigorously. Property dualism feels like a concession to the world science leaves us.

              The question is, what’s the alternative? Panpsychism looks like the properties made universal, which gets rid of the question of why only brains would have them, but the epiphenomenal issue seems to remain. (Unless the intrinsic nature of matter thing can be shown to work.) And if I understand most idealists, they still see the need to account for the instrumental success of science.

              Ha! I’ve learned that defining words, and redefining them, is what philosophers do. The bad ones just refuse to admit it.

              Like

            3. I think interaction dualism is preferable because it takes the full problem into account, even if it fails to solve it. Most of the panpsychist papers I’ve read seem to presume physicalism, which I always found strange, but now I can see why if the starting point is property dualism. Anyway, all this has inspired a blog post, so I’ll write about it soon.

              Like

    1. We discussed Quining Qualia a while ago at my blog, under “Unconsciousness Unexplained,” so you’ll already know my views. Dennett’s TL;DR argument is that, because qualia are not accessible to public inspection, they do not exist. The unspoken premise is that everything that exists must be accessible to public inspection. Defenders of qualia generally insist that there are things in our reality that are not accessible to public inspection.

      Now, this does raise the serious problem of public and private, a problem succinctly illustrated by Wittgenstein’s “beetle in a box.” In this classic thought experiment, “Suppose everyone has a box that only they can see into, and no one can see into anyone else’s box: we call it a ‘beetle’. No one can look into anyone else’s box, and everyone says he [or she] knows what a beetle is only by looking at his [or her] beetle.” Wittgenstein’s point is that “it would be quite possible for everyone to have something different in their box. One might even imagine such a thing constantly changing.” His question is how we manage to talk about beetles.

      Some have read this thought experiment as an argument against “private experience” as a coherent concept. Dennett appears to be one of them. His argument in Quining Qualia is, in effect, about whether there really are beetles. But Wittgenstein, in rejecting the Cartesian theatre of a “beetle in a box,” is inviting us to consider a functionalist alternative. “But suppose the idea ‘beetle’ had a use in these people’s language? If so it would not be used as the name of a thing.” It is the use of the word, what people do with it, that matters; to focus on a referent, on a thing that the word supposedly points to, is to stray into confusion.

      I have the greatest respect for functionalism of this sort, but I confess to being a little confused about the functionalist account of consciousness. As I understand it, functionalists stop short of saying there are no qualia. Their position seems to be that there might as well be no qualia. They do leave a little room for them, in the form of a causal role. On examination, however, their causality is seen to vanish into other causalities, usually having to with brain science. Because the appeal to causality seems to be undermined in a self-owned sort of way, critics complain that functionalism might as well be illusionism.

      My complaint is a little different. The functionalist interest in brain science seems to be at odds with the stated interest in what conscious experiences do, rather than what they are. If we’re actually interested in what the experience of red does, we will talk about how it affects our behaviour. Perhaps it helps us to notice cherries. Brain science need not enter into such an analysis. It seems to me that functionalists really want to explain experience in terms of what it is, namely, a phenomenon associated with, and ultimately explicable as, a set of referents, things we can point to in or around the brain. To their credit, the functionalists afford this token more recognition than the illusionists, but it remains strangely useless; what it “is” is, TL;DR, the brain, and what it “does” is no more and no less than what the brain does.

      But then I tend to come at this orthogonally. It’s as if there’s a room where an animated discussion is going on, and I’m overhearing from another.

      Liked by 2 people

      Reply

      1. Very well put. I enjoyed your explanation and analysis of Wittgenstein in this context. That was illuminating. The questions you ask here are ones I’ve been wondering about myself. It seems to me functionalists are nearly always reductive physicalists in that they identify mind (what it’s like to be) with brain, but only when it comes to us. In other cases consciousness can be multiply realized. Which gives functionalism a BS flavor; it seems like a convenient add-on for exceptional cases that aren’t afforded by type identity physicalism.

        Like

        Reply

          1. Yeah, I realize now that I’m re-reading my comment that I may have been a bit harsh, and also unclear. I didn’t mean functionalism itself is BS, though I still don’t have a clear idea of what it is. What I meant was I don’t hear many functionalists talking as functionalists, but instead as type-identity physicalists, for the most part. Which makes me wonder what functionalism is doing within their view.

            Honestly, the way functionalism itself is described—’things are what they do’, or what something does is crucial to understanding its essential nature—sounds strangely similar to a theory of forms. I certainly have nothing against that and find it interesting. But when such a view is espoused by those who wouldn’t take kindly to such a comparison, I have to wonder if they really mean something else or if I’m misunderstanding something.

            Like

            Reply

            1. Hope you guys are okay with me jumping in, but I have views. 🙂

              What I see as the difference between type functionalism and type identity theory, is that functionalism adds an additional explanatory layer between the mental and the physical.

              So a common type identity theorist toy example is to say that pain is c-fiber firing. A functionalist would be more like to say something like pain is a negative evaluation of signals from a body part. That could include c-fiber firing (in mammals), but would almost certainly be implemented in other ways in different species.

              You could call functionalism a multilevel identity theory, with an identity relationship between an experience and functionality, and then a multi-realizable one between the functionality and physics. But usually identity theories focus on just the mental and physical, which I personally find incomplete.

              Can’t say I understand the forms comparison, but that may be due to an impoverished understanding of all the ways the term can be used. What I will say though is functionalism is structural realism, just with a focus on causal relations.

