"Untitled Post" - Views: 360 · Hits: 360 - Type: Public

Pearce: “David, clearly one or more of our background assumptions are different. I'm just trying to pinpoint where exactly. When you say "physical properties", what have you mind? Classical physics? The universal wavefunction?”

Me: “When I say ‘physical properties,’ I mean those properties in virtue of which an object or force has the causal dispositions it has. Usually, these are presumed to be geometric structural properties. So if you look at the floor next to you, on a usual account the floor doesn't do any of the physical things that it does *in virtue of* the fact that it's blue; it's blue (at least partly) *because of* the interactions of geometric structures (in this case, physically determining the structurally definable frequencies of light it absorbs versus reflecting). 

The problem is that an account like this renders the qualitative aspects of reality in their qualitative aspects as such epiphenomenal, and a conclusion of epiphenomenalism renders modus tollens against any premises that entail epiphenomenalism, because epiphenomenalism is decisively refuted by the fact that we can think and talk about those qualitative aspects in their qualitative aspects as such at all.

Yet, on the other hand, the structural properties and the qualitative aspects can't be considered 'identical,' either (as is the usual materialist means of trying to rescue mental causation), as proven by the fact that zombie worlds or even worlds in which panpsychism are false are even conceptually imaginable.

Hence the trajectory from the relationship between qualitative aspects of reality and the physical, structural, causal disposition-explaining properties going from eliminativism of the qualitative aspects to "identification" of the two as just the same exact literal ‘thing’ to the facade of "emergence" of the former from the latter to "epiphenomenalism" of the former with respect to the latter to—finally—direct interactionism between the two. Logically—and the panpsychist authors themselves confirm this—the pathway to holding forward panpsychism as a plausible hypothesis is as a potential solution to the failures of exactly these positions. 

The flaw I charge panpsychism with is that the claim (that structural causally disposing mathematically describable properties and qualitative aspects mutually coexist everywhere) does not actually address how the two hang together. There are a finite number of possibilities, and panpsychism actually has no options other than those open to the materialist account - so despite the first appearance of a plausible solution, if materialism fails then panpsychism actually fails for any reason materialism would (depending on the particular formulation). Thus, panpsychism either fails or else has no legitimate motivation, depending on where your intuitions or conclusions lie on those primary, fundamental underlying ontological questions.

Mine is to agree with the panpsychist assessment that they fail—but I therefore think the panpsychist alternative fails on the same exact points for the same ultimate reasons.”

Pearce: “Thanks David. How might an advocate of the claim consciousness discloses the intrinsic nature of the physical respond?

First, if consciousness discloses the intrinsic nature of the physical, then p-zombies are impossible – conceptually impossible, not just impossible in the real world. By contrast, explaining why we're not p-zombies is a problem - IMO an insoluble problem - for materialists and epiphenomenalists. On the other hand, explaining why we're not so-called micro-experiential zombies is a problem both for physicalistic panpsychists / Strawsonian physicalists and materialists / epiphenomenalists alike insofar as we make the standard neuroscientific assumption that membrane-bound neurons are discrete, decohered, and thus effectively classical.

Secondly, if consciousness discloses the intrinsic nature of the physical, then strictly speaking all consciousness, and only consciousness, has causal efficacy: it's the “fire” in the equations that makes there a world for us to describe. But there is a difference between contexts where the particular subjective texture of consciousness is functionally relevant, such the causal capacity of organic minds to talk about the nastiness of phenomenal pain, and other contexts where the particular intrinsic experiential nature of the physical would seem incidental. However, “incidental' is not synonymous with “epiphenomenal” in the strict sense of the term. Perhaps compare semiconductors made of silicon and those made of gallium arsenide. Software can run perfectly well on computers with chips made of either. Incidental implementation details don't matter - for our purposes, at any rate. But this substrate-neutrality doesn't mean epiphenomenalism is true: it's still the intrinsic physical properties of each information-processing system in question that do all the causal work.

So how are organic minds functionally different from digital computers for which the physical implementation details are indeed incidental to the functioning of the system? How is intentionality - both subjective intrinsic intentionality and a functional approximation to extrinsic intentionality - naturalistically possible for conscious organic minds within a (non-materialist) physicalist framework?”