              Like

            2. By all means jump in, it’s your blog, and this is about your ideas after all.

              To Tina first: I don’t think functionalism resembles the Platonic theory of forms. Quite the contrary. Platonists seek to define what a thing is in terms of static essences. Functionalists reject this approach. They prefer to define what a thing is in terms of its practical effects. Buckminster Fuller seems to have entertained functionalism when he reflected that “I seem to be a verb.”

              However, I think you’re right that the functionalist explanation of consciousness finds itself tangled up with identity physicalism in unavoidable ways. This brings me to Mike’s point.

              If pain is identical to c-firing in some cases, and to other physical phenomena in others, and if pain is also identical to negative evaluation, then two things follow. First, c-firing must be identical to any other physical phenomenon that is identical with pain, which is obviously a difficulty, since physically they are clearly different. Second, negative evaluation must be identical to physical phenomena. The relation between mental and physical may be unsatisfactory, but adding a layer of evaluation doesn’t help. “Negative evaluation” becomes exactly same thing as “pain,” that is, c-firing.

              One way out of this is to qualify the meaning of “identical,” which in its pure form is prone to ship-of-Theseus problems. In place of just saying A is the same as B, one has to ask further: the same what? This brings in the application of appropriate ontologies, which I’ve been on about elsewhere in this thread.

              Another way out is to embrace the psychic quality of c-firings and other physical phenomena, so that “pain” or “negative evaluation” is more plausibly identifiable with them. But this is too close to panpsychism for the comfort of those with more classical views of reality.

              Like

            3. “I don’t think functionalism resembles the Platonic theory of forms. Quite the contrary. Platonists seek to define what a thing is in terms of static essences. Functionalists reject this approach. They prefer to define what a thing is in terms of its practical effects.”

              I get what you’re saying in that Platonism takes forms to exist in a timeless reality. What I’m comparing functionalism to is really my interpretation of Plato, which is definitely disputable, and I’m considering all the forms in relation to each other and to the idea of the good. What reminded me of this is something Mike described a while back when he was explaining functionalism, but I have no idea where. I think he even talked about car engines, this very same example. Anyway, here’s what I wrote in my senior thesis about Plato’s forms before I had ever heard of functionalism:

              “As we have learned, philosophy makes use of a “science” called the dialectic in which the philosopher attempts to come up with the definition of every essence or form. But how does definition relate to teleological order? My supposition is that when the dialectician attempts to define all forms, he must describe the form’s function. If essence is function, and function is goodness (or usefulness), then the dialectician describes the goodness of each form as it relates to the good itself.

              If we consider a car engine as an example, we might get a clearer picture of the kind of teleology that I think Plato puts forth. Pistons in a car engine produce the power to make the car move, but the pistons rely on spark plugs, which ignite gas to keep the pistons moving. Spark plugs rely on the distributor to distribute a certain amount of spark to the spark plugs at a certain time. Each part of the car engine relies on something else, such that the car could not move if one of these parts were missing. In other words, each part has a function which works to make the car move. But in order to know how any part of an engine functions, we must first understand what an engine is to be used for; otherwise, we could not explain why any part should be necessary to the whole.

              Perhaps the idea of the good is to the moving car as the forms are to the engine parts. If this analogy is not far from how Plato conceived of it, each form has a function that relies on all the other forms, such that they are completely interconnected. Presumably, there is no idea that is completely separated from the whole, for everything exists in an organic unity such that no part can be removed without destroying the whole. If my interpretation is correct, then each form’s function (goodness) is grasped only after knowing the idea of the good. In this way, every form would have the idea of the good as its teleological end, and would thereby be infused with goodness to the extent that they partake in it.”

              Of course you confine functionalist thinking to the physical world and say nothing else exists, then no, definitely not Plato. But the manner of thinking is what I’m considering in my comparison, which I don’t see as needing to be confined to a particular kind of object. Does that make sense? In other words, there can be a kind of teleological functionalism.

              By the way, I know nothing about cars. I had to ask my dad. 🙂

              Like

            4. I see what you’re getting at with function and teleology.

              One way of defining “spark plug” would be to draw an exploded view of its parts and their physical relationships. Another way, the functional definition, would be to explain what it’s used for, or what it does, that is, generate sparks to ignite fuel. The physical implementation is less important.

              You could certainly discuss participation in Platonic forms in this spirit. If there is a form for a spark plug (and why not?), it could be taken either as the perfect expression of what in this world is always a flawed physical manifestation—form as blueprint—or as the ideal expression of its purpose, which in this world is compromised by things like entropy—one might say, form as “mission statement.” Which way to read Plato is a matter for Plato scholars, although I think the “blueprint” story is the usual one.

              Like

            5. “One way of defining “spark plug” would be to draw an exploded view of its parts and their physical relationships. Another way, the functional definition, would be to explain what it’s used for, or what it does, that is, generate sparks to ignite fuel. The physical implementation is less important.”

              Yes, exactly.

              My interpretation of Plato—and yeah, I’m sure many scholars will disagree—is that there are levels of forms, all of which you might call realizations of the Good. These forms get their definitions from their relationships to each other and to the Good. This is the realm where I think entities are best understood by their function (or mission statement). I imagine there are levels here in Plato Cloud Cuckooland with a great non-buzzing of poof-ball ideas that are all interconnected, the transcendental Form of which is the Good (which I think of as Reason itself, but of the qualitative-aesthetic-moral kind). Then there’s the physical world full of particulars. These are on the whole a copy of the forms on the whole, but the physical world has some some other power, a bastard-reasoning sort of non-being pulling it away from the forms. Like an anti-form. Call it entropy, people seem to like that word, entropy. I think we get the blueprint idea between form-world and physical world. But I could be wrong. Maybe there’s a blueprint story involved in the pure realm of forms too. It’s hard to say. There’s not a lot to go by, as you know.