 Well, this is far closer to a response to the crux of the actual issues I’ve been raising that were the crux of my own eventual turn against panpsychism (I’m not an outsider producing ex–post–facto rationalizations, here!) than anything I felt I’ve got so far, so I’m actually quite relieved, this time. Now it sounds like we’re actually—finally—getting into the core of where our differences might lie and at the very least gaining the opportunity to explore the exact nature of what those differences are further. It feels, from here, like it’s conceivable that we could actually: (1) identify where my reasoning fell off, in which case it’s entirely possible panpsychism could become a live option for me again (frankly, I’d like it to!); (2) identify where your reasoning is muddled and either (2a) it should be refined or (2b) I actually have had insight into something here that renders panpsychism quite problematic; or (3) at the very least we can come to a more precise understanding of the exact underlying points where our intuitions and/or conclusions split off and agree to disagree at least with an actual understanding of what it is that we’re most ultimately agreeing to disagree about. I appreciate the fact that you’ve continued giving my arguments a chance long enough for us to reach this point. 

Your responses seem to alternative between the tacks they’d like to take on my argument (not that that’s a flaw in them—we just need to keep very clear how every argument fits into the overarching dialectic). Your first response is one I can only read as a clearly attempted defense of a panpsychist “identity theory.” The second one is a little more difficult: the statement that “all consciousness, and only consciousness, has causal efficacy” immediately seems to be a statement of the “conscious substance, emergent physical properties” version of panpsychism I mention. But the undercutting of epiphenomenalism would only apply to my arguments against the “property dualist” variant—so it’s not perfectly clear whether you want to defend the “conscious substance, emergent physical properties” view and add that THIS view would not entail epiphenomenalism, or attack the implication of epiphenomenalism in order to render the “property dualist” variant of panpsychism an added live option.

 Of the two approaches, however, I consider the first by far the least promising. You suggest that if panpsychism should turn out to be true, “then p-zombies are impossible – conceptually impossible, not just impossible in the real world.” 

I’m not sure how one would actually even defend this line of approach, and the paragraph doesn’t provide any clear sketch. The problem, which seems to me to be immediately decisive against any approach you could even possibly try to devise, is this:: Suppose I say antecedently that it is epistemically possible that given what I know about my mailman, and given what I know about the man I see at the bar on Friday nights, that someone could be responsible for doing everything I have seen my mailman doing—and not be the man I see at the bar on Friday nights. 

Granting this analogy, which phrases the “identity” claim in a way that is as much in the identity theorist’s favor as possible[1] , suppose it turns out that the mailman and the man at the bar on Friday nights are in fact identical. Now, a full understanding of everything that the mailman does will include “going home, changing clothes, and heading out to the bar” and a full understanding of everything that the man at the bar does will include “coming into the bar after a hard day of work slinging mail.” Now, it turns out that they are, after all, identical. And a FULL description of what both actually are and do will indeed, after all, reveal this within the very description itself.

Does this have any implication for my original considerations, however? Is there any reason whatsoever why this would undermine the original case that {given what I knew about who my mailman was at the time}, it is conceptually possible—conceivable—that someone could do what I see him doing, and {given what I knew about the man at the bar at the time}, it would not follow by necessity that someone who, in general, did the former things would have to be identical with someone who did the latter? 

Would it refute my antecedent awareness that the CONCEPT of doing what I had previously seen my mailman doing is not identical to the CONCEPT of doing what I had until that point seen the man at the bar on Friday nights doing? The answer seems to be a clear and clearly inescapable “no.” If there is any promise for explicating an “identity theory” in here, it doesn’t lie in denying that the concepts are, indeed, utterly distinct concepts, It lies in finding some complex additional story which bridges those concepts despite their obvious and undeniable distinctions. 

It would not lie by any stretch of the wildest imagination in saying “it turns out that the concept of putting mail in my mailbox is identical with the concept of drinking White Russians on Friday nights at the local bar.” To make sure we link the comparison explicitly—it absolutely would not lie in saying “it turns out that a mailman–zombie (who puts mail in my mailbox but doesn’t drink at the bar on Friday nights) is impossible — conceptually impossible, not just possible in the real world.” (Or, to wit, “[If panpsychism is true] then p-zombies are impossible – conceptually impossible, not just impossible in the real world.”) Even in the most favorable case, this would still be just plain false. 