              Like

            6. I’m not a Plato scholar, just an interested bystander, armed with the Collected Dialogues and its excellent index. If the early Plato is concerned with Socratic questions about how the forms or Ideas relate to our earthly existence, the later Plato (of the Parmenides, Philebus, Sophist, and Statesman) starts exploring how the forms relate to one another. Socrates is no longer the spokesman for these speculations. In fact it’s the Sophist who wants to “consider their several natures and then how they stand in respect of being capable of combination with one another” (Sophist 254c)—and we know how Plato felt about the Sophists.

              His respect for Parmenides, on the other hand, is acknowledged; but Parmenides leaves Socrates reeling with, among many questions, his enquiry into “those forms which are what they are with reference to one another” (Parmenides 133c). I’m not sure what Plato thought he was doing in his later years (besides apparently fulminating to his critics that the Sophist had to be dense and boring), but the product reads like Hegelian metaphysics. (Yet more evidence that the history of philosophy consists of footnotes to Plato!)

              I can’t imagine that the later Plato was being deliberately and perversely destructive toward the Socratic doctrine of forms, but he does seem to be struggling with the meta-doctrine, and anticipating the kinds of paradox that Russell would explore more formally in the logic of classes.

              Like

            7. It’s funny, I was just thinking about whether Plato thought the forms were really real or a metaphorical model to explain why we perceive the world as becoming (in motion, with differentiated objects) rather than being, which I imagine he would have favored. I suspect he did think the forms were a kind of metaphor that somehow relates to the nature of being, though I can’t say why. I have a vague memory of making a (possibly erroneous and completely free-associated) connection to the notion of forms-as-metaphor with some mystical Pythagorean mathematical doctrine I must have stumbled across on the internet. I think it had something to do with the way certain numbers collapse into one. But what I’m saying now is probably nonsense, at best a vague memory of a connection I started to make but never fully formed.

              I’ll be the first to admit I find a lot of Plato’s writing to be dense and boring. Even the Republic gets boring at times. Plato comes alive for me more in retrospect, or at least on a second reading, after I’ve formed my hypothesis of what he’s getting at and looking for clues in the text.

              Like

            8. Plato was a poet. His best dialogues show a playful sense of human interaction.

              I could imagine him trafficking in metaphor as a mode of explanation. Surely the vision toward the end of the Republic is not meant literally.

              Like

            9. I think the easiest way to address the concern you raise about functionality and identities, is to remember that any individual instance of pain is a particular physical process, like c-firing and all the rest. But the category of experiences we regard as pain is a functional role, one that can be played by different physical systems.

              Consider a mouse trap. Any one mouse trap has an identity relation to some configuration of matter. But the device type: mouse trap, can be implemented in a large variety of ways. Likewise, a book has a particular physical configuration in a hardback copy, but a completely different one as an ebook on my phone.

              So it seems like any difficulties we’re going to have with a functional conception of an experience, we should also have with other things like cars, planes, coffee makers, and stock markets, all of which are multi-realizable, but individual instantiations have an identity relationship with a particular physical configuration.

              Like

            10. We can talk about a class of mouse traps, which have in common the function of trapping mice. However, there is no entity corresponding to “trapping mice.” If we want to identify something called “trapping mice” with the many types of mouse trap, we would have to map each individual type of mouse trap to a corresponding type of trapping mice. Otherwise, by all being identical to a thing called “trapping mice,” the mousetraps would be identical with one another.

              So it is with pain. If there are multiple physical instantiations of pain, and pain is identical with a physical instantiation, then there must be a class of pains, mapped-one-to one with the class of physical instantiations. Either that or all the physical instantiations are identical with “pain,” and therefore identical with one another.

              Then we can talk about a class of pains, which have something in common—perhaps negative evaluation. But again, if “negative evaluation” is identical to all these different pains, then the pains are all identical; and by a direct mapping, their physical states are all identical..

              The problem can be traced to the free invocation of “identity.” It’s safer to say that, if pain is common to
              a variety of physical instantiations, it cannot also be the same thing as the physical instantiations. Type identity avoids this by limiting pain to one instantiation. Functionalism, if I’m following all this, tries to avoid it by allowing multiple instantiations, but making them all the same as “negative evaluation.” But this fails for transitive reasons.

              Like

            11. “We can talk about a class of mouse traps, which have in common the function of trapping mice. However, there is no entity corresponding to “trapping mice.” If we want to identify something called “trapping mice” with the many types of mouse trap, we would have to map each individual type of mouse trap to a corresponding type of trapping mice. Otherwise, by all being identical to a thing called “trapping mice,” the mousetraps would be identical with one another.”

              I think you might have explained some of my confusion here. I might have been thinking along these lines: “trapping mice” is a form-entity in which mouse trap types “participate”. “Trapping mice” must be something separate in order to be multiply realizable. I could very well have been assuming something like this about functionalism without ‘realizing’ it.

              Maybe you could say trapping mice x is realized by x trap, trapping mice y is realized by y trap. But then I wouldn’t think you could say trapping mice x is realized by y trap.

              It does seem to me that in order to say consciousness can be instantiated by human brains as well as by computers, you’d need a form-entity separate from, not identical to, both. Otherwise you end up with computer consciousness as something different from human consciousness, and I’m not sure what that would be, but it doesn’t sound very exciting.

              Honestly I can only barely grasp functionalism non-Platonically, and only for a few seconds at a time. Any longer and I fear I’ll become multiply realized.

              Liked by 1 person

            12. Tina, I’d say if thinking about it Platonically helps you get there, go with it. I noted elsewhere that I think platonism (small “p” version) amounts to a personal philosophy. I don’t think it’s necessary, but if it helps in conceptualizing things, I don’t think it’s an issue.

              Liked by 1 person

            13. AJ, I’m not following all the logic steps here. For example, why would you say that trapping mice wouldn’t be an identity with a process or capability? It seems like you’re thinking if it isn’t a static thing, then we can do identities. I have to admit I’m not well read on identity philosophy, but that doesn’t seem right to me.

              I will say that pain is not a simple thing. It’s a complex collection of mechanisms (or at least cohere with that collection). I did a post a while back on Jennifer Corns’ book on pain, which gets at some of the complexities, and might be enlightening, at least for getting at where I’m coming from.

              The complex composition of pain

              Like

            14. That’s my fault. When I said ‘there is no entity corresponding to “trapping mice,” I meant that there is no single instantiation of a mouse trap that is identical to ‘trapping mice.’ Bad phrasing, and I apologize.

              “Trapping mice” could indeed be identical to a process or capability. The trouble arises when we “identify” that process with a material instantiation. There are many instantiations of mouse trap. If every one of them is identical with its function, then in the strict sense of identity, they are all identical with one another.

              That’s why strict identity is an awkward thing to claim.
              How can two things be one thing? “There is some x such that 2x = 1x”. The solution, of course, is 1. However many multiplicities you have in mind, you can’t get past 1.

              But using “identity” in some other, looser sense is meaningless, or worse. We can observe differences between various instantiations of a function. We need to know what it is about the function that makes it different from the instances of the multitude.

              This is beginning to sound very Platonic!

              Like

            17. I forgot to consider the point about pain.

              If a doctor asks me what my pain is like, I’m not likely to tell her it’s a complex collection of mechanisms. It’s not that this is untrue; it’s just inappropriate to the situation. What I mean by “pain” at the doctor’s is a fairly simple and immediate sensation, which unfortunately I can’t communicate to her directly.

              To say this is identical to pain, understood as a complex collection of mechanisms, is to mix up the situations and their appropriate functionalities.

              Like

            18. Hey, no apology necessary. It’s all friendly discussion. I’m not always as clear as liked to be myself.

              On the identities, maybe I shouldn’t have conceded that functionalism could be described as a type of identity theory. I seem to have walked into an unnecessary quagmire. Interestingly, a lot of the points you’re making seem similar to criticisms I remember seeing for type identity theories. In any case, I think the relation is hazy with words often referring to a category of phenomena, even if we think of it as something simple and primal.

              On pain and doctor visits, I’ve actually been struggling with dental pain for the last couple of years. Part of the problem is identifying exactly where the pain is coming from. It’s made diagnosis pretty elusive. I’m told it’s not an unusual problem due to the way the nerves are wired in the jaw. It’s been a reminder in just how complex the process of the brain’s evaluation of where a signal is coming from, and what its state is. I relayed another example in that post, of a relative who had phantom back pain due to an opioid addiction.

              Sometimes the supposed simplicity of the experience can be a barrier.

              Like

            19. I hope you find the cause of the dental pain so it can be fixed..

              Phantom pain obviously can’t be in the limb, so it’s an interesting case. I’ve never looked into it, but I’d guess it’s caused by signals at the site of amputation, which as far as the brain knows ought to be coming from farther along those nerve paths.

              So where is the pain located? If not in the limb, then it’s “in the brain.” But by extension, everything the brain senses must then be “in the brain,” and in some sense, your house might as well be a phantom house.

              That’s the first sort of weirdness that occurs to me. I suppose the literature must be full of them.

              Like

            20. On the dental pain, thanks. I do too.

              When everything is working right (adaptively), pain seems to be a combination of an interoceptive signal from a part of the body (including the c-fiber firing often cited), and an affect evaluation in the brain. But you can have pain without any damage to the region in question, and you can have damage without pain. A lot of chronic pain is more a brain situation than a peripheral nervous system one.

              I do think we can that the experience of pain is in the brain. There are definitely weird cases that complicate the picture.

              Like

            21. We could make the same general sorts of remark about red, and of course there can be phantom experiences of red. It makes me think of Berkeley. For a realist, the functional use of “location” has to recognize something else— not that I can put my finger on what. Concepts like “interspace” and “betweenness” keep coming up, but I don’t know how helpful they are really,

              Interesting discussion, thanks.

              Like

            24. “sounds strangely similar to a theory of forms”

              I was looking for something in Pete Mandik’s book on the philosophy of mind, came across this passage in a review of the history of functionalism, and remembered your point about forms. The quote is in reference to the Aristotelian rather than Platonic version, but still an interesting resonance with your observation Thought you might find it interesting that you’re in the same vicinity as other thinkers..