Rather, it would lie in first admitting the overwhelmingly clear conceptual distinction, and then providing explicit bridging principles to perform the active work of linking them: a full definition of who the mailman actually is, and not just what I had seen of him antecedently, will include description of the fact that it turns out he changes in to just those same exact clothes and then goes and sits in just exactly that chair on Friday nights. 

An account of something like this general form is what the identity theorist would owe us if he wanted to defend any actual “identity theory” worth naming. But the problem is that whereas the conceivability of a “zombie–mailman” who puts my mail in my mailbox but doesn’t go outto the bar on Fridays defeated any possibility of saying they were conceptually identical (an incredibly awkward way to verbally formulate such a simple, basic, everyday point! Sometimes philosophical language has the virtues of refined clarity for dealing with the topics in the neighborhood, but sometimes it’s simply unnecessarily opaque), when it comes to the mind–body problem, it goes further to show that no “bridging” account will ever be possible, because ANY attempt to bridge anything from the side of physical properties will always, in principle, simply add more geometric structures and mathematically describable causal dispositions into our account. And thus, any attempt to even begin trying to formulate the most basic sketch of an account of what the equivalent “bridging” explanation might look like can never even bump off of the ground for the briefest second—because we will always, in principle, be dealing with still yet more of just exactly the same old stuff that we perfectly well can conceive of happening—because the ideas are conceptually distinct—without conscious experience of any form ‘coming along for the ride.’ 

In any case, what contingent or empirical details might happen to hold true in any given world seems absolutely irrelevant for questions about what {{ideas}} do or do not represent distinct {{concepts}}. The idea of a watermelon is a distinct concept from the idea of depression, for example. Those being distinct concepts, even God could not create a world in which watermelons and depression were literally identical. He couldn’t create a world made of watermelons and get depression without adding it in as an extra, and he couldn’t create a world made of depression and get watermelons without adding them in as an extra. Not even {{God}} could violate rules of identity. 

And so here: if zombie persons, or even zombie atoms, are even something we can imagine in theory, then the concepts are distinct—and if the concepts are distinct, then not even God could create a world in which geometric, structural properties and qualitative, subjective experiences were simply “identical” any more than he could create a world where watermelons and depression were simply “identical” (not even if he created a world where watermelons had binded consciousnesses and experienced lifelong depressions would they actually be “identical!” They would merely be held by a contingent, added law to be universally correlated—which would still not make them “identical.”)

[1] Since, on this one, after we discover EVERYTHING about the mailman and the man at the bar, this kind of example allows the approach of trying to appropriate Kripke and say that it’s an “a posteriori necessity” that the two are identical even if it wasn’t an “a priori necessity” that they be—with the conceivability argument only establishing the lack of an “a priori necessity.” I think there are specific fallacies in the way that Kripke gets adopted here, and especially which render his actual idea inapplicable to the mind–body problem entirely (Kripke himself explicitly opposed “identity theory” approaches to it, no less!), but I grant an analogy that plays into this approach here to be as charitable to the identity theorist’s intuitions as possible (even if I consider them abhorrently deeply flawed).

Finally, we come to your second approach. I mentioned earlier that it was difficult to see exactly which approach to relating the structural properties and the qualitative and subjective aspects of conscious experience you were endorsing. I’ll work around that by filling in the details myself about the two different notions you might be trying to imply, and address each in turn.

In the opening sentences, you say that “[If panpsychism is true] then strictly speaking all consciousness, and only consciousness, has causal efficacy….” Now, granting we agree for all the reasons above that we should reject a claim of any literal identity between the raw subjective feels that an electron internally has and its velocity (etc.) [1], and granting that there is indeed a conceptual distinction, this can only mean that it is in fact the electron’s qualitative sensation AS SUCH that actually has causal efficacy. This returns us, once again, to my “physical properties emergent from conscious base” conception of panpsychism, and my response to this one was that it leads to an equally absurd, inverted “hard problem” in which we would need to explain how an electron’s physical property of velocity emerges from the subjective, qualitative details of its private phenomenal experience. But if the explanatory gap condemns materialism, then it condemns this approach too—and if that is the line you actually wish to take, then that criticism—and not the threat of epiphenomenalism—is the one you actually need to address in order to try to defend this option. The explanatory gap doesn’t get any easier—indeed, it does not change a whit—if we try to bridge it by jumping across the expanse of its canyon from the east tower over to the west tower instead of jumping from the west tower to the one to the east: it’s an identical and equally impassable distance to gap either way. 