              8.10 An important idea in the philosophy of Aristotle (384–322 BC) is the idea of the form of a thing: that which contributes to the performance of the thing’s defining function or purpose. The form of a sword enables it to cut, and the form of an eye enables it to see. One crucial aspect of Aristotle’s thinking about forms is that they are not a thing separate from the thing that has them. The form of a sword is not one thing and the substance of the sword a second thing. These are two aspects united in a single thing—the sword itself. Aristotle resisted the view of Plato (427–347 BC) that the soul is an immaterial thing distinct from the body (a kind of substance dualism, as we discussed in chapter 2). Instead, according to Aristotle, the human soul is not separable and distinct from the body—it’s the form of the body. Keep in mind that form encompasses more than mere shape, but also the functioning of the body and its parts. As Aristotle puts the point, “If the eye were an animal, then sight would be its soul.”

              Mandik, Pete. This Is Philosophy of Mind: An Introduction (p. 106). Wiley. Kindle Edition.

              Like

            25. Thanks for that quote. I’m glad you brought this up since Aristotle is the more obvious choice when talking about form as function. During my oral exam the outside examiner asked me why I didn’t choose to write about Aristotle instead of Plato, and I really didn’t have a good answer. I said something like, “Because Plato’s more fun.” The one thing that might not make Aristotle the best parallel is the issue of multiple realizability. Plato’s forms aren’t constrained by anything since they exist independently. I’m not sure what sort of constraints Aristotle has in his version. It’s been a long time since I read him, but I recall finding it difficult to get a clear sense of the difference between his forms and Plato’s forms from the reading. It does seem on the face of it that Aristotles forms would appeal to you more in so far as they must be empirically grounded.

              Liked by 1 person

      2. I recall most of Dennett’s approach in the QQ paper coming from the first person perspective, basically pointing out that the concepts don’t even work just looking at it from that direction.  But it’s a long paper and I don’t recall every line of reasoning. I do agree if someone just said that something is invalid because it doesn’t meet their worldview, that’s specious reasoning.  A worldview should be constantly tested and adjusted just like any other theory.  Otherwise it’s just a rigid ideology.

        The main functionalist move is that to explain what something is, is to explain what it does.  It’s a recognition that the only thing meaningful we can say about something is how it relates to everything else.  In that sense, functionalism and structural realism are two sides of the same coin.  

        Of course, it’s conceivable there is an acausal essense to things, but if so, our talk about it seems hopelessly disconnected from whatever the reality might be.  So its existence or non-existence becomes a metaphysical assumption.  Parsimony seems to weigh in on non-existence, but as is often pointed out to me, parsimony is only a heuristic.

        So I think for a functionalist, whether there are qualia amounts to what effects they have.  Qualia that amounts to representational details fits this idea.  Qualia of the type that could be inverted or absent without any effects, fall into the acausal essence category.  

        Like

        Reply

        1. If I get this, then the question of whether you see “green” where I see “red” is an unanswerable and pointless way of thinking about qualia. As Wittgenstein says, it’s possible for everyone to have something different in the box. But unlike Dennett’s illusionism, which claims there might as well be nothing in the box, the functionalist explanation allows for something that we recognize through its effects.

          The functional effect of what we call “red” might be to convey importance,as in the case of fire or blood. (I have my doubts about this line of argument, but we can set them aside.) But one of the effects of this quale, if I understand you, is representational detail. Whatever “red” happens to look like to you or me, it is in any case a representation having a bearing on the relationship between things.

          Normally when we speak of representation, we have a transitive function in mind: there is something represented, and something to which it is represented. But the relationships between things can be understood in terms of simple mechanical forces, of which neither need be aware, so that no “representation” in the interesting sense of “qualia” need be involved. In fact this is the usual understanding.

          The idea of representation introduces the possibility of awareness, but at some cost. Either the representation is thought to be external to the mechanical relationship, leading us down the path of Cartesianism, or it is understood as somehow fundamental to the relationship. In the latter case the term “mechanical” becomes inappropriate. On the other hand, the notion of “causal” takes on fresh interest, for now the causality is not merely mechanical, but intentional.

          Like

          Reply

          1. Just to clarify, I don’t think the only role of red is to signal importance.  There’s also distinctiveness.  Reds, yellows, and greens are more distinct than the shades of blue at equal distances from each other on the electromagnetic spectrum.  Which is to say the light variations in the 530-570 nm wavelength range seem more important to our nervous system than the rest of the visible spectrum.  I think the reason is evolution, since that’s where our ancestors’ evolutionary affordances lay (such as ripe fruit).  

            There are almost certainly other roles as well.  The trick is to look at the causes and effects, particularly the ones prior to our introspective awareness of the experience.  And the aearly pre-conscious conclusion of redness also triggers a host of associations we’ve learned over our lifetime.  We’re not aware of those being triggered, but they cumulatively add to the general way it feels to experience red.

            Your point about the word “representation” implying that there is something to which it is presented is a concern I’ve long had with that word.  I used it above because it’s the standard philosophical term.  But I prefer “model” or “schema”, because it’s easier to understand those as part of an overall process.  A philosopher on X/Twitter, Peter Schulte summed it up in a nice way: “Representationalism about perception says that perceptual states are representational, not that we “perceive representations”. Big difference!”

            https://x.com/DrPeterSchulte/status/1697253424591720694

            We might have different takes on the meaning of “mechanism”.  For me, if it’s a system of causally interacting components then it’s mechanistic.  So I see intentionality as mechanistic, albeit a mechanism of very high complexity. 

            Like

            Reply

            1. “Representation” is an awkward word, but saying “Qualia that amounts to a model or a schema” is subject to the same concerns. First and foremost, let’s set aside the intuition to speak of “perceiving representations.” I think we both agree that this can lead to Cartesian places we want to avoid. If a representation, or presentation, or model, or schema, amounts to a quale, then I would understand by this, if no observer is to be inserted, that the model itself “amounts to” whatever quality we have in mind.

              A good question to ask here is whether some models (or whatever word we use for these functional complexes) have this quality and others don’t. We now deal with atoms quite successfully on a functional level, without ever having evolved qualia for them. We all have a functional model of a car, but we haven’t had several million years to evolve a “car” quale. What this shows is that a model doesn’t need to “amount to a quale” before it can function effectively.

              If highly successful functionality is possible without any mention of guiding qualia, then we have to ask why they might need to be brought in. The immediate answer is that qualia such as “red” have to be explained somehow, and the argument from functional modelling looks like a promising hook, at first glance anyway—the more so since we can invoke millions of years of evolution to make the picture a little hazier (although there must have been one particular day and hour when “red” popped into existence, before which, on this argument, its function was presumably not there at all).

              What goes unspoken in all this is that, in so far as they have to be explained, qualia embarrass us constantly with their existence. In this connection, Dennett’s illusionism is pure madness, but at least consistent. Functionalism has a more sane attitude, but if it is to make qualia part and parcel of functional relationships in a way that avoids the quagmire of “perceiving representations,” it either has to be very generous with the concept, and insist on qualia wherever it finds functional relationships; or it has to show why some models need qualia to work, while others don’t. My own inclination is to claim that no mechanism needs qualia to work, no matter how high its complexity. The mechanical approach brings in qualia not because they are indispensable to a mechanical account, but because they have to be explained somehow.

              Like

            2. On a first day when red popped up, I actually think it was more gradual than that. All that was needed initially was a mutation that led to a quiver of a reaction, an infinitesimally small increased tendency to notice things that reflected that wavelength. But that tendency would have increased the primate’s chance of finding a high calorie food, and its probability of survival and procreation, preserving the genes in the mutation. That would have created a feedback loop, where higher tendencies to notice that wavelength led to ever higher chances of finding food. Higher abilities to detect it would have been selected, from a slight noticing of a difference in shade, to modeling it as an increasingly distinct impression.

              Interestingly, this is a symbiotic relationship between the fruit plant and the primates. The primate eating its fruit increased the chances that its seed will be pooped out far away, increasing the plant’s genetic progeny.

              A side effect for the primate would have been that it noticed flowers more. But the flowers are the colors they are for a symbiotic relationship they have with insects, who the flowers depend to cross pollinate. So the insects evolved their own ability to detect distinct colors, which the plants co-evolved to signal to them that here is high calorie nectar.

              (Confession: the above is a quick summary of one Dennett provides in Consciousness Explained, but resonates with a lot of other stuff I’ve read about it.)

              I don’t think functionalism has to posit that all functionality has qualia, anymore than we have to regard all functionality as the video game Tetris. It takes pretty specific functionality to have Tetris. Likewise, I think it takes fairly specific functionality to have conscious perception.

              Your final point, an inclination to think that no mechanism needs qualia, I think is what gets us into the version illusionists deny. But as I mentioned in the post, we’re left with something that we can’t describe, analyze, or scientifically investigate. One problem I didn’t address is how something like that could evolve. Natural selection can only select for things that make a difference. Even if we’re talking about all matter having it, it seems like something has to lead to its more concentrated form in animals.

              And we should think about phenomena like blindsight. Often the descriptions of it highlight what people can still do, such as detecting the absence or presence of objects in forced choice tests. But it’s worth thinking about the abilities they do lose, such as object discrimination, categorization, the ability to articulate what’s there, etc. In other words, lack of qualia, in the real world, does seem to make a functional difference.

              Like

            3. It’s hard to avoid concepts like “noticing” in these discussions, as when a primate (more likely something earlier in the evolutionary chain) “notices” red at some infinitesimal level. I don’t think we would say that a plant “notices” that having red berries is beneficial to its propagation. For the plant, there is a mechanism that works without any sensitivity that we would call “noticing,” yet it works just fine. It’s not hard to imagine a similar mechanism in the primate, where things work in such a way that functionally it enjoys the advantage of discovering berries, but it needn’t notice anything. If a blueberry-picking machine were to be enhanced with AI so that it selected the best ones by way of their wavelengths, we wouldn’t have to say it notices anything. And a primate, after all, is only a machine—albeit a highly complex one. The concept of “noticing” is there for us to draw upon in these cases, but it’s too easily taken for granted.

              It may be that not all functionality develops the ability to notice. This would begin to explain why plants and blueberry harvesters don’t notice colours, but primates do. Yet as an explanation, this seems rather ad hoc. Noticing is observed to be present, and must be accounted for; happily, we can invoke it as part of the explanation, if we want. Even this overstates the case, for we never observe “noticing” in anything but our own immediate experience; in everything else, we infer it based on reasonable assumptions (for example, I infer that you notice colour, but that AI blueberry harvesters don’t.) And once we discount our own immediate experience as problematically private, and therefore possibly non-existent, we have nothing to go on at all. “Noticing” becomes an empty concept, a wheel doing no work.

              This is where we end up if we start with an idea of mechanism that does not need qualia, because it relies entirely on a conception of the universe that doesn’t need them, and then try to come up with an explanation for them. To me it feels like the sort of fly-bottle that Wittgenstein wanted to show us the way out of. This is why I tend to avoid the tireless and tiresome analytic arguments that dominate the field (as Tina has lamented), and look for some other way to approach things. Both Tina’s views on the limits of scientific discourse, and Matti’s views on ontological pragmatism, seem to me promising.