([1] If we don’t, then the discussion ought to stay at the previous step anyway, and this whole thread of the discussion is moot unless and until those points are resolved one way or another first. The “identity” claim is still the train’s first logical stop, in other words.)

Alternatively, you might be read as trying to defuse the threat that the “property dualist” formulation of panpsychism would lead to epiphenomenalism in order to defend that option. In that case, however, a property dualist panpsychism combined with causal closure of the physical (e.g., mathematically describable structural and spatiomotory causally disposing) dimensions of reality entails that the only causality that is actually possible in the world is owed to those mathematically describable, abstract, geometric properties. 

Crucially, if this is the route you wish to pick, then it simply is not within your rights to say that “all consciousness, and only consciousness, has causal efficacy….” What you defend, in this case, is a vision on which causal efficacy is not owed to the subjective, qualitative aspects as such, but to the mathematically describable spatiostructural dimensions and spatiomotory dispositions of any given entity. You have to commit—you must either defend this line of approach, or you must choose the line of approach which I’ve argued leads to the disqualifyingly absurd “hard problem” of the “strong emergence” of physical properties out of the ontological base of subjective qualia. 

In any case, this returns us to my original account, on which the physical forces that make up your floor don’t {do} any of the things that they physically {do} in {virtue of} the fact that they’re blue; rather, they appear blue to you (at least in large part) because of their physical, structural composition—but it is all and only the mathematically describable {aspects} of these entities to which the actual procession of causality is owed. 

However, for this kind of property dualism in general (whether wedded to the claim of panpsychism, which has it that these properties exist in some degree everywhere, or not), the problems are two–fold: (1) We are presumably here because we rejected, and thus sought a suitable replacement for, the previously considered identity theory. Given that identity theory will fail immediately, in response to nothing more than simple conceptual clarification, and given that “strong emergence” fails for the same reasons that the identity theory fails, if the identity theory fails, why something like subjective experience should tag along for the spatiotemporal ride with the mathematically describable structural and spatiomotory causally disposing properties is still left both (a) unexplained and (b) classified within a status that would seem to leave us with the reasonable expectation that an explanation should be found. 

But more importantly, (2) — this addresses your attempt to counter the threat of epiphenomenalism, because the threat of epiphenomenalism that looms over the attempt to formulate a “property dualist” panpsychism simply does not follow from considerations of substrate–independency, but rather from this:: 

We have—again—arrived here in the attempt to identify a suitable replacement for the immediately disqualifying definitional failure of “identity theories.” If we have arrived here in the first place, we accept that the subjective, qualitative aspects of reality are not just literally identical with the mathematically describable, structural and spatiomotory causally disposing properties. We also accept—and have arrived here precisely BECAUSE we accept—that the latter do not, and in principle can not, “emerge” from the former. Thus, the subjective, qualitative aspects of reality are ontological “extras.” But all of the physical events that actually take place in reality—per the assumption of causal closure—are owed to those mathematically describable properties, and NOT to the subjective, qualitative aspects of reality *as such*—IN PRINCIPLE! And there is no way out of that conclusion without some fundamental reworking of our premises, because they lead directly to epiphenomenalism without escape.

If it helps, I’ll quote the full length of my address of epiphenomenalism in my entry (IV) [http://zombiemeditations.com/2015/03/26/consciousness-iv/] to show exactly why I think it is such a hopelessly disqualifying conclusion that it immediately refutes any premises that would seem to entail it: 

“Causal closure of the physical domain is a principle which almost all physicalists will accept in some form. In Jaegwon’s words, what the principle states is that: “if we trace the causal ancestry of a physical event we need never go outside the physical domain.” What Jaegwon Kim realized was that if we combine this claim with the realization that subjective experience can’t be reduced to or accounted for in terms of physical mechanism, then we end up with a description of reality known as epiphenomenalism, on which—roughly—experiences more or less dangle off the edges of the world before simply falling off (I’ll explain this more in a minute). Jaegwon’s description of the state of play was thus that the choices are to either claim that subjective experience can be reduced to physical description (which is what he had, by then, saw the same compelling reasons to reject which I am outlining here), reject the principle of causal closure, or else accept epiphenomenalism—and so, refusing to reject the principle of causal closure and hoping to find in it “something near enough” to physicalism to rescue him from the stark position which admitting the clear point that subjective experience cannot be reduce to physical description had placed him in, Kim settled for epiphenomenalism. It isn’t physicalism strictly speaking, because conscious experiences—which seem to dangle awkwardly somehow off of some causally irrelevant metaphysical edge of the world—can’t be explained, but isn’t it still “something close enough”?