              But there is hope yet for mechanism, if instead of treating “qualia” (really we should be using scare-quotes for this reification) as something we can’t describe, analyze, or scientifically investigate in the explanatory terms we project onto our universe, we accept them as right in front of our noses, and begin to imagine a different sort of mechanism, in which they play a coherent role. I read the efforts of Whitehead and (to my surprise) Heisenberg along these lines.

              Like

            4. Nothing really hangs on my use of “notice” in the description. I could have used “detect”, as I did for some of it. And I’m not sure we wouldn’t use “notice” for a plant, such as a vine that notices a fence and grows into it. Of course, most of us in that case wouldn’t mean conscious noticing. And I think we’d be tempted to use it for any AI that has to select which parts of its input are relevant, such as noticing the right details to recognize a face. Again, most of us wouldn’t mean conscious noticing at this point.

              On a different sort of mechanism, that might be interesting, if it eventually makes a difference. If not, it seems simpler to me to just take the fork of causal consciousness. But in the end, no one can demonstrate that a psycho-physical parallel reality doesn’t exist. In any case, I’ll be interested to see what you found with Whitehead and Heisenberg.

              Like

            5. We say a geiger counter detects radiation, but we don’t say it notices radiation. The rods and cones in our eyes detect red, but we don’t argue that they notice red. It’s clearer to say that we notice it, after they detect it on our behalf. So I’d argue that quite a lot actually hangs on the use of “notice” here. A plant may sense a fence, and a harvester may sense a berry, but to say they notice these things is to attribute something more to them. The question between us is whether it’s possible to notice something besides detecting it.

              With a functionalist account, we can rest with the detection of red and some ensuing functional effects. With luck, we can avoid asking whether red is also the sort of thing that can be noticed; that is, experienced. I have the impression that functionalist accounts prefer to retreat from such problematic words, in hopes of finding more neutral ways to discuss, or rather not discuss, the matter of experience.

              A psycho-physical reality assumes the primacy of a scientific-materialist physical reality that, for some reason, needs to have a psychic reality along for the ride, but has despaired of connecting them. Neither Whitehead nor Heisenberg subscribe to it. Their philosophies begin by questioning the primacy of a scientific-materialist physical reality. I was aware of the similarities between them when writing about Heisenberg’s Physics and Philosophy, but my exegesis of Heisenberg has worn out its welcome, and for the moment I’ve moved on to something completely different. I could provide a couple of links to get you started, if you like.

              Like

            6. My last post, about noticing and detecting, had a bit of a tone. I think I was put off by the mention of psycho-physical parallelism, which felt like a condescending dismissal, and not even to the point.

              On reflection, I think I see where the reference might have come from. I’ve been arguing that for a functional explanation, it suffices that something is detected. It’s not necessary to say it’s noticed, in the sense of an attended, conscious experience. But then noticing becomes “a wheel that does no work.”

              Someone like Dennett would just drop it, and treat noticing as an illusion. The functionalist account is in a different predicament, because while it can rely on detection alone, it wants to keep consciousness; indeed, it proposes to explain it functionally. But since it’s a wheel that does no work, this noticing consciousness is left dangling, and the working model for that is psycho-physical parallelism. So I can understand why it would occur to you. The thing is, it actually seems to be a problem for functionalism itself, arising from the conflicting goals of keeping consciousness around while explaining it in terms that don’t need it.

              Liked by 1 person

            7. Sorry, didn’t mean to imply anything with the psycho-physical parallelism reference. I was just trying to be precise about my concern that if the different sort of mechanisms amount to that, it wouldn’t seem like a fruitful path to me. Philip Goff talks about psycho-physical parallelism, and to your point, I know some scientists have gone there. So although it’s not a concept I find promising, I didn’t anticipate it sounding dismissive.

              I would agree that “notice” does have an implication of selective attention, one that “detect” doesn’t necessarily, so I can see where you’re coming from. On the other hand, attention is usually understood as a functional process.

              But overall, my thinking is the functional account for an aspect of experience is often not obvious. It may take work to figure out what it is, or at least come up with a hypothesis for it. In that sense, I think both illusionists and non-functionalists give up too quickly. But I’ll admit it depends on what kind of explanations will satisfy us.

              No worries on the Heisenberg stuff. All of this is based on what interests us, while it interests us.

              Like

            8. Philip Goff has probably said something about psycho-physical parallelism at some point, but he’s a panpsychist. So am I. Panpsychism proposes that the primary reality is process, from which material physics arises.

              If I may venture a theory I’m just forming, thanks to this discussion, a functional account might depend on the ontology one chooses. This is related to Matti’s (and James’ and Rorty’s) messy pluralism, and to Tina’s (and Heisenberg’s and Wittgenstein’s) ideas about the limits of a discourse. The designer of a stove is interested in the physical behaviour of the materials used, how the parts are assembled, and so on. For this, an ontology of materials and their properties is required. The user of a stove is interested in the controls, the heat ranges, the number of burners; also, probably, the properties of eggs, flour, sugar, oil. For this, an ontology of cooking, food, and cooking instruments is required.

              It occurs to me in a vague way that this dovetails with Heisenberg’s view that the answer you get depends on the question you ask.

              In your case, I suspect that an ontology of perception is required, but it sits uncomfortably with the ontology of parts and materials you’ve chosen for the explanation.

              Like

            10. This might seem implausible, but it really doesn’t sit uncomfortably for me. As I noted, it can involve some thought, but I wouldn’t expect anything else for something like this. If it was obvious, we wouldn’t find it an interesting topic.

              But I’ve never had the strong intuition that the functional account is incomplete. (Although I acknowledge that many do.) It might help that I see it as far more than just parts and materials, but those in motion, the processes they undergo. The dance is more important than the specific parts.

              Like

            11. I didn’t mean that it sits uncomfortably for you. I meant that the ontologies sit uncomfortaby together. They are incompatible.

              “Parts and materials” is a simplification for expository purposes. Of course there are processes, but they are the processes appropriate to the ontology. The dance is influenced by the kinds of parts it contains, and the kinds of things they can do.

              This is actually old ground for me; I have a whole section that explains this view of ontology (inspired by an encounter with Web Ontology Language or OWL, as it happens.) But up to now I hadn’t considered its application to the functionalist view of consciousness.

              Liked by 1 person

    2. I think it’s probably more of a difference in language than a real disagreement, and I’d add that panpsychism seems about equivalent too, from what I can tell. The illusionist agrees with the panpsychist that the whole universe is in fact equally conscious, in the phenomenal sense. I guess the difference is whether there’s anything to that merely phenomenal aspect, but that seems like a moot point considering without the self awareness kind of consciousness, that phenomenal kind cannot be meaningfully perceived (which is roughly what epiphenomenalists say is the case, if I remember correctly). The functionalist doesn’t bother with the merely phenomenal consciousness, because it seems utterly inert (like how there’s no way of knowing you’re not a philosophical zombie, since qualia supposedly has no causal power).

      Liked by 1 person

      Reply

      1. I agree that the difference between illusionism and functionalism is language preference, communication tactics. (Dennett even admits this in one of the footnotes in his 1988 paper.)

        I also agree that illusionism and panpsychism are empirically equivalent. Both see no difference between the physics of the brain and everywhere else. It still feels like there’s a fact of the matter difference, but one that may never be resolvable. It also seems like which view someone holds shapes which scientific theories they find more plausible.

        Like

        Reply

    3. BTW, I recall reading on SEP that Jackson (he of the “Mary, the shockingly mistreated blind scientist” thought experiment fame) later recanted and decided to support a form of functionalism.

      Liked by 1 person

      Reply

      1. Thanks. I knew he had recanted and gone physicalist, although not sure I knew it was functionalist. Philip Goff, a big fan of the knowledge argument, in a recent Mind Chat interview, tried to convince him back to his earlier conclusions, but no go.

        Like

        Reply

    4. Nothing exists except waves of energy in strings, that can appear and disappear. Properties in the way we see and experience them are assembled in brains. Nothing is red or blue unless someone is there to see it. Just like a tree falling in the forest. Everything is quite and colorless until someone is there.

      Liked by 1 person

      Reply

        1. Well, yes. I you can see the change in color, then you know the fruit is ripe and ready to eat. And toy know when the fire is by the color. Lots of things an animal needs to know for survival depend on color, sound, and smell. Keeps the animal from eating rotten food and so on.

          Liked by 1 person

          Reply

        1. Interesting that you connect “irreducibility” to it. You might be right about that.

          By analogy, think of a pond. A drop of rain hits in one section. That is only place the drop falls, but the waves that move across the pond contain information about the size, speed, and shape of the drop.

          Liked by 1 person

          Reply

          1. Right. “Sharing” in a neural network is more about propagating effects throughout the network.

            Of course, the propagation depends on what other effects are propagating at the time. Some may reinforce the source region’s effects, others may hamper it. The battle of attention.

            Liked by 1 person

            Reply

    6. By the way, I can’t seem to reply directly to the point you made earlier, but something has been nagging me about the SEP article you quoted and I realize I never responded to it:

      “Platonism is the view that there exist abstract (that is, non-spatial, non-temporal) objects (see the entry on abstract objects). Because abstract objects are wholly non-spatiotemporal, it follows that they are also entirely non-physical (they do not exist in the physical world and are not made of physical stuff) and non-mental (they are not minds or ideas in minds; they are not disembodied souls, or Gods, or anything else along these lines). In addition, they are unchanging and entirely causally inert — that is, they cannot be involved in cause-and-effect relationships with other objects.

      https://plato.stanford.edu/archives/spr2016/entries/platonism/#1“

      At first I thought, “That’s a very bad understanding of Plato”. But now see what the problem is (after clicking on the link). Plato’s philosophy is not the same as contemporary Platonism. Plato would say abstract objects are non-spatiotemporal, but he would never say abstract objects exist outside of our minds or Mind itself. Everything, including us, including spatial and temporal objects, derives its being from Mind. This idea of abstract objects being causally inert—definitely NOT Plato!

      Okay. I feel better now. I was really thinking how appalling it was that SEP could get Plato so wrong. What a relief!

      Liked by 1 person

      Reply

      1. Sorry for the confusion. I should have been more clear that I was talking about platonism (lowercase ‘p’), the contemporary philosophy, rather than Platonism (uppercase ‘P’), Plato’s original outlook. I actually don’t know much about Platonism, other than what I’ve picked up from you and a few other sources.

        But hopefully now you see why I toyed with the idea of phenomenal properties being platonic (small ‘p’) ones.

        Like

        Reply

    Your thoughts?