… I think Kim’s description of the state of play is absolutely correct. However, I also think—as I will show in a moment—that epiphenomenalism is a viewpoint that we can conclusively refute and rule out completely. …

One of the easiest ways to explain an epiphenomenalist relationship is by example. If you stand in front of a mirror and jump up and down, your reflection is an epiphenomena of your actual body. What this means is that your body’s jump is what causes your reflection to appear to jump—your body’s jump is what causes your real body to fall—and your body’s fall is what causes your reflection to fall. It may seem to be the case that your reflection’s apparent jump is what causes your reflection to appear to fall, but this is purely an illusion: your reflection doesn’t cause anything in this story; not even its own future states. If we represent physical states with capital letters, states of experience with lower–case letters, and causality with arrows, then a diagram would look something like this: [image]

… Thomas Huxley, not the first to espouse the view but the first to give it a name, described it by saying that consciousness is like the steam–whistle sound blowing off of a train that contributes nothing to the continued motion of the train itself. We shouldn’t fail to realize how extreme the dehumanization of this view is, even still, despite the fact that it acknowledges conscious experiences as real: if this is true, then nobody ever chooses a partner because they are experiencing love; nobody ever fights someone because they are experiencing anger; nobody ever even winces because they are experiencing pain. Rather, a blind inert physical state moves by causal necessity from one state to the next; and it is the meaningless motion of these blind inert forces by causal necessity that explains everything—conscious experiences just happen to incidentally squirt out over the top of these motions as a byproduct, and you are, in effect, a prisoner locked inside the movie in your head with your arms and legs removed and absolutely no influence or control whatsoever over what does or does not happen inside of it. In the words of Charles Bonnett writing in 1755, “the soul is a mere spectator of the movements of its body.”

I would ask you to contemplate the severity of what might result if someone were to actually take this proposal seriously and really honestly begin to look at life and their own conscious existence in this horrific and dehumanized way, but according to the claim of epiphenomenalism, believing that epiphenomenalism is true never has any causal effect on anyone’s physical behavior—nor on any of their future mental states—in the first place either. A series of blind, inert physical events leads to their brain responding physically to the input of symbols and lines (and it is only a mere epiphenomena of this that they have any experience of “understanding their meaning,” but any “ideas” contained therein—as such—would simply in principle have no ability to play any further causal role in anything further whatsoever, either of the individual’s future conscious beliefs or their future physical behavior); and from here a purely physical sequence of physical causation leads to further physical states (which then happen to give off more epiphenomena in turn). On this view, the fact that pain even feels “painful” is a mere coincidence; for it is not because we feel pain and dislike it that we ever recoil away from a painful stimuli: one physical brain event produces another, and it is only a mere unexplained coincidence that what the first physical brain event happens to give off like so much irrelevant steam is a feeling that just so happens to be painful in particular. 

It literally could just as well have been the case that slicing into our skin with a knife would produce the sensation that we currently know in the world as it is as “the taste of strawberries,” and the physical world (according to epiphenomenalism) would proceed in just exactly the same way as it does now. This would be true because: (1) epiphenomenalism admits that conscious experiences are something over and above physical events, and we do not know why particular conscious experiences are linked with particular physical events (since the former are not logically predictable from the latter given that claims that it “emerges” are acknowledged by definition by epiphenomenalism to fail), and (2) none of them play any causal role in anything anyway. Our conscious lives could have consisted of one long feeling orgasm, or one long miserable experience of pain, or one long sounding “C” note combined with the taste of blueberries and a feeling of slight melancholy, and again, everything in the physical universe would have proceeded in exactly the same way it does now. And it is only a coincidence of whatever extra rule specifies that particular conscious experiences superfluously ‘squirt out’ and dissipate into the cosmic aether like steam that our world happens to be otherwise. …

… Unfortunately, while most people—including philosophers—are content to stop here and reject the view for sheer counter–intuitiveness alone, philosophy of mind has been somewhat lazy at producing actual logical objections to it. Actual refutations of epiphenomenalism often aren’t very well known, but there is one that is absolute and undeniable and refutes even the possibility that anything like epiphenomenalism could possibly be true completely once and for all. That is: if epiphenomenalism were true, no one would ever be able to write about it. In fact: no one would ever be able to write—nor think—about consciousness in general. No one would ever once in the history of universe have had a single thought about a single one of the questions posed by philosophy of mind. Not a single philosophical position on the nature of consciousness, epiphenomenalist or otherwise, would ever have been defined, believed, or defended by anyone. No one would even be able to think about the fact that conscious experiences exist.

And the reason for that, in retrospect, is quite plain to see: on epiphenomenalism, our thoughts are produced by our physical brains. But our physical brains, in and of themselves, are just machines—our conscious experiences exist, as it were in effect, within another realm, where they are blocked off from having any causal influence on anything whatsoever (even including the other mental states existing within their realm, because it is some physical state which determines every single one of those). But this means that our conscious experiences can never make any sort of causal contact with the brains which produce all our conscious thoughts in the first place. And thus, our brains would have absolutely no capacity to formulate any conception whatsoever of their existence—and since all conscious thoughts are created by brains, we would never experience any conscious thoughts about consciousness. For another diagram, if we represent causality with arrows, causal closure with parentheses, physical events with the letter P and experiences with the letter e, the world would look something like this:

… e1 ⇠ (((P⇆P))) ⇢ e2 …

Everything that happens within the physical world—illustrated by (((P⇆P)))—would be wholly and fully kept and contained within the physical world, where conscious experiences as such do not reside; the physical world is Thomas Huxley’s train which moves whether the whistle on top blows steam or not. And e1 and e2 float off of the physical world—for whatever reason—and then merely dissipate into nothingness like steam, with no capacity in principle for making any causal inroads back into the physical dimension of reality whatsoever. This follows straightforwardly as an inescapable conclusion of the very premises which epiphenomenalism defines itself by. But since the very brains which produce all our experienced thoughts are contained within (((P⇆P))), in order to have any experienced thought about conscious experience itself, these (per epiphenomenalism) would have to be the epiphenomenal byproducts of a brain state that is somehow reflective or indicative of conscious experience. But brain states, again because per epiphenomenalism they belong to the self–contained world inside (((P⇆P))) where no experiences as such exist, are absolutely incapable in principle of doing this.

To refer back to our original analogy whereby epiphenomenalism was described by the illustration of a person jumping up and down in front of a mirror, then: it would be as if the mirror our brains were jumping up and down in front of were shielded inside of a black hole in a hidden dimension we couldn’t see. Our real bodies [by analogy, our physical brains] would never be able to see anything happening inside that mirror. And therefore, they would never be able to think about it or talk about it. And therefore, we would never see our reflections [by analogy, our consciously experienced minds] thinking or talking about the existence of reflections, because our reflections could only do that if our real bodies were doing that, and there would be absolutely no way in principle that our real bodies ever could.

The fact that we do this, then—the fact that we do think about consciousness as such, and the fact that we write volumes and volumes and volumes and volumes philosophizing about it, and the very fact that we produce theories (including epiphenomenalism itself) about its relation to the physical world in the first place—proves absolutely that whatever the mechanism may be, conscious experiences somehow most absolutely do in fact have causal influence over the world. What we have here is a rare example of a refutation that proceeds solely from the premises of the position itself, and demonstrates an internal inconsistency.

But Jaegwon Kim has identified the possible options for us: either experiences and physical events are just literally identical (which even Kim himself rejects, for good reasons we have outlined here), or else epiphenomenalism is true (which Jaegwon Kim accepts, but which the simple argument outlined just now renders completely inadmissible)—or else the postulate of the causal closure of the physical domain is false—and conscious experience is both irreducible to and incapable of being explained in terms of blind physical mechanisms, and possesses unique causal efficacy over reality all in its own right.”